sotomayor cases

516
Pursuant to Federal Rule of Appellate Procedure 43(c)(2), Governor David Paterson is * automatically substituted for former Governor Eliot Spitzer as a defendant in this case. 07-0581-cv Maloney v. Cuomo UNITED STATES COURT OF APPEALS FOR THE SECOND CIRCUIT ______________ August Term, 2008 (Argued: December 15, 2008 Decided: January 28, 2009) Docket No. 07-0581-cv ______________ JAMES M. MALONEY, Plaintiff-Appellant, —v.— ANDREW CUOMO, in his official capacity as Attorney General of the State of New York, DAVID PATERSON, in his official capacity as Governor of the State of New York, KATHLEEN A. RICE, in her official capacity as District Attorney of the County of Nassau, and their successors, * Defendants-Appellees. ______________ B e f o r e: POOLER, SOTOMAYOR, and KATZMANN, Circuit Judges. ______________ Appeal from a judgment of the United States District Court for the Eastern District of New York (Spatt, J.) dated January 17, 2007, granting defendants-appellees Andrew Cuomo and David Paterson’s motion to dismiss and defendant-appellee Kathleen A. Rice’s motion for judgment on the pleadings, and from an order dated May 14, 2007, denying plaintiff-appellant’s motion for reconsideration. Affirmed.

Upload: maldef

Post on 05-Dec-2014

3.660 views

Category:

Business


1 download

DESCRIPTION

 

TRANSCRIPT

Page 1: Sotomayor Cases

Pursuant to Federal Rule of Appellate Procedure 43(c)(2), Governor David Paterson is*

automatically substituted for former Governor Eliot Spitzer as a defendant in this case.

07-0581-cvMaloney v. Cuomo

UNITED STATES COURT OF APPEALS

FOR THE SECOND CIRCUIT

______________

August Term, 2008

(Argued: December 15, 2008 Decided: January 28, 2009)

Docket No. 07-0581-cv

______________

JAMES M. MALONEY,

Plaintiff-Appellant,

—v.—

ANDREW CUOMO, in his official capacity as Attorney General of the State of New York, DAVID

PATERSON, in his official capacity as Governor of the State of New York, KATHLEEN A. RICE, inher official capacity as District Attorney of the County of Nassau, and their successors,*

Defendants-Appellees.

______________

B e f o r e:

POOLER, SOTOMAYOR, and KATZMANN, Circuit Judges.

______________

Appeal from a judgment of the United States District Court for the Eastern District ofNew York (Spatt, J.) dated January 17, 2007, granting defendants-appellees Andrew Cuomo andDavid Paterson’s motion to dismiss and defendant-appellee Kathleen A. Rice’s motion forjudgment on the pleadings, and from an order dated May 14, 2007, denying plaintiff-appellant’smotion for reconsideration. Affirmed.

Page 2: Sotomayor Cases

There are two sections of the New York Penal Law numbered 265.00(14).1

2

______________

JAMES M. MALONEY, appearing pro se, for Plaintiff-Appellant.

KAREN HUTSON, Deputy County Attorney (Lorna B.Goodman, County Attorney, on the brief) for Defendant-Appellee Kathleen A. Rice, Nassau County DistrictAttorney, Mineola, N.Y.

______________

PER CURIAM:

Plaintiff-appellant James Maloney was arrested at his home on August 24, 2000, and

charged with possessing a chuka stick in violation of N.Y. Penal Law § 265.01(1). A “chuka

stick” (or “nunchaku”) is defined as

any device designed primarily as a weapon, consisting of two or more lengths of arigid material joined together by a thong, rope or chain in such a manner as toallow free movement of a portion of the device while held in the hand and capableof being rotated in such a manner as to inflict serious injury upon a person bystriking or choking.

Id. § 265.00(14). This charge was dismissed on January 28, 2003, and Appellant pleaded guilty1

to one count of disorderly conduct. As part of the plea, he agreed to the destruction of the

nunchaku seized from his home.

Appellant filed the initial complaint in this action on February 18, 2003, and then an

amended complaint on September 3, 2005, seeking a declaration that N.Y. Penal Law §§ 265.00

through 265.02 are unconstitutional insofar as they punish possession of nunchakus in one’s

home. The district court dismissed the amended complaint as against the New York State

Page 3: Sotomayor Cases

Appellant makes no argument in his brief concerning the district court’s dismissal of his2

claims against the Attorney General and the Governor. We therefore deem any challenges to thataspect of the district court’s judgment waived. See Yueqing Zhang v. Gonzales, 426 F.3d 540,541 n.1 (2d Cir. 2005).

3

Attorney General and the Governor for lack of standing, concluding that neither official is

responsible for enforcing the statutes at issue. The district court granted defendant Nassau

County District Attorney Kathleen Rice’s motion for judgment on the pleadings in relevant part

because the Second Amendment does not apply to the States and therefore imposed no

limitations on New York’s ability to prohibit the possession of nunchakus. Appellant moved for

reconsideration on the ground that the district court had failed to consider certain other claims

raised in his amended complaint; the district court denied that motion.

On appeal, Appellant challenges only the district court’s dismissal of his claims against

Rice. He argues, inter alia, that New York’s statutory ban on the possession of nunchakus2

violates (1) the Second Amendment because it infringes on his right to keep and bear arms, and

(2) the Fourteenth Amendment because it lacks a rational basis. Neither of these arguments has

any merit.

The Second Amendment provides: “A well regulated Militia, being necessary to the

security of a free State, the right of the people to keep and bear Arms, shall not be infringed.”

U.S. Const. amend. II. The Supreme Court recently held that this confers an individual right on

citizens to keep and bear arms. See District of Columbia v. Heller, 128 S. Ct. 2783, 2799 (2008).

It is settled law, however, that the Second Amendment applies only to limitations the federal

government seeks to impose on this right. See, e.g., Presser v. Illinois, 116 U.S. 252, 265 (1886)

(stating that the Second Amendment “is a limitation only upon the power of congress and the

Page 4: Sotomayor Cases

4

national government, and not upon that of the state”); Bach v. Pataki, 408 F.3d 75, 84, 86 (2d

Cir. 2005) (holding “that the Second Amendment’s ‘right to keep and bear arms’ imposes a

limitation on only federal, not state, legislative efforts” and noting that this outcome was

compelled by Presser), cert. denied, 546 U.S. 1174 (2006). Heller, a case involving a challenge

to the District of Columbia’s general prohibition on handguns, does not invalidate this

longstanding principle. See Heller, 128 S. Ct. at 2813 n.23 (noting that the case did not present

the question of whether the Second Amendment applies to the states). And to the extent that

Heller might be read to question the continuing validity of this principle, we “must follow

Presser” because “[w]here, as here, a Supreme Court precedent ‘has direct application in a case,

yet appears to rest on reasons rejected in some other line of decisions, the Court of Appeals

should follow the case which directly controls, leaving to the Supreme Court the prerogative of

overruling its own decisions.’” Bach, 408 F.3d at 86 (quoting Rodriguez de Quijas v.

Shearson/Am. Express, Inc., 490 U.S. 477, 484 (1989)) (alteration marks omitted); see also State

Oil Co. v. Khan, 522 U.S. 3, 20 (1997). Thus, N.Y. Penal Law §§ 265.00 through 265.02 do not

violate the Second Amendment.

The Fourteenth Amendment similarly provides no relief for Appellant. “Legislative acts

that do not interfere with fundamental rights or single out suspect classifications carry with them

a strong presumption of constitutionality and must be upheld if ‘rationally related to a legitimate

state interest.’” Beatie v. City of New York, 123 F.3d 707, 711 (2d Cir. 1997) (quoting City of

Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 440 (1985)). We will uphold legislation if

we can identify “some reasonably conceivable state of facts that could provide a rational basis for

the legislative action. In other words, to escape invalidation by being declared irrational, the

Page 5: Sotomayor Cases

5

legislation under scrutiny merely must find some footing in the realities of the subject addressed

by the law.” Id. at 712 (internal quotation marks and citations omitted).

The legislative history of section 265.00 makes plain that the ban on possession of

nunchakus imposed by section 265.01(1) is supported by a rational basis. Indeed, as Appellant

concedes, when the statute was under consideration, various parties submitted statements noting

the highly dangerous nature of nunchakus. For example, New York’s Attorney General, Louis J.

Lefkowitz, asserted that nunchakus “ha[ve] apparently been widely used by muggers and street

gangs and ha[ve] been the cause of many serious injuries.” Mem. from Attorney Gen. Louis J.

Lefkowitz to the Governor (Apr. 8, 1974). And the sponsor of the bill, Richard Ross, stated that

“[w]ith a minimum amount of practice, [the nunchaku] may be effectively used as a garrote,

bludgeon, thrusting or striking device. The [nunchaku] is designed primarily as a weapon and

has no purpose other than to maim or, in some instances, kill.” See N.Y. Penal Law § 265.00,

practice commentary, definitions (“Chuka stick”) (quoting Letter of Assemblyman Richard C.

Ross to the Counsel to the Governor (1974)).

Appellant does not dispute that nunchakus can be highly dangerous weapons. Rather, his

principal argument is that section 265.01(1) prevents martial artists from using nunchakus as part

of a training program. But the fact that nunchakus might be used as part of a martial-arts training

program cannot alter our analysis. Where, as here, a statute neither interferes with a fundamental

right nor singles out a suspect classification, “we will invalidate [that statute] on substantive due

process grounds only when a plaintiff can demonstrate that there is no rational relationship

between the legislation and a legitimate legislative purpose.” Beatie, 123 F.3d at 711. Appellant

has not carried this burden. Consequently, in light of the legislature’s view of the danger posed

Page 6: Sotomayor Cases

6

by nunchakus, we find that the prohibition against the possession of nunchakus created by N.Y.

Penal Law § 265.01(1) is supported by a rational basis.

We have considered Appellant’s remaining arguments and find them to be without merit.

Accordingly, for the foregoing reasons, the judgment of the district court is hereby AFFIRMED.

Appellant’s pending motions to strike defendant Kathleen Rice’s brief and material in her July

28, 2008 Rule 28(j) letter are hereby DENIED.

Page 7: Sotomayor Cases

06-4996-cv

Ricci v. DeStafano

UNITED STATES COURT OF APPEALS

FOR THE SECOND CIRCUIT

August Term, 2007

(Argued: December 10, 2007 Decided: June 9, 2008)

Docket No. 06-4996-cv

FRANK RICCI, MICHAEL BLATCHLEY, GREG BOIVIN, GARY CARBONE, MICHAELCHRISTOFORO, RYAN DIVITO, STEVEN DURAND, WILLIAM GAMBARDELLA,BRIAN JOOSS, JAMES KOTTAGE, MATTHEW MARCARELLI, THOMAS J. MICHAELS,SEAN PATTON, CHRISTOPHER PARKER, EDWARD RIORDAN, KEVIN ROXBEE,TIMOTHY SCANLON, BENJAMIN VARGAS, JOHN VENDETTO AND MARKVENDETTO,

Plaintiffs-Appellants,

-v-

JOHN DESTEFANO, KAREN DUBOIS-WALTON, THOMAS UDE JR., TINA BURGETT,BOISE KIMBER, MALCOM WEBER, ZELMA TIRADO AND CITY OF NEW HAVEN,

Defendants-Appellees.

Before: POOLER, SACK and SOTOMAYOR, Circuit Judges.

AFFIRMED.

Page 8: Sotomayor Cases

2

KAREN LEE TORRE, Attorney, New Haven, CT, for Plaintiffs-Appellants.

RICHARD A. ROBERTS, Attorney, Cheshire, CT (Nicole C. Chomiak, Stacey L. Pitcher, andTodd J. Richardson, on the brief), for Defendants-Appellees.

PER CURIAM:

We withdraw our Summary Order of February 15, 2008. Ricci v. DeStefano, 2008 U.S.

App. LEXIS 3293, 2008 WL 410436 (2d Cir. Feb. 15, 2008).

Plaintiffs appeal from a judgment of the United States District Court for the District of

Connecticut (Arterton, J.) granting the defendants’ motion for summary judgment on all counts.

We affirm, for the reasons stated in the thorough, thoughtful, and well-reasoned opinion

of the court below. Ricci v. DeStefano, 2006 U.S. Dist. LEXIS 73277, 2006 WL 2828419 (D.

Conn., Sept. 28, 2006). In this case, the Civil Service Board found itself in the unfortunate

position of having no good alternatives. We are not unsympathetic to the plaintiffs’ expression

of frustration. Mr. Ricci, for example, who is dyslexic, made intensive efforts that appear to have

resulted in his scoring highly on one of the exams, only to have it invalidated. But it simply does

not follow that he has a viable Title VII claim. To the contrary, because the Board, in refusing to

validate the exams, was simply trying to fulfill its obligations under Title VII when confronted

with test results that had a disproportionate racial impact, its actions were protected.

CONCLUSION

The judgment of the district court is AFFIRMED.

Page 9: Sotomayor Cases

06-4996-cvRicci v. DeStefano

- - - - - - - - - - - - - - - - - - - -X

- v.-

- - - - - - - - - - -X

Docket No. 06-4996-cv

Decided: June 9, 2008Order issued: June 12, 2008)

(Argued: December 10, 2007

UNITED STATES COURT OF APPEALS

August Term, 2007

FOR THE SECOND CIRCUIT

Plaintiffs-Appellants,

Defendants-Appellees

FRANK RICCI, MICHAEL BLATCHLEY, GREGBOIVIN, GARY CARBONE, MICHAELCHRISTOFORO, RYAN DIVITO, STEVENDURAND, WILLIAM GAMBARDELLA, BRIANJOOSS, JAMES KOTTAGE, MATTHEWMARCARELLI, THOMAS J. MICHAELS, SEANPATTON, CHRISTOPHER PARKER, EDWARDRIORDAN, KEVIN ROXBEE, TIMOTHY SCANLON,BENJAMIN VARGAS, JOHN VENDETTO AND MARKVENDETTO,

JOHN DESTEFANO, KAREN DUBOIS-WALTON,THOMAS UDE JR., TINA BURGETT, BOIKIMBER, MALCOM WEBER, ZELMA TIRADO ANDCITY OF NEW HAVEN,

1234567

89

10111213141516171819202122232425262728293031323334353637

38 FOR PLAINTIFFS-APPELLANTS: KAREN LEE TORRE, New Haven, CT3940 FOR DEFENDANTS-APPELLEES: RICHARD A. ROBERTS {Nicole C.

zgorelik
Typewritten Text
06-4996-cv Ricci v. DeStefano
Page 10: Sotomayor Cases

123456

7

Chomiak, Stacey L. Pitcher, andTodd J. Richardson,~~~), Cheshire, CT

ORDER

After disposition of this appeal by summary order dated

8 February 15, 2008, an active judge of Court requested a

9 poll on whether to rehear the case in banco A poll on

10 whether to rehear case in banc was conducted among the

11 active judges of the Cou After the poll was concluded,

12 on June 9, 2008, the original three-judge panel withdrew the

13 summary order and filed a per curiam opinion; no subsequent

14 poll has been requested. Because a majority of the

15 court's act judges voted to deny rehearing in banc,

16 rehearing is hereby DENIED.

17 Judges Calabresi, Straub, Pooler, Sack, Sotomayor,

18 Katzmann, and B.D. ker concur in the denial of rehearing

19 in banco Chief Judge Jacobs and Judges Cabranes, Raggi,

20 Wesley, Hall and vingston dissent from t denial of

21 rehearing in banco

22 With this order, Judge Katzmann is filing a

23 concurring opinion, in which Judges Pooler, Sack, Sotomayor

24 and B.D. Parker joins; Judge B.D. Parker is filing a

25 concurring opinion, in which Judges Calabresi, Pooler, Sack

2

Page 11: Sotomayor Cases

1 and Sotomayor join; and Judge Cabranes is filing a

2 dissenting opinion, in which Chief Judge Jacobs and Judges

3 Raggi, Wesley, Hall and Judge vingston join.

4 Other opinions may be filed with respect to this case,

5 concurring or dissenting in the denial of in bane review.

3

Page 12: Sotomayor Cases

KATZMANN, Circuit Judge, with whom Judge POOLER, Judge SACK, Judge SOTOMAYOR, andJudge B.D. PARKERjoin, concurring in the denial of rehearing en banc:

I concur in the denial ofrehearing en bane, consistent with our Circuit's longstandingtradition of general deference to panel adjudication a tradition which holds whether or not thejudges of the Court agree with the panel's disposition of the matter before it. Throughout ourhistory, we have proceeded to a full hearing en banc only in rare and exceptional circumstances.See Wilfred Feinberg, Unique Customs and Practices o/The Second Circuit, 14 Hofstra L. Rev.297, 311-12 (1986). The Supreme Court now has before it a petition for certiorari in this case,which I recognize presents difficult issues. As the Supreme Court de~ides whether to grantcertiorari, it has for its review the district court's opinion, the panel's per curiam opinion, andopinions concurring with and dissenting from the decision denying rehearing en bane. The issuesare therefore sharply defined for the Supreme Court's consideration of whether to grantcertiorari.

Page 13: Sotomayor Cases

BARRINGTON D. PARKER. Circuit Judge. with whom Judge CALABRESI. Judge POOLER. JudgeSACK, and Judge SOTOMAYOR join. concurring in the denial of rehearing en bane:

At the heart of the dissent from the denial of rehearing en bane is the assertion that there

was no Supreme Court or circuit law to guide this district court. or future district courts faced

with similar claims. I disagree. The district court correctly observed that this case was unusual.

Nonetheless. the district court also recognized that there was controlling authority in our

decisions-among them. Hayden v. County ofNassau. 180 F.3d 42 (2d Cir. 1999) and Bushey v.

N. Y. State Civil Servo Comm 'n, 733 F.2d 220 (2d Cir. 1984). cert. denied, 469 U.S. 1117 (1985).

These cases clearly establish for the circuit that a public employer, faced with a prima facie case

ofdisparate-impact liability under Title VII, does not violate Title VII or the Equal Protection

Clause by taking facially neutral, albeit race-conscious, actions to avoid such liability.

Insofar as the dissent suggests that the plaintiffs produced evidence of a racial

classification or the imposition of a quota, I think it entirely mistaken. Although the City acted

out of a concern that certifying the exam results would have an adverse impact on minority

candidates-and although, as the panel noted in its decision, the result was understandably

frustrating for applicants who passed the test-the City's response, to decline to certify any of the

exams, was facially race-neutral. The City did not classify or confer any actual benefit on

applicants on the basis of race. The dissent's citations to Adarand Constructors, Inc. v. Pena,

515 U.S. 200(1995), and CityofRichmondv. J.A. Croson Co., 488 U.S. 469 (1989), are

1

Page 14: Sotomayor Cases

therefore inapposite.! See Hayden, 180 F.3d at 49 (distinguishing those cases as "concerned with

select affinnative action tools, such as quota systems, set-aside programs, and differential scoring

cutoffs, which utilize express racial classifications and which prevent non-minorities from

competing for specific slots or contracts.").

Because there was no racial classification, the plaintiffs bore the burden of persuasion on

the issue of discriminatory purpose. Jana-Rock Constr., Inc. v. N. Y State Dep 't ofEcon. Dev.,

438 F.3d 195,204 (2d Cir. 2006). Here, however, there was no evidence of a discriminatory

purpose; according to the record evidence, the City was motivated by a desire to comply with,

and avoid liability under, Title VII and its implementing regulations. See Bushey, 733 F.2d at

226 ("It is settled that voluntary compliance is a preferred means of achieving Title VII's goal of

eliminating employment discrimination." (internal quotation marks and alteration omitted)); see

also Hayden, 180 F.3d at 51 ("A desire to reduce the adverse impact on [minority]

applicants ... is not analogous to an intent to discriminate against non-minority candidates.").

I think the dissent also quite unfairly caricatures the district court's evaluation of the

plaintiffs' Title VII claim: "Under the District Court's rationale, it appears that any race-based

employment decision undertaken to avoid a threatened or perceived Title VII lawsuit is itself

I It may be worth noting that the Croson Court based its decision partly on the fact that"[t]here [was] nothing approaching a prima facie case of a constitutional or statutory violation byanyone in the Richmond construction industry." Croson, 488 U.S. at 500. Here, by contrast, theCity was faced with a prima facie case of a violation of Title VII. See Gulino v. N. Y State Educ.Dep't, 460 F.3d 361,382 (2d Cir. 2006) (defining prima facie case of disparate-impact liabilityunder Title VII); Robinson v. Metro-North Commuter R.R. Co., 267 F.3d 147, 160 (2d Cir. 2001)(same); see also 42 U.S.c. § 2000e-2(k) (codifying the disparate-impact theory ofliability andlegislatively overruling Wards Cove Packing Co. v. Atonio, 490 U.S. 642, 659 (1989)).

2

Page 15: Sotomayor Cases

immune from scrutiny under Title VII." This is simply not the case. Prior to reaching its

conclusion, the district court assessed whether the examination results demonstrated a

statistically disproportionate adverse racial impact under the EEOC Guidelines and whether the

City had presented evidence to support its belief that less discriminatory alternatives to this

particular test existed. This analysis shows that, contrary to the dissent's suggestion, the district

court did not rubber stamp the City's proffered non-discriminatory reason for not certifying the

exam results.

Moreover, I hardly think that in order to decline to certify the exam results, the City was

required to prove, through a validation study or some other means, that its own tests were not

"job related for the position in question and consistent with business necessity," 42 U.S.C.

§ 2000e-2(k)(I)(A)(i) (defining affirmative defense to prima/acie case of disparate impact

violation). In fact, our case law explicitly rejects that proposition. See Bushey, 733 F.2d at 226

(disagreeing with the assertion that "before adopting remedial measures" the employer must

"prove that [the] prima face case [ofa disparate-impact Title VII violation] was not rebuttable

through job-related explanations").

I also disagree with the dissent's view that en banc review is warranted because the

district court analyzed the plaintiffs claims using the McDonnell Douglas pretext test rather than

the Price Waterhouse mixed-motive test. See Price Waterhouse v. Hopkins, 490 U.S. 228

(1989); McDonnell Douglas Corp. v. Green, 411 U.S. 792 (1973). As the dissent is well aware,

the plaintiffs did not argue the mixed-motive theory; a non-party raised it in an amicus brief.

"Although an amicus brief can be helpful in elaborating issues properly presented by the parties,

3

Page 16: Sotomayor Cases

it is normally not a method for injecting new issues into an appeal, at least in cases where the

parties are competently represented by counsel." Universal City Studios, Inc. v. Corley, 273 F.3d

429,445 (2d Cir. 2001); see also Bano v. Union Carbide Corp., 273 F.3d 120, 127 n.5 (2d Cir.

2001) (same).

Finally, the dissent suggests that the panel's per curiam opinion inappropriately adopted

the reasoning set forth in the district court's opinion, one that the panel recognized was

"thorough, thoughtful, and well-reasoned." The adherence of a Court of Appeals to the decision

and reasoning of a district court is anything but novel. In fact, the practice pre-dates the formal

establishment of this Court in 1891 by at least fifty years. United States v. Libellants &

Claimants ofthe Schooner Amistad, 40 U.S. 518,590 (1841) ("The Circuit Court, by a mere pro

forma decree, affirmed the decree of the [Connecticut] District Court .... And from that decree

the present appeal has been brought to this Court."). This Court has followed this practice on

numerous occasions in appeals covering myriad issues. See, e.g., In re Bankers Trust Co., 450

F.3d 121, 123 (2d Cir. 2006) (per curiam); Murphy ex rei. Estate ofPayne v. United States, 427

F.3d 158,159 (2d Cir. 2005) (per curiam); In re Red Dot Scenic, Inc., 351 F.3d 57, 58 (2d Cir.

2003) (per curiam); United States v. Gluzman, 154 F.3d 49,50 (2d Cir. 1998); Trans World

Airlines, Inc. v. Sinicropi, 84 F.3d 116, 116 (2d Cir.) (per curiam), cert. denied, 519 U.S. 949

(1996).

The plaintiffs were entitled to a careful and thoughtful review of their claims. The panel

decided that the district court had given them just that, and thus adopted the district court's

reasoning in its per curiam opinion. Nothing more is required.

4

Page 17: Sotomayor Cases

JOSE A. CABRANES, CircuitJudge, with whom ChiefJudge JACOBS, Judge RAGGI, Judge WESLEY, Judge

HALL, and Judge LIVINGSTON join, dissenting:

This appeal raises important questions of fIrst impression in our Circuit-and indeed, in the

nation-regarding the application of the Fourteenth Amendment's Equal Protection Clause and Title

VII's prohibition on discriminatory employment practices. At its core, this case presents a straight­

forward question: Maya municipal employer disregard the results of a qualifying examination, which

was carefully constructed to ensure race-neutrality, on the ground that the results of that examination

yielded too many qualified applicants of one race and not enough of another? In a path-breaking

opinion, which is nevertheless unpublished, the District Court answered this question in the

affirmative, dismissing the case on summary judgment. A panel of this Court affirmed in a summary

order containing a single substantive paragraph. Ricci v. DeStifano, No. 06-4996-cv (2d Cir. Feb. 15,

2008).1 Three days prior to the filing of this opinion, the panel withdrew its summary order and filed a

per CIInam opinion adopting hI toto the reasoning of the District Court, thereby making the District

Court's opinion the law of the Circuit. See Ricci v. DeStefimo, _ F.3d _ (2d Cir. 2008).

The use ofper CIInam opinions of this sort, adopting in full the reasoning of a district court

without further elaboration, is normally reserved for cases that present straight-forward questions that

do not require explanation or elaboration by the Court of Appeals. The questions raised in this appeal

cannot be classified as such, as they are indisputably complex and far from well-settled. These

questions include: Does the Equal Protection Clause prohibit a municipal employer from discarding

examination results on the ground that "too many" applicants of one race received high scores and in

the hope that a future test would yield more high-scoring applicants of other races? Does such a

1 Reproduced as .-\ppendix A.

1

Page 18: Sotomayor Cases

practice constitute an unconstitutional racial quota or set-aside? Should the burden-shifting framework

applicable to claims of pretextual discrimination ever apply to a claim of explicit race-based

discrimination in violation of Title VII? If a municipal employer claims that a race-based action was

undertaken in order to comply with Title VII, what showing must the employer make to substantiate

that claim? Presented with an opportunity to address en bane questions of such "exceptional

importance," Fed. R. App. P. 35(a)(2), a majority of this Court voted to avoid doing so.

I respectfully dissent from that decision, without expressing a view on the merits of the

questions presented by this appeal, in the hope that the Supreme Court will resolve the issues of great

significance raised by this case.

BACKGROUND

In late 2003, 118 applicants took a written and oral examination administered by the New

Haven Fire Department ("NHFD") for promotion to the ranks of Captain and Lieutenant. Forty-one

applicants took the Captain examination, of whom twenty-five were white, eight black, and eight

Hispanic. Based on the examination results and New Haven's protocol for civil service promotions, it

appeared, at the time that the tests were scored, that "no blacks and at most two Hispanics would be

eligible for promotion" to Captain. Ricci v. DeStefano, No. 04cvll09, at 3 (D. Conn. Sept. 28,2006).2

\Vith respect to the Lieutenant examination, the racial composition of the seventy-seven applicants was

as follows: forty-three whites, nineteen blacks, and fifteen Hispanics. The examination results indicated

that no blacks or Hispanics would be promoted to the rank of Lieutenant. Between January and March

2004, the New Haven Civil Service Board ("CSB") held hearings to determine whether to certify the

examination results and confer promotions according to those results. Despite the substantial efforts

undertaken by the examination designer to ensure that it would be race-neutral, the City of New Haven

2 Reproduced as .,-\ppendix B.

2

Page 19: Sotomayor Cases

(the "City") frankly stated its fear that, if the results were certified, it would face an employment

discrimination lawsuit from non-white applicants who were not promoted. The CSB did not certify the

examination results, and no promotions were made.

Eighteen candidates-seventeen whites and one Hispanic-brought an action in the U.S.

District Court for the District of Connecticut. They alleged in their complaint that the City and several

municipal officials-acting in violation of Ticle VII of the Civil Rights Act of 1964,42 U.S.c. § 2000e et

seq., the Equal Protection Clause of the Fourteenth Amendment to the U.S. Constitution, and other

provisions of federal and state law-disregarded the results of two promotional examinations that

produced "too many" eligible white candidates and "too few" eligible non-white candidates. On cross­

motions for summary judgment, the District Court (Janet Bond Arterton, Judge) granted defendants'

motion for summary judgment, denied plaintiffs' motion, and directed the Clerk of Court to close the

case.

In a forty-eight page opinion, the District Court observed that (1) "[P]laintiffs' evidence-and

defendants' own arguments-show that the City's reasons for advocating non-certification [of the

examination results] were related to the racial distribution of the results" and (2) "[a] jury could infer

that the defendants were motivated by a concern that too many whites and not enough minorities

would be promoted were the [eligibility] lists to be certified." Ricci, No. 04cv1109, at 20-21. The

District Court recognized the exceptional circumstances presented by the case, noting that it "presents

the opposite scenario of the usual challenge to an employment or promotional examination, as

plaintiffs attack not the use of allegedly racially discriminatory exam results, but defendants' reason for

their reftsal to use the results." ld at 22. Applying the burden-shifting framework set forth in

McDonnell Douglas Corp. v. Gree1l, 411 U.S. 792 (1973), the District Court held that "[d]efendants'

motivation to avoid making promotions based on a test with a racially disparate impact, even in a

3

Page 20: Sotomayor Cases

political context, does not, as a matter of law, constitute discriminatory intent, and therefore such

evidence is insufficient for plaintiffs to prevail on their Title VII claim." Ricci, No. 04cvll 09, at 39-40

(footnote omitted).

The District Court further concluded that defendants had not violated plaintiffs' rights under

the Equal Protection Clause by, as plaintiffs alleged, "employing a race-based classification system for

promotion or, alternatively, by applying facially neutral promotion criteria in a racially discriminatory

manner." Id. at 40, 44. Although it is not disputed that the decision to discard the examination results

was based on racial considerations, the District Court determined as a matter of law that no racial

discrimination had occurred "because [all ofj the test results were discarded and nobody was

promoted," id. at 42, and because "[n]othing in the record in this case suggests that the City defendants

or CSB acted 'because of discriminatory animus toward plaintiffs or other non-minority applicants for

promotion," 11. at 43. The District Court also rejected plaintiffs' civil rights conspiracy and First

Amendment claims and declined supplemental jurisdiction over a state law tort claim.

On appeal, the parties submitted briefs of eighty-six pages each and a six-volume joint appendi.x

of over 1,800 pages; plaintiffs' reply brief was thirty-two pages long. Two amici briefs were ftled and

oral argument, on December 10,2007, lasted over an hour (an unusually long argument in the practice

of our Circuit). More than two months after oral argument, on February 15, 2008, the panel affirmed

the District Court's ruling in a summary order containing a single substantive paragraph. The operative

portion of the summary order read as follows:

We affirm, substantially for the reasons stated in the thorough, thoughtful, and well-reasonedopinion of the court below. In this case, the Civil Service Board found itself in the unfortunateposition of having no good alternatives. We are not unsympathetic to the plaintiffs' expressionof frustration. Mr. Ricci, for example, who is dyslexic, made intensive efforts that appear tohave resulted in his scoring highly on one of the exams, only to have it invalidated. But itsimply does not follow that he has a viable Title VII claim. To the contrary, because the Board,in refusing to validate the exams, was simply trying to fulfill its obligations under Title VII

4

Page 21: Sotomayor Cases

when confronted with test results that had a disproportionate racial impact, its actions wereprotected.

The judgment of the district court is AFFIRl\1ED.

See App. A. Four months later, and three days prior to the publication of this opinion, the panel

withdrew its summary order and published a per curiam opinion that contained the same operative text

as the summary order, with the addition of a citation to the District Court's opinion in the \Xlestlaw and

LexisNexis databases. This per cunam opinion adopted in toto the reasoning of the District Court,

without further elaboration or substantive comment, and thereby converted a lengthy, unpublished

district court opinion, grappling with significant constitutional and statutory claims of first impression,

into the law of this Circuit. It did so, moreover, in an opinion that lacks a clear statement of either the

claims raised by the plaintiffs or the issues on appeal. Indeed, the opinion contains no reference

whatsoever to the constitutional claims at the core of this case, and a casual reader of the opinion could

be excused for wondering whether a learning disability played at least as much a role in this case as the

alleged racial discrimination. This perfunctory disposition rests uneasily \vith the weighty issues

presented by this appeal.3

Prior to the entry of the per curiam opinion and in light of the "question[s] of exceptional

importance," Fed. R. App. P. 35(a)(2), raised in this appeal, the Court considered a motion for en bam:

reVIew. A majority of this Court declined to take up the appeal.

DISCUSSION

3 Judge Parker's observation that "[t]he adherence of a Court of Appeals to the decision and reasoning of adistrict court is anything but novel" cannot be gainsaid. In appropriate cases, such a disposition is entirelyunobjectionable. \X'here significant questions of unsettled law are raised on appeal, however, a failure to address thosequestions--or even recognize their existence-should not be the approved modus operandi of the U.S. Court ofAppeals.

5

Page 22: Sotomayor Cases

The core issue presented by this case-the scope of a municipal employer's authority to

disregard examination results based solelY on the race of the successful applicants-is not addressed by

any precedent of the Supreme Court or our Circuit. Plaintiffs alleged that the City's actions violated,

inter alia, their rights under the Equal Protection Clause and Title VII. The District Court disagreed,

but did so without the benefit of pertinent guidance from a higher court. The questions raised by the

instant appeal clearly merit further review.4

A. The Equal Protection Clause

Plaintiffs claim that the City's decision to discard the examination results was race-based

discrimination in violation of the Equal Protection Clause because it was undertaken solely to reduce

the number of high-scoring white applicants and increase the number of eligible non-white candidates.

Defendants contend that their decision, though race-based, was necessary because compliance with

federal anti-discrimination laws required them to reduce the number of eligible white candidates. See

Ricci, No. 04cv1109, at 20-21; Appellee Br. at 15-20, 30-31. The Supreme Court has addressed a

government entity's claim that race-based decisions were necessary to redress a racial imbalance in the

closely analogous context of government contracts. In Ciry qfRichmond v. J. A. Croson Co., the Supreme

Court held that: "[w]hile there is no doubt that the sorry history of both private and public

discrimination in this country has contributed to a lack of opportunities for black entrepreneurs, this

observation, standing alone, cannot justify a rigid racial quota in the awarding of public contracts ...."

488 U.S. 469, 499 (1989). The Court further observed that:

[W]hen a legislative body chooses to employ a suspect classification, it cannot rest upon ageneralized assertion as to the classification's relevance to its goals. A governmental actorcannot render race a legitimate proxy for a particular condition merely by declaring that thecondition exists. The history of racial classifications in this country suggests that blind judicial

4 Indeed, in his opinion concurring in the denial of en bane review, Judge Katzmann recognizes as much,observing that this appeal "presents difficult issues."

6

Page 23: Sotomayor Cases

deference to legislative or executive pronouncements of necessity has no place in equalprotection analysis.

Id. at 500-01 (internal citations omitted). More recently, the Supreme Court has identified "three

general propositions with respect to governmental racial classifications." Adarand Constmdors, Inc. v.

Pena, 515 U.S. 200, 223 (1995). They are:

First, skepticism: Any preference based on racial or ethnic criteria must necessarily receive amost searching examination. Second, consistency: The standard of review under the EqualProtection Clause is not dependent on the race of those burdened or benefited by a particularclassification, i.e., all racial classifications reviewable under the Equal Protection Clause must bestrictly scrutinized. And third, congruence: Equal protection analysis in the Fifth Amendmentarea is the same as that under the Fourteenth Amendment. Taken together, these threepropositions lead to the conclusion that at!} person, ofwhatever race, has the right to demand that aI!}governmental actor su,?ject to the Constitutionjustifj af!) racial classification su,?jecting that person to unequaltreatment under the strictestjudicial scmtif!).

Id. at 223-24 (quotation marks, internal citations, and brackets omitted) (emphasis added).

Whether the District Court's judgment comports with these propositions is a question of

immense importance that is not addressed in the panel's per curiam opinion. The District Court's ruling

rested in part on the premise that "where a test is administered and scored in the same manner for all

applicants, plaintiffs cannot make out a claim that the exam was a facially neutral test used in a

discriminatory manner." Ricci, No. 04cv1109, at 42. Neutral administration and scoring--even against

the backdrop of race-conscious design of an employment examination, see Hqyden v. Coun!} ofNassau, 180

F.3d 42, 50 (2d Cir. 1999)-is one thing. But neutral administration and scoring that is followed by

race-based treatment of examination results is surely something else entirely. Where, as here,

examination results are disregarded on the ground that too many candidates of one race qualified for

promotion on the basis of those results, the fact of neutral administration and scoring may not

necessarily immunize defendants from the claims of civil rights violations brought by plaintiffs. If it

did, municipal employers could reject the results of an employment examination whenever those results

7

Page 24: Sotomayor Cases

failed to yield a desired racial outcome-i.e., failed to satisfy a racial quota. Croson and Adarand est.'lblish

that racial quotas are impermissible under the Constitution absent specific findings of past

discrimination that are not in the record here. Whether Croson and Adaraltd preclude the actions

challenged in this case, or whether Hqydm can fairly be read to compel judgment in defendants' favor

as a matter of law, are questions that admit no easy answer. As such, they require the careful analysis of

a full opinion of an appellate court, not abbreviated disposition.

The District Court held that the test was administered in the same manner for all applicants

because the City discarded the scores of all exam-takers. Insofar as the decision to not certify the

results was based on the race of the high-scoring applicants, however, it is arguable that the deck was

stacked against applicants of that race: If too many white applicants obtained high scores, the City

stood ready to nullify the results in the hope that non-white applicants would score relatively higher on

a subsequent examination.5 Whether such action amounts to an impermissible racial quota was not

addressed in the District Court's opinion or in the decisions issued by the panel, which do not even

note that this action arises under the Equal Protection Clause of the Fourteenth Amendment. See App.

A (summary order of Feb. 15, 2008); per curiam opinion filed on June 9, 2008.

The District Court also held as a matter oflaw that none of the City's reasons for disregarding

the examination results amounted to intentional discrimination because the City had

acted based on the following concerns: that the test had a statistically adverse impact onAfrican-American and Hispanic examinees; that promoting off of this list would underminetheir goal of diversity in the Fire Department and would fail to develop managerial role modelsfor aspiring firefighters; that it would sulyect the City to public criticism; and that it \vould likelysubject the City to Title VII lawsuitsfrom minority applicants that, for political reasons, the City did 110twant to difend.

5 In his opinion concurring in the denial of en bane review, Judge Parker concludes that the City "did nor ...confer any actual benefit on applicants on the basis of race." It is, at the very least, an open question whether discardingexamination results on the basis of race so that members of certain races could have a "second chance" to competeconstitutes the conferral or denial of a benefit on the basis of race.

8

Page 25: Sotomayor Cases

Ricci, No. 04cv1109, at 43 (emphasis added). Leaving aside the propriety of the District Court's

evaluation, on summary judgment, of the City's motives-a quintessential question of fact, see, e.g., Hunt

v. Cromartie, 526 U.S. 541, 552-53 (1999)-it is at least arguable that the District Court failed to subject

the City's justifications to the "most searching examination" prescribed by the Adarand Court. See 515

U.S. at 223. The record suggests that the District Court took the City's justifications at face value, as it

appears Judge Parker has done in his opinion concurring in the denial of en banc review. An appellate

court ought to consider whether this level of scrutiny is consistent with Justice O'Connor's

observation, in Croson, that "[a]bsent searching judicial inquiry into the justification for such race-based

measures, there is simply no way of determining what classifications are 'benign' or 'remedial' and what

classifications are in fact motivated by illegitimate notions of racial inferiority or simple racial politics."

488 U.S. at 493 (plurality opinion). Justice O'Connor's cautionary note on "racial politics" is

particularly relevant in light of the District Court's observation that fear of "public criticism" and other

"political reasons" factored into the City's decision. Whether the District Court subjected the City's

claims to sufficient scrutiny-and whether the City's claims could have withstood such scrutiny-are

vital "question[s] of exceptional importance," Fed. R. App. P. 35(a)(2), that warrant further review,

both for the proper resolution of this case and for the guidance of other courts and municipalities in

future cases.

B. Title VII

Plaintiffs urge that the City's race-based action also violated Title VII's prohibition of

employment discrimination. See 42 U.S.c. § 2000e-2. The District Court dismissed plaintiffs' Title VII

claim by applying the three-step burden-shifting framework for adjudicating claims of pretextual

discrimination established by McDonnell Douglas. The dismissal of the Title VII claim on this basis raises

two significant questions: (1) whether the McDonnell Douglas test for prete-"'dual discrimination should be

9

Page 26: Sotomayor Cases

applied to claims of discrimination that is overt, and (2) whether a race-based decision allegedly made to

avoid perceived liability for racial discrimination is exempt from scrutiny under Title VII and, if not,

what quantum of proof is required to substantiate such a defense.

Courts generally apply McDonnell Douglas in cases where plaintiffs "presentD no direct evidence

of discriminatory treatment." Woodman v. WWOR-TV, Inc., 411 F.3d 69, 76 (2d Cir. 2005); see also

Graves tJ. Finch Pruyn & Co. Inc, 457 F.3d 181, 187 (2d Cir. 2006). "If a plaintiff can convince the trier of

fact that an impermissible criterion in fact entered into the employment decision, [however,] a

somewhat different analysis takes place." 7)ler 11. Bethlehem Steel Corp., 958 F.2d 1176, 1181 (2d Cir.

1992). In that kind of "mixed-motive" case, the burden-shifting analysis set forth in Price Waterhouse v.

Hopkins, 490 u.s. 228 (1989), governs the claim. Under this framework,

the plaintiff ... must focus his proof directly at the question of discrimination and prove thatan illegitimate factor had a motivating or substantial role in the employment decision. If theplaintiff convinces the factfinder that the illegitimate factor played such a role, the employeehas proved that the decision was made at least in part because of the illegitimate factor. At thispoint the employee is entitled to succeed subject only to the employer's opportunity to proveits affIrmative defense; that is, that it would have reached the same decision as to theemployee's employment even in the absence of the impermissible factor.

7)ler, 958 F.2d at 1181 (internal citations, quotation marks, and brackets omitted); see also Sista v. CDC

IxiJ N. Am., I,IC., 445 F.3d 161, 173-74 (2d Cir. 2006); Raskin v. u::jatt Co., 125 F.3d 55,60-61 (2d Cir.

1997) ("Evidence potentially warranting a Price Waterhouse burden shift includes, inter alia, policy

documents and evidence of statements or actions by decisionmakers that may be viewed as directly

reflecting the alleged discriminatory attitude." (internal quotation marks omitted)).

The Ricci plaintiffs offered evidence that an impermissible factor-their race-motivated

defendants to discard the results of the employment examination. As the District Court itself candidly

observed: "[P]laintiffs' evidence-and defendants' own arguments-show that the City's reasons for

advocating non-certification [of the examination results] were related to the racial distribution of the

10

Page 27: Sotomayor Cases

results" and "[a] jury could infer that the defendants were motivated by a concern that too many whites

and not enough minorities would be promoted were the [eligibility] lists to be certified." Rim, No.

04cvl109, at 20-21. The District Court's application of the McDonnell Douglas test for pretextual

discrimination, its conclusion that plaintiffs cannot pass that test as a matter of law, and its failure to

consider the possibility that defendants themselves might bear a burden of proof under the analysis set

forth in Price Waterhouse, all raise serious concerns left unaddressed by the panel in its per a/riam opinion

and by the full Court, which declined en banc review of the appeaL

Assuming arguendo that a claim of overt racial discrimination is ever appropriately evaluated

under the McDonnell Douglas framework for pretextual discrimination, the application of that framework

to this case required a "reversal" of the usual roles assigned to plaintiffs and defendants in such cases.

As the District Court observed:

[T]his case presents the opposite scenario of the usual challenge to an employment orpromotional examination. . .. Ordinarily, as contemplated by the statute, the "complainingparty" bears the burden of proving a disparate impact, and the "respondent" bears the burdenof "demonstrat[ing] that the challenged practice is job related for the position in question andconsistent with business necessity," or, alternatively, the "complaining party" may prevail byshowing that an alternative employment practice with less disparate impact existed and that therespondent failed to utilize it. Here, the roles of the parties are in essence reversed, with thedefendants, normally reflecting a "respondent" role in the Title VII disparate impact analysis,contending that use of the promotional exams, if they had been certified, would have had anadverse impact, and the plaintiffs, normally the "complaining party," arguing that the testresults were sufficiently job-related to be defensible under the law.

Ricci, No. 04cv11 09, at 22 (alteration in original) (internal citations omitted). Unlike the Court of

Appeals, the District Court answered the exceptional, and difficult, questions presented, concluding

that the City's expressed desire to comply with "the letter and the spirit of Title VII," id. at 22,

constituted a non-pretextual reason for its action, id. at 39-40, and therefore no employment

discrimination occurred. Under the District Court's rationale, it appears that any race-based

employment decision undertaken to avoid a threatened or perceived Title VII lawsuit is immune from

11

Page 28: Sotomayor Cases

scrutiny under Title VII." This appears to be so, moreover, regardless of whether the employer has

made any efforts to verify that a valid basis exists for the putative Title VII suit. Applying this

rationale, the District Court concluded that the City, which had not conducted any study to determine

whether latent racial bias had tainted the results of the promotion examination, could discard the

results of the examination, id. at 25-26, in the hope that a future test would yield a preferable racial

distribution, id. at 36. Regardless of how one may decide the matter, there can be little doubt that a

decision of this Court thus sanctioning race-based employment decisions in the name of compliance

with Title VII raises novel questions that are indisputably of "exceptional importance."

CONCLUSION

It is arguable that when an appeal raising novel questions of constitutional and statutory law is

resolved by an opinion that tersely adopts the reasoning of a lower court-and does so without further

legal analysis or even a full statement of the questions raised on appeal-those questions are insulated

from further judicial review. It is arguable also that the decision of this Court to deny C11 banc review of

this appeal supports that view. What is not arguable, however, is the fact that this Court has failed to

grapple with the questions of exceptional importance raised in this appeal. If the Ricci plaintiffs are to

obtain such an opinion from a reviewing court, they must now look to the Supreme Court. Their

claims are worthy of that review.

6 Despite Judge Parker's assertion to the contrary, I do not charge the District Court with applying a "rubberstamp" to the Ciry's race-based decisions. I simply question whether the Court of A.ppeals has set forth a standard fordetermining when such action is acceptable and when it violates the constitutional and statutory rights of citizens. If anyfault is to be levied in this regard, it falls on our Court for failing to provide guidance, and not on the District Courtwhich endeavored to confront this question of exceptional importance.

12

Page 29: Sotomayor Cases

Appendix A

Page 30: Sotomayor Cases

SUMMARY ORDER

Defendants-Appellees.

06-4996-cvRicci v. DeStefano

UNITED STATES COURT OF APPEALSFOR THE SECOND CIRCUIT

(06-4996-cv)

Plaintiffs-Appellants,

-v-

ROSEMARY S. POOLER,ROBERT D. SACK,SONIA SOTOMAYOR,

Circuit Judges,

RULINGS BY SUMMARY ORDER DO NOT HAVE PRECEDENTIAL EFFECT. CITATIONTO SUMMARY ORDERS FILED AFTER JANUARY 1,2007, IS PERMITTED AND ISGOVERNED BY THIS COURT'S LOCAL RULE 32.1 AND FEDERAL RULE OF APPELLATEPROCEDURE 32.1. IN A BRIEF OR OTHER PAPER IN WHICH A LITIGANT CITES ASUMMARY ORDER, IN EACH PARAGRAPH IN WHICH A CITATION APPEARS, AT LEASTONE CITATION MUST EITHER BE TO THE FEDERAL APPENDIX OR BE ACCOMPANIEDBY THE NOTATION: "(SUMMARY ORDER)." A PARTY CITING A SUMMARY ORDERMUST SERVE A COpy OF THAT SUMMARY ORDER TOGETHER WITH THE PAPER INWHICH THE SUMMARY ORDER IS CITED ON ANY PARTY NOT REPRESENTED BYCOUNSEL UNLESS THE SUMMARY ORDER IS AVAILABLE IN AN ELECTRONICDATABASE WHICH IS PUBLICLY ACCESSIBLE WITHOUT PA YMENT OF FEE (SUCH ASTHE DATABASE AVAILABLE AT HTTP://WWW.CA2.USCOURTS.GOVI). IF NO COpy ISSERVED BY REASON OF THE AVAILABILITY OF THE ORDER ON SUCH A DATABASE,THE CITATION MUST INCLUDE REFERENCE TO THAT DATABASE AND THE DOCKETNUMBER OF THE CASE IN WHICH THE ORDER WAS ENTERED.

Present:

At a stated Term of the United States Court of Appeals for the Second Circuit, held at theDaniel Patrick Moynihan United States Courthouse, 500 Pearl Street, in the City of New York,on the 15th day of February, two thousand and eight.

FRANK RICCI, MICHAEL BLATCHLEY, GREG BOIVIN, GARY CARBONE, MICHAELCHRISTOFORO, RYAN DIVITO, STEVEN DURAND, WILLIAM GAMBARDELLA,BRIAN JOOSS, JAMES KOTTAGE, MATTHEW MARCARELLI, THOMAS J. MICHAELS,SEAN PATTON, CHRISTOPHER PARKER, EDWARD RIORDAN, KEVIN ROXBEE,TIMOTHY SCANLON, BENJAMIN VARGAS, JOHN VENDETTO AND MARKVENDETTO,

JOHN DESTEFANO, KAREN DUBOIS-WALTON, THOMAS UDE JR., TINA BURGETT,BOISE KIMBER, MALCOM WEBER, ZELMA TIRADO AND CITY OF NEW HAVEN,

123456789

10111213141516171819202122232425262728293031323334353637383940414243444546474849

Page 31: Sotomayor Cases

The judgment of the district court is AFFIRMED.

By:. _

FOR THE COURT:Catherine O'Hagan Wolfe, Clerk

Plaintiffs appeal from a judgment of the United States District Court for the District of .Connecticut (Arterton, J.) granting the defendants' motion for summary judgment on all counts.

Karen Lee Torre, New Haven, CT.Appearing for Plaintiffs-Appellants:

Appearing for Defendants-Appellants:

UPON CONSIDERATION WHEREOF, IT IS HEREBY ORDERED, ADJUDGED,AND DECREED that the judgment of the distri~t court is AFFIRMED.

Richard A. Roberts (Nicole C. Chomiak,Stacey L. Pitcher, and Todd 1. Richardson

__________________~on:.:..:::.th.:.::e:...:b::.:.n.:.::·e::Jf),Cheshire, CT.

We affirm, substantially for the reasons stated in the thorough, thoughtful, and well-reasonedopinion of the court below. In this case, the Civil Service Board found itself in the unfortunateposition of having no good alternatives. We are not unsympathetic to the plaintiffs' expressionof frustration. Mr. Ricci, for example, who is dyslexic, made intensive efforts that appear tohave resulted in his scoring highly on one of the exams, only to have it invalidated. But itsimply does not follow that he has a viable Title VII claim. To the contrary, because the Board,in refusing to validate the exams, was simply trying to fulfill its obligations under Title VII whenconfronted with test results that had a disproportionate racial impact, its actions were protected.

I23456789

1011121314151617181920212223242526272829

2

Page 32: Sotomayor Cases

Appendix B

Page 33: Sotomayor Cases

UNITED STATES DISTRICT COURTDISTRICT OF CONNECTICUT

FRANK RICCI, et al.,Plaintiffs,

v.

JOHN DESTEFANO, et al.,Defendants.

Civil No. 3:04cvll09 (JBA)

RULING ON CROSS-MOTIONS FOR SUMMARY JUDGMENT[DOCS. ii 52, 60]

In March 2004 the New Haven Civil Service Board (UCSB")

refused to certify the results of two promotional exams for the

positions of Lieutenant and Captain in the New Haven Fire

Department. This lawsuit arises from the circumstances leading

to that decision and its consequences.

Plaintiffs are seventeen white candidates and one Hispanic

candidate who took the promotional exams, on which they fared

very well, but received no promotion because without the CSB's

certification of the test results, the promotional process could

not proceed. Defendants are the City of New Haven, Mayor John

DeStefano, Chief Administrative Officer Karen Dubois-Walton,

Corporation Counsel Thomas Ude, Director of Personnel Tina

Burgett, and the two members of the CSB, Malcolm Weber and Zelma

Tirado, who voted against certification. Plaintiffs assert that

defendants' actions in voting or arguing against certification of

the examination results violated their rights under Title VII of

1

Page 34: Sotomayor Cases

the Civil Rights Act of 1964, 42 U.S.C. § 2000e et seq., the

Equal Protection Clause, the First Amendment, and 42 U.S.C. §

1985; plaintiffs also allege a common law claim of intentional

infliction of emotional distress. The parties have cross-moved

for summary judgment on the Title VII and Equal Protection

claims, and defendants additionally move for summary judgment on

plaintiffs' other claims.

For the reasons that follow, defendants' motion for summary

judgment [Doc. # 52] will be granted as to plaintiffs' federal

claims; plaintiffs' cross-motion for summary judgment [Doc. # 60]

will be denied; and the Court will decline jurisdiction over

plaintiffs' state law claim. 1

I. Factual Background

While the parties strenuously dispute the relevance and

legal import of, and inferences to be drawn from, many aspects of

this case, the underlying facts are largely undisputed. In

November and December 2003, the New Haven Fire Department

administered written and oral examinations for promotion to

Lieutenant and Captain. The City's Department of Human Resources

issued a Request for Proposal for these examinations; as a result

of which I/O Solutions (\\IOS"), a seven-year-old Illinois company

that specializes in entry-level and promotional examinations for

lDefendants also moved to strike portions of plaintiffs'Local Rule 56(a)2 Statement, which motion was denied. See RulingDenying Motion to Strike [Doc. # 130].

2

Page 35: Sotomayor Cases

public safety (police and fire) departments, designed the

examinations. Pl. Ex. IV(C) at 8. Under the contract between

the City and the New Haven firefighters' union, the written exam

result counted for 60% of an applicant's score and the oral exam

for 40%. Those with a total score above 70% on the exam would

pass.

Forty-one applicants took the Captain exam, of whom 25 were

white, 8 black, and 8 Hispanic. Twenty-two of those applicants

passed, of whom 16 were white, 3 black, and 3 Hispanic. Pl. Ex.

Vol. I, at 43. Given that there were 7 Captain vacancies in the

department when the tests were administered, and that the "Rule

of Three" in the City Charter mandates that a civil service

position be filled from among the three individuals with the

highest scores on the exam, it appeared at that time that no

blacks and at most two Hispanics would be eligible for promotion,

as the top 9 scorers included 7 whites and 2 Hispanics. 2

Seventy-seven applicants took the Lieutenant exam, of whom

43 were white, 19 black, and 15 Hispanic. Thirty-four passed, of

whom 25 were white, 6,black and 3 Hispanic. Id. There were 8

vacancies, but because of the top 10 scorers were white, it

appeared that no blacks or Hispanics would be promoted. 3

2Hispanics ranked 7, 8 and 13; blacks ranked 16, 19 and 22.Pl. Ex. Vol. I, at 43.

3Hispanics ranked 27, 28 and 31; blacks ranked 14, 15, 16,20, 22, and 24. PI. Ex. Vol. I, 43.

3

Page 36: Sotomayor Cases

Certified promotional lists remain valid for two years.

The CSB held five hearings between January and March 2004 on

the issue of whether to certify the test results. The issue

appears to have been raised by New Haven's Corporation Counsel,

Thomas Ude. At the initial hearing on January 22, 2004, Mr. Ude

characterized the exam results as ~a very significant disparate

impact . . . that caused us to think this was something we should

bring to you, the Civil Service Board, to evaluate and to be part

of and to ultimately make a decision about the process." Pl. Ex.

Vol. IV(A) at 32. While it is disputed whether Mr. Ude already

had made up his mind to advise the CSB against certifying the

results, his comments "emphasize [dl . that the case law does

not require that the City find that the test is indefensible in

order to take action that it believes is appropriate to remedy

.. disparate impact from examination." Id. at 34-35. He

advised that "federal law does not require that you [the CSB]

make a finding that this test . . . was not job-related, which is

another way of saying it wasn't fair. A test can be job-related

and have a disparate impact on an ethnic group and still be

rejected because there are less discriminatory alternatives for

the selection process." Id. at 36.

During the hearings, the tests results were not released by

name, and therefore none of the firefighters knew where they had

placed. The only information provided to the CSB and the public,

4

Page 37: Sotomayor Cases

including plaintiffs, was the scores by race and gender.

Nonetheless, several firefighters, although they did not know

where they had placed, spoke in favor of certifying the results.

Plaintiff Frank Ricci stated that the questions on the test were

drawn from "nationally recognized" books and New Haven's own

Rules and Regulations and Standard Operating Procedures. Pl. Ex.

Vol. 1V(A) at 88. He stated that he "studied 8 to 13 hours a day

to prepare for this test and incurred over $1,000 in funds [sic]

to study for this test," including purchasing the books and

paying an acquaintance to read them on tape because he is

dyslexic and learns better by listening. Other firefighters, who

believed the tests were fair, ·also spoke in support of the

certifying the results. See,~, Testimony of Michael

Blatchley, id. at 75 ("[N]one of those questions were not in that

study material. Everyone of those questions came from the

material.") .

During the first hearing, the CSB also took statements from

several New Haven firefighters who complained that some of the

questions were not relevant to knowledge or skills necessary for

the positions (see, ~, Statement of James Watson, at 85

("I think this test was unfair. We don't use ~ lot of things

that were on that test" such as whether to park a firetruck

facing "uptown" or "downtown"», or that the study materials were

difficult to obtain (see Testimony of Gary Kinney, id. at 77

5

Page 38: Sotomayor Cases

(~The only books that most of us had in front of us in the fire

houses were Essentials of Fire Fighting.... [T]hese books [on

the syllabus] were never in the fire houses."».

At the second hearing on February 5, Patrick Egan, president

of the firefighters' union, urged the CSB to conduct a validation

study to determine the job-relatedness of the test, referring

generally, although not by name, to the EEOC's ~Uniform

Guidelines of Employee Selection Procedures." Pl. Ex. Vol. IV(B)

at 11-12. Plaintiffs' counsel in the present case also ~poke and

urged certification.

On the other side, Donald Day, a representative of the

Northeast Region of the International Association of Black

Professional Firefighters, argued against certification, stating

that previous promotional examinations in 1996 and 1999 had black

and Latino firefighters ranked sufficiently high to have a

realistic opportunity for promotion, whereas ~there was something

inherently wrong with this test" because minorities did not score

as highly. Id. at 33-34. He suggested that the CSB speak with

the director of the civil service in Bridgeport ~to find out what

Bridgeport is doing different [sic] than New Haven," as they have

more diversity in their firefighter ranks. rd. at 35. In

particular, he stated that Bridgeport had "changed the relative

weights" among the portions of the exam, such that the written

test counts for 30% of the total score, the oral test for 65%,

6

Page 39: Sotomayor Cases

and seniority 5%. Id. at 36-37. Ronald Mackey, the Internal

Affairs Officer for the Northeast Region of the International

Association of Black Professional Firefighters, supported Patrick

Egan's suggestion of obtaining a validation study, and also

suggested that New Haven could "adjust the test" as Bridgeport

had done, in order to "meet the criteria of having a certain

amount of minorities get elevated to the rank of Lieutenant and

Captain." Id. at 43-45.

On February 11, 2004, the CSB heard from Chad Legel, Vice

President of lOS, who was the "project manager" in charge of

developing the exams at issue. He stated that lOS had prepared

"both an entry-level exam and a physical ability test for the

firefighter position" in New Haven, but had not previously

prepared a New Haven promotional exam. Id. at 10. However, in

recent years his company had worked with similarly-sized public

safety departments with demographics similar to New Haven,

including Lansing, Michigan, Orange County, Florida, and the

North Miami Police Department, among others. Id. at 9.

Legel described the way in which the test was developed.

First, the company interviewed a random sample of current New

Haven Fire Department Lieutenants, Captains and Battalion Chiefs

to determine basic information concerning the structure of the

department, the tasks required of individuals at each rank, and

the materials the department generally utilizes for training.

7

Page 40: Sotomayor Cases

Based on the interviews, lOS developed a written job analysis

questionnaire ("JAQ") that asked all incumbents in the positions

of Lieutenant and Captain "to provide information about how

important they feel a specific task, knowledge area, skill or

ability is. "Id. at 17. The JAQ asked how important each

task was to successful performance on the job and how frequently

it was necessary to perform it. Importance and frequency were

merged into a metric called "criticality or essentiality." Id.

at 19. Tasks above a certain criticality threshold were

designated for testing on the written and oral portions of the

exam. In response to the question of whether he has generally

found a difference between information tested in various

departments "based on the racial content of the city and the

force," Legel stated, "definitely no." rd. at 21. The one

difference among the New Haven firefighters of similar rank that

Legel noted was different levels of training in certain

specialized fields such as hazardous materials; such variation

"throws up a red flag" indicating that lOS should not ask "high-

level questions about hazardous materials. "Id. at 22.

Legal further stated that all the questions were firmly

rooted in the study materials on the syllabus, which was

distributed with the promotion applications. See Def. Ex. 16

("Written Examination Reference List"). Once the test was

completed, an "independent reviewer;" a Battalion Chief from the

8

Page 41: Sotomayor Cases

Cobb County, Georgia, Fire Department, "reviewed the written exam

for content and fidelity to the source material." Pl. Ex. Vol.

IV(B) at 24-25. Another independent reviewer, a retired Fire

Chief from outside Connecticut, reviewed the oral exam questions.

Id. at 26. lOS refrained from utilizing reviewers from

Connecticut because the RFP had specified that examiners must

come from outside Connecticut, due to concerns that utilizing

internal personnel could potentially facilitate cheating on the

test.

Likewise, lOS selected the panelists for the oral

examination panels from departments outside Connecticut, making

an effort "to gain maximum diversity." Id. at 32. All but one

panel had one African-American, one Hispanic and one white

assessor, and a standby panel had two African-Americans and one

white. Id. The assessors were trained on how to grade the oral

exam scenarios consistently, utilizing checklists of desired

criteria. Each panelist also held at least an equal rank (if not

superior) to the position being tested, in order to be able to

identify an answer that was good but not quite the best answer

outlined in the checklist. at 33-34, 37.

Legel concluded by "implor[ing] anyone that had ...

concerns [about disparate impact] to review the content of the

exam. In my professional opinion, it's facially neutral."

Id. at 49.

9

Page 42: Sotomayor Cases

Noelia Marcano, Chief Examiner for the City of-New Haven and

Secretary to the CSB, explained the process by which the RFP was

developed and lOS chosen. She further explained that the job

applications for the Lieutenant and Captain positions contained a

job description, employment application, and "the actual study

list in final form," and that when questions arose concerning

conflicting information in some of the study books, lOS sent a

letter to all applicants that they would not be asked questions

on material where the sources differed. Id. at 78.

At the next hearing on March 11, 2004, the CSB heard from

Christopher Hornick, Ph.D., an industrial/organizational

psychologist from Texas who runs a consulting business in

competition with IOS. 4 See Pl. Ex. Vol. IV(D} at 7, 12. Dr.

Hornick stated that he had "not had time to study the test at

length or in detail." at 13. However, he reviewed

stati~tics provided by the City and concluded that "we're seeing

relatively high adverse impact" from the lOS tests. Id. at 11.

He opined that his company finds "significantly and dramatically

less adverse impact in most of the test procedures that we

4Plaintiffs argue that Dr. Hornick's non-sworn, hearsaystatement at the CSB hearing is inadmissible as non-disclosedexpert evidence. Plaintiffs' argument is rejected becausedefendants proffer Dr. Hornick's not for the truth of hisconclusion that the tests had a racially disparate impact, but toshow that defendants had a good faith belief, based in part onDr. Hornick's testimony, that such a disparate impact existed andjustified the decision not to certify the exams.

10

Page 43: Sotomayor Cases

design." Id. at 12. He stated:

Normally, whites outperform ethnic minorities on themajority of standardized testing procedures. That is,in , the case with the data that we've seen in NewHaven.

I'm a little surprised at how much adverse impact thereis in these tests. And I hope at some point here we'llbe talking in detail about that. But my conclusion isthat we did have significant adverse impact. Some ofit is fairly typical of what we've seen in other areasof the countries (sic) and other tests that people havedeveloped. But in other ways it is somewhat worse thanwhat we're typically seeing in the profession practicedby others.

Id. at 11-12. Dr. Hornick acknowledged that he had not looked at

specific statistics from previous promotional examinations in New

Haven to compare their results with the 2003 exam results. Id.

at 14.

When asked about the reasons behind any possible disparate

impact, Dr. Hornick answered, "I'm not sure that r can explain

it," but suggested that perhaps the 60%/40% breakdown mandated by

the collective bargaining agreement could be responsible, and

further suggested that there were "perhaps different types of

testing procedures that are much more valid in terms of

identifying the best potential supervisors in your fire

department." rd. at 15. He stated that based on his interviews

with firefighters, "we know that" a written test is "not as valid

as other procedures that exist." rd. at 16. He also suggested

that "[b]y not having anyone from within the department review

the items [on the test] you inevitably get things in there" that

11

Page 44: Sotomayor Cases

are not relevant to the specific department. rd. at 17-18.

Finally, Dr. Hornick identified as an alternative to traditional

written and oral testing processes ~an assessment center process,

which is essentially an opportunity for candidates to demonstrate

their knowledge of the . . . standard operating procedures, to

demonstrate how they would address a particular problem as

opposed to just verbally saying it or identifying the correct

option on a written test. For example, there's concepts of

situation judgment tests that can be developed and designed,

customized within organizations that demonstrate dramatically

less adverse impacts. . . " Id. at 22-23.

At the same hearing, Vincent M. Lewis, a Fire Program

Specialist for the u.S. Department of Homeland Security, and a

retired career firefighter from Michigan, testified that he

believed the test was appropriate. He stated that he had looked

~extensively at the Lieutenant's exam and a little less at the

Captain's exam," and believed that the candidates ~should know

that material." Id. at 34-35. His one comment was that ~a

number of questions in the Lieutenant's exam dealt with issues

that an apparatus driver needed to know," and a candidate who had

not had such training would be disadvantaged on those questions.

rd. at 34, 41. However, he generally ~felt the questions were

relevant for both exams," and believed that the New Haven

applicants were advantaged over examinees in other locations

12

Page 45: Sotomayor Cases

because they were instructed exactly which chapters from the

study materials would .be on the tests. Id. at 36. He stated

that he would not have changed anything about the way in which

the tests were developed, and opined that any disparate impact

could be due to a general pattern that ~usually whites outperform

some of the minorities on testing,H or that ~more whites ...

take the exam. H Id. at 37-38.

The last expert witness was Dr. Janet Helms, a professor of

counseling psychology and the Director of the Institute for the

Study and Promotion of Race and Culture at Boston College. Her

area of expertise "is not with firefighters per se but is more in

the general area of how race and culture influence test

performance more generally.H Id. at 43. She did not examine the

specific tests at issue. Id. at 55. However, she offered

several potential explanations for racially disparate impact on

the tests. First, "[w]e know for a fact that regardless of what

kind of written test we give in this country that-we can just

about predict how many people will pass who are members of under­

represented groups. And your data are not that inconsistent

with what predictions would say were the case. H Id. at 44

(emphasis supplied). Second, Dr. Helms suggested that because

67% of the respondents in the JAQ survey were white, the

questions may have been skewed toward their job knowledge~ as

"most of the literature on firefighters show that the different

13

Page 46: Sotomayor Cases

[racial. and gender] groups perform the job differently." Id. at

46. Relying on information she had read in newspaper accounts of

the situation in New Haven, she stated that the difference in

performance may have been due to differences in opportunities for

training and "informal mentoring" available to mirtorities. Id.

at 48. With respect to the oral exam, Dr. Helms suggested that

people who are bilingual or "speak acc~nted speech" may elicit

more negative reactions from evaluators. Id. at 49-50. As

general concerns, Dr. Helms mentioned that test takers may score

lower if they are expected not to perform well, or if tests focus

on "traditional ways of doing the job and the test-taker, in

fact, uses innovative approaches." rd. at 51. Additionally, she

suggested that "removing" "socioeconomic status" from test scores

"reduces the disparate impact to some extent." Id. at 57.

At the final hearing on March 18, 2004, defendant Ude, the

Corporation Counsel, strongly advocated against certifying the

exam results. He concluded: "You have a choice. It is my

opinion that promotions under our rules as a result of these

tests would not be consistent with federal law, would not be

consistent with the purposes of our.Civil Service Rules or our

Charter, nor is it in the best interests of the firefighters and

Lieutenants who took the exams." Pl. Ex. Vol. IV(E) at 15-16.

As a primary reason not to certify the results, Ude argued that

the "results of previous exams in this department and in other

14

Page 47: Sotomayor Cases

departments have not had this kind of a result, which is one of

the reasons why these results were so startling when they came

down. These results were different." Id. at 19. He argued that

Dr. Hornick's statements to the CSB, standing alone, were

"sufficient" reason not to certify, and advised the board "that

it is the employer's burden to justify the use of the

examination" if a Title VII suit were brought. Id. at 21.

Defendant Walton spoke "on behalf of the Mayor" and also

advocated discarding the test results, primarily because the

eligibility list, when combined with the Rule of Three and the

number of vacancies then available, would "create a situation in

which African-Americans are excluded from promotional opportunity

on both the Captain and Lieutenant positions and Latinos are

excluded from promotional opportunity on the Lieutenant

examination." Id. at 30. She questioned whether there were

"other ways of making the selection," that would be l~s~

"discriminatory." Id. at 31-32.

The board split two to twoS on the question of certifying

each exam, see id. at 70-73, as a result of which the promotional

lists were not certified.

Plaintiffs allege that the non-certification vote was due to

SThe fifth member of the CSB, Barbara Tinney Jennings, wasrecused because her brother, Lt. Gary Tinney, was a candidate forpromotion on the Captain's examination. She did not attend thehearings concerning these promotional exams.

15

Page 48: Sotomayor Cases

political pressure, particularly by defendant Rev. Boise Kimber,

a vocal African-American minister who, it is acknowledged by all

parties, is a political supporter and vote-getter for Mayor

DeStefano. Plaintiffs' theory is that the defendants urged the

CSB not to certify the results in the interest of pleasing

minority voters and other constituents in New Haven whose

priority was increasing racial diversity in the ranks of the Fire

Department. Plaintiffs further argue that this pattern of

political manipulation is in keeping with prior actions by the

City of New Haven disregarding the Charter-mandated Rule of-Three

in promotional decisions in the City's police and re

departments. In support of this argument, plaintiffs proffer

evidence regarding prior litigation in the Connecticut Superior

and Appellate Court, the substance and outcome of which is

largely admitted by defendants,6 and which resulted in sharp

rebukes against the City for violating the civil service rules.

See Pl. L.R. 56(a)1 Stmt. ~~ 64-90; Def. Am. L.R. 56(a)2 Stmt. ~~

64-90. Plaintiffs argue that the apparent racial disparity in

the results of the Lieutenant and Captain exams was due to the

fact that hiring into, and promotion within, the Fire Department

historically has been based on political patronage and promotion

6Defendants challenge the relevance of this evidence;however, as the Court held in its ruling on defendants' motion tostrike, such evidence is relevant as background information tothe present case.

16

Page 49: Sotomayor Cases

of racial diversity rather than merit; and they argue that the

higher-scoring firefighters simply studied harder. In addition,

they argue that the evident disparity was not appreciably worse

on the 2003 examinations than other past promotional

examinations.

Defendants argue that "the decision not to certify [the

test] results was mandated by anti-discrimination federal," state

and local laws." Def. Mem. in Support of Mot. for Summary

Judgment [Doc. # 52] at 4. Alternatively, they argue that they

had a good faith belief that Title VII mandated non-certification

of the examinations, and they cannot be liable under Title VII

for attempting to comply with that very statute. Defendants

additionally argue that plaintiffs lack standing to bring their

Equal Protection claim, or, if they do have standing, the claim

lacks merit because all firefighters were treated the same,

regardless of race, as no orie was promoted as ~ result of the

contested exams.

Plaintiffs counter that a "good faith belief" that

certifying the test results would violate Title VII does not

constitute a defense, as a matter of law, to an allegation of

Title VII or Equal Protection violations against the plaintiffs.

II. Standard

Summary judgment is appropriate where "there is no genuine

issue as to any material fact and ... the moving party is

17

Page 50: Sotomayor Cases

entitled to a judgment as a matter of law." Fed. R. Civ. P.

56(c). An issue of fact is "material" if it "might affect the

outcome of the suit under the governing law," and is "genuine" if

"the evidence is such that a reasonable jury could return a

verdict for the nonmoving party." Anderson v. Liberty Lobby,

Inc., 477 U.S. 242, 248 (1986).

On cross-motions for summary judgment "neither side is

barred from asserting that there are issues of fact, sufficient

to prevent the entry of judgment, as a matter of law, against it.

When faced with cross-motions for summary judgment, a district

court is not required to grant judgment as a matter of law for

one side or the other." Heublein, Inc. v. United States, 966

F.2d 1455, 1461 (2d Cir. 1993) (citing Schwabenbauer v. Board of

Educ. of Olean, 667 F.2d 305, 313 (2d Cir. 1981». "Rather, the

court must evaluate each party's motion on its own merits, taking

care in each instance to draw all reasonable inferences against

the party whose motion is under consideration." Schwabembauer,

667 F.2d at 314.

III. Discussion

A. Title VII

Plaintiffs argue that defendants' decision and/or advocacy

against certifying the exam results amounted to intentional

discrimination against plaintiffs, 17 of whom are white and one

of whom is Hispanic, in favor of Hispanic and African-American

18

Page 51: Sotomayor Cases

examinees who were favored due to their race and their alleged

political support of Mayor DeStefano, via the Rev. Boise Kimber.

Plaintiffs essentially argue that defendants' professed desire to

comply with Title VII's anti-disparate-impact requirements was in

fact a pretext for intentional discrimination against white

candidates. Because plaintiffs allege intentional

discrimination, the familiar McDonnell Douglas three-prong

burden-shifting test applies.

1. Burden-Shifting Framework

Under that framework, plaintiffs first must establish a

prima facie case of discrimination on account of race.

Weinstock v. Columbia Univ., 224 F.3d 33, 42 (2d Cir. 2000). To

do so, they must prove: (1) membership in a protected class; (2)

qualification for the position; (3) an adverse employment action;

and (4) circumstances giving"rise to an inference of

discrimination on the basis of membership in the protected class.

"See, ~, McDonnell Douglas Corp. v.Green, 411 U.S. 792, 802

(1973), Graham v. Long Island R.R., 230 F.3d 34, 38 (2d Cir.

2000). "A plaintiff's burden of establishing a prima facie case

is de minimis." Abdu-Brisson v. Delta Airlines, Inc., 239 F.3d

456, 467 (2d Cir. 2001). Defendants do not dispute the first

three prongs of the test, but argue that plaintiffs cannot

establish an inference of discrimination because all applicants

were treated the same, as nobody was promoted off the examination

19

Page 52: Sotomayor Cases

lists.

Proof of a prima facie case shifts the burden to defendant

"to produce evidence that the plaintiff was [terminated] for a

legitimate, nondiscriminatory reason. This burden is one of

production, not persuasion; it can involve no credibility

assessment." Reeves v. Sanderson Plumbing, 530 U.S. 133, 142

(2000). (internal citations, quotations, and alterations omitted).

Defendant's burden is satisfied if the proffered evidence "'taken

as true, would permit the conclusion that there was a

nondiscriminatory reason for the adverse action.'" Schnabel v.

Abramson, 232 F.3d 83, 88 (2d Cir. 2000) (quoting St. Mary's

Honor Ctr. v. Hicks, 509 U.S. 502, 509 .(1993». In this case,

defendants proffer a good faith attempt to comply with Title VII

as their legitimate nondiscriminatory reason for refusing to

certify the exams.

If the employer articulates a neutral reason for the

plaintiff's termination, the burden shifts back to the plaintiff

to show pretext. That is, the plaintiff "may attempt.to

establish that he was the victim of intentional discrimination by

showing that the employer's proffered explanation is unworthy of

credence." Reeves, 530 U.S. at 143.

2. Prima Facie Case

Plaintiffs' evidence- and defendants' own arguments - show

that the City's reasons for advocating non-certification were

20

Page 53: Sotomayor Cases

related to the racial distribution of the results. As the

transcripts show, a number of witnesses at the CSB hearings,

including Kimber, mentioned "diversityH as a compelling goal of

the promotional process. Ude, Marcano, and Burgett specifically'

urged the CSB not to certify the results because, given the

number of vacancies at that time, no African-Americans would be

eligible for promotion to either Lieutenant Dr Captain, and no

Latinos would be eligible for promotion to Captain. They

believed this to be an undesirable outcome that could subject the

City to Title VII litigation by minority firefighters,- and the

City's leadership to political consequences. Had the tests not

yielded what defendants perceived as racially disparate results,

defendants would not have advocated rejecting the tests, and

plaintiffs would have had an opportunity to be promoted ..

A jury could infer that the defendants were motivated by a

concern that too many whites and not enough minorities would be

promoted were the lists to be certified. Given their minimal

prima facie burden, the Court will assume arguendo that

plaintiffs have proffered sufficient evidence to satisfy the

fourth prong of the prima facie case, given defendants'

acknowledgment that racial concerns, i.e. the disparate impact of

the test results on minority firefighters, provided the impetus

for their actions.

21

Page 54: Sotomayor Cases

3. Pretext/Discriminatory Intent

Defendants proffer as their legitimate non-discriminatory

reason that they desired to comply with the letter and the spirit

of Title VII. Plaintiffs deride this ~feigned desire to 'comply'

with Title VII," Pl. Mem. of Law [Doc. I 81] at 3, arguing that

defendants in fact violated that statute, and their actions were

a mere pretext for promoting the interests of African-American

firefighters and political supporters of the mayor.

As plaintiffs point out, this case presents the opposite

scenario of the usual challenge to an employment or promotional

examination, as plaintiffs attack not the use of allegedly

racially discriminatory exam results, but defendants' reason for

their refusal to use the results. See Pl. Mem. of Law at 32, 34

35. Ordinarily, as contemplated by the statute, the "complaining

party" bears the burden of proving a disparate impact, and the

"respondent" bears the burden" of "demonstrat[ing] that the

challenged practi~e is job related fbr" the positio~ in question

and consistent with business necessity," or, alternatively, the

"complaining party" may prevail by showing that an alternative

employment practice with less disparate impact existed and that

the respondent failed to utilize"it. See 42 U.S.C. § 2000e-2(k).

Here, the roles of the parties are in essence reversed, with the

defendants, normally reflecting a "respondent" role in the Title

VII disparate impact analysis, contending that use of the

22

Page 55: Sotomayor Cases

promotional exams, if they had been certified, would have had an

adverse impact, and the plaintiffs, normally the "complaining

party," arguing that the test results were suff iently job-

related to be defensible under the law.

a. Existence of Racially Disparate Impact

Although the parties dispute the exact racial breakdown of

candidates passing the Captain's test,' plaintiffs do not dispute

that the results showed a racially adverse impact on African-

American candidates for both the Lieutenant and Captain

positions, as judged by the EEOC Guidelines. Pl. L.R. 56 Stmt. ~

246; Def. L.R. 56 Stmt. ~ 246. Thus, it is necessarily

undisputed that, had minority firefighters challenged the results

the examinations, the City would have been in a position of

defending tests that, under applicable Guidelines, presumptively

had a disparate racial impact.

Specifically, the EEOC "four-fifths rule" provides that a

selection tool that yields "[al selection rate for any race, sex,

or ethnic group which is less than· four-fifths (4/5) (or eighty

percent) of the rate for the group with the highest rate will

generally be regarded by the Federal enforcement agencies as

evidence of adverse impact, while a greater than four-fifths rate

'Plaintiffs assert that 32% of African-American examineespassed the Captain's examination, while defendants assert thefigure is 37.5%. See Marcano Aff., Def. Ex. 4, ~ 21; Pl. L.R.56(a) Stmt. ~~ 244-47.

23

Page 56: Sotomayor Cases

will generally not be regarded by Federal enforcement agencies as

evidence of adverse impacL" 29 C.F.R. § 1607.4(D).

Here, the evidence shows that on the 2003 Lieutenant's exam

the pass rate for whites was 60.5%, for African-Americans 31.6%

and Hispanics 20%. The four-fifths score would be 48%. In other

words, African-Americans had a pass rate that was about half the

pass rate for Caucasians, yielding an adverse impact ratio

("AIR") of 0.59, significantly below the AIR of 0.80 that is

presumed to not evidence adverse impact under the'EEOC

Guidelines. See Pl. L.R. 56(a) Stmt. ~ 246; Def. L.R. 56(a)

Stmt. ~ 246. While- the parties dispute the Captain's exam pass

rate for African-Americans and Hispanics (see supra note 7), the

pass rate far Caucasians was 88%, which is more than double that

of minorities and thus by either party's statistic an AIR far

below the four-fifths guideline is yielded.

Plaintiffs argue that these AIRs were not appreciably

different from those on past promotional exams, and therefore

defendants' stated concern with avoiding adverse impact must be

pretextual. The parties agree that the AIRs on the 1999

promotional examinations would have failed the four-fifths rule

as well. The AIR for African~Americans on the 1999 Lieutenant's

exam was 0.58, compared to 0.59 on the 2003 test. See Pl. L.R.

56(a) Stmt. ~ 246; Def. L.R. 56(a) Stmt. ~ 246. The 1999 Captain

examination had an AIR of 0.45 on African-American test-takers.

24

Page 57: Sotomayor Cases

See Pl. Ex. Vol. I, 40 (1999 scores).

However, it is also undisputed that, because of the Rule of

Three, the pass rate is not synonymous with the promotion rate,

because only the top three scorers may be considered for each

vacant position. Thus, the rank of the minority applicants is

also a key factor. In 2003, given the number of vacancies, it

appeared that at most two Hispanics and no African-Americans

would have the opportunity to be promoted to Captain, and no

minor~ties would have the opportunity to be promoted to

Lieutenant. Although the record lacks specification, witnesses

at the CSB hearings testified to the effect that in 1999 more

minority candidates had scored toward the top of the lists, and

therefore had more promotional opportunities.

In any event, in 2003 defendants' concern was with the

absence of minority candidates potentially eligible to be

promoted, and with the diversity of the Fire Department's

management in general. Thus, the fact that the 1999 exams also

had a statistically adverse impact yet were certified, while the

2003 results were not, is insufficient in itself to show that

defendants' concerns about complying with Titl~ VII were

pretextual.

b. Validation Study and Less DiscriminatoryAlternatives

Plaintiffs additionally argue that defendants' decision was

pretextual because they failed to complete a validation study to

25

Page 58: Sotomayor Cases

test whether the 2003 exams could be defended as adequately job­

related. Going further, plaintiffs argue that defendants were

legally required to conduct such a validation study before

rendering a decision on cert cation of the results.

Title VII provides: nNotwithstanding any other provision

. it shall not be an unlawful employment practice for . . . an

employer to give and to act upon the results of any

professionally developed 'ability test prov~ded,that such test~

its administration or action upon the results is not designed,

intended or used to discriminate because of race, color,

religion, sex or national ~rigin." 42 U.S.C. § 2000e-2(h). As

plaintiffs concede, this section nprovides that professionally

developed and properly validated tests ~re a defense to a claim

of disparate impact." Def. Mem. of Law at 32 (emphasis

supplied). The statute itself does not require employers to

implement or continue to use any test simply because it is

professionally developed, nor does it provide a defense to an

employer who "use[s]" a test with a discriminatory impact where

other less-discriminatory, equally effective, alternatives are

available. 42 U.S.C. § 2000e-21h).

Although plaintiffs argue that EEOC guidelines mandated that

defendants conduct a validation study before deciding not to

certify the exams, the language of the guidelines does not

support such a requirement. A validation .study is a method for

26

Page 59: Sotomayor Cases

determining whether a test is sufficiently related to the

position for which the test or other criterion is administered.

The EEOC's Uniform Guidelines for Employee Selection Procedures

create a presumption that "[t]he use of any selection procedure

which has an adverse impact on the hiring, promotion, or other

employment or membership opportunities of members of any race,

sex, or ethnic group will be considered to be discriminatory and

inconsistent with these guidelines, unless the procedure has been

validated in accordance with these guidelines." 29 C.F.R. §

l607.3(A). The Guidelines further state:

Where two or more selection procedures are availablewhich serve the user's legitimate interest in efficientand trustworthy workmanship, and which aresubstantially equally valid for a given purpose, theuser should use the procedure which has beendemonstrated to have the lesser adverse impact.Accordingly, whenever a validity study is called for bythese guidelines, the user should include, as a part ofthe validity study, an investigation of suitablealternative selection procedures and suitablealternative methods of using the selection procedurewhich have as little adverse impact as possible, todetermine the appropriateness of using or validatingthem in accord with these guidelines.

Id. at § l607.3(B) .

. Where a selection procedure results in an adverseimpact on a race, sex, or ethnic group . . . and thatgroup is a significant factor in the relevant labormarket, the user generally should investigate thepossible existence of unfairness for that group if itis technically feasible to do so. The greater theseverity of the adverse impact on a group, the greaterthe need to investigate the possible existence ofunfairness.

29 C.F.R. § l607.l4(B) (8) (b) The Guidelines provide technical

27

Page 60: Sotomayor Cases

guidance for three types of studies: criterion-related validity

studies, content validity studies, and construct validity

studies. See generally 29 C.F.R. § 1607.14.

The Guidelines are written on the assumption that the

employer would be defending a certain test and seeking to

validate such test in response to a disparate impact challenge

from protected group employees. They do not address the

tuation in the present case of an employer rejecting a test

without conducting a validation study. Nonetheless, it is

evident from the language of the guidelines that a validation

study is contemplated as one method by which an employer can

defend its use of a test or other selection method it desires to

utilize by demonstration that it is sufficiently job-related -to

pass muster under the statute, despite a racially adverse impact.

The guidelines do not require or mandate a validity study where

an employer decides using a certain selection procedure

that manifests this impact and plaintiff's argument that

defendants violated Title VII by refusing to conduct a validity

study before rejecting testing results is thus unpersuasive.

Plaintiffs argue that the CSB did not have extensive

evidence of the existence of other, less-discriminatory, and

equally-effective selection measures. Dr. Hornick telephonically

testified that other tests, particularly ones he had developed,

generally yield less adverse impact, and mentioned that an

28

Page 61: Sotomayor Cases

~assessment center approach" might benefit New Haven, without

specifically explaining what that approach entailed. As

plaintiffs argue, there was no testimony that an ~assessment

center" approach has a demonstrably less adverse impact, and

there is some evidence in the record in this case, including from

Dr. Hornick's website, that such an approach may still have some

adverse impact. Dr. Hornick acknowledged that he had not had

time to review the exams carefully, and his comments illustrated

lack of famil rity with the methods lOS utilized to develop the

tests. He suggested that lack of internal review by members of

the New Haven Fire Department could have yielded questions that

were less relevant to the particular department, but offered no

explanation of why such a circumstance would have an adverse

impact on minority candidates in particular. Dr. Helms from

Boston College testified that the racial disparity 6n the exams

at issue were not significantly different from the statistical

disparities apparent on standardized tests nationwide. Mr.

Lewis, the arson specialist from the Department of Homeland

Security, stated that he believed the tests were fair and focused

on material that a Lieutenant or Captain should kno~.

On the other hand, Dr. Hornick and representatives of the

black firefighters' union suggested that the 60/40 weighting

system for the oral and written examinations could have produced

an adverse impact. The testimony suggested that changing the

29

Page 62: Sotomayor Cases

weighting system yielded increased minority pass rates and

diversity in the ranks of Bridgeport fLrefighters and officers.

Dr. Helms suggested that because different employees have

different ways of doing the same job, the fact that approximately

2/3 of those interviewed for the JAQwere white could have

unintentionally introduced a bias into the test instrument. She

and Mr. Lewis also suggested that differences in the availability

of formal training and informal mentoring may have created the

disparate effect apparent in the results.

Plaintiffs purport to counter this argument with affidavits

emphasizing how much they studied and sacrificed to perform well

on the exams, compared to their observations of the efforts of

some other examinees, and point to the availability of study

groups and informal mentoring in the department.

It appears that the reasons for testing disparities remain

elusive. Dr. Helms testified that many theories exist, but

experts on standardized testing nationwide have been unable to

satisfactorily fully explain the reasons for the disparity in

. performance observed on many tests.

Plaintiffs' argument boils down to the assertion that if

defendants cannot prove that the disparities on the Lieutenant

and Captain exams were due to a particular flaw inherent in those

exams, then they should have certified the results because there

was no other alternative in place. Notwithstanding the

30

Page 63: Sotomayor Cases

shortcomings in the evidence on existing, effective alternatives,

it is not the case that defendants must certify a test where they

cannot pinpoint its deficiency explaining its disparate impact

under the four-fifths rule simply because they have not yet

formulated a better selection method.

c. Diversity Rationale

The real crux of plaintiffs' argument is that defendants

refused to explore alternatives or conduct a validity study

because they had already decided that they did not like the

inevitable promotional results if the process continued to its

expected conclusion,8 and that their udiversityH rationale is

prohibited as reverse discrimination under Title VII.

In Haydenv. County of Nassau, 180 F.3d 42 (2d Cir. 1999),

the Second Circuit held that race-conscious configuration of an

entry-level police department exam did not violate Title VII or

the Equal Protection Clause. In that case, the Nassau County

Police Department was operating under several cons~nt decrees

prohibiting it from engaging in discrimination in its selection

of police officers, and particularly from utilizing examinations

with disparate impact on minority applicants. Following

development of a test by the county and Department of Justice

8Plaintiffs present evidence in the form of emails from theMayor's staff suggesting they desired to convince the CSB not tocertify, and further suggesting that if the CSB had certified,the Mayor would have announced his intention to refuse to forwardthe lists to the Fire Department for promotion.

31

Page 64: Sotomayor Cases

advisors, a validity analysis was conducted to determine which

configuration of the test was sufficiently job-related "yet

minimized the adverse impact on minority applicants. Of the

twenty-five sections administered to the applicants, the

[technical report] recommended that Nassau County use nine

sections as the . test." Id. at 47. A class of White and

Latino officers challenged use of the adjusted test under Title

VII and the Fourteenth Amendment, inter alia, contending that the

deliberate design of the test to reduce adverse impact on

African~American candidates necessarily discriminated against

them on the basis of race." The Court of Appeals rejected the

plaintiffs' contentions, finding plaintiffs were "mistaken in

treating racial motive.as a synonym for a constitutional

.violation" and observing that "[e]veryantidiscrimination statute

aimed at racial discrimination, and every enforcement measure

taken under such a statute, reflect a concern with race. That

does not make such enactments or actions unla~ftil or

automatically suspect ... " Id. at 48-49 (quoting Raso v. Lago,

135 F.3d II, 16 (1st Cir.») (internal quotation marks omitted).

The Hayden court further held that the construction of the Nassau

County test for the purpose of minimizing adverse impact on

minorities was not intentional "reverse discrimination" against

whites because the same nine test sections were used for all

applicants, so it was "simply not analogous to a quota system or

32

Page 65: Sotomayor Cases

33

Page 66: Sotomayor Cases

claim of discrimination to serve as a predicate for a voluntary

compromise containing race-conscious remedies." Id. at 1130.

The Second Circuit expanded Kirkland in Bushey v. New York

State Civil Service Commission, 733 F. 2d 220 (2d Cir. 1984).

There, the civil service commission had administer.ed a

promotional examination that had a significant adverse impact,

with non-minority applicants passing at almost twice the rate of

minority ap.plicants. The defendants race-normed the scores for

each group, increasing the pass rate of the minority group to the

equivalent of the non~minority group, and effectively making an

additional 8 minority individuals eligible for promotion, without

taking any non-minorities off the list. The Court of Appeals

held that the initial results, particularly ~the score

distributions of minority and nonminority candidates, were

sufficient to establish a prima facie showing of adverse impact,"

id. at 225, and, consistent with Kirkland, ~a showing of a prima

facie case of employment discrimination through a statistical

demonstration of disproportional racial impact constitutes a

sufficiently serious claim of discrimination to serve as a

predicate for employer-initiated, voluntary race-conscious

remedies," id. at 228. In other words, a prima facie case is one

way that a race-conscious remedy is justified, but it is not

required: all that is required is ~a sufficiently serious claim

of discrimination" to warrant such a remedy. Id. at 228; see

34

Page 67: Sotomayor Cases

also id. at 226 n. 7.

In this case, the parties agree that the adverse impact

ratios for African-American and Hispanic test-takers on both the

Lieutenant and Captain exams were too low to pass muster under

the EEOC's "four-fifths rule." As Kirkland and Bushey held, a

statistical showing of discrimination, and particularly a pass

rate below the "four-fifths rule," is sufficient to make out a

prima facie case of discrimination, and therefore sufficient to

justify voluntary race-conscious remedies. 9 Here, defendants'

remedy is "race conscious" at most because their actions

reflected their intent not to implement a promotional process

based on testing results that had an adverse impact on African-

Americans and Hispanics. The remedy chosen here was decidedly

less "race conscious" than the remedies in Kirkland and Bushey,

9Plaintiffs denigrate reliance on Kirkland and Bushev on thegrounds that the "race-norming" procedures utilized in thosecases would be unlawful under the 1991 amendments to the CivilRights Act. 42 U.S.C. § 2000e-2(1) ("It shall be an unlawfulemployment practice for a respondent, in connection with theselection or referral of applicants or candidates for employmentor promotion, to adjust the scores of, use different cutoffscores for, or otherwise alter the .results of, employment relatedtests on the basis of race, color, religion, sex, or nationalorigin."). See also Hayden, 180 F.3d at 53 (this provision was"intended to prohibit 'race norming' and other methods of usingdifferent cut-offs for different races or altering scores basedon race.") (emphasis in original). While plaintiffs are correctthat Title VII now prohibits race-norming, none is alleged tohave happened here and the 1991 amendments do not affect thereasoning and holding of either case, namely, that a showing of a"sufficiently serious claim of discrimination" is adequate tojustify race conscious, remedial measures.

35

Page 68: Sotomayor Cases

because New Haven did not race-norm the scores, they simply

decided to start over, to develop some new assessment mechanism

with less disparate impact. Thus, while the evidence shows that

race was taken into account in the decision not to certify the

test results, the result was race-neutral: all the test results

were discarded, no one was promoted, and firefighters of every

race will have to participate in another selection process to be

considered for promotion. Indeed, there is a total absence of

any evidence of discriminatory animus towards plaintiffs - under

the reasoning of Hayden, 180 F.3d at 51, "nothing in our

jurisprudence precludes the use of race-neutral means to improve

racial and gender representation. [T]he intent to remedy

the disparate impact of the prior exams is not equivalent to an

intent to discriminate against non-minority applicants. H1o

lOTaxman v. Bd. of Educ. of T'wp of Piscataway, 91 F.3d1547, 1558 (3d Cir. 1996) (en banc) , cert. dismissed, 522 U.S.1010 (1997), relied on by plaintiffs, is readily distinguishable.There, the board of education relied on an affirmative actionplan to defend its decision to layoff a white teacher instead ofa black teacher with equal seniority, and the Third Circuit heldthat promoting racial diversity on the faculty, absent a historyof past discrimination, was insufficient justification for layingoff the white teacher because of her race and violated Title VII.Here, defendants had ample statistical evidence that the testshad an adverse impact on minority candidates and importantly didnot opt to select black applicants over white applicants forpromotion, but rather decided to select nobody at all. Williamsv. Consolo City of Jacksonville, No. 00cv469, 2002 U.S. Dist.LEXIS 27066 (D. Fla. July 5, 2002), can similarly bedistinguished as that case did not concern a decision not tocertify test results, but rather a post-certification decisionnot to create the positions which would result in plaintiffs'promotions because plaintiffs were not African-American.

36

Page 69: Sotomayor Cases

Plainti contend that Hayden is distinguishable by the

fact that the remedy approved there was pursuant to previous

consent decrees; they do not explain why they view this

distinction as significant. As Bushey held, it would contravene

the remedial purpose of Title VII if an employer were required to

await a lawsuit before voluntarily implementing measures with

less discriminatory impact. Bushey, 733 F.2d at 227 (rejecting

the plaintiffs' argument that the remedial measures in

Kirkland were only permissible as part of a settlement in that

case, because that "would create an anomalous situation. It

would require an employer. '.. ' to issue a presumptively

discriminatory eligibility list, wait to be sued by minority

candidates, and only then seek a settlement .... Such an approach

would serve no purpose other than to impede the process of

voluntary compliance with Title VII and cause the proliferation

of litigation in all such cases ... . ff).

Plaintiffs also attempt to distinguish Hayden on the grounds

that the challengers to that test, which was constructed from the

nine most job-related sections with the least disparate impact,

were not injured or disadvantaged, whereas "the instant

plaintiffs have been both injured, as they were deprived of

promotions, and disadvantaged as they will now be forced to

compete once again." Pl. Mem. of Law at 58. Plaintiffs take

this language from Hayden out of context. In holding that the

37

Page 70: Sotomayor Cases

Hayden plaintiffs did not prove disparate impact on nonminority

applicants, the Court of Appeals held that because "appellants

continued to score higher than black candidates, on average, the

exam did not impair or disadvantage these appellants in favor of

African-American applicants. Thus, appellants are unable to set

forth a claim that they endured any disparate impact as a result

of the design and administration of the . examination."

Hayden, 180 F:3d at 52. Here, plaintiffs. allege disparate

.treatment, not disparate impact. Nor do they have a viable claim

of disparate impact because the decision to disregard the test

results affected all applicants equally, regardless of race - all

applicants will have to participate in a new test or selection

procedure. 11

Furthermore, plaintiffs were not "deprived of promotions."

As the parties agree, under New Haven's civil service ru~es, no

applicant is entitled .to promotion unless and until the CSB

certifies the results. Even then, application of the Rule of

Three would give top scorers an opportunity for promotion,

depending on the number of vacancies, but no guarantee of

llWhile plaintiffs, who describe their considerable effortsto perform well on this infrequently given promotion exam, areunderstandably disappointed and frustrated that their successfulstudy efforts have come to naught this time, this result is notevidence of being disadvantaged because of their race norevidence of disparate impact because it does not show injury ordisadvantage, only uncertainty as to their performance in theCity's next promotion selection process.

38

Page 71: Sotomayor Cases

promotion; it is even conceivable that the applicant with the

highest score never would be promoted. See United States v. City

of Chicago, 869 F.2d 1033, 1038 (7th Cir. 1989) (where state law

permitted promotion from among five highest-ranked individuals on

eligibility list, challenger had no property right to promotion:

ft a roster ranking may create an expectation of promotion, but an

officer has no entitlement to a particular roster position or to

promotion."); Bridgeport Firebird Society v. City of Bridgeport,

686 F. Supp. 53, 58 (ftAt best, the provisions of the City Charter

[mandating a Rule of One for promotions] provide the refighters

ranked on the eligibility list only with a mere expectation

of promotion, which does not rise to the level of a legally

protected interest, especially in the face of 'presumptively

discriminatory employment practices.'") (quoting Kirkland, 711 F.

2d at 1126».

Thus, while the facts of Hayden were slightly different than

those here, the Court finds the holding quite relevant and

instructive. Defendants' motivation to avoid making promotions

based on a test with a racially disparate impact, even in a

political context,12 does not, as a matter of law, constitute

12Assuming arguendo that political favoritism or motivationsmay be shown to have been intertwined with the race concern, thatdoes not suffice to establish a Title VII violation. See,~,

EEOC v. Flasher Co., Inc., 986 F.2d 1312, 1321 (10th Cir. 1992)(pretext is not shown merely because ft some less seemly reasonpersonal or political favoritism, a grudge, random conduct, anerror in the administration of neutral rules actually accounts

39

Page 72: Sotomayor Cases

discriminatory intent, and therefore such evidence is

insufficient for plaintiffs to prevail on their Title VII claim.

Accordingly, the Court will grant defendants' motion and deny

plaintiffs' motion for summary judgment on this claim.

B. Equal Protection Claim

Plaintiffs argue that defendants violated the Equal

Protection Clause either by employing a race-based classification

system for promotion or, alternatively, by applying facially

neutral promotion criteria in a racially discriminatory manner.

Defendants counter that they did not employ any racial

classi cations because every applicant was treated the same when

the CSB decided that nobody would be promoted off the lists, and

there was no discriminatory intent against whites motivating

their non-certification decision. Additionally, defendants argue

that plaintiffs lack standing to bring an Equal Protection claim.

1. Standing

Defendants acknowledge, as they must, that non-minorities

have been found to be in a protected group for purposes of

standing under the Equal Protection Clause. See,~, Adarand

Constructors, Inc. v. Pena, 515 U.S. 200, 210 (1995) (holding

that non-minority-owned business' ~allegation that it has lost a

contract in the past because of a [minority set~aside]

subcontractor compensation clause of course entitles it to seek

for the decision") .

40

Page 73: Sotomayor Cases

damages for the loss of that contract."). However, defendants

argue that because plaintiffs have not suffered any harm, and

specifically because plaintiffs were not "passed over for a

benefit that was given to an allegedly less deserving minority,"

Def. Reply Mem. at 37, they lack standing.

Defendants confuse standing with the merits of the case.

The constitutional injury plaintiffs claim here is not failure to

be promoted, but failure to be treated equally on the basis of

race. Plaintiffs have standing to bring such a claim. Comer

v. Cisneros, 37 F.3d 775~' 791 (2d Cir. 1994) (plaintiff had

standing to bring equal protection claim where she alleged that

the defendant's Section 8 housing subsidy program "rules and

regulations, in their administration, violate the Constitution

because they erect a barrier that makes it more difficult for

economically disadvantaged blacks to obtain a housing benefit

than it was for rion-minorities") .

2. Racial Classification/Discriminatory Intent

Plaintiffs' Equal Protection claim, however, lacks merit,

with respect to both the racial classification and disparate

treatment arguments. As the Second Circuit held in Hayden when

rejecting plaintiffs' classification argument, if an exam is

"administered and scored in an identical fashion for all

applicants," there is no racial classification. Hayden, 180 F.3d

at 48. Further, a "desire" "to design an entrance exam which

41

Page 74: Sotomayor Cases

would diminish the adverse impact on black applicants

not constitute a 'racial classification.'" Id. Here, all

does

applicants took the same test, and the result was the same for

all because the test results were discarded and nobody was

promoted. This does not amount to a facial classification based

on race. I3 Likewise, where a test is administered and scored in

the same manner for all applicants, plaintiffs cannot make out a

claim that the exam was a facially neutral test used in a

discriminatory manner. at 50.

Plaintiffs argue that their equal protection rights were

violated because they passed the tests and therefore were not

similarly-situated to minority applicants who failed~Plainti

argue that if a black employee "shows up for work and works a

full day" and a white employee does,not, and the black employee

13Therefore, plaintiffs' reliance on Berkley v. United'287 F.3d 1076 (Fed. Cir. 2002), is unavailing. In that

case, the Air Force employed facially different criteria forselecting women and minority employees for layoff compared towhite male employees, and the Federal Circuit held that such aprogram should be subjected to strict scrutiny (without ruling onthe merits). Likewise, in Dallas Fire Fighters Assoc. v City ofDallas, 150 F.3d 438 (5th Cir. 1998), also relied on byplaintiffs, the city followed an affirmative action plan thatspecifically called for promoting African-American, Hispanic andfemale firefighters out of rank, ahead of white and NativeAmerican male fighters with higher test scores. Here, noclassification system was employed, as the test results werediscarded for every examinee regardless of race. Whiledefendants clearly were concerned with achieving diversity in thedepartment by enhancing minority promotional opportunity,plaintiffs offer no evidence that defendants employed an actualrace-based affirmative action plan that advantaged minority overwhite applicants for promotion.

42

Page 75: Sotomayor Cases

complains "that he was due his wages,H the employer cannot be

heard to defend the complaint on the ground that the employees

were treated the same because neither was paid. Pl. Mem. in Opp.

at 64. Plaintiffs' analogy is faulty because performing well on

the exam does not create an entitlement to promotion, whereas

working entitles an employee to be paid. Second, a presumptively

. flawed test result may not be a proper measure for determining

whether anyone should be promoted.

Finally, plaintiffs cannot show that defendants acted out of

an intentionally discriminatory purpose. "Discriminatory purpose

'implies that the decisionmaker ... selected or reaffirmed a

particular course of action at least in part 'because of,' not

merely 'in spite of,' its adverse effects upon an identifiable

group.'H (quoting Personnel Administrator v. Feeney, 442 U.S.

256, 279 (1979». Nothing in the record in this case suggests

that the City defendants or CSB acted "because of~' discriminatory

animus toward plaintiffs or other non-minority applicants for

promotion. Rather, they acted based on the following concerns:

that the test had a statistically adverse impact on African­

American and Hispanic examinees; that promoting off of this list

would undermine their goal of diversity in the Fire Department

and would fail to develop managerial role models for aspiring

firefighters; that it would subject the City to public criticism;

and that it would likely subject the City to Title VII lawsuits

43

Page 76: Sotomayor Cases

from minority applicants that, for political reasons" the City

did not want to defend. "[T]he intent to remedy the disparate

impact of [the tests] is not equivalent to an intent to

discriminate against non-minority applicants." Hayden, 180 F.3d

at 51. None of the defendants' expressed motives could suggest

to a reasonable juror that defendants acted,"because of" animus

again~t non-minority firefighters who took the Lieutenant and

Captain exams.

Accordingly, defendants' motion for summary judgment on this

claim will be granted and plaintiffs' motion will be denied. 14

C. Civil Rights Conspiracy

Title 42 U.S.C. § 1985(3) permits recovery of damages if a

plaintiff can prove a conspiracy "for the purpose of depriving,

either directly or indirectly, any person or class of persons of

the equal protection of the laws, or of equal privileges and

immunities under the laws." Because the Court has found that

plaintiffs fail to present sufficient evidence that their equal

protection rights were violated, their § 1985 conspiracy claim

must fail as well. See Mian v. Donaldson, Lufkin & Jenrette

Securities Corp., 7 F.3d 1085, 1088 (2d Cir. 1993) (evidence of

"racial or perhaps otherwise class-based, invidious

discriminatory animus" required to prevail on § 1985 claim).

14For this reason the Court need not reach defendants'arguments that they are entitled to qualified immunity on theEqual Protection claim.

44

Page 77: Sotomayor Cases

Accordingly, defendants' motion for summary judgment on this

claim will be granted.

D. First Amendment

Defendants additionally move for summary judgment on

plaintiffs' First Amendment freedom-of-association claim, which

motion will also be granted.

Plaintiffs do not attempt to rebut defendants' contentions

that plaintiffs have not identified a free speech activity in

which they participated nor claimed that any chilling of speech

resulted. Rather, plaintiffs argue that th~ CSB's non­

certification decision, and the City defendants' advocacy of that

decision, resulted from political pressure by defendant Kimber,

who threatened the CSB with ~political ramifications H if they

voted to certify the results. Plaintiffs argue that ~a jury

could rationally infer that city officials worked behind the

scenes to sabotage the promotional examinations because they knew

that, were the exams certified, the Mayor would incur the wrath

of Kimber and other influential leaders of New Haven's African­

American community." PI. Mem. in Opp. at 73.

While a jury could make such an inference, it would not lead

to the conclusion that plaintiffs' First Amendment right to

freedom of association was violated as. a matter of law. The

evidence shows that Kimber spoke at the first CSB hearing and

strenuously argued against certification, and the City defendants

45

Page 78: Sotomayor Cases

do not dispute that Kimber is a close political ally of the

Mayor. However, there is no evidence in the record to suggest

that the non-certification decision was made in retaliation for

plaintiffs' refusal to "associate with," or their expression of

disagreement with, Kimber. As with the Equal Protection claim,

the fact that defendants desired to avoid the wrath of one group

(in this case African-American firefighters and other political

supporters of Kimber and DeStefano) does not logically lead to

the conclusion that defendants intended to discriminate or

retaliate against plaintiffs because they were not members of

that group. More importantly, there is no evidence in the record

even to. suggest that defendants· knew plaintiffs' political

affiliations, i.e., whether they supported Kimber and/or

DeStefano on any issue other than the certification of these

particular exam re?ults. In sum, in plaintiffs' terms, the

record shows that defendants acted to head off.the potential

adverse impact of the promotion tests on African-American and

Hispanic firefighters in order to curry favor with minority

voters and political leaders in the City, but it does not contain

any evidence of an intent or purpose to target plaintiffs for not

supporting that political coalition or its interests. Thus,

defendants' motion for summary judgment on the First Amendment

claim must be granted.

46

Page 79: Sotomayor Cases

47

Page 80: Sotomayor Cases

distress. The Clerk is directed to close this case.

IT IS SO ORDERED.

lsiJanet Bond ArtertonUnited States Dis Judge

Dated at New Haven, Connecticut this 28th day of September, 2006.

48

Page 81: Sotomayor Cases

- 1 -

06-4996-cv

Ricci v. DeStefano

UNITED STATES COURT OF APPEALSFOR THE SECOND CIRCUIT

____________________________________

August Term, 2007

(Argued: December 10, 2007 Decided: June 9, 2008

En Banc Concurrence Decided: June 13, 2008)

Docket No. 06-4996-cv____________________________________

FRANK RICCI, MICHAEL BLATCHLEY, GREG BOIVIN, GARY CARBONE, MICHAEL CHRISTOFORO, RYAN DIVITO, STEVEN DURAND, WILLIAM GAMBARDELLA,

BRIAN JOOSS, JAMES KOTTAGE, MATTHEW MARCARELLI, THOMAS J. MICHAELS,SEAN PATTON, CHRISTOPHER PARKER, EDWARD RIORDON, KEVIN ROXBEE,

TIMOTHY SCANLON, BENJAMIN VARGAS, JOHN VENDETTO and MARK VENDETTO,

Plaintiffs-Appellants,

v.

JOHN DESTEFANO, KAREN DUBOIS-WALTON, THOMAS UDE JR., TINA BURGETT, BOISE KIMBER, MALCOLM WEBER, ZELMA TIRADO and CITY OF NEW HAVEN,

Defendant-Appellees.

____________________________________

CALABRESI, Circuit Judge, concurring in the denial of rehearing en banc:

1 I join entirely Judge Parker’s opinion concurring in the denial of a rehearing en banc. I also

2 join fully Judge Katzmann’s opinion because, as he points out, going en banc is unnecessary as all

3 that is involved in this case has already been described in the filed opinions. I write today to

Page 82: Sotomayor Cases

For an exceptionally thoughtful and thorough discussion of this area, see Richard A. Primus,1

Equal Protection and Disparate Impact: Round 3, 117 Harv. L. Rev. 494 (2003).

- 2 -

1 emphasize one reason that, I believe, makes it particularly inappropriate for us to exercise our purely

2 discretionary power to review this case en banc.

3 The question of whether a municipality incurs liability when, motivated only by a desire to

4 comply with federal anti-discrimination law, it takes race-neutral actions that have racially

5 significant consequences, is undoubtedly an interesting one. To reach that question one must,1

6 however, first examine whether the municipality’s proffered desire to comply with federal law is in

7 good faith and not a pretext. After that, we would normally ask whether that asserted desire,

8 although in good faith, is not also in part motivated by other, racial, considerations. In this case, the

9 municipality claimed that its actions were grounded solely in the desire to comply with federal law.

10 The plaintiffs alleged instead that this was not the real reason for the city’s actions, and asserted that

11 the city had other less salubrious, and directly racial-political, reasons for what it did.

12 The district court and the panel readily rejected the notion that the city’s stated reason was

13 just a pretext. But neither court went on to consider whether the city was influenced by mixed

14 motives. And that is why Judge Cabranes, in his dissent from the denial of en banc review, suggests

15 that, since the plaintiffs alleged that their race motivated the defendants’ decision, the district court

16 should have undertaken such a mixed motive analysis. He contends, that is, that the courts should

17 have examined the situation as one in which a legitimate motive may have combined with an

18 improper motive to bring about the challenged action. See Price Waterhouse v. Hopkins, 490 U.S.

19 228 (1989). He would be precisely right . . . except for the fact that that type of analysis is not

Page 83: Sotomayor Cases

It is unavailable, that is, unless we reach out and consider a legal theory that the parties have2

eschewed. Sometimes – for example, in matters of life and death – such a reaching out may beappropriate. But generally, and specifically in this case, it is not.

- 3 -

1 available to us in this case. It is not available for the most traditional of legal reasons. The parties

2 did not present a mixed motive argument to the district court or to the panel. 2

3 It is the unavailability of mixed motive analysis that makes this case an especially undesirable

4 one for elective review. The interesting issue the case might present – concerning the obligations

5 of a municipality seeking only to comply with the relevant federal anti-discrimination law – is, in

6 the circumstances before us, clouded by the allegations that something more is going on. Given the

7 plaintiffs’ failure to argue mixed-motive analysis, those allegations cannot be adequately evaluated.

8 But they nevertheless cannot help but affect how we look at the city’s actions. And they may even

9 influence, inappropriately, how we are inclined to rule on the underlying, “interesting” issue.

10 Difficult issues should be decided only when they must be decided, or when they are truly

11 well presented. When they need not be decided – and rehearing en banc is always a matter of choice,

12 not necessity – it is wise to wait until they come up in a manner that helps, rather than hinders, clarity

13 of thought. That is not so in this case.

14 For this reason too, I concur in the denial of rehearing en banc.

15

Page 84: Sotomayor Cases

I have not solicited concurrences for my opinion.1

1

06-4996-cvRicci v. DeStefano

1 DENNIS JACOBS, Chief Judge, dissenting from the denial of

2 rehearing in banc:

3

4 Along with almost half of the members of this Court, I

5 join Judge Cabranes’s dissent, which does the heavy lifting

6 on the procedural merits of in banc review. I write

7 separately to answer respectfully the concurring opinions of

8 Judge Calabresi and Judge Katzmann. 1

9 Judge Katzmann and those of my colleagues who signed

10 his opinion “recognize” that this case “presents difficult

11 issues,” but would leave further review and consideration to

12 the Supreme Court, citing a Circuit “tradition” of deference

13 to panel adjudication. In effect, this has become a Circuit

14 tradition of hearing virtually no cases in banc.

15 The grant or denial of in banc review is governed by

16 Fed. R. App. P. 35, which says that in banc rehearing is

17 disfavored--unless such review is needed for coherence of

Page 85: Sotomayor Cases

In the alternative, Judge Calabresi contends that we2

cannot consider whether the District Court applied the

correct legal standard to plaintiffs’ Title VII claim

because the “parties did not present [that] argument to the

district court or the panel” and we can only consider a

2

1 the Court’s decisions or “the proceeding involves a question

2 of exceptional importance.” Fed. R. App. P. 35(a).

3 Accordingly, the next subdivision of Rule 35 requires the

4 petition to explain why the case falls within one or both of

5 these categories. See Fed. R. App. P. 35(b).

6 This weighing calls for an exercise of discretion.

7 Judge Calabresi’s concurring opinion deprecates this

8 standard as a “purely discretionary power” that is “always a

9 matter of choice” (emphasis added). He nevertheless

10 “join[s] fully” in both Judge Parker’s opinion, which

11 counsels against in banc review as a matter of (plain

12 ordinary) discretion, and Judge Katzmann’s opinion, which

13 decides against in banc review as a matter of tradition. I

14 understand Judge Calabresi to be saying, in effect, that

15 when it comes to in banc review, discretion should be

16 leavened by caprice. As applied to this case, that means

17 that there might be discretionary grounds for denying in

18 banc review were it not already foreclosed by tradition.2

Page 86: Sotomayor Cases

“legal theory that the parties have eschewed” in such

circumstances as “matters of life and death.” Judge

Calabresi provides no authority for this proposition for the

good reason that it is unsound. Writing for a unanimous

Supreme Court, Justice Thurgood Marshall explained that

“[w]hen an issue or claim is properly before the court, the

court is not limited to the particular legal theories

advanced by the parties, but rather retains the independent

power to identify and apply the proper construction of

governing law.” Kamen v. Kemper Fin. Servs., Inc., 500 U.S.

90, 99 (1991); see also Hankins v. Lyght, 441 F.3d 96, 104(2d Cir. 2006) (“We are required to interpret federalstatutes as they are written . . . and we are not bound byparties’ stipulations of law.”); Neilson v. D'Angelis, 409F.3d 100, 105 n.2 (2d Cir. 2005) (“The parties’ apparentagreement on the standard of ‘similarity’ for ‘class of one’cases does not control our judgment, because this court isnot bound by stipulations of law.”); United States v.Pabon-Cruz, 391 F.3d 86, 97 (2d Cir. 2004) (“It is clearthat we have the authority to resolve this question despiteits not having been raised in the District Court proceedingsor in the parties’ initial briefs.”).

3

1 This occluded view of our discretion to sit in banc

2 runs counter to the criteria set down for our guidance in

3 Rule 35. No doubt, the proper exercise of discretion

4 results in the denial of review in the overwhelming number

5 of cases. And the resulting pattern may resemble the

6 pattern of denial that would result from saying “no” by

7 tradition. But the decision to grant or deny in banc review

8 is like any other discretionary decision in the sense that

9 discretion should be exercised, not elided or stuck in a

Page 87: Sotomayor Cases

4

1 default position. See United States v. Campo, 140 F.3d 415,

2 419 (2d Cir. 1998) (holding that “refusal to exercise

3 discretion accorded [the court] by law . . . constitutes an

4 error of law”).

5 The exercise of discretion to hear cases in banc is

6 integral to the judicial process. The advisory notes

7 emphasize that “an en banc proceeding provides a safeguard

8 against unnecessary intercircuit conflicts.” See Fed. R.

9 App. P. 35, Advisory Committee Notes (1998 Amendments). In

10 other words, issues of exceptional importance that may

11 divide the circuits should be subject to in banc review lest

12 a three-judge panel adopt a rule of law that would not

13 command a majority vote of the appeals court as a whole, and

14 thereby provoke an avoidable circuit conflict that the

15 Supreme Court would have to resolve.

16 That is why I respectfully disagree with those of my

17 colleagues who are pleased to defer as a matter of tradition

18 to the ruling of the three-judge panel, and thereby leave

19 further consideration to the Supreme Court. Cf. Landell v.

20 Sorrell, 406 F.3d 159, 167 (2d Cir 2005) (Sack, J., and

Page 88: Sotomayor Cases

5

1 Katzmann, J., concurring) (observing that in banc hearing

2 should be avoided where it “would only forestall resolution

3 of issues destined appropriately for Supreme Court

4 consideration”).

5 I do not think it is enough for us to dilate on

6 exceptionally important issues in a sheaf of concurrences

7 and dissents arguing over the denial of in banc review. If

8 issues are important enough to warrant Supreme Court review,

9 they are important enough for our full Court to consider and

10 decide on the merits. Of course, if an in banc poll

11 discloses broad-based agreement with the panel opinion, in

12 banc review may be a spinning of wheels. Under such

13 circumstances, it may very well be an appropriate exercise

14 of discretion to deny rehearing in banc. But to rely on

15 tradition to deny rehearing in banc starts to look very much

16 like abuse of discretion.

Page 89: Sotomayor Cases

UNITED STATES COURT OF APPEALSFOR THE SECOND CIRCUITAugust Term, 2005(Argued: June 8, 2006 Decided: January 25, 2007)Docket Nos. 04-6692-ag(L), 04-6693-ag(CON), 04-6694-ag(CON), 04-6695-ag(CON), 04-6696-ag(CON), 04-6697-ag(CON), 04-6698-ag(CON), 04-6699-ag(CON)_____________________________________________ RIVERKEEPER, INC., NATURAL RESOURCES DEFENSE COUNCIL, WATERKEEPERALLIANCE, SOUNDKEEPER, INC., SCENIC HUDSON, INC., SAVE THE BAY-PEOPLEFOR NARRAGANSETT BAY, FRIENDS OF CASCO BAY, AMERICAN LITTORALSOCIETY, DELAWARE RIVERKEEPER NETWORK, HACKENSACK RIVERKEEPER,INC., NEW YORK/NEW JERSEY BAYKEEPER, SANTA MONICA BAYKEEPER, SANDIEGO BAYKEEPER, CALIFORNIA COASTKEEPER, COLUMBIA RIVERKEEPER,CONSERVATION LAW FOUNDATION, SURFRIDER FOUNDATION, STATE OF RHODEISLAND, STATE OF CONNECTICUT, STATE OF DELAWARE, COMMONWEALTH OFMASSACHUSETTS, STATE OF NEW JERSEY, STATE OF NEW YORK, APPALACHIANPOWER COMPANY, ILLINOIS ENERGY ASSOCIATION, UTILITY WATER ACT GROUP,PSEG FOSSIL LLC, PSEG NUCLEAR LLC, ENTERGY CORPORATION, Petitioners,– v. – UNITED STATES ENVIRONMENTAL PROTECTION AGENCY, STEPHEN L. JOHNSON, in his official capacity as ADMINISTRATOR OF THE UNITED STATES ENVIRONMENTAL PROTECTION AGENCY, Respondents.____________________________________________Before: STRAUB, SOTOMAYOR, and HALL, Circuit Judges.____________________________________________Petitioners challenge a final rule promulgated by the Environmental ProtectionAgency pursuant to section 316(b) of the Clean Water Act, 33 U.S.C. § 1326(b), that is intendedto protect aquatic organisms from being harmed or killed by cooling water intake structures atlarge, existing power-producing facilities. While we conclude that certain aspects of the rule are

Page 90: Sotomayor Cases

2

based on a reasonable interpretation of the Clean Water Act and supported by substantialevidence in the administrative record, several aspects of the rule are not consistent with thestatute, are not supported by sufficient evidence, or were not properly subject to notice andcomment. We therefore grant in part and deny in part the petitions for review and dismiss in partone aspect of the petitions for lack of jurisdiction because there is no final agency action toreview. REED W. SUPER, Morningside Heights LegalServices, Inc., Environmental Law Clinic, ColumbiaUniversity School of Law (Michelle Avallone, JuliaErrea, Vivian Mills, Ian Dattner, Monique Mendez,Misti Duvall, Devon Knowles, Molly McOwan,Adam Orford, Scott Sneddon, on the brief; P. KentCorell, of counsel), New York, New York, forPetitioners Riverkeeper, Inc., Natural ResourcesDefense Council, Waterkeeper Alliance,Soundkeeper, Inc., Scenic Hudson, Inc., Save theBay–People for Narragansett Bay, Friends ofCasco Bay, American Littoral Society, DelawareRiverkeeper Network, Hackensack Riverkeeper,Inc., New York/New Jersey Baykeeper, SantaMonica Baykeeper, San Diego Baykeeper,California Coastkeeper, Columbia Riverkeeper,Conservation Law Foundation, and SurfriderFoundation.TRICIA K. JEDELE, Special Assistant AttorneyGeneral of Rhode Island, Providence, Rhode Island(Patrick C. Lynch, Attorney General of RhodeIsland; Michael Rubin, Special Assistant AttorneyGeneral, Providence, Rhode Island; RichardBlumenthal, Attorney General of Connecticut,Kimberly Massicotte and Matthew Levine,Assistant Attorneys General, Hartford, Connecticut;Carl C. Danberg, Attorney General of Delaware,Kevin Maloney, Deputy Attorney General,Wilmington, Delaware; Thomas F. Reilly, AttorneyGeneral of Massachusetts, Andrew Goldberg,

Page 91: Sotomayor Cases

3

Assistant Attorney General, Boston, Massachusetts;Zulima V. Farber, Attorney General of New Jersey,Ellen Barney Balint, Deputy Attorney General,Trenton, New Jersey; Eliot Spitzer, AttorneyGeneral of New York, Maureen F. Leary, AssistantAttorney General, Albany, New York, on the brief),for State Petitioners Rhode Island, Connecticut,Delaware, Massachusetts, New Jersey, and NewYork.KRISTY A.N. BULLEIT, Hunton & Williams,Washington, D.C. (James N. Christman, ElizabethE. Aldridge, Hunton & Williams, Richmond,Virginia, on the brief), for Petitioners AppalachianPower Company, Illinois Energy Association, andUtility Water Act Group.KARL S. LYTZ, Latham & Watkins LLP, SanFrancisco, California (Christopher J. McAuliffe,PSEG Services Corporation, Newark, New Jersey;David J. Hayes, Cassandra Sturkie, Latham &Watkins LLP, Washington, D.C., on the brief), forPetitioners PSEG Fossil LLC and PSEG NuclearLLC.CHUCK D. BARLOW, Entergy Services, Inc.,Jackson, Mississippi and ELISE N. ZOLI, GoodwinProcter LLP, Boston, Massachusetts (Robert H.Fitzgerald, U. Gwyn Williams, Goodwin ProcterLLP, Boston, Massachusetts, on the brief), forPetitioner Entergy Corp.DAVID S. GUALTIERI, CYNTHIA J. MORRIS,and JESSICA O’DONNELL, United StatesDepartment of Justice (Sue Ellen Woolridge,Assistant Attorney General, John C. Cruden,Deputy Assistant Attorney General, on the brief;Leslie J. Darman, United States EnvironmentalProtection Agency, of counsel), Washington, D.C.,for Respondents.Lisa Madigan, Attorney General of Illinois,Matthew Dunn, Chief, Environmental Enforcement/

Page 92: Sotomayor Cases

4

Asbestos Litigation Division, Ann Alexander,Environmental Counsel and Assistant AttorneyGeneral, Chicago, Illinois, for Amicus Curiae Stateof Illinois.Jon Bruning, Attorney General of Nebraska,Lincoln, Nebraska (David D. Cookson, SpecialCounsel to the Attorney General, Lincoln,Nebraska; Troy King, Attorney General ofAlabama, Montgomery, Alabama; Gregory D.Stumbo, Office of the Attorney General of theCommonwealth of Kentucky, Frankfort, Kentucky;Wayne Stenehjem, Attorney General of NorthDakota, Bismarck, North Dakota; Paul G. Summers,Attorney General of Tennessee, Nashville,Tennessee; Steve Carter, Office of the IndianaAttorney General, Indianapolis, Indiana, on thebrief), for State Amici Curiae Nebraska, Alabama,Kentucky, North Dakota, Tennessee, and Indiana.Nancy Elizabeth Olinger, Assistant AttorneyGeneral (Greg Abbott, Attorney General of Texas,Barry R. McBee, First Assistant Attorney General,Edward D. Burbach, Deputy Attorney General forLitigation, Karen W. Kornell, Chief, NaturalResources Division, on the brief), Austin, Texas, forAmicus Curiae Texas Commission onEnvironmental Quality. Russell S. Frye, FryeLaw PLLC, Washington, D.C.,for Amicus Curiae American Petroleum Institute. Jonathan F. Lewis, Clean Air Task Force, Boston,Massachusetts, for Amici Curiae Healthlink,Kentucky Resources Council, New England CleanWater Action, The Ohio Environmental Council,and Ohio Valley Environmental Council.Lisa Heinzerling, Georgetown University LawCenter, Washington, D.C., for Amicus Curiae OMBWatch.

Page 93: Sotomayor Cases

1 We refer to statutory provisions mentioned in the text by their section in the CWA andprovide in citations both the section of the Act and the parallel section of the United States Code.5

SOTOMAYOR, Circuit Judge:This is a case about fish and other aquatic organisms. Power plants and otherindustrial operations withdraw billions of gallons of water from the nation’s waterways each dayto cool their facilities. The flow of water into these plants traps (or “impinges”) large aquaticorganisms against grills or screens, which cover the intake structures, and draws (or “entrains”)small aquatic organisms into the cooling mechanism; the resulting impingement and entrainmentfrom these operations kill or injure billions of aquatic organisms every year. Petitioners herechallenge a rule promulgated by the Environmental Protection Agency (“the EPA” or “theAgency”) pursuant to section 316(b) of the Clean Water Act (“CWA” or “the Act”), 33 U.S.C.§ 1326(b),1 that is intended to protect fish, shellfish, and other aquatic organisms from beingharmed or killed by regulating “cooling water intake structures” at large, existing power-producing facilities. For the reasons that follow, we grant in part and deny in part the petitions forreview, concluding that certain aspects of the EPA’s rule are based on a reasonable interpretationof the Act and supported by substantial evidence in the administrative record, but remandingseveral aspects of the rule because they are inadequately explained or inconsistent with thestatute, or because the EPA failed to give adequate notice of its rulemaking. We also dismiss forlack of jurisdiction one aspect of the petitions because there is no final agency action to review. BACKGROUNDOur decision in Riverkeeper, Inc. v. EPA, 358 F.3d 174 (2d. Cir. 2004)

Page 94: Sotomayor Cases

6

(“Riverkeeper I”), which addressed challenges to the EPA’s rule governing cooling water intakestructures at new – as opposed to existing – facilities discusses at length the procedural andfactual background of the rulemaking pursuant to section 316(b). We presume familiarity withRiverkeeper I and provide here only a brief overview of the statute and the various stages of therulemaking. These consolidated petitions for review concern a final rule promulgated by theEPA regarding the water that large, existing power plants withdraw from rivers, lakes, and otherwaterways of the United States to cool their facilities. See 40 C.F.R. § 125.91(a). This coolingprocess requires power plants to extract billions of gallons of water per day from the nation’swaters, thereby impinging and entraining a huge number of aquatic organisms. Riverkeeper I,358 F.3d at 181. Indeed, a single power plant can kill or injure billions of aquatic organisms in asingle year. Id. Cognizant of this problem, Congress in 1972 amended the CWA, 33 U.S.C.§§ 1251-1387, to regulate cooling water intake structures. See Federal Water Pollution ControlAct Amendments of 1972, Pub. L. No. 92-500, 86 Stat. 816 (1972). We have describedCongress’s regulation of such structures as “something of an afterthought,” Riverkeeper I, 358F.3d at 186 n.12, given that the directive appears in a section of the Act addressing the seeminglyunrelated issue of thermal pollution, see CWA § 316(a), 33 U.S.C. § 1326(a). The Act, asamended, provides that “[a]ny standard established pursuant to section 1311 of this title [CWAsection 301] or section 1316 of this title [CWA section 306] and applicable to a point sourceshall require that the location, design, construction, and capacity of cooling water intakestructures reflect the best technology available for minimizing adverse environmental impact.”

Page 95: Sotomayor Cases

2 A “point source” is “any discernible, confined and discrete conveyance . . . from whichpollutants are or may be discharged.” 33 U.S.C. § 1362(14). 7

CWA § 316(b), 33 U.S.C. § 1326(b). The provisions of the Act cross-referenced in section 316(b) direct the EPA toissue rules regulating the discharge of pollution from existing point sources, CWA § 301, 33U.S.C. § 1311, and new point sources, CWA § 306, 33 U.S.C. § 1316.2 As we noted inRiverkeeper I, “[w]hen the EPA established new source performance discharge standard[s] . . . itought then to have regulated . . . intake structures . . . .” 358 F.3d at 185 (internal quotationmarks omitted; emphasis in original). Put differently, section 316(b) required the EPA topromulgate regulations for cooling water intake structures at the same time that it establishedpollution discharge standards pursuant to sections 301 and 306. The EPA’s first attempt atregulation under section 316(b), however, was remanded by the Fourth Circuit in 1977 onprocedural grounds, and years passed without the EPA issuing new rules. Id. at 181 (citingAppalachian Power Co. v. Train, 566 F.2d 451 (4th Cir. 1977)). Environmental groupsultimately sued the EPA and won a consent decree, pursuant to which the Agency established atimetable to issue rules pursuant to Section 316(b) in three “phases.” Id. & n.3. Phase I –addressed in Riverkeeper I – governs new facilities; Phase II – addressed here – covers large,existing power plants; and Phase III will regulate existing power plants not governed by Phase II,as well as other industrial facilities. See Riverkeeper, Inc. v. Whitman, No. 93 Civ. 0314 (AGS),2001 WL 1505497, at *1 n.3 (S.D.N.Y. Nov. 27, 2001).Our interpretation of section 316(b) is informed by the two provisions it cross-references, CWA sections 301 and 306. Section 301 sets forth a framework under which

Page 96: Sotomayor Cases

8

limitations on the discharge of pollutants from existing sources would become more stringentover time. CWA § 301(b), 33 U.S.C. § 1311(b); see Riverkeeper I, 358 F.3d at 185. Section 301(b)(1)(A) required the EPA, beginning in 1977, to set effluent limitations for existingsources based on “the best practicable control technology currently available,” or “BPT.” CWA§ 301(b)(1)(A), 33 U.S.C. § 1311(b)(1)(A). By 1989, existing source effluent limitations were tobe based on the more stringent “best available technology economically achievable,” or “BAT.” CWA § 301(b)(2)(A), 33 U.S.C. § 1311(b)(2)(A). Additionally, section 306 requires the EPA toestablish “standards of performance” for the control of the discharge of pollutants from newsources based on “the best available demonstrated control technology,” a standard that “reflectsthe greatest degree of effluent reduction.” CWA § 306(a)(1), 33 U.S.C. § 1316(a)(1). In section 316(b), Congress established yet another standard to govern coolingwater intake structures, which requires such structures to reflect the “best technology availablefor minimizing adverse environmental impact,” or “BTA.” CWA § 316(b), 33 U.S.C. § 1326(b). We noted in Riverkeeper I that “[a]lthough the EPA is permitted to consider a technology’s costin determining whether it is ‘practicable,’ ‘economically achievable,’ or ‘available,’ it shouldgive decreasing weight to expense as facilities have time to plan ahead to meet tougherrestrictions.” 358 F.3d at 185 (citations omitted). Additionally, we observed that “[b]ecausesection 316(b) refers to sections 301 and 306 but provides a different standard (‘best technologyavailable for minimizing adverse environmental impact’ instead of, for example, ‘best availabledemonstrated control technology’) and does not explicitly provide that regulations pursuant tosection 316(b) are subject to the requirements of sections 301 and 306, we think it is permissiblefor the EPA to look to those sections for guidance but to decide that not every statutory directive

Page 97: Sotomayor Cases

3 As we noted in Riverkeeper I, Cooling water systems fall into three groups. “Once-through” systems take waterin, use it to absorb heat, and return the water to its source at a higher temperature. “Closed-cycle” systems recirculate the water (after allowing it to cool off in areservoir or tower before being reused) and add water to the system only toreplace that which is lost through evaporation. Closed-cycle systems, therefore,withdraw far less water than once-through systems. Dry cooling systems . . . useair drafts to transfer heat, and, as their name implies, they use little or no water.358 F.3d at 182 n.5 (internal citations omitted). 9

contained therein is applicable” to rulemaking under section 316(b). Id. at 187. With thisgeneral background in mind, we consider Phases I and II of the EPA’s rulemaking.I. The Phase I RuleOn December 18, 2001, the EPA issued its first rule (“the Phase I Rule”)governing cooling water intake structures for new – as opposed to existing – facilities. Regulations Addressing Cooling Water Intake Structures for New Facilities; Final Rule, 66 Fed.Reg. 65,256 (Dec. 18, 2001) (codified at 40 C.F.R. pts. 9, 122-25). The Phase I Rule establisheda two-track approach to regulating cooling water intake systems at new facilities, under which anew facility could choose one of two “tracks” to comply with the statute. Track I creatednational intake capacity and velocity standards based on closed-cycle cooling technology,3 whichthe EPA deemed the best technology available for minimizing adverse environmental impacts. See Riverkeeper I, 358 F.3d at 182-83. Track II did not require the use of any specific technologyso long as the facility “can show, in a demonstration study, ‘that the technologies employed willreduce the level of adverse environmental impact . . . to a comparable level to that which’ wouldbe achieved applying Track I’s capacity and velocity requirements.” Id. at 183 (quoting 40C.F.R. § 125.84(d)(1)).

Page 98: Sotomayor Cases

10

Environmental and industry groups challenged certain aspects of the rule,including, inter alia, the part of the Track II procedure allowing power plants to comply withsection 316(b) by undertaking so-called “restoration measures,” such as restocking the waterbodywith fish, reclaiming abandoned mines to reduce drain-off, or removing barriers to fishmigration, to maintain fish and shellfish in a waterbody at certain levels. In Riverkeeper I, weupheld most aspects of the Phase I Rule, but remanded the provisions relating to the Track IIrestoration option. We found that the restoration option was inconsistent with section 316(b)’srequirement that the EPA minimize adverse environmental impacts by regulating the “‘location,design, construction, and capacity of cooling water intake structures’” because this option hasnothing to do with the location, design, construction, or capacity of such structures. Id. at 189(quoting CWA § 316(b), 33 U.S.C. § 1326(b)). Given this, we held that the EPA hadimpermissibly exceeded its authority in allowing Phase I facilities to use these restorationmeasures to comply with regulations implementing the statute. Id. II. The Phase II Rule On July 9, 2004, the EPA issued a final rule, pursuant to the second phase of theconsent decree (“the Phase II Rule” or “the Rule”), that governs cooling water intake structures atlarge, existing power plants. See Final Regulations to Establish Requirements for Cooling WaterIntake Structures at Phase II Existing Facilities, 69 Fed. Reg. 41,576 (July 9, 2004) (codified at40 C.F.R. pts. 9, 122-125). The Phase II Rule covers existing facilities that are “point sources”and that, as their primary activity, “both generate[] and transmit[] electric power, or generate[]electric power but sell[] it to another entity for transmission,” “use[] or propose[] to use coolingwater intake structures with a total design intake flow of 50 million gallons per day (MGD) or

Page 99: Sotomayor Cases

11

more,” and “use[] at least 25 percent of water withdrawn exclusively for cooling purposes.” 40C.F.R. § 125.91. Although we will discuss the specifics of the Rule with respect to eachchallenge, we provide here an overview of the Rule. The Phase II Rule sets forth five compliance alternatives. See 40 C.F.R.§ 125.94(a). Section 125.94(a) requires that a facility select and implement one of the following“for establishing best technology available for minimizing adverse environmental impact”:(1)(i) You may demonstrate to the Director that you have reduced, or will reduce,your flow commensurate with a closed-cycle recirculating system. In this case,you are deemed to have met the applicable performance standards and will not berequired to demonstrate further that your facility meets the impingement mortalityand entrainment performance standards specified in paragraph (b) of thissection. . . .; or(ii) You may demonstrate to the Director that you have reduced, or will reduce, yourmaximum through-screen design intake velocity to 0.5 ft/s or less. In this case, you aredeemed to have met the impingement mortality performance standards and will not berequired to demonstrate further that your facility meets the performance standards forimpingement mortality specified in paragraph (b) of this section and you are not subjectto the requirements in §§ 125.95, 125.96, 125.97, or 125.98 as they apply to impingementmortality. However, you are still subject to any applicable requirements for entrainmentreduction . . . .[;](2) You may demonstrate to the Director that your existing design andconstruction technologies, operational measures, and/or restoration measures meetthe performance standards specified in paragraph (b) of this section and/or therestoration requirements in paragraph (c) of this section[;](3) You may demonstrate to the Director that you have selected, and will installand properly operate and maintain, design and construction technologies,operational measures, and/or restoration measures that will, in combination withany existing design and construction technologies, operational measures, and/orrestoration measures, meet the performance standards specified in paragraph (b)of this section and/or the restoration requirements in paragraph (c) of this section;(4) You may demonstrate to the Director that you have installed, or will install,and properly operate and maintain an approved design and constructiontechnology in accordance with § 125.99(a) or (b); or

Page 100: Sotomayor Cases

4 The “calculation baseline” is “an estimate of impingement mortality and entrainmentthat would occur” at a specific site based on a number of quantitative assumptions regardingintake velocity and the location and design of the site’s intake structures. 40 C.F.R. § 125.93. The regulation also permits a site’s specific historical and current data on impingement and12

(5) You may demonstrate to the Director that you have selected, installed, and areproperly operating and maintaining, or will install and properly operate andmaintain design and construction technologies, operational measures, and/orrestoration measures that the Director has determined to be the best technologyavailable to minimize adverse environmental impact for your facility inaccordance with paragraphs (a)(5)(i) or (ii) of this section. . . .40 C.F.R. § 125.94(a). The Phase II Rule does not require large, existing power plants to install closed-cycle cooling systems, although a facility with such a system (or one whose intake flow iscommensurate with that of a closed-cycle system) will be considered in compliance with theRule. 40 C.F.R. § 125.94(a)(1)(i). The Rule instead references national performance standards,discussed below, that “are based on consideration of a range of technologies that EPA hasdetermined to be commercially available for the industries affected as a whole.” 69 Fed. Reg. at41,598-99. And rather than limiting BTA to technologies based on closed-cycle cooling systems,the EPA designated a “suite” of technologies – including fine- and wide-mesh wedgewirescreens, aquatic filter barrier systems, barrier nets, and fish return systems, among others, id. at41,599; see also 40 C.F.R. § 125.99(a) – as BTA for large, existing power plants.Section 125.94(b) establishes national performance standards to be achievedthrough one of the compliance alternatives set forth in section 125.94(a). With respect toimpingement mortality, it provides that facilities choosing “compliance alternatives in paragraphs(a)(2), (a)(3), or (a)(4) of this section . . . must reduce impingement mortality for all life stages offish and shellfish by 80 to 95 percent from the calculation baseline.”4 Id. § 125.94(b)(1). With

Page 101: Sotomayor Cases

entrainment to serve as a basis for this baseline. Id. 13

respect to entrainment, facilities that choose compliance alternatives in paragraphs (a)(1)(ii), (a)(2), (a)(3), or (a)(4) of thissection . . . must also reduce entrainment of all life stages of fish and shellfish by60 to 90 percent from the calculation baseline if: (i) [the] facility has a capacity utilization rate of 15 percent or greater, and (ii)(A) . . . uses cooling water withdrawn from a tidal river, estuary, ocean, or oneof the Great Lakes; or (B) . . . uses cooling water withdrawn from a freshwaterriver or stream and the design intake flow of your cooling water intake structuresis greater than five percent of the mean annual flow.Id. § 125.94(b)(2). Section 125.94(c) permits facilities to comply with the Rule by implementingrestoration measures “in place of or as a supplement to installing design and control technologiesand/or adopting operational measures that reduce impingement mortality and entrainment.” Id.§ 125.94(c). In order to adopt restoration measures under the Rule, a facility must demonstratethat “meeting the applicable performance standards or site-specific requirements through the useof design and construction technologies and/or operational measures alone is less feasible, lesscost-effective, or less environmentally desirable than meeting the standards . . . through the useof restoration measures.” Id. § 125.94(c)(1). Moreover, the restoration measures implementedby the facility must produce ecological benefits “at a level that is substantially similar” to whatwould be achieved by meeting the national performance standards of section 125.94(b). Id.§ 125.94(c)(2). The compliance provision of section 125.94(a)(5) permits what is in effect a site-specific compliance alternative to the generally applicable performance standards in two

Page 102: Sotomayor Cases

5 The NPDES process is promulgated under CWA section 402(a)(1), 33 U.S.C.§ 1342(a)(1). 14

circumstances. In the first circumstance (“the cost-cost compliance alternative” or “the cost-costvariance”), if a facility demonstrates that its compliance costs “would be significantly greaterthan the costs considered by the Administrator,” the permitting authority must make a site-specific determination of BTA that is “as close as practicable to the applicable performancestandards . . . without resulting in costs that are significantly greater than the costs considered bythe Administrator” in establishing those standards. 40 C.F.R. § 125.94(a)(5)(i). In the secondcircumstance (“the cost-benefit compliance alternative” or “the cost-benefit variance”), thepermitting authority must make a site-specific determination of BTA that is “as close aspracticable” to the national performance standards if a facility demonstrates that its compliancecosts would be “significantly greater than the benefits of complying” with the performancestandards at the facility. Id. § 125.94(a)(5)(ii).For those facilities installing technologies designated as BTA, section 125.94(d)allows the national performance standards set forth in section 125.94(b) to be satisfied bydemonstrating compliance with a technology installation and operation plan (“TIOP”), whichconcerns, inter alia, a facility’s installation, operation and maintenance of BTA. As the Rule isenforced through the permitting process under the National Pollutant Discharge EliminationSystem (“NPDES”),5 section 125.94(d)(1) provides that a facility that uses one of the compliancemethods other than closed-cycle cooling may request that compliance with the nationalperformance standards during the first permit cycle be determined with respect to whether thefacility has complied with the TIOP it submitted with its permit application. Section

Page 103: Sotomayor Cases

6 The parties filed petitions for review here as well as in several of our sister circuits. Thepetitions were consolidated in the Ninth Circuit by order of the judicial panel on multi-districtlitigation pursuant to 28 U.S.C. §§ 1407 and 2112(a)(3). The Ninth Circuit thereafter transferredthe case here pursuant to 28 U.S.C. § 2112(a)(5).15

125.94(d)(2) authorizes facilities to request that compliance during subsequent permit terms bedetermined based on whether a facility remains in compliance with its TIOP, and in accordancewith any necessary revisions, “if applicable performance standards are not being met.” 40 C.F.R.§ 125.94(d)(2).Finally, section 125.94(f) applies solely to nuclear power facilities. It providesthat if a nuclear facility’s compliance with the Rule would conflict with a safety requirementestablished by the Nuclear Regulatory Commission, the EPA must make a site-specificdetermination of BTA that would not conflict with the Commission’s safety requirement. 40C.F.R. § 125.94(f).For purposes of judicial review, the Phase II Rule was promulgated on July 23,2004. See 69 Fed. Reg. at 41,576. Three sets of petitioners, discussed below, brought timelychallenges to the Rule.6

DISCUSSIONI. Standard of Review

We have jurisdiction to review this Rule pursuant to CWA section 509(b)(1), 33U.S.C. § 1369(b)(1). See Riverkeeper I, 358 F.3d at 183 (stating that the Phase I Rule is coveredby the jurisdictional grant of § 1369(b)(1)). As we explained in Riverkeeper I, our substantivereview is twofold. “First, we examine the regulation against the statute that contains the EPA’scharge.” Id. at 184. If Congress “has directly spoken to the precise question at issue” and its

Page 104: Sotomayor Cases

16

intent is clear, we “must give effect to the unambiguously expressed intent of Congress.” Chevron U.S.A., Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837, 842-43 (1984). If,however, the statute is silent or ambiguous, we ask whether “the agency’s answer is based on apermissible construction of the statute.” Id. at 843. Second, if the agency has followed Congress’s unambiguously expressed intent orpermissibly construed an ambiguous statute, “we measure the regulation against the recorddeveloped during the rulemaking, but we ‘hold unlawful’ the agency’s regulation only if it is‘arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.’” Riverkeeper I, 358 F.3d at 184 (quoting 5 U.S.C. § 706(2)(A)). “Normally, we must deemarbitrary and capricious an agency rule where ‘the agency has relied on factors which Congresshas not intended it to consider, entirely failed to consider an important aspect of the problem,offered an explanation for its decision that runs counter to the evidence before the agency, or isso implausible that it could not be ascribed to a difference in view or the product of agencyexpertise.’” Waterkeeper Alliance, Inc. v. EPA, 399 F.3d 486, 498 (2d Cir. 2005) (quoting MotorVehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983) (internal quotationmarks and citations omitted)).Finally, our review has a procedural dimension. The Administrative ProcedureAct (“APA”) requires that notice of proposed rulemaking be published in the Federal Register, 5U.S.C. § 553(b)(3), and that interested parties be allowed an opportunity to comment onproposed rules, id. § 553(c). Where an agency fails to comply with the APA’s notice andcomment provisions, we remand to the agency for further proceedings. See Sprint Corp. v. FCC,315 F.3d 369, 371 (D.C. Cir. 2003) (“Because the [agency] failed to provide adequate notice and

Page 105: Sotomayor Cases

7 Rhode Island, Connecticut, Delaware, Massachusetts, New Jersey, and New York.8 Riverkeeper, Inc., Natural Resources Defense Council, Waterkeeper Alliance,Soundkeeper, Inc., Scenic Hudson, Inc., Save the Bay–People for Narragansett Bay, Friends ofCasco Bay, American Littoral Society, Delaware Riverkeeper Network, Hackensack Riverkeeper,Inc., New York/New Jersey Baykeeper, Santa Monica Baykeeper, San Diego Baykeeper,California Coastkeeper, Columbia Riverkeeper, Conservation Law Foundation, and SurfriderFoundation. 17

opportunity to comment, we grant the petition and remand the case to the [agency].”).II. The Petitions for ReviewThe state petitioners7 and the self-styled environmental petitioners8 challenge theRule on similar grounds. Their petitions contain the following arguments: (1) the EPA exceededits authority in rejecting closed-cycle cooling as BTA for existing facilities generally, and theAgency’s rejection of closed-cycle cooling as BTA for facilities on sensitive waterbodies is notentitled to deference because the decision was made at the direction of the Office of Managementand Budget (“OMB”); (2) the EPA exceeded its authority by establishing ranges of acceptableperformance rather than a single-numeric performance standard; (3) the CWA does not allowrestoration measures as a means of compliance; (4) the EPA failed to give adequate notice that itwould allow site-specific determinations of BTA based on cost-cost analysis, and the EPAimpermissibly construed the statute to allow site-specific determinations of BTA based on cost-benefit analysis; (5) the provision allowing that compliance with the performance standards bedetermined by establishing compliance with a facility’s TIOP is unauthorized and violates therulemaking requirement of notice and comment; and (6) the Agency has classified certain newconstructions as “existing facilities” contrary to the definitions set forth in the Phase I Rulewithout providing adequate notice and opportunity for comment.

Page 106: Sotomayor Cases

9 UWAG petitions this court in conjunction with the Appalachian Power Company andthe Illinois Energy Association. 18

Three groups of industry petitioners, which we will refer to collectively as the“industry petitioners” or individually as Entergy Corporation (“Entergy”), the Utility Water ActGroup (“UWAG”),9 and PSEG Fossil LLC and PSEG Nuclear LLC (“PSEG”), advance variouschallenges to the Phase II Rule. Their challenges raise the following arguments: (1)section 316(b) of the CWA does not apply to existing facilities; (2) the Agency’s definition of“adverse environmental impact” is insufficiently supported by the record; (3) the EPA’sassumption of zero entrainment survival is insufficiently supported by the record; (4) the EPAimproperly requires evaluation of qualitative non-use benefits in site-specific cost-benefitanalyses; (5) the Agency failed to account for the Rule’s disproportionate impact on nuclearfacilities; (6) the EPA gave inadequate notice of the independent-supplier provision; and (7) theAgency provided no notice of its post-rulemaking definition of “Great Lakes.” The industrypetitioners also seek to preserve the right on this petition for review to raise new challenges to theRule if we remand significant aspects of it. We consider first the challenges raised by the stateand environmental petitioners and then will turn to the industry petitioners’ challenges. III. The State and Environmental Petitioners

A. Determination of BTAPerhaps the most significant challenge to the Phase II Rule is the petitioners’contention that the EPA exceeded its authority in rejecting closed-cycle cooling, and selectinginstead the suite of technologies, as the “best technology available” as required by section 316(b),33 U.S.C. § 1326(b), in large part because the Agency engaged in improper cost considerations.

Page 107: Sotomayor Cases

19

This challenge requires us at the outset to determine to what extent, if any, the EPA can considercost when selecting “the best technology available for minimizing adverse environmentalimpact” under the statute.1. Cost Analysis Pursuant to Sections 301 and 306Section 316(b) does not itself set forth or cross-reference another statutoryprovision enumerating the specific factors that the EPA must consider in determining BTA. Thestatute, however, does make specific reference to CWA sections 301 and 306, which we havetaken previously as “an invitation” to look to those sections for guidance in “discerning whatfactors Congress intended the EPA to consider in determining” BTA. Riverkeeper I, 358 F.3d at186. We look to each of these statutes in turn.Section 301(b)(1)(A) established the BPT standard that governed the effluentlimitations applicable to existing sources through 1989. Congress provided that, in determiningBPT, the Agency could consider “the total cost of application of technology in relation to theeffluent reduction benefits to be achieved from such application.” CWA § 304(b)(1)(B), 33U.S.C. § 1314(b)(1)(B). As noted above, however, the CWA created standards that were tobecome increasingly stringent over time, and in 1989, the more lenient BPT standard for existingsources was replaced by the BAT standard of section 301(b)(2)(A), in which Congress providedthat the EPA could consider only “the cost of achieving such effluent reduction.” CWA§ 304(b)(2)(B), 33 U.S.C. § 1314(b)(2)(B). Notably omitted from the list of permissible factorsto which the EPA could look in determining BAT was the cost of technology in relation to thebenefits that technology could achieve.This shift from BPT to BAT fundamentally altered the way in which the EPA

Page 108: Sotomayor Cases

20

could factor cost into its CWA determinations. Indeed, in analyzing BTP and BAT, the SupremeCourt stated that in “assessing BAT[,] total cost is no longer to be considered in comparison toeffluent reduction benefits,” as it had been in assessing BPT. EPA v. Nat’l Crushed Stone Ass’n,449 U.S. 64, 71 (1980). The Court indicated that the less stringent BPT standard had allowed fora “limited cost-benefit analysis” intended to “‘limit the application of technology only where theadditional degree of effluent reduction is wholly out of proportion to the costs of achieving suchmarginal level of reduction.’” Id. at 71 n.10 (quoting Remarks of Senator Muskie reprinted inLegislative History of the Water Pollution Control Act Amendments of 1972 (Committee Printcompiled for the Senate Committee on Public Works by the Library of Congress) Ser. No. 93-1,p. 170 (1973)). In determining BAT, by contrast, the EPA may consider cost as a factor to alimited degree, see id., but only as to whether the cost of a given technology could be reasonablyborne by the industry and not the relation between that technology’s cost and the benefits itachieves, Riverkeeper I, 358 F.3d at 195. Section 306, which governs the effluent limitations that apply to new sources,provides that a “standard of performance” established by the EPA must reflect the “best availabledemonstrated control technology.” CWA § 306(a)(1), 33 U.S.C. § 1316(a)(1). In languageidentical to the text of § 304(b)(2)(B) governing BAT, Congress provided that in establishingstandards of performance, the EPA “shall take into consideration the cost of achieving sucheffluent reduction,” CWA § 306(b)(1)(B), 33 U.S.C. § 1316(b)(1)(B), but did not require theEPA to conduct cost-benefit analysis. Nat’l Wildlife Fed’n v. EPA, 286 F.3d 554, 570 (D.C. Cir.2002) (“[S]ection 306 requires that, when setting the [new source performance standards], theAdministrator must take costs into consideration, but does not require that she conduct a

Page 109: Sotomayor Cases

10 See, e.g., OMB Circular A-94, Guidelines and Discount Rates for Benefit-Cost Analysisof Federal Programs, Appendix A (1992) (defining “benefit-cost analysis” as “[a] systematicquantitative method of assessing the desirability of government projects or policies when it isimportant to take a long view of future effects and a broad view of possible side-effects” and“cost-effectiveness” as “[a] systematic quantitative method for comparing the costs of alternativemeans of achieving the same stream of benefits or a given objective”).21

cost-benefit analysis.”). Sections 301 and 306 of the CWA thus demonstrate that, after 1989,cost is a lesser, more ancillary consideration in determining what technology the EPA shouldrequire for compliance under those sections. The shift from the BPT standard to the more stringent BAT one clearly signaledCongress’s intent to move cost considerations under the CWA from a cost-benefit analysis to acost-effectiveness one. We understand the difference between these two analyses to turn on thedifference between means and ends. Cost-benefit analysis, like BPT, compares the costs andbenefits of various ends, and chooses the end with the best net benefits. By contrast, cost-effectiveness considerations, like BAT, determine which means will be used to reach a specifiedlevel of benefit that has already been established.10 Given the above and considering the parallellanguage of sections 304(b)(2)(B) and 306(b)(1)(B), the reasoning of National Crushed Stonestrongly suggests that cost-benefit analysis is no longer permitted under those sections of theCWA. 2. Cost Analysis Pursuant to Section 316(b)As already noted, section 316(b) does not itself set forth the factors that theAgency can consider in determining the “best technology available for minimizing adverseenvironmental impact.” The BTA standard of section 316(b), however, is linguistically similarto the BAT standard of section 301 and the standard that applies to new sources under

Page 110: Sotomayor Cases

22

section 306, and to the extent that cost-benefit analysis is precluded under those statutes, onemight reasonably conclude that it is similarly not permitted under section 316(b). We concludein any event that the language of section 316(b) itself plainly indicates that facilities must adoptthe best technology available and that cost-benefit analysis cannot be justified in light ofCongress’s directive. We stated in Riverkeeper I that the EPA can consider cost in establishing BTA,but only in a limited fashion and not as a primary consideration. Indeed, “[w]ith respect to costs,‘the Administrator must inquire into the initial and annual costs of applying the technology andmake an affirmative determination that those costs can be reasonably borne by the industry.’”Riverkeeper I, 358 F.3d at 195 (quoting Chem. Mfrs. Ass’n v. EPA, 870 F.2d 177, 262 (5th Cir.1989)) (emphasis added). While the statutory language suggests that the EPA may consider costsin determining BTA, in that a technology that cannot not be reasonably borne by the industry isnot “available” in any meaningful sense, cost-benefit analysis is not similarly supported by thelanguage or purpose of the statute. Section 316(b) expressly requires a technology-driven result,cf. Natural Res. Def. Council, Inc. v. EPA, 822 F.2d 104, 123 (D.C. Cir. 1987) (“[T]he mostsalient characteristic of [the CWA’s] statutory scheme, articulated time and again by its architectsand embedded in the statutory language, is that it is technology-forcing.”), not one driven by costconsiderations or an assessment of the desirability of reducing adverse environmental impacts inlight of the cost of doing so. A selection of BTA based on cost-benefit considerations is thusimpermissibly cost-driven, but a selection based in part on cost-effectiveness considerations,while taking cost into account, remains technology-driven. The statute therefore precludes cost-benefit analysis because “Congress itself defined the basic relationship between costs and

Page 111: Sotomayor Cases

11 This conclusion accords with the analysis in Riverkeeper I that the EPA may considercost in establishing BTA pursuant to section 316(b), but only in a limited way. In our discussionof the EPA’s choice of closed-cycle, rather than dry, cooling as BTA for Phase I facilities, see358 F.3d at 194-95, 194 n.22, we noted that “dry cooling costs more than ten times as much peryear as closed-cycle wet cooling,” but emphasized that “it is estimated to reduce water intake byonly an additional 5 percent relative to once-through cooling.” Id. at 194 (internal footnotesomitted). We acknowledged that dry cooling is both much more effective and much moreexpensive than closed-cycle cooling in absolute terms, but stressed that, as compared to thebaseline of once-through cooling systems, the marginal benefits of dry cooling were small: “it isundeniably relevant that that difference represents a relatively small improvement overclosed-cycle cooling at a very significant cost.” Id. at 194 n.22. In dicta, we characterized thismode of analysis as “relevant” and stated that it “adds a useful perspective,” id., but did not treatit as the fulcrum of our analysis. Ultimately, we deferred to the EPA’s determination insofar as itwas based on the grounds that dry cooling was too expensive for industry reasonably to bear andthat dry cooling has negative environmental effects best left to the considered judgment of theAgency. Id. at 195-96. 23

benefits.” Am. Textile Mfrs. Inst., Inc. v. Donovan, 452 U.S. 490, 509 (1981). Moreover, thisconclusion is further supported by the fact that Congress in establishing BTA did not expresslypermit the Agency to consider the relationship of a technology’s cost to the level of reduction ofadverse environmental impact it produces.11 “When Congress has intended that an agencyengage in cost-benefit analysis, it has clearly indicated such intent on the face of the statute.” Id.at 510. Given the above, the EPA may permissibly consider cost in two ways: (1) todetermine what technology can be “reasonably borne” by the industry and (2) to engage in cost-effectiveness analysis in determining BTA. Thus, the EPA must first determine what is the mosteffective technology that may reasonably be borne by the industry. In making this initialdetermination, the most effective technology must be based not on the average Phase II facilitybut on the optimally best performing Phase II facilities, see, e.g., Kennecott v. United States EPA,780 F.2d 445, 448 (4th Cir. 1985) (“In setting BAT, EPA uses not the average plant, but the

Page 112: Sotomayor Cases

12 We note that the EPA is by no means required to engage in cost-effectiveness analysis.Indeed, to require the Agency to conduct cost-effectiveness analysis would transform suchanalysis into a primary factor in choosing BTA, which clearly is contrary to the technology-forcing principle that animates the CWA. Equally important, we note that the Agency may alsodepart from this performance benchmark because of other permissible considerations aside fromcost, for instance, energy efficiency or environmental impact. See Riverkeeper I, 358 F.3d at195-96 (noting “the EPA was permitted to consider . . . energy efficiency in determining the ‘besttechnology available’” and could also factor in environmental impact). While the EPA hasindicated throughout the record and its briefing before us that its determination of BTA hereincluded such considerations as energy efficiency and production concerns, we nevertheless mustremand, as we explain below, for further explanation from the Agency. Accordingly, we expressno view on the merits of the EPA’s determination regarding the other factors it claims influencedits decision. 24

optimally operating plant, the pilot plant which acts as a beacon to show what is possible.”),although, of course, the EPA must still ascertain whether the industry as a whole can reasonablybear the cost of the adoption of the technology, bearing in mind the aspirational and technology-forcing character of the CWA. This technology constitutes the benchmark for performance. Once this determination has been made, the EPA may then consider other factors, including cost-effectiveness, to choose a less expensive technology that achieves essentially the same results asthe benchmark.12 For example, assuming the EPA has determined that power plants governed bythe Phase II Rule can reasonably bear the price of technology that saves between 100 - 105 fish,the EPA, given a choice between a technology that costs $100 to save 99 - 101 fish and one thatcosts $150 to save 100 - 103 fish (with all other considerations, like energy production orefficiency, being equal), could appropriately choose the cheaper technology on cost-effectivenessgrounds. Cost-benefit analysis, however, is not permitted under the statute because, as noted,Congress has already specified the relationship between cost and benefits in requiring that the

Page 113: Sotomayor Cases

13 For this reason, we reject Entergy’s argument that the Rule is improper because its costexceeds its benefits. 25

technology designated by the EPA be the best available.13 Cf. Am. Textile Mfrs. Inst., 452 U.S. at509-10. The Agency accordingly could not make the policy decision, in the face of Congress’sdetermination that facilities use the best technology available, that an economically feasible levelof reduction of impingement mortality and entrainment is not desirable in light of its cost. Indeed, in the example above, the EPA could not choose the cheaper technology on costconsiderations under section 316(b) if the EPA had first determined that the power plants couldreasonably bear the cost of technology that could save at least 102 fish. We nevertheless acknowledge that the comparable technologies considered by theAgency need not be identically effective for the Agency to engage in cost-effectiveness analysis. Were that the case, all that would be required would be the simple determination of which amongcompeting technologies that achieved the same degree of reduction of adverse environmentalimpacts is the cheapest. Instead, the specified level of benefit is more properly understood as anarrowly bounded range, within which the EPA may permissibly choose between two (or more)technologies that produce essentially the same benefits but have markedly different costs. Withthese considerations in mind, we turn to the Rule as promulgated.3. The Determination of BTA under the Phase II RuleAs noted previously, unlike the Phase I Rule, the Phase II Rule does not requirefacilities to reduce intake flow to a level commensurate with the intake of closed-cycle systems.Instead, the Rule requires facilities to meet the national performance standards associated withthe suite of technologies the EPA identified as BTA. Petitioners’ challenge here has two

Page 114: Sotomayor Cases

14 While the state petitioners frame their argument as a challenge to the site-specificcompliance alternatives, their brief presses the view that closed-cycle cooling is the besttechnology available and that the EPA exceeded its authority in failing to require Phase IIfacilities to adopt it. 26

components. First, the state petitioners contend that closed-cycle cooling is the best technologyavailable and that the EPA has exceeded its authority by promulgating a rule that does not requireclosed-cycle cooling, or the use of technologies producing a commensurate reduction of waterusage for existing facilities in the same manner as the Phase I Rule required for new facilities.14 Second, the environmental petitioners argue that the EPA improperly rejected closed-cyclecooling as BTA for the largest facilities on the most sensitive waterbodies at the direction ofOMB because it sought to maximize net economic benefits rather than to minimize adverseenvironmental impact. They further argue that the BTA standard of section 316(b) requires acommitment of the maximum resources economically feasible to the goal of eliminating adverseenvironmental impacts and that the statute does not permit the EPA to select BTA on the basis ofcost-benefit analysis.For the reasons that follow, we conclude that the statute’s “best technologyavailable” standard permits cost-effectiveness considerations to influence the choice amongtechnologies whose performance does not essentially differ from the performance of the best-performing technology whose cost the industry reasonably can bear, but that the statute does notpermit the EPA to choose BTA on the basis of cost-benefit analysis. As we explain below,however, the record is unclear as to the basis for the EPA’s selection of the suite of technologiesas BTA, and we therefore remand for clarification of the basis for the Agency’s decision andpotentially for a reassessment of BTA.

Page 115: Sotomayor Cases

27

The EPA stated in the Rule’s preamble that the BTA standard should beinterpreted as “best technology available commercially at an economically practicable cost,” andexplained that “an important component of economic practicability” is “the relationship of coststo environmental benefits.” 69 Fed. Reg. at 41,604. The EPA further explained that this inquiryrequired that “there should be some reasonable relationship between the cost of cooling waterintake structure control technology and the environmental benefits associated with its use.” Id.(emphasis added).The EPA took this “economically practicable” concept directly from the text of afloor speech of a single representative – the only specific reference to section 316(b) in thecongressional debates. See Riverkeeper I, 358 F.3d at 186 n.12. We noted in Riverkeeper I thatthe “paucity” of legislative history “counsels against imputing much specific intent to Congressbeyond the section’s words themselves.” Id. Moreover, we find the EPA’s interpretation ofsection 316(b) problematic because its construction significantly resembles the less stringent, andnow obsolete, BPT standard of section 301(b)(1)(A). As noted earlier, in setting forth the factorsfor the EPA to consider in establishing BPT under section 301(b)(1)(A) and the more stringentBAT under section 301(b)(2)(A), Congress made only one distinction: while the Agency couldconsider the relationship between cost and benefits in establishing BPT, CWA § 304(b)(1)(B), 33U.S.C. § 1314(b)(1)(B), it could consider cost insofar as it can be “reasonably borne” by theindustry, but not the relationship between cost and benefits, in establishing BAT, CWA§ 304(b)(2)(B), 33 U.S.C. § 1314(b)(2)(B). Riverkeeper I, 358 F.3d at 195.This difference in how the EPA can consider cost under section 304(b) inestablishing BPT and BAT is directly mirrored by the most significant textual distinction

Page 116: Sotomayor Cases

28

between sections 301(b)(1)(A) and 301(b)(2)(A) – the requirement that a technology be“practicable” under only the less stringent BPT standard. Compare CWA § 301(b)(1)(A), 33U.S.C. § 1311(b)(1)(A) (BPT is the “best practicable control technology”) with CWA§ 301(b)(2)(A), 33 U.S.C. 1311(b)(2)(A) (BAT is the “best available technology”). The use ofthe word “practicable,” therefore, when coupled with the permissible cost considerations undersection 304, signals that Congress intended the EPA to strike a balance between cost and benefitsin determining BPT. But the word “practicable” is missing from the more stringent BATstandard, under which Congress prohibited the EPA from considering the relation of cost tobenefits. This omission is thus significant. See Russello v. United States, 464 U.S. 16, 23 (1983)(“[W]here Congress includes particular language in one section of a statute but omits it inanother section of the same Act, it is generally presumed that Congress acts intentionally andpurposely in the disparate inclusion or exclusion.” (quotation marks and citation omitted;alteration in original)). Because Congress also omitted “practicable” from section 316(b), weare troubled by the Agency’s interpretation of the statute to require “practicability” analysis hereand its implicit corollary that the Agency can undertake a cost-benefit analysis in establishingBTA under section 316(b).Our concern with the EPA's determination with section 316(b) is further deepenedby the Agency’s rejection of closed-cycle cooling and selection of a suite of technologies as thebasis for BTA for existing facilities because the suite of technologies were the most “costeffective” option. 69 Fed. Reg. at 41,667. The EPA explained this decision on several grounds. It first noted that it was rejecting closed-cycle cooling as BTA because of (1) “its generally highcosts (due to conversions),” (2) “the fact that other technologies approach the performance of this

Page 117: Sotomayor Cases

29

option,” and (3) “concerns for energy impacts due to retrofitting existing facilities, and otherconsiderations.” Id. at 41,605. The EPA emphasized that it selected BTA based on itsdetermination that “a national requirement to retrofit existing systems is not the most cost-effective approach and at many existing facilities, retrofits may be impossible or noteconomically practicable.” Id. It further explained that its rejection of closed-cycle cooling asBTA was based on “total social costs” and “lack of cost-effectiveness,” as well as “concernsregarding potential energy impacts.” Id. at 41,606. Given the EPA’s discussion, noted above, of economically practicability, it isunclear whether the Agency improperly weighed the benefits and the costs of requiring closed-cycle cooling. Indeed, a comparison between the cost of closed-cycle cooling and the monetizedbenefits of this technology appears to have played some role in the EPA’s rejection of this optionas BTA. In the preamble to the proposed Rule, for instance, the EPA examined whether torequire closed-cycle cooling on specific large bodies of waters and stated that “the incrementalcosts of [this closed-cycle cooling] option relative to the proposed option ($413 million)significantly outweigh the incremental benefits ($146 million).” 67 Fed. Reg. at 17,158. Otherrecord evidence on the EPA’s rejection of closed-cycle cooling as BTA is a terse EPAmemorandum indicating that a requirement commensurate with closed-cycle cooling for facilitieson sensitive waterbodies would cost three times as much as the option ultimately adopted by theEPA and reduce entrainment, at most, by 1.33 times that option. Given the above indications that the EPA engaged in cost-benefit analysis, weremand for the EPA to explain its conclusions. At the outset, it is difficult to discern from therecord how the EPA determined that the cost of closed-cycle cooling could not be reasonably

Page 118: Sotomayor Cases

15 Indeed, at one point in the Rule’s proposal, the EPA mentions that requiring closed-cycle cooling at 539 existing power plants nationwide subject to the Phase II Rule would costupwards of $2.26 billion and could close nine power plants (about 1.6% of all Phase II facilities). 67 Fed. Reg. at 17,155. There is little discussion about whether the industry could reasonablybear this burden. We note that in Riverkeeper I, it appears the EPA supplied the court with betterdata on this question, noting that of the 83 facilities subject to the Phase I Rule, dry coolingcompliance costs would equal on average more than 4% of revenue for all 83 producers and morethan 10% of the revenue at 12 facilities (nearly 15% of all Phase I facilities), presenting apossibly high barrier to entry for new facilities. See Riverkeeper I, 358 F.3d at 194. We havefound no comparable data in the record, nor has the EPA cited any. 16 Some limited evidence, however, is available in the record in intelligible form. In theRule’s proposal, the EPA noted that closed-cycle systems “generally reduce the water flow from72 percent to 98 percent, thereby using only 2 percent to 28 percent of the water used by once-through systems.” Proposed Regulations to Establish Requirements for Cooling Water IntakeStructures at Phase II Existing Facilities; Proposed Rule, 67 Fed. Reg. 17,122, 17,189 (Apr. 9,2002). The Agency went on to state that “[i]t is generally assumed that this would result in acomparable reduction in impingement and entrainment,” id., indicating that closed-cycle systemsreduce the adverse impacts of impingement and entrainment by 72 to 98 percent. In a technicaldevelopment document (“TDD”) for the proposal, however, the EPA disaggregated these data,stating that closed-cycle cooling systems use 96 to 98 percent less fresh water and 70 to 96percent less salt water than once-through systems. TDD for the Proposed § 316(b) Phase IIExisting Facilities Rule 4-1; see also Riverkeeper I, 358 F.3d at 194 n.22 (“[C]losed-cycle wetcooling systems use 96 to 98 percent less fresh water (and 70 to 96 percent less salt water) thansimilarly situated once-through systems.” (citing 66 Fed. Reg. at 65,273)). The EPA’s analysis inits proposal suggests that the disaggregated data indicate that closed-cycle cooling would reduceimpingement mortality and entrainment by 96 to 98 percent at facilities that use fresh water andby 70 to 96 percent at facilities that use salt water. A 72 to 98 percent reduction of impingementmortality and entrainment, and the corresponding disaggregated percent reductions for facilitiesusing salt and fresh water, differs from the Phase II requirement that facilities reduceimpingement mortality by 80 to 95 percent and entrainment by 60 to 90 percent. Thesedifferences seem potentially significant, especially in determining whether this suite of BTAtechnologies achieve essentially the same result as closed-cycle cooling, but are neither explained30

borne by the industry.15 Additionally, the EPA did not explain its statement that the suite oftechnologies “approach[es]” the performance of closed-cycle cooling. We see no adequatecomparison in the Rule’s proposal, the final Rule or its preamble, or the EPA’s submissions tothis Court of the effectiveness of closed-cycle cooling and the group of technologies whoseeffectiveness provided the basis for the Phase II Rule’s performance standards.16 In a technical

Page 119: Sotomayor Cases

nor adequately compared for purposes of our review here. For instance, there does not appear tobe any discussion regarding the seemingly large differences in the rates of impingement andentrainment reduction between closed-cycle cooling and the Phase II national performancestandards. 17 The Agency, however, did provide some indication of the relative costs of closed-cyclecooling and the suite of technologies identified as BTA. The EPA stated in the preamble to theRule that the “total social cost” of closed-cycle cooling would be “$3.5 billion per year,” 69 Fed.Reg. at 41,605, and that the “final rule will have total annualized social (pre-tax) costs of $389million,” id. at 41,650. 31

area of this sort, it is difficult for judges or interested parties to determine the propriety of theAgency’s action without a justification for the action supported by clearly identified substantialevidence whose import is explained. The record evidence alone here, which consists in large partof a voluminous database compilation of studies that assess the efficacy of various technologiesat different locations, is oblique, complicated, and insufficient to permit us to determine what theEPA relied upon in reaching its conclusion. As the Supreme Court has emphasized, “[o]urrecognition of Congress’ need to vest administrative agencies with ample power to assist in thedifficult task of governing a vast and complex industrial Nation carries with it the correlativeresponsibility of the agency to explain the rationale and factual basis for its decision, even thoughwe show respect for the agency’s judgment in both.” Bowen v. Am. Hosp. Ass’n, 476 U.S. 610,627 (1986) (plurality opinion).The EPA was required to explain its judgment and the basis for it. Because theEPA purported to base its decision in large part on cost-effectiveness considerations, it wasrequired to identify and explain any evidence indicating a minimal performance differencebetween comparable technologies, but it did not do so here. It stated only that the performanceof the technologies it identified as BTA “approach” the performance of closed-cycle cooling.17

Page 120: Sotomayor Cases

32

We therefore find it impossible to judge whether the performance of these technologies isessentially the same as the performance of closed-cycle cooling, or whether they simply arecheaper per percentage point of reduction in entrainment and impingement mortality. That is, onthe record before us, it is impossible to tell whether the EPA based its decision on permissiblecost-effectiveness analysis or exceeded its authority by relying impermissibly upon a cost-benefitanalysis. To the extent that the record does not indicate the EPA’s basis, however, its statementthat “the relationship of costs to environmental benefits is an important component of economicpracticability,” 69 Fed. Reg. at 41,604, indicates that cost-benefit analysis, under the cover ofconsiderations of “practicability,” was central to the Agency’s decisionmaking.In short, the EPA’s failure to explain its decision frustrates effective judicialreview. If the EPA construed the statute to permit cost-benefit analysis, its action was not “basedon a permissible construction of the statute.” Chevron, 467 U.S. at 843. It may also be that theEPA misunderstood or misapplied cost-effectiveness analysis. If so, its decision was arbitraryand capricious because the Agency relied on factors Congress has not intended it to consider. See Waterkeeper Alliance, 399 F.3d at 498. Finally, the EPA may have simply failed either toperform the required analysis or to explain adequately a decision that was within its authority tomake. We cannot opine on this subject, because we must consider only those justifications thatthe EPA offered at the time of the rulemaking. See SEC v. Chenery, 318 U.S. 80, 87-88 (1943). Moreover, while the EPA could rely on factors other than impingement and entrainment inestablishing BTA, such as negative environmental impacts or concerns about energy productionand efficiency, see Riverkeeper I, 358 F.3d at 195-96, we are unable to determine, on the recordbefore us, whether the EPA gave paramount consideration to an improper factor in determining

Page 121: Sotomayor Cases

18 As previously noted, the environmental petitioners, supported by amicus curiae OMBWatch, have challenged the EPA’s decision to reject closed-cycle cooling as BTA for 59facilities on the most sensitive waterbodies in part because of OMB’s participation in therulemaking process. They contend that the EPA’s action is not entitled to deference because theEPA was improperly influenced by OMB in promulgating this aspect of the Phase II Rule. Thepetitioners and the EPA have also sought either to strike from the record or to supplement therecord with certain OMB review documents. Because we have granted the petitioners’ challengeto the EPA’s determination of BTA without considering OMB’s role in interagency review, wedo not reach the petitioner’s arguments regarding OMB’s involvement in the rulemaking anddeny the motions to strike and to supplement as moot. See Waterkeeper Alliance, 399 F.3d at524 n.34 (denying petitioners’ motion to supplement the record with OMB review documents asmoot where the Court granted the petitions without considering the OMB review documents).33

BTA. We therefore remand for clarification of the basis for the Agency’s action and possibly fora new determination of BTA.184. Performance Standards Expressed as RangesThe Phase II Rule establishes performance standards expressed as an 80 to 95percent reduction in impingement mortality and a 60 to 90 percent reduction in entrainment,which existing power plants must achieve, subject to certain exceptions, in order to be consideredin compliance with the Rule. 40 C.F.R. § 125.94(b)(1), (2). The environmental petitionerschallenge the Rule’s “wide and indeterminate ranges” as failing to constitute “precise single-levellimitations based on the best technology available for minimizing adverse environmental impact”and argue that these ranges are inconsistent with Congress’s intent that there be a nationalstandard under section 316(b). We agree in part and, because the EPA in reconsidering itsselection of BTA on remand may alter the suite of technologies it originally selected, therebycausing a coordinate alteration in the performance ranges, we provide some guidance to the EPAinsofar as the petitioners’ challenge touches on the limits of the Agency’s authority. Althoughthe EPA may, in the circumstances to be discussed, set performance standards as ranges, it must

Page 122: Sotomayor Cases

34

require facilities to minimize the adverse environmental impacts attributable to their coolingwater intake structures to the best degree they can. The petitioners note that the EPA has found that certain screens and filter systemscan reduce impingement mortality by up to 99 percent and that similar technologies can produce80 to 90 percent reduction in entrainment. 69 Fed. Reg. at 41,599. They contend that the CWAtherefore requires the EPA to set BTA standards reflecting these best performers, see Texas Oil& Gas Ass’n v. EPA, 161 F.3d 923, 928 (5th Cir. 1998) (“Congress intended these [BAT]limitations to be based on the performance of the single best-performing plant in an industrialfield.” (citation and internal quotation marks omitted)), particularly given the EPA’sacknowledgment that “[t]he higher end of the range is a percent reduction that available datashow many facilities can and have achieved with the available technologies upon which theperformance standards are based.” 69 Fed. Reg. at 41,600. The petitioners emphasize that theRule’s ranges impermissibly fail to require facilities even to attempt to achieve performanceequal to the upper bound of the prescribed ranges. According to the EPA, section 316(b) does not require a single-numeric standardapplicable to all Phase II existing facilities, and expressing the performance standards as rangesis necessary to account for the variables involved in reducing impingement mortality andentrainment under local conditions at particular facilities. The EPA contends that “[b]ecause thePhase II requirements are applied in a variety of settings and to existing facilities of differenttypes and sizes, no single technology is most effective for all facilities subject to the Rule.” TheAgency argues that the technologies do not provide a fixed level of performance at all facilitiesand that their performance is affected by the nature of the waterbody, facility intake

Page 123: Sotomayor Cases

19 The specific provision requires that when a facility’s permit expires, any reissuedpermit must contain a requirement that “the facility . . . reduce impingement mortality andentrainment . . . commensurate with the efficacy at the facility of the installed design andconstruction technologies, operational measures, and/or restoration measures.” 40 C.F.R.§ 125.98(b)(1)(iii). 35

requirements, climatic conditions, and the waterbody’s biology. The EPA argues also that thepermit process requires facilities to reduce impingement mortality and entrainmentcommensurate with the efficacy of the installed technologies, which it claims ensures that theinstalled technologies will be maintained to ensure their utmost efficacy.19 The difficulty withthe EPA’s arguments is that the Rule does not require facilities to choose technologies thatproduce the greatest reduction possible.Our decision in Riverkeeper I sheds some light on the parties’ arguments. In thatcase, we discussed the differences between the two tracks in the Phase I Rule: Track I set forthprecise velocity and capacity requirements while Track II permitted compliance via technologiesthat would achieve at least 90 percent of the reduction in impingement mortality and entrainmentthat compliance with Track I would yield. See 358 F.3d at 182-83. The petitioners in that casechallenged the Track II provision on the ground that it deviated from the statutory requirementthat the EPA establish a single level of performance applicable to all facilities. Id. at 187. TheEPA argued that Tracks I and II reflected the same standard and that 10 percent is an acceptablemargin of error given that measurements of reduction of impingement mortality and entrainmentare necessarily inexact and depend upon natural fluctuations in animal populations and samplingerrors. Id. at 188. In assessing the parties’ arguments, we stated that “the EPA, consistent withCongress’s intention that there be a national standard governing the discharge of pollutants, must

Page 124: Sotomayor Cases

36

promulgate precise effluent limitations under sections 301 and 306 . . . .” Id. (emphasis added). We went on to note, however, that while pollutant concentration and the velocity and volume ofwater withdrawn can be measured accurately, impingement mortality and entrainment “cannotalways be measured directly and with mathematical precision.” Id. at 189. We concluded thatthe EPA acted reasonably in specifying “how much ambiguity it is willing to tolerate inmeasuring compliance and what it considers a reasonable margin of error in comparing theperformance of different technologies.” Id. In short, we acknowledged that the Track IIperformance requirements, unlike the Track I requirements, could not be measured precisely andthat it was therefore reasonable to consider a margin of error in comparing performance under thetwo standards.This case is not entirely similar to Riverkeeper I because of the rationales thatanimate the EPA’s creation of the performance ranges in Phases I and II. The Phase II Rulegenerally require facilities to reduce impingement mortality and entrainment by the specifiedpercent ranges from the calculation baseline. 40 C.F.R. § 125.94(b). These ranges, as explainedby the EPA, are based on the reductions achievable by using various technologies. See 69 Fed.Reg. at 41,599. The EPA explained that it expressed the performance standards “in the form ofranges rather than a single performance benchmark because of the uncertainty inherent inpredicting the efficacy of any one of these technologies.” Id. at 41,600. It stated further that thelower end of the range is the percent reduction it “expects all facilities could eventually achieve ifthey were to implement and optimize available design and construction technologies andoperational measures on which the performance standards are based” and that the higher end ofthe range “is a percent reduction that available data show many facilities can and have achieved

Page 125: Sotomayor Cases

20 Nothing in Hooker Chems. & Plastics Corp. v. Train, 537 F.2d 620 (2d Cir. 1976),suggests the contrary. While we did announce in that case that performance ranges did notcomport with the effluent limitations established by the CWA, id. at 630, the dischargelimitations at issue there were easily measured, unlike the reductions in impingement andentrainment at issue in both this case and Riverkeeper I, which “cannot always be measureddirectly and with mathematical precision,” Riverkeeper I, 358 F.3d at 189. 37

with the available technologies upon which the performance standards are based.” Id. UnlikeRiverkeeper I, therefore, a margin of error from a relatively precise benchmark that is tolerablegiven measurement difficulties is not at issue here. Instead, the performance standards reflect therange of performance associated with various technologies identified as BTA. That performance,in turn, depends in part on local conditions and natural fluctuations. Id.Record evidence supports the EPA’s conclusion that the percent reduction ofimpingement mortality and entrainment is not completely within the control of a facility andtherefore may not be precisely achieved by a facility. See TDD for the Final § 316(b) Phase IIExisting Facilities Rule 4-3. Reducing these adverse environmental impacts is not as easilymeasured and controlled as are the discharge of pollutants and the capacity and flow rate of waterintake.20 We therefore acknowledge that in many cases it may be difficult, as a practical matter,for the EPA or other permitting authority to predict which plants will be able to achieve theupper, as opposed to the lower, end of the ranges. This uncertainty, however, does not justify arule that permits even those facilities that could achieve the upper end of a range to be deemed incompliance if they reach only the lower end, particularly when the EPA has acknowledged thatmany facilities “can and have” achieved reductions at the high end of the range. 69 Fed. Reg. at41,600. Congress’s use of the superlative “best” in the statute cannot be read to mean that afacility that achieves the lower end of the ranges, but could do better, has complied with the law.

Page 126: Sotomayor Cases

21 Although it may be difficult to know ex ante which plants can achieve the upper end ofthe range, this is at least in part because the technologies preferred by the EPA are somewhat newand untested. In future permitting cycles, permitting authorities will be likely to have a clearernotion of which facilities can feasibly achieve the upper end of the range.Finally, the upper end of the range established by the EPA, should it retain thisprovision on remand, should not be set at a level that many facilities “have achieved” with theinstallation of one or more of the technologies determined to be BTA but, as we noted earlier, atthe best possible level of impingement and entrainment reduction the EPA determines thesetechnologies can achieve. See, e.g., Kennecott, 780 F.2d at 448 (“In setting BAT, EPA uses notthe average plant, but the optimally operating plant, the pilot plant which acts as a beacon toshow what is possible.”). If, at a particular Phase II facility, the adoption of BTA technologiescan achieve a 95% reduction in entrainment and impingement, it is unclear why, under ourjurisprudence and the clear dictates of the CWA, the EPA could establish a performance standardthat has placed the ceiling at the 90% threshold which “many” Phase II facilities “can and have”achieved with the same technology. See Am. Iron & Steel Institute v. EPA, 526 F.2d 1027, 1051(3d Cir. 1975) (“It will be sufficient, for the purpose of setting the level of control underavailable technology, that there be one operating facility which demonstrates that the level can beachieved.”) (internal quotation marks and citations omitted). This would not require every PhaseII facility to meet the upper end of the ranges, but only that each Phase II facility achieve thehighest reduction it can with the installation of technologies determined by the EPA to be BTA. The performance ranges, if retained on remand, should accordingly reflect this understanding. 38

The statutory directive requiring facilities to adopt the best technology cannot be construed topermit a facility to take measures that produce second-best results, see Chevron, 467 U.S. at 843,especially given the technology-forcing imperative behind the Act, Natural Res. Def. Council,822 F.2d at 123. Insofar as the EPA establishes performance standards instead of requiringfacilities to adopt particular technologies, it must require facilities to choose the technology thatpermits them to achieve as much reduction of adverse environmental impacts as istechnologically possible.21 For this reason, the EPA on remand should address these concerns ifin its BTA determination, it retains performance ranges.B. Restoration Measures

The Phase II Rule allows a facility to meet the national performance standards set

Page 127: Sotomayor Cases

39

forth in 40 C.F.R. § 125.94(b) through the use of restoration measures such as restocking fishkilled by a cooling water system and improving the habitat surrounding the intake structure inorder, as the EPA explains, “to provide additional flexibility to facilities in complying with therule by eliminating or significantly offsetting the adverse environmental impact caused by theoperation of a cooling water intake structure.” 69 Fed. Reg. at 41,609; 40 C.F.R. § 125.94(c). The state and environmental petitioners contend that the EPA exceeded its authority by allowingcompliance with section 316(b) through restoration measures because Riverkeeper I held that thestatute’s meaning is plain and that restoration measures cannot substitute for the “best technologyavailable for minimizing adverse environmental impact” in cooling water intake structures. TheEPA contends that its interpretation of the statute to permit restoration measures as a means ofcompliance is entitled to deference because it defined certain statutory terms in the Phase II Rulethat it had not defined in the Phase I Rule. The EPA also relies on the Supreme Court’s holdingin National Cable & Telecommunications Association v. Brand X Internet Services, 545 U.S.967, 125 S. Ct. 2688 (2005), and our statement in Riverkeeper I limiting the decision’s reach tothe Phase I Rule, to argue that our prior interpretation of the statute does not trump the Agency’sconstruction. We agree with the petitioners that Riverkeeper I held that the Agency’s decision topermit restoration measures in the Phase I Rule was not “based on a permissible construction ofthe statute,” Chevron, 467 U.S. at 843, and that this holding applies equally here.We began Riverkeeper I by noting that we were remanding the Phase I restorationprovision because it “contradicts Congress’s clearly expressed intent.” 358 F.3d at 181. Wewent on to state that “however beneficial to the environment, [restoration measures] have nothingto do with the location, the design, the construction, or the capacity of cooling water intake

Page 128: Sotomayor Cases

40

structures, because they are unrelated to the structures themselves.” Id. at 189. “Restorationmeasures correct for the adverse environmental impacts of impingement and entrainment,” wenoted, but “they do not minimize those impacts in the first place.” Id. (emphasis added). For thisreason and others, we concluded in Riverkeeper I that the EPA had exceeded its authority inpromulgating the Phase I Rule by allowing compliance with section 316(b) through restorationmeasures because this Rule was “plainly inconsistent” with the statute’s text and Congress’sintent. Id. at 189, 191. The EPA’s argument that Riverkeeper I is not binding on this issue here has threecomponents. First, the EPA contends that our rejection of the restoration measures at issue inRiverkeeper I did not turn on the statute’s text, but instead was based on various other indicatorsof Congressional intent. The Agency makes much of Riverkeeper I’s brief discussion ofCongress’s rejection of a proposed amendment to section 316(b) that would have explicitlyallowed restoration measures and of the EPA’s support of that amendment because in its opinion,the existing language did not authorize restoration measures. Id. at 190-91. Second, the EPAargues that its interpretation of section 316(b) in the Phase II Rule is entitled to deferencebecause the Rule defined certain statutory terms it had not defined in the previous rulemakingphase and that its reasonable interpretation of these terms is entitled to deference. Specifically,the Agency “defined” three statutory terms in the preamble: it (1) read the phrase “minimiz[e]adverse environmental impact” to let facilities “minimize adverse environmental impact byreducing impingement and entrainment, or to minimize adverse environmental impact bycompensating for those impacts after the fact,” 69 Fed. Reg. at 41,628; (2) interpreted “reflect” toauthorize it to consider the full range of technologies, including restoration measures, that

Page 129: Sotomayor Cases

41

minimize adverse environmental impact; and (3) viewed “restoration measures as part of the‘design’ of a cooling water intake structure,” and “one of several technologies that may beemployed . . . to minimize adverse environmental impact,” 69 Fed. Reg. at 41,637. Finally, theEPA relies on our statement in Riverkeeper I that our ruling on the Phase I Rule was not meant to“predetermine the factors and standard[s] applicable to Phases II and III of the rulemaking,” 358F.3d at 186 n.13, and on the Supreme Court’s holding in Brand X that “[a] court’s prior judicialconstruction of a statute trumps an agency construction otherwise entitled to Chevron deferenceonly if the prior court decision holds that its construction follows from the unambiguous terms ofthe statute and thus leaves no room for agency discretion.” 545 U.S. at —, 125 S. Ct. at 2700. We reject each of the EPA’s contentions. First, our primary conclusion inRiverkeeper I was that restoration measures are “plainly inconsistent” with the statute’s text, 358F.3d at 189, and our statements regarding the legislative history of a proposed amendment, whichwe offered as ancillary, but not dispositive, support for our construction of the statute, in no waydiminish the force of our conclusion that Congress unambiguously expressed its intent in thestatute. See Chevron, 467 U.S. at 842-43 (“If the intent of Congress is clear, that is the end of thematter; for the court, as well as the agency, must give effect to the unambiguously expressedintent of Congress.”). Second, as to the EPA’s claim that its construction of the statute is entitledto deference because it has now interpreted certain statutory terms, our holding in Riverkeeper Iwas and remains clear: restoration measures contradict the unambiguous language ofsection 316(b). The EPA’s promulgation of the Phase II Rule obviously did nothing to alter thetext of section 316(b), and the Agency cannot create ambiguity where none otherwise exists bydefining statutory terms contrary to their plain meaning. Finally, as the foregoing analysis

Page 130: Sotomayor Cases

42

suggests, our decision in Riverkeeper I foreclosed the EPA from interpreting section 316(b) inthe Phase II Rule to permit restoration measures as a means of complying with the statute, and,therefore, nothing in Brand X undermines the precedential value of our prior holding. Ourstatement in Riverkeeper I that we did not “mean to predetermine the factors and standardapplicable to Phases II and III of the rulemaking,” was made in the narrow context of identifying“one reasonable reading” of particular statutory language relating to the standard for new andexisting sources, not the restoration measures. 358 F.3d at 186 n.13. Where we held that thestatutory language is unambiguous, Riverkeeper I is binding. Even assuming arguendo that we did not consider ourselves bound by RiverkeeperI, we are persuaded by its reasoning as applied here. Restoration measures are not part of thelocation, design, construction, or capacity of cooling water intake structures, Riverkeeper I, 358F.3d at 189, and a rule permitting compliance with the statute through restoration measuresallows facilities to avoid adopting any cooling water intake structure technology at all, incontravention of the Act’s clear language as well as its technology-forcing principle. As wenoted in Riverkeeper I, restoration measures substitute after-the-fact compensation for adverseenvironmental impacts that have already occurred for the minimization of those impacts in thefirst instance. Id. The Agency’s attempt to define the word “minimize” to include“compensati[on] . . . after the fact,” 69 Fed. Reg. at 41,628, is simply inconsistent with thatword’s dictionary definition: “to reduce to the smallest possible extent,” Webster’s Third NewInt’l Dictionary 1438 (1986). Accordingly, the EPA impermissibly construed the statute by allowingcompliance with section 316(b) via restoration measures, and we remand that aspect of the Rule.

Page 131: Sotomayor Cases

43

C. Site-Specific Compliance AlternativesAs noted earlier, the Phase II Rule includes two site-specific compliancealternatives or variances from the generally applicable requirements. The cost-cost alternativeauthorizes a site-specific determination that “data specific to [a] facility demonstrate that thecosts of compliance under . . . this section would be significantly greater than the costsconsidered by the Administrator . . . in establishing the applicable performance standards,” 40C.F.R. § 125.94(a)(5)(i), while the cost-benefit alternative authorizes a site-specificdetermination that “data specific to [a] facility demonstrate that the costs of compliance under . . .this section would be significantly greater than the benefits of complying with the applicableperformance standards.” Id. § 125.94(a)(5)(ii). If a facility makes either showing, the permittingauthority “must make a site-specific determination of the best technology available” and impose“site-specific alternative requirements” that are “as close as practicable to the applicableperformance standards.” Id. § 125.94(a)(5)(i), (ii).Petitioners challenge the cost-cost compliance alternative because, inter alia, theyclaim as a threshold matter that the Agency failed to comply with the APA’s notice and commentrequirements by disclosing cost data for specific facilities that would be used in determiningwhether a facility qualifies for the cost-cost compliance alternative only at the time the final Rulewas issued. Petitioners also challenge the cost-benefit compliance alternative on two substantivegrounds. They contend that this alternative (1) impermissibly allows compliance with the statuteto be based on cost-benefit analysis and (2) is analogous to a water-quality standard, which theAct permits only for thermal pollution. CWA § 316(a), 33 U.S.C. § 1326(a). We address each

Page 132: Sotomayor Cases

22 Section 316(b) does not specifically authorize the Agency to allow variances fromgenerally applicable requirements. In Riverkeeper I, we accepted the argument that the EPA canallow variances even in the absence of explicit statutory authority. 358 F.3d at 193. Althoughthe statutes to which section 316(b) refers contained variance provisions, we did not read them torequire that section 316(b) permit (or by its silence prohibit) variances. Instead, we stated that[s]ection 316(b)’s silence with respect to variances does not . . . equal anunambiguous prohibition. In the absence of such a statutory bar, we think,consistent with precedent, that it is reasonable for the EPA to allow variancesfrom regulations promulgated pursuant to section 316(b), for a regulatory systemwhich allows flexibility, and a lessening of firm proscriptions in a proper case, canlend strength to the system as a whole.Id. at 193 (citation and internal quotation marks omitted). Riverkeeper I thus rejected the view that the EPA’s authority to grant a variancefrom the generally applicable requirements of a rule promulgated pursuant to section 316(b)derives from the statutory variance provisions in other sections of the Act. Instead, Riverkeeper Ifound that the variance provision would not be authorized if it left alternative requirements to theAgency’s “unfettered discretion” but would be authorized if it “guide[d] the permitting authorityto consider appropriate factors and allow[ed] relaxation of the Rule’s uniform technologyrequirements only insofar as necessary to account for unusual circumstances not considered bythe Agency during its rulemaking.” Id. at 193-94 (internal quotation marks omitted). 44

of these arguments in turn.221. Cost-Cost Compliance AlternativeAs already noted, a variance may be available to a facility pursuant to 40 C.F.R.§ 125.94(a)(5)(i) if the facility’s compliance costs would be “significantly greater than” the costsconsidered by the Agency “in establishing the applicable performance standards.” This variancerequires a calculation of compliance costs based on the suite of BTA technologies that the EPAhas identified and promulgated in the final Rule. 69 Fed. Reg. at 41,644-46. We remand thisprovision because (1) the EPA did not give interested parties the requisite notice and opportunityto challenge the variance by failing to identify cost data for actual, named facilities, as opposed tomodel facilities, until after the notice and comment period had ended, Sprint Corp., 315 F.3d at

Page 133: Sotomayor Cases

23 This variance requires a comparison between the actual costs of compliance and thosecontemplated by EPA under the Rule. Because the Agency has calculated the costs it believesspecific facilities will incur in adopting the appropriate BTA technologies (as currently defined)and then promulgated these costs in the final Rule, any change in the selection of BTA onremand will necessarily alter these costs. 45

371, and (2) the variance is expressly premised on the validity of the BTA determination,23 whichitself has been remanded for further explanation, see, e.g., Solite Corp. v. U.S. EPA, 952 F.2d473, 494-95 (D.C. Cir. 1991) (remanding rule where the underlying grounds for its promulgationhad been remanded to the EPA for procedural defects); cf. Chenery, 318 U.S. at 87-88 (a rulemay only be upheld on the grounds that the agency proffers). In the Rule’s proposal, the EPA indicated that it had estimated compliance costsfor 539 “model plants” based on factors such as “fuel source, mode of electricity generation,existing intake technologies, waterbody type, geographic location, and intake flow.” 67 Fed.Reg. at 17,144. An accompanying technical development document set forth the Agency’s costcalculation methodology for these model plants and listed the compliance cost estimates for eachof the 539 model plants. The proposal indicated that a facility must “determine which modelplant [it] most closely” resembles in order to identify the costs considered by the Agency inestablishing the national performance standards. See id. The EPA subsequently published in theFederal Register a so-called Notice of Data Availability (“NODA”) in which it explained that ithad changed its methodology for estimating the model plants’ compliance costs. ProposedRegulations to Establish Requirements for Cooling Water Intake Structures at Phase II ExistingFacilities; Notice of Data Availability; Proposed Rule, 68 Fed. Reg. 13,522, 13,527 (Mar. 19,2003). Accompanying documents explained in greater detail the costing methodology and costdata underlying the revised approach. The revised proposal, however, did not depart from the

Page 134: Sotomayor Cases

46

“model plant” approach. The final Rule, by contrast, assigned cost estimates to specific, namedfacilities rather than model facilities. 69 Fed. Reg. at 41,670-82. The Agency explained in thepreamble to the final Rule that the EPA will adjust facility-specific costs pursuant to a multiple-step calculation formula to arrive at a final estimated cost the EPA considers a comparison forpurposes of the cost-cost variance. Id. at 41,644-47. The EPA acknowledges that it did not disclose in the proposal or the NODAspecific facility names in connection with cost data and explains that it failed to do so because itneeded to protect certain confidential business information (“CBI”) and had not developed duringthe proposal stage a means to protect that information while still providing cost data to thepublic. We accept the EPA’s argument that masking the facility names did not prevent interestedparties from commenting on the methodology and general cost data underlying the EPA’sapproach because the NODA explained the costing methodology and because the general costdata, while not identified by the Agency as relating to actual, specific facilities, was madeavailable to interested parties. Nat’l Wildlife Fed., 286 F.3d at 564-65 (holding that the EPAcannot be faulted for lack of notice in not releasing CBI data). We are persuaded, however, thatthe release of information and request for comments on the EPA’s new approach to developingcompliance cost modules via the NODA did not afford adequate notice of the costs associatedwith specific facilities promulgated in the final Rule.We have previously stated that “[n]otice is said not only to improve the quality ofrulemaking through exposure of a proposed rule to comment, but also to provide fairness tointerested parties and to enhance judicial review by the development of a record through thecommentary process.” Nat’l Black Media Coalition v. FCC, 791 F.2d 1016, 1022 (2d Cir. 1986).

Page 135: Sotomayor Cases

24 The D.C. Circuit has phrased the test somewhat differently, stating that “[w]hether the‘logical outgrowth’ test is satisfied depends on whether the affected party ‘should haveanticipated’ the agency’s final course in light of the initial notice.” Covad Comms. Co. v. FCC,450 F.3d 528, 548 (D.C. Cir. 2006) (citation omitted). It has also indicated that “[t]he ‘logicaloutgrowth’ doctrine does not extend to a final rule that finds no roots in the agency’s proposalbecause something is not a logical outgrowth of nothing, nor does it apply where interestedparties would have had to divine the agency’s unspoken thoughts because the final rule wassurprisingly distant from the Agency’s proposal.” Envtl. Integrity Project v. EPA, 425 F.3d 992,996 (D.C. Cir. 2005) (citations, internal quotation marks, and alteration omitted). 47

“While a final rule need not be an exact replica of the rule proposed in the Notice, the final rulemust be a ‘logical outgrowth’ of the rule proposed.” Id. “The test that has been set forth iswhether the agency’s notice would fairly apprise interested persons of the subjects and issues” ofthe rulemaking. Id. (citation and internal quotation marks omitted).24 Agencies accordingly arenot permitted “to use the rulemaking process to pull a surprise switcheroo.” Envtl. IntegrityProject, 425 F.3d at 996.Here, only the final Rule identified facilities by name in estimating compliancecosts. Interested parties therefore could not comment on the basis for particular facilities’ costfigures that the EPA established. This is problematic because the availability of a variance turnson the relationship between the costs estimated in the Rule and those that a specific facilityestablishes in a permit proceeding. The EPA focuses on the notice it gave of its intendedmethodology for calculating the costs the Agency considered, but ignores the overridingimportance of the cost estimates for a particular facility in determining whether a site-specificcost-cost variance is appropriate. Thus, the EPA should have afforded notice and an opportunityto challenge the cost estimates for specific facilities and not simply an opportunity to comment

Page 136: Sotomayor Cases

25 Because we remand on this procedural ground, we do not consider the provision on themerits. We presume that after comment on remand, however, the EPA will reevaluate the meritsof the cost-cost variance provision, particularly in light of any reevaluation of BTA in which theAgency may engage. While we do not decide the issue here, we note our discomfort with the“significantly greater than” standard of 40 C.F.R. § 125.94(a)(5)(i), given the historicalapplicability of a “wholly disproportionate to” standard and the use of the latter standard in thePhase I Rule. We have not found entirely persuasive the EPA’s position that [t]his difference in standards for new and existing facilities is based on (1) thegreater flexibility available to new facilities for selecting the location of theirintakes and installing technologies at lower costs relative to the costs associatedwith retrofitting existing facilities and (2) the desire to avoid economicallyimpracticable impacts on energy prices, production costs, and energy productionthat could occur if large numbers of Phase II existing facilities incurred costs thatwere more than “significantly greater” than but not “wholly out of proportion” tothe costs in EPA’s record.68 Fed. Reg. at 13,541. The EPA would presumably consider each of these two factors inestablishing BTA for existing facilities, and need not further consider them in determiningwhether a particular facility warrants a variance from the generally applicable BTA. Becausecost is not supposed to be a paramount consideration in determining BTA, see Riverkeeper I, 358F.3d at 185, the “significantly greater than” standard poses substantial concerns.26 In contrast to the cost-cost variance, the cost-benefit compliance alternative will notchange on remand for the reconsideration of BTA. We thus reach the merits here. 48

on the EPA’s methodology and general cost data.25 We remand this variance for inadequatenotice and because of our remand of the BTA determination.2. Cost-Benefit Compliance Alternative26If a facility requests that it be permitted to demonstrate compliance with the PhaseII Rule through the site-specific cost-benefit provision of 40 C.F.R. § 125.94(a)(5)(ii), the facilitymust submit with its application a Comprehensive Cost Evaluation Study, Benefits ValuationStudy, and Site Specific Technology Plan. 40 C.F.R. § 125.95(b)(6). As part of the BenefitsValuation Study, the facility must indicate the monetized value of commercial, recreational, andecological benefits of compliance with the generally applicable national performance standards

Page 137: Sotomayor Cases

49

as well as a qualitative assessment of any so-called “non-use” benefits that cannot be monetized. 40 C.F.R. § 125.95(b)(6)(ii)(A), (E). Ultimately, the facility must demonstrate that itscompliance costs are “significantly greater than” the benefits of compliance. The petitionerscontend that this alternative impermissibly focuses on cost-benefit considerations, contrary toCongress’s directive, and is analogous to the kind of water-quality-based standard we found to beinconsistent with the statute in Riverkeeper I. 358 F.3d at 190. For both reasons, we arepersuaded that the EPA exceeded its authority in permitting site-specific cost-benefit variances. In light of this conclusion, we do not reach the industry petitioners’ claim that the provisionimpermissibly requires consideration of qualitative non-use benefits in the cost-benefit analysis.As we discussed previously in analyzing the EPA’s determination of BTA, cost-benefit analysis is not consistent with the requirement of § 316(b) that cooling water intakestructures “reflect the best technology available for minimizing adverse environmental impact.” Indeed, the statutory language requires that the EPA’s selection of BTA be driven by technology,not cost. The Agency is therefore precluded from undertaking such cost-benefit analysis becausethe BTA standard represents Congress’s conclusion that the costs imposed on industry inadopting the best cooling water intake structure technology available (i.e., the best-performingtechnology that can be reasonably borne by the industry) are worth the benefits in reducingadverse environmental impacts. Cf. Am. Textile Mfrs. Inst., 452 U.S. at 509 (noting that whereCongress has defined the basic relationship between costs and benefits, a regulatory standard thatstrikes a different balance is inconsistent with the statute). Just as the Agency cannot determineBTA on the basis of cost-benefit analysis, it cannot authorize site-specific determinations of BTAbased on cost-benefit analysis.

Page 138: Sotomayor Cases

27 As the petitioners note, section 316(a) permits consideration of the quality of thereceiving water for purposes of granting variances with respect to the rules concerning thermalpollution. Section 316(a) provides that the Administrator may issue a variance to the rulesgoverning “thermal discharges” to a facility that establishes that those rules “require effluentlimitations more stringent than necessary to assure the pro[t]ection and propagation of abalanced, indigenous population of shellfish, fish, and wildlife in and on the body of water intowhich the discharge is to be made.” CWA § 316(a), 33 U.S.C. § 1326(a). As the petitioners alsonote, section 316(b) does not similarly permit consideration of the quality of the receiving water,and this distinction is significant. See Russello, 464 U.S. at 23. Moreover, as we discussed inRiverkeeper I, the heat-pollution provision of section 316(a) is a “notable exception” to theCWA, which “otherwise relies on limitations on what a source can put into the water, not theultimate effect of that discharge.” 358 F.3d at 190. Nothing in the statute, therefore, supports theview that the EPA can consider the quality of the receiving water in granting variances from50

The cost-benefit variance also impermissibly authorizes the EPA to consider thedegraded quality of waterways in selecting a site-specific BTA. We stated in Riverkeeper I thatin enacting the CWA, Congress rejected regulation by reference to water quality standards. 358F.3d at 189-90. Before 1972, Congress “regulated point sources based on their effect on thesurrounding water and allowed sources to discharge pollutants provided the discharge did notcause water quality to dip below an acceptable level.” Id. at 189. Congress changed its approachin 1972, in part because a plaintiff attempting to prove a violation of the law faced a nearlyimpossible burden of showing that a particular polluter had caused the water quality to dip belowthe regulatory standards. Id. at 189-90. The Act now regulates discharges from point sourcesrather than water quality. We thus concluded in Riverkeeper I that water-quality standardscannot be considered under section 316(b). Id. at 190. Of course, “water quality” in the contextof the Act is generally understood to refer to pollutant concentration. As we noted inRiverkeeper I, however, for purposes of section 316(b), which regulates water intake rather thanthe discharge of pollutants, water quality is “measured by wildlife levels.” Id. at 189. Thisanalysis in Riverkeeper I is, thus, equally applicable here.27

Page 139: Sotomayor Cases

generally applicable rules promulgated pursuant to section 316(b).51

The challenged provision of the Phase II Rule apparently would permit a facilityto argue that, based on water quality (i.e., the level of aquatic wildlife in a particular body ofwater), the cost of complying with the national performance standards is not justified. TheAgency explained in the preamble to the Rule that “in a waterbody that is already degraded, veryfew aquatic organisms may be subject to impingement or entrainment, and the costs ofretrofitting an existing cooling water intake structure may be significantly greater than thebenefits of doing so.” 69 Fed. Reg. at 41,604. This kind of water-quality-based regulation is notauthorized by the CWA because it would exempt facilities from meeting the mandatedperformance standards simply because wildlife levels in the waterbody were already low, and aswe held in Riverkeeper I, the CWA does permit the EPA to consider water quality in makingBTA determinations. Finally, we note that to the extent that facilities on highly degradedwaterbodies with relatively low wildlife levels face high compliance costs to achieve the nationalperformance standards, those facilities may qualify for the cost-cost variance if such variance isretained on remand. Because the EPA exceeded its authority under section 316(b) by permitting (1)cost-benefit analysis and (2) assessment of the quality of the receiving water (i.e., the receivingwater’s wildlife levels) in determining whether a variance is warranted, we do not need to deferto the Agency’s construction of the statute. We therefore remand this aspect of the Rule.

D. TIOP ProvisionThe Phase II Rule’s TIOP provision, 40 C.F.R. § 125.94(d), permits a facility to

Page 140: Sotomayor Cases

52

comply with the national performance standards determined on the basis of whether the facilityhas “complied with the construction, operational, maintenance, monitoring, and adaptivemanagement requirements of a Technology Installation and Operation Plan.” Id. § 125.94(d)(1). The petitioners contend that this provision impermissibly allows a facility’s compliance to bedetermined not by reference to the performance standards themselves, but by evaluating whethera facility has complied with a plan to achieve the performance standards. In other words, theyargue that the TIOP provision essentially allows for an unauthorized margin of error. Thepetitioners also argue that the EPA denied the public an opportunity to comment on theprovision. Like the cost-cost compliance alternative, remand is appropriate here on two grounds: (1) the EPA did not give adequate notice regarding the provisions in section 125.94(d)(2); and(2) the record justification for the TIOP provision depends on the EPA’s selection of a suite oftechnologies as BTA, a selection which has been remanded for further explanation. Given this,we remand the TIOP provision without reaching the merits here.The Rule provides that during the first permit term, a facility may request that itscompliance be determined based on whether it has complied with its TIOP, which must bedesigned to meet the performance standards, 40 C.F.R. § 125.94(d)(1), and submitted with apermit application, id. § 125.95(b)(4)(ii). During subsequent permit terms, if a facility hascomplied with its TIOP but is not meeting the performance standards, the facility may requestthat its compliance with the standards during the following term be based on whether it remainsin compliance with its TIOP, revised in accordance with the facility’s adaptive management plan. 40 C.F.R. § 125.94(d)(2). The EPA explained in the Rule’s preamble that it is difficult todetermine reductions in impingement mortality and entrainment relative to what would have

Page 141: Sotomayor Cases

53

occurred in the absence of control technologies given natural variability and the vagaries ofsampling methods. 69 Fed. Reg. at 41,613. The EPA explained further that it established theTIOP compliance options to account for these variabilities on the ability of a technology to meetthe performance standards consistently over time. Id. at 41,613-14. As previously noted, a “final rule must be a ‘logical outgrowth’ of the ruleproposed.” Nat’l Black Media Coalition, 791 F.2d at 1022. The final rule must have roots in theproposal, Envtl. Integrity Project, 425 F.3d at 996, which must “fairly apprise interested personsof the subjects and issues” involved in the rulemaking, Nat’l Black Media Coalition, 791 F.2d at1022. An agency cannot “pull a surprise switcheroo” on interested parties between a proposaland the issuance of a final rule. See Envtl. Integrity Project, 425 F.3d at 996.Although the Rule’s proposal notified interested parties that the Agency wasconsidering a provision that would give facilities time to achieve the performance standards afterimplementing new technologies, the EPA gave inadequate notice of the potentially indefinitescope of this provision. Specifically, the EPA failed to provide notice of the Rule codified at 40C.F.R. § 125.94(d)(2), which permits a facility to be deemed in compliance with the Phase IIRule in subsequent permit terms if it continues to adhere to its TIOP. In the Rule’s proposal, the EPA stated only that it was considering “the need forregulatory language that would allow facilities time to come into compliance [with theperformance standards] if they choose to install technologies to meet the performance standards.” 68 Fed. Reg. at 13,586. The proposal indicated further that the EPA was “evaluating andconsidering allowing six months, one year, two years, or five years (one permit term) for afacility to come into compliance after issuance of its permit.” Id. The TIOP provision in the

Page 142: Sotomayor Cases

54

final Rule, by contrast, does not simply allow facilities additional time, up to one permit term, tocome into compliance with the performance standards. Instead, it appears to permit a facility tosatisfy the Rule’s requirements in subsequent permit terms, for an indefinite period, without everdemonstrating compliance with the performance standards, so long as the facility has adhered toits TIOP. 40 C.F.R. § 125.94(d)(2) (stating that “[d]uring subsequent permit terms” a facility“may request that compliance . . . be determined based on whether [it] remain[s] in compliancewith” its TIOP). This aspect of the TIOP provision appears then not to be a “logical outgrowth”of the proposal, see Nat’l Black Media Coalition, 791 F.2d at 1022, because interested partieswould not have divined from the proposal that facilities could be given an indefinite period tocome into compliance with the national performance standards.We thus remand the rule for failure to provide notice and comment and because therecord justification for the TIOP provision depends on the EPA’s selection of BTA, which hasbeen remanded.E. Definitions of “New Facility” and “Existing Facility”

The environmental petitioners challenge the reclassification in the Phase II Rulepreamble of certain new constructions as “existing facilities,” thereby rendering them subject tothe Phase II Rule rather than the more stringent Phase I requirements that apply to new facilities. We agree with the petitioners that the Agency interpretively modified a definition appearing inthe Phase I Rule via statements in the preamble to the Phase II Rule without providing interestedparties notice and an opportunity for comment.“An agency’s interpretation of its own . . . regulation must be given controlling

Page 143: Sotomayor Cases

55

weight unless it is plainly erroneous or inconsistent with the regulation.” Fowlkes v. Adamec,432 F.3d 90, 97 (2d Cir. 2005) (citations and internal quotation marks omitted). Although wetypically owe considerable deference to an agency’s construction of its own regulation, Udall v.Tallman, 380 U.S. 1, 16 (1965), “[u]nder settled principles of statutory and rule construction, acourt may defer to administrative interpretations of a statute or regulation only when the plainmeaning of the rule itself is doubtful or ambiguous,” Pfizer, Inc. v. Heckler, 735 F.2d 1502, 1509(D.C. Cir. 1984) (emphasis in original). “Deference to agency interpretations is not in order ifthe rule’s meaning is clear on its face.” Id. Implicit in the rule that an agency cannot interpret aregulation contrary to its unambiguous meaning is the requirement that “an agency must adhereto its own rules and regulations.” Reuters Ltd. v. FCC, 781 F.2d 946, 950 (D.C. Cir. 1986). Anagency may modify a regulation that has already been promulgated, therefore, only through theprocess of notice and comment rulemaking. See Alaska Prof’l Hunters Ass’n, Inc. v. FAA, 177F.3d 1030, 1034 (D.C. Cir. 1999); see also 5 U.S.C. § 551(5) (defining “rule making,” which isgoverned by the notice and comment requirements of 5 U.S.C. § 553, as the “agency process forformulating, amending, or repealing a rule”); Shalala v. Guernsey Memorial Hosp., 514 U.S. 87,100 (1995) (noting in dicta that APA rulemaking is required where an agency interpretation“adopt[s] a new position inconsistent with . . . existing regulations”). The Phase I Rule defined “new facility” as any structure whose constructioncommenced after January 17, 2002 that meets both the definition of “new source” in 40 C.F.R.§ 122.29 (discussed below) and one of two other requirements: the structure must use either (1)“a newly constructed cooling water intake structure” or (2) “an existing cooling water intakestructure whose design capacity is increased to accommodate the intake of additional cooling

Page 144: Sotomayor Cases

28 The Phase I Rule defined facilities meeting either of the first two tests ofsection 122.29(b) as “greenfield” facilities and facilities meeting the third test as “stand-alone”facilities. 40 C.F.R. § 125.83 (“A greenfield facility is a facility that is constructed at a site atwhich no other source is located, or that totally replaces the process or production equipment atan existing facility. A stand-alone facility . . . is constructed on property where an existingfacility is located and whose processes are substantially independent of the existing facility at thesame site.”). 56

water.” See 40 C.F.R. § 125.83. A “new source” under section 122.29 is a facility that (1) “isconstructed at a site at which no other source is located,” (2) “totally replaces the process orproduction equipment that causes the discharge of pollutants at an existing source,” or (3)undertakes “processes . . . substantially independent of an existing source at the same site.”28 40C.F.R. § 122.29(b) (emphasis added). Section 122.29(b) provides further that, in determiningwhether a facility is “substantially independent” of an existing source, the director shouldconsider “the extent to which the new facility is integrated with the existing plant; and the extentto which the new facility is engaged in the same general type of activity as the existing source.” 40 C.F.R. § 122.29(b)(iii). In determining whether a new construction qualifies as a “new facility” forpurposes of the Phase I Rule, therefore, the permitting authority must perform a two-part, but notnecessarily sequential, analysis. It must determine whether the construction uses a new coolingwater intake structure or an existing structure whose capacity has been increased. The permittingauthority must also determine whether the new construction qualifies as a “new source.” Failureto meet either part of this analysis precludes the new construction from qualifying as a “newfacility,” and thus from falling under the Phase I Rule’s purview.The Phase I Rule stated that “new facilities” meeting the foregoing requirements

Page 145: Sotomayor Cases

29 A facility would clearly find it advantageous to classify a new construction on itsgrounds as one added for purposes of the same general industrial operation rather than as a stand-alone facility. A stand-alone facility is treated as an existing facility only if it uses an existingintake structure whose capacity is not increased; a “new unit,” by contrast, is treated as part of anexisting facility even if it requires the construction of a new intake structure or an increase in anexisting intake structure’s capacity. 57

include a “stand-alone” facility, which the Phase I Rule defined as “a new, separate facility that isconstructed on property where an existing facility is located and whose processes aresubstantially independent of the existing facility at the same site.” Id. § 125.83. This is becausesuch a facility, by definition, essentially qualifies as a “new source.” Compare id. with § 122.29(defining “new source” as including a facility that undertakes “processes . . . substantiallyindependent of an existing source at the same site”). The Phase I Rule clarified, however, that“new facility” does not include “new units that are added to a facility for purposes of the samegeneral industrial operation (for example, a new peaking unit at an electrical generating station).” Id. § 125.83. This is presumably because such units do not qualify as “new sources” in that theyare not substantially independent of existing sources. See id. § 122.29(b)(1)(iii) (setting forth thefactors to be considered in determining substantial independence, including “the extent to whichthe new facility is integrated with the existing plant; and the extent to which the new facility isengaged in the same general type of activity as the existing source”). The Phase I Rule thusappears to have left regulation over the following to a subsequent rulemaking phase: (1) newstand-alone facilities that use existing intake structures whose design capacity is not increasedand (2) new units that are added to a facility for purposes of the same general industrial operationeven if they require either an increase in the intake structure design capacity or the constructionof a new cooling water intake structure altogether. Id.29

Page 146: Sotomayor Cases

30 We find no merit to the EPA’s argument that the state and environmental petitionershere are ostensibly challenging the Phase II Rule’s definition of “existing facility” but areactually seeking review of the meaning of “new facility” under the Phase I Rule – for which theirclaim would be time-barred. Though the EPA claims that “[n]othing in the Phase II Rule alteredor amended the definition of ‘new facility,’” it is clear from the discussion above that thepreamble to the Phase II Rule eliminates without notice or comment the analysis the Phase I Rulehad required. 58

The Phase II Rule defines “existing facility” as any facility whose constructioncommenced on or before January 17, 2002, “and any modification of, or any addition of a unit atsuch a facility that does not meet the definition of a new facility at § 125.83.” Id. § 125.93. Thus, from this definition, it appears that new stand-alone facilities that use existing, unmodifiedintake structures and new units added to a facility for purposes of the same industrial operation,regardless of their impact on the facility’s cooling water intake structure, (i.e., the two kinds ofnew constructions left unregulated by the Phase I Rule) are considered “existing facilities” andgoverned by the Phase II Rule.The parties’ dispute concerns statements in the preamble to the Phase II Rule thatpurportedly narrow, by way of interpretation, the Phase I Rule’s definition of “new facility”30

without the required procedures of notice and comment. In the preamble to the Phase II Rule, theEPA states that “the Phase I [R]ule treated almost all changes to existing facilities for purposes ofthe same industrial operation as existing facilities.” 69 Fed. Reg. at 41,579. The preamble thenappears to distinguish stand-alone facilities from new units that are part of the same industrialoperation, thereby defining the latter as existing facilities without reference to the definition of“new source” or the “substantial independence” test of 40 C.F.R. § 122.29. Id. at 41,579 n.2a. The preamble states that the “substantial independence” test does not apply where there is anaddition to an existing facility for purposes of the same industrial operation, such as the “addition

Page 147: Sotomayor Cases

59

of new generating units at the same site” as an existing facility, id. at 41,579, because suchadditions “are categorically treated as ‘existing facilities’ regardless of any other considerationsunless they completely replace an existing facility and its cooling water design intake capacity isincreased,” id. at 41,579 n.2a. These comments are contrary to the plain meaning of the relevantportion of the Phase I Rule.The Phase I Rule unambiguously stated that “new facility” means any structurethat is a “new source,” as defined by 40 C.F.R. § 122.29, subject to certain other requirements. Under this provision, a source is considered “new” if, inter alia, “[i]ts processes are substantiallyindependent of an existing source at the same site.” 40 C.F.R. § 122.29(b)(1)(iii). A permittingauthority could not classify a source constructed at the site of an existing source as new orexisting for purposes of the Phase I Rule, therefore, without reference to the “substantialindependence” test. It is plain, then, that the Phase I Rule distinguished between “stand-alone”facilities and “new units,” where the new construction is not built at an empty site and does nottotally replace an existing source, by reference to the definition of a “new source.” A stand-alonefacility is “substantially independent” of an existing facility, and therefore a new source; a newunit that is part of the same industrial operation as an existing facility is not substantiallyindependent of an existing facility, and therefore not a new source. It is impossible to determinewhich classification applies to a particular construction under the Phase I Rule without referringto the definition of “new source,” i.e., whether it satisfies the “substantial independence” test. Put differently, the touchstone of the definition of “new facility” in the Phase I Rule is whether asource is a “new source.” The Phase I Rule’s plain terms thus indicate that a unit that is“substantially independent” of an existing facility is not “part of the same general industrial

Page 148: Sotomayor Cases

60

operation” as the existing facility. Any elimination of the “substantial independence” inquiry,therefore, strikes at the heart of the Phase I Rule and its classification of what facilities aresubject to its requirements. The EPA claims that the Phase II Rule has in no way eliminated the “substantialindependence” test and that the Rule’s preamble merely makes clear that the fifth sentence insection 125.83 exempts “new units” from regulation under the Phase I Rule. This argument failsbecause the Phase I Rule provides no way to distinguish between stand-alone facilities and newunits where the construction is built on a site where a source is already located and does nottotally replace the existing source except by reference to the “substantial independence” test, i.e.,without assessing the factors set forth at 40 C.F.R. § 122.29(b)(1)(iii) in order to determinewhether the source is new or existing. Just as “stand-alone facility” has no intrinsic meaning,neither does “new unit.” The Phase I Rule defines each by reference to the “substantialindependence” test of section 122.29(b). Thus, while an existing facility can be repowered withnew generating units and remain an “existing facility” for regulatory purposes undersection 316(b), the determination can only be made by reference to whether a particular newgenerating unit is a stand-alone facility or a new unit that is part of the same general industrialoperation as an existing facility. In fact, a permitting authority must first determine whether asource is “new” within the meaning of 40 C.F.R. § 122.29(b) before it can conclude that thesource is a stand-alone facility or a new unit added to an existing facility for purposes of the samegeneral industrial operation. Because the Phase I Rule was not ambiguous, we do not owe deference to theAgency’s interpretation of the Phase I Rule in the preamble to the Phase II Rule. See Fowlkes,

Page 149: Sotomayor Cases

61

432 F.3d at 97; Pfizer, 735 F.2d at 1509. By permitting the Agency to determine that a newconstruction is not subject to the Phase II Rule without any definitional guidance and incontravention of the Phase I Rule, the EPA has expanded the scope of what may be classified asa “new unit” while narrowing the Phase I definition of “stand-alone” facility. Moreover, byincluding a potentially expansive definition of “new unit” in the preamble to the Phase II Rule, the EPA has interpretively modified the definitions that appeared in the Phase I Rule withoutproviding interested parties an opportunity for notice and comment. Accordingly, we direct the EPA on remand to adhere to the definitions set forth inthe Phase I Rule, see Reuters, 781 F.2d at 950-52, or to amend those definitions following noticeand comment, see Alaska Prof’l Hunters, 177 F.3d at 1034.

IV. The Industry PetitionersA. Applicability of Section 316(b) to Existing Facilities

Entergy argues that the EPA lacks authority to apply CWA section 316(b) toexisting, as opposed to new, facilities. We disagree and conclude that, at the very least, the EPApermissibly interpreted the statute to cover existing facilities and that its interpretation istherefore entitled to deference under Chevron.Entergy’s argument turns primarily on the statutory language that the besttechnology available be reflected in the “location, design, construction, and capacity of coolingwater intake structures” – a collection of words Entergy contends indicates Congress’s intent to

Page 150: Sotomayor Cases

31 Entergy cites a number of federal statutory and regulatory provisions using similarwords and argues that each provision applies only in the pre-construction context. Theenvironmental petitioners contend that the provisions upon which Entergy relies explicitly limittheir application to new facilities and that, as a result, the words “location, design, constructionand capacity,” standing alone, do not apply solely to new facilities; otherwise, the environmentalpetitioners’ argument goes, there would be no need to limit expressly the provisions’applicability to new facilities. For example, 49 U.S.C. § 60103, which governs safety standardsfor liquefied natural gas pipelines, by its express terms does not apply to certain existingfacilities. 49 U.S.C. § 60103(c) (“Except as provided in paragraph (2) of this subsection, adesign, location, installation, construction, initial inspection, or initial testing standard prescribedunder this chapter after March 1, 1978, does not apply to an existing liquefied natural gaspipeline facility . . . .”). These are additional reasons for the holding discussed in this opinion. 62

regulate only new facilities.31 Entergy argues further that the EPA has authority to approvecooling water intake structures only before construction and cannot regulate these structuresthrough the NPDES permits issued pursuant to CWA section 402(a)(1), 33 U.S.C. § 1342(a)(1),which allows the Administrator to “issue a permit for the discharge of any pollutant,” but not, inEntergy’s view, to cover existing intake structures.The EPA emphasizes that section 316(b) cross-references section 301, whichapplies to existing facilities, and that the Agency’s historical practice of applying section 316(b)to existing facilities effectuates Congress’s objectives in enacting the CWA. The EPA contendsthat Congress’s link between effluent limitations established pursuant to section 301 and BTAunder section 316(b) indicates an intent to regulate cooling water intake structures at existingfacilities. The EPA thus argues that the intake-structure standard is to be applied whenever anintake structure is present at a point source of pollutant discharge, whether that point source isnew or existing. As to Entergy’s argument concerning the ability to assess intake structuresduring NPDES permitting proceedings, the EPA argues that its decision to implement the PhaseII requirements through NPDES permits has a strong textual basis. It notes that section 402

Page 151: Sotomayor Cases

32 We are not persuaded by Entergy’s selective definitions of the terms “location, design,construction, and capacity.” While we agree that words in a statute are to be accorded “theirordinary, contemporary, common meaning,” Harris v. Sullivan, 968 F.2d 263, 265 (2d Cir.1992), it is clear that the EPA’s interpretation of the statute does not deviate from thisrequirement. See, e.g., Black’s Law Dictionary 958 (8th ed. 1999) (“[L]ocation” is “[t]hespecific place or position of a person or thing.”); id. at 478 (“[D]esign” means “a plan or scheme”or “[t]he pattern or configuration of elements in something.”); id. at 332 (“[C]onstruction” is“[t]he act of building,” but also “the thing so built.”); Webster’s Third New Int’l Dictionary 330(1986) (“[C]apacity” means “the power or ability to hold, receive or accommodate” and “acontaining space: a measure of content for gas, liquid, or solid.”). Nothing in the abovedefinitions suggests the EPA could not have reasonably interpreted such words to includeexisting facilities. 63

provides for the issuance of “a permit for the discharge of any pollutant” so long as the dischargemeets “all applicable requirements under sections 1311 . . . [and] 1316,” sections 301 and 306 ofthe CWA respectively. CWA § 402(a)(1), 33 U.S.C. § 1342(a)(1). EPA contends that the PhaseII requirements are “requirements under sections 1311 . . . [and] 1316” and therefore that coolingwater intake structures may be regulated via permits issued pursuant to section 402. The EPA has the better of both arguments. As to whether section 316(b) appliesto existing facilities, we find Entergy’s textual argument, while superficially appealing,ultimately to be unavailing. Nothing in section 316(b) indicates that because it applies to the“location, design, construction, and capacity” of a facility’s cooling water intake structure, thesection is therefore limited to new facilities and does not require existing facilities either tomodify existing intake structures or to construct new intake structures in order to come intocompliance with the EPA’s Rule.32 In fact, given the cross-references in section 316(b) toprovisions governing both new and existing facilities, the EPA’s reading is far more reasonablethan Entergy’s. See E.I. du Pont de Nemours & Co. v. Train, 430 U.S. 112, 121 (1977) (“Section301(b) [to which section 316(b) expressly refers] defines the effluent limitations that shall be

Page 152: Sotomayor Cases

64

achieved by existing point sources . . . .”); id. at 136 (holding that “301 does authorize the[Agency] to promulgate effluent limitations for classes and categories of existing point sources”). The cross-reference to section 301 is particularly significant given that Congress, having madethis explicit cross-reference, did not then limit section 316(b)’s application to new facilities –which would have been a simple task to do. At the very least, the EPA’s view that section 316(b)applies to existing facilities is a reasonable interpretation of the statute, and we therefore accordit deference. See Chevron, 467 U.S. at 842-43.Entergy’s argument concerning the permitting process presents a closer question,but it is ultimately not persuasive and does not undermine our conclusion that section 316(b), onits face, applies to existing facilities. The textual basis for the EPA to regulate cooling waterintake structures during the periodic permitting process applicable to the discharge of pollutantsis not immediately apparent. Section 402 conditions the issuance of a permit on the circumstancethat a “discharge will meet . . . all applicable requirements under sections 1311 . . [and] 1316.” CWA § 402(a)(1), 33 U.S.C. § 1342(a)(1) (emphasis added). While the Phase II requirementsare “requirements” under sections 301 and 306, they do not apply to the discharge of pollutants,and section 402 says nothing about conditioning a permit on compliance with other requirementsof sections 301 and 306, i.e., requirements not relating to the discharge of pollutants. Despitethis textual hiccup, the EPA’s decision to use the NPDES process to enforce section 316(b) is notunreasonable. Insofar as the provision applies to existing facilities – and the cross-reference insection 316(b) to section 301 provides a clear textual basis for that conclusion – the EPA couldenforce it only through some permit process following the issuance of an initial constructionpermit. Moreover, the structure of section 316(b) supports the view that its requirements are to

Page 153: Sotomayor Cases

65

be enforced through the same process used to enforce the effluent limitations of sections 301 and306. As noted, the statute requires that any standard established by the EPA to govern thedischarge of pollutants from existing facilities must also regulate cooling water intake structures. CWA § 316(b), 33 U.S.C. § 1326(b) (providing that “[a]ny standard established pursuant tosection 1311 . . . or section 1316” shall also regulate the cooling water intake structures of pointsources). In light of this language, it is at least reasonable to conclude that Congress intended therequirements of section 316(b) to be part and parcel of any regulation of, and therefore anypermit issuance relating to, the discharge of pollutants. See Riverkeeper I, 358 F.3d at 185-86. Itis a fair conclusion that section 402 implicitly requires permitting authorities to ensurecompliance with section 316(b) as a permit condition. See U.S. Steel Corp. v. Train, 556 F.2d822, 850 (7th Cir. 1977) (“[Section] 402(a)(1) implicitly requires the Administrator to insurecompliance with § 316(b) as one of the permit conditions.”), overruled on other grounds by Cityof West Chicago, Ill. v. U.S. Nuclear Regulatory Comm’n, 701 F.2d 632, 644 (7th Cir. 1983)(abandoning the view that 5 U.S.C. § 558(c) independently provides that formal adjudicatoryhearings must be held when requested by a license applicant under CWA § 402). Section 402thus does not undermine the deference to which the Agency’s interpretation of section 316(b) isentitled under Chevron. Because section 316(b) plainly applies to existing facilities and Congress intendedthe requirements of section 316(b) to apply in tandem with the effluent limitations establishedpursuant to sections 301 and 306, we conclude that the EPA may regulate cooling water intakestructures via the NPDES permit process. Otherwise, Congress’s intent to regulate the intake

Page 154: Sotomayor Cases

33 Entergy also contends that section 316(b) cannot be enforced via a citizen suit pursuantto section 505(a)(1), 33 U.S.C. § 1365(a)(1), because that provision covers only violations of “aneffluent standard or limitation” or “an order issued by the Administrator or a State with respect tosuch a standard or limitation,” and that the statute therefore applies only in the pre-constructioncontext. Because the case before us is not a citizen suit, we are not directly presented with thequestion of whether a citizen can sue to enforce section 316(b) and do not decide the question. We are not, however, persuaded by Entergy’s argument. The citizen suit provision states that acitizen may commence a suit against any person who is alleged to be in violation of “an effluentstandard or limitation under this chapter.” CWA § 505(a)(1), 33 U.S.C. § 1365(a)(1). Viewed inlight of this language alone, it is difficult to characterize a violation of regulations promulgatedpursuant to section 316(b) as a violation of an “effluent standard or limitation.” The statute,however, defines “effluent standard or limitation” to include “an effluent limitation or otherlimitation under section 1311” and a “standard of performance under section 1316.” CWA§ 505(f), 33 U.S.C. § 1365(f). Arguably, therefore, a limitation established pursuant tosection 316(b) is an “other limitation under section 1311” and a “standard of performance undersection 1316,” given the cross-reference to sections 301 and 306 in section 316(b), and thus fallswithin the scope of the citizen-suit provision. While we do not decide the question here, we donot read the citizen suit provision to undermine the deference we accord the EPA’s reasonableinterpretation pursuant to Chevron that section 316(b) applies to existing facilities.66

structures of existing facilities could not be effectuated.33 Accordingly, we reject this aspect ofEntergy’s challenge.B. Definition of “Adverse Environmental Impact”

In the Phase II Rule, as in the Phase I Rule, the EPA has interpreted the statutorydirective of section 316(b) to minimize “adverse environmental impact” (“AEI”) to require thereduction of “the number of aquatic organisms lost as a result of water withdrawals associated”with cooling water intake structures. 69 Fed. Reg. at 41,586. This interpretation reflects the factthat section 316(b) is a somewhat unusual provision of the CWA in that it governs theenvironmental effects of large scale withdrawals from waters of the United States rather than therelease of pollutants into receiving water. As did the industry petitioners in Riverkeeper I, PSEGargues that the EPA arbitrarily defined AEI to include any loss of or harm to aquatic organisms

Page 155: Sotomayor Cases

34 PSEG relies on the Draft Guidance for Evaluating the Adverse Impact of CoolingWater Intake Structures on the Aquatic Environment, prepared by the EPA in 1977. While theEPA indicated in this document that “[a]dverse environmental impacts occur whenever there willbe entrainment or impingement damage as a result of the operation of a specific cooling waterintake structure,” PSEG focuses on the Agency’s statement that the “critical question” is “themagnitude of any adverse impact.”35 The environmental petitioners contend that PSEG has missed the mark here entirely. They argue that when organisms die of natural causes they remain available as food for toppredators, but that when organisms are entrained, and potentially disintegrated, they areconsumed by lower organisms. They argue also that entrained organisms are not available toconsume organisms lower on the food chain. The environmental petitioners thus contend thatone of the primary adverse environmental impacts of impingement mortality and entrainment istheir effect on the ecosystem as a whole by disrupting the food chain. The EPA discussed theseand other environmental impacts attributable to impingement mortality and entrainment in thepreamble to the Phase II Rule, concluding that “[d]ecreased numbers of aquatic organisms candisrupt aquatic food webs and alter species composition and overall levels of biodiversity.” 69Fed. Reg. at 41,586. 67

due to impingement mortality and entrainment rather than only more severe “population-leveleffects.” It contends that the EPA has historically focused on the “population dynamics” ofaquatic organisms and the fact that many of these organisms are “density dependent,” i.e., theyproduce large numbers of offspring, only a few of which survive to adulthood.34 On the theorythat the “vast majority” of entrained organisms would have died of natural causes in any event,PSEG argues that reduction of impingement mortality and entrainment will have only a marginalpositive environmental impact.35 PSEG contends that because the Agency has changed coursefrom its longstanding interpretation of AEI without giving a clear and reasoned justification forits decision, the EPA’s Rule is entitled to less deference than it otherwise would receive. PSEGalso argues that the Phase II Rule’s exclusion from the entrainment-reduction requirement offacilities withdrawing small amounts of water, facilities withdrawing water from lakes other thanthe Great Lakes, and facilities that have a small capacity utilization rate undercuts the EPA’s

Page 156: Sotomayor Cases

68

determination that impingement mortality and entrainment are per se adverse environmentalimpacts. We are not persuaded by PSEG’s arguments.We agree with the EPA that the Phase II Rule is based on substantially the samerecord evidence of impingement mortality and entrainment relied upon in promulgating the PhaseI Rule and that we rejected substantially the same arguments advanced here by the industrypetitioners in Riverkeeper I, 358 F.3d at 197 (“The EPA considered all of the factors that UWAGnow raises, and we are inclined to defer to the EPA’s judgment of how best to define andminimize ‘adverse environmental impact.’” (internal footnote omitted)). Riverkeeper I thuscontrols this issue. In Riverkeeper I, we rejected the arguments that some species are nuisances andrequire eradication, that other species respond to population losses by increasing theirreproduction, and that removing large numbers of aquatic organisms from waterbodies is not inand of itself an adverse impact. 358 F.3d at 196. We specifically rejected the view that “theEPA should only have sought to regulate impingement and entrainment where they havedeleterious effects on the overall fish and shellfish populations in the ecosystem, which can onlybe determined through a case-by-case, site-specific regulatory regime.” Id. We emphasized that“the EPA’s focus on the number of organisms killed or injured by cooling water intake structuresis eminently reasonable.” Id. We reiterated that Congress had “rejected a regulatory approachthat relies on water quality standards,” analogizing the argument pressed there as urging what isessentially a water quality standard that focuses on fish populations and consequential

Page 157: Sotomayor Cases

36 We make an additional observation here, which we cited in Riverkeeper I as a reasonfor rejecting restoration measures as impermissible under the statute. It is significant that insection 316(a), which governs thermal discharges, Congress permits the EPA to vary the standardapplicable to a point source “by considering the particular receiving waterbody’s capacity todissipate the heat and preserve a ‘balanced, indigenous’ wildlife population.” 358 F.3d at 190. Itis also significant that Congress “did not include that [water quality or population level] approach(or make any reference to it) in the very next subsection,” id., since “where Congress includesparticular language in one section of a statute but omits it in another section of the same Act, it isgenerally presumed that Congress acts intentionally and purposely in the disparate inclusion orexclusion.” Id. (quoting Bates v. United States, 522 U.S. 23, 29-30 (1997) (internal quotationmarks omitted)). The statutory structure thus indicates that Congress did not intend to limit“adverse environmental impact” in section 316(b) to population-level effects.69

environmental harm.36 Id. at 196-97. Given that the record evidence on this issue has notchanged in any meaningful way since the Phase I rulemaking, we are both persuaded and boundby our statements on this issue in Riverkeeper I. Were we considering the issue in the first instance, however, we would beinclined to defer to the EPA’s judgment in any event. The EPA explained that it has set“performance standards for minimizing adverse environmental impact based on a relatively easyto measure and certain metric – reduction of impingement mortality and entrainment.” 69 Fed.Reg. at 41,600. It explained further that it chose this approach “because impingement andentrainment are primary, harmful environmental effects that can be reduced through the use ofspecific technologies” and stated that “where other impacts at the population, community, andecosystem levels exist, these will also be reduced by reducing impingement and mortality.” Id. We see no reason to second-guess this judgment, given the Agency’s consideration of the variousenvironmental consequences of cooling water intake structures. See Nat’l Wildlife Fed’n, 286F.3d at 570 (noting that courts afford the EPA considerable discretion to weigh and balance

Page 158: Sotomayor Cases

37 We also find no merit in PSEG’s claim that the final Rule gave no notice of thesupposed change in the EPA’s view of “adverse environmental impact.” Not only did theproposed Rule seek comment on several competing definitions, one of which referred to gaugingsuch impact in part by examination of the “absolute damage” wrought by impingement andentrainment, but in which the EPA noted that “there will be adverse environmental impactwhenever there is entrainment or impingement ‘damage’ as a result of a cooling water intakestructure.” 67 Fed. Reg. at 17,162. While the EPA did not ultimately adopt any of the definitionsit noticed, it is clear that PSEG was “fairly apprise[d]” that the EPA was considering an approachto AEI that looked to entrainment and impingement at the individual level. Nat’l Black MediaCoalition, 791 F.2d at 1022. 70

various factors in determining how to establish performance standards).37 C. Zero Entrainment Survival Assumption

The Phase II Rule requires a reduction of impingement mortality, but a reductionof entrainment generally. See 40 C.F.R. § 125.94(b). The industry petitioners contend that theEPA improperly presumed that all entrained organisms are killed. They argue that the Rule’sassumption that no organisms survive entrainment is contrary to the evidence which, in theirview, indicates a survival rate of significantly more than zero. We conclude that in light ofuncertain record evidence, the EPA acted within its discretion in assuming zero entrainmentsurvival.The EPA explained in the preamble to the Phase II Rule that it assumed zeropercent entrainment survival because it “does not have sufficient data to establish performancestandards based on entrainment survival for the technologies used as the basis for today’s rule.” 69 Fed. Reg. at 41,620. It explained further that it “believes the current state of knowledge doesnot support reliable predictions of entrainment survival that would provide a defensible estimatefor entrainment survival above zero at a national level.” Id. It also stated that the performancestandard would likely have been higher had it incorporated entrainment survival into its

Page 159: Sotomayor Cases

71

conclusions. Id. While impingement mortality can be readily quantified, the EPA contends,entrainment mortality cannot because many entrained organisms are small, fragile, and prone todisintegrate during entrainment. The EPA also contends that death from entrainment sometimesoccurs immediately but other times only after an organism is discharged back into the waterbody. In view of these factors, the EPA claims that it reasonably concluded that the available data didnot support an estimate of entrainment survival at the national level. None of the peer reviewers accepted the EPA’s assumption of zero percentsurvival. For instance, peer reviewer Dr. Mark Bain was “not convinced by the argumentspresented that fish do not survive entrainment in significant numbers” and concluded that there is“very strong evidence that entrainment survival is not zero.” Another peer reviewer, Dr. CharlesHocutt, concluded that the EPA’s assumption is based on inference and innuendo and does notstatistically refute opposing views. Although peer reviewer Dr. Greg Garman stated that thestudy submitted by the power industry was “very clearly biased” and “too seriously flawed toprovide a serious challenge to the EPA position,” he also noted that “EPA’s position is onlyslightly more defensible” given the lack of rigorous statistical analyses. No peer reviewer, however, expressed the belief that reliable national statistics onentrainment survival exist. Hocutt found “it difficult for the layman and professional alike todraw confident conclusions without a detailed analysis” of the evidence, and Garman suspected alack of sufficient data “to conduct a definitive and statistically valid test of the EPA zero survivalassumption.” While these comments do not, on balance, support the EPA’s assumption, neitherdo they reflect any meaningful agreement on the incidence of entrainment survival.Given the statutory directive to set national standards and the well-documented

Page 160: Sotomayor Cases

72

uncertainty in the entrainment data, the EPA was well within its authority to determine that itcould not provide a reasonable estimate of entrainment survival on a national basis. Judicialreview is considerably deferential when “the agency’s decision rests on an evaluation of complexscientific data within the agency’s technical expertise.” Texas Oil & Gas Ass’n, 161 F.3d at 934;City of Waukesha v. EPA, 320 F.3d 228, 247 (D.C. Cir. 2003) (stating that an agency is entitledto “an extreme degree of deference . . . when it is evaluating scientific data within its technicalexpertise” (citation and internal quotation marks omitted)). Moreover, “it is within EPA’sdiscretion to decide that in the wake of uncertainty, it would be better to give the values aconservative bent rather than err on the other side.” Am. Iron & Steel Inst. v. EPA, 115 F.3d 979,993 (D.C. Cir. 1997). Indeed, one peer reviewer expressly noted that the EPA had adopted a“conservative approach” by its assumption of zero entrainment survival in the Rule. It is thusclear that the EPA acted well within its discretion in presuming zero entrainment survival afterthe Agency had reviewed a substantial body of complex scientific data, and acknowledging thatthe evidence is inconclusive, it adopted a conservative approach. For these reasons and those stated in Riverkeeper I, we therefore “defer to theEPA’s judgment of how best to define and minimize ‘adverse environmental impact.’” 358 F.3dat 197.

D. Nuclear PlantsEntergy contends that the Phase II Rule fails to account for its purportedlydisproportionate impacts on nuclear power plants and is therefore arbitrary and capricious.Entergy argues that nuclear facilities face unique safety concerns associated with the stable flowof cooling water to ensure safe reactor operation and shutdown. Any change in water intake or

Page 161: Sotomayor Cases

73

obstruction of water intake systems due to, for example, the clogging of screens, it argues further,affects nuclear power facilities in specific and serious ways. Entergy takes the position that theEPA failed to account for these issues in the Phase II Rule. We disagree because the recorddemonstrates adequate consideration by the EPA of nuclear plants’ particular concerns. The EPA considered and responded to comments from nuclear facilities duringthe rulemaking process. Most importantly, the Agency considered whether the Rule’srequirements presented any concerns relating to the safety of nuclear facilities. 69 Fed. Reg. at41,585 (noting that the EPA had coordinated with the Nuclear Regulatory Commission to ensurethat there would not be a conflict between the EPA Rule and safety requirements applicable tonuclear facilities). The EPA ultimately included in the Phase II Rule a provision that accountsfor this concern by providing for a site-specific compliance alternative for nuclear facilities. Thisprovision states that if a nuclear facility “demonstrate[s] to the [EPA] based on consultation withthe Nuclear Regulatory Commission that compliance with this subpart would result in a conflictwith a safety requirement established by the Commission, the [EPA] must make a site-specificdetermination of best technology available for minimizing adverse environmental impact thatwould not result in a conflict.” 40 C.F.R. § 125.94(f). We defer to the EPA’s determination that this compliance alternative ensures thatany safety concerns unique to nuclear facilities will prevail over application of the general PhaseII requirements. See Nat’l Wildlife Fed’n, 286 F.3d at 570 (noting that appellate courts give anagency considerable discretion when it has weighed and balanced the appropriate factors); BPExploration & Oil, Inc. v. EPA, 66 F.3d 784, 802 (6th Cir. 1995) (“The overriding principle inour review of the Final Rule is that the agency has broad discretion to weigh all relevant factors

Page 162: Sotomayor Cases

38 The Phase I Rule provided that[u]se of a cooling water intake structure includes obtaining cooling water by anysort of contract or arrangement with an independent supplier (or multiplesuppliers) of cooling water if the supplier or suppliers withdraw(s) water fromwaters of the United States. . . . [t]his provision is intended to prevent74

during rulemaking. The CWA does not state what weight should be accorded to the relevantfactors; rather, the Act gives EPA the discretion to make those determinations.”). Moreover, weare persuaded that the generous cost-cost compliance alternative, which we remand for lack ofnotice but do not address on the merits, may further account for Entergy’s concerns. Accordingly, we deny the petition for review insofar as it challenges the Rule’sapplication to nuclear facilities.

E. Independent SuppliersThe Phase II Rule provides that a large, existing facility is subject to the Rule’srequirements even when it obtains cooling water from an independent supplier that is not itself aPhase II existing facility. 40 C.F.R. § 125.91(c). Put differently, a Phase II facility can purchasecooling water only from suppliers whose intake structures are in compliance with the Phase IIRule. The provision is intended to prevent circumvention of the Rule by purchasing coolingwater from non-Phase II entities. UWAG argues that the EPA gave inadequate notice of thescope of this so-called third-party or independent-supplier Rule. We agree.UWAG contends that while the Phase I Rule included an independent-supplierprovision similar to the one at issue here, the parallel provision of the Phase I Rule applies onlyto third-party suppliers that are not point sources, and thus do not need discharge permits and arenot subject to any rule promulgated pursuant to section 316(b).38 UWAG argues that the Rule’s

Page 163: Sotomayor Cases

circumvention of these requirements by creating arrangements to receive coolingwater from an entity that is not itself a point source.40 C.F.R. § 125.81(b). 75

proposal indicated that the Phase II provision would track the language of the Phase I provision. UWAG contends, therefore, that the EPA gave inadequate notice of the more broadly phrasedprovision that was actually promulgated: a rule applying both to the intake structures of third-parties that are not point sources, and therefore not subject to direct regulation undersection 316(b), as well as to facilities with intake structures that will be governed by the Phase IIIrule, which will encompass smaller power plants and other facilities. The issue is thus whetherthe EPA provided adequate notice that the Rule ultimately promulgated would impose Phase IIrequirements on Phase III facilities that supply cooling water to Phase II facilities. As we have noted, our inquiry into whether an agency has provided adequatenotice of its rulemaking as required by the APA is guided by the “logical outgrowth” test. Nat’lBlack Media Coalition, 791 F.2d at 1022. That is, we must determine whether the Agency’snotice fairly apprised interested parties of the rulemaking, id., or whether the final Rule wassufficiently remote or distant from the Agency’s proposal to constitute a “surprise switcheroo,”Envtl. Integrity Project, 425 F.3d at 996.The proposed Rule published in the Federal Register stated that “[u]se of acooling water intake structure includes obtaining cooling water by any sort of contract orarrangement with an independent supplier (or multiple suppliers) of cooling water if the supplieror suppliers withdraw(s) water from waters of the United States.” 67 Fed. Reg. at 17,220. Thenotice then stated that the provision was intended “to prevent circumvention of these

Page 164: Sotomayor Cases

39 The final Rule provides that “[u]se of a cooling water intake structure includesobtaining cooling water by any sort of contract or arrangement with one or more independentsuppliers of cooling water if the supplier withdraws water from waters of the United States but isnot itself a Phase II existing facility . . . . This provision is intended to prevent circumvention ofthe requirements by creating arrangements to receive cooling water from an entity that is notitself a Phase II existing facility.” 40 C.F.R. § 125.91(c). 76

requirements by creating arrangements to receive cooling water from an entity that is not itself apoint source,” id. (emphasis added), indicating that the provision would track the Phase I Rule’slanguage. The initial language of the provision appearing in the proposal – defining“independent supplier” as any provider that withdraws water from waters of the United States –is virtually identical to the language in the final Rule.39 The final Rule, however, states that it “isintended to prevent circumvention of these requirements by creating arrangements to receivecooling water from an entity that is not itself a Phase II existing facility.” 40 C.F.R. § 125.91(c)(emphasis added). This difference, while small, is not insignificant. As noted, under theproposed Rule, Phase II facilities could purchase cooling water from facilities complying withthe Phase I, II, or III requirements. Under the final Rule, by contrast, Phase II facilities canpurchase cooling water only from facilities complying with the Phase II Rule. The proposaltherefore apprised affected parties of the general subject, but not the scope, of the Rule ultimatelypromulgated. Interested parties were thus given notice that the EPA sought to preventcircumvention of its rules and an opportunity to comment on this general policy, but could nothave anticipated the final Rule’s scope. Indeed, while the final provision has roots in theproposal, it clearly reaches further than the proposed provision and even the more stringent PhaseI Rule. Affected parties would therefore have had no reason to anticipate the Agency’s final

Page 165: Sotomayor Cases

77

course in light of the initial notice. Because the EPA provided inadequate notice of the scope ofthe Phase II Rule’s independent-supplier provision, we conclude that the Rule ultimatelypromulgated is not a “logical outgrowth” of the proposed Rule. Nat’l Black Media Coalition,791 F.2d at 1022. Our conclusion rests in large part on the similarity between the Phase I Rule’sindependent-supplier provision and the proposed Phase II Rule’s parallel provision. Given thatthe requirements under the Phase I Rule are more stringent than those imposed under the Phase IIRule, the proposal provided no notice that the Phase II Rule’s independent-supplier provisionwould be more stringent than the Phase I Rule’s provision. We therefore remand this aspect ofthe Rule.

F. Definition of “Great Lakes”The entrainment performance standard of 40 C.F.R. § 125.94(b)(2) applies tofacilities that use cooling water “withdrawn from . . . one of the Great Lakes.” 40 C.F.R.§ 125.49(b)(2)(ii)(A). “Great Lakes,” however, is not defined in the Rule. An affidavit ofUWAG’s counsel states that UWAG members

have been told by state regulators that they have been told by EPA Headquartersthat EPA would apply the Great Lakes national performance standards regardingimpingement and entrainment to the cooling water intake structures located in theGreat Lakes connecting channels or in waterways with open fish passage to aGreat Lake and within 30 miles from a Great Lake. UWAG argues that interested parties were given no notice of this interpretation of the Phase IIRule or opportunity to comment on it and that the interpretation has no record support. UWAGalso remarks that the EPA mentioned in the preamble to the Rule that in assessing the national

Page 166: Sotomayor Cases

78

environmental benefits of its final Phase II Rule, it had evaluated the benefits in seven studyregions. The EPA defined the Great Lakes region for that purpose as follows:The Great Lakes region includes all facilities in scope of the Phase II rule thatwithdraw water from Lakes Ontario, Erie, Michigan, Huron, and Superior or arelocated on a waterway with open fish passage to a Great Lake and within 30 milesof the lake.

69 Fed. Reg. at 41,655. UWAG contends that the EPA never indicated that this definition wouldapply to § 125.94(b)(2) and seeks either a ruling that “Great Lakes” has its ordinary meaning or aremand for an express definition of what the term means. Notwithstanding UWAG’s proffer that the EPA has informally interpreted “GreatLakes,” the Rule itself does not define what is encompassed by the term, and UWAG hasprovided no documentary evidence that the EPA has issued a formal and binding definition oreven applied a particular definition in a permitting proceeding. There is, therefore, no finalagency action for us to review. We agree with the holding of the Seventh Circuit in AmericanPaper Institute, Inc. v. Environmental Protection Agency, that in the absence of a formal andbinding rule or some other final agency action, judicial review is not available at this time. 882F.2d 287, 289 (7th Cir. 1989). We therefore dismiss for lack of jurisdiction so much of thepetition for review as challenges the purported definition of “Great Lakes.”

G. Preemptive Preservation of IssuesUWAG has purported to “preserve” the right to raise new challenges to the PhaseII Rule if we remand significant aspects of it to the EPA because UWAG views the Rule as anintegrated whole. If certain aspects of the Rule are remanded, UWAG suggests, previouslyunobjectionable provisions may become, in its view, unacceptable. We are not sure what

Page 167: Sotomayor Cases

79

challenges UWAG seeks to preserve. Of course, UWAG will have the right to challenge any rulethe EPA may promulgate on remand through a new petition for review. See 33 U.S.C. § 1369(b). Insofar as UWAG seeks to raise further challenges following the issuance of our ruling butbefore the EPA acts on remand, however, it waived those challenges by failing to raise them inthe briefs it has already submitted to this Court. See Norton v. Sam’s Club, 145 F.3d 114, 117(2d Cir. 1998) (noting that issues not argued in a party’s briefs are considered waived). We thusreject UWAG’s claim to have preserved the right to raise new challenges to the Rule currentlybefore us on this petition for review. CONCLUSION

For the foregoing reasons, the state and environmental petitioners’ petitions aregranted in part and denied in part, and the industry petitioners’ petitions granted in part, denied inpart, and dismissed in part for lack of jurisdiction. We remand to the EPA the provisionestablishing BTA so that it may provide either a reasoned explanation of its decision or a newdetermination of BTA based on permissible considerations. We further remand the site-specificcost-cost variance and the TIOP provision because the cost-cost variance and subpart (d)(2) ofthe TIOP provision provided inadequate notice and both depend on the BTA determination,which we remand today. We remand as based on impermissible constructions of the statute thoseprovisions that (1) set performance standards as ranges without requiring facilities to achieve thegreatest reduction of adverse impacts they can; (2) allow compliance through restorationmeasures; and (3) authorize a site-specific cost-benefit variance as impermissible under thestatute. We further remand for notice and comment the independent suppliers provision. We

Page 168: Sotomayor Cases

80

also direct the EPA on remand to adhere to the definition of “new facility” set forth in the Phase IRule or to amend that definition by rulemaking subject to notice and comment. Finally, wedismiss for lack of jurisdiction so much of the petitions as challenges the purported definition of“Great Lakes” and deny as moot the motions to strike certain material from the record and tosupplement the record with other material.

Page 169: Sotomayor Cases

1

UNITED STATES COURT OF APPEALSFOR THE SECOND CIRCUIT

August Term, 2004

(Argued: January 5, 2005 Decided: February 16, 2006)

Docket No. 04-0743-cv_____________________________________________

JOHN PAUL HANKINS,

Plaintiff-Appellant,

v.

ERNEST S. LYGHT and NEW YORK ANNUAL CONFERENCE OF THE UNITEDMETHODIST CHURCH,

Defendants-Appellees,

STONY BROOK COMMUNITY CHURCH,

Defendant.____________________________________

SOTOMAYOR, Circuit Judge, dissenting:

The Religious Freedom and Restoration Act (“RFRA”) is not relevant to this dispute.

First, appellees have unambiguously indicated that they do not seek to raise a RFRA defense, and

the statute’s protections, even if otherwise applicable, are thus waived. Second, the statute does

not apply to disputes between private parties. Third, we should affirm the judgment of the

district court without reaching the RFRA issue on the ground that Supreme Court and Second

Circuit precedent compels a finding that the Age Discrimination in Employment Act (ADEA), 29

U.S.C. § 621 et seq., does not govern disputes between a religious entity and its spiritual leaders.

Page 170: Sotomayor Cases

1 In addition to ignoring most of the language in the appellees’ brief relating to waiver,the majority opinion makes two factually erroneous claims regarding the content of thesupplemental letter-briefs. First, the majority writes that “appellant argues that the RFRA is

2

The majority’s opinion thus violates a cardinal principle of judicial restraint by reaching

unnecessarily the question of RFRA’s constitutionality. For these reasons, I respectfully dissent.

A.

Because the parties’ original submissions to this Court mentioned RFRA without

providing a detailed analysis of either the Act’s constitutionality or its relevance to this case, we

ordered supplemental briefing. The letter-briefs submitted in response to our order make clear

that appellees have waived any RFRA defense.

In several portions of appellees’ supplemental brief that the majority neglects to mention,

appellees state plainly that they do not intend to raise a RFRA defense. Appellees’ supplemental

brief explains that “the reference to RFRA in Appellees’ [original] brief was for the limited

purpose of providing an example of how critically the question of ‘entanglement’ was viewed”

by Congress. In other words, appellees’ aim was not to rely on the statute as a defense against

appellant’s claims, but merely to illustrate Congress’s agreement with the proposition that

“entanglement of the Government in church affairs [was] prohibited by the First Amendment.”

(emphasis added). In fact, appellees explicitly reject the application of RFRA to their claims

because they believe that the statute does not apply to suits between private parties, and “the case

at bar is a matter relating to a private employment situation and does not involve actions by the

government.” The letter-brief concludes: “We do not think this issue [RFRA] is determinative in

the matters raised by this case.” While the majority might find appellees’ position unwise or

“supris[ing],” Maj. Op. at 12, appellees’ letter-brief clearly waives any RFRA defense.1

Page 171: Sotomayor Cases

inapplicable only because it is unconstitutional.” Maj. Op. at 12. On the contrary, appellantargues also that RFRA is inapplicable because “[t]here is no substantial burden to the freeexercise of religion that could result from a ruling by this court” that appellees violated theADEA. The majority also contends that “[t]he parties have not briefed the issue of how [RFRA]impacts the merits of this case.” Maj. Op. at 24. Both parties, however, have addressed thequestion of RFRA’s relevance to this suit. Appellant argues that RFRA is inapplicable because“[t]here is no substantial burden to the free exercise of religion” in this case, and that, in anyevent, “application of RFRA to federal law is unconstitutional.” Appellees, in turn, argue thatRFRA is constitutional but should not affect the outcome of this case.

3

The majority does not contest that RFRA’s protections are generally waivable. Maj. Op.

at 13; see United States v. Amer, 110 F.3d 873, 879 n.1 (2d Cir. 1997); see also In re Watson,

403 F.3d 1, 7 (1st Cir. 2005) (holding that RFRA argument was forfeited); Bethesda Lutheran

Homes & Servs., Inc. v. Leean, 122 F.3d 443, 449 (7th Cir. 1997) (holding that RFRA argument

was waived); Cochran v. Morris, 73 F.3d 1310, 1317 n.3 (4th Cir. 1996) (holding that RFRA

claim was waived). In the majority’s view, however, because appellees’ arguments relate to

rights protected under RFRA—namely, First Amendment religious rights—appellees have “[i]n

substance” relied on RFRA and thus have not, despite their explicit disclaimer, waived its

protections. Maj. Op. at 14.

The majority’s refusal to recognize appellees’ waiver in this case is at odds with RFRA’s

text, which provides that individuals “may assert” a RFRA defense when challenging a

substantial burden on their religious rights, not that they must assert a RFRA defense when

religious rights are burdened. 42 U.S.C. § 2000bb-1(c) (emphasis added). Moreover, the

majority’s insistence on the viability of a RFRA defense despite appellees’ waiver leads the

Court to assess RFRA’s constitutionality unnecessarily. See Cutter v. Wilkinson, 125 S. Ct.

2113, 2118 n.2 (2005) (noting that the Supreme Court has “not had occasion to rule” whether

RFRA “remains operative as to the Federal Government”); see also City of Boerne v. Flores, 521

Page 172: Sotomayor Cases

2 The other “cardinal rule[]” cited in Brockett is that federal courts should “never . . .formulate a rule of constitutional law broader than is required by the precise facts to which it is tobe applied.” 472 U.S. at 501 (citation and internal quotation marks omitted).

3 Like RFRA, RLUIPA prohibits the government from imposing substantial burdens onreligion even where the burden results from a neutral law of general applicability. See 42 U.S.C.§ 2000cc. RLUIPA’s remedial provision is virtually identical to RFRA’s. Compare 42 U.S.C. § 2000cc-2(a) (“A person may assert a violation of this Act as a claim or defense in a judicialproceeding and obtain appropriate relief against a government.”), with 42 U.S.C. § 2000bb-1(c)(“A person whose religious exercise has been burdened in violation of this section may assertthat violation as a claim or defense in a judicial proceeding and obtain appropriate relief against agovernment.”). RLUIPA is simultaneously more broad and more narrow than RFRA, however. RLUIPA is more broad because it still reaches state law. See Cutter v. Wilkinson, 125 S. Ct.2113 (2005). It is more narrow because it applies only to certain government actions involvingland use regulations and correctional facilities. See 42 U.S.C. §§ 2000cc, 2000cc-1.

4

U.S. 507, 532-36 (1997) (invalidating RFRA as applied to state law). By going out of its way to

reach this constitutional question, the majority violates one of the “cardinal rules governing the

federal courts,” namely, “never to anticipate a question of constitutional law in advance of the

necessity of deciding it.” Brockett v. Spokane Arcades, Inc., 472 U.S. 491, 501 (1985) (citation

and internal quotation marks omitted).2

The majority’s approach is also inconsistent with our case law, which has recognized

waiver of statutory religious rights even where a litigant raises claims under the Free Exercise

Clause. In Fifth Avenue Presbyterian Church v. City of New York, 293 F.3d 570 (2d Cir. 2002),

for example, the plaintiff argued before this Court that its religious rights had been violated under

both the First Amendment and the Religious Land Use and Institutionalized Persons Act

(RLUIPA)—a statute virtually identical to RFRA in all aspects relevant to the issue of waiver in

the instant case.3 Although we ruled on the merits of the plaintiff’s Free Exercise claim in Fifth

Avenue Presbyterian Church, we refused to reach the RLUIPA issue because the plaintiff had

raised it for the first time on appeal. See id. at 576. It is impossible to square our refusal to

Page 173: Sotomayor Cases

5

consider plaintiff’s belated RLUIPA claim in Fifth Avenue Presbyterian Church with our refusal

to recognize the defendant’s voluntary waiver of a RFRA defense in the instant case. There is no

meaningful difference between RFRA and RLUIPA that could justify such inconsistent results.

The most troublesome aspect of the majority’s ruling on waiver, however, is that it

fundamentally misconstrues the nature of RFRA and First Amendment rights, and, in doing so,

directly contradicts Supreme Court precedent. The majority holds that because appellees invoke

the First-Amendment-based “ministerial exception” and allege interference with their rights

under the Free Exercise and Establishment Clauses, they have effectively “ask[ed] us to apply the

RFRA, but not to mention it.” Maj. Op. at 14. This is incorrect. RFRA and the First

Amendment do not provide identical protections, and the invocation of First Amendment

rights—whether under the Free Exercise or the Establishment Clause—does not necessarily

implicate RFRA.

As interpreted by the Supreme Court, for example, the Free Exercise Clause does not

normally “inhibit enforcement of otherwise valid laws of general application that incidentally

burden religious conduct,” Cutter, 125 S. Ct. at 2118 (citing Employment Div., Dep’t of Human

Res. v. Smith, 494 U.S. 872, 878-82 (1990)), such as the ADEA. RFRA, in contrast, requires

strict scrutiny of such laws where the incidental burden on religion is substantial. See 42 U.S.C.

§ 2000bb-1. Indeed, the fact that RFRA’s protections sweep more broadly than those of the Free

Exercise Clause provided the principal basis for the Supreme Court’s holding in City of Boerne

v. Flores that RFRA could not be considered “preventive” or “remedial” legislation under

Section Five of the Fourteenth Amendment. 521 U.S. at 532. The Court found RFRA’s

protections “so out of proportion to a supposed remedial or preventive object that [the statute]

Page 174: Sotomayor Cases

4 I express no view on whether RFRA is constitutional as applied to federal law because itis unnecessary for us to reach this question.

5 Before Boerne, a reasonable argument could have been made that all Free ExerciseClause claims required scrutiny under RFRA. The Tenth Circuit, for example, held in Werner v.McCotter, 49 F.3d 1476 (10th Cir. 1995), that RFRA applied to all Free Exercise claims, evenwhere the parties had not raised a claim or defense under the statute. In a subsequent en bancopinion, however, the Tenth Circuit recognized that Boerne had undermined its earlierconclusion:

[I]n Werner, decided prior to City of Boerne, we were laboring under the falseunderstanding that RFRA “legislatively overturned a number of recent SupremeCourt [free exercise] decisions” and that it created a new rule of constitutionallaw. Thus, we concluded that because the language of RFRA made it applicableto “all cases where free exercise of religion is substantially burdened,” its standardought to control a Free Exercise Clause claim even when not raised. Because theSupreme Court has made clear that the Werner court’s assumptions about RFRAwere faulty, its rationale is no longer convincing.

United States v. Hardman, 297 F.3d 1116, 1125 n.15 (10th Cir. 2002) (en banc) (alteration inoriginal) (citations omitted).

6

cannot be understood as responsive to, or designed to prevent, unconstitutional behavior.” Id.

Because RFRA went so far beyond what the First Amendment required, the Boerne Court

understood the statute as “attempt[ing] a substantive change in constitutional protections”—a

change that Congress was not authorized to make. Id. Although Boerne does not resolve the

issue of RFRA’s constitutionality as applied to federal law, as opposed to state law,4 the case

does firmly establish that RFRA and the Free Exercise Clause create different standards for the

protection of religion and that RFRA’s substantive protections extend far beyond what the Free

Exercise Clause requires. Thus, the majority’s suggestion that a claim alleging unconstitutional

interference with the free exercise of religion is “[i]n substance” a RFRA claim flies in the face

of Boerne.5 See also Kaufman v. McCaughtry, 419 F.3d 678, 681 (7th Cir. 2005) (noting that

RFRA provides protections beyond those guaranteed by the First Amendment); Brzonkala v. Va.

Page 175: Sotomayor Cases

6 As recently emphasized by a plurality of Justices, the Supreme Court has not applied theLemon test with much consistency. See Van Orden v. Perry, 125 S. Ct. 2854, 2860-61 (2005)(plurality opinion). I am unaware of any application of the Establishment Clause, however, thatwould invalidate a neutral, generally applicable law imposing an incidental but substantialburden on religion.

7

Polytech. Inst. & State Univ., 169 F.3d 820, 881-82 (4th Cir. 1999) (“The [RFRA] created a right

of religious exercise that was more generous than that right protected by the Constitution . . . .”),

aff’d sub nom. United States v. Morrison, 529 U.S. 598 (2000).

Nor can the majority plausibly argue that appellees’ Establishment Clause defense

necessarily implicates RFRA. To satisfy the Establishment Clause: (1) the statute must have “a

secular legislative purpose”; (2) the statue’s “principal or primary effect must be one that neither

advances nor inhibits religion”; and (3) “the statute must not foster an excessive government

entanglement with religion.” Lemon v. Kurtzman, 403 U.S. 602, 612-13 (1971) (citation and

internal quotation marks omitted). Thus, like the Free Exercise Clause, the Establishment Clause

imposes less stringent requirements on federal statutes than RFRA, which mandates strict

scrutiny even of neutral, generally applicable laws that incidentally impose substantial burdens

on religion.6 Furthermore, Congress made clear in enacting RFRA that the statute was not

intended to have any effect on Establishment Clause claims. See 42 U.S.C. § 2000bb-4

(“Nothing in this chapter shall be construed to affect, interpret, or in any way address that portion

of the First Amendment prohibiting laws respecting the establishment of religion.”).

The majority’s assertion that appellees have presented a RFRA defense in “all but name”

would be more plausible if something in appellees’ briefs indicated that they sought protection

beyond that which the Constitution guarantees. Nothing in the briefs, however, supports such a

conclusion. Appellees’ briefs rely heavily on the Free Exercise Clause, the Establishment

Page 176: Sotomayor Cases

7 The closest appellees come to making a RFRA argument, as opposed to a FirstAmendment argument, is a statement in their original brief that application of the ADEA would“substantially burden the free exercise rights of the United Methodist Church.” This is the onlyoccasion, however, in which appellees employ RFRA-like language by referring to the allegedintrusion on their rights as a “substantial[] burden,” and it is clear from context that the statementformed part of appellees’ Establishment Clause argument that application of the ADEA wouldfoster an excessive entanglement with religion. The brief did not purport to raise a separatedefense under RFRA. In any event, even if appellees’ mention of a “substantial[] burden” intheir original brief could be generously construed as an attempt to present a RFRA defense,appellees’ subsequent letter-brief makes clear that this was not their intent and that they do notseek to invoke RFRA’s protections.

8

Clause, and case law interpreting those provisions. Nowhere do they ask that the Court apply a

standard stricter than what the First Amendment requires.7 On the contrary, appellees’

supplemental brief explicitly disclaims any intent to rely on RFRA.

In sum, because appellees’ religious freedom argument relies only on the Free Exercise

and Establishment Clauses, and because the substance of the protections afforded by these

constitutional provisions differs considerably from the protections afforded by RFRA, as

interpreted by the Supreme Court, I cannot agree with the majority’s conclusion that appellees

have “[i]n substance” relied on RFRA. Maj. Op. at 14.

The majority’s refusal to recognize appellees’ clear waiver of any RFRA defense appears

to rest, in part, on its disagreement with the reasons underlying appellees’ decision not to pursue

such a defense. Specifically, the majority takes issue with appellees’ conclusion that RFRA does

not apply to suits between private parties. See Maj. Op. at 10-11. I am unaware of any other case

in which this Court, after ordering supplemental briefing to allow a party to discuss a waivable

statutory defense, refused to recognize the party’s subsequent waiver of that defense on the

ground that the Court disagreed with counsel’s reasons for declining to rely on the statute. Cf.

DeLuca v. Lord, 77 F.3d 578, 588 (2d Cir. 1996) (observing that where defense counsel in a

Page 177: Sotomayor Cases

9

criminal case has made “a considered decision, after investigation, not to pursue” a particular

defense, this Court should be “extremely reluctant to second-guess that decision”). Even if such

second-guessing of a party’s decision not to pursue a particular defense is appropriate in certain

limited circumstances, the majority’s refusal to acknowledge the clear waiver in this case is

improper, given that appellees are adequately represented by counsel and based their waiver on a

reasonable interpretation of the law. Indeed, the majority concedes that it is unable to find a

single holding that contradicts appellees’ view that RFRA does not apply to suits between private

parties. See Maj. Op. at 25 n.4.

Quoting the Supreme Court, the majority argues that “[w]hen an issue or claim is

properly before the court, the court is not limited to the particular legal theories advanced by the

parties, but rather retains the independent power to identify and apply the proper construction of

governing law.” Maj. Op. at 14 (quoting Kamen v. Kemper Fin. Servs., Inc., 500 U.S. 90, 99

(1991)). This certainly is true, but it only begs the question of whether the “issue or claim is

properly before the court.” Id. Given appellees’ clear indication that they do not seek to rely on

RFRA, the applicability of that statute is not before us. The majority’s disagreement with

appellees’ reasoning does not change that fact.

B.

Even assuming, arguendo, that appellees’ clear disclaimer of any RFRA defense does not

suffice to waive such a defense, I would find it improper to remand the case to the district court

for consideration of RFRA’s implications because I disagree with the majority’s conclusion

regarding RFRA’s applicability. RFRA by its terms does not apply to suits between private

parties.

Page 178: Sotomayor Cases

8 There are two plausible ways to reconcile section 2000bb-1(b) of RFRA with themajority opinion in this case. The first would be to require government intervention in everyprivate suit where one of the parties asserts that a law has—even incidentally—imposed asubstantial burden on religious freedom. Absent a clear statement that Congress intended such aresult, it is not the role of this Court to mandate such widespread and automatic federalintervention in lawsuits between private parties. Moreover, were we to read the statute to requiregovernment intervention, this would surely underscore the wisdom in recognizing appellees’explicit waiver of any RFRA defense. The second would be to force private parties to bear theburden RFRA places on the government. The statute gives no indication that Congress intendedprivate parties to bear such a burden, nor would it be appropriate to require private parties tosatisfy the stringent burden RFRA places on the government.

10

Two provisions of the statute implicitly limit its application to disputes in which the

government is a party. Section 2000bb-1(c) states that “[a] person whose religious exercise has

been burdened in violation of this section may assert that violation as a claim or defense in a

judicial proceeding and obtain appropriate relief against a government” (emphasis added). In the

majority’s view, we should read this provision as “broadening, rather than narrowing, the rights

of a party asserting the RFRA.” Maj. Op. at 11. This interpretation would be questionable even

if Section 2000bb-1(c) were the only provision of the statute affecting the question of whether

RFRA applies to private suits. When read in conjunction with the rest of the statute, however, it

becomes clear that this section reflects Congress’s understanding that RFRA claims and defenses

would be raised only against the government. For instance, section 2000bb-1(b) of RFRA

provides that where a law imposes a substantial burden on religion, the “government” must

“demonstrate[] . . . that application of the burden” is the least restrictive means of furthering a

compelling governmental interest (emphasis added). The statute defines “demonstrate” as

“meet[ing] the burdens of going forward with the evidence and of persuasion.” 42 U.S.C. §

2000bb-2(3). Where, as here, the government is not a party, it cannot “go[] forward” with any

evidence.8 In my view, this provision strongly suggests that Congress did not intend RFRA to

Page 179: Sotomayor Cases

9 All of the examples cited in the Senate and House Reports on RFRA involve actual orhypothetical lawsuits in which the government is a party. See S. Rep. No. 103-111 (1993); H.R.Rep. 103-88 (1993). The lack of even a single example of a RFRA claim or defense in a suitbetween private parties in these Reports tends to confirm what is evident from the plain languageof the statute: It was not intended to apply to suits between private parties.

11

apply in suits between private parties.9

I recognize that according to RFRA’s “applicability” section, the statute applies “to all

Federal law.” 42 U.S.C. § 2000bb-3. This provision, however, is not inconsistent with a finding

that the statute does not apply to suits between private parties. Read in conjunction with the rest

of the statute, the provision simply requires courts to apply RFRA “to all Federal law” in any

lawsuit to which the government is a party.

The majority objects that this interpretation makes RFRA’s protections improperly

dependent on whether a private party, as opposed to the EEOC, brings suit under the ADEA.

“[T]he substance of the ADEA’s prohibitions,” the majority argues, “cannot change depending

on whether it is enforced by the EEOC or an aggrieved private party.” Maj. Op. at 11. The

majority does not explain, however, why this is so. If RFRA amends all federal statutes as they

apply to suits in which the government is a party, then the substance of the ADEA’s prohibitions

most certainly can change depending on who enforces it. Although the majority evidently finds

this unsatisfactory from a policy perspective, there is no acceptable reading of the statute that

would yield the kind of consistency the majority desires.

Finally, as noted above, the majority concedes that it is unable to locate a single court

holding that directly supports its novel application of RFRA to a suit between private parties.

Page 180: Sotomayor Cases

10 The majority cites dicta from district court opinions in Indiana and Arizona butconcedes that those courts “assumed” that RFRA could apply without analyzing the issue in anydepth. See Maj. Op. at 25-26 n.4.

12

See Maj. Op. at 25 n.4.10 This is telling, for Congress enacted RFRA over twelve years ago. The

plain language of the statute, its legislative history, and its interpretation by courts over the past

twelve years demonstrate that RFRA does not apply to suits between private parties.

C.

Even if appellees had not waived the RFRA defense, and even if RFRA applied to suits

between private parties, I would still find it unnecessary to reach the RFRA issue, or to analyze

the statute’s constitutionality, because Supreme Court and Second Circuit precedent compel the

conclusion that the ADEA does not apply to this dispute. Because the ADEA does not apply,

there is no “substantial burden” on religion, and RFRA, even if constitutional, is irrelevant.

In analyzing the ADEA’s applicability to this case, we find guidance in the principles

articulated by the Supreme Court in NLRB v. Catholic Bishop of Chicago, 440 U.S. 490 (1979).

To determine whether the National Labor Relations Act (NLRA) authorized the National Labor

Relations Board to regulate labor relations between a parochial school and its faculty, the

Catholic Bishop Court considered two principal questions. See id. at 501. First, it considered

whether this application of the NLRA raised First Amendment concerns. The Court concluded

that it did, explaining that judicial oversight of labor relations at a parochial school would risk

excessive entanglement between secular and religious authorities in violation of the

Establishment Clause. Id. at 501-04. Second, the Court examined whether Congress expressed

an intention to apply the statute to religious institutions despite these constitutional concerns.

Because the Court discerned no such congressional intent, it construed the NLRA in a manner

Page 181: Sotomayor Cases

11 The Court reached this conclusion even though the NLRA did not expressly includereligious institutions in its list of eight types of employers exempted from the act. See CatholicBishop, 440 U.S. at 511 (Brennan, J., dissenting) (citing 29 U.S.C. § 152(2)).

13

that avoided the constitutional difficulty, holding that the statute did not apply to labor disputes

between parochial schools and their employees.11 Id. at 504-07; see id. at 500 (citing the

longstanding principle that acts of Congress “ought not be construed to violate the Constitution if

any other possible construction remains available”) (citing Murray v. The Charming Betsy, 6

U.S. (2 Cranch) 64, 118 (1804)); see also Hsu By & Through Hsu v. Roslyn Union Free Sch.

Dist. No. 3, 85 F.3d 839, 854 (2d Cir. 1996) (noting this Court’s “consistent . . . practice of

avoiding constitutional questions where possible”).

Distinguishing Catholic Bishop, we concluded in DeMarco v. Holy Cross High Sch., 4

F.3d 166 (2d Cir. 1993), that the ADEA, unlike the NRLA, generally applies to religious

institutions. Id. at 172. Specifically, we held that a former lay teacher could bring an ADEA

action against a parochial school even though the teacher performed some religious duties. Id. at

168-72. In so holding, we observed that the ADEA, unlike the NRLA, does not pose the risk of

“extensive or continuous administrative or judicial intrusion into the functions of religious

institutions.” Id. at 170. Instead, the ADEA involves “‘routine regulatory interaction’” and

requires “‘no inquiries into religious doctrine, no delegation of state power to a religious body,

and no detailed monitoring [or] close administrative contact between secular and religious

bodies.’” Id. at 170 (quoting Hernandez v. Comm’r, 490 U.S. 680, 696-97 (1989) (internal

quotation marks omitted)); see also id. (“In age discrimination cases, the EEOC’s authority

extends only to the investigation and attempted conciliation or resolution of individual or group

complaints; it is limited in time and scope.” (citation and internal quotation marks omitted)).

Page 182: Sotomayor Cases

12 As discussed below, DeMarco also found the ADEA distinguishable from the NLRAbecause Congress clearly intended the ADEA to apply to religious institutions. See 4 F.3d at172.

14

These factors distinguished the ADEA from the NLRA.12

As a general rule, federal courts may decide civil disputes, including employment

discrimination disputes, between a religious institution and its employees without violating the

First Amendment. See Merkos L’Inyonei Chinuch, Inc. v. Otsar Sifrei Lubavitch, Inc., 312 F.3d

94, 99-100 (2d Cir. 2002) (citing Martinelli v. Bridgeport Roman Catholic Diocesan Corp., 196

F.3d 409, 431 (2d Cir. 1999); Gargano v. Diocese of Rockville Ctr., 80 F.3d 87, 90 (2d Cir.

1996); DeMarco, 4 F.3d at 172; cf. Employment Div., Dep’t of Human Res. v. Smith, 494 U.S.

872, 879 (1990) (“[T]he right of free exercise does not relieve an individual of the obligation to

comply with a valid and neutral law of general applicability on the ground that the law proscribes

(or prescribes) conduct that his [or her] religion prescribes (or proscribes).” (internal quotation

marks omitted)). The instant case, however, presents the more difficult question of whether this

general rule applies in the narrow context of a forced-retirement dispute between a religious body

and a member of its clergy.

As we noted in DeMarco, the relationship between a religious institution and certain of

its employees may be “so pervasively religious that it is impossible to engage in an age-

discrimination inquiry without serious risk of offending the Establishment Clause.” Id. at 172.

This risk is particularly serious in employment disputes between religious institutions and their

spiritual leaders where the enforcement of statutes like the ADEA might threaten the “power of

religious bodies to decide for themselves, free from state interference, matters of church

government as well as those of faith and doctrine.” Serbian E. Orthodox Diocese for the U.S. &

Page 183: Sotomayor Cases

15

Can. v. Milivojevich, 426 U.S. 696, 722 (1976) (internal quotation marks and alteration omitted).

“A church’s selection of its own clergy” is a “core matter of ecclesiastical self-governance” at the

“heart” of the church’s religious mission. Bollard v. Cal. Province of the Soc’y of Jesus, 196

F.3d 940, 946 (9th Cir. 1999). Federal court entanglement in matters as fundamental as a

religious institution’s selection or dismissal of its spiritual leaders risks an unconstitutional

“trespass[] on the most spiritually intimate grounds of a religious community’s existence.”

EEOC v. Roman Catholic Diocese of Raleigh, N.C., 213 F.3d 795, 800 (4th Cir. 2000).

In light of these serious constitutional concerns, we must ask whether Congress intended

to apply the ADEA to religious institutions in their selection of spiritual leaders. See Catholic

Bishop, 440 U.S. at 504. We concluded in DeMarco that Congress “implicitly expressed an

intention to apply the ADEA to religious institutions.” 4 F.3d at 172. We based this conclusion,

in part, on the ADEA’s similarity to Title VII of the Civil Rights Act, 42 U.S.C. § 2000e et seq.

“Given that Congress intended to apply Title VII to religious institutions, and that Congress

modelled the ADEA’s coverage upon that of Title VII,” we were “convinced that [Congress] also

intended to apply the ADEA to such institutions.” Id. at 173.

DeMarco, however, involved an employment dispute between a religious institution and a

math teacher who, despite having some religious duties, served primarily non-religious functions

in a parochial school. Here, in contrast, the dispute is between a minister with primarily religious

duties and a church that no longer wishes him to serve as pastor of a congregation. That

Congress intended the ADEA and Title VII to apply under the circumstances described in

DeMarco does not indicate an intention that those statutes should apply in all circumstances.

Nothing in the text, structure, or legislative history of the ADEA indicates an intention to extend

Page 184: Sotomayor Cases

13 This conclusion is consistent with the holdings of at least seven of our sister Circuits,which have adopted a limited “ministerial exception” that exempts religious institutions on FirstAmendment grounds from employment discrimination suits brought by clergy members or otheremployees serving primarily religious roles. See Roman Catholic Diocese of Raleigh, N.C., 213F.3d at 800, 805; Gellington v. Christian Methodist Episcopal Church, Inc., 203 F.3d 1299, 1304(11th Cir. 2000); Bollard, 196 F.3d at 949; Combs v. Central Tex. Annual Conf. of the UnitedMethodist Church, 173 F.3d 343, 351 (5th Cir. 1999); EEOC v. Catholic Univ. of Am., 83 F.3d455, 463 (D.C. Cir. 1996); Young v. N. Ill. Conf. of United Methodist Church, 21 F.3d 184, 187(7th Cir. 1994); Scharon v. St. Luke’s Episcopal Presbyterian Hosps., 929 F.2d 360, 363 (8thCir. 1991). Most circuits have reached the constitutional question directly and have held that theFirst Amendment bars adjudication of ministerial employment disputes. See, e.g., Gellington,203 F.3d at 1304; Combs; 173 F.3d at 351; Young, 21 F.3d at 187. Here, in contrast, I wouldapply Catholic Bishop’s principles of statutory construction so as to avoid making definitivepronouncements on the constitutional question. See 440 U.S. at 507; see also Scharon, 929 F.2dat 361-63 (applying the Catholic Bishop analysis to an employment discrimination action broughtby a priest). Despite this difference, my conclusion is substantially the same as that of otherCircuits: courts may not adjudicate employment discrimination lawsuits brought by clergymembers challenging a religious body’s refusal to select or retain them as spiritual leaders.

16

its provisions to a religious body’s selection or dismissal of its ministers. See Catholic Bishop,

440 U.S. at 504; DeMarco, 4 F.3d at 169, 172-73. Accordingly, I believe that the ADEA does

not apply to the case at bar.13 Because the ADEA does not apply, there is no substantial burden

on religion that could trigger RFRA.

The majority suggests that reliance on Catholic Bishop (and DeMarco) is unwarranted,

because “RFRA [is] the full expression of Congress’s intent with regard to the religion-related

issues before us and displace[s] earlier judge-made doctrines that might have been used to

ameliorate the ADEA’s impact on religious organizations and activities.” Maj. Op. at 8. Even if

RFRA applied to private suits and had not been waived in this case, I would disagree with the

majority’s suggestion that the statute completely displaces the Catholic Bishop analysis.

Although the Catholic Bishop rule and RFRA serve similar purposes, they require courts to

undertake different inquiries. See Univ. of Great Falls v. NLRB, 278 F.3d 1335, 1347 (D.C. Cir.

Page 185: Sotomayor Cases

14 I take no issue, however, with the analysis of the ADEA’s procedural requirements insection (a) of the majority’s opinion. See Maj. Op. at 5-8.

17

2002) (holding that the court need not address a university’s RFRA argument because the

university was entitled to an exemption under Catholic Bishop, and observing that “RFRA

presents a separate inquiry from Catholic Bishop”). Catholic Bishop requires courts, where

possible, to interpret statutes in ways that would avoid raising serious constitutional concerns. In

some cases, no such interpretation will be reasonably available. In those cases, RFRA may

provide an independent avenue both for protecting religious rights and for avoiding definitive

resolution of constitutional questions. Thus, RFRA should not be read to supplant the Catholic

Bishop inquiry, but to supplement it. Indeed, given that RFRA’s express purpose was to enhance

protection for religion, see 42 U.S.C. § 2000bb, it makes little sense to read the statute as

eliminating the protection afforded by the Catholic Bishop rule.

D.

I believe that a remand is a wasteful expenditure of judicial resources and an unnecessary

and uninvited burden on the parties. The district court is in no better position than we are to

decide either the statutory or constitutional questions presented in this case. In my view, the

most appropriate disposition of this case would be to affirm the district court’s dismissal of

appellant’s claims on the ground that the ADEA does not apply to employment suits brought

against religious institutions by their spiritual leaders. Because the majority’s contrary approach

disregards a clear and voluntary waiver, conflicts with RFRA’s text and with binding precedent,

and unnecessarily resolves a contested constitutional question, I respectfully dissent.14

Page 186: Sotomayor Cases

1

UNITED STATES COURT OF APPEALS12

FOR THE SECOND CIRCUIT34

August Term, 200456

(Argued: January 5, 2005 Decided: February 16, 2006)78

Docket No. 04-0743-cv9101112

JOHN PAUL HANKINS,1314

Plaintiff-Appellant,1516

v.1718

ERNEST S. LYGHT and NEW YORK ANNUAL CONFERENCE OF THE UNITED19METHODIST CHURCH,20

21Defendants-Appellees,22

23STONY BROOK COMMUNITY CHURCH,24

25Defendant.26

272829

B e f o r e: WINTER, SOTOMAYOR, and B.D. PARKER, Circuit30Judges.31

32Appeal from the dismissal of a minister’s age discrimination33

action against his church in the Eastern District of New York34

(Denis R. Hurley, Judge). We hold that the Religious Freedom35

Restoration Act of 1993 is constitutional as applied to federal36

law. It therefore amended the ADEA and governs the merits of37

this action. We vacate and remand for reconsideration in light38

of the RFRA.39

Judge Sotomayor dissents in a separate opinion.40

41

Page 187: Sotomayor Cases

2

BRUCE MILES SULLIVAN, Stony Brook,1New York, for Plaintiff-Appellant.2

3FREDERICK K. BREWINGTON, Hempstead,4New York, for Defendants-Appellees.5

6WINTER, Circuit Judge:7

8John Paul Hankins appeals from the dismissal by Judge Hurley9

of his age discrimination action. Hankins was a clergy member10

ordained by appellee New York Annual Conference of the United11

Methodist Church ("NYAC"). He was forced into retirement when he12

attained the age of 70. Appellee Ernest S. Lyght is the Bishop13

of the NYAC and has the power to appoint clergy to NYAC churches.14

Hankins claims that the NYAC’s mandatory retirement policy15

violates the Age Discrimination in Employment Act of 196716

("ADEA"), 29 U.S.C. § 621 et seq. We hold that the Religious17

Freedom Restoration Act of 1993 ("RFRA"), 42 U.S.C. § 2000bb et18

seq., is constitutional as applied to federal law; it therefore19

amended the ADEA and governs the merits of the principal issue20

raised by the parties. We vacate the dismissal of Hankins’21

complaint and remand for a determination of whether application22

of the ADEA to Hankins’ relationship with the NYAC and Lyght23

violates the RFRA.24

BACKGROUND25

We assume the existence of the facts as alleged in the26

complaint. Hankins was ordained by the NYAC and served as a27

clergy member from 1962 to July 1, 2003. He turned 70 on28

Page 188: Sotomayor Cases

3

November 5, 2002, and was forced into retirement on July 1,1

2003, as prescribed by paragraph 356 of the Methodist Book of2

Discipline. 3

According to a statement by the Methodist Church's Council4

of Bishops, the Book of Discipline is neither "sacrosanct" nor5

"infallible, but . . . is the most current statement of how6

United Methodists agree to live together" as "an inclusive7

society without regard to ethnic origin, economic condition,8

gender, age, or the disabilities of its constituents." The9

complaint alleges that the Book of Discipline contains "subject10

matters that are sectarian and ecclesiastical in nature[,] being11

related to the nature of the Deity and the Trinity, the12

scriptures, the tenets of the United Methodist Church, the13

theological grounding of biblical faith, the teachings of John14

Wesley and/or other religious principles or values (. . .15

‘religious considerations')," as well as "subject matters that16

are secular, temporal and/or civil in nature[,] not being17

determined, controlled or influenced by any religious18

considerations." The complaint further claims that paragraph19

356, under which Hankins was mandatorily retired, "is a secular,20

temporal, and/or civil subject matter, not being determined,21

controlled or influenced by any religious considerations." 22

Bishop Lyght told Hankins and other members of the Church23

that he had the authority to reappoint Hankins as pastor, despite24

Page 189: Sotomayor Cases

4

the fact that Hankins is over 70 years old. However, Bishop1

Lyght also stated that it is his "personal policy (as2

distinguished from the policy set forth in the Book of3

Discipline) never to reappoint members of the clergy who have4

attained age seventy to the church out of which they were5

retired." 6

Appellant brought an age discrimination charge to the Equal7

Employment Opportunity Commission ("EEOC") on March 19, 2003. 8

The EEOC issued a Notice of Right to Sue on April 11, 2003. 9

Appellant also filed a Verified Complaint with the New York10

Division of Human Rights on June 11, 2003; that Complaint was11

dismissed for administrative convenience on July 1, 2003. 12

Appellant filed the instant suit on July 3, 2003. 13

Appellant’s complaint claimed that the mandatory retirement14

policy violated the ADEA, the New York Human Rights Law, and the15

NYAC's covenant with him (Counts I, II, and IV); and that Bishop16

Lyght's personal policy against reappointing retired clergy17

violated the ADEA and Human Rights Law (Count III).1 18

Appellees moved to dismiss for lack of subject matter19

jurisdiction and for failure to state a claim upon which relief20

could be granted, under Rules 12(b)(1) and (6) respectively. The21

district court, ruling orally, declined to decide the 12(b)(1)22

motion, which was apparently based on deficiencies in the EEOC’s23

review of appellant’s charge. Instead, the court granted the24

Page 190: Sotomayor Cases

5

12(b)(6) motion based on a “ministerial exception” to the ADEA --1

a rule adopted by several circuits that civil rights laws cannot2

govern church employment relationships with ministers without3

violating the free exercise clause because they substantially4

burden religious freedom. See, e.g., McClure v. Salvation Army,5

460 F.2d 553, 560 (5th Cir. 1972) (applying Title VII to church-6

minister relationship "would result in an encroachment by the7

State into an area of religious freedom into which it is8

forbidden to enter" by the Free Exercise Clause). The court9

dismissed the complaint under Rule 12(b)(6). 10

DISCUSSION11

Appellant argues that the ministerial exception should not12

insulate a church’s non-religious regulations that discriminate13

against ministers on the basis of age. Appellees assert that14

this action is barred by EEOC errors. Alternatively, they15

continue to rely upon "the ministerial exception," the Free16

Exercise clause, and the Establishment Clause, claiming that17

applying the ADEA to the church-minister relationship would18

substantially burden religion. In that regard, appellees note19

that "for this very reason" Congress passed the RFRA. We address20

the alleged EEOC errors before turning to the main issue: 21

whether the RFRA amended the ADEA.22

a) Completion of Administrative Proceedings23

Appellees argue that the district court lacked jurisdiction24

Page 191: Sotomayor Cases

6

because the EEOC issued appellant’s Notice of Right to Sue fewer1

than sixty days after his charge was filed.2 We disagree. 2

Appellant satisfied all statutory requirements for bringing3

this private action under the ADEA. He filed an age4

discrimination charge with the EEOC on March 19, 2003; the EEOC5

issued a Notice of Right to Sue on April 11, 2003. Under 296

U.S.C. § 626(d) and (e), appellant had to file the instant suit7

more than sixty days after filing his EEOC complaint and within8

ninety days of his receipt of the EEOC Notice. Hankins complied9

with both requirements by filing suit on July 3, 2003 -- more10

than 60 days after March 19, and 83 days after April 11. 11

Furthermore, contrary to appellees’ arguments, the instant suit12

was not barred by appellant’s June 11, 2003 filing of a Complaint13

with the New York Division of Human Rights because the Division14

dismissed the complaint on July 1, 2003, before appellant filed15

this suit. See 29 U.S.C. § 633(b) (ADEA prohibits bringing suit16

before 60 days after commencement of state proceedings, "unless17

such proceedings have been earlier terminated"). 18

Appellees rely for their jurisdictional contention on two19

Title VII cases: Martini v. Fed. Nat’l Mortgage Ass’n, 178 F.3d20

1336 (D.C. Cir. 1999), and Rodriguez v. Connection Tech. Inc., 6521

F. Supp. 2d 107 (E.D.N.Y. 1999). These cases inferred from the22

language of 42 U.S.C. § 2000e-5(f)(1)3 that the EEOC lacks23

authority to issue right-to-sue notices based on Title VII claims24

Page 192: Sotomayor Cases

7

before 180 days after a charge is filed. E.g., Martini, 178 F.3d1

at 1347 (“[T]he EEOC’s power to authorize private suits within2

180 days undermines its express statutory duty to investigate3

every charge filed, as well as Congress’s unambiguous policy of4

encouraging informal resolution of charges up to the 180th5

day.”). We have not decided whether the regulation allowing6

early issuance of right-to-sue notices, 29 C.F.R. §7

1601.28(a)(2), is a permissible construction of Section 2000e-5. 8

We express no opinion on the issue here, although we note that9

two circuits and several district courts within this circuit have10

disagreed with Martini and Rodriguez. Sims v. Trus Joist11

MacMillan, 22 F.3d 1059, 1061-63 (11th Cir. 1994) (early issuance12

of right-to-sue letter by EEOC does not bar a Title VII suit);13

Saulsbury v. Wismer & Becker, Inc., 644 F.2d 1251, 1257 (9th Cir.14

1980) (same); Commodari v. Long Island Univ., 89 F. Supp. 2d 353,15

381-83 (E.D.N.Y. 2000) (same); Palumbo v. Lufthansa German16

Airlines, 1999 U.S. Dist. LEXIS 11412, No. 98 Civ. 5005, 1999 WL17

540446, at *2 (S.D.N.Y. July 26, 1999) (same); Figueira v. Black18

Entm’t Television, Inc., 944 F. Supp. 299, 303-08 (S.D.N.Y. 1996)19

(same).20

The key fact in the present matter is that the language of21

29 U.S.C. § 626, which authorizes suits under the ADEA, differs22

significantly from that of Section 2000e-5(f)(1). Section 62623

provides that “[n]o civil action may be commenced by an24

Page 193: Sotomayor Cases

8

individual under this section until 60 days after a charge1

alleging unlawful discrimination has been filed with the [EEOC]." 2

Id. § 626(d). Appellant complied with this provision by waiting3

sixty days after filing his EEOC charge before bringing the4

instant suit. The fact that the EEOC terminated its proceedings5

prior to the expiration of sixty days was irrelevant to the6

district court’s authority to entertain the case. This is7

especially so because Section 626, unlike Section 2000e-5,8

explicitly contemplates early termination of EEOC investigations. 9

Id. § 626(e) (“If a charge filed with the [EEOC] under this10

chapter is dismissed or the proceedings of the [EEOC] are11

otherwise terminated by the [EEOC], the [EEOC] shall notify the12

person aggrieved.”). This suit was therefore properly before the13

district court.14

b) The Religious Freedom Restoration Act15

In our view, the dispositive issue in this matter concerns16

the application of the RFRA. The statute's substantive17

provisions state:18

(a) In general. Government shall not substantially19burden a person's exercise of religion even if the20burden results from a rule of general applicability,21except as provided in subsection (b).22(b) Exception. Government may substantially burden a23person's exercise of religion only if it demonstrates24that application of the burden to the person--25(1) is in furtherance of a compelling governmental26interest; and27(2) is the least restrictive means of furthering that28compelling governmental interest.29

30

Page 194: Sotomayor Cases

9

42 U.S.C. § 2000bb-1.1

The test set out in Subsection (b)(1) and (2) “applies to2

all Federal law, and the implementation of that law, whether3

statutory or otherwise, and whether adopted before or after4

November 16, 1993.” Id. § 2000bb-3(a). The RFRA's remedial5

provision states that “[a] person whose religious exercise has6

been burdened in violation of this section may assert that7

violation as a claim or defense in a judicial proceeding and8

obtain appropriate relief against a government.” Id. §9

2000bb-1(c). "[G]overnment" is in turn defined to include any10

"branch, department, agency, instrumentality, and official (or11

other person acting under color of law) of the United States." 12

Id. § 2000bb-2(1). 13

The present action is a suit against a church and an14

official of that church. The suit claims that the defendants15

violated a federal statute, the ADEA, and seeks judicial16

remedies; appellees claim that application of the statute would17

substantially burden the exercise of their religion. If the18

RFRA's test for evaluating burdens on religious activity --19

Subsections (b)(1) and (2) -- is not met, appellees can arguably20

assert a violation of the RFRA as a complete defense. 21

The district court dismissed the case based on a22

"ministerial exception" that some courts had read into various23

anti-discrimination laws -- an unresolved issue in this circuit -24

Page 195: Sotomayor Cases

10

- including the ADEA. Whatever the merits of that exception as1

statutory interpretation or policy, it has no basis in statutory2

text, whereas the RFRA, if applicable, is explicit legislation3

that could not be more on point. Given the absence of other4

relevant statutory language, the RFRA must be deemed the full5

expression of Congress’s intent with regard to the religion-6

related issues before us and displace earlier judge-made7

doctrines that might have been used to ameliorate the ADEA’s8

impact on religious organizations and activities. City of9

Milwaukee v. Illinois, 451 U.S. 304, 314 (1981) ("Federal common10

law is a necessary expedient, and when Congress addresses a11

question previously governed by a decision rested on federal12

common law the need for such an unusual exercise of lawmaking by13

federal courts disappears.") (internal quotation marks and14

citations omitted).15

There is little caselaw addressing the issue whether the16

RFRA applies to an action by a private party seeking relief under17

a federal statute against another private party who claims that18

the federal statute substantially burdens his or her exercise of19

religion.4 The RFRA's language surely seems broad enough to20

encompass such a case. The statutory language states that it21

"applies to all federal law, and the implementation of that law,"22

42 U.S.C. § 2000bb-3(a), and that a defendant arguing that such a23

law substantially burdens the exercise of religion "may assert [a24

Page 196: Sotomayor Cases

11

violation of the RFRA] as a . . . defense in a judicial1

proceeding." Id. § 2000bb-1(c). This language easily covers2

the present action. The only conceivably narrowing language is3

the phrase immediately following: "and obtain appropriate relief4

against a government." Id. However, this language would seem5

most reasonably read as broadening, rather than narrowing, the6

rights of a party asserting the RFRA. The narrowing7

interpretation -- permitting the assertion of the RFRA as a8

defense only when relief is also sought against a governmental9

party -- involves a convoluted drawing of a hardly inevitable10

negative implication. If such a limitation was intended,11

Congress chose a most awkward way of inserting it. The12

legislative history is neither directly helpful nor harmful to13

that view.14

We need not, however, decide whether the RFRA applies to a15

federal law enforceable only in private actions between private16

parties. The ADEA is enforceable by the EEOC as well as private17

plaintiffs, and the substance of the ADEA's prohibitions cannot18

change depending on whether it is enforced by the EEOC or an19

aggrieved private party. See United States v. Brown, 79 F.3d20

1550, 1559 n.16 (11th Cir. 1996) ("The meaning of the statutory21

words 'scheme to defraud' does not change depending on whether22

the case is Civil RICO or criminal."). An action brought by an23

agency such as the EEOC is clearly one in which the RFRA may be24

Page 197: Sotomayor Cases

12

asserted as a defense, and no policy of either the RFRA or the1

ADEA should tempt a court to render a different decision on the2

merits in a case such as the present one. Indeed, appellant3

argues that the RFRA is inapplicable only because it is4

unconstitutional.5

1. Waiver6

First, however, we must address whether appellees have7

waived or forfeited reliance upon the RFRA. In their original8

brief, as noted, appellees argued that the ADEA was an unlawful9

burden on their religious activities and that Congress has10

enacted the RFRA, a statute that applied to all federal laws,11

"for this very reason." Appellant’s Brief at 28. Believing that12

this reference to a seemingly dispositive but otherwise13

unmentioned statute needed some elaboration and unconvinced that14

appellant's claim that the Supreme Court had held the RFRA15

unconstitutional in all circumstances was correct, we asked for16

further briefing.17

Somewhat to our surprise, appellees’ post-argument letter-18

brief states that, although all pertinent portions of the RFRA19

are constitutional, the statute is inapplicable because “the case20

at bar is a matter relating to a private employment situation and21

does not involve actions by the government.” Nevertheless,22

appellees continue to rely upon the "ministerial exception" and23

the Free Exercise and Establishment Clauses.24

Page 198: Sotomayor Cases

13

In our view, as discussed above, the RFRA’s provisions are1

directly on point, and allow parties who, like appellees, claim2

that a federal statute, like the ADEA, substantially burdens the3

exercise of their religion to assert the RFRA as a defense to any4

action asserting a claim based on the ADEA. The issue then is5

whether their post-argument letter-brief constitutes a waiver or6

forfeiture of that defense.7

A party may certainly waive or forfeit a RFRA defense by8

failing to argue that a law or action substantially burdens the9

party’s religion. For example, in United States v. Amer,10

appellant had forfeited the defense that his child kidnaping11

conviction violated the RFRA, because “[a]t no point during the12

pretrial, trial, or sentencing proceedings did [appellant] argue13

that his act of removing and retaining the children was14

religiously mandated or inspired.” 110 F.3d 873, 879 & n.1 (2d15

Cir. 1997). Where a party fails to assert a substantial burden16

on religious exercise before a district court, therefore, the17

party may not raise that issue -- an inherently fact-based one --18

for the first time on appeal.19

However, appellees argued in the district court and here --20

and continue to argue -- that application of the ADEA to the21

relationship between their church and appellant substantially22

burdens their religion. They continue to assert the "ministerial23

exception," which in their view tracks the Free Exercise clause24

Page 199: Sotomayor Cases

14

of the Constitution and the Establishment Clause as well. 1

Appellees’ Brief at 4-15; see Elvig v. Calvin Presbyterian2

Church, 397 F.3d 790, 790 (9th Cir. 2005) ("[T]he 'ministerial3

exception' to Title VII is carved out from the statute based on4

the commands of the Free Exercise and Establishment Clauses of5

the First Amendment."). In substance, therefore, they ask us to6

apply the RFRA, but not to mention it.7

Appellees' position that the RFRA does not apply to suits8

between private parties is not determinative of our analysis,9

given that they have vigorously pursued and preserved the10

substance of the issue. We are required to interpret federal11

statutes as they are written -- in this case the ADEA as amended12

by the RFRA -- and we are not bound by parties’ stipulations of13

law. Becker v. Poling Transp. Corp., 356 F.3d 381, 390 (2d Cir.14

2004); see also Kamen v. Kemper Fin. Servs. Inc., 500 U.S. 90, 9915

("When an issue or claim is properly before the court, the court16

is not limited to the particular legal theories advanced by the17

parties, but rather retains the independent power to identify and18

apply the proper construction of governing law."). We are not in19

the business of deciding cases according to hypothetical legal20

schemes, particularly when the hypothetical scheme posed by a21

party tracks the actual law in all but name.22

2. Constitutionality23

In addressing the constitutional issues raised by appellant24

Page 200: Sotomayor Cases

15

with regard to the RFRA, we first describe the statutory1

background.2

The RFRA was passed in response to Employment Div. v. Smith,3

494 U.S. 872 (1990). The Supreme Court held there that "the4

right of free exercise does not relieve an individual of the5

obligation to comply with a valid and neutral law of general6

applicability on the ground that the law proscribes (or7

prescribes) conduct that his religion prescribes (or8

proscribes)." Id. at 879 (internal quotation marks and citation9

omitted). Smith limited the applicability of the "compelling10

state interest" test the Court had previously applied to neutral11

laws before allowing them to place a substantial burden on12

religious practice. Id. at 883-84 (limiting test to mean that13

"where the State has in place a system of individual exemptions,14

it may not refuse to extend that system to cases of ‘religious15

hardship' without compelling reason").516

Congress enacted the RFRA pursuant to two sources of17

authority, Section 5 of the Fourteenth Amendment and the18

Necessary and Proper Clause of the Constitution. See H.R. Rep.19

No. 103-88, at 17 (1993) (“Finally, the Committee believes that20

Congress has the constitutional authority to enact [the RFRA]. 21

Pursuant to Section 5 of the Fourteenth Amendment and the22

Necessary and Proper Clause of the Constitution, the legislative23

branch has been given the authority to provide statutory24

Page 201: Sotomayor Cases

16

protection for a constitutional value . . . .”). The Supreme1

Court held that the RFRA could not be enacted under Section 5 of2

the Fourteenth Amendment, which empowers Congress to enforce the3

Amendment's other provisions against the states. City of Boerne4

v. Flores, 521 U.S. 507, 519 (1997) (“Congress does not enforce a5

constitutional right by changing what that right is.”). The RFRA6

is therefore unconstitutional as applied to state law.7

However, the RFRA applies by its terms not only to the8

states but also to “all Federal law, and the implementation of9

that law, whether statutory or otherwise, and whether adopted10

before or after November 16, 1993.” 42 U.S.C. § 2000bb-3(a); see11

also id. § 2000bb-2(1) (“‘[G]overnment’ includes a branch,12

department, agency, instrumentality, and official (or other13

person acting under color of law) of the United States.”). 14

Boerne could not have addressed whether the RFRA was validly15

enacted under the Necessary and Proper Clause because the only16

issue before the Court was the denial of a building permit to a17

church by local zoning authorities. 521 U.S. at 512. Since18

Boerne, “[e]very appellate court that has squarely addressed the19

question has held that the RFRA governs the activities of federal20

officers and agencies." O'Bryan v. Bureau of Prisons, 349 F.3d21

399, 401 (7th Cir. 2003); Guam v. Guerrero, 290 F.3d 1210, 122122

(9th Cir. 2002); Henderson v. Kennedy, 265 F.3d 1072, 1073 (D.C.23

Cir. 2001); Kikumura v. Hurley, 242 F.3d 950, 960 (10th Cir.24

Page 202: Sotomayor Cases

17

2001); Christians v. Crystal Evangelical Free Church (In re1

Young), 141 F.3d 854, 856 (8th Cir. 1998); see also Madison v.2

Riter, 355 F.3d 310, 315 (4th Cir. 2003). 3

We join the other circuits in holding that the RFRA is4

constitutional as applied to federal law under the Necessary and5

Proper Clause of the Constitution. As presented in this case,6

the issue is simply whether Congress had the authority to amend7

the ADEA to include the RFRA standard. See In re Young, 141 F.3d8

at 861 (the RFRA “has effectively amended the Bankruptcy Code,9

and has engrafted the additional clause to § 548(a)(2)(A) that a10

recovery that places a substantial burden on a debtor's exercise11

of religion will not be allowed unless it is the least12

restrictive means to satisfy a compelling governmental13

interest.”).14

Congress enacted the ADEA pursuant to its Commerce Clause15

powers under Article I. Kimel v. Fla. Bd. of Regents, 528 U.S.16

62, 78 (2000) (“the ADEA constitutes a valid exercise of17

Congress' power ‘[t]o regulate Commerce . . . among the several18

States’”) (citing EEOC v. Wyoming, 460 U.S. 226, 243 (1983))19

(alterations in original); McGinty v. New York, 251 F.3d 84, 9120

(2d Cir. 2001); see U.S. Const., Art. I, § 8, cl. 3 (“The21

Congress shall have power . . . [t]o regulate commerce with22

foreign Nations, and among the several States, and with the23

Indian Tribes."). Furthermore, the Necessary and Proper Clause24

Page 203: Sotomayor Cases

18

authorizes Congress “[t]o make all Laws which shall be necessary1

and Proper for carrying into Execution” its Article I powers,2

including its Commerce Clause powers. U.S. Const. art. I, § 8,3

cl. 18. The Clause allows all legitimate legislation “plainly4

adapted” to a constitutional end. M’Culloch v. Maryland, 17 U.S.5

(4 Wheat.) 316, 421 (1819) (“Let the end be legitimate, let it be6

within the scope of the constitution, and all means which are7

appropriate, which are plainly adapted to that end, which are not8

prohibited, but consist with the letter and spirit of the9

constitution, are constitutional.”). Finally, the "plainly10

adapted" standard requires only “that the effectuating11

legislation bear a rational relationship to a permissible12

constitutional end.” United States v. Wang Kun Lue, 134 F.3d 79,13

84 (2d Cir. 1998).14

It is obvious to us that because Congress had the power to15

enact the ADEA, it also had the power to amend that statute by16

passing the RFRA. The RFRA was authorized by the Necessary and17

Proper Clause because its purpose -- to protect First Amendment18

rights as interpreted by the Congress, see S. Rep. No. 103-111,19

at 14 (1993), reprinted in 1993 U.S.C.C.A.N. 1892, 1903 -- was20

permissible. “When Congress acts within its sphere of power and21

responsibilities, it has not just the right but the duty to make22

its own informed judgment on the meaning and force of the23

Constitution.” Boerne, 521 U.S. at 535. 24

Page 204: Sotomayor Cases

19

The RFRA was also proper as applied to the ADEA in1

particular because, as noted, Congress had authority to enact2

that statute under the Commerce Clause. See INS v. Chadha, 4623

U.S. 919, 941 (1983) ("’Congress has plenary authority in all4

cases in which it has substantive legislative jurisdiction, so5

long as the exercise of that authority does not offend some other6

constitutional restriction.’”) (quoting Buckley v. Valeo, 4247

U.S. 1, 132 (1976)) (internal citation omitted); Guerrero, 2908

F.3d at 1220 (“Congress derives its ability to protect the free9

exercise of religion from its plenary authority found in Article10

I of the Constitution; it can carve out a religious exemption11

from otherwise neutral, generally applicable laws based on its12

power to enact the underlying statute in the first place.”); In13

re Young, 141 F.3d at 861 (“[W]e can conceive of no argument to14

support the contention[] that Congress is incapable of amending15

the legislation that it has passed.”).616

In his post-argument letter-brief, appellant argues that17

application of the RFRA to federal law violates separation of18

powers principles and the Establishment Clause of the19

Constitution.7 We address these issues in turn.20

Appellant's separation of powers challenge is that because21

the RFRA mandates evaluation of laws and actions that burden22

religion by a standard different from that prescribed by the23

Supreme Court, it is a Congressional usurpation of judicial24

Page 205: Sotomayor Cases

20

power. However, we agree with the Eighth Circuit that “[t]he key1

to the separation of powers issue in this case is . . . not2

whether Congress disagreed with the Supreme Court’s3

constitutional analysis, but whether Congress acted beyond the4

scope of its constitutional authority in applying RFRA to federal5

law.” In re Young, 141 F.3d at 860; United States v. Marengo6

County Comm'n, 731 F.2d 1546, 1562 (11th Cir. 1984)7

(“[C]ongressional disapproval of a Supreme Court decision does8

not impair the power of Congress to legislate a different result,9

as long as Congress had that power in the first place.”). 10

Indeed, “Congress has often provided statutory protection of11

individual liberties that exceed the Supreme Court’s12

interpretation of constitutional protection.” In re Young, 14113

F.3d at 860 (collecting examples); Guerrero, 290 F.3d at 122114

(“Certainly Congress can provide more individual liberties in the15

federal realm than the Constitution requires without violating16

vital separation of powers principles.”). That the RFRA provides17

more protection from federal actors and statutes than may be18

required by the First Amendment hardly undermines separation of19

powers principles.20

With respect to appellant's Establishment Clause argument,21

the Clause provides that "Congress shall make no law respecting22

an establishment of religion." U.S. Const. amend. I. The23

Supreme Court has established a three-prong test to determine24

Page 206: Sotomayor Cases

21

whether a statute violates the Clause. 1

First, the statute must have a secular legislative2purpose; second, its principal or primary effect must3be one that neither advances nor inhibits religion;4finally, the statute must not foster an excessive5government entanglement with religion. 6

7

Lemon v. Kurtzman, 403 U.S. 602, 612-13 (1971) (quotations and8

citations omitted). Applying this test, the Court has held that9

exempting religious organizations from compliance with neutral10

laws does not violate the Constitution. E.g., Corp. of the11

Presiding Bishop of the Church of Jesus Christ of Later-Day12

Saints v. Amos, 483 U.S. 327, 338-40 (1987) (exemption from13

federal antidiscrimination laws for religious organizations does14

not violate Establishment Clause); see also Gillette v. United15

States, 401 U.S. 437, 460 (1971) (exemption from military draft16

for religious conscientious objectors does not violate17

Establishment Clause); Walz v. Tax Comm'n, 397 U.S. 664, 68018

(1970) (state property tax exemption for religious organizations19

does not violate Establishment Clause).20

Given these holdings, appellant faces an unwinnable battle21

in claiming that the RFRA -- a limited exemption for religious22

organizations from compliance with neutral laws -- violates the23

Establishment Clause. The RFRA had a secular legislative purpose24

within the meaning of Lemon -- namely, to protect individual25

First Amendment rights as interpreted by the Congress. As noted,26

this purpose was not only permissible but was also required by27

Page 207: Sotomayor Cases

22

Congress’s duty to interpret the Constitution. Boerne, 521 U.S.1

at 535. A “secular legislative purpose” need not be “unrelated2

to religion”; rather, Lemon’s first prong aims to prevent3

Congress “from abandoning neutrality and acting with the intent4

of promoting a particular point of view in religious matters.” 5

Amos, 483 U.S. at 335; Gillette, 401 U.S. at 454 ("'Neutrality'6

in matters of religion is not inconsistent with 'benevolence' by7

way of exemptions from onerous duties, so long as an exemption is8

tailored broadly enough that it reflects valid secular9

purposes.") (citation omitted). The RFRA reflected no purpose to10

promote a particular religious point of view.11

The RFRA also satisfies the other two prongs of the Lemon12

test. Its principal effect neither advances nor inhibits13

religion within the meaning of Lemon. “For a law to have14

forbidden ‘effects’ under Lemon, it must be fair to say that the15

government itself has advanced religion through its own16

activities and influence,” rather than simply by granting an17

exemption to religious organizations. Amos, 483 U.S. at 337-3818

(“Where . . . government acts with the proper purpose of lifting19

a regulation that burdens the exercise of religion, we see no20

reason to require that the exemption come packaged with benefits21

to secular entities.”). Although the RFRA certainly provides22

some benefit to religious organizations, “a law is not23

unconstitutional simply because it allows churches to advance24

Page 208: Sotomayor Cases

23

religion, which is their very purpose.” Id. at 337. Finally,1

there is no question that the RFRA decreases rather than fosters2

government entanglement with religion, as required by the third3

prong of Lemon. Amos, 483 U.S. at 339 (An exemption “effectuates4

a more complete separation of [church and state] and avoids . . .5

intrusive inquiry into religious belief.”).6

We note in general that the Supreme Court approved of and7

invited legislative enactments of religious exceptions to neutral8

laws in Smith itself. 494 U.S. at 890. The court pointed to9

state exceptions to drug laws for sacramental peyote use and10

noted with approval that “a society that believes in the negative11

protection accorded to religious belief can be expected to be12

solicitous of that value in its legislation as well.” Id. 13

(“[T]o say that a nondiscriminatory religious-practice exemption14

is permitted, or even that it is desirable, is not to say that it15

is constitutionally required.”). We therefore hold that the16

RFRA, as applicable to federal law, does not violate the17

Establishment Clause of the Constitution.18

Having found the portions of the RFRA applicable to the19

federal government and federal law constitutional, we have little20

difficulty finding those portions severable from the RFRA’s21

unconstitutional sections. A court must sever the invalid parts22

of a statute from the valid parts “unless it is evident that the23

Legislature would not have enacted those provisions which are24

Page 209: Sotomayor Cases

24

within its power, independently of that which is not.” Chadha,1

462 U.S. at 931-32 (internal quotation marks, citations, and2

alterations omitted); Alaska Airlines, Inc. v. Brock, 480 U.S.3

678, 684 (1987) (“A court should refrain from invalidating more4

of the statute than is necessary.”) (alteration omitted). We5

know of no evidence that Congress would not have applied the RFRA6

to the federal government unless it could also be applied to7

state and local governments. We therefore hold the portion of8

the RFRA applicable to the federal government severable from its9

unconstitutional portions. See Kikumura, 242 F.3d at 959-6010

(finding federal portions of the RFRA severable); In re Young,11

141 F.3d at 859 (same).12

CONCLUSION13

The RFRA is an amendment to the ADEA and, as such, is14

constitutional. The parties have not briefed the issue of how it15

impacts the merits of this case. The district court did not16

apply the RFRA, relying instead on the “ministerial exception” to17

the ADEA. We believe that, while the RFRA's application is a18

matter of law, it would be appropriate to hear from the district19

court first, rather than seek yet further briefing in this court. 20

We therefore vacate and remand for reconsideration under the21

RFRA standards.22

23

24

Page 210: Sotomayor Cases

25

1. Appellant initially moved for a preliminary injunction

requiring appellees to restore his active status, but he withdrew

the motion after the NYAC and Lyght appointed another clergy

member to fill his vacant position.

2. The district court did not address this issue, but because it

raises purely legal questions, we do so here. See McGinty v. New

York, 251 F.3d 84, 90 (2d Cir. 2001) (addressing question not

decided by district court where facts were undisputed and legal

question was briefed).

3. Section 2000e-5(f)(1) provides in pertinent part:

If a charge filed with the Commission . . . isdismissed by the Commission, or if within one hundredand eighty days from the filing of such charge . . .the Commission has not filed a civil action . . . orthe Commission has not entered into a conciliationagreement to which the person aggrieved is a party, theCommission . . . shall so notify the person aggrievedand within ninety days after the giving of such noticea civil action may be brought against the respondentnamed in the charge. . . .

4. No court appears to have addressed the issue squarely, but

some suggestive caselaw exists. Some courts seem to have assumed

FOOTNOTES

Page 211: Sotomayor Cases

26

without discussion that the RFRA may be asserted as a defense by

a private party against another private party. See, e.g., Guinan

v. Roman Catholic Archdiocese of Indianapolis, 42 F. Supp. 2d

849, 853 (S.D. Ind. 1998) (permitting the private party defendant

to assert a RFRA defense but rejecting it after first finding

that the ministerial exception negated the need to discuss the

RFRA defense); Urantia Found. v. Maaherra, 895 F. Supp. 1335,

1336-37 (D. Ariz. 1995) (permitting the defendant to raise a RFRA

defense but rejecting it because the defendant did not contest

the constitutionality of the trademark and copyright laws in

general or as applied to her). Bankruptcy courts have also

generally permitted a private-party defendant to assert a RFRA

defense against a Chapter 7 trustee. See Christians v. Crystal

Evangelical Free Church (In re Young), 82 F.3d 1407, 1418-19 (8th

Cir. 1996) (permitting a defendant to assert a RFRA defense and

recover debtors’ tithes to the church because “the government

action in question meaningfully curtails, albeit retroactively, a

religious practice”), vacated, 521 U.S. 1114 (1997), reaff’d, 141

F.3d 854 (8th Cir. 1998); see also In re Tessier, 190 B.R. 396

(Bankr. D. Mont. 1995); Newman v. Midway Southern Baptist Church

(In re Newman), 183 B.R. 239 (Bankr. D. Kan. 1995), aff’d, 203

B.R. 468 (D. Kan. 1996). A bankruptcy trustee is arguably

"acting under color of law" and therefore falls within the RFRA's

definition of "government." 42 U.S.C. § 2000bb-2(1). United

Page 212: Sotomayor Cases

27

States trustees are part of the executive branch and protect the

interests of the United States in the liquidation. See 28 U.S.C.

§ 586(a); 11 U.S.C. §§ 701(a)(1), 703(b)-(c) and 704(9); In re

Shoenewerk, 304 B.R. 59, 62-63 (Bankr. E.D.N.Y. 2003).

5. The RFRA’s stated purposes include "restor[ing] the compelling

interest test as set forth in Sherbert v. Verner, 374 U.S. 398

(1963) and Wisconsin v. Yoder, 406 U.S. 205 (1972)." 42 U.S.C. §

2000bb(b)(1). The Supreme Court noted that "Congress enacted

RFRA in direct response to the Court’s decision in" Smith. City

of Boerne v. Flores, 521 U.S. 507, 512 (1997).

6. We find no principled constitutional distinction between

Congress’s ability to amend statutes on an individual basis and

its power to do so in a wholesale manner through an enactment

such as the RFRA. See Guerrero, 290 F.3d at 1221 n.18.

7. Appellant also argues that Boerne explicitly invalidated all

of the RFRA due to separation of powers concerns. Specifically,

appellant relies upon the statement that the “RFRA contradicts

vital principles necessary to maintain separation of powers and

the federal balance.” Boerne, 521 U.S. at 536. The argument is

entirely unconvincing. The quoted language simply explained why

Congress could not enact the RFRA pursuant to its Section 5

Page 213: Sotomayor Cases

28

power. The quoted phrase reads in full as follows: “Broad as

the power of Congress is under the Enforcement Clause of the

Fourteenth Amendment, RFRA contradicts vital principles necessary

to maintain separation of powers and the federal balance.” Id.

This analysis has no application to any separation of powers

concerns raised by the RFRA’s enactment and application to the

federal government under the Necessary and Proper Clause. See

Guerrero, 290 F.3d at 1220 (Boerne’s “discussion of the

separation of powers doctrine was entirely within the framework

of its section 5 analysis -- not an independent rationale.”).

Page 214: Sotomayor Cases

1We direct the Clerk of Court to change the official caption to comport with this

decision.

1

04-5711(L)-cv, 04-5943-cvWalczyk v. Rio

UNITED STATES COURT OF APPEALS

FOR THE SECOND CIRCUIT

August Term, 2005

(Argued: April 3, 2006 Decided: August 1, 2007)

Docket Nos. 04-5711-cv(L), 04-5943-cv(XAP)

THOMAS WALCZYK, ELIZABETH WALCZYK, MAXIMINA WALCZYK, EACH

INDIVIDUALLY AND AS P.P.A. FOR MICHELLE WALCZYK, A MINOR CHILD,1

Plaintiffs-Appellees-Cross-Appellants,

—v.—

JAMES RIO, BRIAN KILLIANY, JAMES JEPSEN, WILLIAM TYLER,

ANGELA DESCHENES, AND SHAWN BROWN,

Defendants-Appellants-Cross-Appellees.

Before:

CABRANES, SOTOMAYOR, and RAGGI, Circuit Judges.

Interlocutory appeal from so much of an order of the United States District Court for

the District of Connecticut as (1) denied defendants qualified immunity on plaintiffs’ federal

Page 215: Sotomayor Cases

2

and state constitutional challenges to the execution of arrest and search warrants. Cross-

appeal from so much of the order as (2) denied plaintiff Elizabeth Walczyk summary

judgment on the liability element of her illegal search claim and (3) granted defendants

summary judgment on Thomas Walczyk’s claim of excessive bail.

REVERSED AND REMANDED on part of defendants’ qualified immunity appeal.

AFFIRMED in all other respects.

Judge Sotomayor concurs in a separate opinion.

THOMAS R. GERARDE (John J. Radshaw, III, on the brief), Howd & Ludorf,

LLC, Hartford, Connecticut, for Defendants-Appellants-Cross-

Appellees.

JON L. SCHOENHORN (Jennifer L. Bourn, on the brief), Jon L. Schoenhorn &

Associates, Hartford, Connecticut, for Plaintiffs-Appellees-Cross-

Appellants.

REENA RAGGI, Circuit Judge:

In 2001, plaintiff Thomas Walczyk (“Walczyk”) was convicted after a jury trial in

Connecticut on state law charges of disorderly conduct, see Conn. Gen. Stat. § 53a-182(a)(2);

reckless endangerment, see id. § 53a-64(a); and improper firearm storage, see id. § 29-37i.

On appeal, the Connecticut Appellate Court reversed, holding that Walczyk’s conviction

violated federal and state law because it was based on incriminating evidence obtained

through search warrants that were not supported by probable cause. See State v. Walczyk,

76 Conn. App. 169, 182, 818 A.2d 868, 876 (Conn. App. Ct. 2003). Thereafter, Walczyk,

Page 216: Sotomayor Cases

3

his wife Maximina, his minor child Michelle, and his mother Elizabeth initiated this civil

action, suing defendants, all members of the Farmington, Connecticut Police Department,

in the United States District Court for the District of Connecticut (Robert N. Chatigny, Chief

Judge), pursuant to 42 U.S.C. §§ 1983 and 1988 and Connecticut law for money damages

arising from events relating to Walczyk’s reversed conviction.

Although the district court granted defendants’ motion for summary judgment with

respect to some of plaintiffs’ claims, defendants now pursue an interlocutory appeal from so

much of the district court’s order, entered on September 30, 2004, as denied them qualified

immunity from plaintiffs’ unlawful arrest and search claims. See Walczyk v. Rio, 339 F.

Supp. 2d 385, 389-91 (D. Conn. 2004). Not surprisingly, plaintiffs defend that denial. At

the same time, Elizabeth Walczyk cross-appeals the district court’s denial of her motion for

summary judgment on the liability element of her challenge to the search of her home. See

id. at 391. Meanwhile, Thomas Walczyk cross-appeals the award of summary judgment to

defendants on his Eighth Amendment claim that he was detained on excessive bail. See id.

at 390.

For the reasons discussed herein, we conclude that the arrest of Thomas Walczyk and

the search of the home he shared with his wife and daughter were supported by probable

cause. We reverse so much of the district court’s order as concluded otherwise, and we

remand with directions to enter summary judgment in favor of defendants on that part of

plaintiffs’ complaint. With respect to defendants’ search of Elizabeth Walczyk’s home, we

Page 217: Sotomayor Cases

2Lucien Walczyk is not a party to this lawsuit, having died prior to its commencement.

4

affirm the district court’s denial of qualified immunity to defendants because the warrant

authorizing that search was procured on the basis of plainly stale information and questions

of fact remain as to whether any or all defendants acted knowingly or recklessly in

misleading the issuing magistrate as to the currency of that information. Those same

questions of fact prompt us to affirm the district court’s denial of summary judgment to

Elizabeth Walczyk on the liability element of her unlawful search claim. Finally, with

respect to Walczyk’s excessive bail claim, we affirm the award of summary judgment in

favor of defendants on the ground of absolute immunity.

I. Factual Background

A. The Underlying Land Dispute

1. Barberino Realty Acquires the Land

This case has its origins in a longstanding property dispute between the Walczyk

family and Barberino Realty and Development Corporation (“Barberino”). The property —

undeveloped land adjacent to the Farmington homes of Thomas Walczyk at 28 Tunxis Street

and of his parents, Elizabeth and Lucien Walczyk,2 at 27 Tunxis Street — was acquired by

Barberino in 1973. Over the next two decades, Barberino encountered various difficulties

developing the land, only some of which are relevant to this appeal.

2. 1981: Walczyk Brandishes a Rifle at Barberino Workers

Sometime in 1981, a Barberino work crew entered onto the undeveloped land to drill

Page 218: Sotomayor Cases

5

for soil samples preliminary to development. Thomas Walczyk, who was licensed to possess

numerous firearms, brandished a rifle at the workmen and challenged their actions. The

workmen sought police assistance, after which their work proceeded apparently without

interruption and without any official action being taken against Walczyk.

The incident nevertheless prompted Barberino’s counsel to seek assurances from

Elizabeth and Lucien Walczyk that there would be no further attempts to hinder development

of the land. In response, an attorney for the elder Walczyks advised that his clients were

claiming title to the undeveloped land by adverse possession. The claim was based on the

Walczyks’ long use of a portion of the undeveloped land for vegetable gardening and cattle

grazing.

3. 1988: Walczyk Again Brandishes a Gun at a Barberino Worker

Despite these 1981 events, the relationship between the Walczyks and Barberino

appears to have remained uneventful until January 1988 when, in response to a Barberino

demand that the Walczyks remove certain items from the property, the elder Walczyks

reiterated their adverse possession claim.

A few months later, in March 1988, a Barberino worker equipped with a bulldozer

attempted forcibly to remove various items from the disputed property. Once again, Thomas

Walczyk confronted the worker with a licensed firearm, specifically, an AR-15 automatic

assault rifle loaded with thirty rounds of ammunition, and ordered him off the property.

Police responded to the scene and directed Walczyk to drop his weapon. Walczyk initially

Page 219: Sotomayor Cases

3In addition to holding that Walczyk failed to adduce evidence sufficient to establish

any of the elements of adverse possession, the court ruled that his claim was barred by the

equitable doctrines of unclean hands and laches because he had “initiated and directed” his

parents’ 1988 suit against Barberino and made the strategic choice to pursue that claim and

its settlement only in their name. See Walczyk v. Barberino Realty, Inc., No. cv-

950465712S, 1997 Conn. Super. LEXIS 715, at *35 (Conn. Super. Ct. Mar. 14, 1997) (“The

plaintiff, by his failure to disclose his own claim in a timely fashion, effectively misled the

6

ignored several such directives, “yell[ing] about trespassing and some statute.” Police Rpt.,

Mar. 24, 1988, at 2. When Walczyk finally put down the rifle, a “wrestling match[]” ensued

as he tried to prevent the police from taking him into custody. Id. at 3. Charged with

threatening, reckless endangerment, and interfering with police, Walczyk eventually pleaded

guilty to the lesser infraction of creating a public disturbance.

4. The Walczyks’ Lawsuits Claiming Adverse Possession

The following month, in April 1988, Elizabeth and Lucien Walczyk sued for adverse

possession of the undeveloped property. The action settled in 1991 with Barberino paying

the elder Walczyks $20,000 and granting them a perpetual agricultural easement over a

portion of the disputed land. In return, Elizabeth and Lucien Walczyk signed a quitclaim of

any right, title, or interest in the property and promised not to oppose Barberino’s

development plans before the town planning and zoning commission.

Four years later, in January 1995, Thomas Walczyk sued Barberino, as well as his

parents, claiming that he held title to the disputed property through adverse possession. On

March 14, 1997, the Connecticut Superior Court rejected Walczyk’s claim as a matter of

law.3 In granting judgment to Barberino and quieting title in its favor, the Connecticut court

Page 220: Sotomayor Cases

defendants [in the settlement of that claim] . . . and now seeks to take unfair advantage of a

strategic decision he made, with the advice of counsel, to prosecute the first adverse

possession claim in his parents’ names only.”).

7

stated: “Thomas Joseph Walczyk has no estate, interest in or encumbrance of said real

property or any part thereof.” Walczyk v. Barberino Realty & Dev. Corp., No. cv-

950465712S, 1997 Conn. Super. LEXIS 718, at *2 (Conn. Super. Ct. Mar. 14, 1997). This

judgment was affirmed on direct appeal, see Walczyk v. Barberino Realty & Dev. Corp., 48

Conn. App. 911, 719 A.2d 1233 (Conn. App. Ct. 1998), and the Connecticut Supreme Court

declined further review, see Walczyk v. Barberino Realty & Dev. Corp., 245 Conn. 904, 719

A.2d 1165 (1998).

B. Events Relating to the Challenged Arrest and Searches

1. Walczyk’s April 1999 Threat to “Take Matters Into My Own Hands”

Despite the state courts’ unequivocal rejection of his adverse possession claim,

Walczyk persisted in asserting a superior interest in the disputed land. On April 9, 1999, he

visited the Farmington Police Department to complain about Barberino’s development

efforts, insisting to defendant Captain James Rio that he (Walczyk) “had a common law right

to the land because he had been farming and maintaining it for some time.” Arrest Warrant

Aff. at 2. When Rio explained that the police had received notice of the court order to the

contrary, Walczyk stated that he expected to secure reversal of that judgment based on

witness perjury and judicial misconduct. More significantly for purposes of the issues raised

on this appeal, Walczyk told Rio that, “[i]f you guys don’t comply with what I’m telling you

Page 221: Sotomayor Cases

4Walczyk acknowledged the comment at his deposition in this case: “I said well

here’s the law. It’s got a thousand years of history behind it and it’s clearly designed to

prevent blood shed and disputes over property. Now if a blood bath ensues, are you going

to be responsible, being the police, because they refused to uphold the law.” Walczyk Dep.

at 77.

8

I’ll take matters into my own hands.” Id. Rio warned Walczyk not to do anything illegal,

but Walczyk stated that he would “do what [he] had to do to protect his property.” Id.

(alteration in original).

2. Walczyk’s August 1999 Reference to a Potential “Bloodbath”

Some five months later, on August 30, 1999, Walczyk called the Farmington police

to complain again that Barberino personnel were trespassing on the disputed property.

Responding to the scene, Officer David Hebert explained to Walczyk that the police could

not act on his trespass complaint without some documentary support for his property claim.

In his report of the encounter, Hebert noted that Walczyk made “some off color com[m]ents

that the police were not taking the action needed to avoid a ‘blood bath.’” Police Rpt., Aug.

30, 1999, at 1.4

At Walczyk’s subsequent criminal trial, Hebert explained that, although he considered

this remark offensive, he did not immediately place Walczyk under arrest because the officer

did not feel any direct threat to himself. Nevertheless, he did understand Walczyk’s

comment as a threat toward “the Barberino Corporation and who[m]ever they were going to

have down there working.” Trial Tr. vol. 1, 56, Mar. 23, 2001. Indeed, Hebert informed

Barberino of the bloodbath statement, prompting its counsel to contact Captain Rio to request

Page 222: Sotomayor Cases

5Although Walczyk was initially charged with threatening, the state’s attorney

ultimately declined prosecution.

6Originally charged with unlawful discharge of a firearm, cruelty to animals, and

conspiracy to commit cruelty to animals, Walczyk ultimately pleaded guilty to breach of the

peace.

7John Walczyk told police that he did not fear imminent physical injury during this

argument in which no firearms were actually displayed; nevertheless, he felt nervous because

he knew his brother owned guns and could act on his threat. No charges were filed in

connection with this incident.

9

police protection at the disputed property site during any work periods.

3. The Challenged Arrest and Search Warrants

Soon thereafter, Rio reviewed Hebert’s report of his August 30, 1999 encounter with

Walczyk. Rio was, of course, aware of Walczyk’s earlier statement that, if police did not

assist him in his land dispute, he would take matters into his “own hands” and do what he

“had to do” to protect his rights. Moreover, Rio knew that Walczyk was the licensed owner

of a variety of firearms and that, over the years, he had responded to various situations by

displaying, discharging, or threatening to discharge a firearm. In addition to the 1981 and

1988 incidents, detailed supra at [5-6], during which Walczyk had brandished firearms

specifically at Barberino workers, these situations included a 1990 road-rage incident during

which Walczyk, armed with a loaded AK-47 rifle, confronted an angry motorist who had

followed him home;5 a 1992 complaint by neighbors that Walczyk shot a cat in his backyard;6

and a 1996 argument during which Walczyk threatened to shoot his brother John for using

the undeveloped land for driving practice.7

Page 223: Sotomayor Cases

8At the time of the events in question, section 53a-62(a) stated:

A person is guilty of threatening when: (1) By physical threat, he intentionally

places or attempts to place another person in fear of imminent serious physical

injury, or (2) he threatens to commit any crime of violence with the intent to

terrorize another, to cause evacuation of a building, place of assembly, or

facility of public transportation, or otherwise to cause serious public

inconvenience, or (3) he threatens to commit such crime in reckless disregard

of the risk of causing such terror or inconvenience.

Conn. Gen. Stat. § 53a-62(a) (1999) (amended 2001 & 2002). All citations herein are

to this version of the provision.

10

Viewing the bloodbath comment in this larger context, Rio concluded that the events

of August 30, 1999, demonstrated probable cause to arrest Walczyk for the Connecticut Class

A misdemeanor of threatening, see Conn. Gen. Stat. § 53a-62(a),8 and to search Walczyk’s

home and that of his parents for firearms that could be used as instrumentalities of the

bloodbath threat. Rio did not speak with Officer Hebert before making this determination,

but he did consult with a state’s attorney, who concurred in the captain’s probable cause

assessment.

Rio then communicated the relevant circumstances to defendant Sergeant William

Tyler and directed him to prepare the paperwork necessary to procure an arrest warrant for

Walczyk. After Tyler completed and signed the arrest warrant, he and Rio used the same

information to prepare search warrant applications for the homes of Thomas Walczyk and

his parents. These papers were then given to defendants Corporal Angela Deschenes and

Officer Shawn Brown, who acted as affiants for the search warrant applications. On

Page 224: Sotomayor Cases

9 We use the term “magistrate” as a general term for a judicial officer in the State of

Connecticut.

11

September 4, 1999, a magistrate authorized the arrest and search warrants.9

4. Facts Supporting the Warrants

Because plaintiffs claim, inter alia, that the warrant affidavits, on their face, fail to

establish probable cause to support the challenged arrest and searches, we here reproduce

the facts as detailed in the affidavits:

On 08-30-99, Officer Hebert of the Farmington Police Department responded

to [a] trespassing complaint made by Mr. Thomas Walczyk . . ., 27 Tunxis

Street, Farmington, CT. Walczyk complained of trespassing by employees of

the Barberino Realty & Development Corporation on property located on

Tunxis Street. Walczyk has had a long standing dispute over that property and

has made claims in the past that he has common law rights to the land.

Walczyk told Officer Hebert that the Farmington Police were not taking the

action needed to avoid a “bloodbath.”

Officer Hebert reported Walczyk’s complaint to Mr. [Stephen] Barberino Jr.,

the owner of the land in question. As a result of Walczyk’s threat of a

“bloodbath,” Atty Robert Reeve, representing Barberino, contacted Capt.

James Rio of the Farmington Police Department. Reeve expressed concerns

for the safety of employees during imminent construction work planned for the

Tunxis Street property. He requested extra police presence during work

periods.

In the early spring of 1999, the Farmington Police Department received a letter

from Barberino Jr.’s attorney along with a copy of a March 14, 1997 decision

made by the State of Connecticut Superior Court regarding the issue of the

land in question. The decision by Judge Christine E. Keller was in favor of

Barberino Realty & Development Corp., and stated that “Thomas Joseph

Walczyk has no estate, interest in or encumbrance of said real property or any

part thereof.”

During late winter of 1998 and early spring of 1999 Walczyk came to the

Page 225: Sotomayor Cases

12

Farmington Police Department to speak with Capt. Rio about the land dispute

and the impending land development. He told Capt. Rio at that time that he

had a common law right to the land because he had been farming and

maintaining it for some time. Walczyk said that he was in the process of

getting a Superior Court ruling to reverse the one made in Barberino’s favor.

He claimed that Stephen Barberino Jr. had perjured himself and the presiding

Judge had acted inappropriately. Capt. Rio explained that the Farmington

Police had been advised of the ruling in favor of Barberino and that until we

were officially notified otherwise, all parties and the police department would

have to abide by the last court ruling. Walczyk responded that, “If you guys

don’t comply with what I’m telling you I’ll take matters into my own hands.”

Capt. Rio advised Walczyk against any illegal actions to which Walczyk

responded that he would, “ . . . do what [he] had to do to protect his property.”

The Farmington Police Department has investigated Walczyk on previous

occasions for incidents involving threatening during which times he has either

threatened the use of or displayed a gun.

On 09-15-96 Officer Charette of the Farmington Police Department

investigated a disturbance on Tunxis Street involving Walczyk and his brother.

The argument was over property on Tunxis Street for which Thomas Walczyk

was suing his parents. His brother claimed that Walczyk pushed him and

threatened to shoot him. His brother felt no imminent threat but was

concerned because he knew Walczyk owned numerous guns. Walczyk denied

making the threat and no arrest was made.

On 02-15-92 Walczyk was arrested in Farmington for cruelty to animals and

unlawful discharge of a firearm. Neighbors reported seeing him shoot a cat on

his property with a handgun.

On 12-08-90 neighbors complained that Walczyk was shooting guns on

property at the end of Tunxis Street. He was shooting but was not in violation.

On 07-30-90 Walczyk was arrested by the Farmington Police for threatening.

A motorist followed Walczyk home to complain about the way he was driving.

Walczyk went inside and came back with a A-K assault rifle and an argument

ensued. The assault rifle was taken as evidence. It was loaded with twenty

rounds of ammunition.

On 03-24-88 the Farmington Police Department responded to a disturbance at

Page 226: Sotomayor Cases

10The arrest warrant affidavit contained identical facts, but did not include the last

paragraph describing the Tunxis Street residences.

13

Tunxis Street. The disturbance was over the same land dispute between

Walczyk and Barberino employees. Barberino employees were working on the

land. Walczyk approached them carrying an AR 15 assault rifle ordering them

to get off his property. One of the employee[]s complained that Walczyk had

pointed the gun directly at him. Walczyk was arrested for threatening, reckless

endangerment, and interfering with police. He denied actually pointing the

gun at anyone. The gun was seized as evidence. It contained one .223 round

in the chamber and twenty-nine rounds in the magazine. Walczyk fought with

officers prior to being arrested.

As of 09-03-99, Walczyk, according to Connecticut State Police records, has

the following handguns registered in his name.

1. Colt model 1903, 32 caliber, ser. #: 354507.

2. Colt Govt. Model, 45 caliber, ser. #: 40562G70.

3. Smith & Wesson model 629, 44 caliber, ser. #: N872450.

4. Walthers Woodsman model, .22 caliber, ser. #: 142639.

He also has the following assault weapons registered in his name.

1. Non-classified, 86S, ser. #: A000316.

2. Ruger (SR), Mini-14, ser. #: 18465824.

3. Colt, AR15-A2 H-BAR, ser. #: 325465.

4. Non-classified, MAK-90, ser. #: 9362979.

That a review of Farmington Police records indicate[s] that Walczyk has

maintained residences at both 27 and 28 Tunxis St., Farmington, CT. That

town of Farmington property records show that 27 Tunxis Street is owned by

Lucian Walzak [sic], and 28 Tunxis Street is owned by Thomas Walzak [sic].

Search Warrant Aff. and App. at 2-3.10

5. Execution of the Warrants

The challenged warrants were executed on September 7, 1999. On that date,

defendant Sergeant James Jepsen contacted Thomas Walczyk and, on the pretense of wishing

to discuss the land dispute, proposed a meeting at the police station. Upon Walczyk’s arrival,

Page 227: Sotomayor Cases

11Under Connecticut law, police officers are empowered to set temporary bail. See

Conn. Gen. Stat. § 54-63c(a) (discussed infra at [43-45]).

12Walczyk was acquitted on four other counts of improper firearm storage; risking

injury to a child, see Conn. Gen. Stat. § 53-21; and threatening, the charge that had initially

prompted his arrest and the search of his and his parents’ homes.

13Although plaintiffs note that this decision was authored by former Connecticut

Supreme Court Chief Justice Ellen A. Peters, they do not argue that this ruling is dispositive

in this case. See Norton v. Sam’s Club, 145 F.3d 114, 117 (2d Cir. 1998) (noting that

14

Sergeant Tyler placed him under arrest and detained him in a cell block on $10,000 bail.11

Walczyk remained in custody until later that day, when his mother posted bail.

While Walczyk was in custody, a team of officers, including Sergeant Jepsen and

defendant Detective Brian Killiany executed the challenged search warrants, seizing nearly

60 licensed firearms from Thomas Walczyk’s home and 18 licensed firearms from his

parents’ home, as well as approximately 2,600 rounds of ammunition, gun clips, ammunition

belts, and other items related to firearms’ use.

C. Connecticut’s Invalidation of the Challenged Warrants

After a Connecticut jury found Walczyk guilt of disorderly conduct, see Conn. Gen.

Stat. § 53a-182(a)(2); reckless endangerment, see id. § 53a-64(a); and two counts of

improper firearm storage, see id. § 29-37i,12 he was sentenced to pay a fine of $100 for each

count of conviction.

The Appellate Court of Connecticut reversed Walczyk’s conviction, holding that the

search warrant that resulted in seizure of the charged guns was not supported by probable

cause. See State v. Walczyk, 76 Conn. App. at 180-82, 818 A.2d at 875-76.13 In reaching

Page 228: Sotomayor Cases

argument not raised on appeal is deemed waived). Nor is it likely that such an argument

would be convincing in light of numerous decisions declining to hold individual state

officials bound, in their individual capacities, by determinations adverse to the state in prior

criminal cases. See Jenkins v. City of New York, 478 F.3d 76, 85-86 (2d Cir. 2007) (holding

that ruling in New York state criminal proceeding that defendant’s arrest was not supported

by probable cause did not collaterally estop police from relitigating question when defendant

sued them under § 1983); Bilida v. McCleod, 211 F.3d 166, 170-71 (1st Cir. 2000) (holding

that collateral estoppel did not bar Rhode Island officers sued under § 1983 from asserting

legality of searches found unconstitutional in earlier criminal proceedings: “[T]he interests

and incentives of the individual police [officers] . . . are not identical to those of the state, and

the officers normally have little control over the conduct of a criminal proceeding”); McCoy

v. Hernandez, 203 F.3d 371, 374-75 (5th Cir. 2000) (same re: suit against Texas officers);

Kinslow v. Ratzlaff, 158 F.3d 1104, 1105-06 (10th Cir. 1998) (same re: Oklahoma officers);

Kraushaar v. Flanigan, 45 F.3d 1040, 1050-51 (7th Cir. 1995) (reaching same result

regarding an Illinois officer); Duncan v. Clements, 744 F.2d 48, 51-52 (8th Cir. 1984) (same

under Missouri law); Davis v. Eide, 439 F.2d 1077, 1078 (9th Cir. 1971) (same with respect

to California officers). While we need not conclusively decide the issue, there is no reason

to think Connecticut law would support a different conclusion regarding estoppel in this case.

See, e.g., State v. Fritz, 204 Conn. 156, 173, 527 A.2d 1157, 1166 (1987) (observing that

privity necessary to trigger collateral estoppel is not established “from the mere fact that

persons may happen to be interested in the same question or in proving or disproving the

same facts. While the concept of privity is difficult to define precisely, it has been held that

a key consideration for its existence is the sharing of the same legal right by the parties

allegedly in privity.” (internal quotation marks and citation omitted)), overruled on other

grounds by State v. Crawford, 257 Conn. 769, 779-80, 778 A.2d 947, 954 (2001); accord

Tevolini v. Tevolini, 66 Conn. App. 16, 22 n.6, 783 A.2d 1157, 1163 n.6 (Conn. App. Ct.

2001).

15

this conclusion, the court faulted the warrant’s supporting affidavit for failing “to reconcile

a construction of the ‘bloodbath’ statement as a threat . . . with the statement . . . of what the

defendant actually had said.” Id. at 180, 818 A.2d at 875. Specifically, the court held that

“[a] statement to a police officer that the police needed to act to avoid a ‘bloodbath’ cannot

be the basis of probable cause to believe that the defendant, at that time or in the immediate

future, would engage in threatening behavior.” Id. at 181-82, 818 A.2d at 876 (emphasis in

Page 229: Sotomayor Cases

16

original). The Connecticut court also noted that the affidavit failed to establish probable

cause because it did not state that any of Walczyk’s earlier misconduct had resulted in his

“conviction of threatening or of any other crime,” distinguish “between recent incidents and

those that ha[d] become stale,” or state that his firearm possession was in any way unlawful.

Id. at 180, 818 A.2d at 875 (emphasis in original). Nor did the court think that Walczyk’s

prior statement to Captain Rio — that “he would take matters into his own hands and do what

he had to do to protect his property” — established probable cause because Rio advised

Walczyk “not to do so,” and, on August 30, 1999, Walczyk “followed instructions to report

any possible trespass to the police.” Id. at 181, 818 A.2d at 876 (emphasis in original).

D. The District Court Action

On August 30, 2002, plaintiffs commenced this action, charging defendants with (1)

violating their federal and state constitutional rights to have arrests and searches supported

by probable cause, (2) depriving them of their federal rights to equal protection of the laws

and to free expression, (3) violating their federal and state constitutional rights to bear arms,

and (4) holding Walczyk on excessive bail. Defendants moved for summary judgment,

which the district court granted with respect to all claims except those challenging Thomas

Walczyk’s arrest, the searches of his and his parents’ homes, and Walczyk’s First

Amendment claim, which apparently had been previously abandoned.

We need not here discuss the district court rulings with respect to the plaintiffs’ equal

protection or right to bear arms claims because neither is challenged on this appeal. As for

Page 230: Sotomayor Cases

17

Walczyk’s excessive bail claim, the district court ruled “as a matter of law” that “when a

police officer sets temporary bail under Conn. Gen. Stat. § 54-63c, he performs a judicial

function and hence has absolute immunity from suit.” Walczyk v. Rio, 339 F. Supp. 2d at

390.

With respect to plaintiffs’ unlawful arrest and search challenges, the district court

concluded that, although the supporting warrants were “facially valid,” defendants were not

entitled to summary judgment on the ground of qualified immunity because questions of fact

existed as to whether they had “knowingly and deliberately, or with reckless disregard of the

truth, made material misstatements or omissions in the warrant affidavit[s] that were

necessary to the finding of probable cause.” Id. at 389; see Franks v. Delaware, 438 U.S.

154, 155-56 (1978). The court identified three such material omissions: the affidavits’

failure to disclose that (1) the defendants “had not spoken with Officer Hebert about his

conversation” with Walczyk, (2) none of Walczyk’s previous arrests “had resulted in a

conviction for threatening,” and (3) Walczyk “had not lived at 27 Tunxis Street for seven

years.” Walczyk v. Rio, 339 F. Supp. 2d at 389. The court concluded that “[a] reasonable

juror could find that the omission of the first two items of information was critical to the

finding of probable cause for the arrest, and that the omission of all three items was critical

to the finding of probable cause for the searches of the houses and the seizures of the

firearms.” Id. Thus, it ruled that defendants were not entitled to summary judgment on the

basis of qualified immunity under either federal or state law (assuming arguendo that

Page 231: Sotomayor Cases

14We need not decide on this appeal whether Connecticut affords qualified immunity

in cases of unlawful searches or seizures. See Rustici v. Malloy, No. cv-970164460S, 2004

Conn. Super. LEXIS 1734, at *46 n.26 (Conn. Super. Ct. July 1, 2004) (assuming that

“qualified immunity appl[ies] to state constitutional claims”). Our holding that probable

cause supports the arrest of Walczyk and the search of his home, see infra at [31-38],

obviates the need for an immunity shield — state or federal — on those claims. See, e.g.,

Weyant v. Okst, 101 F.3d 845, 852 (2d Cir. 1996) (“The existence of probable cause to arrest

constitutes justification and is a complete defense to an action for false arrest, whether that

action is brought under state law or under § 1983.” (internal quotation marks and citation

omitted)); Beinhorn v. Saraceno, 23 Conn. App. 487, 492, 582 A.2d 208, 210-11 (Conn.

App. Ct. 1990) (stating that claim for false arrest under Connecticut law requires proof that

arresting officer lacked probable cause). As to Elizabeth Walczyk’s unlawful search claim,

because we identify factual issues, see infra at [41-43], that could moot defendants’ qualified

immunity claim, we conclude that the availability of state law immunity need not be

addressed unless and until these issues are resolved favorably to the defendants.

15To the extent Captain Rio claimed that the evidence was insufficient to ascribe

supervisory responsibility to him for any warrant deficiencies, the district court concluded

that sufficient evidence had been adduced to support a jury finding that Rio was actually

“involved in preparing the warrant affidavit.” Walczyk v. Rio, 339 F. Supp. 2d at 390.

Because we agree with this conclusion, see Hayut v. State Univ. of New York, 352 F.3d 733,

753 (2d Cir. 2003) (noting that supervisory liability is supported by “evidence of a

supervisory official’s personal involvement in the challenged conduct” (internal quotation

marks omitted)), it is unnecessary for us to distinguish between Rio and the other defendants

in discussing the probable cause question that is the crux of plaintiffs’ unlawful search and

arrest challenges.

18

Connecticut would afford immunity to state constitutional claims14) because, when the record

was viewed in the light most favorable to the plaintiffs, “a jury could find that the defendants

lacked even arguable probable cause to believe that [Walczyk] had committed the crime of

threatening or that the firearms to be seized were connected with criminal activity.” Id. at

390 (noting that ruling was consistent with Connecticut Appellate Court’s determination that

affidavits “fell well short of establishing probable cause”).15 Nevertheless, the district court

Page 232: Sotomayor Cases

19

denied Elizabeth Walczyk’s cross-motion for summary judgment on the liability phase of her

search warrant challenge, concluding that she had not shown that no reasonable juror could

find defendants’ acts lawful.

II. Discussion

A. Jurisdiction and the Standard of Review

Because the denial of a motion for summary judgment is not a final judgment, it is

generally not immediately appealable. See, e.g., Jones v. Parmley, 465 F.3d 46, 54 (2d Cir.

2006). An exception obtains, however, when the denied motion was based on a claim of

immunity, at least to the extent the immunity claim presents a “purely legal question.”

Mitchell v. Forsyth, 472 U.S. 511, 530 (1985) (recognizing jurisdiction to review “purely

legal question on which . . . claim of immunity turns”); see O’Bert ex rel. Estate of O’Bert

v. Vargo, 331 F.3d 29, 38 (2d Cir. 2003) (observing that “[u]nder the collateral order doctrine

. . . the denial of a qualified-immunity-based motion for summary judgment is immediately

appealable to the extent that the district court has denied the motion as a matter of law,

although not to the extent that the defense turns solely on the resolution of questions of

fact”); accord Jones v. Parmley, 465 F.3d at 54. The rationale for this exception is the law’s

recognition that immunity shields a defendant from suit itself, not merely from liability. See

Saucier v. Katz, 533 U.S. 194, 199 (2001) (“The privilege is ‘an immunity from suit rather

than a mere defense to liability; . . . it is effectively lost if a case is erroneously permitted to

go to trial.’” (quoting Mitchell v. Forsyth, 472 U.S. at 526 (emphasis in original))). In this

Page 233: Sotomayor Cases

20

case, defendants’ appeal from the district court’s denial of qualified immunity on plaintiffs’

search and arrest claims can be decided as a matter of law; accordingly, our jurisdiction is

established.

Although an interlocutory appeal would not be available from either the denial of

Elizabeth Walczyk’s motion for summary judgment on her unlawful search claim or the

district court’s dismissal of Walczyk’s excessive bail claim on the ground of absolute

immunity, we elect to exercise pendent jurisdiction over both. “[W]e may exercise pendent

jurisdiction over . . . issues that are not ordinarily subject to interlocutory review only when:

(1) they are ‘inextricably intertwined’ with the determination of qualified immunity; or (2)

their resolution is ‘necessary to ensure meaningful review’ of the district court’s ruling on

qualified immunity.” Jones v. Parmley, 465 F.3d at 64 (quoting Swint v. Chambers County

Comm’n, 514 U.S. 35, 51 (1995)). As we explain further below, see infra at [42-43],

Elizabeth Walczyk’s claim is “inextricably intertwined” with the determination of

defendants’ entitlement to qualified immunity in that the same disputed factual issues that

preclude a finding of qualified immunity on this claim at this stage also make summary

judgment inappropriate. As to Walczyk’s excessive bail claim, if we were to determine that

the district court improperly granted defendants absolute immunity, they might nonetheless

be entitled to qualified immunity, and thus review of this determination is also inextricably

intertwined with our resolution of defendants’ entitlement to qualified immunity. Cf. Clynch

Page 234: Sotomayor Cases

21

v. Chapman, 285 F. Supp. 2d 213, 219 n.6 (D. Conn. 2003) (characterizing absolute

immunity in excessive bail claim as a “cousin issue” to qualified immunity).

We review de novo defendants’ legal challenge to the district court’s qualified

immunity ruling, see Jones v. Parmley, 465 F.3d at 55, as well as Elizabeth and Thomas

Walczyks’ cross-appeals from other summary judgment rulings, see, e.g., Root v. Liston, 444

F.3d 127, 130 (2d Cir. 2006).

B. Plaintiffs’ Search and Arrest Claims

1. The Qualified Immunity Standard

When a defendant officer charged with violations of federal constitutional rights

invokes qualified immunity to support a motion for summary judgment, a court must first

consider a threshold question: Do the facts, viewed in the light most favorable to the

plaintiff, show that the officer’s conduct violated a constitutional right? If the answer to this

question is no, “there is no necessity for further inquiries concerning qualified immunity.”

Saucier v. Katz, 533 U.S. at 201; see X-Men Sec., Inc. v. Pataki, 196 F.3d 56, 66 (2d Cir.

1999) (observing that resolution of this first question favorable to defendant “moots” further

inquiry into qualified immunity). The reason for this rule is that, where there is no viable

constitutional claim, defendants have no need of an immunity shield. See generally Farrell

v. Burke, 449 F.3d 470, 499 n.14 (2d Cir. 2006) (“Because we have found no cognizable

violation of [p]laintiff’s rights in this case, we need not reach the question of qualified

immunity.”); Holcomb v. Lykens, 337 F.3d 217, 223-25 (2d Cir. 2003) (declining to decide

Page 235: Sotomayor Cases

22

qualified immunity question and affirming summary judgment on ground that, as a matter

of law, defendants did not violate plaintiff’s due process rights).

Only if the answer to the first question is yes must a court proceed to the inquiry for

qualified immunity: Was the right at issue clearly established at the time of the defendant’s

actions? As the Supreme Court has explained, this question is not answered by reference to

how courts or lawyers might have understood the state of the law: “The relevant, dispositive

inquiry in determining whether a right is clearly established is whether it would be clear to

a reasonable officer that his conduct was unlawful in the situation he confronted.” Saucier

v. Katz, 533 U.S. at 202 (emphasis added). If the right at issue was not clearly established

by then existing precedent, then qualified immunity shields the defendant. Even if the right

at issue was clearly established in certain respects, however, an officer is still entitled to

qualified immunity if “officers of reasonable competence could disagree” on the legality of

the action at issue in its particular factual context. Malley v. Briggs, 475 U.S. 335, 341

(1986); accord Iqbal v. Hasty, – F.3d –, 2007 WL 1717803, at * 20 (2d Cir. June 14, 2007).

Cerrone v. Brown, 246 F.3d 194, 202 (2d Cir. 2001); Lennon v. Miller, 66 F.3d 416, 420 (2d

Cir. 1995); see Saucier v. Katz, 533 U.S. at 208 (holding officer entitled to qualified

immunity if “[a] reasonable officer in [his] position could have believed that [the challenged

conduct] was within the bounds of appropriate police responses”). In this respect, the

Supreme Court has observed that qualified immunity protects “all but the plainly incompetent

or those who knowingly violate the law.” Malley v. Briggs, 475 U.S. at 341 (quoted

Page 236: Sotomayor Cases

16Our concurring colleague takes exception to these standards at the same time that

she acknowledges their reiteration of well established precedent. Her concern that a

bifurcation of the “clearly established” inquiry might allow a defendant to secure qualified

immunity for conduct already held unconstitutional “in the particularized sense,” post at [SS

Concurrence at 6-7] is, in fact, unwarranted. If controlling authority has already established

the unlawfulness of the challenged conduct in the particularized circumstances presented in

the pending case, then no reasonable officer could think otherwise and, thus, qualified

immunity would not shield the defendant. See, e.g., Groh v. Ramirez, 540 U.S. 551, 564

(2004).

Judge Sotomayor’s further criticism — that determining clearly established law by

reference to disagreements among reasonably competent officers, as indicated in Malley v.

Briggs, 475 U.S. at 341, is “more permissive of defendants” than a single reasonable officer

standard, post at [SS Concurrence at 8]— might merit attention if Malley contemplated

“officers of reasonable competence” disagreeing based on unreasonable views of existing

law. In fact, neither Malley nor today’s decision supports that conclusion. Instead, what

Malley does is provide courts with a useful tool for assessing when pre-existing law that did

not recognize the invoked right in the particularized context at issue, nevertheless, “must”

have alerted the defendant to the unlawfulness of his action. See Anderson v. Creighton, 483

U.S. 635, 640 (1987) (holding that, where particular action at issue has not previously been

held unlawful, defendant is not entitled to qualified immunity if, “in the light of pre-existing

law the unlawfulness must [have] be[en] apparent”). By instructing courts to focus on

whether “officers of reasonable competence could disagree” about the illegality of the

challenged conduct, Malley sounds a useful reminder: because law enforcement work relies

on probabilities and reasonable suspicions in an almost infinite variety of circumstances,

many requiring prompt action, there can frequently be a range of responses to given

situations that competent officers may reasonably think are lawful. Within this range, an

officer enjoys qualified immunity for “reasonable mistakes.” Saucier v. Katz, 533 U.S. at

205, 206.

To the extent Judge Sotomayor’s objection to Malley’s formulation relies on recent

Supreme Court habeas jurisprudence, we note that the considerations informing limitations

on habeas review are sufficiently distinct from those prompting recognition of qualified

immunity to preclude easy analogy. See, e.g., Harlow v. Fitzgerald, 457 U.S. 800, 807

(1982) (recognizing qualified immunity because of “need to protect officials who are

required to exercise their discretion and the related public interest in encouraging the

vigorous exercise of official authority”). Nothing in the Supreme Court’s qualified immunity

jurisprudence signals a narrowing of these considerations or an abandonment of Malley’s

analysis. To the contrary, as we note in text, Saucier v. Katz cites approvingly to Malley’s

23

approvingly in Saucier v. Katz, 533 U.S. at 202).16

Page 237: Sotomayor Cases

observation that qualified immunity is intended to shield “‘all but the plainly incompetent or

those who knowingly violate the law.’” 533 U.S. at 202 (quoting Malley v. Briggs, 475 U.S.

at 341).

Finally, insofar as Judge Sotomayor expresses some concern about courts

contemplating persons reaching different reasonable conclusions about the same facts, we

note simply that courts do so routinely in upholding jury verdicts as long as “any rational trier

of fact” could have reached the challenged result. United States v. MacPherson, 424 F.3d

183, 187 (2d Cir. 2005) (and cases cited therein).

Accordingly, we hold that courts may continue to rely on Malley in resolving qualified

immunity disputes.

17“The right of the people to be secure in their persons, houses, papers, and effects,

against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue,

but upon probable cause, supported by Oath or affirmation, and particularly describing the

place to be searched, and the persons or things to be seized.” U.S. Const. amend. IV.

18“The people shall be secure in their persons, houses, papers and possessions from

unreasonable searches or seizures; and no warrant to search any place, or to seize any person

or things, shall issue without describing them as nearly as may be, nor without probable cause

supported by oath or affirmation.” Conn. Const. art. First, § 7.

“No person shall be arrested, detained or punished, except in cases clearly warranted

by law.” Id. art. First, § 9.

24

2. The Legality of the Challenged Arrest and Searches

Applying these principles to this case, we consider first whether defendants’ actions

violated plaintiffs’ rights under both the United States and Connecticut Constitutions to be

free from unreasonable searches and arrests. See U.S. Const. amend. IV;17 Conn. Const. art.

First, §§ 7, 9.18 Ordinarily, an arrest or search pursuant to a warrant issued by a neutral

magistrate is presumed reasonable because such warrants may issue only upon a showing of

probable cause. See Franks v. Delaware, 438 U.S. at 171; United States v. Awadallah, 349

F.3d 42, 64 (2d Cir. 2003); Golino v. City of New Haven, 950 F.2d 864, 870 (2d Cir. 1991);

Page 238: Sotomayor Cases

25

see also United States v. Leon, 468 U.S. 897, 913-14 (1984). Plaintiffs nevertheless insist

that the presumption is defeated in this case because (1) the warrant affidavits, on their face,

fail to demonstrate probable cause, see generally United States v. Leon, 468 U.S. at 923; and

(2) the issuing magistrate was, in any event, misled into finding probable cause by material

omissions for which defendants were knowingly or recklessly responsible, see Franks v.

Delaware, 438 U.S. at 155-56; Golino v. City of New Haven, 950 F.2d at 870-71.

a. The Probable Cause Standard

Before discussing these two contentions, we observe that federal and Connecticut law

are identical in holding that probable cause to arrest exists when police officers have

“knowledge or reasonably trustworthy information of facts and circumstances that are

sufficient to warrant a person of reasonable caution in the belief that the person to be arrested

has committed or is committing a crime.” Weyant v. Okst, 101 F.3d 845, 852 (2d Cir. 1996);

see State v. James, 261 Conn. 395, 415, 802 A.2d 820, 833 (2002) (“Probable cause exists

when the facts and circumstances within the knowledge of the officer and of which he has

reasonable trustworthy information are sufficient in themselves to warrant a man of

reasonable caution to believe that a [crime] has been committed.” (internal quotation marks

omitted)). Similarly, under both federal and state law, probable cause to search is

demonstrated where the totality of circumstances indicates a “fair probability that contraband

or evidence of a crime will be found in a particular place.” Illinois v. Gates, 462 U.S. 213,

238 (1983); accord United States v. Gaskin, 364 F.3d 438, 456 (2d Cir. 2004); State v.

Page 239: Sotomayor Cases

26

Vincent, 229 Conn. 164, 171, 640 A.2d 94, 98 (1994) (noting that valid search requires

“probable cause to believe that the particular items to be seized are connected with criminal

activity or will assist in a particular apprehension or conviction” and “that the items sought

to be seized will be found in the place to be searched”); State v. Orellana, 89 Conn. App. 71,

80, 872 A.2d 506, 516 (Conn. App. Ct. 2005) (citing Illinois v. Gates, 462 U.S. at 231-32)).

Accordingly, we need not separately discuss federal and state law in assessing probable cause

for the challenged arrest and searches.

As the Supreme Court has famously observed, probable cause is “a fluid concept . .

. not readily, or even usefully, reduced to a neat set of legal rules.” Illinois v. Gates, 462 U.S.

at 232; see United States v. Gaskin, 364 F.3d at 456. While probable cause requires more

than a “mere suspicion,” of wrongdoing, Mallory v. United States, 354 U.S. 449, 455 (1957),

its focus is on “probabilities,” not “hard certainties,” Illinois v. Gates, 462 U.S. at 231. In

assessing probabilities, a judicial officer must look to “‘the factual and practical

considerations of everyday life on which reasonable and prudent men, not legal technicians,

act.’” Id. (quoting Brinegar v. United States, 338 U.S. 160, 175 (1949)); accord United

States v. Gaskin, 364 F.3d at 456. “Finely tuned standards such as proof beyond a reasonable

doubt or by a preponderance of the evidence, useful in formal trials, have no place” in a

probable cause determination. Illinois v. Gates, 462 U.S. at 235. Nor can probable cause be

analogized to a prima facie case. See id. (observing that “‘only the probability, and not a

prima facie showing, of criminal activity’” is necessary to establish probable cause to search

Page 240: Sotomayor Cases

27

or arrest (quoting Spinelli v. United States, 393 U.S. 410, 419 (1969))). In sum, probable

cause does not demand any showing that a good-faith belief be “correct or more likely true

than false.” Texas v. Brown, 460 U.S. 730, 742 (1983). It requires only such facts as make

wrongdoing or the discovery of evidence thereof probable.

It has long been recognized that, where there is no dispute as to what facts were relied

on to demonstrate probable cause, the existence of probable cause is a question of law for the

court. See Stewart v. Sonneborn, 98 U.S. 187, 194 (1878) (observing that whether facts

alleged to show probable cause are true is a matter of fact, “but whether, supposing them to

be true, they amount to a probable cause, is a question of law” (internal quotation marks

omitted)); accord Director Gen. of R.R.’s v. Kastenbaum, 263 U.S. 25, 28 (1923) (observing

that where facts are in dispute, court submits the question of probable cause to the jury, but

“with instructions as to what facts will amount to probable cause if proved”); Sanders v.

Palmer, 55 F. 217, 220 (2d Cir. 1893) (holding that question whether facts, “supposing them

to be true, . . . amount to a probable cause, is a question of law”; when the facts are disputed,

“it is the duty of the court to instruct the jury what facts, if established, will constitute a

probable cause . . . and to submit to them only the question as to the existence of those

facts”); see also United States v. Awadallah, 349 F.3d at 65 (distinguishing between de novo

review of legal question whether undisputed facts support probable cause and deferential

review of factual question whether, in case of misstated or omitted material facts, affiant’s

actions were deliberate or reckless). In this case, there can be no dispute as to what facts the

Page 241: Sotomayor Cases

28

defendants relied on to establish probable cause for the challenged arrest and searches; they

are memorialized in warrant affidavits. Thus, whether the affidavits, on their face,

demonstrate probable cause, is a question of law. In answering that question, however, a

reviewing court must accord considerable deference to the probable cause determination of

the issuing magistrate, see Illinois v. Gates, 462 U.S. at 238-39 (holding that duty of

reviewing court “is simply to ensure that the magistrate had a substantial basis” for probable

cause determination (internal quotation marks omitted)); accord Velardi v. Walsh, 40 F.3d

569, 574 n.1 (2d Cir. 1994), mindful of the well established principle that a showing of

probable cause cannot be negated simply by demonstrating that an inference of innocence

might also have been drawn from the facts alleged, see United States v. Webb, 623 F.2d 758,

761 (2d Cir. 1980); see also United States v. Forero-Rincon, 626 F.2d 218, 222 (2d Cir.

1980).

To the extent plaintiffs argue, in addition to their facial challenge, that material

omissions infected the issuing magistrate’s probable cause determination, there is no dispute

between the parties as to what purported omissions are appropriately considered. They are

the three non-disclosures identified by the district court, i.e., that (1) defendants had not

spoken with Officer Hebert about his understanding of Walczyk’s bloodbath statement; (2)

none of Walczyk’s prior arrests had resulted in a conviction for threatening, and (3) Walczyk

had not lived at his parents’ home for seven years. See Walczyk v. Rio, 339 F. Supp. 2d at

389. The materiality of these omissions presents a mixed question of law and fact. See

Page 242: Sotomayor Cases

29

Velardi v. Walsh, 40 F.3d at 574. Whether omitted information is “relevant to the probable

cause determination” is a question of law that we review de novo. Id. If we identify

relevancy, then questions of fact may arise as to what “weight . . . a neutral magistrate would

likely have given such information,” id., and whether defendants acted “deliberately or

recklessly” in omitting the information from the warrant affidavits, United States v.

Awadallah, 349 F.3d at 65 (internal quotation marks omitted). Even in such circumstances,

however, a court may grant summary judgment based on qualified immunity where “the

evidence, viewed in the light most favorable to the plaintiffs, discloses no genuine dispute

that a magistrate would have issued the warrant on the basis of the ‘corrected affidavits.’”

Velardi v. Walsh, 40 F.3d at 574 (emphasis in original).

Mindful of these principles, we proceed to consider plaintiffs’ warrant challenges,

focusing first on the arrest of Thomas Walczyk and the search of his 28 Tunxis Street

residence and then on the search of Elizabeth Walczyk’s 27 Tunxis Street home.

b. Thomas Walczyk

(1) The Facial Challenge

We reject as without merit Thomas Walczyk’s contention that the challenged warrant

affidavits, on their face, fail to state probable cause for his arrest or the search of his 28

Tunxis Street home. The facts alleged establish probable cause to believe (1) that Walczyk

had violated Connecticut law by “threaten[ing] to commit . . . [a] crime [of violence] in

reckless disregard of the risk of causing . . . terror” to another person, Conn. Gen. Stat. § 53a-

Page 243: Sotomayor Cases

19Although the warrant affidavits note that Walczyk denied pointing the rifle at

Barberino workers, the magistrate judge had probable cause to conclude otherwise given that

one of the workers specifically “complained that Walczyk had pointed the gun directly at

him.” Arrest Aff. at 3; Search Warrant Aff. and App. at 3.

30

62(a)(3), and (2) that Walczyk maintained in his residence firearms that, in light of past use,

were relevant evidence that he intended to threaten violence and recklessly disregarded the

threat’s terrorizing effect.

The facts reveal that, on August 30, 1999, Walczyk complained to Officer Hebert that

the police were not taking the action necessary to avoid a bloodbath. A reasonable person

would understand the bloodbath reference as a prediction of probable violence between

Walczyk and Barberino. More to the point, a reasonable person would understand from other

facts alleged in the affidavits that Walczyk would likely be the person initiating any such

violence. A few months earlier, Walczyk had stated to Captain Rio that, if the police did not

assist him in his property dispute with Barberino, he would take matters into his “own

hands,” doing whatever he “had to do” to protect his property rights. Rio knew that what

Walczyk frequently put into his hands to resolve disputes were loaded firearms that he stored

in his home. In the past, Walczyk had brandished firearms retrieved from his home at

various individuals, including Barberino workers on two occasions. On one of those

occasions, the brandished weapon was a loaded automatic rifle and,19 when police

intervened, Walczyk initially defied their orders to put down the weapon and resisted arrest.

Moreover, the affidavits demonstrated that Walczyk plainly knew how to fire his weapons;

Page 244: Sotomayor Cases

31

he had used them to kill a cat on his property. Rio further knew that Walczyk’s efforts to

vindicate his property rights peaceably through the courts had failed. Under the totality of

these circumstances, the issuing magistrate certainly had a substantial basis to conclude that,

when Walczyk told police that their continued failure to assist him in his property dispute

with Barberino would result in a bloodbath, he was effectively threatening to employ

violence against Barberino employees with reckless disregard for the terror such a threat

would cause when communicated to the intended victim.

We are, of course, mindful that a Connecticut appellate court has ruled otherwise.

Observing that Walczyk’s bloodbath statement was made to secure police assistance, that

court concluded: “A statement to a police officer that the police needed to act to avoid a

‘bloodbath’ cannot be the basis of probable cause to believe that the defendant, at the time

or in the immediate future, would engage in threatening behavior.” State v. Walczyk, 76

Conn. App. at 181-82, 818 A.2d at 876 (emphasis in original). We respectfully disagree.

Walczyk may have desired police assistance in his land dispute, but how he sought to compel

that assistance was by threatening violence. Walczyk was, after all, the only person to have

used an instrument of violence in connection with the land dispute. Given his prior

brandishing of loaded firearms, it was certainly probable that Walczyk’s bloodbath statement

was a threat to use violence against Barberino workers if the police did not intervene in his

favor (something they could not do in light of state court rulings). Whether Walczyk would,

in fact, have acted on his threat is not determinative of whether it was probable that he had

Page 245: Sotomayor Cases

20Walczyk does not fault the police for communicating his bloodbath statement to

Barberino, much less suggest that he did not intend or foresee such communication. Indeed,

given the totality of circumstances, it may well have been irresponsible of the police not to

have communicated the statement.

32

made the threat with reckless disregard of the terror it would cause Barberino.20 We

conclude that the affidavits, on their face, state facts reasonably supporting such a finding

by the issuing magistrate.

Walczyk submits that the search warrant affidavit nevertheless failed to demonstrate

that there was any connection between his present lawful possession of firearms and the

alleged crime of threatening. We are persuaded that the warrant affidavit states probable

cause to believe that a search of Walczyk’s home for firearms would produce evidence

relevant to demonstrating that Walczyk had committed the offense of threatening. At the

time the search warrant affidavit was prepared, Walczyk’s apparent possession of firearms

constituted relevant evidence which could suggest that his intent in making the bloodbath

remark was, in fact, to threaten violence. See State v. Crudup, 81 Conn. App. 248, 260 n.14,

838 A.2d 1053, 1062 n.14 (Conn. App. Ct. 2004) (discussing intent element of threatening).

Specifically, a seizure of firearms from Walczyk’s home could have shown that, at the time

Walczyk made the bloodbath remark, he had the actual capacity to cause bloodshed.

Moreover, such a seizure following the authorized search could have served to corroborate

witness accounts that Walczyk had used weapons against Barberino workers and others in

the past, which in turn could have helped establish his reckless disregard of the bloodbath

Page 246: Sotomayor Cases

21We also reject as without merit plaintiffs’ argument that the search warrant, which

simply sought “Firearms,” was insufficiently particular and that the seizure of ammunition,

gun clips, ammunition belts, and other items not named in the warrant violated the Fourth

Amendment. Although requests to search for “evidence of a crime” violate the proscription

against general warrants, see Groh v. Ramirez, 540 U.S. 551, 557-58 (2004), defendants’

application to search for “Firearms” was sufficiently particular because any firearms in

Walczyk’s possession were relevant evidence that his bloodbath remark was an intentional

threat of violence. Because we reach the same relevancy conclusion with respect to the

ammunition, gun clips, and related firearms paraphernalia found in Walczyk’s home, we

conclude that these items were properly seized under the “plain view” doctrine. United

States v. $557,933.89, 287 F.3d 66, 81 (2d Cir. 2002).

33

remark’s terrorizing effect. In sum, Walczyk’s possession of firearms was evidence relevant

to the mens rea element of the crime because a factfinder could reasonably infer from such

possession and from Walczyk’s past use of firearms that his bloodbath statement was not idle

hyperbole, but an intentional threat of violence made with reckless disregard of its potential

to cause terror. As the search warrant affidavit makes clear, the police were aware that

Walczyk had previously used his home to store the firearms he brandished in confrontations

with others, including Barberino workers, and thus they had probable cause to believe that

evidence relevant to his alleged threatening would turn up in a search of his home.

Accordingly, we hold that plaintiffs’ facial challenge to the warrant affidavits in this

case necessarily fails as a matter of law.21

(2) Purported Omissions

Walczyk asserts that two of the three identified material omissions misled the issuing

magistrate into erroneously finding probable cause to support his arrest and the search of his

home.

Page 247: Sotomayor Cases

34

(a) Failure to Speak With Officer Hebert

First, the district court pointed to defendants’ failure to disclose that no officer had

spoken directly with Officer Hebert, who would have revealed that he did not himself feel

threatened by Walczyk’s bloodbath statement. The conclusion is unconvincing both as a

matter of law and fact.

Preliminarily, we observe that the law permitting one law enforcement officer to rely

on the report of another in applying for a warrant nowhere requires direct consultation to

ensure that the officer reviewing the report ascribes no more weight to the described facts

than the reporter intended. See generally Panetta v. Crowley, 460 F.3d 388, 395 (2d Cir.

2006); Martinez v. Simonetti, 202 F.3d 625, 634 (2d Cir. 2000); Velardi v. Walsh, 40 F.3d

at 574. Indeed, we have specifically ruled that “a police officer is not required to explore and

eliminate every theoretically plausible claim of innocence before making an arrest.”

Martinez v. Simonetti, 202 F.3d at 635 (internal quotation marks omitted). Thus, we reject

the suggestion that a law enforcement officer is guilty of a material omission when, in

applying for a warrant, he fails to disclose that he has not spoken directly with a fellow

officer on whose report he relies to establish probable cause.

Even if we were to assume, however, that the applicant officer’s failure to ascertain

that the reporting officer did not consider words he heard to be threatening was relevant to

a determination of probable cause, that is not this case. Officer Hebert did understand

Walczyk’s bloodbath statement as a threat. At Walczyk’s criminal trial, Hebert testified that

Page 248: Sotomayor Cases

35

he did not understand Walczyk to be threatening him, but he most certainly did understand

Walczyk to be threatening any Barberino employees who came onto the disputed property.

As we have already noted, this understanding finds ample support in the totality of facts

recounted in the warrant affidavits. In sum, because Hebert’s subjective view of Walczyk’s

statement, if it had been solicited by defendants and reported to the issuing magistrate, would

actually have reinforced rather than undermined probable cause, we conclude that a

“corrected” warrant affidavit would raise no genuine dispute as to the magistrate’s issuance

of warrants for the arrest of Walczyk or the search of his home.

(b) The Lack of a Prior Conviction for Threatening

A second purported omission is the defendants’ failure to disclose that none of

Walczyk’s prior conduct had resulted in a conviction for threatening. This omission is hardly

relevant. Certainly, nothing in the challenged affidavits wrongly insinuates that Walczyk had

such a conviction. Absent such conduct, we expect that when a magistrate, mindful of the

government’s burden to demonstrate probable cause, reviews a warrant application that does

not report a prior conviction for a particular crime, the magistrate assumes for purposes of

determining whether the government has carried its burden that no such conviction exists.

See generally Minnesota v. Dickerson, 508 U.S. 366, 376 (1993) (noting that probable cause

requirement ensures against government action based on speculation). Moreover, as this

court observed in Brown v. D’Amico, the law does not demand that an officer applying for

a warrant “volunteer every fact that arguably cuts against the existence of probable cause,”

Page 249: Sotomayor Cases

36

as long as he does “not omit circumstances that are critical” to its evaluation, 35 F.3d 97, 99

(2d Cir. 1994). It was the particulars of Walczyk’s past conduct in using or threatening to

use firearms to resolve disputes that was critical to the determination of the probability that

his bloodbath statement constituted a threat of violence. This probability is in no way

undermined by the lack of a prior conviction for threatening.

Because we identify no merit in Thomas Walczyk’s facial challenge to the warrant

affidavits authorizing his arrest and the search of his home, and because we determine as a

matter of law that no alleged omissions were material to the issuance of these warrants, we

conclude that he (as well as his wife and daughter) fails to demonstrate a viable unlawful

search or arrest claim under federal or state law. Accordingly, we reverse the district court

order denying defendants’ summary judgment with respect to these plaintiffs’ unlawful

search and arrest claims, and we remand with directions to enter such a judgment.

c. Elizabeth Walczyk

(1) The Lack of Probable Cause

According to defendants, the theory for searching Elizabeth Walczyk’s residence was

that it probably contained firearms accessible to her son, constituting some further evidence

that his bloodbath statement was a threat of violence. To the extent Elizabeth Walczyk joins

in her son’s facial challenge to the warrant affidavits and to his charged material omissions

regarding Officer Hebert and Walczyk’s criminal record, we have already explained why we

reject these arguments. The district court, however, identified another omission that raises

Page 250: Sotomayor Cases

37

greater concern with respect to the search of Elizabeth Walczyk’s home.

The warrant affidavit reported that Thomas Walczyk was licensed to possess various

firearms and that he maintained two neighboring residences where such firearms would likely

be found: “[A] review of Farmington Police records indicate[s] that Walczyk has maintained

residences at both 27 and 28 Tunxis St., Farmington, CT.” Search Warrant Aff. and App.

at 3. The implication was that Walczyk had maintained the residences recently. What the

affidavit omitted, however, was the apparently undisputed fact that Walczyk had not resided

at his mother’s 27 Tunxis Street residence for more than seven years.

There can be no question that the omitted information was relevant to any assessment

of probable cause. In evaluating probable cause, a magistrate is always required to consider

whether the facts adduced in the warrant application “appear[] to be current, i.e., true at the

time of the application,” or whether they have “become stale.” Rivera v. United States, 928

F.2d 592, 602 (2d Cir. 1991). The law sensibly draws no bright-line rule for staleness.

Rather, a magistrate is expected to consider the age of the facts in light of the conduct at

issue with a view toward ensuring that probable cause exists at the time the warrant is to be

executed, not simply at some past time. See id.; see also United States v. Martino, 664 F.2d

860, 867 (2d Cir. 1981) (observing that, in circumstances of continuing or ongoing conduct,

as contrasted with isolated illegal acts, “the passage of time between the last described act

and the presentation of the application becomes less significant”). Thus, where information

is seven years old, a magistrate must be alerted to that fact to make a reasonable probable

Page 251: Sotomayor Cases

38

cause determination.

As we have already observed, the question of what weight a magistrate would have

given omitted relevant evidence is generally a question for the finder of fact. See Velardi v.

Walsh, 40 F.3d at 574. In this case, however, we can conclude as a matter of law that non-

disclosure of the staleness of the dual residency allegation was fatal to a demonstration of

probable cause. Not only was the allegation seriously outdated, it was the sole support for

a search of Elizabeth Walczyk’s home. A comparison best makes this point. With respect

to 28 Tunxis Street, the issuing magistrate could have inferred that Walczyk maintained guns

at that address because he was licensed to possess such weapons and, on at least one

occasion, he was actually seen retrieving a loaded assault rifle from that premises to brandish

at a person with whom he had a dispute. Further, neighbors had reported seeing him fire a

handgun on that property. By contrast, no facts were alleged indicating that Walczyk had

ever stored or retrieved firearms from his parents’ home, much less that he had done so in

the seven years since last residing there. Defendants urged that inference simply from

Walczyk’s license to possess firearms and his purported residence at 27 as well as 28 Tunxis

Street. Whatever questions might be raised about the strength of such an inference in any

circumstance, it could not be drawn from a dual residency allegation that was seven years

old.

Defendants submit that the search of Walczyk’s 28 Tunxis Street residence “would

have been meaningless” if he were “free to cross the street to his parents[’] home, where he

Page 252: Sotomayor Cases

39

formerly lived for many years, and have free access to other weaponry.” Appellant’s Br. at

30. This argument overlooks the fact that the warrant affidavit fails to allege any facts —

apart from Walczyk’s dual residency — suggesting that anyone ever stored guns at 27 Tunxis

Street. The affidavit makes no mention of the fact that Lucien or Elizabeth Walczyk was

licensed to possess guns. As we have recently reiterated, probable cause to search must be

based on particularized information about the place to be searched, not simply on a target’s

“‘mere propinquity to others independently suspected of criminal activity.’” United States

v. Martin, 426 F.3d 86, 88 (2d Cir. 2005) (quoting Ybarra v. Illinois, 444 U.S. 85, 91 (1979)).

Once the dual residency allegation is “corrected,” we can conclude as a matter of law that the

affidavit is devoid of any particularized information establishing probable cause to search

Elizabeth Walczyk’s home.

(2) Defendants’ Claim of Qualified Immunity;

Elizabeth Walczyk’s Motion for Summary Judgment on

the Issue of Liability

Despite our ruling that the search of Elizabeth Walczyk’s home was not supported by

probable cause, defendants might still be entitled to claim qualified immunity from liability

for damages if the search was supported by “arguable probable cause.” Escalera v. Lunn,

361 F.3d 737, 743 (2d Cir. 2004) (holding that, even in the absence of probable cause, officer

“will still be entitled to qualified immunity from a suit for damages if he can establish that

there was ‘arguable probable cause’”). “Arguable probable cause exists ‘if either (a) it was

objectively reasonable for the officer to believe that probable cause existed, or (b) officers

Page 253: Sotomayor Cases

22As the Supreme Court has explained:

[P]olice officers called upon to aid other officers in executing . . . warrants are

entitled to assume that the officers requesting aid offered the magistrate the

information requisite to support an independent judicial assessment of

40

of reasonable competence could disagree on whether the probable cause test was met.’” Id.

(quoting Golino v. City of New Haven, 950 F.2d at 870); see also Caldarola v. Calabrese,

298 F.3d 156, 162 (2d Cir. 2002). Like the district court, we conclude that questions of

disputed fact preclude a judicial resolution of this issue in favor of either side.

While no competent officer who knew that Thomas Walczyk had not resided in his

parents’ home for more than seven years could reasonably think that the stale allegation of

dual residency established probable cause to search that premises for firearms, it cannot be

determined from the present record which officers, if any, possessed — or even should have

possessed — such knowledge. Further record development and factfinding are necessary to

determine, among other things, (1) which of the defendants, if any, actually reviewed the

police records that purportedly established Walczyk’s residence at both 27 and 28 Tunxis

Street; (2) whether these records would have alerted a reasonable officer to the staleness of

the dual residency allegation; (3) which defendants, if any, possessed independent knowledge

that the dual residency statement was not accurate; (4) the circumstances under which the

dual residency allegation was communicated along the chain of defendants; and (5) whether

any defendant’s reliance on such communication without further inquiry was reasonable in

light of the totality of the circumstances.22 Because a resolution of some of these matters in

Page 254: Sotomayor Cases

probable cause. Where, however, the contrary turns out to be true, an

otherwise illegal arrest [or search] cannot be insulated from challenge by the

decision of the instigating officer to rely on fellow officers to make the arrest

[or search].

Whiteley v. Warden, 401 U.S. 560, 568 (1971); see also Varrone v. Bilotti, 123 F.3d 75, 81

(2d Cir. 1997) (concluding that subordinate officers carrying out search directive of superior

were entitled to qualified immunity although none had determined basis for order or

reliability of information on which it was based).

23The statute, in pertinent part, requires a police officer “promptly [to] order release

of the arrested person upon the execution of a written promise to appear or the posting of

such bond as may be set by the police officer, except that no condition of release set by the

court or a judge thereof may be modified by such officer.” Conn. Gen. Stat. § 54-63c(a)

(emphasis added). An officer may set bail only after conducting an interview with the

individual concerning the terms and conditions of release, at which interview counsel may

be present. See id.

41

favor of Elizabeth Walczyk could preclude one or more defendants from claiming that they

acted with arguable probable cause in searching her home, the district court correctly

concluded that defendants did not yet establish their entitlement to qualified immunity. On

the other hand, because a resolution favorable to one or more defendants could afford them

the benefit of a qualified immunity shield, the district court also correctly denied summary

judgment on liability to Elizabeth Walczyk. Accordingly, we affirm both these rulings.

C. Walczyk’s Excessive Bail Claim

Connecticut law allows state police to set temporary bail in certain cases. See Conn.

Gen. Stat. § 54-63c(a).23 The district court dismissed Walczyk’s excessive bail claim as a

matter of law, holding that “when a police officer sets temporary bail” pursuant to this

statute, “he performs a judicial function and hence has absolute immunity from suit.”

Page 255: Sotomayor Cases

42

Walczyk v. Rio, 339 F. Supp. 2d at 390. Other district courts in Connecticut have similarly

ruled. See Sanchez v. Doyle, 254 F. Supp. 2d 266, 269-73 (D. Conn. 2003); accord Machuca

v. Canning, No. 3:00-cv-1722, 2006 WL 2828160, at *6 (D. Conn. Sept. 29, 2006); Minney

v. Kradas, No. 3:01-cv-1543, 2004 U.S. Dist. LEXIS 5520, at *9-11 (D. Conn. Mar. 31,

2004); Bacciocchi v. Chapman, No. 3:02-cv-1403, 2004 U.S. Dist. LEXIS 1077, at *18-20

(D. Conn. Jan. 26, 2004); Clynch v. Chapman, 285 F. Supp. 2d at 219-23. Citing Sanchez

and Clynch by analogy, this court recently concluded that absolute judicial immunity shielded

a prosecutor who ordered a defendant’s bond increased pursuant to Conn. Gen. Stat. § 54-

63d because the prosecutor was performing a judicial function. See Root v. Liston, 444 F.3d

at 132 (observing that courts apply “functional approach to immunity questions”). Following

Root, we now hold what the citations to Sanchez and Clynch implied: police officers setting

bail under Conn. Gen. Stat. § 54-63c(a) are engaged in a judicial function that affords them

absolute immunity.

“It is . . . well established that officials acting in a judicial capacity are entitled to

absolute immunity against § 1983 actions, and this immunity acts as a complete shield to

claims for money damages.” Montero v. Travis, 171 F.3d 757, 760 (2d Cir. 1999) (extending

absolute immunity to parole board officials performing a quasi-judicial function in making

parole decisions); see also Butz v. Economou, 438 U.S. 478, 511 (1978) (granting absolute

immunity to administrative hearing examiners performing adjudicatory functions within

federal agencies). As the Supreme Court has explained, it is “the nature of the function

Page 256: Sotomayor Cases

43

performed, not the identity of the actor who performed it, that inform[s] our immunity

analysis.” Forrester v. White, 484 U.S. 219, 229 (1988) (holding that judges do not enjoy

absolute immunity when performing administrative, legislative, or executive functions).

Following this “functional approach to immunity questions,” this court in Root v. Liston

observed that “[o]rdinarily, it is judges who set bail, and judges enjoy absolute immunity

when they do so.” 444 F.3d at 132 (internal citations omitted). In short, because the setting

of bail is a judicial function, see Cleavinger v. Saxner, 474 U.S. 193, 205 (1985), absolute

immunity extends to police officers when they perform that function pursuant to statute.

Accordingly, we affirm the district court’s dismissal of Walczyk’s excessive bail

claim.

III. Conclusion

To summarize:

1. The unlawful search and arrest challenges of Thomas, Maximina, and Michelle

Walczyk are without merit as a matter of law because the warrants for Walczyk’s arrest and

for the search of these plaintiffs’ home were supported by probable cause. Defendants are

entitled to have summary judgment entered in their favor on these claims.

2. The warrant authorizing the search of Elizabeth Walczyk’s home was plainly not

supported by probable cause; nevertheless, the district court correctly denied summary

judgment to both Elizabeth Walczyk and defendants because disputed questions of fact must

be resolved before it can be determined whether defendants’ actions are shielded by qualified

Page 257: Sotomayor Cases

44

immunity or whether plaintiff is entitled to have a liability judgment entered in her favor.

3. Summary judgment was correctly entered in favor of defendants on Thomas

Walczyk’s claim of excessive bail because police officers, when setting bail pursuant to

Conn. Gen. Stat. § 54-63c(a), perform a judicial function, which affords them absolute

immunity from suit for money damages.

That part of the district court order denying defendants’ summary judgment motion

with respect to the unlawful search and arrest claims of plaintiffs Thomas, Maximina, and

Michelle Walczyk is hereby REVERSED and the case is REMANDED for entry of such a

judgment. In all other respects the appealed summary judgment rulings of the district court

are AFFIRMED.

Page 258: Sotomayor Cases

1 To the extent the majority holds that courts may continue to rely on this language1

from Malley, Maj. Op. at [25 n.16], I note that such a holding is unnecessary and serves no2

purpose in this case except to complicate the law further.3

2 Although I also disagree with the majority’s use of the term “arguable probable1

cause” and its reliance on whether “officers of reasonable competence could disagree,” Maj.2

Op. at [41], I join Part II(B)(2)(c)(2) of the majority opinion because I agree with its3

conclusion that questions of disputed fact preclude judicial resolution of whether the officers4

are entitled to qualified immunity for their search of Elizabeth Walczyk’s house.5

45

Sotomayor, J., concurring:

I agree fully with the outcome of this case, and I concur with most of the majority’s

reasoning; however, I disagree with its description of the qualified immunity standard we should

apply and its related discussion of “arguable probable cause.” A long line of decisions of this Court

features the same doctrinal misstatements, and it is time we stopped repeating uncritically this

particular language and gave it the attention it deserves.1 I join all of the majority opinion except

Part II(B)(1),2 and I write separately to call the Court’s attention to our collective failure to

harmonize our qualified immunity analysis with the Supreme Court’s directives.

The portion of the majority’s qualified immunity discussion that I find objectionable reads

as follows: “If the right at issue was not clearly established by then existing precedent, then qualified

immunity shields the defendant. Even if the right at issue was clearly established in certain respects,

however, an officer is still entitled to qualified immunity if ‘officers of reasonable competence could

disagree’ on the legality of the action at issue in its particular factual context.” Maj. Op. at [23]

(quoting Malley v. Briggs, 475 U.S. 335, 341 (1986)). These two sentences and the citation to

Malley reveal the two flaws I see in this circuit’s approach to qualified immunity. First, our

approach splits the single question of whether a right is “clearly established” into two distinct steps,

Page 259: Sotomayor Cases

46

contrary to Supreme Court precedent. Second, we demand a consensus among all hypothetical

reasonable officers that the challenged conduct was unconstitutional, rather than positing an

objective standard of reasonableness to which defendant officers should be held, as the Supreme

Court has repeatedly instructed us to do. I address both of these points in turn.

The Supreme Court has made clear that “[t]he relevant, dispositive inquiry in determining

whether a right is clearly established is whether it would be clear to a reasonable officer that his [or

her] conduct was unlawful in the situation he [or she] confronted.” Saucier v. Katz, 533 U.S. 194,

202 (2001). That is, whether a right is clearly established is the same question as whether a

reasonable officer would have known that the conduct in question was unlawful. This Court’s case

law, in contrast, bifurcates the “clearly established” inquiry into two steps. See, e.g., Cerrone v.

Brown, 246 F.3d 194, 199 (2d Cir. 2001) (“A police officer is entitled to qualified immunity from

liability for his [or her] discretionary actions if either (1) his [or her] conduct does not violate clearly

established statutory or constitutional rights of which a reasonable person would have known, or (2)

it was objectively reasonable for him [or her] to believe that his [or her] actions were lawful at the

time of the challenged act.” (emphasis added; internal quotation marks and citation omitted)). By

splitting the “relevant, dispositive inquiry” in two, we erect an additional hurdle to civil rights claims

against public officials that has no basis in Supreme Court precedent.

Whether a reasonable officer would know his or her conduct to be unlawful requires an

inquiry into the state of the law at the time of the conduct and “in light of the specific context of the

case.” Saucier, 533 U.S. at 201. If the right at issue has not previously been articulated, or had been

addressed only in a factual context that is “distinguishable in a fair way,” id. at 202, a reasonable

Page 260: Sotomayor Cases

3 The Supreme Court has stated, however, that the “clearly established” standard does1

not mean that “an official action is protected by qualified immunity unless the very action in2

question has previously been held unlawful,” Anderson v. Creighton, 483 U.S. 635, 6403

(1987), nor does the standard necessarily require that the facts of earlier cases be “materially4

similar” to the case under consideration, Hope v. Pelzer, 536 U.S. 730, 739-41 (2002). The5

standard is one of “fair warning,” id. at 741, such that “unlawfulness must be apparent” in6

light of pre-existing law, Anderson, 483 U.S. at 640. 7

47

officer might not have known that the challenged conduct was unlawful.3 See also id. (“‘[T]he right

the official is alleged to have violated must have been “clearly established” in a more particularized,

and hence more relevant, sense: The contours of the right must be sufficiently clear that a reasonable

official would understand that what he [or she] is doing violates that right.’” (quoting Anderson v.

Creighton, 483 U.S. 635, 640 (1987)). Contrary to what our case law might suggest, the Supreme

Court does not follow this “clearly established” inquiry with a second, ad hoc inquiry into the

reasonableness of the officer’s conduct. Once we determine whether the right at issue was clearly

established for the particular context that the officer faced, the qualified immunity inquiry is

complete.

Wilson v. Layne, 526 U.S. 603 (1999), illustrates the inquiry that the Supreme Court

contemplates for qualified immunity. The Wilsons brought suit against law enforcement officers

who permitted members of the media to accompany them in an early morning raid of the Wilsons’

home. Id. at 607. Having concluded that the officers’ actions violated the Fourth Amendment, the

Court considered whether the officers were entitled to qualified immunity. Id. at 614. The Court

explained that “what ‘clearly established’ means in this context depends largely ‘upon the level of

generality at which the relevant “legal rule” is to be identified.’” Id. at 614 (quoting Anderson, 483

U.S. at 639). That is, to be clearly established, “[t]he contours of the right must be sufficiently clear

Page 261: Sotomayor Cases

48

that a reasonable official would understand that what he [or she] is doing violates that right.” Id. at

615 (internal quotation marks omitted). The Court concluded that the officers were entitled to

qualified immunity because it was “not obvious from the general principles of the Fourth

Amendment,” id. at 615-16, or judicial decisions that the presence of the media was unlawful, and

“[g]iven [the] undeveloped state of the law, the officers . . . cannot have been expected to predict the

future course of constitutional law,” id. at 617 (internal quotation marks omitted).

Wilson confirms that whether an officer’s conduct was objectively reasonable is part and

parcel of the inquiry into whether the law was clearly established at the time of the challenged

conduct and for the particular context in which it occurred. To ask whether an officer’s violation

of an individual’s right was objectively reasonable after we have found that the right was clearly

established in the particularized sense finds no warrant in Wilson, Saucier, or any other recent

Supreme Court discussion of qualified immunity. See also Brosseau v. Haugen, 543 U.S. 194, 199-

200 (2004) (per curiam); Groh v. Ramirez, 540 U.S. 551, 563 (2004); Hope v. Pelzer, 536 U.S. 730,

739-46 (2002).

I suspect that our bifurcation of the “clearly established” analysis derives from the eminently

reasonable principle that whether a right is clearly established “is not answered by reference to how

courts or lawyers might have understood the state of the law.” Maj. Op. at [23]. We do not expect

law enforcement officers to keep abreast of every development in the case law or to recognize every

implication of legal precedent for police conduct that courts have not previously considered. See

Saucier, 533 U.S. at 205 (“It is sometimes difficult for an officer to determine how the relevant legal

doctrine . . . will apply to the factual situation the officer confronts.”). But our bifurcated approach

Page 262: Sotomayor Cases

49

makes too much of this principle by divorcing the reasonableness inquiry from the state of the law

at the time of the conduct in question. The inquiry described by the Supreme Court already

incorporates a recognition that police officers should not be expected to anticipate every application

of legal principles because it requires that the right be clearly established with particularity for the

conduct at issue.

In this case, the particularity requirement means that our “clearly established” inquiry is not

complete upon reaching the indisputable conclusion that an individual has the right to be free from

arrest, search, or seizure absent probable cause. See Anderson, 483 U.S. at 640-41. Rather, we must

determine whether it was clearly established that the situation the officer confronted did not give rise

to probable cause. See id. at 641 (holding that officials who “reasonably but mistakenly conclude

that probable cause is present . . . should not be held personally liable”); see also Saucier, 533 U.S.

at 206 (“[E]ven if a court were to hold that the officer violated the Fourth Amendment by conducting

an unreasonable, warrantless search, Anderson still operates to grant officers immunity for

reasonable mistakes as to the legality of their actions.”). In other words, based on the law at the time

the conduct occurred, would a reasonable officer have known that his or her actions were not

supported by probable cause, and therefore were in violation of the Fourth Amendment?

The majority opinion takes this question wholly outside of the “clearly established” inquiry

and asks whether the officers had “arguable probable cause.” See Maj. Op. at [41]. This Court has

used the term “arguable probable cause” to describe the standard for finding that a defendant officer

is entitled to qualified immunity for his or her reasonable but mistaken determination that probable

cause existed in a particular context. See Caldarola v. Calabrese, 298 F.3d 156, 162 (2d Cir. 2002)

Page 263: Sotomayor Cases

4 Other courts of appeals have also used the term “arguable probable cause” in a1

similar way as this Court. See, e.g., Williams v. Jaglowski, 269 F.3d 778, 781-82 (7th Cir.2

2001); Jones v. Cannon, 174 F.3d 1271, 1283 & n.3 (11th Cir. 1999). 3

50

(“[I]n the context of a qualified immunity defense to an allegation of false arrest, the defending

officer need only show arguable probable cause. This is because at its heart, [t]he concern of the

immunity inquiry is to acknowledge that reasonable mistakes can be made as to the legal constraints

on particular police conduct.” (internal quotation marks and citations omitted; alteration in original));

Cerrone, 246 F.3d at 203.4 We have also stated that “arguable probable cause” falls under the

objective reasonableness determination of our qualified immunity test. See Jenkins v. City of New

York, 478 F.3d 76, 87 (2d Cir. 2007). Yet reasonableness—and therefore the existence of “arguable

probable cause”—are considerations that properly fall within the clearly established inquiry as the

Supreme Court has described it. See Anderson, 483 U.S. at 640-41; Brosseau, 543 U.S. at 199-201.

It is not surprising, then, that “arguable probable cause” finds no mention in any Supreme Court

opinion; the need for a separate term to describe this concept arises only once we have improperly

splintered the “clearly established” inquiry. Because I believe “arguable probable cause” is both

imprecise and an outgrowth of the first flaw in our qualified immunity analysis, I do not agree with

the majority’s use of the term.

I recognize that the distinction I am drawing is a fine one, but I believe it has real

consequences. Our approach does not simply divide into two steps what the Supreme Court treats

singly, asking first, whether the right is clearly established as a general proposition, and second,

whether the application of the general right to the facts of this case is something a reasonable officer

could be expected to anticipate. Instead, we permit courts to decide that official conduct was

Page 264: Sotomayor Cases

5 This language also appears in the majority opinion as part of the explanation of1

“arguable probable cause.” Maj. Op. at [41]. Notably, however, by stating that “[a]rguable2

probable cause exists if either (a) it was objectively reasonable for the officer to believe that3

probable cause existed, or (b) officers of reasonable competence could disagree on whether4

the probable cause test was met,” id. (emphasis added; internal quotation marks and citation5

omitted), this passage seems to support the argument that whether “officers of reasonable6

competence could disagree” is not an objective reasonableness test. In addition, by adding7

two steps to the qualified immunity analysis beyond whether the particular right was clearly8

established, the majority’s discussion of “arguable probable cause” further splinters our9

qualified immunity test.10

6 Although Saucier does not specifically refer to the reasonable officer’s competence,1

I have no quarrel with the assumption that a “reasonable officer” is also a competent officer.2

51

“reasonable” even after finding that it violated clearly established law in the particularized sense.

By introducing reasonableness as a separate step, we give defendants a second bite at the immunity

apple, thereby thwarting a careful balance that the Supreme Court has struck “between the interests

in vindication of citizens’ constitutional rights and in public officials’ effective performance of their

duties.” Anderson, 483 U.S. at 639 (quoting Davis v. Scherer, 468 U.S. 183, 195 (1984)).

My second objection to the majority’s formulation of the qualified immunity standard is that

it treats objective reasonableness as turning on whether “officers of reasonable competence could

disagree.”5 Maj. Op. at [23]. This language, which our cases frequently recite, see, e.g., Iqbal v.

Hasty, – F.3d –, 2007 WL 1717803, at *20 (2d Cir. June 14, 2007); Cerrone, 246 F.3d at 202;

Lennon v. Miller, 66 F.3d 416, 420 (2d Cir. 1995), derives from the Supreme Court’s 1986 decision

in Malley, 475 U.S. at 341. Whether reasonably competent officers could disagree about the

lawfulness of the conduct at issue, however, is not the same question the Supreme Court has

repeatedly instructed us to consider: whether “it would be clear to a reasonable officer that his [or

her] conduct was unlawful in the situation he [or she] confronted.”6 Saucier, 533 U.S. at 202

Page 265: Sotomayor Cases

52

(emphasis added); see also Brosseau, 543 U.S. at 199 (quoting Saucier); Groh, 540 U.S. at 563

(same); Hope, 536 U.S. at 746 (same). As with our bifurcation of the “clearly established” inquiry,

our requirement of consensus among all reasonable officers departs from Supreme Court dictates and

unjustifiably raises the bar to liability for violations of constitutional rights.

Reasonable person standards are familiar constructs in the law. They define the level of

prudence, care, or knowledge that the law will require of a defendant called to task for his or her

actions. In the qualified immunity context, the reasonable officer embodies the minimum degree of

judgment and awareness of the law that courts expect law enforcement officials to exercise in the

conduct of their duties. That is, the reasonable officer standard sets the threshold beyond which a

defendant officer will not be entitled to immunity. As I have discussed, to determine what conduct

a reasonable officer should have known to be unlawful in the situation presented, a court must decide

whether the law was sufficiently clear regarding the conduct at issue, such that the reasonable officer,

and thus the defendant, would have had “fair notice that [his or] her conduct was unlawful.”

Brosseau, 543 U.S. at 198.

Asking whether “officers of reasonable competence could disagree” shifts this inquiry subtly

but significantly. Instead of asking whether the defendant’s conduct was beyond the threshold of

permissible error, as the reasonable officer standard does, this inquiry affords a defendant immunity

unless a court is confident that a range of hypothetical reasonably competent officers could not

disagree as to whether the defendant’s conduct was lawful. This standard is not only more

permissive of defendants seeking to justify their conduct; it also takes courts outside their traditional

domain, asking them to speculate as to the range of views that reasonable law enforcement officers

Page 266: Sotomayor Cases

53

might hold, rather than engaging in the objective reasonableness determination that courts are well-

equipped to make.

The Supreme Court has specifically criticized the conflation of an objective reasonableness

standard with a requirement of unanimous consensus in the context of a petition for a writ of habeas

corpus. In Williams v. Taylor, 529 U.S. 362 (2000), the Court interpreted the statutory provision

allowing a federal court to grant review of a petition for a writ of habeas corpus when a state court

judgment “involved an unreasonable application of[] clearly established Federal law,” 28 U.S.C.

§ 2254(d)(1). The Fourth Circuit had previously held that a state court’s adjudication involved an

“unreasonable application” of federal law only if “the state court has applied federal law ‘in a manner

that reasonable jurists would all agree is unreasonable.’” Williams, 529 U.S. at 409 (quoting Green

v. French, 143 F.3d 865, 870 (4th Cir. 1998) (emphasis added)). Both the majority and plurality

Supreme Court opinions rejected the Fourth Circuit’s interpretation of the “unreasonable

application” standard, explaining that whether an application of the law is objectively unreasonable

is a different, less stringent standard than one that asks whether reasonable jurists would

unanimously find an application of law unreasonable. See id. at 409-10 (majority opinion) (“[A]

federal habeas court making the ‘unreasonable application’ inquiry should ask whether the state

court’s application of clearly established federal law was objectively unreasonable. The federal

habeas court should not transform the inquiry into a subjective one by resting its determination

instead on the simple fact that at least one of the Nation’s jurists has applied the relevant federal law

in the same manner the state court did in the habeas petitioner’s case.”); id. at 378 (plurality opinion)

(“As Congress is acutely aware, reasonable lawyers and lawgivers regularly disagree with one

Page 267: Sotomayor Cases

7 This caution against transforming a reasonableness inquiry into a consensus1

requirement is instructive notwithstanding the plurality’s statement that the particular statute2

at issue was not meant to codify the standard for qualified immunity into the law of habeas3

review. See Williams, 529 U.S. at 380 n.12 (plurality opinion).4

54

another. Congress surely did not intend that the views of one such judge who might think that relief

is not warranted in a particular case should always have greater weight than the contrary, considered

judgment of several other reasonable judges.”).7 Our Court similarly has adopted an unjustifiably

stringent standard in the qualified immunity context by prohibiting liability for constitutional

violations where a court believes that one reasonably competent officer would find the conduct at

issue lawful, even if the overwhelming majority would not.

Finally, I note that although we repeat Malley’s “officers of reasonable competence” test with

regularity, and it appears frequently in the decisions of other federal courts of appeals, see, e.g.,

Brittain v. Hansen, 451 F.3d 982, 988 (9th Cir. 2006); Armstrong v. City of Melvindale, 432 F.3d

695, 701 (6th Cir. 2006); Wollin v. Gondert, 192 F.3d 616, 625 (7th Cir. 1999), it has not appeared

a second time in any majority opinion of the Supreme Court. It seems curious that we would

continue to rest our qualified immunity standard on language the Supreme Court has carefully

eschewed for over twenty years since Malley was decided.

In sum, the Supreme Court has struck a careful balance between the vindication of

constitutional rights and government officials’ ability to exercise discretion in the performance of

their duties. Our case law, in subtle but important ways, has altered this balance in favor of

defendants by adding another analytic step to the qualified immunity analysis and equating objective

reasonableness with unanimity among “officers of reasonable competence.” In the vast majority of

Page 268: Sotomayor Cases

55

cases, including this one, the particular phrasing of the standard will not alter the outcome of the

qualified immunity analysis. There is no doubt in this case that a reasonable officer would believe

that the arrest of Thomas Walcyzk, as well as the search of his home and the seizure of firearms

found there, were lawful. Yet the effect in future cases may not always be so benign. What is more,

the majority’s framework introduces unnecessary complications into an already complicated

qualified immunity analysis. It is time to eliminate these complications and reconcile our qualified

immunity analysis with the Supreme Court’s most recent, authoritative jurisprudence.

Page 269: Sotomayor Cases

1

02-4611-ag, 02-4629-ag, 03-40837-ag Lin et. al. v. U.S. Dept. of Justice

1 UNITED STATES COURT OF APPEALS2 FOR THE SECOND CIRCUIT

3 August Term, 2006

4 (Argued: March 7, 2007 Decided: July 16, 2007)

5

6 Docket Nos. 02-4611-ag, 02-4629-ag, 03-40837-ag

7 _____________________

8 SHI LIANG LIN,9

10 Petitioner,1112 v.1314 UNITED STATES DEPARTMENT OF JUSTICE; ATTORNEY GENERAL GONZALES,1516 Respondents;17 _____________________1819 ZHEN HUA DONG,2021 Petitioner,2223 v.2425 UNITED STATES DEPARTMENT OF JUSTICE; ATTORNEY GENERAL GONZALES

2627 Respondents;28 _____________________2930 XIAN ZOU,3132 Petitioner,3334 v.3536 ATTORNEY GENERAL GONZALES,

Page 270: Sotomayor Cases

2

1 Respondent.2 _____________________

3 Before: JACOBS, Chief Judge, CALABRESI, CABRANES, STRAUB, POOLER, SACK, SOTOMAYOR,

4 KATZMANN, PARKER, RAGGI, WESLEY, and HALL, Circuit Judges.

5 Petitions for review of orders of the Board of Immigration Appeals denying applications6 for asylum, withholding of removal, and relief under the Convention Against Torture. The7 petition for review of Zhen Hua Dong is DENIED. Xian Zou’s and Shi Liang Lin’s petitions are8 DISMISSED for lack of jurisdiction.9

10 Judge PARKER delivered the opinion of the Court, in which JACOBS, C.J., and CABRANES,11 SACK, RAGGI, WESLEY, and HALL, JJ., joined.1213 Judge KATZMANN filed a concurring opinion, in which STRAUB, POOLER, and14 SOTOMAYOR, JJ., joined.1516 Judge SOTOMAYOR filed a concurring opinion, in which POOLER, J., joined.1718 Judge CALABRESI filed an opinion concurring in part and dissenting in part.192021 BRUNO JOSEPH BEMBI, Hempstead, NY, for Petitioners Shi22 Liang Lin and Zhen Hua Dong.2324 ALEKSANDER MILCH, Christophe & Associates, P.C., New25 York, NY, for Petitioner Xian Zou.2627 KATHY S. MARKS, Assistant United States Attorney, (Sara28 L. Shudofsky, Assistant United States Attorney, of counsel)29 for Michael J. Garcia, United States Attorney for the30 Southern District of New York, New York, NY, for31 Respondents the United States Department of Justice and32 Attorney General Gonzales.3334 BARRINGTON D. PARKER, Circuit Judge:3536 In 1997 the Board of Immigration Appeals (“BIA”) held in Matter of C-Y-Z-, 21 I. & N.

37 Dec. 915 (B.I.A. 1997) (en banc) that an individual whose spouse has been forced to abort a

38 pregnancy, undergone involuntary sterilization, or been persecuted under a coercive population

Page 271: Sotomayor Cases

See In re Shi Liang Lin, No. A70 895 638 (B.I.A. Sept. 29, 2002), aff’g No. A70 8951

638 (Immig. Ct. N.Y. City May 9, 2000); In re Zhen Hua Dong, No. A 7 293 661 (B.I.A. Sept.25, 2002), aff’g No. A77 293 661 (Immig. Ct. N.Y. City Oct. 12, 2000); In re Zou, No. A77 322595 (B.I.A. Aug. 27, 2002), aff’g No. A77 322 595 (Immig. Ct. N.Y. City Apr. 4, 2002).

3

1 control program could automatically qualify for asylum as a “refugee” under § 601(a) of the

2 Illegal Immigration Reform and Immigrant Responsibility Act of 1996 (“IIRIRA”) (amending 8

3 U.S.C. § 1101(a)(42), Immigration and Nationality Act (“INA”) § 101(a)(42)). See In re S-L-L-,

4 24 I. & N. Dec. 1, 3 (B.I.A. 2006) (en banc) (“In Matter of C-Y-Z-, . . . we held that a husband

5 whose wife was forcibly sterilized could establish past persecution under this amendment to

6 section 101(a)(42) of the [INA].”). This appeal considers whether the BIA’s interpretation of the

7 statute was correct. We conclude it was not.

8 Petitioners Shi Liang Lin, Zhen Hua Dong, and Xian Zou are citizens of the People’s

9 Republic of China and unmarried partners of individuals allegedly victimized by China’s

10 coercive family planning policies. Each seeks review of an order of the BIA summarily

11 affirming the denial of an application for asylum based, in part, on the BIA’s holding in C-Y-Z. 1

12 We remanded these petitions to the BIA to afford it the opportunity to explain its rationale in C-

13 Y-Z- for reading § 601(a) to say that the spouses of those directly victimized by coercive family

14 planning policies are per se eligible for asylum as if they were directly victimized themselves

15 and also to clarify the status of boyfriends and fiancés under that statute. See Lin v. U.S. Dep’t of

16 Justice, 416 F.3d 184, 187 (2d Cir. 2005). We retained jurisdiction. Id.

17 On remand, the BIA reaffirmed its holding in C-Y-Z- that spouses are entitled to

18 automatic eligibility under § 601(a) but limited this per se eligibility to legally married

19 applicants. S-L-L-, 24 I. & N. Dec. 1. Eschewing a text-based analysis, the BIA elected to

Page 272: Sotomayor Cases

Subsequent to oral arguments, we learned that Lin has had no contact with his attorneys2

since “early 2004,” and that his attorney believes that he has either returned to China or isdeceased. (Aff. of Yee Ling Poon, ¶¶ 3-5). Accordingly, Lin’s case is moot. See 8 C.F.R. §

4

1 interpret the forced abortion and sterilization clause of the section “in light of the overall purpose

2 of the amendment” to include both parties to a marriage. Id. at 8. The Board reaffirmed the

3 dismissal of the appeals of petitioners Lin, id., and Dong, In re Zhen Hua Dong, No. A77 293

4 661 (B.I.A. Nov. 27, 2006), and remanded Zou’s petition for a determination of whether he

5 qualified for asylum based on the “other resistance to a coercive population control program”

6 clause in § 601(a), In re Xian Zou, No. A73 178 541 (B.I.A. Nov. 21, 2006).

7 Following the BIA’s decision, we ordered rehearing en banc to consider two issues:

8 First, whether § 601(a)’s provisions are ambiguous, so that the BIA’s construction of them

9 warrants Chevron deference; and second, whether the BIA reasonably construed § 601(a) to

10 extend automatic asylum eligibility to a petitioner whose legally married spouse was subjected to

11 an involuntary abortion or sterilization but not to a domestic partner or fiancé whose claim is

12 derivative unless the petitioner engaged in “other resistance” to a coercive population control

13 policy. Lin v. U.S. Dep’t of Justice, Nos. 02-4611, 02-4629, 03-40837 (2d Cir. Nov. 13, 2006)

14 (order) (“En banc order”). See S-L-L-, 24 I. & N. Dec. 1; Chevron U.S.A. Inc. v. Natural Res.

15 Def. Council, Inc., 467 U.S. 837 (1984).

16 We now conclude that the BIA erred in its interpretation of 8 U.S.C. § 1101(a)(42) by

17 failing to acknowledge language in § 601(a), viewed in the context of the statutory scheme

18 governing entitlement to asylum, that is unambiguous and that does not extend automatic refugee

19 status to spouses or unmarried partners of individuals § 601(a) expressly protects. Accordingly,

20 the petition of Zhen Hua Dong is denied. The petition of Shi Liang Lin is dismissed as moot. 2

Page 273: Sotomayor Cases

1208.8.

We no longer have jurisdiction over Zou’s petition because the BIA has remanded the3

case to the immigration court for further findings. See In re Xian Zou, No. A77 322 295 (B.I.A.Nov. 21, 2006).

A number of our sister circuits have deferred to the BIA’s interpretation of § 601(a). 4

See, e.g., Zhang v. Gonzales, 434 F.3d 993, 1001 (7th Cir. 2006); Huang v. Ashcroft, 113 Fed.App’x 695, 700 (6th Cir. 2004) (unpublished opinion); He v. Ashcroft, 328 F.3d 593, 604 (9thCir. 2003); Li v. Ashcroft, 82 Fed. App’x 357, 358 (5th Cir. 2003) (unpublished per curiamopinion). While the Third Circuit had questioned the BIA’s reading of the plain language of theamendment, stating that “[i]t takes some effort to reconcile [the BIA’s] interpretation with thelanguage of the 1996 amendment, since the phrase ‘a person who has been forced to abort apregnancy or to undergo involuntary sterilization’ is most naturally read as referring only to aperson who has personally undergone one of those procedures,” Chen v. Ashcroft, 381 F.3d 221,226 (3d Cir. 2004) (Alito, J.), a divided panel of the Third Circuit recently validated the BIA’sinterpretation of § 601(a) over a vigorous dissent. See Sun Wen Chen v. U.S. Att’y Gen., – F.3d–, 2007 WL 1760658, at *4-*6 (3d Cir. June 20, 2007).

The circuits are already split over whether § 601(a) provides protection for individualswho marry in traditional ceremonies not recognized by their government and later seek asylumbased on the forced abortion or sterilization of their “common law spouses.” The Seventh andNinth Circuits have held that the statute covers spouses from traditional marriage ceremonies,see Zhang, 434 F.3d at 999; Zhu v. Gonzales, 465 F.3d 316, 321 (7th Cir. 2006); Ma v. Ashcroft,361 F.3d 553, 559-61 (9th Cir. 2004). In contrast, the Third Circuit, in Chen, supra, held that theamendment does not cover unmarried partners, even when they have been prevented frommarrying by their government’s family planning policy. 381 F.3d at 232-34; see also Chen v.Gonzales, 418 F.3d 110, 111 (1st Cir. 2005) (acknowledging circuit split on the issue). Whilethey have not reached the issue of traditional marriage ceremonies, the Fifth and EleventhCircuits have declined to extend IIRIRA § 601 to cover boyfriends of individuals who have beensubjected to a forced abortion or sterilization. See Zhang v. Ashcroft, 395 F.3d 531, 532 (5th Cir.2004); Wang v. U.S. Att’y Gen., 152 Fed. App’x 761, 767 (11th Cir. 2005) (unpublishedopinion).

5

1 The petition of Xian Zou is dismissed for lack of jurisdiction. We recognize that this decision3

2 creates a split among the circuits.4

3 I. BACKGROUND

4 Congress has given the Attorney General the discretionary authority to grant asylum to

5 an alien who qualifies as a “refugee” because he or she “is unable or unwilling to avail himself

Page 274: Sotomayor Cases

6

1 or herself of the protection of [his or her native country] because of persecution or a well-

2 founded fear of persecution on account of race, religion, nationality, membership in a particular

3 social group, or political opinion.” 8 U.S.C. § 1101(a)(42). A showing of past persecution gives

4 rise to a rebuttable presumption of a well-founded fear of future persecution. 8 C.F.R. §

5 1208.13(b)(1).

6 In 1996, Congress passed IIRIRA § 601(a), which amended 8 U.S.C. § 1101(a)(42) by

7 broadening its definition of “refugee,” adding the following language:

8 [A] person who has been forced to abort a pregnancy or to undergo involuntary9 sterilization, or who has been persecuted for failure or refusal to undergo such a

10 procedure or for other resistance to a coercive population control program, shall11 be deemed to have been persecuted on account of political opinion, and a person12 who has a well founded fear that he or she will be forced to undergo such a13 procedure or subject to persecution for such failure, refusal, or resistance shall be14 deemed to have a well founded fear of persecution on account of political15 opinion.1617 8 U.S.C. § 1101(a)(42).

18 The next year, the BIA held that “past persecution of one spouse can be established by

19 coerced abortion or sterilization of the other spouse,” so that spouses of individuals directly

20 victimized by coercive family planning policies are per se eligible for asylum pursuant to §

21 1101(a)(42). See Matter of C-Y-Z-, 21 I. & N. Dec 915, 917-18 (B.I.A. 1997) (en banc). The

22 BIA gave no reasons for reading the statute to compel this result.

23 Petitioner Lin entered the United States in January 1991 and filed an application for

24 asylum and withholding of removal in June 1993. According to Lin’s application, he had sought

25 the required governmental permission to marry his girlfriend and have children with her, but she

26 was too young under Chinese law. After his girlfriend became pregnant and was forced to have

Page 275: Sotomayor Cases

7

1 an abortion, Lin left China. His girlfriend remained in China because she was too weak to travel.

2 Following a hearing, the IJ found Lin credible, but concluded that he did not qualify for asylum

3 based on his girlfriend’s forced abortion and denied the petition. The BIA affirmed without

4 opinion. See In re Shi Liang Lin, No. A70 895 638 (B.I.A. Sept. 29, 2002), aff’g No. A70 895

5 638 (Immig. Ct. N.Y. City May 9, 2000).

6 Petitioner Dong attempted to enter the United States in October 1999, and was detained

7 by INS officials. When the INS commenced removal proceedings, Dong requested asylum,

8 withholding of removal, and relief under the Convention Against Torture (“CAT”). His asylum

9 petition was based on a claim that his fiancée (who continued to reside in China) had been forced

10 to undergo two abortions and that he would be jailed and fined for having left China illegally

11 were he to be deported. The IJ denied Dong’s petition, finding that, although he was credible, he

12 did not qualify for refugee status as a fiancé of a woman who had undergone forced abortions,

13 and had not established other grounds for relief. The BIA affirmed the IJ’s decision. See In re

14 Zhen Hua Dong, No. A77 293 661 (B.I.A. Sept. 25, 2002), aff’g No. A77 293 661 (Immig. Ct.

15 N.Y. City Oct. 12, 2000).

16 Petitioner Zou was taken into custody by the INS when he attempted to enter the United

17 States in September 2000. After removal proceedings commenced, Zou filed for asylum. Under

18 Chinese law, Zou and his girlfriend had been too young to marry. He claimed that she had been

19 forced to have an abortion and he had been threatened with arrest after protesting to family

20 planning officials. An IJ denied the application. The IJ found Zou’s testimony concerning his

21 threatened arrest incredible, doubting that Zou would be able to evade the Chinese authorities

22 and travel to the United States with an outstanding warrant of arrest from the Chinese

Page 276: Sotomayor Cases

8

1 government. Left only with Zou’s claim that his girlfriend had undergone a forced abortion, the

2 IJ held that Zou did not qualify for relief under § 1101(a)(42) on that basis. Zou appealed to the

3 BIA, which affirmed the decision of the IJ. See In re Zou, No. A77 322 595 (B.I.A. Aug. 27,

4 2002), aff’g No. A77 322 595 (Immig. Ct. N.Y. City Apr. 4, 2002). The three petitioners sought

5 review here and the cases were heard in tandem.

6 Three different interpretations of the application of “refugee” status to the facts in these

7 cases have been proposed by the parties throughout the litigation. When we heard the petitions

8 in tandem, the Department of Justice argued that § 601(a) of IIRIRA should be understood to

9 confer per se refugee status on spouses of individuals who had undergone forced abortions or

10 involuntary sterilizations, but not to boyfriends or fiancés, who were covered, if at all, under the

11 “other resistance” language of the amendment. Second, the petitioners argued, both before this

12 Court and before the BIA, that the distinction between spouses and unmarried partners was

13 arbitrary and that the statute should be interpreted to extend per se refugee status to the

14 committed partners of individuals who have been forced to undergo an abortion or involuntarily

15 sterilization.

16 The third, a text-based interpretation, was put forth before the BIA by the Department of

17 Homeland Security (“DHS”), which is “charged with the administration and enforcement of . . .

18 laws relating to the immigration and naturalization of aliens,” 8 U.S.C § 1103(a)(1). In its brief

19 to the BIA, the DHS insisted that no support existed in the statute for the BIA’s automatic

20 extension of § 601(a) to spouses. Specifically, the DHS asserted that the BIA’s interpretation

21 was “at odds with the most natural reading of the statute and with established principles of

22 asylum law.” Brief of DHS on Remand at 5, In re S-L-L-, 24 I. & N. Dec. 1 (B.I.A. 2006) (en

Page 277: Sotomayor Cases

9

1 banc) (No. A 70 895 638) (“DHS brief”). Carefully considering the text of § 601(a), the DHS

2 concluded that a rule conferring per se refugee status on spouses of individuals persecuted under

3 coercive family planning policies contradicted the unambiguous language of the amendment.

4 Further, the DHS stated that such a rule was at odds with the legislative history of the

5 amendment, as well as “a fundamental precept of U.S. refugee law . . . that, to qualify for

6 protection, an applicant must demonstrate that he will be targeted for his own protected belief or

7 characteristic.” DHS brief at 8-9, 11. The DHS understood that the statute required a “case-by-

8 case” application to individuals who have not themselves been victims of a forced abortion or

9 involuntary sterilization. Under this approach, spouses, boyfriends, and fiancés would have to

10 demonstrate their qualification for refugee status under the “for other resistance to a coercive

11 population control program” provision of § 601(a).

12 On remand, in S-L-L-, 24 I. & N. Dec 1, the BIA affirmed its earlier decision in C-Y-Z

13 that under § 601(a) an applicant whose spouse was forced to undergo an abortion or sterilization

14 procedure is automatically entitled to asylum, while limiting its interpretation to individuals who

15 were opposed to their legal spouse’s abortion or sterilization. Id. at 4. The BIA reasoned that §

16 601(a) provided “no clear or obvious answer to the scope of the protections . . . to partners of

17 persons forced to submit to an abortion or sterilization.” Id. As a result of this perceived

18 ambiguity, the majority chose to ground its interpretation “in the context of the history and

19 purposes of the Act as a whole,” finding that

20 absent evidence that the spouse did not oppose an abortion or sterilization21 procedure, we interpret the forced abortion and sterilization clause of section22 101(a)(42) of the Act, in light of the overall purpose of the amendment, to include23 both parties to a marriage.24

Page 278: Sotomayor Cases

10

1 Id. at 8. While the decision rested on “the forced abortion and sterilization clause of section

2 101(a)(42),” the Board noted that it was applying “general principles regarding nexus and level

3 of harm.” Id. at 5.

4 The majority based its conclusion on a number of policy-based factors, including the

5 positions taken by the litigants and the fact that C-Y-Z-, as a ten-year-old decision, constituted

6 long-standing precedent. The majority was also influenced by the shared responsibility of

7 married couples regarding family planning decisions under Chinese law and the “profound

8 impact” that a forced abortion or sterilization has on both parties to a marriage. Id. at 6-8.

9 The majority stopped short of extending a per se presumption to boyfriends, fiancés, and

10 other unmarried partners. It concluded that they were not comparable to husbands because “the

11 sanctity of marriage and the long term commitment reflected by marriage place the husband in a

12 distinctly different position from that of an unmarried father.” Id. at 9. The BIA dismissed the

13 appeals of Lin, id., and Dong, In re Zhen Hua Dong, No. A77 293 661 (B.I.A. Nov. 27, 2006),

14 and remanded Zou’s case to the immigration court “for further evidence on the issues of

15 resistance and harm” based on the “other resistance” clause of § 1101(a)(42), In re Xian Zou,

16 No. A73 178 541 (B.I.A. Nov. 21, 2006).

17 Board Member Pauley concurred. While deferring to the BIA’s prior decision in C-Y-Z

18 as long-standing and widely accepted precedent, he conceded that:

19 [w]ere we writing on a clean slate, I would adopt the lately arrived at position of20 the Department of Homeland Security . . . that whether or not the spouse of a21 forcibly sterilized or aborted individual can be found to have been persecuted22 depends on a case-by-case assessment of whether that spouse was persecuted on23 account of “other resistance” to a coercive population control system, because the24 language of the Act does not support extending refugee status to any person other25 than the one sterilized or aborted, aside from the “other resistance” ground.

Page 279: Sotomayor Cases

While here, as throughout the opinion, we refer to a male petitioner with a wife or5

girlfriend who has been forced to undergo an abortion or sterilization, our reasoning applies withequal force to the perhaps more uncommon situation in which a female petitioner’s male spouseor boyfriend has been forced to undergo sterilization.

11

1 Id. at 13 (Pauley, B.M., concurring).

2 Board Member Filppu, joined by Board Member Cole, concurred in the result but

3 dissented from the majority’s reaffirmation of C-Y-Z. They reasoned that the language of the

4 statute was unambiguous in “focus[ing] on ‘a person’ who has been forced to abort a pregnancy,

5 not on a ‘couple,’ let alone a married couple . . . .” Id. at 16 (Filppu, B.M., concurring and

6 dissenting). Understanding that “statutory interpretation must begin with reference to the

7 language and structure of the statute,” id. at 15, and “Congress expresses its intent through the

8 language it chooses,” id. at 19 (quoting INS v. Cardoza-Fonseca, 480 U.S. 421, 432 n.12

9 (1987)), the dissenters concluded that an individual who has not been subjected to a forcible

10 abortion or sterilization procedure can qualify for refugee status in relation to a coercive family

11 planning policy only if he proves that he was persecuted or has a well-founded fear of future

12 persecution “for other resistance to a coercive population control program.” Id.5

13 Following the BIA’s decision, we ordered an en banc rehearing to determine whether §

14 601(a) is ambiguous, whether the BIA’s interpretation of “refugee” should be accorded Chevron

15 deference, and “[w]hether the BIA reasonably construed IIRIRA § 601’s definition of “refugee”

16 to: (a) include a petitioner whose legally married spouse was subjected to an involuntary

17 abortion or sterilization . . . ; and (b) not include a petitioner whose claim is derivatively based

18 on any other relationship with a person who was subjected to such a procedure, unless the

19 petitioner has engaged in ‘other resistance’ to a coercive population control program . . . .” See

Page 280: Sotomayor Cases

We announced our intention to reach this question in our order that this case be reheard6

en banc. In that order we instructed the parties to address the BIA’s interpretation of § 601 as itrelated to both spouses and non-married partners of individuals subjected to an involuntaryabortion or sterilization. All members of this en banc panel joined the order, including thosewho now express confusion as to why we reach the question. We are particularly perplexed bythe position taken by our colleagues Judge Katzmann and Judge Sotomayor, who contend we areoverreaching by considering whether the BIA’s per se rule survives review under Chevron stepone, but who then proceed to assess, and declare valid, the same rule under Chevron step two. Op. of Judge Katzmann at 16, Op. of Judge Sotomayor at 1 & n.1.

12

1 En banc order. We now conclude that the BIA’s interpretation of the statute is not correct.

2 II. DISCUSSION

3 When reviewing the BIA’s interpretation of statutes that it administers, we apply the

4 Chevron principles. We first ask if Congress has spoken. “If the intent of Congress is clear, that

is the end of the matter; for the court, as well as the agency, must give effect to the5

unambiguously expressed intent of Congress.” Chevron, 467 U.S. at 842-43; see also INS v.6

7 Cardoza-Fonseca, 480 U.S. 421, 447-48 (1987). Only if the statute is silent or ambiguous do we

8 turn to the second inquiry – whether the BIA’s interpretation constitutes “a permissible

9 construction of the statute.” Chevron, 467 U.S. at 843.

10 While the petitioners in this case are unmarried partners, and not spouses, of individuals

11 who have been subjected to forced abortions, we review the BIA’s interpretation of § 601, as

12 articulated in its decision below, to extend a per se presumption of persecution to spouses, but

13 not to non-married partners, of individuals who have been involuntarily subjected to an abortion

14 or sterilization. It is the existence of this spousal policy that the petitioners argue is an arbitrary6

15 and capricious interpretation of the statute. If the BIA’s policy is at odds with the plain language

16 of the statute, it makes little sense to consider only whether it can reasonably be limited to

17 couples who are formally married.

Page 281: Sotomayor Cases

13

1 Accordingly, we start our Chevron analysis of § 601(a) by asking whether Congress has

2 spoken directly to the question of whether an individual can establish past persecution based

3 solely on his spouse or partner’s forced abortion or sterilization. 467 U.S. at 842-43. In S-L-L-,

4 the BIA, without elaboration or explanation, observed that “[t]here is no clear or obvious answer

5 to the scope of the protections afforded by the amendment to partners of persons forced to

6 submit to an abortion or sterilization.” 24 I. & N. Dec. at 4. We disagree. We conclude that

7 Congress has spoken to this issue and that it has done so unambiguously.

8 In the past, this Court, when following the BIA’s holding in C-Y-Z-, has deferred to the

9 BIA’s interpretation without performing a threshold Chevron analysis of the ambiguity of the

10 statute. In Yuan v. U.S. Dep’t of Justice, for example, we stated of IIRIRA § 601 that, “[b]y its

11 plain language, the law would seem to extend refugee status only to actual victims of persecution

12 – for example, a woman who was ‘forced to abort a pregnancy,’ but not her husband.” 416 F.3d

13 192, 196 (2d Cir. 2005). Despite this recognition of the amendment’s plain language, we went

14 on to defer to the interpretation of the BIA, stating that, “we have held that spouses of people

15 actually subject to persecution under coercive family planning policies are per se eligible for

16 asylum. . . . In doing so, however, we did not explain why. Rather, we followed the lead of the

17 BIA.” Id. at 196-97 (internal citation omitted); see also Zhang v. I.N.S., 386 F.3d 66, 73 (2d Cir.

18 2004); Qiu v. Ashcroft, 329 F.3d 140, 149 (2d Cir. 2003). To the extent that deference implicit

19 in these cases can be read to say that deference is due, they are overruled.

20 The amendment provides:

21 [(1)] a person who has been forced to abort a pregnancy or to undergo involuntary22 sterilization, or [(2)] who has been persecuted for failure or refusal to undergo23 such a procedure or [(3)] for other resistance to a coercive population control

Page 282: Sotomayor Cases

14

1 program, shall be deemed to have been persecuted on account of political opinion,2 and [(4)] a person who has a well founded fear that he or she will be forced to3 undergo such a procedure or [(5)] subject to persecution for such failure, refusal, 4 or [(6)] resistance shall be deemed to have a well founded fear of persecution on5 account of political opinion.67 8 U.S.C. § 1101(a)(42). The language of the first clause, which refers to “a person who has been

8 forced to abort a pregnancy or to undergo involuntary sterilization,” could not be more clear in

9 its reference to “a person,” rather than “a couple,” who has been subjected to a forced abortion or

10 involuntary sterilization. This interpretation follows two cardinal rules: first that “[s]tatutory

11 construction must begin with the language employed by Congress and the assumption that the

ordinary meaning of that language accurately expresses the legislative purpose,” Park ‘N Fly,12

13 Inc. v. Dollar Park & Fly, Inc., 469 U.S. 189, 194 (1985); and, second, that “we begin with the

14 understanding that Congress says in a statute what it means and means in a statute what it says

15 there,” Hartford Underwriters Ins. Co. v. Union Planters Bank, N.A., 530 U.S. 1, 6 (2000)

16 (internal quotation marks omitted).

17 As the statute is written, “a person who has been forced to abort a pregnancy”

18 unambiguously refers to a woman who has been physically subjected to a forced abortion, and “a

19 person who has been forced . . . to undergo involuntary sterilization” means an individual who

20 has physically undergone an involuntary medical procedure intended to result in infertility. Had

21 Congress intended this clause to refer to a spouse or partner of someone who has been physically

22 subjected to a forced procedure, “it could simply have said so.” Id. at 7.

23 Similarly, the second clause of the amendment refers to “a person” who “has been

24 persecuted for failure or refusal to undergo [an abortion or involuntary sterilization.]” 8 U.S.C. §

Page 283: Sotomayor Cases

No member of the en banc Court disputes this reading of IIRIRA § 601(a). Instead, our7

colleagues assert that other language in 8 U.S.C. § 1101(a)(42) – language predating adoption ofthe amendment – might (in the case of our colleague, Judge Calabresi) or does (in the case of ourcolleagues Judge Katzmann and Judge Sotomayor) somehow support the BIA’s per se rule.

15

1 1101(a)(42). Like the preceding clause, this language refers to individuals who have failed or

2 refused to undergo (i.e., “submit to”) a procedure affecting their own bodies. Under the

3 language used by Congress, having someone else, such as one’s spouse, undergo a forced

4 procedure does not suffice to qualify an individual for refugee status.

5 A parallel analysis governs the categories of § 601(a) relating to the establishment of a

6 well-founded fear of future persecution. The fourth category covers “a person who has a well

7 founded fear that he or she will be forced to undergo [an abortion or involuntary sterilization].”

8 This category plainly excludes an individual who does not fear that she herself will be subjected

9 to a forced abortion or sterilization. Not only does it refer to “a person,” and indicate that that

10 person will “undergo” the procedure his- or herself, the use of the pronouns “he” and “she”

11 reinforces the intention of Congress to limit the application of the clause to individuals who are

12 themselves physically forced to undergo an abortion or sterilization.

13 Finally, the fifth category of individuals refers to “a person who has a well founded fear

14 that he or she will be . . . subject to persecution for such failure [or] refusal [to undergo an

15 abortion or involuntary sterilization].” Like the language of the previous clauses, this phrase is

16 unambiguous in its reference to “a person” who fears that “he or she” will be persecuted for his

17 or her reaction to a threat of “undergo[ing]” an abortion or sterilization to his or her own body.

18 This section, like the previous three categories discussed, cannot be read reasonably to cover an

19 individual’s fears arising from a coercive procedure performed on someone else. 7

Page 284: Sotomayor Cases

These assertions are incorrect, we believe, for reasons discussed infra. By contrast, the Third Circuit’s recent decision in Sun Wen Chen, which our concurring

colleagues Judge Katzmann and Judge Sotomayor cite extensively even though they apparentlydisagree with its analysis, asserts that § 601(a) contains an ambiguity that the BIA is empoweredto fill. See Sun Wen Chen, 2007 WL 1760658, at *4 (“The C-Y-Z- rule . . . fleshes out an issuegermane to the application of [§ 601(a)] that was not addressed by Congress, and so poses noChevron step one problem.”). But see Sun Wen Chen, 2007 WL 1760658, at *10 (McKee, J.,dissenting) (“Rather than accept the language [of § 601(a)] as drafted, the majority concludesthat the absence of ‘spouse’ in the statute creates a vacuum that the Attorney General may rushin and fill, even though this results in amending the statute.”); id. at *11 (McKee, J., dissenting)(“Our analysis should therefore begin and end with the language of [§ 601(a)]. There is no roomhere for a step two inquiry under Chevron. . . . I believe Congress meant what it said, and I donot assume that the omission of any reference to a ‘spouse’ is accidental or insignificant.”).

16

1 In S-L-L-, the BIA noted that it was applying “general principles regarding nexus and

2 level of harm,” 24 I. & N. Dec. at 5 – principles derived from the general refugee definition

3 subsection of § 1101(a)(42) – to reach its holding. We believe that this approach was not

4 correct. The general definition of a “refugee” under § 1101(a)(42) permits “any person” who

5 experiences “persecution or a well-founded fear of persecution on account of race, religion,

6 nationality, membership in a particular social group, or political opinion” to obtain asylum. 8

7 U.S.C. § 1101(a)(42). Nothing in the general definition of refugee would permit “any person”

8 who has not personally experienced persecution or a well-founded fear of future persecution on a

9 protected ground to obtain asylum, as the BIA’s per se rule would permit. Indeed, an

10 examination of the overall statutory scheme reassures us that, pursuant to Chevron, we must

11 conclude that Congress has clearly and unambiguously spoken to the issue at hand. See FDA v.

12 Brown & Williamson Tobacco Corp., 529 U.S. 120, 132-33 (2000) (“In determining whether

13 Congress has specifically addressed the question at issue [under Chevron step one], a reviewing

14 court should not confine itself to examining a particular statutory provision in isolation. . . . A

Page 285: Sotomayor Cases

Judge Katzmann contends that we have employed this well-known canon of statutory8

construction to conclude that IIRIRA § 601(a) restricts, rather than expands, the availability ofasylum relief. Op. of Judge Katzmann at 5. This contention is misplaced. We acknowledge thatIIRIRA § 601(a) expands the availability of asylum, but find it significant that it does so only tospecifically-enumerated categories of asylum applications – categories that do not include thebeneficiaries of the BIA’s per se rule.

17

1 court must . . . interpret the statute as a symmetrical and coherent regulatory scheme.” (citation

2 and internal quotation marks omitted)).

3 It is apparent to us that when Congress rejected the BIA’s view in Matter of Chang, 20 I.

4 & N. Dec. 38 (B.I.A. 1989), of birth control policies in other nations as an avenue for asylum, it

5 did so in clear and unmistakable language. It identified those to whom asylum could be granted

6 and reaffirmed the need for direct personal persecution. Congress’s specific designation of some

7 persons (i.e., those who fear, resist, or undergo particular medical procedures) is incompatible

8 with the view that others (e.g., their spouses) should also be granted asylum per se because of

9 birth control policies. The inclusion of some obviously results in the exclusion of others. See

10 TRW Inc. v. Andrews, 534 U.S. 19, 28 (2001); Leatherman v. Tarrant County Narcotics

Intelligence and Coordination Unit, 507 U.S. 163, 168 (1993).811

12 The language of § 601(a) does nothing to alter the pre-IIRIRA definition of “political

13 opinion” in § 1101(a)(42), and this further demonstrates the exclusivity of the group of persons

14 entitled to asylum per se under § 601(a). Congress could have announced that the term “political

15 opinion” included any reproductive act in violation of a coercive population control program, but

16 instead it chose to create a specific exception to the general statutory requirement that a person

17 claiming refugee status based on past persecution has the burden of demonstrating that the

18 particular conduct experienced by him rose to the level of persecution and the persecution had a

Page 286: Sotomayor Cases

To “deem” is “[t]o treat (something) as if (1) it were really something else, or (2) it has9

qualities that it doesn’t have,” or, in the alternative, “[t]o consider, think, or judge.” Black’s LawDictionary 446 (8th ed. 2004). Black’s Law Dictionary further notes that the word deem “hasbeen traditionally considered to be a useful word when it is necessary to establish a legal fictioneither positively by 'deeming' something to be what it is not or negatively by 'deeming'something not to be what it is. . . . All other uses of the word should be avoided.” Id. (quotingG.C. Thornton, Legislative Drafting 99 (4th ed. 1996)).

18

1 specified impermissible nexus. See 8 U.S.C. § 1158(b)(1)(B)(i) (stating that “[t]he burden of

2 proof is on the applicant to establish that the applicant is a refugee”). IIRIRA § 601(a) states that

3 certain individuals affected by coercive population control programs “shall be deemed”

4 persecuted by reason of political opinion. In using the word “deem” in this context, § 601(a)9

5 makes clear that those who benefit from the amendment would not be entitled to per se political

6 opinion asylum relief absent the amendment. In other words, their political opinion exists de

7 jure rather than as a matter of fact on which the applicant bears the burden of proof. For an

8 asylum applicant who does not fall within this limited exception, the burden remains on the

9 applicant – and the opportunity remains open – to demonstrate, in light of the particular facts of

10 the case, that he has (i) a well-founded fear of personal persecution (ii) based on political opinion

11 or some other impermissible ground.

12 This is consistent with what we know: While it is plain that suffering a forced medical

13 procedure can be a persecution if it is on account of a protected ground, the conception of a child

14 is no more an expression of political opinion than birth, death, sleep, or the taking of

15 nourishment. If the language of § 601(a) indicates that the woman who is subjugated to the

16 outrage of a forced abortion has not herself been persecuted for the “political opinion” of

17 conceiving a child under pre-IIRIRA § 1101(a)(42), then so much less the man who has

Page 287: Sotomayor Cases

19

1 impregnated her; but unlike his wife or partner, he is not “deemed” under § 601(a) to hold a

2 political opinion and he must prove the existence of a political opinion or other protected ground

3 under § 1158(b)(1)(B)(i). Accordingly, we conclude that the statutory scheme unambiguously

4 dictates that applicants can become candidates for asylum relief only based on persecution that

5 they themselves have suffered or must suffer. See Sun Wen Chen, 2007 WL 1760658, at *11

6 (McKee, J., dissenting) (“Congress could have easily drafted [§ 601(a)] to extend to ‘married

7 couples who have been subjected to a forced abortion or involuntary sterilization.’ So drafted,

8 an actual victim of persecution under a coercive population control program, as well as his/her

9 spouse, would qualify for relief under the statute. However, Congress did not draft the statute in

10 this way, and we can not rewrite the statute's explicit text to achieve that result.”).

11 Indeed, the critical defect in the BIA’s policy of according per se refugee status to

12 spouses of individuals explicitly protected by § 601(a) is its creation of an irrebuttable

13 presumption of refugee status for a new class of persons. This policy effectively absolves large

14 numbers of asylum applicants of the statutory burden to prove that they have (i) a well-founded

15 fear of persecution (ii) based on an impermissible nexus. Such a presumption is contrary to the

16 text of 8 U.S.C. § 1158(b)(1)(B), which specifies that “[t]he burden of proof is on the applicant,”

17 that “the applicant must establish that . . . political opinion was or will be at least one central

18 reason for persecuting the applicant,” and that an applicant’s testimony may be sufficient to meet

19 this burden only if it “refers to specific facts sufficient to demonstrate that the applicant is a

20 refugee.” The law is clear that “an agency is not free to ignore statutory language by creating a

21 presumption on grounds of policy to avoid the necessity for finding that which the legislature

Page 288: Sotomayor Cases

20

1 requires to be found.” United Scenic Artists v. NLRB, 762 F.2d 1027, 1034 (D.C. Cir. 1985).

2 The creation of such a presumption “is beyond the [agency’s] statutory authority.” Id. at 1035;

3 see also Cerrillo-Perez v. INS, 809 F.2d 1419, 1426-27 (9th Cir. 1987) (holding that the “BIA

4 cannot adopt a general presumption” unfavorable to applicant but instead “must consider the

5 specific facts and circumstances of each case”). Thus, the BIA lacks authority to adopt a policy

6 that presumes that every person whose spouse was subjected to a forced abortion or sterilization

7 has himself experienced persecution based on political opinion.

8 Our concurring colleagues Judge Katzmann and Judge Sotomayor suggest that the

9 political nexus prong of this presumption is, in fact, inherent in Congress’s 1996 amendments.

10 Op. of Judge Katzmann at 15; op. of Judge Sotomayor at 7. However, this contention is

11 irreconcilable with the language of § 601(a), in which Congress created this presumption for

12 specifically identified persons – those who were themselves subjected to or threatened with a

13 forcible abortion or sterilization. To the extent that the amendments overruled Matter of

14 Chang’s categorical rejection of such a nexus, 20 I. & N. Dec. 38, this means that an asylum

15 applicant is no longer foreclosed from relying on evidence of a forced abortion or sterilization –

16 whether personal or spousal – in attempting to demonstrate persecution based on political

17 opinion. However, the fact remains that Congress has relieved only persons who actually

18 experienced, or are threatened with, a forcible abortion or sterilization from the burden of

19 proving a political nexus in their particular cases.

20 We do not deny that an individual whose spouse undergoes, or is threatened with, a

21 forced abortion or involuntary sterilization may suffer a profound emotional loss as a partner and

Page 289: Sotomayor Cases

If this conclusion is inconsistent with Congress’s intentions, it can, if it so chooses, of10

course, amend the statute, as it did when it adopted IIRIRA § 601(a) in response to the BIA’sdecision in Matter of Chang, 20 I. & N. Dec. 38 (B.I.A. 1989).

21

1 a potential parent. But such a loss does not change the requirement that we must follow the

2 “ordinary meaning” of the language chosen by Congress, according to which an individual does

3 not automatically qualify for “refugee” status on account of a coercive procedure performed on

4 someone else. See Moskal v. United States, 498 U.S. 103, 108 (1990) (stressing the importance

5 of “giving the ‘words used’ their ‘ordinary meaning.’”); Cardoza-Fonseca, 480 U.S. at 453

6 (“Where the language of [a] law[] is clear, we are not free to replace it with an unenacted

7 legislative intent.”).10

8 Under Chevron, once it is apparent that the statute is unambiguous, our inquiry stops. “If

9 the intent of Congress is clear, that is the end of the matter,” and we are required to refrain from

10 deferring to an agency’s contradictory interpretation. 467 U.S. at 842-43; see, e.g., Barnhart v.

11 Sigmon Coal Co., Inc., 534 U.S. 438, 462 (2002); Brown & Williamson Tobacco Corp., 529 U.S.

12 at 160-61; Nat’l Credit Union Admin. v. First Nat. Bank & Trust Co., 522 U.S. 479, 500 (1998);

13 MCI Telecommunications Corp. v. American Tel. & Tel. Co., 512 U.S. 218, 226-28 (1994);

14 Cardoza-Fonseca, 480 U.S. at 447-48. Congress’s choices of language in the phrases, “a person

15 who has been forced to abort a pregnancy or to undergo involuntary sterilization,” “[a person]

16 who has been persecuted for failure or refusal to undergo such a procedure,” and “a person who

17 has a well founded fear that he or she will be forced to undergo such a procedure or subject to

18 persecution for such failure [or] refusal,” is uniformly unambiguous in its reference to an

19 individual who is subjected to, or threatened with, an involuntary abortion or sterilization

Page 290: Sotomayor Cases

While no party before us argues that the rule in S-L-L- fails under Chevron step one,11

DHS did argue this point before the BIA, and the BIA considered and rejected the argument overa persuasive dissent by two members of the Board. In any event, we cannot defer to theDepartment of Justice’s argument (opposed below by DHS, the agency charged with enforcingimmigration laws) that the rule in S-L-L- survives review under Chevron step one if the rulefinds no support in the statutory text. Accordingly, we assume that the Solicitor General willtake appropriate action to recommend or assure that the views of DHS and this Court will berepresented in any future proceedings.

22

1 affecting his or her own body, and the statutory scheme of § 1101(a)(42) only reinforces that

2 conclusion.

3 As a result, we conclude that the statute does not provide that a spouse – and a fortiori, a

4 boyfriend or fiancé – of someone who has been forced to undergo, or is threatened with, an

5 abortion or sterilization is automatically eligible for “refugee” status. Instead, to qualify for

6 refugee status under the amendment, such an individual must turn to the two remaining

7 categories of § 601(a), which provide protection to petitioners who demonstrate “other resistance

8 to a coercive population control program” or “a well founded fear that he or she will be . . .

9 subject to persecution for such . . . resistance . . . .” 8 U.S.C. § 1101(a)(42). 11

10 In S-L-L, the BIA was influenced by the fact that C-Y-Z- “is a precedent of long standing

11 at this point . . . and numerous court decisions have deferred to the holding.” S-L-L-, 24 I. & N.

12 Dec. at 4; see also id. at 14 (Pauley, B.M., concurring) (“[N]otwithstanding my belief that Matter

13 of C-Y-Z-, . . . was wrongly decided, I would not overrule it now, clearly a decade later and in the

14 aftermath of thousands of decisions applying it . . . .”). However, the Supreme Court has made

15 clear that “[s]tare decisis is not an inexorable command; rather, it is a principle of policy and not

16 a mechanical formula of adherence to the latest decision.” Payne v. Tennessee, 501 U.S. 808,

17 828 (1991) (internal quotation marks omitted).

Page 291: Sotomayor Cases

23

1 While stare decisis is undoubtedly of considerable importance to questions of statutory

2 interpretation, the Supreme Court “ha[s] never applied stare decisis mechanically to prohibit

3 overruling . . . earlier decisions determining the meaning of statutes.” Monell v. Dep’t of Social

4 Servs., 436 U.S. 658, 695 (1978). We should not do so either. The fact that we have failed to

5 follow the plain language of a law of Congress for ten years does not require that we do so

6 indefinitely. That would “place on the shoulders of Congress the burden of the Court's own

7 error.” Girouard v. United States, 328 U.S. 61, 70 (1946).

8 Given the clarity of the statute, there is no need to resort to legislative history, which is a

9 tool of construction that we employ only if the statutory text at issue in the context of the statute

10 as a whole is ambiguous. However, were we to examine the statute’s legislative history, we

11 would find that our interpretation of § 601(a) comports with Congress’s stated purpose in

12 passing the amendment. The House Report accompanying the passage of the amendment states

13 that its

14 primary intent . . . is to overturn several decisions of the Board of Immigration15 Appeals, principally Matter of Chang and Matter of G- . . . which . . . hold that a16 person who has been compelled to undergo an abortion or sterilization, or has17 been severely punished for refusal to submit to such a procedure, cannot be18 eligible on that basis for refugee or asylee status unless the alien was singled out19 for such treatment on account of factors such as religious belief or political20 opinion.2122 H.R. Rep. 104-469(I) at 173-74 (1996) (emphasis added).

23 The report mentions as examples of victims of coercive family planning policies women

24 who have been subjected to involuntary abortions, men and women who are forcibly sterilized,

25 and “couples with unauthorized children [who are] subjected to excessive fines,” but not spouses

Page 292: Sotomayor Cases

Judge Katzmann asserts that Congress, when it adopted IIRIRA § 601(a) in 1996,12

intended to accomplish the same goal as a 1993 order of former Attorney General Barr that wasnever enacted and never reviewed by any court. That order would have explicitly grantedasylum to spouses of coercive family planning policy victims. Op. of Judge Katzmann at 8-9. Judge Katzmann’s assertion is squarely contradicted by the plain language of IIRIRA § 601(a),which – unlike Attorney General Barr’s order – does not grant asylum to spouses of persecutionvictims, as well as by the legislative history recounted above, which emphasizes that IIRIRA §601(a) was not intended to make asylum available to those not explicitly protected by theamendment. Even more precarious is Judge Katzmann’s reliance on the various messages he

24

1 of individuals who have been subjected to forced abortions or sterilizations. Id. While Congress

2 disapproved of coercive family planning policies as a whole, the amendment was meant to

3 provide protection for individuals who were subjected to persecution themselves. As the report

4 goes on to state:

5 The Committee emphasizes that the burden of proof remains on the6 applicant, as in every other case, to establish by credible evidence that he7 or she has been subject to persecution - in this case, to coercive abortion or8 sterilization - or has a well-founded fear of such treatment. The Committee9 is aware that asylum claims based on coercive family planning are often

10 made by entire groups of smuggled aliens, thus suggesting that at least11 some of the claims, if not the majority, have been "coached." Section12 [601(a)] is not intended to protect persons who have not actually been13 subjected to coercive measures or specifically threatened with such14 measures . . . .15 16 Id. at 174 (emphasis added). There is nothing in the legislative history that leads us to question

17 “the strong presumption that Congress expresses its intent through the language it chooses.”

18 Cardoza-Fonesca, 480 U.S. at 432 n.12. Here, the language Congress employed in § 601(a)

19 demonstrates that it wanted to cover “a person,” not “a couple,” not a “significant other” and not

20 an “intimate friend.” Moreover, Congress’s “emphasi[s]” on its intention that “the burden of

21 proof remains on the applicant” demonstrates that no other subsection of § 1101(a)(42) could

22 support the BIA’s interpretation of the statute. See Sun Wen Chen, 2007 WL 1760658, at *1612

Page 293: Sotomayor Cases

hears in the sounds of Congressional silence. Op. of Judge Katzmann at 13.

In her concurring opinion, Judge Sotomayor suggests that today’s holding casts doubt13

on Jorge-Tzoc v. Gonzales, 435 F.3d 146 (2d Cir. 2006) (remanding asylum claim to BIA ondetermination that a “combination of circumstances” experienced by applicant in Guatemala as adependent child – including the massacre of close family members – “could well constitute[past] persecution” of the child). We do not address this concern because today’s decision doesnot preclude the BIA from considering the totality of circumstances in any particular case todetermine if an asylum applicant has carried his statutory burden.

25

1 (McKee, J., dissenting) (“The House Report . . . expresses a congressional intent to restrict

2 asylum to the ‘person’ who undergoes the coercive procedure just as clearly as the text of the

3 statute.”).

4 This reading of the statute is further supported by the Supreme Court’s pronouncement

5 about what “refugee” means. In INS v. Elias-Zacharias, the Supreme Court held that under the

6 plain language of the general refugee definition subsection of § 1101(a)(42), “‘persecution on

7 account of . . . political opinion’ in [§ 1101(a)(42)] is persecution on account of the victim’s

8 political opinion,” 502 U.S. 478, 482 (1992) (emphasis in original), not persecution on account

9 of a political opinion held by someone else. Similarly, a spouse who has not demonstrated that

10 he himself is a victim of persecution cannot be entitled to asylum under this section of the

11 statute. It would be unreasonable to conclude otherwise, whether under § 601(a), or under §

12 1101(a)(42) as a whole.13

13 Our conclusion that Congress never intended § 601 – or § 1101(a)(42) – to apply

14 automatically to spouses is reinforced by the fact that Congress already provides for family

15 members elsewhere in the statute by authorizing derivative asylum status for spouses and

16 children of individuals who qualify as “refugees.” 8 U.S.C. § 1158(b)(3)(A). Specifically, under

Page 294: Sotomayor Cases

26

1 § 1158(b)(3)(A), an individual whose spouse or parent has been granted asylum on the basis of

2 having undergone or been threatened with the prospect of a forced abortion or sterilization is

3 automatically eligible for derivative asylum: “[a] spouse or child . . . of an alien who is granted

4 asylum under this subsection may, if not otherwise eligible for asylum under this section, be

5 granted the same status as the alien if accompanying, or following to join, such alien.”

6 What Congress did by providing in § 1158(b)(3)(A) for automatic asylum for spouses of

7 individuals who have been granted “refugee” status as a result of having been forced to undergo

8 an abortion or involuntary sterilization reflects a policy decision to encourage the preservation of

9 families. Under § 1158(b)(3)(A) the benefits of asylum are extended first to the person Congress

10 understood to be most deserving of protection – the direct victim. Once the victim gains asylum,

11 so does the spouse, and so do their children. This structure encourages couples to remain

12 together, or, in circumstances where this is not possible, facilitates reunion.

13 T he BIA’s interpretation of the statute in S-L-L cuts in a different direction since it has

14 the perverse effect of creating incentives for husbands to leave their wives. As hundreds of

15 cases in the courts illustrate, the policy allows a married man to “capitalize on the persecution of

16 his wife to obtain asylum even though he has left his wife behind and she might never join him

17 and he might intend that she not do so,” Chen v. Ashcroft, 376 F.3d 215, 223 n.2 (3d Cir. 2004).

18 It is highly unlikely – indeed, inconceivable – that Congress would approve of, much less

19 intended, any of this. These counterintuitive results reinforce our conclusion that in § 601(a)

20 Congress intended to grant automatic asylum to an individual directly victimized by a coercive

21 birth control policy, and that no part of § 1101(a)(42) extends this automatic asylum to spouses.

Page 295: Sotomayor Cases

For an analysis of what “resistance” might mean when someone has not been forcibly14

sterilized himself, see Li v. Ashcroft, 356 F.3d 1153, 1159-61 (9th Cir. 2004) (en banc)(analyzing the meaning of the “other resistance” clause and holding that it applies to a womanwho announced her opposition to government population control policies and was thereaftersubjected to a forced gynecological exam and threatened with future abortion, sterilization of herboyfriend, and arrest).

27

1 Congress did not, at the same time it adopted § 601(a), intend to gut or to undermine §

2 1158(b)(3)(A), which already contained a synchronous, sensible way of addressing the

3 compelling problems faced by spouses and children of direct victims.

4 Although we conclude that Congress has spoken unambiguously to whether an asylum

5 applicant is per se eligible for asylum if his spouse or partner has suffered as a result of a

6 coercive population control program, the phrase “other resistance” is ambiguous and leaves

7 room for the BIA’s reasonable interpretation where the applicant relies on something beyond his

8 spouse’s or partner’s persecution. See Yang v. U.S. Att’y Gen., 418 F.3d 1198, 1203 (11th Cir.14

9 2005) (“There is very little case law analyzing the ‘other resistance’ clause in the asylum statute.

10 Additionally, a review of the legislative history behind the 1996 Amendment does not reveal any

11 clear intent from Congress on the scope of the ‘other resistance clause.’”).

12 In its decision, the BIA held that an applicant claiming persecution for “other resistance”

13 must demonstrate (1) “resistance” to a coercive family planning policy, which can “cover[] a

14 wide range of circumstances, including expressions of general opposition, attempts to interfere

15 with enforcement of government policy in particular cases, and other overt forms of resistance to

16 the requirements of the family planning law”; and (2) that the applicant has “suffered harm

17 amounting to persecution on account of that resistance.” S-L-L-, 24 I. & N. Dec. at 10. An

18 individual whose spouse or partner has been subjected to a forced abortion or involuntary

Page 296: Sotomayor Cases

Judge Calabresi asserts first that our holding conflicts—if not in actuality, then in15

“spirit,” Op. of Judge Calabresi at 6 —with the Supreme Court’s recent decisions in INS v.Orlando Ventura, 537 U.S. 12 (2002) (per curiam), and Gonzales v. Thomas, 547 U.S. 183, 126S. Ct. 1613 (2006) (per curiam), and second that we rush to reach a particular result. Wedisagree with both of these assertions. First, Orlando Ventura and Thomas held that a reviewingcourt should ordinarily remand rather than pass upon a matter that is (1) primarily committed tothe BIA’s discretion, and (2) has not yet been considered by the BIA. See Orlando Ventura, 537U.S. at 16-17; Thomas, 126 S. Ct. at 1615. Neither of these conditions is present in this case. We, rather than the BIA, have primary authority under Chevron to determine whether aparticular agency interpretation is consistent with the unambiguously expressed intent of

28

1 sterilization can therefore qualify for “refugee” status under this interpretation if that individual

2 can prove past persecution or a fear of future persecution for “resistance” that is directly related

3 to his or her own opposition to a coercive family planning policy.

4 Whatever interpretation the BIA chooses to give to the meaning of “resistance,” it is clear

5 that the fact that an individual’s spouse has been forced to have an abortion or undergo

6 involuntary sterilization does not, on its own, constitute resistance to coercive family planning

7 policies. See Zhang, 395 F.3d at 532 (“[M]erely impregnating one's girlfriend is not alone an act

8 of ‘resistance.’”). Nor could the resistance of an individual’s spouse or partner to a family

9 planning policy – whether by failure or refusal to undergo a procedure, or for “other resistance”

10 – constitute, on its own, “resistance” under regulations implementing § 1101(a)(42). Instead, as

11 the DHS has argued, “where the applicant himself has not resisted [coercive family control

12 policies], he would need to demonstrate, though persuasive direct or circumstantial evidence,

13 that his wife’s, fiancee’s, or girlfriend’s resistance has been or will be imputed to him.” DHS

14 brief at 17 (citing Singh v. INS, 134 F.3d 962, 970 (9th Cir. 1998)). The fact that someone’s

15 spouse has been subjected to a forced abortion or sterilization would not be irrelevant to such an

16 analysis, it simply could not provide for asylum status per se. 15

Page 297: Sotomayor Cases

Congress. See, e.g., Cardoza-Fonseca, 480 U.S. at 447-48. Second, the BIA has had ample opportunity to consider the statutory interpretation

question in the first instance. The per se rule that we now invalidate was first announced by theBIA in 1997, in its opinion in C-Y-Z-, 21 I. & N. Dec. at 915. In 2005, we remanded this case tothe BIA to give it the opportunity to reconsider whether the rule in C-Y-Z- could find support inthe language of § 601(a). See Lin, 416 F.3d at 187. On remand, DHS explicitly argued to theBIA that its per se rule was foreclosed by the plain language of § 601(a) and the statutoryscheme. A majority of the BIA considered and rejected this argument over the forceful, andpersuasive, objections of a minority of the Board. See S-L-L-, 24 I. & N. Dec. at 15-21 (Filppu,B.M., concurring and dissenting). We see no reason to remand yet again – ping pong style – when the BIA has had ten years and several opportunities to reconsider a rule that has no basis instatutory text.

29

1 Before turning to the dispositions of the petitioners’ claims, we address some practical

2 implications of our decision. We affirm the result of the BIA’s decision in S-L-L- denying per

3 se refugee status to boyfriends or fiancés of individuals who have been persecuted directly under

4 coercive family planning policies. A necessary predicate for this result is our conclusion that §

5 601 does not confer automatic asylum eligibility on spouses, whether legal spouses or spouses

6 from a traditional marriage, but only on individuals who themselves have undergone or been

7 threatened with coercive birth control procedures. Thus, although none of the petitioners before

8 us is legally married, we understand that our reading of the statute would necessarily exclude

9 spouses of those directly victimized from per se asylum eligibility as well.

10 We emphasize that our holding today should not be read to presage the reopening of

11 cases of aliens who have already been granted asylum based on the BIA’s interpretation of § 601

12 in C-Y-Z-. Under 8 U.S.C. § 1158(c)(2)(A), the DHS is permitted to seek the termination of

13 asylum when an alien no longer qualifies for refugee status “because, owing to a fundamental

14 change in circumstances relating to the original claim, the alien's life or freedom no longer would

15 be threatened on account of race, religion, nationality, membership in a particular social group,

Page 298: Sotomayor Cases

30

1 or political opinion in the country from which deportation or removal was withheld.” 8 C.F.R. §

2 208.24(b)(1) (emphasis added). Just as a change in United States asylum law does not qualify as

3 a “change in circumstances” sufficient to reopen an asylum case under 8 C.F.R. §

4 1003.2(c)(3)(ii) (permitting motion to reopen “based on changed circumstances arising in the

5 country of nationality or in the country to which deportation is ordered”), see Azanor v. Ashcroft,

6 364 F.3d 1013, 1022 (9th Cir. 2004), a change in the BIA’s interpretation of section 601(a) as a

7 result of our decision should not be seen as a “fundamental change in circumstances relating to

8 the original claim” under 8 C.F.R. § 208.24(b)(1) so as to allow the termination of an asylum

9 claim that has already been granted. See S-L-L-, 24 I. & N. Dec. at 21 n.2 (Filppu, B.M.,

10 concurring and dissenting) (“We are not now concerned with reopening past cases.”).

11 III. PETITIONERS’ CLAIMS

12 We agree with the BIA that none of the petitioners can qualify for automatic refugee

13 status as a result of the treatment of their girlfriends or fiancées. Instead, each petitioner must

14 demonstrate “other resistance to a coercive population control program” or “a well founded fear

15 that he . . . will be . . . subject to persecution for such . . . resistance . . . .” 8 U.S.C. §

16 1101(a)(42).

17 Petitioner Dong’s application for asylum was based upon his fiancée’s two forced

18 abortions and threats from family planning officials that they would fine and sterilize Dong if his

19 fiancée became pregnant again. But Dong failed to demonstrate that he acted in a manner that

20 could constitute “resistance” or opposition to a coercive family control program. Nor can we

21 find that Dong has a fear of future persecution as a result of the threat that the Chinese

Page 299: Sotomayor Cases

31

1 government would sterilize him if his fiancée became pregnant again. Dong submitted evidence

2 to the immigration court that his fiancée had moved to Taiwan to be with her father. The IJ thus

3 found that Dong was unlikely to return to China, and his fear of sterilization was conjectural.

4 Moreover, as the BIA notes, were Dong’s fiancée to return to China, Dong and his fiancée would

5 now meet the age requirements for marriage. See In re Dong, A. 77 293 661 (B.I.A. Nov. 27,

6 2006). In addition, the immigration judge correctly determined that Dong did not demonstrate

7 that he would “more likely than not” be persecuted as grounds for a withholding of removal, or

8 that he would be subjected to torture within the meaning of the Convention Against Torture. See

9 id. Accordingly, we deny Dong’s petition for review.

10 Petitioner Lin’s claim for asylum is that he and his girlfriend were refused permission to

11 marry or have a child out of wedlock, and his girlfriend was forced to undergo an abortion. Lin

12 did not claim before the immigration court, the BIA, or this Court that he had “otherwise

13 resisted” China’s coercive family planning policies. Even if he had, we agree with the BIA that

14 a request, through the appropriate legal channels, for permission to have a child, combined with

15 the subsequent abortion performed on his girlfriend, does not constitute “resistance to a coercive

16 population control program.” S-L-L-, 24 I. & N. at 10-12. However, we do not have jurisdiction

17 over Lin’s petition for review, as we find that his petition is moot. Lin’s attorney has informed

18 this Court that he has never had any contact with the petitioner, and that Lin’s prior attorney has

19 not spoken to him in approximately three years. In addition, an individual from Lin’s village in

20 China has told the attorney that “he heard from other villagers that Lin was terminally ill and had

21 returned to China not too long ago and died.” Aff. of Yee Ling Poon ¶ 5. When the possibility

Page 300: Sotomayor Cases

32

1 of relief is “so remote and speculative that any decision on the merits . . . would amount to a

2 ‘declar[ation of] principles or rules of law which cannot affect the matter in issue in the case

3 before [us],’” United States v. Blackburn, 461 F.3d 259, 262 (2d Cir. 2006) (quoting Mills v.

4 Green, 159 U.S. 651, 653 (1895) (alterations in original)), we would run afoul of Article III were

5 we to decide the claim. Given that we do not retain jurisdiction over Lin’s claim if he has

6 returned to China and has provided no explanation to overcome the presumption that his asylum

7 application has been abandoned, see 8 C.F.R. § 1208.8, or if he is deceased, Lin’s petition is

8 dismissed as moot.

9 Petitioner Zou’s petition has been remanded by the BIA to the immigration court to

10 review its findings of adverse credibility and determine whether Zou qualifies as a refugee for

11 “resistance” to a coercive family planning policy. See In re Xian Zou, No. A77 322 295 (B.I.A.

12 Nov. 21, 2006). Accordingly, we no longer have jurisdiction over Zou’s petition. See 8 U.S.C.

13 § 1252(a)(1) (giving courts of appeals jurisdiction to review only final orders of removal).

14 CONCLUSION

15 For these reasons, the petition for review of Zhen Hua Dong is DENIED. The petitions

16 for review of Shi Liang Lin and Xian Zou are DISMISSED for lack of jurisdiction.

17

18

19

20

21

Page 301: Sotomayor Cases

33

1

2

3

4

5

6

7

8

9

10

11

12

Page 302: Sotomayor Cases

The Department of Homeland Security (“DHS”) advanced a different view before the1

BIA. The majority “assume[s] that the Solicitor General will take appropriate action torecommend or assure that the views of DHS and this Court will be represented in any futureproceedings.” Maj. Op. at 23 n.11. In so doing, the majority ignores the historic independenceof the Office of the Solicitor General in determining the executive branch’s position before theSupreme Court.  When agencies of the executive branch have taken inconsistent positions, asthey have here, the Solicitor General may choose among those positions, or it may adopt anyother available litigation position. The Solicitor General has no obligation to endorse thepreferred legal theory of the court below.  To the extent the majority attempts to influence theposition the Solicitor General will take in future proceedings, that effort is inappropriate.

34

1 KATZMANN, Circuit Judge, with whom Judges Straub, Pooler, and Sotomayor join, concurring in

2 the judgment:

3 With the majority’s emphasis on denying asylum relief to legal spouses under

4 § 101(a)(42) of the Immigration and Nationality Act (the “INA”), it is easy to lose sight of one

5 fact central to the disposition of these cases: Not one of the petitioners in these consolidated

6 cases was married. In each case, it was the plaintiff’s girlfriend, not his wife, who was forced to

7 abort her pregnancy. It is thus unnecessary for us to resolve whether the BIA can legally extend

8 asylum relief to legal spouses; indeed, in doing so, we are addressing not only an issue not

9 presently before us, but also one that the parties in these cases do not even dispute. In their

10 briefs before us, both the petitioners and the Government agree that the statute is ambiguous. 1

11 The question the parties dispute, and the only one that these cases require us to answer, is

12 whether the BIA’s interpretation of the statute as applied to boyfriends and fiancés is reasonable.

13 Every judge on this Court who reaches this issue agrees that it is.

14 Thus, this case could have been resolved simply and nearly unanimously by assuming the

15 reasonableness of the BIA’s construction of the statute as applied to legal spouses and then

16 holding that it was also reasonable as applied to boyfriends and fiancés. See Cai Luan Chen v.

Page 303: Sotomayor Cases

In a related context, we remanded to the BIA to address the proper scope of the term2

“refugee” in the first instance, noting that “it would be unsound for each of the several Courts ofAppeals to elaborate a potentially nonuniform body of law” and describing uniformity as“especially desirable in cases such as these.” Jian Hui Shao v. BIA, 465 F.3d 497, 502 (2d Cir.2006).

The majority finds my decision to engage in this analysis “perplex[ing]” given my belief3

that we need not—and should not—answer today a question that this case does not require us toanswer and the proper resolution of which the parties do not dispute. Although I believe weshould have limited our decision to the BIA’s treatment of boyfriends, the majority has

35

1 Ashcroft, 381 F.3d 221, 227 (3d Cir. 2004) (Alito, J.) (assuming “for the sake of argument” that

2 “C–Y–Z–’s interpretation is permissible” as applied to legal spouses and then determining

3 whether the BIA was reasonable in distinguishing “between married and unmarried couples”).

4 Instead, the majority has gone out of its way to create a circuit split where none need exist, see

5 Maj. Op. at 5 n.4, thereby frustrating the BIA’s uniform enforcement of a national immigration

6 policy. Finding in textual silence an expression of unambiguous congressional intent, the2

7 majority has rejected the BIA’s determination that § 101(a)(42) is ambiguous.

8 When a governmental body with substantial experience in interpreting a complex

9 statutory scheme concludes that a statute is ambiguous, that determination should give us pause.

10 Here, the fact that the BIA concluded that the INA is ambiguous with respect to the question we

11 are called upon to answer suggests that we would do well to probe further, to consider whether

12 the seemingly plain language belies a more complicated meaning. It suggests that we should

13 consider carefully not only the text of the statute, but also the context—both the events that gave

14 rise to that text and the various agency and judicial responses to it. Text without context can

15 lead to confusion and misunderstanding. The majority’s analysis is testament to that

16 proposition. 3

Page 304: Sotomayor Cases

nonetheless chosen to address its treatment of husbands. I would be remiss if I did not discussboth why I believe the majority’s discussion of this issue is unnecessary and also why I believe itis wrong.

Although we remanded to the BIA to “more precisely explain its rationale for4

construing IIRIRA § 601(a)” to protect spouses, Shi Liang Lin v. U.S. Dep’t of Justice, 416 F.3d184, 187 (2d Cir. 2005), nothing in that decision precluded the BIA from looking to the moregeneral language in the statute in determining whether to extend relief to spouses. Indeed, theBIA had the authority to interpret the meaning of 8 U.S.C. § 1101(a)(42) independent of ourremand in Lin. See INS v. Aguirre-Aguirre, 526 U.S. 415, 425 (1999) (“The Attorney General . .. has vested the BIA with power to exercise the discretion and authority conferred upon theAttorney General by law in the course of considering and determining cases before it. Based on

36

1 In answering the first of the questions set out in Chevron, U.S.A., Inc. v. Natural

2 Resources Defense Council, 467 U.S. 837 (1984), that is, whether “the statute is silent or

3 ambiguous with respect to the specific issue,” id. at 843, we must look to the text of § 101(a)(42)

4 of the INA. That text provides, in pertinent part, that a refugee is

5 any person who is outside any country of such person’s nationality6 . . . and who is unable or unwilling to return to . . . that country7 because of persecution . . . on account of . . . political opinion . . . .8 For purposes of determinations under this chapter, a person who has9 been forced to abort a pregnancy or to undergo involuntary

10 sterilization . . . shall be deemed to have been persecuted on account11 of political opinion . . . .1213 8 U.S.C. § 1101(a)(42). The majority focuses its textual analysis on the final sentence of that

14 provision, that is, § 601(a) of the Illegal Immigration Reform and Immigration Responsibility

15 Act of 1996 (“IIRIRA”), concluding that the BIA’s decision “rested on” that sentence, and not

16 the general definition of “refugee” in 8 U.S.C. § 1101(a)(42). Maj. Op. at 10. Yet, as the

17 majority notes, the BIA found that “‘[t]here is no clear or obvious answer to the scope of the

18 protections afforded by the amendment to partners of persons forced to submit to an abortion or

19 sterilization.’” Maj. Op. at 13. Reasoning that the “lack of . . . a reference” to spouses in the4

Page 305: Sotomayor Cases

this allocation of authority, we [have] recognized . . . that the BIA should be accorded Chevrondeference as it gives ambiguous statutory terms concrete meaning through a process of case-by-case adjudication . . . .” (quotation marks and citation omitted)); see also Kuhali v. Reno, 266F.3d 93, 102 (2d Cir. 2001) (granting “Chevron deference to the Board’s construction of theINA, which it . . . administer[s]”).

Even if the BIA construed only the 1996 amendment, it would still be necessary to5

consider the text of the entire statute and the context against which that amendment was enactedto determine whether the 1996 amendment is itself ambiguous within the meaning of Chevron. See Natural Res. Def. Council v. Abraham, 355 F.3d 179, 198 (2d Cir. 2004) (“In interpretingthe plain language of the statute, we must look to the particular statutory language at issue, aswell as the language and design of the statute as a whole, and, where appropriate, its legislative

37

1 1996 amendment “does not necessarily preclude an applicant from demonstrating past

2 persecution based on harm inflicted on a spouse when both spouses are harmed by government

3 acts motivated by a couple’s shared protected characteristic,” the Board looked to the “general

4 principles regarding nexus and level of harm” for guidance. In re S–L–L–, 24 I. & N. Dec. 1, 5

5 (B.I.A. 2006); see also id. at 5 n.5 (citing to the general regulatory framework that defines when

6 an applicant may qualify as a refugee); id. at 6 (considering the “well-established principles

7 regarding nexus and level of harm for past persecution”).

8 Thus, although the majority places great emphasis on its view that the “language in §

9 601(a) . . . is unambiguous and . . . does not extend automatic refugee status to spouses or

10 unmarried partners of individuals § 601(a) expressly protects,” Maj. Op. at 4, I find the focus on

11 the amendment misplaced. I believe we must look instead to the entirety of 8 U.S.C.

12 § 1101(a)(42) to determine whether the statute is ambiguous. See SEC v. Chenery Corp., 332

13 U.S. 194, 196 (1947) (“[A] reviewing court, in dealing with a determination or judgment which

14 an administrative agency alone is authorized to make, must judge the propriety of such action

15 solely by the grounds invoked by the agency.”). That section provides that a “refugee” is any5

Page 306: Sotomayor Cases

history.” (quotation marks omitted)). Because I believe the BIA held that the entire provisionwas ambiguous, it is not necessary to determine here whether the 1996 amendment alone isambiguous. Thus, contrary to Judge Calabresi’s suggestion, Judge Calabresi Op. at 16 n.6, I donot necessarily agree that the BIA could not have relied on § 601(a) to provide asylum relief tospouses.

I agree with Judge Calabresi that this “general language” in 8 U.S.C. § 1101(a)(42)6

means that the BIA, with its expertise in this area, is particularly well-suited to exercise itsdiscretion and decide how that language should be interpreted. Judge Calabresi Op. at 8. JudgeCalabresi and I differ as to whether the BIA has already exercised that discretion.

38

1 individual who cannot return to his or her home country because of “persecution . . . on account

2 of . . . political opinion.” None of these terms is defined in any way, and none explicitly

3 addresses whether the spouses of those who have been forced to undergo an abortion or

4 sterilization are entitled to asylum relief. Hence, the statute, on its face, does not “directly6

5 address[] the precise question at issue.” Chevron, 467 U.S. at 843; cf. Sun Wen Chen v. U.S.

6 Att’y Gen., No. 05-4011, 2007 WL 1760658, at *4 (3d Cir. June 20, 2007) (“[T]he C–Y–Z– rule

7 thus fleshes out an issue germane to the application of § 1101(a)(42)(B) that was not addressed

8 by Congress, and so poses no Chevron step one problem.”).

9 Indeed, the majority points to no language in the statute that explicitly denies asylum

10 relief to these spouses, or that precludes the BIA from extending it to them. Cf. 8 U.S.C. §

11 1101(a)(42) (explicitly precluding from the definition of “refugee” individuals “who ordered,

12 incited, assisted, or otherwise participated in the persecution of any person on account of race,

13 religion, nationality, membership in a particular social group, or political opinion”). Rather, the

14 majority takes the position that “Congress’s specific designation of some persons (i.e., those who

15 fear, resist, or undergo particular medical procedures) is incompatible with the view that others

16 (e.g., their spouses) should also be granted asylum per se because of birth control policies. The

Page 307: Sotomayor Cases

The majority’s recognition that § 601(a) expanded the availability of asylum relief does7

nothing to explain why this language, which by its terms only expands relief, should also be read

39

1 inclusion of some obviously results in the exclusion of others.” Maj. Op. at 17-18. But where

2 the majority reads the language of the 1996 amendment and sees it as a limitation on the

3 availability of asylum relief, I see it, in the context in which it was enacted, as an expansion of

4 that relief.

5 I believe one question is fundamental: What was Congress’s purpose in enacting the 1996

6 amendment? Cf. Ai Feng Yuan v. U.S. Dep’t of Justice, 416 F.3d 192, 197 (2d Cir. 2005) (noting

7 the “canon of statutory construction that requires us to reconcile a statute’s plain language with

8 its purpose”). In answering that question, we should not limit our inquiry to the language of the

9 amendment; instead, we must look at the statutory scheme of which that amendment is a part and

10 the legislative activity that led to its enactment. Congress’s intent in enacting IIRIRA § 601(a)

11 was to clarify that, contrary to the BIA’s prior rulings, the imposition of some aspects of China’s

12 family planning policy can constitute persecution on the basis of political opinion, and that

13 certain victims of that persecution are entitled to protection under our asylum laws. Nothing in

14 the amendment suggests that Congress intended to prevent the BIA from extending relief to

15 victims other than those explicitly identified in the amendment. See Sun Wen Chen, 2007 WL

16 1760658, at *6 (“We are not convinced that Congress, in expanding asylum to include more

17 reproductive rights-based claims, intended to define the outer limits of relief in such cases.”).

18 There is thus some irony in the majority’s approach: By giving short shrift to context, it infers an

19 intent to limit the availability of asylum relief; had it looked more closely at context, it would

20 have recognized Congress’s intent was only to expand that relief. Context makes clear what7

Page 308: Sotomayor Cases

to have limited the BIA’s preexisting authority to further expand it. Although the majoritysuggests that I rely on “messages [I] hear[] in the sounds of Congressional silence,” Maj. Op. at26 n.12, all that I actually find in silence, especially given the context, is ambiguity. SeeChevron, 467 U.S. at 843 (noting that we turn to Chevron step two “if the statute is silent orambiguous with respect to the specific issue”); see also Sun Wen Chen, 2007 WL 1760658, at *3(“Chevron deference embodies the judgment that agencies, rather than courts, ought to serve asgap-fillers in situations of statutory silence.”). It is the majority that appears to find in silenceclear evidence of Congress’s intent.

These amendments to the asylum regulations appear to contemplate the possibility that8

asylum relief would be available to the spouses of those who were subject to forced abortion orsterilization. See Refugee Status, 55 Fed. Reg. at 2805 (to be codified at C.F.R. § 208.5(b)(2))(“An applicant who establishes that the applicant (or applicant’s spouse) has refused to abort apregnancy or to be sterilized in violation of a country’s family planning policy, and who has a

40

1 text alone fails to convey.

2 The INA provides asylum relief to individuals who have been “persecut[ed] . . . on

3 account of . . . political opinion,” but does not define those terms. In Matter of Chang, 20 I. &

4 N. Dec. 38 (B.I.A. 1989), the BIA held that whatever else “persecution” might mean, it did not

5 encompass any retribution visited upon individuals who violated China’s “one couple, one child”

6 policy. The BIA ruled that the policy “is [not] on its face persecutive” and does not “persecut[e]

7 any portion of the Chinese citizenry on account of one of the reasons enumerated in section

8 101(a)(42)(A) of the Act.” Id. at 43, 44.

9 Others in the executive branch took a different view. The next year, the Department of

10 Justice issued “interpretative rules and general statements of policy for establishing statutory

11 eligibility for asylum or withholding of deportation on the basis of political opinion for aliens

12 who express a fear of coercive population control policies in their homeland.” Refugee Status,

13 Withholding of Deportation and Asylum; Burden of Proof, 55 Fed. Reg. 2803, 2804 (Jan. 29,

14 1990). President George H.W. Bush reaffirmed his Administration’s support of the interim rule8

Page 309: Sotomayor Cases

well-founded fear that he or she will be required to abort the pregnancy or to be sterilized orotherwise persecuted if the applicant were returned to such country may be granted asylum.”);see also id. (to be codified at C.F.R. § 242.17(c)) (“Eligibility for withholding of deportation onaccount of political opinion is established by the respondent who establishes that he or she (orrespondent’s spouse) will be required to abort a pregnancy or to be sterilized . . . .”).

41

1 with the promulgation of Executive Order 12,711, which provided for “enhanced consideration

2 under the immigration laws for individuals from any country who express a fear of persecution

3 upon return to their country related to that country’s policy of forced abortion or coerced

4 sterilization.” Exec. Order No. 12,711, 55 Fed. Reg. 13,897, 13,897 (Apr. 11, 1990).

5 Although the INS, in July 1990, set forth a final rule that did not address this issue, see

6 Aliens and Nationality; Asylum and Withholding of Deportation Procedures, 55 Fed. Reg.

7 30,674 (July 27, 1990), the Office of the General Counsel of the INS subsequently sent a letter to

8 Regional and District Counsel stating that “Department of Justice and INS ‘policy with respect

9 to aliens claiming asylum or withholding of deportation based upon coercive family planning

10 policies is that the application of such coercive policies does constitute persecution on account of

11 political opinion.’” Xin-Chang Zhang v. Slattery, 55 F.3d 732, 740 (2d Cir. 1995), superseded

12 by 8 U.S.C. § 1101(a)(42) (quoting INS letter to Regional and District Counsel).

13 In January 1993, Attorney General William P. Barr signed a final rule that would have

14 made this view law. It provided, in pertinent part, that “[a]n applicant (and the applicant’s

15 spouse, if also an applicant) shall be found to be a refugee on the basis of past persecution on

16 account of political opinion if the applicant establishes that, pursuant to the implementation . . .

17 of a family planning policy . . . the applicant has been forced to abort a pregnancy or to undergo

18 sterilization or has been persecuted for failure or refusal to do so.” AG Order No. 1659-93, at 14

Page 310: Sotomayor Cases

42

1 (Jan. 15, 1993) (to be codified at 8 C.F.R. 208) (“January 1993 Rule”). Although this rule was

2 sent to the Federal Register, where it was made available for public inspection and scheduled for

3 publication, it was never published due to the change in presidential administrations. Xin-Chang

4 Zhang, 55 F.3d at 741. In February, additional regulations pertaining to asylum were published,

5 but these made no mention of the January rule. Id.

6 Against the background of these conflicting BIA decisions and administrative

7 regulations,

8 we were asked to determine whether asylum relief was available to victims of China’s family

9 planning policy. We held that such relief was not available, explaining that “[e]ven were we to

10 accept [the] view that the ‘administrative cacophony’ surrounding the issue justified diminished

11 deference to Chang, our result would not change. It is difficult to frame a result different from

12 the holding of Chang that would be ‘reasonable’ under both Elias-Zacarias and the existing

13 immigration laws.” Id. at 752; see also INS v. Elias-Zacarias, 502 U.S. 478 (1992). We did not

14 stop there, however. Instead, we called upon Congress and the President to determine whether

15 relief should be available to the victims of China’s family planning policy: “No doubt, the

16 President and the Congress acting together have power to create an exception to the existing

17 immigration laws for PRC citizens . . . .” Id.

18 The next year, Congress enacted IIRIRA § 601(a). The House Committee Report

19 explained that Congress’s “primary intent” in amending the definition of refugee was “to

20 overturn several decisions of the [BIA], principally Matter of Chang and Matter of G–,” H.R.

21 Rep. No. 104-469(I), 1996 WL 168955, at *173 (1996), in which the BIA had held that “the

Page 311: Sotomayor Cases

As previously noted, Attorney General Barr had attempted to “supersede the [BIA]9

decision in Matter of Chang,” January 1993 Rule, at 4-5, but his efforts fell victim to a change inpresidential administrations. See Xin-Chang Zhang, 55 F.3d at 741. Noting that the nextadministration had thus far failed to take action, Congress stepped in to accomplish the samegoal through legislation. See H.R. Rep. No. 104-469(I), 1996 WL 168955, at *174 (“[T]heAdministration, which has the authority to overrule the BIA decisions through regulation orthrough decision of the Attorney General, has not done so. Nor has it offered adequate relief topersons who have undergone such coercion.”).

43

1 Chinese Government’s implementation of its family planning policies is not on its face

2 persecutive and does not by itself create a well-founded fear of persecution on account of one of

3 the five grounds delineated in the Act, even to the extent that involuntary sterilization may

4 occur,” Matter of G-, 20 I. & N. Dec. 764, 778 (B.I.A. 1993); Matter of Chang, 20 I. & N. Dec.

5 at 43-44; see also Junshao Zhang v. Gonzales, 434 F.3d 993, 995 (7th Cir. 2006) (“The impact

6 of that amendment was to overrule Chang, and allow for the granting of asylum applications in

7 cases in which the claim of persecution stemmed from the enforcement of China’s coercive

8 population control policies.”); H.R. Rep. No. 104-469(I), 1996 WL 168955, at *174 (criticizing

9 the BIA decisions for “effectively preclud[ing] from protection persons who have been

10 submitted to undeniable and grotesque violations of fundamental human rights”); id. (noting that

11 “the BIA’s rationale for these opinions—that policies of coercive family planning are ‘laws of

12 general application’ motivated by concerns over population growth, and thus are not

13 ‘persecutory’—is unduly restrictive”).9

14 When Congress stated that “[f]or purposes of determinations under this [Act], a person

15 who has been forced to abort a pregnancy or to undergo involuntary sterilization . . . shall be

16 deemed to have been persecuted on account of political opinion,” 8 U.S.C. § 1101(a)(42)(a), it

17 was not providing an exhaustive list of those who could claim asylum relief because they were

Page 312: Sotomayor Cases

That two definitions of the term exist suggests, at the very least, that there is ambiguity10

in Congress’s use of the term, and the fact that Black’s Law Dictionary may favor one definitionis hardly sufficient to dispel that ambiguity. Indeed, although Black’s Law Dictionary maydescribe the majority’s preferred usage as the more “traditional[]” one and may discourage otheruses, the word “deem” has long been used to mean “consider, think, or judge.” See, e.g., U.S.Const. art. V (“The Congress, whenever two thirds of both Houses shall deem it necessary, shall

44

1 victimized by China’s family planning policy. Rather, it was expressing a congressional

2 determination that, contrary to the BIA’s prior rulings, China’s “one couple, one child” policy is

3 on its face persecutory, and victims of that policy who experienced persecution should be able to

4 qualify for asylum relief without making an additional showing of their own political opinion.

5 The majority takes the position that § 601(a) does not “alter the pre-IIRIRA definition of

6 ‘political opinion’ in § 1101(a)(42)” and that, for the individuals identified in § 601(a), “their

7 political opinion exists de jure rather than as a matter of fact on which the applicant bears the

8 burden of proof.” Maj. Op. at 18, 19. It notes, in this regard, that the amendment provides that

9 those who have been subjected to forced procedures “‘shall be deemed’ persecuted by reason of

10 political opinion,” Maj. Op. at 18-19, and that “[t]o ‘deem’ is ‘[t]o treat (something) as if (1) it

11 were really something else, or (2) it has qualities that it doesn’t have,’” Maj. Op. at 18-19 n.9

12 (quoting Black’s Law Dictionary 446 (8th ed. 2004)). But, as the majority acknowledges, there

13 is more than one definition of the word “deem”; it may also mean “[t]o consider, think, or

14 judge.” Black’s Law Dictionary 446 (8th ed. 2004). Thus, in this context, Congress’s use of the

15 word “deemed” may mean that these individuals should be “judged” as having been persecuted

16 on account of political opinion, just as the proposed 1993 rule provided that these applicants

17 “shall be found to be . . . refugee[s] on the basis of past persecution on account of political

18 opinion.” January 1993 Rule, at 14. Indeed, the legislative history suggests that Congress was10

Page 313: Sotomayor Cases

propose Amendments to this Constitution . . . .” (emphasis added)); 15 U.S.C. § 80b-3(j) (“TheCommission is authorized to adopt rules, regulations, and orders . . . as it deems appropriate toimplement this subsection.” (emphasis added)).

Indeed, Congress has repeatedly used the term “deem” in the INA itself to mean“consider, think, or judge.” See, e.g., 8 U.S.C. § 1101(a)(27)(J)(i) (defining the term “specialimmigrant” to include “an immigrant who is present in the United States . . . who has beendeclared dependent on a juvenile court located in the United States or whom such a court haslegally committed to, or placed under the custody of, an agency or department of a State and whohas been deemed eligible by that court for long-term foster care due to abuse, neglect, orabandonment” (emphasis added)); id. § 1103(a)(3) (providing that the Secretary of HomelandSecurity may “perform such other acts as he deems necessary for carrying out his authority underthe provisions of this chapter” (emphasis added)). A “normal rule of statutory construction”provides that “identical words used in different parts of the same act are intended to have thesame meaning.” Beharry v. Ashcroft, 329 F.3d 51, 61 (2d Cir. 2003) (quoting Gustafson v.Alloyd Co., 513 U.S. 561, 570 (1995) (quotation marks omitted)). Even if this rule does notconclusively resolve any ambiguity created by the existence of the two definitions, Congress’suse of the allegedly disfavored definition elsewhere in the INA renders the meaning of “deemed”in § 1101(a)(42), at the very least, ambiguous.

45

1 not attempting to create an exception to Matter of Chang, but to overrule it; and if Congress were

2 attempting to carve out an exception to the normal requirement that applicants must establish

3 that they have faced persecution on account of one of the protected grounds, it could have done

4 so in much plainer language. In any event, the statute is, at a minimum, ambiguous, and it is the

5 BIA, not the courts, that is charged with construing the statute in the face of that ambiguity.

6 Other circuits to have considered this issue have held that when Congress enacted the

7 1996 amendment it intended to protect both members of couples that are targeted under China’s

8 family planning policy. See, e.g., Junshao Zhang, 434 F.3d at 999 (“Congress passed §

9 601(a)(1) of the IIRIRA to ensure that families who are victims of forced abortion and

10 sterilization under China’s population control policy would receive asylum . . . .” (emphasis

11 added)); Kui Rong Ma v. Ashcroft, 361 F.3d 553, 559 (9th Cir. 2004) (identifying “Congress’s

12 goal in passing the amendments—to provide relief for ‘couples’ persecuted on account of an

Page 314: Sotomayor Cases

Although “[g]eneral language of a statutory provision . . . will not be held to apply to a11

matter specifically dealt with in another part of the same enactment,” D. Ginsberg & Sons, Inc. v.Popkin, 285 U.S. 204, 208 (1932), that rule has no applicability here where the specific provisiondoes not address whether spouses should be entitled to relief. In the 1996 amendment, Congressidentified a particular category of individuals entitled to relief, but left to the BIA the task ofdetermining whether to expand upon that relief, just as the BIA routinely defines whatindividuals are entitled to asylum relief in a myriad of other contexts.

Indeed, there is some language in the legislative history which suggests that Congressmay have considered the possibility that individuals other than those who had been forced toundergo an abortion or sterilization might qualify for asylum. The House Committee Report onthe amendment noted that “[d]etermining . . . whether the actual or threatened harm rises to thelevel of persecution is a difficult and complex task, but no more so in the case of claims based oncoercive family planning than in cases based on other factual situations. Asylum officers andimmigration judges are capable of making such judgments.” H.R. Rep. No. 104-469(I), 1996WL 168955, at *174. If only those subjected to the procedures could claim asylum relief,immigration judges and the BIA would never have needed to consider whether an applicant’sharm rises to the level of persecution, given that the amendment explicitly provides that a forcedabortion or sterilization does rise to this level.

46

1 ‘unauthorized’ pregnancy and to keep families together” (emphasis added)). We need not go as

2 far as these courts to affirm the BIA’s decision here. Even if Congress did not specifically

3 intend to protect “couples,” there is nothing in the text of the amendment, or the context that

4 gave rise to it, that indicates that it intended to preclude the BIA from extending asylum relief to

5 both members of a couple. See Sun Wen Chen, 2007 WL 1760658, at *5 (granting Chevron

6 deference to the BIA’s interpretation in In re S–L–L–). 11

7 Just as nothing in the language or history of the amendment indicates a congressional

8 intent to foreclose the extension of relief to spouses, Congress has done nothing to indicate such

9 an intent in the years since the amendment’s enactment, notwithstanding that the BIA interpreted

10 § 1101(a)(42) to cover spouses a decade ago and numerous courts of appeals have upheld this

11 interpretation as reasonable. See, e.g., Yuan, 416 F.3d at 197; Junshao Zhang, 434 F.3d at 999;

12 Kui Rong Ma, 361 F.3d at 559 (“The BIA and the courts have uniformly applied the statute’s

Page 315: Sotomayor Cases

47

1 protections to husbands whose wives have undergone abortions or sterilization procedures, as

2 well as to the wives themselves.”); In re C–Y–Z–, 21 I. & N. Dec. 915, 918-19 (B.I.A. 1997) (en

3 banc). There are obscure areas of public policy, largely hidden from public attention and

4 concern, in which it makes little sense to ascribe meaning to the absence of congressional

5 response to administrative and judicial interpretations of a statute. Immigration is hardly one of

6 those areas. To the contrary, immigration—and the issue of the appropriate scope of asylum

7 relief—have consistently been on Congress’s radar. Immigration is frequently in the news, and

8 Congress has repeatedly legislated in this area. Indeed, as recently as 2005, Congress revisited

9 this very provision and removed the annual cap on the number of asylees who could be admitted

10 under it. See REAL ID Act of 2005, Pub. L. No. 109-13, Div. B, §§ 101(g)(2), (h)(5), 119 Stat.

11 231, 305-06 (May 11, 2005). While the fact that Congress, in the course of its active attention to

12 immigration issues and legislation, has not amended 8 U.S.C. § 1101(a)(42) in light of the

13 interpretation it has been given by the BIA and the courts does not definitively mean that

14 Congress intended to protect spouses, it does suggest, at the very least, that it was not Congress’s

15 intent to foreclose that relief.

16 The majority nonetheless holds that it was “not correct” for the BIA to construe the

17 general definition of “refugee” to allow for the provision of this relief because “the statutory

18 scheme unambiguously dictates that applicants can become candidates for asylum relief only

19 based on persecution that they themselves have suffered or must suffer.” Maj. Op. at 19. I agree

20 that an individual must have personally experienced persecution to be entitled to asylum relief,

21 but that statement begs the question of what constitutes persecution.

Page 316: Sotomayor Cases

48

1 The statute does not, in either the 1996 amendment or in its general definition of the term

2 “refugee,” prescribe exactly how much harm or what kind of harm an individual must experience

3 to have been “persecuted” within the meaning of the statute. See Ivanishvili v. U.S. Dep’t of

4 Justice, 433 F.3d 332, 340 (2d Cir. 2006) (noting that the term “persecution” is “not defined by

5 the Immigration and Nationality Act”); see also Matter of Acosta, 19 I. & N. Dec. 211, 222

6 (B.I.A. 1985), overruled on other grounds by Matter of Mogharrabi, 19 I. & N. Dec. 439 (B.I.A.

7 1987) (defining persecution as “either a threat to the life or freedom of, or the infliction of

8 suffering or harm upon, those who differ in a way regarded as offensive” (emphasis added)); see

9 also Ivanishvili, 433 F.3d at 341 (acknowledging the BIA’s definition of “persecution” as “the

10 infliction of suffering or harm upon those who differ on the basis of a protected statutory

11 ground.”). The 1996 amendment states that when one is forced to undergo an abortion or

12 sterilization, the harm thereby experienced is sufficient to qualify for asylum, but it hardly makes

13 that level of harm necessary.

14 The majority may view the 1996 amendment as providing that only individuals who have

15 undergone a forced abortion or sterilization have experienced “persecution.” If so, we again

16 differ on our interpretation of the significance of the 1996 amendment. I believe Congress

17 enacted the 1996 amendment not primarily to define the term “persecution,” but to clarify what it

18 means to be persecuted “on account of political opinion.” As noted above, Congress sought to

19 make clear, contrary to the BIA’s earlier decision in Matter of Chang, that the imposition of

20 penalties through the implementation of China’s family planning policy can constitute

21 persecution “on account of political opinion” by effectively adopting the position taken in the

Page 317: Sotomayor Cases

49

1 commentary to the 1993 rule that, “[a]lthough . . . prosecution and punishment under a law of

2 general applicability will not ordinarily constitute persecution ‘on account of’ one of the

3 statutory grounds, . . . [p]ersecution on account of political opinion encompasses persecution of

4 people whose violation of laws may not be motivated by their political opinions but is regarded

5 by the state as political disloyalty.” January 1993 Rule, at 8.

6 Thus, I do not think that § 601(a) unambiguously defines the term persecution, and the

7 majority has pointed to nothing in the statute that suggests that the emotional and psychological

8 harm one suffers when one’s spouse is forced to undergo an abortion or sterilization is not severe

9 enough to constitute persecution. Nor does anything in the statute preclude the BIA from

10 considering the effect that China’s family planning policies may have on a couple’s shared right

11 to reproduce and raise children. Because Congress did not specifically address these issues, the

12 statute is ambiguous. It therefore falls to the BIA to determine whether the harm an individual

13 experiences when his or her spouse is subjected to a forced abortion or sterilization is sufficient

14 to constitute persecution. See, e.g., Kuhali, 266 F.3d at 102 (granting “Chevron deference to the

15 Board’s construction of the INA, which it . . . administer[s]”); cf. Sun Wen Chen, 2007 WL

16 1760658, at *5 (noting that the BIA “exercised its delegated gap-filling authority reasonably”

17 when it recognized the harms an individual experiences as a result of the forced abortion or

18 sterilization of his spouse). By holding that persecution cannot encompass such individuals, the

19 majority, as Judge Sotomayor cogently explains, usurps the BIA’s task of giving meaning to

20 ambiguous statutory terms. Further, by suggesting that the BIA is creating a presumption which

21 allows individuals to be granted asylum without proving that they were “persecuted . . . on

Page 318: Sotomayor Cases

The majority’s assertion that § 601 “relieve[s] . . . persons who actually experienced a12

forcible abortion or sterilization from the burden of proving a political nexus” is similarlyproblematic. Maj. Op. at 21. In my view, Congress did not relieve anyone of the burden ofproving political nexus. Rather, it determined that an applicant can meet this burden byestablishing proof of persecution pursuant to a coercive family planning policy. Given themajority’s agreement that any asylum applicant may “rely[] on evidence of a forced abortion orsterilization—whether personal or spousal—in attempting to demonstrate persecution based onpolitical opinion,” Maj. Op. at 21 (emphasis added), I fail to see what additional evidence anapplicant whose spouse was subjected to a forced abortion or sterilization would have to show tocarry his burden of proving political nexus.

50

1 account of political opinion,” the majority errs. The BIA here created no presumption; rather,

2 fulfilling a basic responsibility, it simply discerned the meaning of those ambiguous terms. 12

3 Finally, the majority finds support for its reading of the statute in its perception that there

4 is some tension between this rule and the provision of derivative asylum status for spouses of

5 individuals who qualify as “refugees.” Maj. Op. at 26-28. But in adopting this rule, the BIA

6 provided a basis by which individuals could claim asylum relief in their own right for harm they

7 suffered as a result of their spouses’ forced abortion or sterilization. There is no apparent tension

8 in providing derivative asylum status to spouses who have not themselves suffered any harm and

9 providing an additional basis of relief to those spouses who have, that is, those who have

10 themselves suffered harm when their partners were subjected to a forced abortion or sterilization.

11 See 8 U.S.C. § 1158(b)(3)(A) (providing derivative asylum to spouses who are “not otherwise

12 eligible for asylum under this section”); cf. Junshao Zhang, 434 F.3d at 1001 (noting that “it

13 would be particularly perverse for courts to treat a subsequent break-up of the marriage as

14 somehow lessening the impact of [the prior] persecution [of the husband]”). Because there is no

15 tension in providing these two separate, distinct forms of relief, it does not seem to me that the

16 availability of derivative asylum relief unambiguously precludes the BIA from providing an

Page 319: Sotomayor Cases

51

1 additional basis of relief to those whose spouses have been subjected to a forced abortion or

2 sterilization. See Sun Wen Chen, 2007 WL 1760658, at *4 (“We . . . do not believe that the

3 existence of derivative asylum status under a statute implies that Congress intended to foreclose

4 additional pathways to asylum specific to spouses.”).

5 At bottom, the majority seems to be motivated by a concern that the BIA’s rule will “gut

6 or . . . undermine” the availability of derivative asylum relief or have “the perverse effect of

7 creating incentives for husbands to leave their wives.” Maj. Op. at 28, 27. But nothing in the

8 BIA’s rule denies relief of any kind to husbands who come to this country with their wives;

9 moreover I think it very likely that husbands will find it advantageous to come with their wives

10 when possible because doing so will buttress the credibility of their claims. In the end, however,

11 my views of the policy consequences of the BIA rule are immaterial; so, too, are those of the

12 majority. Once we determine that the statute is ambiguous, we must defer to the BIA’s

13 interpretation of the statute if it is reasonable, whatever our own personal policy preferences.

14 See, e.g., Udall v. Tallman, 380 U.S. 1, 16 (1965) (“To sustain the Commission’s application of

15 this statutory term, we need not find that its construction is the only reasonable one, or even that

16 it is the result we would have reached had the question arisen in the first instance in judicial

17 proceedings.” (quotation marks omitted)).

18 In considering the reasonableness of the BIA’s interpretation under step two of Chevron,

19 our charge is not to determine whether the BIA’s interpretation of the statute is the one we would

20 have adopted in the first instance. Instead, “the question for the court is whether the agency’s

21 answer is based on a permissible construction of the statute”; we must defer to “legislative

Page 320: Sotomayor Cases

52

1 regulations . . . unless they are arbitrary, capricious, or manifestly contrary to the statute.”

2 Chevron, 467 U.S. at 843, 844. Here, in determining whether to extend asylum relief to spouses,

3 the BIA reasonably considered the general principles underlying the definition of persecution

4 and concluded that a husband is persecuted “when the government forces an abortion on a

5 married couple.” In re S–L–L–, 24 I. & N. Dec. at 6; see also id. (“When the government

6 intervenes in the private affairs of a married couple to force an abortion or sterilization, it

7 persecutes the married couple as an entity.”). I see no reason why the BIA could not reasonably

8 conclude that one has suffered harm or injury sufficiently severe to constitute persecution when

9 one’s spouse is forced to undergo an abortion or sterilization. Indeed, this determination finds

10 support in the decisions of a number of courts that have explicitly recognized that non-physical

11 harm may support a finding of past persecution in at least some circumstances. See Junshao

12 Zhang, 434 F.3d at 1001 (rejecting explicitly the “notion that [a husband] suffers no persecution

13 independent of his wife, as the result of the forcible abortion of his child” and holding that

14 “[a]lthough his wife was certainly a very direct victim of China’s population control measures,

15 Zhang was a victim as well. The forcible abortion has deprived him of his unborn child, of the

16 ability to realize the family that his wife and he had desired, and forever deprived him of the

17 ability to become a parent to that unborn son or daughter with his wife”); see also Ouk v.

18 Gonzales, 464 F.3d 108, 111 (1st Cir. 2006) (noting that “[u]nder the right set of circumstances,

19 a finding of past persecution might rest on a showing of psychological harm” (quotation marks

20 omitted)); Mashiri v. Ashcroft, 383 F.3d 1112, 1120 (9th Cir. 2004) (“Persecution may be

21 emotional or psychological, as well as physical.”); Abay v. Ashcroft, 368 F.3d 634, 642 (6th Cir.

Page 321: Sotomayor Cases

Because petitioner Dong, the only petitioner whose claim we address on this appeal,13

had not participated in a traditional marriage ceremony, I need not determine now whether theBIA’s rule would also be reasonable as applied to individuals who were not old enough to marryunder Chinese law and who participated in such a ceremony. Compare Junshao Zhang, 434F.3d at 999 (holding that the BIA rule is unreasonable in this context), and Kui Rong Ma, 361F.3d at 560 (same), with Cai Luan Chen, 381 F.3d at 231 (holding that the BIA rule isreasonable).

53

1 2004) (holding that the applicant was entitled to asylum “based on her fear that her daughter will

2 be forced to undergo female genital mutilation” because her “fear of . . . being forced to witness

3 the pain and suffering of her daughter is well-founded”).

4 The BIA also determined that there were not “convincing reasons to extend the nexus and

5 level of harm attributed to a husband who was opposed to his wife’s forced abortion to a

6 boyfriend or fiancé.” In re S–L–L–, 24 I. & N. Dec. at 9. Recognizing that “marriage place[s]13

7 the husband in a distinctly different position from that of an unmarried father,” id., the BIA noted

8 that unmarried fathers do not bear the same legal and societal responsibility for violations of

9 family planning policies. Indeed, because their relationships with their partners are not

10 registered with the government and may not even be known within the community, the

11 government may often be unaware of their identities. See id. at 9-10. The BIA thus presumed

12 that the family planning officials target legal spouses for persecution to a greater extent than

13 boyfriends and fiancés. See id. Furthermore, “[p]roof or presumption of paternity . . . may be

14 considerably more difficult when a boyfriend claims to have fathered a child who was forcibly

15 aborted by government officials.” Id. at 10. Although, as the BIA itself acknowledges,

16 “drawing the line at marriage is not” perfect, id. at 9, and reasonable policymakers could differ

17 as to how to draw the line, I cannot say that, under the deferential standard which guides us, the

Page 322: Sotomayor Cases

Because petitioner Dong was not married and has not otherwise established his14

eligibility for asylum relief, I agree with the majority that his petition for review should bedenied. I also agree that the petitions of Lin and Zou should be dismissed.

The BIA, too, remains free to revisit its decision. Although Judge Calabresi suggests15

that my approach “preclude[s] the agency from thinking deeply and fully about the matter,”Judge Calabresi Op. at 16, I do not understand why this would be so. Unlike the majority, I donot purport to offer an authoritative constructive of 8 U.S.C. § 1101(a)(42); I would hold onlythat the statute is ambiguous and that the BIA’s interpretation is reasonable. As Judge Calabresielsewhere acknowledges, id. at 2, the BIA may always change its own interpretation of statutorylaw, so long as the change is not inconsistent with that law. See, e.g., Nat’l Cable & Telecomms.Ass’n v. Brand X Internet Servs., 545 U.S. 967, 981 (2005) (“For if the agency adequatelyexplains the reasons for a reversal of policy, change is not invalidating, since the whole point ofChevron is to leave the discretion provided by the ambiguities of a statute with the implementingagency.” (quotation marks omitted)).

54

1 BIA’s reading is not based on a permissible construction of the statute. See Chevron, 467 U.S. at

2 843. If Congress disagrees with the BIA’s interpretation, it can overturn the decision. 14 15

3 This case presents difficult and challenging questions at the heart of our immigration

4 laws. How we respond will affect the hopes and dreams of human beings seeking to live in

5 freedom. In enacting the INA, Congress established a framework for determining when asylum

6 relief should be provided to such individuals, and in doing so, it delegated considerable authority

7 to the BIA to fill in statutory gaps and define the broad language used in the INA. It is in

8 situations such as these that we should be particularly mindful of the views of the agency

9 charged by Congress with administering the statute, views that will reflect the agency’s

10 considerable experience and expertise. We should recognize that in such circumstances what is

11 advanced as the obvious answer may not be the right one. Here, the meaning of the text

12 becomes much less clear when one examines context, and the BIA, recognizing that ambiguity,

13 has offered a reasonable interpretation of the statute. I would defer to that interpretation.

Page 323: Sotomayor Cases

The majority claims to be perplexed by my concern that today’s decision reaches a1

question it need not, particularly because, the majority reasons, all judges – including myself –who agreed to hear these cases en banc joined an order instructing “the parties to address theBIA’s interpretation of § 601 as it related to both spouses and non-married partners.” Maj. Op.at 13 n.6. My concern, however, is not a quibble over the semantics of the en banc order, butrather the majority’s unnecessary but apparently pressing need to decide a question which thefacts of petitioners’ appeals simply do not present. Like Judge Katzmann, I engage the questionthe majority answers because I would be remiss in not voicing my profound disagreement withthe majority’s conclusions.

55

1 SOTOMAYOR, Circuit Judge, with whom Judge POOLER joins, concurring in the judgment:

2 Today’s decision marks an extraordinary and unwarranted departure from our

3 longstanding principles of deference and judicial restraint. Instead of answering the limited

4 question before us – whether the BIA’s denial of asylum to the unmarried partners of women

5 forced to undergo abortions or sterilization was unreasonable – the majority has chosen to go far

6 beyond it to address an issue that is unbriefed, unargued, and unnecessary to resolve this appeal.

7 Indeed, the cases before us, which involve only unmarried petitioners, are inappropriate vehicles

8 through which to opine on the merits of the BIA’s position with respect to spouses under

9 8 U.S.C. § 1101(a)(42). See Carducci v. Regan, 714 F.2d 171, 177 (D.C. Cir. 1983) (Scalia, J.)1

10 (“The premise of our adversarial system is that appellate courts do not sit as self-directed boards

11 of legal inquiry and research, but essentially as arbiters of legal questions presented and argued

12 by the parties before them.”); see also Turner Broad. Sys., Inc. v. FCC, 520 U.S. 180, 224 (1997)

13 (declining to reach question which received little or no attention from the parties and noting

14 “prudence dictates that we not decide this question based on such scant argumentation”).

15 Moreover, as Judge Katzmann’s concurrence, in which I fully join, cogently notes, today’s

16 holding simply ignores the context animating § 601’s enactment and further upends

Page 324: Sotomayor Cases

56

1 congressional intent by reading the statute too narrowly. Section 601 was, after all, expressly

2 enacted to expand, not contract, the availability of asylum under § 1101(a)(42) in the context of

3 coercive population control programs.

4 I will not reiterate what Judge Katzmann has already ably stated. I write separately to

5 highlight the potentially ill-considered breadth of the majority opinion, which appears to cast

6 doubt on our own circuit’s caselaw, as well as to create further circuit conflicts when such

7 outcomes are easily avoided. Most importantly, however, I write because the majority’s zeal in

8 reaching a question not before us requires the unprecedented step of constricting the BIA’s

9 congressionally delegated powers – a decision whose ramifications we are ill-prepared, given the

10 procedural posture of this case, to understand or appreciate fully.

11 The majority analyzes § 601 within the broader framework of the INA and concludes that

12 “[n]othing in the general definition of refugee would permit ‘any person’ who has not personally

13 experienced persecution or a well-founded fear of future persecution . . . to obtain asylum.” Maj.

14 Op. at 17. Again, I agree fully with Judge Katzmann’s discussion of the fatal flaws in this

15 analysis, and I seek only to emphasize the majority’s apparent failure to appreciate that this

16 deceptively simple proposition may unduly and inappropriately limit the BIA not merely in cases

17 under § 601 but in others as well.

18 In coming to its conclusion, the majority endorses the view that “persecution” can only

19 be direct and personal, by which it appears to mean that the granting of asylum can never be

20 based on, in whole or in part, harm to others, no matter how closely related the harm or the

21 person harmed is to the applicant or whether harm to another is directed in whole or in part

Page 325: Sotomayor Cases

In another portion of the majority opinion, the Court states “we conclude that the statutory2

scheme unambiguously dictates that applicants can become candidates for asylum relief onlybased on persecution that they themselves have suffered or must suffer.” Maj. Op. at 19. By thispronouncement, the opinion suggests that harm to others cannot form a part of the rationale forgranting asylum. I further agree with Judge Katzmann’s analysis of the majority’s misplaced reliance on 83

U.S.C. § 1158 to support its conclusion here.

57

1 toward the applicant. The majority tries to anchor this limiting principle to the text of the2

2 statute, but such a reading is unwarranted and unsupportable. It is pellucidly clear from the text

3 of § 1101(a)(42) that Congress did not define nor intend to define “persecution” to exclude

4 harms “not personally” suffered by an applicant. The statute instead reads that “any person”

5 who “because of persecution or a well-founded fear of persecution” is “unable or unwilling” to

6 return to his or her country is entitled to asylum. There is no indication whatsoever of how

7 personal or direct the harm or injury must be, only that persecution to an individual can merit

8 asylum protection. We should, moreover, eschew the limiting construction of § 601 and3

9 § 1101(a)(42) urged in today’s opinion because such a construction could lead to absurd results.

10 United States v. Dauray, 215 F.3d 257, 264 (2d Cir. 2000) (“A statute should be interpreted in a

11 way that avoids absurd results.”). If government officials shot and killed an asylum applicant’s

12 child to force him or her to convert to another religion, would that harm, which the majority

13 would ostensibly label “not personal,” be insufficient in itself to demonstrate persecution of that

14 applicant? Or what if the parent of an adult applicant was kidnapped and tortured to force the

15 applicant to renounce an opposition political party or endorse a government candidate? In the

16 end, I see no unambiguous language in the text of § 1101(a)(42) that compels the limiting

17 construction of the INA that the majority now divines.

Page 326: Sotomayor Cases

I note that this and other circuits have found “persecution” to be an ambiguous term in other4

asylum cases. See, e.g., Mirzoyan v. Gonzales, 457 F.3d 217, 220 (2d Cir. 2006) (per curiam)(finding that the INA does not “unambiguously explain[] what the word ‘persecution’ means” inthe economic context); Corado v. Ashcroft, 384 F.3d 945, 947 (8th Cir. 2004) (per curiam)(“The BIA is entitled to deference in interpreting ambiguous statutory terms such as‘persecution.’”); cf. Singh v. INS, 134 F.3d 962, 967 (9th Cir. 1998) (noting that the INA “doesnot define ‘persecution’ or specify what acts constitute persecution”).

58

1 Requiring an applicant’s eligibility for asylum to rest only on instances where he or she

2 suffers persecution “personally” merely begs the question of what personal harm is and how to

3 define it. As with any ambiguous statutory term, it is for the BIA to determine within its

4 expertise what exactly constitutes “persecution” so long as its interpretation is reasonable. 4

5 Chevron, U.S.A., Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837, 843-44 (1984). From its

6 decision in In re S-L-L-, 24 I. & N. Dec. 1 (B.I.A. 2006), the BIA clearly construed

7 “persecution” as not only entailing the spouse forced to undergo the procedure but also including

8 the other spouse who, while physically unharmed, was nevertheless also targeted by the

9 government for punishment and persecution. The BIA reached this conclusion by utilizing its

10 traditional tests of nexus and level of harm, id. at 5, that is, by examining how the procedure

11 affected each spouse’s respective health and emotional well-being as well as the couple’s interest

12 in procreation and child-rearing. Perhaps most importantly of all, the BIA also considered to

13 whom the government’s actions were directed.

14 It is this last factor that is crucial. The majority concedes that both spouses suffer a

15 “profound emotional loss” as a result of a forced abortion or sterilization, but it never sufficiently

16 explains why the harm of sterilization or abortion constitutes persecution only for the person

17 who is forced to undergo such a procedure and not for that person’s spouse as well. Maj. Op. at

18 21. This conclusion rests on two principal conclusions, both equally flawed. First, the majority

Page 327: Sotomayor Cases

59

1 clings to the notion that the persecution suffered is physically visited upon only one spouse, but

2 this simply ignores the question of whom exactly the government was seeking to persecute when

3 it acted. On this point, the harm is clearly directed at the couple who dared to continue an

4 unauthorized pregnancy in hopes of enlarging the family unit. Indeed, the majority’s conclusion

5 disregards the immutable fact that a desired pregnancy in a country with a coercive population

6 control program necessarily requires both spouses to occur, and that the state’s interference with

7 this fundamental right “may have subtle, farreaching and devastating effects” for both husband

8 and wife. See Skinner v. Oklahoma ex rel. Williamson, 316 U.S. 535, 541 (1942). The

9 termination of a wanted pregnancy under a coercive population control program can only be

10 devastating to any couple, akin, no doubt, to the killing of a child. Similarly, as to sterilization,

11 the Ninth Circuit has aptly observed that:

12 In addition to the physical and psychological trauma that is common to many13 forms of persecution, sterilization involves drastic and emotionally painful14 consequences that are unending: The couple is forever denied a pro-creative life15 together. As the BIA explained,1617 The act of forced sterilization should not be viewed as a discrete onetime18 act, comparable to a term in prison, or an incident of severe beating or19 even torture. Coerced sterilization is better viewed as a permanent and20 continuing act of persecution that has deprived a couple of the natural21 fruits of conjugal life, and the society and comfort of the child or children22 that might eventually have been born to them.2324 Qili Qu v. Gonzales, 399 F.3d 1195, 1202 (9th Cir. 2005) (quoting In re Y-T-L-, 23 I. & N. Dec.

25 601, 607 (B.I.A. 2003)). Viewed in this light, the harm here is directed as much at the husband

26 as at the wife. By its action, after all, the state is preventing both members of the couple from

27 procreating as a unit, and as the BIA found in In re Y-T-L-, such harm is not limited in time to

Page 328: Sotomayor Cases

The majority incorrectly asserts that the Third Circuit’s analysis in Sun Wen Chen is5

incongruent with my own analysis because the court there held that § 601(a) contains anambiguity. Maj. Op. at 16 n.7. The Third Circuit found that § 601(a) is ambivalent on thetreatment of spouses and that the existence of derivative asylum was not “intended to forecloseadditional pathways to asylum specific to spouses.” Sun Wen Chen, --- F.3d ---, 2007 WL1760658, at *4 (emphasis added); see also id. (“[W]e see nothing in the statute evincingCongressional intent to establish a particular policy regarding spousal eligibility.”). The ThirdCircuit then reasoned that because § 601(a) “establishes that forced abortion and sterilizationconstitute persecution,” id., it is entirely within the BIA’s authority to “interpret[] the scope ofthat persecution,” id., including its applicability to spouses. It is my contention that we shoulddefer to the BIA’s interpretation regarding the broader scope of persecution under § 1101(a)(42)and not impose, as the majority here does, an unfounded requirement that persecution be directand personal and that harm to another, even if directed at the applicant, is never sufficient for thepurposes of § 1101(a)(42). These analyses are not incongruent because they both center on thedeference we owe to the BIA on defining persecution.

The majority notes that its decision corrects the “perverse effect of creating incentives for6

husbands to leave their wives” inherent in the BIA’s determination in In re S-L-L-, 24 I. & N.Dec. 1 (B.I.A. 2006). Maj. Op. at 27. I note first that one of the petitioners before us was not, in

60

1 the moment of sterilization, but is an ongoing harm that affects a married couple indefinitely. 23

2 I. & N. Dec. at 607. As the Third Circuit recently observed in Sun Wen Chen v. Attorney

3 General of the United States, --- F.3d ---, 2007 WL 1760658, at *5 (3d Cir. June 20, 2007), the

4 “persecution of one spouse can be one of the most potent and cruel ways of hurting the other

5 spouse.” In the end, I fail to understand how the majority can claim that the harm caused by a5

6 spouse’s forced abortion or sterilization is not a personal harm to both spouses – either or both of

7 whom can be sterilized for violations of the population control programs – especially given the

8 unique biological nature of pregnancy and special reverence every civilization has accorded to

9 child-rearing and parenthood in marriage. I similarly fail to understand how the majority

10 justifies limiting the BIA’s ability to take this special and egregious harm into consideration and

11 to determine within its expertise that such acts constitute persecution against both a wife and

12 husband. 6

Page 329: Sotomayor Cases

fact, attempting to flee alone; he and his fiancée planned to leave China together but she wasunfortunately caught before she could escape. More importantly, however, the majority’sassertion here is based on nothing but speculation as to the decisionmaking in which couples,persecuted by coercive population control programs, must engage before attempting to flee. Wesimply have no foundation on which to conclude that all couples have the financial resources toescape at the same time, and as the government stated at oral argument, it is not uncommon forChinese couples to separate and have one spouse go abroad in order to amass the necessaryresources to bring over the other spouse. I believe the majority here is opining on a subject –imbued with potentially significant cultural differences – with which it has no expertise orempirical evidence.

61

1 Second, the majority argues that the BIA has impermissibly created an irrebuttable

2 presumption that relieves applicants from the statutory burden of proving that they have a well-

3 founded fear of persecution based on an impermissible nexus. As Judge Katzmann correctly

4 observes, however, the presumption argument is merely a red herring. In enacting § 601,

5 Congress was not creating a presumption but merely expanding the BIA’s unduly restrictive

6 definition in Matter of Chang of the terms “persecution” and “political opinion.” Thus, § 601

7 defined “persecution” and “political opinion” to include an individual’s forced abortion or

8 sterilization under a restrictive population control policy. No presumption was created, however,

9 as the applicant still bears the burden of establishing that he or she was subject to the conduct

10 that qualifies under this expanded definition of persecution. And, while the majority places great

11 weight on the fact that § 601 supposedly creates a presumption that “relieved only persons who

12 actually experienced a forcible abortion or sterilization from the burden of proving a political

13 nexus,” and not those married to such persons, Maj. Op. at 21, his conclusion ignores the clear

14 congressional intent of § 601, expressed in the legislative history, that

15 [n]othing in [§ 601] is intended to lower the evidentiary burden of proof for any16 alien, no matter how serious the nature of the claim. The Committee emphasizes17 that the burden of proof remains on the applicant, as in every other case, to18 establish by credible evidence that he or she has been subject to persecution-in

Page 330: Sotomayor Cases

62

1 this case, to coercive abortion or sterilization-or has a well-founded fear of such2 treatment. 34 H.R. Rep. No. 104-469(I), 1996 WL 168955, at *174 (1996). Quite simply, there is no way to

5 read in § 601 the presumption the majority discusses, nor a limitation on the BIA in extending

6 § 601’s reach to spouses. Either the persecution occurred or it did not, and the applicant retains

7 the burden of proving such circumstances. The BIA accordingly did not err in interpreting

8 ambiguous terms in the INA to determine that either spouse may qualify as a refugee where one

9 of them has in fact undergone forced abortion or sterilization.

10 Finally, if adopted, the majority’s limiting construction may have significant, unintended

11 consequences, broader than the Court today acknowledges. By claiming categorically that an

12 applicant cannot rely on the harm inflicted on others in § 601 cases – no matter how closely

13 related the harm or the person harmed is to the applicant or whether the harm is directed in

14 whole or in part towards the applicant – to establish persecution or entitlement to asylum, this

15 Court suggests that the BIA is precluded from ever considering harm to others as evidence of

16 persecution to the applicant. While I agree that there are certainly limits as to when harm to

17 another may inform persecution or a well-founded fear of persecution of an applicant, I cannot

18 endorse the majority’s apparent per se conclusion. As noted above, this is a question of statutory

19 construction properly answered by the BIA, which, in its administrative expertise, may interpret

20 the ambiguity inherent in “persecution” to determine when a harm is direct or personal enough to

21 be considered persecution of an applicant. And, in certain limited contexts pertinent to this

22 appeal, the BIA has done precisely this, examining the harm to family members in determining

23 whether an asylum applicant has in fact suffered past persecution, particularly where an

Page 331: Sotomayor Cases

63

1 immediate family member has been subjected to significant and enduring mistreatment. In

2 Matter of Chen, 20 I. & N. Dec. 16, 19-21 (B.I.A. 1989), the seminal BIA decision recognizing

3 the availability of humanitarian asylum for victims of severe past persecution, the BIA recited a

4 litany of horrific acts visited on and suffered by Chen’s parents during the Cultural Revolution,

5 and such evidence – when coupled with the Chinese government’s treatment of the petitioner

6 himself – supported the BIA’s conclusion that “the respondent has clearly established that he and

7 his family were severely persecuted in the past in China.” Id. at 21. Similarly, in In re H-, 21 I.

8 & N. Dec. 337, 345 (B.I.A. 1996), where an applicant had testified about the severe physical

9 beatings he had suffered as the member of a subclan in Somalia, the BIA based its finding of past

10 persecution in part on the applicant’s testimony that his father and brother, also members of the

11 same subclan, were beaten and killed. In examining the allegations concerning the deaths of his

12 father and brother, the BIA specifically noted that “evidence of treatment of persons similarly

13 situated is persuasive of an applicant’s claim of political persecution.” Id. at 345 (quoting Matter

14 of Mogharrabi, 19 I. & N. Dec. 439 (B.I.A. 1987)); see also In re N-M-A-, 22 I. & N. 312, 326

15 (B.I.A. 1998) (finding in the context of humanitarian asylum that the applicant had suffered past

16 persecution in part because of “the disappearance and likely death of his father”). The BIA has

17 thus identified specific situations in which the harm to close family members could be central to

18 the finding of persecution and the granting of refugee status. The majority’s misguided exercise

19 in statutory interpretation, however, undermines this agency determination and suggests that

20 because the years-long harassment and egregious mistreatment of Chen’s parents or the deaths of

Page 332: Sotomayor Cases

The regulations governing the claims under the Convention Against Torture explicitly7

recognize that torture encompasses not only physical harm to the individual but also “mentalpain or suffering” that results from the threat of infliction of physical pain or suffering onanother person. See 8 C.F.R. § 1208.18(a)(4)(iv). By analogy, this regulation further supportsthe position that persecution is not limited to direct and physical harm upon an individual but canencompass harm inflicted on others as well. See, e.g., Yan Chen v. Gonzales, 417 F.3d 268, 275(2d Cir. 2005) (“Certainly . . . torture can constitute persecution . . . .”).

The passing statement in a footnote in Melgar de Torres v. Reno, 191 F.3d 307, 313 n.2 (2d8

Cir. 1999), that the death of the applicant’s uncle did not constitute political persecution of her isnot to the contrary. As noted in Jorge-Tzoc, the petitioner in Melgar de Torres “was an adultwho offered no objective evidence that her uncle’s killing was politically motivated.” 435 F.3d at150. To the extent that Melgar de Torres suggested, furthermore, that even if the petitioner hadestablished the link between her uncle’s killing and his political activities, such killing couldthen not be considered part of her past persecution, this suggestion was clearly dicta.

The majority professes no opinion on the continued vitality of our holding in Jorge-Tzoc9

and claims that nothing in today’s decision “preclude[s] the BIA from considering the totality ofcircumstances in any particular case to determine if an asylum applicant has carried his statutory

64

1 H-’s brother and father were “not personally experienced” by the applicants, neither Chen nor H-

2 could base their asylum applications on such harm after today’s decision. 7

3 The holding today also calls into question our own caselaw – as well that of other circuits

4 – in which appellate panels have recognized that harm inflicted upon one individual may give

5 rise to, or at least help establish, persecution of another in certain circumstances. In Jorge-Tzoc

6 v. Gonzales, 435 F.3d 146 (2d Cir. 2006) (per curiam), this Court, acknowledging that petitioner

7 had not been “victimized directly” when as a young boy, his sister and her family were killed for

8 their political activities, nevertheless remanded the case to the BIA for further proceedings to

9 determine whether his age, coupled with the harm to his family members, helped to establish

10 past persecution. Id. at 150 (internal quotation marks omitted). While the decision rested in8

11 great part on the petitioner’s age, this decision illustrates another category of asylum cases where

12 it might be appropriate to consider harm to others in determining past persecution. Similarly,9

Page 333: Sotomayor Cases

burden.” Maj. Op. at 26 n.13. Yet, the majority fails to explain why the totality ofcircumstances may not be applied in the context of married couples who suffer under coercivepopulation control programs. If the BIA could consider the direct harm to Jorge-Tzoc’s familymembers in determining whether Jorge-Tzoc himself had been persecuted, it should be able toconsider the targeting of and effect on an individual when his or her spouse is forced to undergoan abortion or sterilization.

65

1 other circuits have confronted situations where they found persecution relying in whole or in part

2 on harm to others in certain circumstances. In Sun Wen Chen, the Third Circuit upheld In re S-

3 L-L-, the very BIA determination the majority strikes down today, in part by acknowledging that

4 the physical harm to one’s spouse is nevertheless harm to both spouses in the family planning

5 context. --- F.3d ---, 2007 WL 1760658, at *5 (“In a great many cases, forced abortion or

6 involuntary sterilization of one spouse will directly affect the reproductive opportunities of the

7 other spouse . . . . And persecution of one spouse can be one of the most potent and cruel ways of

8 hurting the other spouse . . . .”). The Sixth Circuit in Abay v. Ashcroft, 368 F.3d 634, 642 (6th

9 Cir. 2004), determined that an applicant was entitled to asylum because she had fled Ethiopia

10 with her teenage daughter to protect the teenager from undergoing forced genital mutilation.

11 The Abay court specifically noted that derivative asylum under § 1158 was not available to Abay

12 – as she was neither a spouse nor a child of a persecuted individual – but granted her asylum,

13 observing that several oral IJ and BIA decisions “suggest a governing principle in favor of

14 refugee status in cases where a parent and protector is faced with exposing her child to the clear

15 risk of being subjected against her will to a practice that is a form of physical torture causing

16 grave and permanent harm.” Id. at 642.

17 Having carefully weighed the law and arguments presented in this appeal, I must concur

18 in the judgment for the reasons already stated ably by Judge Katzmann. I agree in particular

Page 334: Sotomayor Cases

While I ultimately agree with Third Circuit’s recent pronouncement in Sun Wen Chen10

that the BIA properly interpreted an ambiguity in § 601 and § 1101(a)(42) to extend asylumprotection to spouses of individuals forced to abort pregnancies or undergo sterilization, SunWen Chen involves a question we need not reach here because the petitioner here is not married. Had the majority dealt only with the question presented, I would adopt the approach as outlinedby then-Judge Alito in Cai Luan Chen.

Unlike my esteemed colleague Judge Calabresi, I do not find that the BIA limited its analysis11

to § 601 in In re S-L-L-, but rather was grounding its interpretation in both the specific languageof § 601 and the more general definition of “refugee” in § 1101(a)(42). Because the statute isambiguous, I would defer to the BIA’s interpretation here.

66

1 with Judge Katzmann that the Third Circuit in Cai Luan Chen v. Ashcroft, 381 F.3d 221 (3d Cir.

2 2004) (Alito, J.), did what we should have done here. In that case, then-Judge Alito found no

3 need to reach the question of whether § 601 or § 1101(a)(42) were ambiguous because Chen,

4 who was not married to his fiancée on whom the forced abortion was performed, could prevail

5 only if the BIA’s distinction between married and unmarried couples was unreasonable. Id. at10

6 227. Judge Alito ultimately ruled that the distinction was reasonable and denied the petition. Id.

7 at 235. This analysis should control our own very similar cases here. 11

Given the above, the majority should never have reached the question it has taken upon

itself to resolve, particularly in the immigration context where the Supreme Court has long

recognized “that judicial deference to the Executive Branch is especially appropriate . . . where

officials ‘exercise especially sensitive political functions that implicate questions of foreign

relations.’” INS v. Aguirre-Aguirre, 526 U.S. 415, 425 (1999) (quoting INS v. Abudu, 485 U.S.

94, 110 (1988)). In reaching this question, the majority has, I fear, started a domino effect that

may have significant and unforeseen repercussions. Finally, the majority, in noting that “[i]f this

conclusion is inconsistent with Congress’s intentions, [Congress] can, if it so chooses, of course,

amend the statute,” Maj. Op. at 22 n.10, seems to take comfort that its conclusion, if wrong, may

Page 335: Sotomayor Cases

67

be simply overturned. But for those petitioners who fled a draconian population control program

because their spouses had been forced to undergo an abortion or sterilization, the majority’s

caveat must be cold comfort indeed.

Page 336: Sotomayor Cases

68

CALABRESI, Circuit Judge, concurring in part and dissenting in part:

What is remarkable about this case is that essentially everyone on this court

agrees that IIRIRA § 601(a), codified at 8 U.S.C. § 1101(a)(42)(B), cannot be read to grant per

se asylum to spouses. This is remarkable because it is in direct conflict with every other circuit,

the BIA, and ten years of rulings. See Maj. Op. at 5 & n.4. Yet we all agree. And we are

correct. Moreover, it follows that, because § 601(a) does not grant that kind of asylum to

spouses, it also cannot be read as granting asylum to non-spouses — like the petitioners in this

case. This part of the majority’s analysis is admirable, and I join it.

Because § 601(a), in unambiguous terms, grants per se “refugee” status only to the

“person who has been forced to abort a pregnancy or to undergo involuntary sterilization,” the

panel in this case — consisting of the author of the majority opinion, the principal concurrence,

and me — sent the case back to the BIA, to allow the agency to reexamine its decision in In re

C-Y-Z- (C-Y-Z-), 21 I. & N. Dec. 915 (B.I.A. 1997) (en banc) (construing § 601 as granting per

se “refugee” status to spouses). The BIA stuck to its decision, see In re S-L-L- (S-L-L-), 24 I. &

N. Dec. 1 (B.I.A. 2006) (en banc), but has not convinced us that C-Y-Z-’s rule can be squared

with the plain text of § 601(a), and so we now appropriately say that the BIA was wrong.

Unfortunately, both the majority and concurrences are not willing to stop with that,

which was the issue clearly before us and fully considered by the BIA. For reasons that are quite

understandable, but nonetheless wrong — both in terms of results and in terms of what the

Supreme Court has said about our relationship to the BIA — the majority and the concurrences

Page 337: Sotomayor Cases

69

go further. They do so in different directions, and that fact is, to me, simply additional evidence

that going further was inappropriate.

I

The majority says that if the BIA were to construe the general definition of “refugee”

found in 8 U.S.C. § 1101(a)(42)(A) as granting per se refugee status to certain categories of

people — e.g., spouses or non-spouses — that would be an impermissible reading of §

1101(a)(42)(A). This seems to me to be mistaken on several counts.

A

First, the majority relies on the fact that § 601(a) was enacted to overturn Matter of

Chang, 20 I. & N. Dec. 38, 44 (B.I.A. 1989) (adopting the rule that victims of a coercive family

planning regime could claim refugee status only if the victims demonstrated that the family-

planning policy had been “selectively applied” to them on the basis of a protected ground). See

Maj. Op. at 17. The majority posits that § 601(a) was not intended to do more than overturn

Chang, and that Chang is therefore left in place as to spouses and partners who are not

themselves forcibly aborted or involuntarily sterilized. See Maj. Op. at 18 (“The inclusion of

some obviously results in the exclusion of others.”). And, under Chang, spouses and partners

are not entitled to per se refugee status.

Fair enough, but Chang is not an opinion of the Supreme Court, or even of a Court of

Appeals; it is an interpretation of underlying statutory law by the BIA. As such, the agency is

perfectly free to change it — so long as the change is not inconsistent with the underlying law.

Page 338: Sotomayor Cases

See Chevron U.S.A., Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837 (1984).1

70

Thus, any suggestion that the BIA could not, because of Chang, now grant per se status to

spouses pursuant to § 1101(a)(42)(A) is a non sequitur, plain and simple.

B

Second, the logical consequences of what the majority seems to be saying appear to me

to be untenable. Suppose the BIA were to issue an interpretation of § 1101(a)(42)(A) that said,

categorically, that any child who sees his parents tortured and murdered before him by a

totalitarian government — say, the Nazis — is persecuted, and therefore eligible for asylum.

Would such a ruling be invalid under § 1101(a)(42)(A)’s broad definition of refugee? If that is

what the majority is saying, it is, in my judgment, manifestly absurd. There is nothing in the

language or history of § 1101(a)(42)(A) that suggests the BIA could not adopt such a per se rule.

But, if the BIA could adopt the kind of per se rule I described above — and I believe a

majority of our court would agree with me that such a rule would indeed be proper — then it is

improperly premature to say — as today’s governing opinion does — that the agency could not

adopt an analogous per se rule with respect to individuals in the situation of the petitioners in

this case. It may be that if the BIA did adopt such a per se rule, I would ultimately agree with

the majority that, in the context of coercive family planning laws, such an interpretation of §

1101(a)(42)(A) is “unreasonable” at Chevron’s Step Two. But once it is admitted that some1

categorical per se asylum rules — like the one involving my hypothetical children — might be

valid under § 1101(a)(42)(A) (i.e., would get by Chevron Step One) — it is, I believe,

impermissible to say that an equivalent per se interpretation dealing with spouses would

necessarily be invalid if it were adopted — which is in effect what the majority’s holding

Page 339: Sotomayor Cases

71

amounts to. It is impermissible given the Supreme Court’s unanimous decisions in INS v.

Orlando Ventura, 537 U.S. 12 (2002) (per curiam), and Gonzales v. Thomas, 547 U.S. 183, 126

S. Ct. 1613 (2006) (per curiam).

1

In Ventura, the Supreme Court held that, “[g]enerally speaking, a court of appeals should

remand a case to an agency for decision of a matter that statutes place primarily in agency

hands.” 537 U.S. at 16. “This principle,” the Court explained, “has obvious importance in the

immigration context,” id. at 16-17, because “[w]ithin broad limits the law entrusts the agency to

make the basic asylum eligibility decision here in question.” Id. at 16. Subsequent case law has

only strengthened Ventura’s reasoning. The “ordinary remand rule” was recently reaffirmed by

a unanimous Supreme Court in Thomas, 126 S. Ct. 1613, and has been followed by our court in a

series of cases, most notably in Ucelo-Gomez v. Gonzales, 464 F.3d 163, 168-70 (2d Cir. 2006).

Yet, despite the “obvious importance” of the ordinary remand rule in the immigration context,

the majority insists on precluding the BIA from interpreting § 1101(a)(42)(A)’s general

provisions in the first instance. In my view, this aspect of the majority’s holding is dangerously

in tension with Ventura’s command.

In Ventura — much as in the case before us — the Ninth Circuit reversed a holding of

the BIA, and then “went on to consider an alternative argument that the Government had made

before the Immigration Judge,” but which “the BIA itself had not considered . . . .” 537 U.S. at

13. Specifically, the Ninth Circuit reversed the BIA’s holding that the petitioner was not

persecuted “on account of” a “political opinion,” but then, rather than remanding to the BIA for

Page 340: Sotomayor Cases

72

further proceedings, the court evaluated for itself, and rejected, the government’s alternative

argument that the petitioner failed to qualify for asylum because of changed country conditions

in Guatemala. Id. In reversing the Ninth Circuit’s judgment, the Supreme Court found that the

court of appeals

seriously disregarded the agency’s legally-mandated role. Instead, it

independently created potentially far-reaching legal precedent about . . . a highly

complex and sensitive matter. And it did so without giving the BIA the

opportunity to address the matter in the first instance in light of its own expertise.

Id. at 17.

More recently, the Court in Thomas reversed a Ninth Circuit decision which had

decided, without first remanding the issue to the BIA, “that in principle ‘a family may constitute

a social group for the purposes of the refugee statutes,’ . . . [and] that the particular family at

issue . . . fell within the scope of the statutory term ‘particular social group.’” 126 S. Ct. at 1614

(quoting Thomas v. Gonzales, 409 F.3d 1177, 1187, 1189 (9th Cir. 2005) (en banc)) (emphasis

added). Quoting Ventura — and echoing the basic principle of SEC v. Chenery Corp. (Chenery

I), 318 U.S. 80 (1943), that “an appellate court cannot intrude upon the domain which Congress

has exclusively entrusted to an administrative agency,” id. at 88 — the Thomas Court reiterated

that “[a] court of appeals is not generally empowered to conduct a de novo inquiry into the

matter being reviewed and to reach its own conclusions based on such an inquiry.” Thomas, 126

S. Ct. at 1615 (quoting Ventura, 537 U.S. at 16 (internal quotation marks omitted)).

Page 341: Sotomayor Cases

73

In Ucelo-Gomez, a panel of this court concluded that Thomas and Ventura establish the

rule that “where (as here) the agency has yet to decide whether a group, a thing, or a situation

falls within the ambit of a statutory term, the proper course is for the reviewing court to remand

the matter to the agency in accordance with the well-worn ordinary remand rule.” Ucelo-Gomez,

464 F.3d at 169 (internal quotation marks omitted). Moreover, the panel in Ucelo-Gomez

asserted that “the agency interpretation required by Thomas and Ventura is ‘in the first instance’

a particularized interpretation by the agency.” Id. (emphasis omitted).

As a purely formal matter, the approach taken by the majority today is perhaps

reconcilable with Ventura and Thomas. But it is fundamentally incompatible with the spirit of

those cases. Even if the majority is convinced that C-Y-Z-’s rule would be an unreasonable

construction of § 1101(a)(42)(A), the correct approach would be to allow the agency to make a

determination on that matter first. Instead, the majority opinion — perhaps realizing that it could

not, at this time, authoritatively speak on the question of C-Y-Z-’s reasonableness as a

construction of § 1101(a)(42)(A) — by a preemptive strike strips the BIA of its capacity to

consider the issue under § 1101(a)(42)(A). In so doing, the majority precludes the BIA from

examining thoroughly this “highly complex and sensitive matter,” Ventura, 537 U.S. at 17, and

“independently create[s] . . . far-reaching legal precedent . . . . without giving the BIA the

opportunity to address the matter in the first instance in light of its own expertise.” Id.

Significantly, Ventura and Thomas are designed to prevent just such judicial preemption of BIA

positions, even when that preemption reaches what is arguably the correct result.

2

Page 342: Sotomayor Cases

The question of whether, as a matter of Chevron Step Two “reasonableness” review, the2

BIA could base its C-Y-Z- decision on § 1101(a)(42)(A), is arguably neither a pure question offact, nor of statutory interpretation. And the extent to which such mixed questions may beresolved by a Court of Appeals, without first remanding to the agency for its consideration, hasnot been clearly settled by the Supreme Court. Compare Thomas, 126 S. Ct. at 1615 (“[T]heproper course, except in rare circumstances, is to remand to the agency for additionalinvestigation or explanation.” (quoting Ventura, 537 U.S. at 16) (internal quotation marksomitted)) with id. (requiring remand, and observing that “[t]he matter requires determining thefacts and deciding whether the facts as found fall within a statutory term”); Hussain v. Gonzales,477 F.3d 153, 157-58 (4th Cir. 2007) (distinguishing between factual issues not considered bythe BIA, and statutory issues, and reasoning that Ventura and Thomas were directed only towardfactual issues); Fernandez-Ruiz v. Gonzales, 466 F.3d 1121, 1132-35 (9th Cir. 2006)(considering a variety of factors — including the fact that the issue being decided would,following a remand to the BIA, have been reviewed by the court de novo anyway — inconcluding that the Thomas-Ventura remand rule did not apply to the particular issue inquestion); Ucelo-Gomez, 464 F.3d at 170 (“[I]f a reviewing court can state with assuredconfidence (absent agency guidance as to its protectability under the INA) that a group would orwould not under any reasonable scenario qualify as a ‘particular social group,’ it need notremand, and may rule on the issue in the first instance.”).

74

Moreover, even if the majority were not required — as I believe it was — to remand

Zhen Hua Dong’s case to the BIA, it should have remanded his case as a matter of wise2

discretion. Cf. Jian Hui Shao v. Bd. of Immigration Appeals, 465 F.3d 497, 501, 503 (2d Cir.

2006) (concluding that “the BIA is better situated than we are to decide the statutory

interpretation question in the first instance,” and noting that “[o]ur decision to remand this

question of law to the BIA for resolution in the first instance is supported by recent decisions of

the Supreme Court of the United States and our Court” (emphases added)); Yuanliang Liu v. U.S.

Dep’t of Justice, 455 F.3d 106, 116 (2d Cir. 2006) (“Because we conclude, as a matter of

discretion, that it is prudent and useful for us to remand the issue of frviolousness, we need not

address the more complicated question of when remands to the BIA are required by elementary

principles of administrative law.”).

Page 343: Sotomayor Cases

75

I believe that the majority’s haste in narrowly construing § 1101(a)(42)(A) — and,

therefore, in cabining the BIA’s discretion — is particularly troubling given the circumstances of

this case. At stake is a rule of the BIA that has been in place, and relied upon, for over ten years.

See S-L-L-, 24 I. & N. Dec. at 14 (Board Member Pauley, concurring) (“[N]otwithstanding my

belief that Matter of C-Y-Z- . . . was wrongly decided, I would not overrule it now, nearly a

decade later and in the aftermath of thousands of decisions applying it to grant asylum on a

derivative basis.”). In addition, the invalidation of C-Y-Z-’s rule will have sweeping

ramifications for this court’s immigration law docket; by one estimate, “70-80 percent of the

[petitioners in our court] are Chinese seeking asylum to escape their homeland’s family planning

policies.” BIA Appeals Remain High in 2nd and 9th Circuits, The Third Branch: Newsletter of

the Fed. Cts. (Admin. Office of the U.S. Cts. Office of Pub. Affairs, D.C.), Feb. 2005, available

at http://www.uscourts.gov/ttb/feb05ttb/bia/index.html (citing statement of Elizabeth Cronin).

Given all this, our court should have approached the question of C-Y-Z-’s permissibility — either

as it was or with nuanced modifications — not with haste, but with trepidation. For the truth is

that we cannot foretell how the BIA would have interpreted the general definition of §

1101(a)(42)(A), had it been asked to focus on that language.

By trying to decide something that is not yet before us, the majority bars the BIA from

bringing its expertise to bear on this sensitive issue. In the process, the majority does not only

preclude the BIA from reenacting the per se rule of C-Y-Z- — a rule which, at the proper time, I

might well have concluded was “unreasonable” at Chevron Step Two, for all of the reasons the

majority recites. The majority also prevents the agency from interpreting the general language

Page 344: Sotomayor Cases

To cite just one of the many possibilities which the majority prematurely forecloses: had3

the BIA not relied on § 601(a)’s automatic persecution rule, but instead focused on the generalnotion of “persecution,” the agency might have interpreted § 1101(a)(42)(A) as providing (1)that partners who had tried to marry, and were prevented from doing so, but who stayed together,are jointly eligible for asylum (which conclusion would both (a) promote the congressionalpolicy of keeping families together, and (b) extend asylum eligibility to individuals not alreadycovered by § 601(a)); but (2) that husbands who are legally married at the time of a wife’s forcedabortion, but who choose to leave their wives behind for good, are not.

I am mindful that the Supreme Court has cautioned that respect for the role and4

expertise of agencies does not “require that we convert judicial review of agency action into aping-pong game,” and that, therefore, remand is not required when it “would be an idle anduseless formality.” NLRB v. Wyman-Gordon Co., 394 U.S. 759, 766 n.6 (1969); see also Li ZuGuan v. INS, 453 F.3d 129, 135-38 (2d Cir. 2006) (discussing futility standards); Alam v.Gonzales, 438 F.3d 184, 187-88 (2d Cir. 2006) (per curiam) (same). Moreover, and relatedly,the Supreme Court has clarified that a reviewing court must “uphold a decision of less than idealclarity if the agency’s path may reasonably be discerned.” Bowman Transp., Inc. v. Arkansas-Best Freight Sys., Inc., 419 U.S. 281, 286 (1974); see also Motor Vehicle Mfrs. Ass’n v. StateFarm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983) (same).

But, regardless of whether these cases, which limit the necessity to remand, areunderstood to be “exceptions” to the Chenery and Ventura-Thomas requirements, or merely areflection of the deeper truth that formulaic statements cannot substitute for sound judgment inparticular cases, see Li Hua Lin v. U.S. Dep’t of Justice, 453 F.3d 99, 112 (2d Cir. 2006), itremains clear on which side of the line the case before us falls. We simply do not know — and,because the majority and concurring opinions make it almost impossible for the BIA to considerthe general language of § 1101(a)(42)(A) in the first instance, in relation to spouses and partnersof directly victimized persons, we are not likely to learn — how the BIA would have interpreted§ 1101(a)(42)(A) had it been asked to do so. This is not a case in which the agency’s path, whilenot perfectly clear, can “reasonably be discerned”; nor is it a case in which the agency’s likelyresponse to a remand can be predicted with confidence. Rather, it is a case in which (1) the BIAhas not yet spoken — at all, and certainly not clearly — on § 1101(a)(42)(A)’s breadth in thisarea, and (2) our court has, unfortunately, chosen to make further inquiry impossible.

76

of § 1101(a)(42)(A) in ways which might have suffered from none of the problems the majority

properly associates with the current per se rule — ways, incidentally, which might truly have

promoted congressional policy goals. In this respect, the majority opinion keeps the agency3

from doing what administrative agencies do best, namely, using their expertise to convert

general statutes into specific rules that best reflect an underlying legislative intent.4

* * * *

Page 345: Sotomayor Cases

77

Ironically, it was precisely because of the above reasons that the panel in Shi Liang Lin

— comprised, as I mentioned earlier, of the author of the majority opinion, the principal

concurrence, and me — sent it back originally. Yet if the majority’s reasoning were valid, then

there would have been no reason for the panel to do so. Nonetheless, we sent it back then, and

the BIA ruled only as to whether per se refugee status could be granted directly under § 601. See

infra Part II.B. If the case were sent back again, to allow the agency to consider whether to

extend per se protection under § 1101(a)(42)(A), it is possible that the BIA would have agreed

with the majority that no such protection should be adopted. Or the BIA might have adopted a

more sensible rule. Under the majority’s approach, we will never know.

Accordingly, I respectfully, partially, dissent from the majority opinion.

II

But I cannot join the concurrences either. They act as if the BIA, because it mentioned

“nexus” in passing, made a ruling under § 1101(a)(42)(A). It didn’t. Since the agency has yet to

interpret the broad language of that section, it is wrong for us to say — as the concurrers do —

that the agency expressed views to which we owe deference. And this is so, regardless of

whether such a ruling, had it been made, would have passed the requirements of Chevron Step

Two.

A

In SEC v. Chenery Corp. (Chenery I), 318 U.S. 80 (1943), and SEC v. Chenery Corp.

(Chenery II), 332 U.S. 194 (1947), the Supreme Court articulated, and then reaffirmed, “a simple

but fundamental rule of administrative law”: “[A] reviewing court, in dealing with a

Page 346: Sotomayor Cases

78

determination or judgment which an administrative agency alone is authorized to make, must

judge the propriety of such action solely by the grounds invoked by the agency.” Chenery II,

332 U.S. at 196. And “[i]f those grounds are inadequate or improper, the court is powerless to

affirm the administrative action by substituting what it considers to be a more adequate or proper

basis.” Id. The reason for this rule is obvious: “If an order is valid only as a determination of

policy or judgment which the agency alone is authorized to make and which it has not made, a

judicial judgment cannot be made to do service for an administrative judgment,” because “an

appellate court cannot intrude upon the domain which Congress has exclusively entrusted to an

administrative agency.” Chenery I, 318 U.S. at 88.

The Chenery decisions also recognized “an important corollary of the foregoing rule”: “If

the administrative action is to be tested by the basis upon which it purports to rest, that basis

must be set forth with such clarity as to be understandable.” Chenery II, 332 U.S. at 196. As the

Court explained, “[i]t will not do for a court to be compelled to guess at the theory underlying

the agency’s action; nor can a court be expected to chisel that which must be precise from what

the agency has left vague and indecisive.” Id. at 196-97. If it were otherwise, an appellate court

could impose its own policy judgments under the guise of “review.”

Our court has repeatedly recognized and applied these fundamental rules of

administrative law: (1) we may only review that which an agency itself has stated; and (2) the

agency must make those statements in clear terms. See, e.g., Riverkeeper, Inc. v. EPA, 475 F.3d

83, 105 (2d Cir. 2007) (“We cannot opine on this subject, because we must consider only those

justifications that the [agency] offered at the time of the rulemaking.”); Singh v. U.S. Dep’t of

Page 347: Sotomayor Cases

79

Justice, 461 F.3d 290, 294 n.3 (2d Cir. 2006) (“[W]e cannot, on appeal, substitute an argument

— even one the BIA made in another context — for those that the BIA actually gave to support

the conclusion . . . dispute[d] on appeal.”); Cao He Lin v. U.S. Dep’t of Justice, 428 F.3d 391,

400 (2d Cir. 2005) (“[W]e will limit our review of the [agency’s] decision to the reasons [it]

actually articulates . . . . To assume a hypothetical basis for the [agency’s] determination, even

one based in the record, would usurp [the agency’s] role.”); Shi Liang Lin v. U.S. Dep’t of

Justice, 416 F.3d 184, 192 (2d Cir. 2005) (“The government suggests that we may simply supply

our own rationale for the BIA’s decision in C-Y-Z- and then act accordingly. But the Supreme

Court has made clear that ‘[i]t will not do for a court to be compelled to guess at the theory

underlying [a particular] agency’s action; nor can a court be expected to chisel that which must

be precise from what the agency has left vague and indecisive.’ It is not difficult to understand

why. Were courts obliged to create and assess ex-post justifications for inadequately reasoned

agency decisions, courts would, in effect, be conscripted into making policy.” (quoting Chenery

II, 332 U.S. at 196-97)).

B

I recite these well-known tenets of administrative law because I believe that they

preclude us from taking the route advocated by the concurring opinions. Because the BIA’s

opinion in S-L-L- is lacking in clarity, it is certainly possible, with some creativity, to construe

the decision as having been based on rationales which the BIA itself did not invoke. But we are

not empowered to invoke those reasons. The BIA is required to speak for itself.

1

Page 348: Sotomayor Cases

80

The precise basis of the BIA’s decision in C-Y-Z- was anything but clear. But the BIA

and this court have in the past stated that it was based on a construction of § 601(a). See Shi

Liang Lin, 416 F.3d at 188 (noting that, in C-Y-Z-, “the BIA held that, under IIRIRA § 601(a),

the forced sterilization or abortion of one spouse is an act of persecution against the other spouse

. . . .” (emphasis added)); see also id. at 191 (“[A] fresh look at C-Y-Z- reveals that the BIA

never adequately explained how or why, in the first instance, it construed IIRIRA § 601(a) to

permit spouses of those directly victimized by coercive family planning policies to become

eligible for asylum themselves.” (emphasis added)); see S-L-L-, 24 I. & N. Dec. at 3 (“In Matter

of C-Y-Z-, supra, we held that a husband whose wife was forcibly sterilized could establish past

persecution under this amendment [i.e., IIRIRA § 601(a)]” (emphasis added)).

In Shi Liang Lin, the panel

remand[ed] the instant petitions to the BIA so that the BIA [could]: (a) moreprecisely explain its rationale for construing IIRIRA § 601(a) to provide that the‘forced sterilization of one spouse on account of a ground protected under the Actis an act of persecution against the other spouse’ and that, as a result, the spousesof those directly victimized by coercive family planning policies are per se aseligible for asylum as those directly victimized themselves; and (b) clarifywhether, when, and why boyfriends and fiancés may or may not similarly qualifyas refugees pursuant to IIRIRA § 601(a).

Shi Liang Lin, 416 F.3d at 192 (emphases added).

Thus, in remanding, the Shi Liang Lin panel plainly assumed that the BIA’s ruling

in C-Y-Z- was based on a construction of § 601(a), and accordingly, requested that the BIA

explain how § 601(a) might plausibly be read in such a manner. Consistent with these

instructions, the BIA’s response in S-L-L- focused on the scope of § 601(a). See S-L-L-, 24 I. &

N. Dec. at 1 (“The United States Court of Appeals for the Second Circuit has remanded this case

Page 349: Sotomayor Cases

The “IIRIRA amendment” refers, of course, to § 601(a).5

81

with a request that we further explain our rationale in Matter of C-Y-Z-, ‘for construing IIRIRA §

601(a) to provide that the “forced sterilization of one spouse on account of a ground protected

under the Act is an act of persecution against the other spouse” . . . .’” (internal citation

omitted)); id. at 4 (“[W]e reaffirm our holding in Matter of C-Y-Z- . . . .”).

Perhaps sensing that § 601(a)’s text had little to offer, however, the BIA’s

decision in S-L-L- provided little analysis of that text. Instead, the BIA asserted, conclusorily,

that its decision in C-Y-Z- “reflects the significant tensions inherent in the IIRIRA amendment”5

and that “[t]here is no clear or obvious answer to the scope of the protections afforded by the

amendment to partners of persons forced to submit to an abortion or sterilization.” S-L-L-, 24 I.

& N. Dec. at 4. The BIA did not hint at what these “tensions” might plausibly be, or why the

plain language of § 601(a) should not be taken as establishing the proper scope of “the IIRIRA

amendment.” Instead, the BIA “reaffirm[ed] [its] holding in Matter of C-Y-Z-,” id., largely on

the basis of stare decisis and Congress’s supposed acquiescence.

It is the BIA’s reticence to engage with § 601(a)’s text — and the resulting

ambiguity in the BIA’s opinion in S-L-L- — that the concurring opinions now seek to convert

into an argument that S-L-L- was based, not on § 601(a), but on the general definition of

“refugee” found in § 1101(a)(42)(A). It is true that, at one point in S-L-L-, the BIA obscurely

remarked that “[a]lthough there is no specific reference in the statutory definition of a refugee to

a husband’s claim based on harm inflicted upon his wife, the general principles regarding nexus

and level of harm apply in determining such a claim.” S-L-L-, 24 I. & N. Dec. at 5.

Page 350: Sotomayor Cases

82

But this phrase cannot, I believe, establish — as the concurring opinions would

have it — that the BIA’s decision in S-L-L- was based on the general terms “persecution” and

“political opinion” found in § 1101(a)(42)(A). And, even if it did, it would not do so clearly (as

required by Chenery II). Indeed, one can say, as to that: manifestly not.

Notably, in its very next breath, after using the nexus phrase relied on by the

concurrers, the BIA in S-L-L- stated that it was applying “general principles requiring nexus and

level of harm for past persecution in assessing a claim under the IIRIRA amendment.” Id.

(emphasis added). It is permissible to read § 601(a) in this way, the BIA argued, because

“[a]lthough the wife is obviously the individual subjected to the abortion procedure, Congress

was concerned not only with the offensive assault upon the woman, but also with the obtrusive

government interference into a married couple’s decisions regarding children and family.” Id. at

6. Therefore, the BIA concludes, “[w]hen the government intervenes in the private affairs of a

married couple to force an abortion or sterilization, it persecutes the married couple as an

entity.” Id. (emphasis added). It seems to me patent that the BIA reached this conclusion under

§ 601(a), and not § 1101(a)(42)(A).

2

In my view, then, the BIA’s decisions in C-Y-Z- and S-L-L- were grounded in a

(mistaken) belief that, based on an “entity theory” of persecution, spouses of those directly

victimized by coercive family planning policies could themselves become directly eligible for

asylum under § 601(a). And it is not enough for the concurring opinions to cast doubt on my

conclusion; Chenery II’s “clarity corollary” requires that the agency make clear its decision to

Page 351: Sotomayor Cases

83

rest upon a purported ground. Thus, to restate my problem with the concurring opinions: They

would use the fact that the BIA refused to engage clearly with the text of § 601(a) as a basis for

concluding that the BIA was relying on something else. But the incompatibility of this approach

with Chenery II is apparent.

Moreover, the (at best) ambiguousness of the BIA’s decision in S-L-L- results in

precisely the problems adverted to in Chenery II. For it is far from clear that, had the BIA

focused on the general definition of § 1101(a)(42)(A), the agency would have preserved C-Y-Z-

’s rule in its current form. That is, had the BIA been asked to examine, not § 601(a)’s automatic-

eligibility rule, but instead the more general definition of “refugee,” it is quite possible that the

BIA would have come up with a different per se rule, and perhaps even one that would have

avoided the many problems inherent in its C-Y-Z- approach. See supra at 8-9.

Under the concurring opinions’s approach, we are unlikely to know. For, by

reading the agency’s opinion as deciding that which it did not decide — and certainly did not

decide clearly — the concurring opinions, in effect, preclude the agency from thinking deeply

and fully about the matter. And that is the very thing which the clarity requirement of Chenery

II is meant to make the agency do.

III

In the end, as at the beginning, the BIA read us to ask — what we in fact asked:

whether C-Y-Z-’s rule could be based upon § 601(a)’s text, and if so, what its reasons were. The

agency could, under our remand, have turned more broadly to § 1101(a)(42)(A). It didn’t, and it

certainly didn’t do so clearly. Today, we properly reject the BIA’s ruling interpreting the

Page 352: Sotomayor Cases

The majority, attempting to answer my opinion, says, at footnote 15, that remanding6

Zhen Hua Dong’s case to the BIA would be engaging in useless “ping pong.” With greatrespect, the majority in that footnote simply repeats its conflation of two quite separate things. Itis certainly true that the BIA has had multiple occasions to consider the “spousal” question under§ 601(a), and has answered (incorrectly, we all agree) that per se persecutee status is available tospouses under that section. But it has never been asked what the status of spouses or of peoplesituated like Zhen Hua Dong would be under the general terms of § 1101(a)(42)(A), if § 601(a)did not cover spouses. And, in view of its consistent – but incorrect, we today hold – rulings that§ 601(a) did apply to spouses, the BIA never had any reason to address that question on its own. The concurrers, nevertheless, act as if the BIA had addressed the question and had validly givenspouses per se persecutee status under § 1101(a)(42)(A). The majority holds that even if the BIAwere to consider the question, it could not validly say that spouses et al. were covered per se. Both the majority and the concurrers seem to me to overstep, and for precisely the reasonsindicated in Ventura and Thomas. It is not proper for appellate courts to speak for the BIA andto decide the validity of that “speech,” before the agency has had a full and focused opportunityto make its position clear. On § 1101(a)(42)(A), the agency has not yet had that opportunity. Itis not ping pong when only one player has been invited to the relevant table.

84

coverage of § 601(a). But in the spirit of Ventura, Thomas, and our own tradition of sending

things back to the BIA for a first reading, we should now ask the BIA something that it has never

been asked by any court: What would you do under § 1101(a)(42)(A), given that § 601(a) does

not give you the authority to do what you did in C-Y-Z- and S-L-L-?6

We do not know what answer the BIA would give to that question for the simplest

of reasons. The agency has never been specifically asked. And we should not, indeed cannot

properly, assume that what it would say in response — one way or another — would be either a

reasonable or an unreasonable interpretation of the statute. Moreover, since it is possible that

such interpretation might have covered Zhen Hua Dong, I cannot concur with the majority and

concurrences that his case is now hopeless.

The sad thing is that, in their rush to reach a result in terms of who gets asylum

and who does not, both the majority and the concurrers sanction bad law and bad practices with

Page 353: Sotomayor Cases

85

respect to our relationship with the BIA. The reason they do this is certainly understandable.

But it is all unnecessary. It’s just being in a hurry.

* * * *

For all these reasons, while I concur with the majority opinion insofar as it (1)

dismisses the petition of Xian Zou for lack of jurisdiction; (2) denies the petition of Shi Liang

Lin as moot; and (3) persuasively interprets 8 U.S.C. § 1158(c)(2)(A) as being limited to a

“fundamental change” in country conditions, I must respectfully dissent from the premature

denial of Zhen Hua Dong’s petition.

Page 354: Sotomayor Cases

69E.E.O.C. v. J.B. HUNT TRANSPORT, INC.Cite as 321 F.3d 69 (2nd Cir. 2003)

cal in directing that any lack of claritymust be resolved in favor of the insured.See Ferraiolo Const. Co., 584 A.2d at 609(‘‘Any ambiguity must be resolved in favorof a duty to defend.’’) (Maine law); Wil-kin Insulation Co., 161 Ill.Dec. 280, 578N.E.2d at 930 (‘‘All doubts and ambiguitiesmust be resolved in favor of the insured.’’)(Illinois law); Lime Tree Vill. Cmty. ClubAss’n, 980 F.2d at 1405 (‘‘If the allegationsof the complaint leave any doubt as to theduty to defend, the question must be re-solved in favor of the insured.’’) (Floridalaw). Regardless of which of the threestate’s laws applied, the potential for cov-erage existed at the time CI refused todefend Auto Europe. The duty to defendwas therefore ‘‘clear’’ and, accordingly, thedistrict court properly awarded attorney’sfees.

VI. Conclusion

The district court properly concludedthat this insurance coverage disputeshould be heard in Maine and resolvedpursuant to Maine law. Because CI’s dutyto defend was clear, the district courtproperly awarded attorney’s fees to AutoEurope.

The judgment of the district court istherefore affirmed.

,

EQUAL EMPLOYMENT OPPOR-TUNITY COMMISSION,

Plaintiff–Appellant,

v.

J.B. HUNT TRANSPORT, INC.,Defendant–Appellee.

Docket No. 01–6084.

United States Court of Appeals,Second Circuit.

Argued: Jan. 9, 2002.

Decided: Feb. 5, 2003.

Equal Employment Opportunity Com-mission (EEOC) commenced action pursu-ant to the Americans with Disabilities Act(ADA), alleging that truckload motor carri-er violated the ADA by discriminatingagainst over-the-road truck drivers whoused certain prescription medications. Onparties’ cross-motions for summary judg-ment, the United States District Court forthe Northern District of New York, Nor-man A. Mordue, J., 128 F.Supp.2d 117,granted summary judgment in favor ofcarrier, and EEOC appealed. The Court ofAppeals, F.I. Parker, Circuit Judge, heldthat applicants perceived as unsuitable forposition of over-the-road truck driverswere not perceived as substantially limitedin major life activity of working, as wouldestablish ‘‘disability’’ under the ADA.

Affirmed.

Sotomayor, Circuit Judge, dissentedand filed opinion.

leaves some ambiguity on whether allegationsof intentional conduct eliminate the duty todefend pursuant to an intentional acts policyexclusion even when facts could be developedat trial to support judgment for the plaintiff

based on non-intentional conduct. See, e.g.,Applestein, 377 So.2d at 231 (holding thatallegations of malice and deliberate ‘‘ ‘attemptto discredit’ ’’ negated coverage).

Page 355: Sotomayor Cases

70 321 FEDERAL REPORTER, 3d SERIES

1. Federal Courts O776, 802

The Court of Appeals reviews a dis-trict court’s grant of summary judgmentde novo, construing the evidence presentedbelow in the light most favorable to thenon-moving party.

2. Federal Courts O759.1, 766

While the Court of Appeals may af-firm a district court’s grant of summaryjudgment on any ground with adequatesupport in the record, it may not affirmsummary judgment where any evidence inthe record would support a reasonable in-ference in favor of the opposing party.

3. Civil Rights O173.1

Applicants’ perceived unsuitability forposition of over-the-road truck drivers,based on their use of prescription medi-cations with side effects that could impairdriving ability, was not a perceived inabili-ty to perform broad range or class of jobs,but rather was limitation on particular jobwithin larger group of jobs, and thus appli-cants failed to establish that they wereperceived as substantially limited in majorlife activity of working, as would establish‘‘disability’’ under the ADA. Americanswith Disabilities Act of 1990, § 3(2), 42U.S.C.A. § 12102(2); 29 C.F.R.§ 1630.2(j)(3)(i).

4. Civil Rights O173.1

Truckload motor carrier did not viewapplicants who were perceived unsuitablefor position of over-the-road truck driversbased on their use of prescription medi-cations with side effects that could impairdriving ability as unable to drive anytrucks, so as to regard them as disabledwithin meaning of ADA, but rather per-ceived applicants as unfit to perform spe-cific job of long-distance, freight-carrying,tractor-trailer driving. Americans withDisabilities Act of 1990, § 3(2), 42 U.S.C.A.§ 12102(2); 29 C.F.R. § 1630.2(j)(3)(i).

5. Civil Rights O173.1

Comments made by people other thanultimate hiring authorities suggesting thatcertain applicants for position of over-the-road truck drivers were not suited to anyform of professional driving, based on theiruse of prescription medications with sideeffects that could impair driving ability,were not sufficient to indicate that truck-load motor carrier thought applicants weremore broadly limited in major life activityof working, so as to regard them as dis-abled within meaning of ADA, where carri-er had its own safety requirements aboveand beyond those of federal standards, andit did hire some applicants on medicationsat issue. Americans with Disabilities Actof 1990, § 3(2), 42 U.S.C.A. § 12102(2); 29C.F.R. § 1630.2(j)(3)(i).

6. Civil Rights O173.1

A finding of perceived disability, forpurposes of a discrimination action underthe ADA, may not rest merely on a singleemployer’s failure to hire a candidate.Americans with Disabilities Act of 1990,§ 2 et seq., 42 U.S.C.A. § 12101 et seq.

7. Civil Rights O240(2)

For purposes of a discrimination ac-tion under the ADA, courts will not pre-sume a mistaken assumption of disabilitybased only on an employer’s decision notto hire certain candidates. Americanswith Disabilities Act of 1990, § 2 et seq.,42 U.S.C.A. § 12101 et seq.

8. Civil Rights O173.1

Applicants for over-the-road truckdriver positions who suffered from condi-tions treated with prescription medicationswith side effects that could impair drivingability were not regarded as substantiallylimited in major life activity of workingbased on underlying condition itself, aswould establish ‘‘disability’’ under theADA; rather, employer perceived appli-

Page 356: Sotomayor Cases

71E.E.O.C. v. J.B. HUNT TRANSPORT, INC.Cite as 321 F.3d 69 (2nd Cir. 2003)

cants as unfit for positions based on use ofmedications with dangerous side effects.Americans with Disabilities Act of 1990,§ 3(2), 42 U.S.C.A. § 12102(2); 29 C.F.R.§ 1630.2(j)(3)(i); 49 C.F.R. § 391.41.

Julie L. Gantz, Equal Opportunity Em-ployment Commission (Nicholas M. Inzeo,Acting Deputy General Counsel, Philip B.Sklover, Associate General Counsel, Vin-cent J. Blackwood, Assistant GeneralCounsel, on brief), for Plaintiff–Appellant.

James H. Hanson, Scopelitis, Garvin,Light & Hanson, Indianapolis, IN (LaurieT. Baulig, Scopelitis, Garvin, Light & Han-son, Washington, DC, Thomas J. Grooms,Bond Schoeneck & King, Syracuse, NY, onbrief), for Defendant–Appellee.

Before: JACOBS, F.I. PARKER,SOTOMAYOR, Circuit Judges.

F.I. PARKER, Circuit Judge.

J.B. Hunt Transport, Inc. chose not toemploy over-the-road truck drivers whoused prescription medications with side ef-fects that might impair driving ability.The Equal Employment Opportunity Com-mission argued that under the Americanswith Disabilities Act, Hunt’s decision vio-lated the rights of job applicants usingthose medications. We disagree.

I.

Plaintiff–Appellant Equal EmploymentOpportunity Commission (‘‘EEOC’’) ap-peals from the February 8, 2001 decisionof the United States District Court for theNorthern District of New York (NormanA. Mordue, Judge ) granting defendantJ.B. Hunt Transport Inc.’s (‘‘Hunt’’ ’s) mo-tion for summary judgment and denyingplaintiff EEOC’s cross-motion for sum-mary judgment. The district court found

that the applicants in question had beendenied over-the-road driving positions withHunt because of their use of medicationswith potentially harmful side effects, andnot as a result of an actual or perceiveddisability or a record of disability as con-templated by the Americans with Disabili-ties Act of 1990, 42 U.S.C. § 12101, et seq.(‘‘ADA’’). On appeal, the EEOC arguesexclusively that Hunt regarded the reject-ed applicants as disabled, i.e., substantiallylimited from a major life activity, as de-fined by 42 U.S.C. § 12102(2)(C) becauseof their use of certain medications. Be-cause we find that Hunt did not regard theapplicants as disabled as defined by theADA, we affirm the decision of the districtcourt.

II.

J.B. Hunt Transportation, Inc. (‘‘Hunt’’)is the nation’s largest publicly held motorcarrier company. Hunt operates for-hireproperty transport services in the forty-eight contiguous states, the District of Co-lumbia, Canada, and Mexico. Its fleet in-cludes 8,000 tractors, and it employs ap-proximately 12,000 individuals to drive thetrucks. Of these employees, approximate-ly 10,000 are the over-the-road (‘‘OTR’’)drivers whose positions are at issue in thiscase. These OTR drivers operate vehiclesweighing approximately 80,000 poundsover irregular routes under particularlydifficult work conditions, including sleepdeprivation, irregular work and rest cycles,inclement weather, long driving periods,long layovers, irregular meal schedules,tight delivery schedules, en route delays,night driving, accumulated fatigue, stress,and extended periods of loud noise andvibrations. According to Hunt, the largevehicle size and extreme driving conditionsfaced by its OTR drivers warrant height-ened safety evaluations of those OTR driv-ers.

Page 357: Sotomayor Cases

72 321 FEDERAL REPORTER, 3d SERIES

Like other motor carriers, Hunt is sub-ject to federal regulation under the De-partment of Transportation’s Federal Mo-tor Carrier Safety Act Regulations(‘‘FMCSAR’’). 49 C.F.R. § 301, et seq.(2001). These regulations establish mini-mum qualifications for any person drivinga commercial motor vehicle, as well asminimum duties for motor carriers usingOTR drivers. The regulations specificallyallow an operator to require and enforce‘‘more stringent requirements relating tosafety of operation and employee safetyand health’’, 49 C.F.R. § 390.3(d), and re-quire operators to restrict drivers fromoperating vehicles ‘‘while the driver’s abili-ty or alertness is so impaired, or so likelyto become impaired, through fatigue, ill-ness, or any other cause, as to make itunsafe for him/her to begin or continue tooperate the commercial motor vehicle.’’ Id.at § 392.3 (2001). A motor carrier is re-quired to ensure that drivers do not oper-ate unless they are in compliance with theDOT regulations. 49 C.F.R. §§ 391.11,392.3, 392.4(b)(2001).

A. The Drug Review List

Between September 1993 and May1994, in an effort to comply with theFMCSAR in its hiring processes, Huntcreated a Drug Review List (‘‘DRL’’) ofmedications known to have side effectsthat might impair driving ability. Thelist, thirty-seven pages in length and in-cluding over 836 medications, was com-piled by Hunt’s Safety Department Di-rector of Compliance, David Whiteside

(‘‘Whiteside’’), based entirely on notationsin the 1993 edition of the Physician’s DeskReference (‘‘PDR’’). Whiteside dividedthe DRL into six columns labeled ‘‘name,’’‘‘class,’’ ‘‘comment,’’ ‘‘restriction,’’ ‘‘treats,’’and ‘‘1993 PDR page number.’’ In the‘‘restriction’’ column, Whiteside indicatedthe impact a particular drug might haveon an applicant’s eligibility. Whitesidedesignated five categories of restrictions:‘‘Rule Out Side [E]ffects,’’ ‘‘Not Permit-ted,’’ ‘‘Unsafe [E]ffects,’’ ‘‘Heart Condi-tion,’’ and ‘‘Disqualifying Condition.’’ 1 Anapplicant whose medication had a ‘‘RuleOut Side Effects’’ notation was required toobtain a release from the prescribing doc-tor certifying that the applicant couldsafely drive a tractor trailer truck whileusing the medication. An applicant takinga ‘‘Not Permitted,’’ ‘‘Unsafe Effects,’’‘‘Disqualifying Condition,’’ or ‘‘Heart Con-dition’’ medication could not drive forHunt while using the indicated medi-cation.2 The notation ‘‘Unsafe Effects’’ in-dicated either that the PDR cautionedusers against operating heavy equipmentor driving automobiles while taking thedrug (noted as ‘‘warning on driving’’ in thecomment column) or that the drug causeddrowsiness, sedation, or a high incidenceof dizziness. A ‘‘Rule Out Side Effects’’notation indicated that a medication couldcause side effects similar to, but less per-vasive than, those warranting an ‘‘UnsafeEffects’’ label. Finally, ‘‘Heart Condition’’indicated that the medication was general-ly used for heart problems that could dis-qualify drivers under DOT regulations.

1. The Court will use the corrected labels‘‘Rule Out Side Effects’’ for ‘‘Rule Out SideAffects’’ and ‘‘Unsafe Effects’’ for ‘‘Unsafe Af-fects’’ throughout the opinion.

2. The district court found that Whiteside mis-takenly believed that the DOT prohibiteddrivers from using any Schedule II–V medi-cations, rather than only Schedule I medi-cations, and that he therefore included all of

these medications in the ‘‘Not Permitted’’ cat-egory. Equal Employment OpportunityComm’n v. J.B. Hunt Transp., Inc., 128F.Supp.2d 117, 120 n. 2 (N.D.N.Y.2001); see49 C.F.R. § 391.42(b)(12)(i) (prohibiting useof Schedule I drugs, amphetamines, narcot-ics, and other habit-forming drugs); 49C.F.R. § 392.2 (same).

Page 358: Sotomayor Cases

73E.E.O.C. v. J.B. HUNT TRANSPORT, INC.Cite as 321 F.3d 69 (2nd Cir. 2003)

B. The Hunt Hiring Process

Upon receiving an application for a com-mercial driving position, Hunt forwardedthe application to its Corporate DriverPersonnel Department in Lowell, Arkan-sas for screening of motor vehicle, crimi-nal, and prior employment records and fora review of listed references. If an appli-cant passed this first level of screening andreceived a conditional employment offer,the applicant underwent medical screen-ing, including questioning regarding theapplicant’s use of prescription medicationfor the last five years. Hunt used non-medical personnel to conduct these screen-ings. If the applicant indicated use of aprescription drug, the reviewing employeeconsulted Hunt’s medical guidelines 3 andthe DRL to determine the applicant’s med-ical eligibility.

C. The EEOC Claim

EEOC claims that Hunt improperly re-jected 546 applicants in violation of theADA on the basis of a ‘‘blanket’’ exclusion-ary policy. EEOC admits, however, thatHunt hired several applicants who wereusing drugs prohibited under the DRL—in1995, two applicants using drugs labeled‘‘Disqualifying Condition’’ and eleven usingdrugs labeled ‘‘Unsafe Effects,’’ and in1996 and 1997, one applicant taking a ‘‘Dis-qualifying Condition’’ drug and thirteenusing drugs with ‘‘Unsafe Effects.’’ Priorto commencing work, each of these newemployees provided Hunt with medicaldocumentation from a treating physicianor health care provider certifying that heor she did not suffer from the potentiallyproblematic side effects and could operatea truck safely while taking the drug.

III.

EEOC filed its complaint in the UnitedStates District Court for the NorthernDistrict of New York on October 24, 1997.Both sides moved for summary judgment.EEOC alleged that Hunt violated the ADAby discriminating against individuals withdisabilities or ‘‘perceived’’ disabilities.Hunt alleged that the DRL was a safety-related qualification standard addressingserious business concerns. The districtcourt granted summary judgment forHunt and denied EEOC’s summary judg-ment motion based on (1) its conclusionthat ADA protections did not extend to theexcluded driver-applicants because the ap-plicants were not, by virtue of their use ofcertain medications, disabled within themeaning of the ADA, and (2) its findingthat the EEOC had failed to contradictHunt’s assertion that its use of the DRLas a safety measure was reasonable withinDOT guidelines. Equal Employment Op-portunity Comm’n. v. J.B. Hunt Transp.,Inc., 128 F.Supp.2d 117, 135–36 (N.D.N.Y.2001). On appeal, EEOC abandoned itsargument that the excluded applicantswere ‘‘disabled’’ under the ADA, claimingonly that the district court erred by grant-ing summary judgment to Hunt when theevidence supported the conclusion thatHunt regarded the applicants as disabledbecause of their use of medications on theDRL.

IV.

[1, 2] We review a district court’sgrant of summary judgment de novo, con-struing the evidence presented below inthe light most favorable to the non-movingparty. Manning v. Utils. Mut. Ins. Co.,254 F.3d 387, 391 (2d Cir.2001). While

3. As the district court found, Hunt maintaineda restrictive policy on the use of drugs forpsychological conditions separate from theDRL. An applicant was not eligible to drive

for Hunt unless he or she had been off suchdrugs for at least thirty days before commenc-ing work. Hunt Medical Guidelines, April 11,1996.

Page 359: Sotomayor Cases

74 321 FEDERAL REPORTER, 3d SERIES

this Court may affirm on any ground withadequate support in the record, we maynot affirm summary judgment where anyevidence in the record would support areasonable inference in favor of the oppos-ing party. See McCarthy v. Am. Int’lGroup, Inc., 283 F.3d 121, 124 (2d Cir.2002); VKK Corp. v. Nat’l FootballLeague, 244 F.3d 114, 119 (2d Cir.2001).

A. The Statutory Framework and theDefinition of ‘‘Disability.’’

The ADA provides a deceptively simpledefinition of disability:

The term ‘‘disability’’ means, with re-spect to an individual—

(A) a physical or mental impairmentthat substantially limits one or moreof the major life activities of suchindividual;(B) a record of such an impairment;or(C) being regarded as having such animpairment.

42 U.S.C. § 12102(2)(1995). EEOC regu-lations further develop this definition, ex-plaining ‘‘physical or mental impairment’’as:

(1) Any physiological disorder, or condi-tion, cosmetic disfigurement, or anatomi-cal loss affecting one or more of thefollowing body systems: neurological,musculoskeletal, special sense organs,respiratory (including speech organs),cardiovascular, reproductive, digestive,genito-urinary, hemic and lymphatic,skin, and endocrine; or(2) Any mental or psychological disor-der, such as mental retardation, organicbrain syndrome, emotional or mental ill-ness, and specific learning disabilities.

29 C.F.R. § 1630.2(h) (2001).

Although EEOC initially challengedHunt’s reliance on the DRL under allthree prongs of the statutory definition of

‘‘disability,’’ on appeal, EEOC alleges onlythat the rejected OTR driver applicantswere ‘‘regarded as’’ disabled by Huntbased on their use of certain medications,invoking the statutory definition of disabil-ity under § 12102(2)(C). As the SupremeCourt explained in Sutton v. United AirLines, Inc., ‘‘[t]here are two apparentways in which individuals may fall withinthis [§ 12102(2)(C) ] statutory definition:(1) a covered entity mistakenly believesthat a person has a physical impairmentthat substantially limits one or more majorlife activities, or (2) a covered entity mis-takenly believes that an actual, nonlimitingimpairment substantially limits one ormore major life activities.’’ 527 U.S. 471,489, 119 S.Ct. 2139, 144 L.Ed.2d 450(1999).

Evaluating the evidence before the dis-trict court, this Court agrees with thecourt below that EEOC failed to put forthevidence sufficient to demonstrate that therejected applicants were ‘‘disabled’’ withinthe meaning of the ADA. Specifically,EEOC failed to set forth evidence suffi-cient to establish that Hunt perceived re-jected applicants as substantially limited intheir ability to perform a major life activi-ty.

B. The Evidence Is Insufficient ToSupport the Inference that HuntRegarded Applicants as Having a‘‘Substantial Limitation’’ on a ‘‘Ma-jor Life Activity.’’

To qualify for ADA protections, a per-son’s ‘‘impairment’’ must ‘‘substantiallylimit’’ a ‘‘major life activit[y].’’ 42 U.S.C.§ 12102(2). Major life activities may in-clude ‘‘caring for oneself, performing man-ual tasks, walking, seeing, hearing, speak-ing, breathing, learning,’’ and, pertinent tothis appeal, ‘‘working.’’ 29 C.F.R.§ 1630.2(i). An activity is ‘‘substantiallylimited’’ when an individual cannot per-

Page 360: Sotomayor Cases

75E.E.O.C. v. J.B. HUNT TRANSPORT, INC.Cite as 321 F.3d 69 (2nd Cir. 2003)

form the activity that an average person inthe general population could perform orfaces significant restrictions in the ‘‘condi-tion, manner, or duration under which theindividual can TTT perform [the] activity.’’29 C.F.R. § 1630.2(j)(i)-(ii). The activityof ‘‘working’’ is further defined by theregulations as follows:

With respect to the major life activity ofworking—(i) The term substantially limits meanssignificantly restricted in the ability toperform either a class of jobs or a broadrange of jobs in various classes as com-pared to the average person having com-parable training, skills and abilities.The inability to perform a single, partic-ular job does not constitute a substantiallimitation in the major life activity ofworking.

29 C.F.R. § 1630.2(j)(3)(i). Thus, unlessHunt perceived the applicants in questionas limited from a class of jobs or a broadrange of jobs, the EEOC’s claim must fail.

1. Driving 40–Ton, 18–Wheel TrucksOver Long Distances for ExtendedPeriods is Neither a ‘‘Class of Job’’nor a ‘‘Broad Range of Jobs’’ Withinthe Meaning of the ADA.

[3] Driving freight-carrying tractor-trailer trucks over long distances for ex-tended periods of time is neither a ‘‘classof jobs’’ nor a ‘‘broad range of jobs,’’ as theEEOC alleges, but rather a specific jobwith specific requirements. Such a posi-tion requires specific abilities, especiallythe ability to stay alert over long hoursunder difficult conditions. A Hunt OTRdriver’s alertness cannot flag. He or shemust be able to stay alert and withstandthe mesmerizing affect of driving an eigh-teen-wheel vehicle for hours at a stretch,sometimes at night, with continuous vibra-tion over long distances. Given these de-manding requirements, the fact that onemay not be able to perform the specific job

of a Hunt OTR driver does not mean thatone could not successfully engage in othertypes of truck driving, let alone in otherkinds of safety-sensitive work.

In Sutton, the Supreme Court consid-ered the claims of pilots who had beendenied positions as ‘‘global pilots’’ withUnited Airlines. The Supreme Court heldthat the position of ‘‘global pilot’’ was ‘‘asingle job’’ and, therefore, was not suffi-ciently broad to satisfy the ‘‘major lifeactivity requirement’’. Sutton, 527 U.S. at493, 119 S.Ct. 2139. The Court reasonedthat ‘‘there are a number of other positionsutilizing petitioners’ skills, such as regionalpilot and pilot instructor to name a few.’’Id.

Like the limitation that United Airlinesplaced on global airline pilots in Sutton,the limitation that Hunt placed on appli-cants for the position of OTR driver was alimitation on a particular job within a larg-er group of jobs, and not a substantiallimitation on working. See Baulos v.Roadway Express, Inc., 139 F.3d 1147,-1154 (7th Cir.1998) (driving sleeper trucksis a specific job within the broader class oftruck driving jobs). Therefore, the appli-cants’ perceived unsuitability for the posi-tion of OTR driver cannot be characterizedas a perceived inability to perform a broadrange or a class of jobs. This is true evenassuming that truck-driving in general is asufficiently broad range or class of jobs toconstitute a ‘‘major life activity’’, an issuewe do not need to reach. As the dissentreadily acknowledges, persons licensed todrive the types of vehicles driven by HuntOTR drivers are also qualified to drive‘‘various types of small and large trucks,including tractor-trailers, moving trucks,and cargo vans.’’ Dissent page 80.

Accordingly, to show that Hunt per-ceived applicants rejected under the DRLas substantially limited in a major life ac-

Page 361: Sotomayor Cases

76 321 FEDERAL REPORTER, 3d SERIES

tivity, the EEOC must show that Huntviewed such applicants as limited from abroader range or class of jobs than merelyOTR positions at Hunt.

2. The Evidence Is Not Sufficient ToSupport a Reasonable Inference thatHunt Regarded Applicants RejectedUnder the ‘‘Not Permitted’’ and‘‘Unsafe Effects’’ Categories as Sub-stantially Limited in a Broad Rangeor Class of Jobs.

[4] The EEOC argues that Hunt re-garded applicants who took particularmedications as incapable of driving trucks,which according to the EEOC constituteseither a ‘‘class of jobs’’ or a ‘‘broad rangeof jobs.’’ The record, however, only showsthat Hunt saw the applicants as unfit toperform a job for which they were seekingapplicants: long-distance, freight-carrying,tractor-trailer driving. The SupremeCourt has clearly stated that ‘‘[t]he inabili-ty to perform a single, particular job,’’however, ‘‘does not constitute a substantiallimitation in the major life activity of work-ing.’’ Sutton v. United Air Lines, Inc.,527 U.S. 471, 493, 119 S.Ct. 2139, 144L.Ed.2d 450 (1999). Here, Hunt dismissedthe applicants as unable to meet Hunt’sown safety requirements—requirementsabove and beyond the DOT’s industry-widestandards and unique from the require-ments of other trucking companies. SeeCompl. ¶ 8c; Def.’s Statement of MaterialFacts at 7; see also Adair Dep. at 85–86;J.B. Hunt Transp., Inc., 128 F.Supp.2d at129 n. 17 (noting drivers were employed byother trucking companies while takingsame medications).

The evidence suggests that Hunt foundthe applicants unsuited for long-distancedriving of Hunt’s 40–ton trucks on irregu-lar, stressful schedules, but does not indi-cate that Hunt perceived the applicants asmore broadly limited. The fact that Hunt

did not have another, less demanding driv-ing position to offer the candidates doesnot indicate that Hunt perceived the candi-dates as being unqualified for any drivingposition at all. Giordano v. City of NewYork, 274 F.3d 740, 748–50 (2d Cir.2001)(finding inability of the New York PoliceDepartment to offer light duty, non-patrolposition to officer taking anti-coagulationmedication did not demonstrate that offi-cer was substantially limited in workingwhere other security and law enforcementjobs in the area had such positions); seealso Baulos v. Roadway Express Inc., 139F.3d 1147, 1154 (2d Cir.1998) (concludingthat truck drivers unable to operate sleep-er trucks did not show that they wereregarded as disabled where employer didnot offer them less demanding, non-over-night positions that were taken by driverswith more seniority).

[5] EEOC references a few commentsfrom Hunt’s evaluators to candidates sug-gesting that certain candidates were notsuited to any form of professional driving.These comments, made by people otherthan the ultimate hiring authorities, simplyare not sufficient to indicate that Huntthought the applicants were more broadlylimited given the heightened nature ofHunt’s standards and the fact that Huntdid hire some applicants on DRL medi-cations. Although a few evaluators’ com-ments could be more broadly interpreted,there is no evidence that Hunt’s reviewers,relying on Hunt’s own DRL and drug liststo make a judgment on qualification for aposition at Hunt, intended to make anevaluation beyond Hunt’s specific guide-lines. Nor is there sufficient evidence tosupport a finding that Hunt viewed thedriving limitation as extending beyondHunt. Furthermore, as the Supreme Courthas clearly stated, ‘‘[i]t is not enough tosay that if the physical criteria of a singleemployer were imputed to all similar em-

Page 362: Sotomayor Cases

77E.E.O.C. v. J.B. HUNT TRANSPORT, INC.Cite as 321 F.3d 69 (2nd Cir. 2003)

ployers one would be regarded as substan-tially limited in the major life activity ofworking only as a result of this imputa-tion.’’ Sutton, 527 U.S. at 493, 119 S.Ct.2139.

[6] In short, EEOC demonstrated onlythat Hunt refused to hire certain appli-cants according to its own hiring criteria;however, a finding of perceived disabilitymay not rest merely on a single employer’sfailure to hire a candidate. Baulos, 139F.3d at 1154 (‘‘Courts have uniformly heldthat an employer does not necessarily re-gard an employee as handicapped simplyby finding the employee to be incapable ofsatisfying the singular demands of a par-ticular job.’’ (internal citation omitted)).

[7] Thus, we affirm the district court’sgrant of summary judgment in favor ofHunt as to the applicants rejected under atleast the ‘‘Not Permitted’’ and ‘‘UnsafeEffects’’ categories because EEOC hasfailed to demonstrate that Hunt mistaken-ly perceived that the rejected applicants’had impairments that substantially limiteda ‘‘major life activity.’’ Accordingly,EEOC has failed to show that the appli-cants were ‘‘disabled’’ within the meaningof the ADA. In so holding, we emphasizethat this Court will not presume a mistak-en assumption of disability based only on

an employer’s decision not to hire certaincandidates.

3. The Evidence Is Also Not SufficientTo Support a Reasonable Inferencethat Hunt Regarded Applicants Re-jected Under the ‘‘DisqualifyingCondition’’ and ‘‘Heart Condition’’Categories as Substantially Limitedin a Broad Range or Class of Jobs.

As noted above, two of the categories inthe DRL appear, at least superficially, torefer to the condition causing the relianceon a DRL drug, not merely the applicants’use of a DRL medication.4 Although theevidence suggests that these applicantswere, like those in the other categories,often told that they were disqualified onthe basis of the drug they were usingrather than on the basis of the conditionsupporting their use of the drug, we brief-ly consider whether applicants using drugsfrom these two categories warrant a differ-ent legal conclusion. We conclude thatthey do not.

[8] Individuals suffering from the con-ditions treated with the ‘‘Heart Condition’’or ‘‘Disqualifying Condition’’ drugs are po-tentially explicitly barred from truck driv-ing by 49 C.F.R. § 391.41.5 Hunt therefore

4. As noted in the discussion of the pertinentfacts, the DRL contained five categories ofdrugs: ‘‘Rule out Side Effects,’’ ‘‘Not Permit-ted,’’ ‘‘Unsafe Effects,’’ ‘‘Heart Condition,’’and ‘‘Disqualifying Condition.’’ The EEOCdoes not represent in this appeal any appli-cants rejected under the ‘‘Rule Out Side Ef-fects’’ category, thus removing that categoryfrom our consideration. J.B. Hunt Transp.,Inc., 128 F.Supp.2d at 122, n. 7.

5. Review of the DRL reveals that the label‘‘Disqualifying Condition’’ attached to drugstreating Parkinson’s Disease, serious arryth-mia, alcoholism, epilepsy, seizure, migraines,dementia, depression, schizophrenia, diabe-tes, severe arthritis, severe hypertension, opi-ate addiction, subarachnoid hemorrhage, ar-

tery occlusion, and severe headache, whilethe label ‘‘Heart Condition’’ attached to medi-cines treating heart failure, thrombosis, ede-ma, congestive heart failure, ischemia, andventric arrythmia. 49 C.F.R. § 391.41 ap-pears to exclude persons with all of theseconditions from driving a commercial vehiclewhere those conditions are likely to interferewith their ability to safely drive a commercialvehicle. 49 C.F.R. §§ 391.41(b)(3) (diabetes),391.41(b)(4) (‘‘myocardial infarction, anginapectoris, coronary insufficiency, thrombosis,or any other cardiovascular disease of a vari-ety known to be accompanied by syncope,dyspnea, collapse or congestive heart fail-ure’’), 391.41(b)(6) (high blood pressure),391.41(b)(7) (‘‘rheumatic, arthritic, orthope-dic, muscular, neuromuscular, or vascular

Page 363: Sotomayor Cases

78 321 FEDERAL REPORTER, 3d SERIES

potentially regarded applicants using thesedrugs as substantially limited not justfrom driving Hunt vehicles according tothe rules of the DRL and other companyregulations, but as prevented from drivinglegally for any commercial trucking com-pany. As with the other categories in theDRL, however, the restrictions on themedications labeled ‘‘Disqualifying Condi-tion’’ or ‘‘Heart Condition’’ were placed onapplicants taking the drug, not on appli-cants with the underlying condition itself.Even though in some cases, the company,under 49 C.F.R. § 391.41, could have cre-ated a policy excluding the applicant onthe basis of the underlying condition, thebasis for the exclusion from employmentwas the use of a listed drug, not anypotential ‘‘disability’’ created by the treat-ed disease.6

We conclude, therefore, that any claimsarising under the ‘‘Heart Condition’’ and‘‘Disqualifying Condition’’ categories arenot distinguishable from the claims underthe ‘‘Not Permitted’’ and ‘‘Unsafe Effects’’categories and so they must also fail.

V.

Although Hunt admittedly rejected theapplicants for its OTR driving positionsbecause of their use of certain prescriptionmedications, the EEOC cannot succeed inits ADA claim on behalf the rejected appli-cants. The record only shows that Huntregarded the applicants in question as ine-ligible for a specific position within Hunt,not that Hunt regarded them as ‘‘disabled’’within the meaning of the ADA. The appli-cants, through the EEOC, therefore do nothave a valid ADA claim.

For the reasons set forth above, thisCourt affirms the district court’s grant ofsummary judgment to defendant Hunt andits denial of the cross-motion by plaintiffEEOC.

The judgment of the district court isAFFIRMED.

SOTOMAYOR, Circuit Judge,dissenting.

This case is quite straightforward.Based upon a list of drugs and their poten-

disease’’), 391.41(b)(8) (epilepsy or ‘‘any othercondition which is likely to cause loss of con-sciousness’’), 391.41(b)(9) (‘‘mental, nervous,organic, or functional disease or psychiatricdisorder’’), 391.41(b)(12)(i) (controlled sub-stances), 391.41(b)(13) (alcoholism). ‘‘Mi-granes’’ or ‘‘severe headaches,’’ as ‘‘vascularheadache[s],’’ DORLAND’S ILLUSTRATEDMEDICAL DICTIONARY 1042 (28thed.1994), potentially fall within§ 391.41(b)(7)’s restriction on vascular dis-ease.

EEOC alleged that Hunt misinterpreted aDOT report cautioning about the effects ofdrugs used to treat heart conditions, claimingthat the report merely required individual as-sessment of each patient. The DOT subse-quently issued a report clarifying that the useof Coumadin, a anticoagulator previouslyquestioned, was not automatically disqualify-ing. J.B. Hunt Transp., Inc., 128 F.Supp.2dat 120 n. 3 (describing reports). According tothe DRL, however, Coumadin treats thrombo-sis, a condition specifically prohibited by the

current regulations. See 49 C.F.R.§ 391.41(b)(4).

6. For example, Amandtadine Hydrochloride,a drug to which the ‘‘Disqualifying Condi-tion’’ label attaches treats both Parkinson’sDisease and the flu. While an applicant tak-ing the drug for Parkinson’s might be dis-abled on the basis of the disease within themeaning of the ADA, an applicant using thedrug for the flu would not likely so qualify.Hunt, however, would have excluded eitherapplicant because of the drug usage. Fur-thermore, some ‘‘Disqualifying Condition’’drugs treat the same underlying diseases asdrugs given other labels. For example Zoloft,a drug used for treatment of, inter alia, de-pression, is listed as ‘‘Unsafe Effects,’’ whileProzac, also for depression, is listed as ‘‘Dis-qualifying Condition.’’ This further supportsthe idea that the drug, not the condition itself,was the true basis of Hunt’s hiring ban.

Page 364: Sotomayor Cases

79E.E.O.C. v. J.B. HUNT TRANSPORT, INC.Cite as 321 F.3d 69 (2nd Cir. 2003)

tial side effects compiled by David White-side, a Hunt employee with no medicaltraining, and a Medical Guidelines policydeveloped by Michael Gray, a former RedLobster cashier with no medical trainingwho was, nevertheless, Hunt’s Medical Ad-visor, Hunt determined that certain appli-cants were unfit to be truck drivers. TheEEOC has provided substantial evidencethat Hunt believed that these individualswere unfit to drive a truck, or, for thatmatter, to drive at all and were incapableof performing the broad class of jobs thatfall under the classification ‘‘truck driving.’’Based upon this showing, I would vacatethe district court’s grant of summary judg-ment and hold that there is a genuinedispute of material fact with respect towhether the EEOC has established a pri-ma facie case of disability discrimination.I therefore respectfully dissent.

I agree with the majority that the issuein this appeal is whether the applicantswere denied truck driving positions atHunt because of their perceived disabilitywithin the meaning of the ADA. Ignoringsignificant evidence that Hunt perceivedthe applicants as more broadly limited,however, the majority holds that theEEOC has only provided evidence thatHunt perceived the rejected applicants as‘‘ineligible for a specific position withinHunt.’’ Ante at 78. In doing so, the major-ity reasons that long haul trucking is not asufficiently broad class of jobs such that asubstantial limitation on an individual’sability to be a long haul trucker wouldimply that the individual was disabledwithin the meaning of the ADA. See anteat 75–76. The majority asserts that alimitation on an individual’s ability to be along haul truck driver does not substan-tially limit his or her ability to engage inthe major life activity of working, as manyother truck driving jobs are available forthese individuals. See ante at 75–76. Themajority does not, however, hold that

truck driving in general is such a specificclass of jobs that a substantial limitationon truck driving would fail to imply adisability; its holding relies solely upon aninappropriately narrow view that Huntperceived the applicants as limited only intheir ability to work as long haul truckersfor Hunt.

Contrary to the majority’s assertion, theEEOC has produced significant evidencethat Hunt regarded the applicants as sub-stantially limited in the major life activityof working as truck drivers in general. Anemployer perceives an employee to be sub-stantially limited in his or her ability towork if it believes the employee is:

significantly restricted in the ability toperform either a class of jobs or a broadrange of jobs in various classes as com-pared to the average person having com-parable training, skills and abilities.The inability to perform a single, partic-ular job does not constitute a substantiallimitation in the major life activity ofworking.

29 C.F.R. § 1630.2(j)(3)(i); see also Bart-lett v. N.Y. State Bd. of Law Exam’rs, 226F.3d 69, 82–83 (2d Cir.2000). Factors thatmay be considered under this standardinclude the geographical area to which anindividual has reasonable access; the num-ber and types of jobs utilizing similartraining, knowledge, skills or abilities asthe job from which the applicant has beendisqualified; and the number and types ofjobs not utilizing similar training, knowl-edge, skills or abilities from which theapplicant will also be disqualified. 29C.F.R. § 1630.2(j)(3)(ii).

If other jobs utilizing an individual’sskills are available, that person is not sub-stantially limited in a class of jobs, even ifthis alternate employment would not allowthe individual to showcase his or her spe-cial talents. Sutton v. United Air Lines,

Page 365: Sotomayor Cases

80 321 FEDERAL REPORTER, 3d SERIES

Inc., 527 U.S. 471, 492, 119 S.Ct. 2139, 144L.Ed.2d 450 (1999). In Sutton, the Su-preme Court determined that plaintiffs ap-plying for positions as global airline pilotscould use their particular skills to obtainother piloting positions from which theywere not disqualified, so these plaintiffswere not regarded as being shut out froman entire occupational class. Id. at 492–93,119 S.Ct. 2139. In applying this rubric,the Second Circuit has found that practic-ing law is a broad occupational class, seeBartlett, 226 F.3d at 84, but that workingas a policeman is a specific position withinthe class of investigative or security jobs,see Giordano v. City of New York, 274F.3d 740, 749 (2d Cir.2001).

The EEOC has proffered evidence thatthe members of the plaintiff class haveundergone specialized driver training,earned commercial drivers’ licenses,passed road tests and received medicalcertifications pursuant to DOT regulations.The set of jobs that call for these qualifica-tions includes driving various types ofsmall and large trucks, including tractor-trailers, moving trucks, and cargo vans.See Office of Management & Budget, Stan-dard Occupational Classification Manual220 (2000), available at http://www.bls.gov/soc/soc v3d0.htm. The De-partment of Labor classifies truck drivingas a separate occupation within the overallcategory of ‘‘Transportation and MaterialMoving Occupations,’’ as does the Office ofManagement and Budget. Id. The De-partment of Labor estimated that in 2000there were more than 3.3 million jobs thatcame under the heading of ‘‘Truckdriverand Driver/Sales Workers.’’ 1 Bureau of

Labor Statistics, Occupational OutlookHandbook 577 (2002–03), available athttp://www.bls.gov/oco/pdf/ocos246.pdf.This evidence demonstrates that truckdriving is a general field of employmentrather than a specific position. AccordBaulos v. Roadway Express, Inc., 139F.3d 1147, 1154 (7th Cir.1998) (holdingthat driving a sleeper car is a specific jobwithin the class of truck drivers); Best v.Shell Oil Co., 107 F.3d 544, 548 (7th Cir.1997) (holding that truck driving is a classof jobs).

The majority does not reach the ques-tion whether truck driving is a class ofjobs. Instead, the majority argues thatHunt only dismissed the applicants be-cause ‘‘Hunt found the applicants unsuitedfor long-distance driving of Hunt’s 40–tontrucks on irregular, stressful schedules.’’Ante at 76. Such hyperbole is inapposite.Whether long haul trucking is, in fact,different from other types of truck drivingis not the central issue in this appeal;Hunt’s perception of the applicants as sub-stantially limited in their ability to drivetrucks, without further limitation to longhaul truck driving, is the central issue.

Beyond this basic misconception, themajority also misrepresents the record byasserting that the evidence ‘‘does not indi-cate that Hunt perceived the applicants asmore broadly limited.’’ Ante at 77. To thecontrary, the EEOC provided significantevidence that Hunt believed that the appli-cants were unfit to drive trucks. Numer-ous drugs were listed on the DRL as ‘‘NotPermitted,’’ 2 reflecting a belief that the

1. Driver/Sales Workers drive trucks and workas sales agents for the goods they haul; bothof these aspects are integral to their jobs. SeeOccupational Outlook Handbook 576–77(2002–03).

2. The majority discusses the categories ‘‘NotPermitted’’ and ‘‘Unsafe Effects’’ separately

from the two condition-based categories,‘‘Disqualifying Condition’’ and ‘‘Heart Condi-tion.’’ Ultimately, the majority finds no legaldistinction between the ‘‘condition’’ catego-ries and the others. See ante at 78.

Page 366: Sotomayor Cases

81E.E.O.C. v. J.B. HUNT TRANSPORT, INC.Cite as 321 F.3d 69 (2nd Cir. 2003)

applicant was prohibited by DOT regula-tions from driving a commercial vehiclewhile taking that particular medication.Dr. Cooper, Hunt’s physician consultant,testified with respect to one applicant thathe did not feel it was ‘‘in this patient’s bestinterest to pursue this profession.’’ Inter-view records show that the company be-lieved another applicant ‘‘would most likelyhave difficulty functioning in the lifestyleof a trucker.’’ Similarly, Dr. Cooper indi-cated with regard to another applicant thather ‘‘problems with sleep and concentra-tion under stress are not very compatiblewith the lifestyle expected of a driver.’’Applicant Joseph Lisa was told by a Hunt

employee that he would ‘‘never drive foranybody,’’ and numerous other applicantswere told that the medications they weretaking made it unsafe for them to drive atruck, or drive in general. See, e.g., CurtinDecl., Exh. 13 (reviewer told applicant‘‘that she could not be on [the medication]and drive [because] it can cause unsafeaffect [sic]’’); id. (reviewer told applicant‘‘that he could not drive[ ] while on thismedication’’); id. (reviewer told applicant‘‘that she cannot be on [the medication]and drive’’); Curtin Decl., Exh. 20 (drug is‘‘not permitted for driving’’); id. (‘‘[b]othdrugs are not approved for driving’’); id.(reviewer ‘‘informed applicant that he can-

I agree that all four categories of medi-cations on the DRL at issue here should betreated identically. I disagree, however,with the majority’s statement that ‘‘the basisfor the exclusion from employment was theuse of a listed drug, not any potential ‘dis-ability’ created by the treated disease.’’ Anteat 77–78. The EEOC has produced signifi-cant evidence that demonstrates the link be-tween the drug categories and potential un-derlying conditions. For example, one ofHunt’s interviewers noted that the ‘‘applicantdid not indicate the reason he is taking [themedication]. [N]eed to verify why he is onthis medication.’’ Similar comments weremade by reviewers with respect to applicantstaking medications in each of Hunt’s catego-ries. See, e.g., Curtin Decl., Exh. 18 (appli-cant needs to provide ‘‘headach [sic] releaseTTT [and a] statement that she is not taking[the medication] for depression’’) (‘‘Not Per-mitted’’ category); id. (‘‘Sent to Brenda forreview on cardiovasular [sic] condition’’)(‘‘Not Permitted’’ category); Curtin Decl.,Exh. 20 (‘‘the diagnosis and severity of hercondition for which she takes the medicationis considered disqualifying’’) (‘‘Unsafe Ef-fects’’ category); id. (applicant ‘‘will need tocomplete his treatments TTT and send in allrecords when his condition is resolved’’)(‘‘Unsafe Effects’’ category); Curtin Decl.,Exhs. 23, 25 (applicant needs to ‘‘provide astatment [sic] that TTT his condition is finew/out the meds’’) (‘‘Disqualifying Condition’’or ‘‘Heart Condition’’ category); id. (Hunt‘‘need[ed] all records on [applicant’s] condi-tion’’) (‘‘Disqualifying Condition’’ or ‘‘Heart

Condition’’ category). In addition, Hunt’sMedical Guidelines relating to mental andpsychological conditions required that in or-der to qualify for a job, an applicant takingmedication for depression must remain offthe medication for thirty days and submit aletter from a doctor stating that he or she nolonger suffers from the underlying condition.Contrary to the majority’s assertion, this evi-dence provides a direct causal link betweenthe applicants’ underlying conditions andHunt’s perception of the applicants as sub-stantially limited in their ability to work astruck drivers.

To make a further distinction that it ulti-mately finds insignificant, the majority assertsthat ‘‘[i]ndividuals suffering from the condi-tions treated with the ‘Heart Condition’ or‘Disqualifying Condition’ drugs are potentiallyexplicitly barred from truck driving by 49C.F.R. § 391.41.’’ Ante at 76–77 & n. 5 Manyof the conditions listed in this regulation onlydisqualify an individual if the condition is‘‘likely to interfere with his/her ability to con-trol and drive a commercial motor vehiclesafely,’’ 49 C.F.R. § 391.41(b)(5), or otherwisesuggest that an individualized determinationof potential safety concerns is required. Seeid. § 391.41(b)(6)-(12). In relying on theseregulations to support Hunt’s policy, the ma-jority ignores the crucial difference betweenindividualized determinations of driver safetyand Hunt’s explicit policy to create a per sebar from truck driving with respect to theseindividuals. Hunt’s policy simply assumes,without justification, that these individualsare unfit to drive trucks.

Page 367: Sotomayor Cases

82 321 FEDERAL REPORTER, 3d SERIES

not take [the medication] and drive’’);Curtin Decl., Exhs. 23, 25 (reviewer ‘‘in-formed appl[icant] that he could not takethe med[ication] on [the] truck’’); CurtinDecl., Adair Depo. (representative told ap-plicant ‘‘it’s illegal to drive a truck withthat [medication]’’); Curtin Decl., ManningDepo. (representative told applicant ‘‘itwas illegal for a driver to drive while onthis medication’’). The EEOC has provid-ed sufficient evidence to create a factualissue whether Hunt perceived the appli-cants as broadly limited in their ability towork as a truck driver.

The majority explains this evidence bystating:

Although a few evaluators’ commentscould be more broadly interpreted, thereis no evidence that Hunt’s reviewers,relying on Hunt’s own DRL and druglists to make a judgment on qualificationfor a position at Hunt, intended to makean evaluation beyond Hunt’s specificguidelines.

Ante at 76–77. In reviewing a grant ofsummary judgment, however, we do notrefuse to credit a broad, but reasonable,interpretation of the evidence. Giordano,274 F.3d at 749–50. Even if this were thestandard, Hunt’s reviewers stated that oneapplicant would ‘‘never drive for anybody,’’and made similar statements about manyother applicants; it is difficult to imagine aclearer statement that the reviewers in-tended to say that the applicants were, infact, substantially limited in their ability towork as a truck driver for any company.

The majority asserts that becauseHunt’s reviewers were not the ultimatedecision makers, the comments ‘‘simplyare not sufficient to indicate that Huntthought the applicants were more broadlylimited.’’ Ante at 76–77. Again, this is amatter for the factfinder to decide.Hunt’s own employees stated on severaloccasions that applicants were unfit to

drive; a factfinder reasonably could im-pute these statements to Hunt, even ifthese employees were not the ultimate de-cision makers. Hunt proffers no evidencethat these unidentified ‘‘ultimate hiring au-thorities’’ did not share the reviewers’ per-ceptions or rely upon their statementsabout the applicants’ limitations. Indeed,Hunt does not argue otherwise; it simplyargues that its employees’ statements im-plicitly refer only to jobs at Hunt. A fact-finder is certainly allowed to determinewhether the statement that an applicantwould ‘‘never drive for anybody’’ implicitlyrefers only to jobs at Hunt; it is not,however, this Court’s job to do so. Inreviewing whether summary judgment isappropriate, this Court does not make fac-tual determinations or refuse to credit le-gitimate inferences based upon the evi-dence presented, but views the evidence inthe light most favorable to the nonmovingparty. See Giordano, 274 F.3d at 746.

Hunt also argues that the statements ofDr. Cooper should not be imputed to it.The EEOC provides significant evidencethat Hunt relied on Dr. Cooper’s advice,including, for example, a reviewer’s state-ment that the applicant was ‘‘disqualifiedper Dr. Cooper.’’ This suffices to providea direct link between Dr. Cooper’s opin-ions regarding applicants and Hunt’s viewof the applicants as disabled.

Finally, the majority’s argument thatHunt’s policy should not be imputed toother companies in determining whetherthe applicants were perceived as disabledis immaterial. Contrary to the majority’sassertion, this is not a case in which thepotential imputation of Hunt’s policy toother companies would result in the appli-cants being regarded as ‘‘substantially lim-ited in the major life activity of workingonly as a result of this imputation.’’ Sut-ton, 527 U.S. at 493, 119 S.Ct. 2139. It isHunt’s explicit statement that it believed

Page 368: Sotomayor Cases

83CICIO v. DOESCite as 321 F.3d 83 (2nd Cir. 2003)

applicants to be unfit to drive a truck thatsupports Hunt’s perception of these indi-viduals as substantially limited in theirability to drive a truck; no potential impu-tation is required. Thus, the EEOC hasprovided sufficient evidence that a factfin-der could reasonably conclude that Huntregarded the rejected applicants as sub-stantially limited in the major life activityof working, because Hunt regarded themas unfit to be truck drivers.

CONCLUSION

Because I find ample support in therecord for the assertion that Hunt regard-ed the applicants as d substantially limitedin the major life activity of working, andthus, the applicants were disabled withinthe meaning of the ADA, I respectfullydissent.

,

Bonnie CICIO, individually and as Ad-ministratrix of the Estate of Car-

mine Cicio, Plaintiff–Appellant,

v.

John DOES 1–8, Defendants,

Vytra Healthcare, and Brent Spears,M.D., Defendants–Appellees.

Docket No. 01–9248.

United States Court of Appeals,Second Circuit.

Argued: June 20, 2002.

Decided: Feb. 11, 2003.

As Amended: March 12, 2003.

Widow, on behalf of herself and herlate husband’s estate, brought state court

suit against plan administrator for employ-ee benefits plan, its medical director, andothers, alleging state law claims arisingout of decision to deny preauthorization formedical procedure recommended by treat-ing physician. Defendants removed actionand moved to dismiss for failure to stateclaim. Widow moved to remand. The Unit-ed States District Court for the EasternDistrict of New York, Joanna Seybert, J.,208 F.Supp.2d 288, adopting the reportand recommendation of United StatesMagistrate Judge E. Thomas Boyle, grant-ed motion to dismiss on ground of preemp-tion under the Employee Retirement In-come Security Act (ERISA), and widowappealed. The Court of Appeals, Sack, Cir-cuit Judge, held that: (1) negligent delayand misrepresentation claims were remov-able under complete preemption doctrine;(2) court had supplemental jurisdictionover medical malpractice claim; (3) negli-gent delay and misrepresentation claimswere subject to dismissal as conflictpreempted; and (4) on issue of first im-pression, state law medical malpracticeclaim brought with respect to a medicaldecision made in the course of prospectiveutilization review by a managed care orga-nization or health insurer is not preemptedunder ERISA.

Affirmed in part, vacated in part, andremanded.

Calabresi, Circuit Judge, filed an opin-ion dissenting in part.

1. Removal of Cases O107(9)

District court’s denial of a motion toremand is reviewed de novo.

2. Federal Courts O776

District court’s decision to grant amotion to dismiss for failure to state aclaim is reviewed de novo. Fed.Rules Civ.Proc.Rule 12(b)(6), 28 U.S.C.A.

Page 369: Sotomayor Cases

UNITED STATES COURT OF APPEALSFOR THE SECOND CIRCUIT

August Term, 2000

(Argued: June 20, 2001 Decided: September 18, 2002)

Docket No. 00-9488

_________________________________________

VALERIE KRIMSTOCK, CHARLES FLATOW, ISMAEL DELAPAZ, CLARENCEWALTERS, JAMES WEBB, MICHAEL ZURLO, and SANDRA JONES, individually and onbehalf of all other persons similarly situated,

Plaintiffs-Appellants,

v.

RAYMOND W. KELLY, in his official capacity as Commissioner of the New York City PoliceDepartment, PROPERTY CLERK, NEW YORK CITY POLICE DEPARTMENT, and THECITY OF NEW YORK,

Defendants-Appellees.__________________________________________

Before: JACOBS, F.I. PARKER, and SOTOMAYOR, Circuit Judges.

Appeal from a judgment of the United States District Court for the Southern

District of New York (Michael B. Mukasey, Chief Judge) granting defendants’ motion to dismiss

the complaint, which alleges constitutional violations arising from the seizure of motor vehicles

by the police. We vacate, holding that plaintiffs’ due process rights were violated by the post-

seizure, pre-judgment retention of vehicles absent a prompt opportunity to challenge the probable

validity of and justification for that deprivation pendente lite.

Vacated and remanded.

THOMAS M. O’BRIEN, of counsel, The Legal Aid Society,Criminal Defense Division, New York, New York, forplaintiffs-appellants.

Page 370: Sotomayor Cases

1 The New York City Corporation Counsel has delegated to the Property Clerk of theNew York City Police Department the authority, under New York City’s Administrative Code, tobring the civil forfeiture actions at issue in this appeal. See Property Clerk v. Covell, 139 Misc.2d 707, 708-09, 528 N.Y.S.2d 299, 300 (Sup. Ct. N.Y. Co. 1988). We refer to defendantsCommissioner of the New York City Police Department, the Property Clerk of the New YorkCity Police Department, and the City of New York, collectively, as the “City.”

2

JULIE STEINER, Assistant Corporation Counsel, New York, NewYork, for Michael D. Hess, Corporation Counsel of the Cityof New York (Stephen J. McGrath, on the brief), fordefendants-appellees.

SOTOMAYOR, Circuit Judge:

Defendant City of New York seizes the motor vehicles of some of those accused

of driving while intoxicated and of committing other crimes for which a motor vehicle could be

considered an instrumentality.1 The City maintains possession of these vehicles in the hope of

one day gaining title to them by prevailing in civil forfeiture proceedings. These proceedings

generally await the resolution of criminal charges and can take months or even years to be

finalized. The City claims and plaintiffs-appellants (“plaintiffs”) do not contest that the

proceedings are authorized, as is the seizure, by provisions of the City’s Civil Administrative

Code that cede to the City title to property found to be an instrumentality of crime.

Plaintiffs contend that their inability to challenge, promptly after the vehicles are

seized, the legitimacy of and justification for the City’s retention of the vehicles prior to

judgment in any civil forfeiture proceeding violates their constitutional rights. We agree. A car

or truck is often central to a person’s livelihood or daily activities. An individual must be

permitted to challenge the City’s continued possession of his or her vehicle during the pendency

of legal proceedings where such possession may ultimately prove improper and where less

Page 371: Sotomayor Cases

3

drastic measures than deprivation pendente lite are available and appropriate.

We vacate the judgment of the district court and remand for the court to order,

after consultation with the parties, the appropriate injunctive relief.

BACKGROUND

Plaintiffs challenge the seizure and retention of motor vehicles under a section of

the City’s Civil Administrative Code, N.Y.C. Code § 14-140. The City claims and plaintiffs do

not contest that the statute authorizes the City’s Property Clerk to take custody, following

seizure, of, among other things, “all property . . . suspected of having been used as a means of

committing crime or employed in aid or furtherance of crime . . . .” N.Y.C. Code § 14-140(b).

Seized property is retained by the Property Clerk of the New York City Police Department until

the City either loses a future forfeiture suit or decides not to pursue one and someone claims the

seized property. Id. § 14-140(e); 38-A New York City Rules & Regulations (“R.C.N.Y.”) §

12-36. The relevant provision of the Administrative Code states:

Where . . . property . . . ha[s] been used as a means of committingcrime or employed in aid or in furtherance of crime . . . , a personwho so . . . used [or] employed . . . any such . . . property orpermitted or suffered the same to be used [or] employed . . . or whowas a participant or accomplice in any such act, or a person whoderives his or her claim in any manner from or through any suchperson, shall not be deemed to be the lawful claimant entitled to . . . such . . . property . . . .

N.Y.C. Code § 14-140(e)(1). The statute applies to all levels of crime, not just felonies, and to

all types of crimes. Moreover, it applies to all property, both real and personal.

Under the statute, the City can seize a motor vehicle following an arrest for the

state-law charge of driving while intoxicated (“DWI”) or any other crime for which the vehicle

Page 372: Sotomayor Cases

2 As noted below, the remedies available to those who have had their property seizedunder New York state civil forfeiture law (N.Y. C.P.L.R. art. 13-A) do not apply to seizuresunder N.Y.C. Code § 14-140. Other remedies suggested by the City, specifically a Request forJudicial Intervention and an Article 78 proceeding brought under New York state law, do notprovide a prompt retention hearing, as discussed below.

4

could serve as an instrumentality. The arraignment of the defendant in the criminal action

concerns only the prosecution of the criminal charge. A defendant charged with DWI does not

have a right to a post-arrest hearing to determine whether probable cause existed either for his or

her arrest or for the seizure of the vehicle. Unlike a felony charge, for which a “prompt”

probable cause hearing must be held or evidence of probable cause must be presented to a grand

jury, N.Y. Crim. Proc. § 180.10, a misdemeanor charge of DWI requires no post-arrest

determination of probable cause, id. §§ 170.10 et seq.; People v. Green, 96 N.Y.2d 195, 199-200,

726 N.Y.S.2d 357, 361 (2001) (noting, in a DWI case, that whereas the “New York Constitution

requires a Grand Jury indictment for felony offenses . . . , misdemeanor charges may be brought

on a prosecutor’s information”); In re Robert L.,129 Misc. 2d 742, 744, 493 N.Y.S.2d 970, 972

(Fam. Ct. Bronx Co. 1985) (“There is no provision for a preliminary or probable cause hearing

under the [New York] Criminal Procedure Law in misdemeanor cases.”). Moreover, N.Y.C.

Code § 14-140 affords the vehicle owner no opportunity for a prompt post-seizure hearing to test

probable cause for the vehicle’s seizure. Thus, neither New York criminal procedure nor the

City’s civil forfeiture law allows a DWI defendant or the owner of a vehicle driven by a DWI

defendant to challenge promptly the legitimacy of the City’s continued custody of the vehicle.2

That challenge may not be made until the City seeks the vehicle’s forfeiture in a separate civil

proceeding that could take place months or even years after the seizure.

Upon seizing the vehicle, the police issue the arrestee a voucher for the vehicle

Page 373: Sotomayor Cases

3 The applicable rules state that “[i]f a timely demand is made for the return of theproperty before the forfeiture proceeding is instituted, such proceeding shall be brought no laterthan . . . within 25 days after the date of demand. If such proceeding is not commenced withinthis time period, the property clerk shall give written notice to the claimant or the claimant’srepresentative, at his or her last known address . . . that the property will be returned forthwith tothat person.” 38-A R.C.N.Y. § 12-36(a). Absent a demand for the return of the property, theCity may commence forfeiture proceedings on its own initiative, often at the conclusion of thecriminal proceedings. See id. (“If such proceeding is instituted before the termination of criminalproceedings against the claimant, this subchapter shall not be construed to effect [sic] any right ofa party to the forfeiture proceeding to have the forfeiture proceeding stayed for such period as thecourt may determine.”).

4 This provision and several others found in the rules governing forfeitures under thepresent statute were adopted as a result of this Court’s determination that certain aspects of thepredecessor statute were unconstitutional. See McClendon v. Rosetti, 460 F.2d 111, 114-16 (2dCir. 1972) (striking down procedures for reclaiming property after termination of criminalproceedings charging crimes unrelated to seized property). As the district court below indicated,the challenge in the instant case presents claims not considered in our previous review of thestatute. Krimstock v. Safir, No. 99 Civ. 12041, 2000 WL 1702035, at *4 (S.D.N.Y. Nov. 13,2000).

5

and any other seized property. 38-A R.C.N.Y. § 12-32(a). If a claimant makes a formal demand

for the return of the vehicle, the City has twenty-five days in which either to initiate a civil

forfeiture proceeding under the City’s Administrative Code or to release the vehicle. Id. § 12-

36(a).3 Even if the City chooses to commence a civil forfeiture proceeding within the twenty-

five day period, however, the proceeding is commonly stayed until the criminal proceeding

concludes. In a forfeiture proceeding, the City “bear[s] the burden of proving by a preponderance

of the evidence that [it] is legally justified to continue to retain the property.” Id. § 12-36(b).4

Vehicles belonging to the named members of the putative class in this action

were seized by the City between March and May of 1999. The vehicles of six of the seven

named plaintiffs—Valerie Krimstock, Charles Flatow, Ismael Delapaz, Clarence Walters, James

Page 374: Sotomayor Cases

5 The state statute prohibits the operation of a motor vehicle if the driver “has .10 of oneper centum or more by weight of alcohol in the person’s blood.” N.Y. Veh. & Traf. § 1192(2). A violation is a misdemeanor punishable by a fine or by imprisonment for not more than oneyear, or both. Id. § 1193(1)(b).

6

Webb, and Michael Zurlo—were seized after each was arrested for DWI.5 In some cases, the

Property Clerk did not institute civil forfeiture actions against plaintiffs’ vehicles until well over

two months after the seizures. For example, Clarence Walters’ car was seized on March 15,

1999. He had never been arrested for any offense before, and his DWI case in criminal court

ended on June 1, 1999, when he pleaded guilty to the lesser charge of driving while impaired, a

non-criminal violation. N.Y. Veh. & Traf. § 1192(1). Mr. Walters paid a fine, performed

community service, and completed a Drinking Driver program that rendered him eligible for

restoration of his driver’s license. On June 4, more than two-and-a-half months after the seizure

of his vehicle, he was served with a forfeiture complaint. At no time between June 1999 and

May 2001, when his vehicle was finally released, was Mr. Walters given an opportunity to

challenge the City’s retention of the vehicle.

Each of the five other DWI arrestees also pleaded guilty to the lesser charge of

driving while impaired. Valerie Krimstock, who at the age of forty-eight had never been arrested

before, entered her plea to the lesser charge in September 1999—some four months after she had

been served with a forfeiture complaint. It was not until eleven months later, in August 2000,

that a judge dismissed the forfeiture action and ordered that her 1995 Toyota, on which she had

continued to make monthly payments of $273.00, be returned to her.

In the case of Charles Flatow—a retired sales manager whose car was seized on

April 3, 1999 in connection with a first-time DWI arrest—the Property Clerk commenced a

Page 375: Sotomayor Cases

6 Judge Jacobs does not subscribe to the four preceding paragraphs which recount andrely upon the circumstances particular to the named plaintiffs in this putative class action. Hisreasons are as follows: (1) The opinion implicitly generalizes from circumstances of theindividual plaintiffs—for example, that this one had never been arrested before, that this one is aretired person, and so forth. The named plaintiffs in a putative class action are frequentlyselected to be especially sympathetic and appealing, but the rule we make will also governvehicles seized from persons who are habitual drunks. (2) The fact that the six “DWI arrestees

7

forfeiture action on June 15, 1999. Mr. Flatow also pleaded guilty to the lesser charge, paid a

fine, and completed the required community service and Drinking Driver program. Yet by

December 1999, he still had received no hearing in the forfeiture action and his car remained in

police custody. As a result, Mr. Flatow had not been given an opportunity to present evidence

that a prescription anti-depressant medication he was taking at the time of the arrest caused the

Breathalyzer test to exaggerate the percentage of alcohol in his bloodstream.

To take yet another example, the 1995 Plymouth van owned by the seventh named

plaintiff, Sandra Jones, was seized in March 1999 when her estranged husband, to whom she had

lent the vehicle, was arrested for drug and weapon possession. Even though these charges were

later dismissed, the Property Clerk initiated a forfeiture action in May 1999, alleging that Ms.

Jones had “consented, suffered or permitted” her vehicle to be used by her husband in the

commission of crimes. During the pendency of the action, Ms. Jones continued to make monthly

payments of $428.06 on her impounded vehicle. In January 2000, some ten months after the

police had seized it, the Property Clerk returned her van. In sum, at the time this suit was filed in

December 1999, no court had yet ruled on the legitimacy of the retention of any of plaintiffs’

vehicles, all of which remained in the possession of the police even though the criminal cases

underlying the forfeiture actions had concluded and none had resulted in a conviction of a crime

that would serve as a predicate for forfeiture.6

Page 376: Sotomayor Cases

. . . pleaded guilty to the lesser charge of driving while impaired” does not remotely call intoquestion the arresting officer’s probable cause to arrest them for drunk driving. It goes withoutsaying that plea bargains in misdemeanor cases are offered for many reasons other than factualinnocence or even insufficiency of evidence.

7 Even though plaintiffs invoked Fourteenth Amendment protections in their complaint,their claims involve both the City’s probable cause to believe their vehicles were subject toseizure and forfeiture (a Fourth Amendment concern, as applied through the FourteenthAmendment) and the City’s justification for retaining their vehicles pendente lite (a FourteenthAmendment concern). We address both issues.

8 The court did abstain from hearing plaintiffs’ claim that they had a due process right to ameaningful opportunity to request court-appointed counsel. Krimstock, 2000 WL 1702035, at*3. Plaintiffs have not raised this issue on appeal, and we therefore deem it abandoned.

8

Plaintiffs brought this action under 42 U.S.C. § 1983, alleging violations of the

Due Process Clause of the Fourteenth Amendment and seeking a prompt hearing following the

seizure of vehicles, at which the City “must demonstrate probable cause that the car was used in

furtherance of a crime and that it is necessary that the vehicle remain in the City’s custody until

the conclusion of the forfeiture proceeding.”7 Named plaintiffs moved for class certification

pursuant to Fed. R. Civ. P. 23(a) and (b)(2) and for a preliminary injunction under Fed. R. Civ. P.

65. The City cross-moved to dismiss the complaint pursuant to Fed. R. Civ. P. 12(b)(6).

The district court granted the City’s motion to dismiss pursuant to Fed. R. Civ. P.

12(b)(6). Krimstock, 2000 WL 1702035, at *7. The court rejected the City’s request to dismiss

plaintiffs’ constitutional claims on abstention grounds, finding that the civil forfeiture

proceedings under the New York City Administrative Code did not provide an adequate forum

for raising these claims. Id. at *3.8 Applying the balancing test set forth in Mathews v. Eldridge,

424 U.S. 319 (1976), to determine what procedural safeguards are required to satisfy the Due

Process Clause of the Fourteenth Amendment in the circumstances of this case, the district court

Page 377: Sotomayor Cases

9

held that plaintiffs were not entitled to prompt post-seizure hearings on the question of probable

cause or the legitimacy of retention pendente lite. Id. at *6-*7. Specifically, the court held that

“plaintiffs’ due process right to a meaningful hearing at a meaningful time does not require the

additional safeguard of a probable cause hearing.” Id. at *7. The court concluded that the

plaintiffs’ interests were adequately protected by “a probable cause arrest” and the eventual

forfeiture proceeding. Id. at *6-*7. This appeal followed.

DISCUSSION

A federal court may dismiss a complaint for “failure to state a claim upon which

relief can be granted.” Fed. R. Civ. P. 12(b)(6). “We review de novo a district court’s dismissal

of a complaint pursuant to Rule 12(b)(6), accepting all factual allegations in the complaint as true

and drawing all reasonable inferences in the plaintiff’s favor.” Ganino v. Citizens Utils. Co., 228

F.3d 154, 161 (2d Cir. 2000). Under Rule 12(b)(6), we uphold a district court’s dismissal only if

“it appears beyond doubt that the plaintiffs can prove no set of facts in support of his [or her]

claim which would entitle him [or her] to relief.” Id. (internal citation omitted).

Our primary focus today is the City’s continued retention of vehicles after their

warrantless seizure by the police and prior to the ultimate resolution of the forfeiture action in

court. It is this intermediate deprivation, lasting months or sometimes years without any prompt

hearing before a neutral fact-finder, that we deem constitutionally infirm. In the absence of a

showing that continued impoundment constitutes a valid deprivation, seized vehicles must be

released during the pendency of civil proceedings.

We reach this conclusion in light of the dictates of the Fourth and Fourteenth

Amendments. In Part I, we establish a framework for analyzing plaintiffs’ challenge to the

Page 378: Sotomayor Cases

9 Here and throughout this opinion, we use the phrase “innocent owner” as a term of artdenoting a person who has an ownership interest in property threatened with civil forfeiture butwho neither participated in nor permitted or suffered the alleged illegal use of the property, andpersons who claim that status. Our use of the term is not intended to suggest that drivers or otherpersons initially charged with a vehicle-related crime are not presumed innocent until provenguilty in a criminal proceeding or are presumptively unlikely to prevail in a civil forfeitureproceeding.

10

probable validity of the City’s post-seizure, pre-judgment retention of their vehicles, and we

define “probable validity” as a due process concept that in the present case embraces the City’s

probable cause for initially seizing the vehicles. In Parts II and III, we discuss the applicability of

the Fourth and Fourteenth Amendments to seizure and retention of personal property under civil

forfeiture laws. In the course of discussing the Fourteenth Amendment, we give special attention

to three areas of due process concern raised by N.Y.C. Code § 14-140 as applied to the present

facts: the temporal gap that typically exists between vehicle seizure and the eventual forfeiture

proceeding; the plight of innocent owners;9 and the inadequacy of the remedies suggested by the

City for addressing the propriety of continued retention of vehicles seized under § 14-140. In

Part IV, we employ the three-factor inquiry prescribed by the Supreme Court in Mathews v.

Eldridge to determine what procedural safeguards are required to satisfy the Due Process Clause

in this case. Finding that plaintiffs’ right to due process has been violated, we conclude in Part V

by offering general guidance as to the prompt post-seizure retention hearing that we deem to be

constitutionally required.

I. The Probable Validity of Continued Deprivation of Vehicles

Plaintiffs in this action essentially seek an early opportunity to test the City’s

likelihood of success on the merits of the forfeiture action, or what the Supreme Court has

termed the “probable validity” of continued deprivation of a claimant’s property during the

Page 379: Sotomayor Cases

11

pendency of legal proceedings. Cf. Comm’r v. Shapiro, 424 U.S. 614, 629 (1976) (“[A]t least

where irreparable injury may result from a deprivation of property pending final adjudication of

the rights of the parties, the Due Process Clause requires that the party whose property is taken be

given an opportunity for some kind of predeprivation or prompt post-deprivation hearing at

which some showing of the probable validity of the deprivation must be made.”); Fuentes v.

Shevin, 407 U.S. 67, 97 (1972) (“Since the essential reason for the requirement of a prior hearing

is to prevent unfair and mistaken deprivations of property, . . . it is axiomatic that the hearing

must provide a real test. ‘(D)ue process is afforded only by the kinds of “notice” and “hearing”

that are aimed at establishing the validity, or at least the probable validity, of the underlying

claim against the alleged debtor before he can be deprived of his property.’”) (quoting Sniadach

v. Family Fin. Corp., 395 U.S. 337, 343 (1969) (Harlan, J., concurring)).

For present purposes, we treat probable validity as a comprehensive due process

concept that includes the City’s probable cause for initially seizing vehicles. Clearly, the legality

of a warrantless seizure is a component of the larger question of the probable validity of

continued retention of a seized vehicle. If a seizure lacked probable cause, and the City could

offer no untainted post-seizure evidence to justify further retention, the claimant’s vehicle would

ordinarily have to be released during the pendency of proceedings. Cf. Marine Midland Bank,

N.A. v. United States, 11 F.3d 1119, 1125 (2d Cir. 1993) (“[C]ourts in this circuit have ordered

the return of seized property before the commencement of a [federal] forfeiture trial on the

ground that the government lacked probable cause to seize the property at the time of the

seizure.”).

Although there is an obvious overlap between probable cause for a seizure and the

Page 380: Sotomayor Cases

12

probable validity of a retention, the two are not necessarily coextensive. For example, at a

retention hearing, the City might succeed in showing that police officers had probable cause for

seizing the vehicle of a DWI arrestee, yet be unable to establish the probable validity of

continued deprivation pendente lite in the face of proof of innocent ownership or evidence that

the Breathalyzer test had registered inaccurate results. Similarly, the City might establish

probable cause for a seizure but fail to persuade the court that its interest in the accused

instrumentality would not be protected by measures less drastic than continued deprivation.

Conversely, the City might fail to establish probable cause for an initial seizure yet be able to

offer post-seizure evidence showing the probable validity of retention during the pendency of

proceedings. Thus, the legality of a seizure typically will be a subset of the larger due process

question of the legitimacy of continued impoundment pendente lite.

II. The Role of the Fourth Amendment in Civil Forfeiture

The Supreme Court has held that the Fourth Amendment protects claimants

against unreasonable seizures of their property in the civil forfeiture context. See United States v.

James Daniel Good Real Prop., 510 U.S. 43, 49 (1993) (“The Fourth Amendment does place

restrictions on seizures conducted for purposes of civil forfeiture . . . .”); see also One 1958

Plymouth Sedan v. Pennsylvania, 380 U.S. 693, 696, 700 (1965) (holding that the exclusionary

rule under the Fourth Amendment applies to civil forfeiture proceedings); cf. In re Seizure of All

Funds in Accounts in Names Registry Publ’g, Inc., 68 F.3d 577, 580 (2d Cir. 1995) (“In order to

seize property under [federal civil forfeiture law], the government must demonstrate that there

was probable cause to believe that the property is subject to forfeiture.”); United States v.

Daccarett, 6 F.3d 37, 49 (2d Cir. 1993) (“[T]his circuit requires seizures made pursuant to

Page 381: Sotomayor Cases

10 Marine Midland Bank involved the seizure of interbank funds under the federal civilforfeiture laws and upon a warrant issued by a federal magistrate judge, who had determined thatthe government had probable cause to believe that interbank accounts were being used to laundernarcotics proceeds. Marine Midland Bank, 11 F.3d at 1121. The banks moved for return of theseized funds by order to show cause pursuant to Fed. R. Crim. P. 41(e). Holding that “[t]hemagistrate judge’s probable cause determination should be subject to judicial review,” this Courtfound that the warrant had not been supported by probable cause and therefore ordered therelease of certain of the interbank funds. Id. at 1125.

13

[federal civil forfeiture law] to comport with the fourth amendment.”); United States v. $37,780

in U.S. Currency, 920 F.2d 159, 162 (2d Cir. 1990) (same).

The Supreme Court has not said that a probable cause hearing is required after a

warrantless seizure of property and before trial of a government’s claim to title under a civil

forfeiture law. Yet many of our laws are fashioned to ensure that a prompt hearing before a

neutral judicial or administrative officer is held after the warrantless seizure of property and

before full adjudication of the merits of a claim. One example is the federal civil forfeiture

regime as administered by the courts of this Circuit in the context of seizures of illicit proceeds

and instrumentalities. As this Court has recognized, “the seizure and forfeiture of property are

two distinct events under the [federal] civil forfeiture laws.” Marine Midland Bank, 11 F.3d at

1124 (citing Daccarett, 6 F.3d 37 at 46).10 “While both events require the government to have

probable cause, the government is not required to demonstrate probable cause until the forfeiture

trial unless a claimant challenges the validity of the seizure before trial.” Id. (emphasis added).

If the government, once challenged, cannot establish probable cause for the initial seizure or offer

post-seizure evidence to justify continued impoundment, retention of the seized property runs

afoul of the Fourth Amendment. Id. at 1125; see also United States v. U.S. Currency in Amount

of One Hundred Forty-Six Thousand, Eight Hundred Dollars, No. 96-CV-4882, 1997 WL

Page 382: Sotomayor Cases

11 See People v. Earley, 244 A.D.2d 769, 770-71, 666 N.Y.S.2d 223, 224 (3d Dep’t 1997)(holding that arresting officer’s testimony at suppression hearing supported finding of probablecause to arrest defendant for DWI, despite contradictory testimony of defendant and hispassenger and despite minor inconsistencies in officer’s testimony); Grinberg v. Safir, 181 Misc.2d 444, 455, 694 N.Y.S.2d 316, 325 (Sup. Ct. N.Y. Co.) (“The seizure is simultaneous with aDWI arrest for which the police must have probable cause. The arresting officer evaluates anoffense committed in his or her presence. Indicia of alcohol consumption and objective testsconfirming the presence of alcohol minimize the risk of erroneous deprivation.”), aff’d, 266A.D.2d 43, 698 N.Y.S.2d 218 (1st Dep’t 1999); see also People v. Mathison, 287 A.D.2d 384,384, 732 N.Y.S.2d 2, 2-3 (1st Dep’t 2001) (noting that probable cause may be demonstrated byreliable hearsay, including information provided by fellow officer present at the arrest scene).

14

269583, at *3 (E.D.N.Y. Apr. 28, 1997) (“[I]f the claimant [in a federal civil forfeiture action]

properly raises the issue of the government’s probable cause for seizure before the forfeiture trial,

and if the claimant demonstrates that the government lacked probable cause at the time of

seizure, the property may be returned to the claimant until the forfeiture trial is held.”).

We recognize that the likelihood of illegal seizure is reduced in the context of

DWI arrests and that the City’s burden of proving probable cause in such cases is not onerous.11

We cannot agree with the district court, however, that a warrantless arrest is sufficient by itself to

ensure the legality of the initial seizure. Some risk of erroneous seizure exists in all cases, and in

the absence of prompt review by a neutral fact-finder, we are left with grave Fourth Amendment

concerns as to the adequacy of an inquiry into probable cause that must wait months or

sometimes years before a civil forfeiture proceeding takes place. Our concerns are heightened by

the fact that the seizing authority in this case “has a direct pecuniary interest in the outcome of

the proceeding.” James Daniel Good Real Prop., 510 U.S. at 55-56; see also Property Clerk v.

Hyne, 147 Misc. 2d 774, 780, 557 N.Y.S.2d 244, 248 (Sup. Ct. N.Y. Co. 1990) (noting that § 14-

140’s “remedial purposes” include the fact that “revenue is generated and applied toward the cost

of law enforcement”), aff’d, 171 A.D.2d 506, 567 N.Y.S.2d 603 (1st Dep’t 1991).

Page 383: Sotomayor Cases

15

III. The Role of the Fourteenth Amendment in Civil Forfeiture

The government’s seizure and retention of property under civil forfeiture statutes,

in the absence of a meaningful hearing at a meaningful time, raise serious due process concerns

under the Fourteenth Amendment. See James Daniel Good Real Prop., 510 U.S. at 62 (holding

that, absent exigent circumstances, “the Due Process Clause requires the Government to afford

notice and a meaningful opportunity to be heard before seizing real property subject to civil

forfeiture”); Fuentes, 407 U.S. at 80 (holding, in a case involving state prejudgment replevin

statutes that permitted seizure of chattels without a prior opportunity to be heard, that it is

“fundamental that the right to notice and an opportunity to be heard must be granted at a

meaningful time and in a meaningful manner”) (quotation marks omitted).

The fundamental right to notice and a meaningful hearing at a meaningful time

has been recognized in many different contexts. See, e.g., James Daniel Good Real Prop., 510

U.S. at 43 (seizure of real property under federal forfeiture law); Connecticut v. Doehr, 501 U.S.

1 (1991) (state ex parte attachment procedures); Memphis Light, Gas & Water Div. v. Craft, 436

U.S. 1 (1978) (termination of municipal utility service); N. Ga. Finishing, Inc. v. Di-Chem, Inc.,

419 U.S. 601 (1975) (prejudgment garnishment of bank account); Fuentes, 407 U.S. at 67 (state

prejudgment replevin statutes); Sniadach, 395 U.S. at 337 (state wage-garnishment procedure).

Due process is inevitably a fact-intensive inquiry. Doehr, 501 U.S. at 10 (“[D]ue process, unlike

some legal rules, is not a technical conception with a fixed content unrelated to time, place and

circumstances.”) (quotation marks omitted). The “timing and nature of the required hearing will

depend on appropriate accommodation of the competing interests involved.” Logan v.

Zimmerman Brush Co., 455 U.S. 422, 434 (1982) (quotation marks omitted).

Page 384: Sotomayor Cases

16

A. Temporary Deprivations of Property Pendente Lite

Temporary deprivation of real or personal property pendente lite in a forfeiture

action must satisfy the demands of the Fourteenth Amendment. The Supreme Court has held that

even a brief and provisional deprivation of property pending judgment is of constitutional

importance. See Fuentes, 407 U.S. at 84-85 (“[I]t is now well settled that a temporary, nonfinal

deprivation of property is nonetheless a ‘deprivation’ in the terms of the Fourteenth

Amendment.”); see also United States v. Monsanto, 924 F.2d 1186, 1192 (2d Cir. 1991) (noting

that a “temporary and nonfinal” removal of a defendant’s assets, pursuant to a federal criminal

forfeiture statute and pending resolution of the criminal case, “is, nonetheless, a deprivation of

property subject to the constraints of due process”) (quotation marks omitted). Plaintiffs here

have not challenged the procedural safeguards under New York law that guarantee the accuracy

of any final judgment of forfeiture. Instead, they question the legitimacy of and justification for

the intermediate deprivation of their property occasioned after seizure of the vehicle and before

judgment in civil forfeiture proceedings under N.Y.C. Code § 14-140, and, indeed, before those

proceedings are even commenced. See James Daniel Good Real Prop., 510 U.S. at 56 (“The

question in the civil forfeiture context is whether ex parte seizure is justified by a pressing need

for prompt action.”); Fuentes, 407 U.S. at 80-81 (stating that due process is intended “to

minimize substantively unfair or mistaken deprivations of property”).

The district court in this case collapsed the separate issues of probable cause and

due process into a single analysis and, applying the test for due process set forth in Mathews v.

Eldridge, 424 U.S. 319 (1976), concluded that plaintiffs had alleged no facts to suggest that a

“probable cause arrest” is a “procedure” that is “unusually unreliable,” Krimstock, 2000 WL

Page 385: Sotomayor Cases

12 The district court also cited United States v. Von Neumann, 474 U.S. 242, 249 (1986),in declaring that “‘the forfeiture proceeding, without more, provides the postseizure hearingrequired by due process.’” Krimstock, 2000 WL 1702035, at *5. The Supreme Court in VonNeumann, however, was addressing the different issue of what process was due in proceedingsfor remission or mitigation under U.S. customs laws when a claimant could challenge the seizureof his or her property in judicial forfeiture proceedings. Von Neumann, 474 U.S. at 249-50. Wenote that under the customs laws applicable in Von Neumann, the claimant could file a motionunder Fed. R. Crim. P. 41(e) for return of the seized vehicle if he or she “believe[d] the initialseizure was improper,” id. at 244 n.3, and that, in the case of Von Neumann, Customs releasedthe claimant’s vehicle after he had posted a bond pursuant to 19 U.S.C. § 1614, id. at 246.

17

1702035, at *6, and further concluded that “plaintiffs’ due process right to a meaningful hearing

at a meaningful time does not require the additional safeguard of a probable cause hearing,” id. at

*7. In reaching this determination, the court applied the “speedy trial” test as deployed in the

federal customs case of United States v. $8,850, 461 U.S. 555 (1983), and held that plaintiffs’

due process interests are fully protected by the eventual forfeiture proceeding. Id.12

The district court’s analysis resembles the approach taken by the New York

Supreme Court in Grinberg v. Safir, in which a DWI arrestee brought an Article 78 proceeding to

contest the City’s seizure of his 1988 Acura for forfeiture. In response to Grinberg’s Fourth

Amendment challenge to the seizure and retention of his vehicle, the court, citing various

warrantless arrest and seizure exceptions, held that “[o]nce an object is permissibly seized as an

instrumentality during an arrest, no warrant, pretrial hearing or judicial approval is needed for

retention during the criminal action.” Grinberg, 181 Misc. 2d at 452, 694 N.Y.S.2d at 323. The

court also found that Grinberg’s Fourteenth Amendment right to a meaningful hearing at a

meaningful time had not been violated. Citing the speedy trial test as applied in $8,850, the court

reasoned that “[i]f pendency of a criminal action is a legitimate reason for the delayed filing of a

forfeiture proceeding, then retention of the subject vehicle without a hearing, while the criminal

Page 386: Sotomayor Cases

18

action is pending, is also permissible.” Grinberg, 181 Misc. 2d at 456, 694 N.Y.S.2d at 326.

For reasons discussed more fully below, we disagree with these courts’

conclusions. Contrary to the district court’s determination in the present case, a warrantless

arrest by itself does not constitute an adequate, neutral “procedure” for testing the City’s

justification for continued and often lengthy detention of a vehicle which may be owned by the

arrestee or by someone entirely unconnected with the conduct that gave rise to the arrest.

Further, to say that the forfeiture proceeding, which often occurs more than a year after a

vehicle’s seizure, represents a meaningful opportunity to be heard at a meaningful time on the

issue of continued impoundment is to stretch the sense of that venerable phrase to the breaking

point. We also consider it a non sequitur to hold, as the Grinberg court did, that because

postponing the commencement of a forfeiture action pending the underlying criminal proceeding

may not offend due process, retention of the seized vehicle without a hearing throughout that

same period, or longer, is constitutionally permissible. The issues of a speedy trial and a prompt

retention hearing are not parallel in this context, particularly when less restrictive methods for

protecting the City’s interest in the allegedly offending res are available. Cf. Lee v. Thornton,

538 F.2d 27, 32 (2d Cir. 1976) (“Deprivation of means of transportation for [substantial] periods

requires an opportunity to be heard.”); DeBellis v. Property Clerk, 79 N.Y.2d 49, 57, 580

N.Y.S.2d 157, 161 (1992) (“The core principle of the Second Circuit’s McClendon decision is

that, although the government may seize and hold a citizen’s property for a variety of reasons in

connection with a criminal or related proceeding, once those proceedings have terminated or it is

determined that the property is not related to or is otherwise not needed for those proceedings,

due process requires that the property be returned upon demand unless the government can

Page 387: Sotomayor Cases

19

establish a new basis for its detention.”).

In sum, just as in the attachment and seizure cases cited above, the purpose of

requiring due process in the present circumstances “is not only to ensure abstract fair play to the

individual,” but “more particularly, . . . to protect his [or her] use and possession of property

from arbitrary encroachment—to minimize substantively unfair or mistaken deprivations of

property.” James Daniel Good Real Prop., 510 U.S. at 53 (quotation marks and citation

omitted). We conclude that plaintiffs have a right under the Fourteenth Amendment to ask what

“justification” the City has for retention of their vehicles during the pendency of proceedings, cf.

id. at 56, 61, and to put that question to the City at an early point after seizure in order to

minimize any arbitrary or mistaken encroachment upon plaintiffs’ use and possession of their

property.

B. Special Due Process Concerns in the Present Case

Our concern that plaintiffs be provided an early opportunity to test the propriety of

the City’s retention of their vehicles, after seizure pursuant to N.Y.C. Code § 14-140 and prior to

any eventual civil forfeiture judgment, is heightened by several factors. These factors include the

temporal gap that typically exists between seizure of the vehicle and the forfeiture proceeding,

the inability of innocent owners to challenge promptly the City’s retention of their vehicles

pendente lite, and the inadequacy of other suggested remedies for providing prompt post-seizure

review under New York’s administrative and civil codes.

1. The Temporal Gap Between Seizure and Forfeiture Proceedings

N.Y.C. Code § 14-140 and the applicable rules leave a significant temporal gap

between the moment a vehicle is seized and the time the City commences forfeiture proceedings.

Page 388: Sotomayor Cases

13 The 25-day rule under 38-A R.C.N.Y. § 12-36(a) represents a legislative expansion ofthe time period originally ordered in McClendon v. Rosetti, 369 F. Supp. 1391, 1394 (S.D.N.Y.1974), which required the Property Clerk to initiate a forfeiture proceeding within 10 days after aclaimant’s timely demand.

14 In contrast, while the statute here requires only that a forfeiture proceeding be initiatedwithin 25 days after a claim is made, the California controlled substances act requires that, once aperson claiming an interest in seized property files a verified claim, “the [civil] forfeitureproceeding shall be set for hearing on a day not less than 30 days therefrom, and the proceedingshall have priority over other civil cases.” Cal. Health & Safety Code § 11488.5(c)(1) (emphasisadded). Similarly, the civil forfeiture provision of the Wisconsin controlled substances act statesthat “the action shall be set for hearing within 60 days of the service of the answer but may becontinued for cause or upon stipulation of the parties.” Wis. Stat. § 961.555(2)(b); see also Statev. Rosen, 240 N.W.2d 168, 171-72 (Wis. 1976) (holding that the 60-day rule under theWisconsin controlled substances act is mandatory and that failure to set the hearing within thatperiod deprived the trial court of jurisdiction). In South Carolina, when a motorist is convictedof a third or subsequent DUI violation within 10 years, the vehicle may be seized and forfeited,but the vehicle’s registered owner, who must be given notice within 72 hours of the confiscation,has 10 days to request a judicial hearing to determine issues concerning the owner’s knowledgeand authorization. “The hearing must be held within ten days from the date of receipt of therequest.” S.C. Code Ann. § 56-5-6240(A).

20

Although it is possible for the City to initiate a forfeiture action earlier, it need not bring such an

action until twenty-five days after a claim is made for the vehicle. 38-A R.C.N.Y. § 12-36(a) (“If

a timely demand is made for the return of the property before the forfeiture proceeding is

instituted, such proceeding shall be brought no later than . . . within 25 days after the date of

demand.”).13 If no demand is made, the Property Clerk may initiate the action at its discretion.

In the present case, forfeiture proceedings were commenced, at the earliest, three weeks after

seizure of a vehicle, and, at the latest, well over two months after seizure. Thus, there typically

exists a significant period after seizure and before the filing of the forfeiture action when the City

is not held responsible for the legality of the warrantless seizure or the continued retention of the

vehicle. The period between the seizure and the holding of a hearing in the forfeiture action is,

of course, considerably longer.14 It can take months or even years.

Page 389: Sotomayor Cases

Some legislatures have sought to mitigate the depreciation of property values and othercosts incurred through delays of civil forfeiture proceedings. See, e.g., Ariz. Rev. Stat. § 13-4310(I) (“Before staying civil discovery [in a forfeiture proceeding until a related criminal trial isconcluded], the court shall make adequate provision to prevent any loss or expense to any victimor party resulting from the delay, including loss or expense due to maintenance, management,insurance, storage or preservation of the availability of the property or due to depreciation in thevalue of the property.”).

21

Many state forfeiture statutes, unlike N.Y.C. Code § 14-140, provide an early

opportunity to challenge the governmental authority’s probable cause for seizing property or the

legitimacy of its retaining seized property during the pendency of proceedings. Florida’s

contraband forfeiture statute is one example. In upholding the Florida statute in a case involving

police seizure of a vehicle from a public place, the U.S. Supreme Court observed that, although

the police had not needed to obtain a warrant to seize the vehicle, the statute required that “‘the

person entitled to notice is notified at the time of the seizure . . . that there is a right to an

adversarial preliminary hearing after the seizure to determine whether probable cause exists to

believe that such property has been or is being used in violation of the Florida Contraband

Forfeiture Act.’” Florida v. White, 526 U.S. 559, 562 n.2 (1999) (quoting Fla. Stat.

§ 932.703(2)(a)). The Florida statute further provides:

Seizing agencies shall make a diligent effort to notify the personentitled to notice of the seizure. Notice provided by certified mailmust be mailed within 5 working days after the seizure and muststate that a person entitled to notice may request an adversarialpreliminary hearing within 15 days after receiving such notice. . . .The seizing agency shall set and notice the hearing, which must beheld within 10 days after the request is received or as soon aspracticable thereafter.

Fla. Stat. § 932.703(2)(a); see also Cochran v. Harris, 654 So. 2d 969, 972 (Fla. Dist. Ct. App.

1995) (holding that a delay of twenty-three days beyond the ten-day limit for a hearing under

Page 390: Sotomayor Cases

22

§ 932.703 violated the claimants’ right to due process); cf. Ariz. Rev. Stat. § 13-4310(B)

(providing that, upon timely application by an owner of or interest holder in property threatened

with forfeiture, the court “may issue an order to show cause to the seizing agency for a hearing

on the sole issue of whether probable cause for forfeiture of the property then exists”); Cal.

Health & Safety Code § 11488.4(h) (providing that “[i]f there is an underlying or related criminal

action, a defendant may move for the return of the property [threatened with civil forfeiture] on

the grounds that there is not probable cause to believe that the property is forfeitable . . . .”).

Nothing like the procedural safeguards contained in the Florida contraband forfeiture act and

similar state statutes is built into the New York forfeiture law.

In addition, many state statutes afford avenues of interim relief for claimants who

are adversely affected by seizure and retention of property. For example, the Florida contraband

forfeiture statute provides that if the court determines that probable cause exists to seize property,

“the court shall order the property restrained by the least restrictive means to protect against

disposal, waste, or continued illegal use of such property pending disposition of the forfeiture

proceeding.” Fla. Stat. § 932.703(2)(d). These means include “a bond or other adequate security

equivalent to the value of the property.” Id.; cf. Ariz. Rev. Stat. § 13-4306(G) (“An owner of

property seized for forfeiture may obtain the release of the seized property by posting . . . a surety

bond or cash . . . .”); Cal. Health & Safety Code § 11492(c) (providing for various remedies to

preserve the status quo pendente lite, including “a surety bond or undertaking to preserve the

property interests of the interested parties”); N.J. Stat. Ann. § 2C:64-3(g) (providing that persons

with an interest in property seized for forfeiture, except defendants prosecuted in connection with

the seized property, may, after posting a bond, secure release of the property pending the

Page 391: Sotomayor Cases

15 In Bennis v. Michigan, 516 U.S. 442 (1996), the Supreme Court held that a statenuisance-abatement law under which authorities sought the forfeiture of a vehicle used for illegalsexual activity did not violate the Due Process Clause of the Fourteenth Amendment by notpermitting an innocent co-owner of the vehicle to contest the abatement. Id. at 453. Unlike thefederal forfeiture law at issue in James Daniel Good Real Property, however, the statute inBennis did not provide for an affirmative defense of innocent ownership. Moreover, the Court inBennis suggested that the apparent unfairness of the Michigan law was mitigated by the trialcourt’s “remedial discretion” to consider whether forfeiture of a vehicle would leave its owner orowners without transportation and to order payment of one-half of any sale proceeds to aninnocent co-owner. Id. at 444-45, 453. We also note that the nuisance-abatement statute inquestion, Mich. Comp. Laws § 600.3825, authorized “removal” and “sale” of property uponentry of an order of abatement only after the property had been deemed a nuisance in a civilaction. Id. at 444 n.3. Nothing on the face of the provision, or in the Bennis case, suggests thatthe statute permitted seizure and retention of property prior to adjudication of its status as anuisance.

23

forfeiture action). Again, no protections for a claimant’s practical interests in seized property are

provided for under the New York forfeiture law.

2. Seizure of Property of Innocent Owners

With respect to innocent owners, the City’s authority to seize property may be

broader than its authority to cause the forfeiture of the property. In the due process context, the

Supreme Court has shown special concern for the risk of erroneous deprivation posed to innocent

owners. See James Daniel Good Real Prop., 510 U.S. at 55.15 The impact of N.Y.C. Code § 14-

140 on innocent owners is vividly illustrated by the predicament of plaintiff Sandra Jones, whose

Plymouth van, which she had lent to her estranged husband, was seized in connection with her

husband’s arrest on drug and weapon charges. Although these charges were later dismissed, Ms.

Jones was deprived of her vehicle for some ten months while continuing to make monthly auto

payments on the vehicle. Ms. Jones was given no early opportunity to test the probable validity

of the City’s continued impoundment of her vehicle.

The forfeiture provision operates against those persons who “shall not be deemed

Page 392: Sotomayor Cases

16 This cumbersome construction, defining those subject to forfeiture proceedings interms of who is not a “lawful claimant” to the property, appears to reflect the former practiceunder which the burden rested on a claimant to bring a civil action in replevin to obtain the returnof seized property. See Hofferman v. Simmons, 290 N.Y. 449, 454 (1943) (“Since these arereplevin actions, we concern ourselves not so much with the [City’s] right to hold as with the[claimant’s] right to recapture.”). In response to a constitutional challenge to the statute, theburden of proof was shifted from the claimant to establish a lawful entitlement, to the City toestablish the forfeitability of the property. McClendon, 369 F. Supp. at 1394.

17 The relevant portion of the section provides:

Where moneys or property . . . have been used as a means of committing crime oremployed in aid or in furtherance of crime or held, used or sold in violation oflaw, . . . a person who . . . so used, employed, sold or held any such moneys orproperty or permitted or suffered the same to be used, employed, sold or held . . .or a person who derives his or her claim in any manner from or through any suchperson, shall not be deemed to be the lawful claimant entitled to any such moneysor property . . . .

N.Y.C. Code § 14-140(e)(1). The statute also deems not a lawful claimant any “person whoderives his or her claim in any manner from or through any such person.” Id. This formulationevidently could include other potentially innocent owners and interest holders, such as securedlenders, mortgagees, and lienholders, as well as purchasers, inheritors, and other transferees ofproperty interests.

24

to be the lawful claimant” to the property that has been seized by the police department. N.Y.C.

Code § 14-140(e)(1).16 The statute identifies two principal groups of those who are not “lawful

claimant[s]” and whose property therefore is forfeitable. One is the person who has “used” the

property “as a means of committing crime or employed [it] in aid or in furtherance of crime”; the

other is the person who “permitted or suffered the same to be used or employed.” Id.17

Thus, the seizure provision authorizes the Property Clerk to take custody,

following seizure, of “all property or money suspected of having been used as a means of

committing crime,” N.Y.C. Code § 14-140(b), without regard to whether or not an owner who

took no part in the crime “permitted or suffered” the vehicle to be used as an instrumentality of

Page 393: Sotomayor Cases

18 It is not entirely clear whether the City has the burden of proving that the owner“permitted or suffered” a vehicle to be used as an instrumentality of crime, or whether innocentownership is an affirmative defense to be raised by the claimant. Without deciding that question,we note that we have found one New York state court opinion that has held that the City has theburden. See Property Clerk v. Pagano, 170 A.D.2d 30, 35, 573 N.Y.S.2d 658, 661 (1st Dep’t1991) (“[A] proceeding brought under § 14-140 differs substantially from one brought pursuantto a forfeiture statute which expressly places the burden on a claimant to establish his innocentstate of mind. Clearly, § 14-140 contains no such language and, on the contrary, must beconstrued by its own terms to place an affirmative burden on [the City] . . . of proving thatrespondent ‘permitted or suffered’ the illegal use of the property.”).

19 Many state statutes, while permitting seizure of forfeitable property, are much moreexplicit than N.Y.C. Code § 14-140 in limiting or exempting the forfeiture liability of innocentowners. These statutes generally provide for affirmative defenses by innocent owners or makeproof of the owners’ culpable knowledge part of the seizing authority’s burden. For example, aMaine statute that permits forfeiture of a motor vehicle upon a defendant’s simultaneousconviction for operating under the influence (“OUI”) and operating after suspension for a priorOUI conviction, requires that the defendant be the “sole owner-operator of that vehicle.” Me.Rev. Stat. Ann. tit. 29-A, § 2421(1); see also State v. One Blue Corvette, 732 A.2d 856 (Me.1999) (holding that the statutory requirement that defendant be the “sole owner-operator of thatvehicle” precluded forfeiture of a vehicle jointly owned by defendant’s wife). Other state statutescontain express exceptions for a variety of innocent parties: Ariz. Rev. Stat. § 13-4304(1)-(2),(4)-(5) (common carriers; owners whose vehicles were “unlawfully in the possession of a personother than the owner” during commission or omission of an act; owners who “did not know andcould not reasonably have known of the act or omission”; and bona fide purchasers for value “notknowingly taking part in an illegal transaction” and “reasonably without notice of the act oromission giving rise to forfeiture”); Cal. Health & Safety Code §§ 11470(e)-(h) (persons otherthan defendant with a community property interest in the vehicle; bona fide purchasers andencumbrancers), 11488.5(e) (owners who did not have “actual knowledge that the seizedproperty would be or was used for a purpose for which forfeiture is permitted and consented tothat use”); Fla. Stat. § 932.703(6)-(7) (spouses, lienholders, lessors, joint owners, and otherparties lacking knowledge of the criminal activity alleged); N.J. Stat. Ann. § 2C:64-5(a)-(b)(unwitting lessors and security-interest holders; owners who show that they were “not involvedin or aware of the unlawful activity and . . . had done all that could reasonably be expected toprevent the proscribed use of the property by an agent”); La. Rev. Stat. Ann. § 14:98(D)(2)(b)(unwitting owners; owners of stolen vehicles); Minn. Stat. § 169A.63(7) (unwitting owners, bonafide security-interest holders, and lessors). Notably, New York state civil forfeiture statutes

25

the crime.18 A statute that authorizes the police to seize property to which the government has

not established a legal right or claim, and that on its face contains no limitation of forfeiture

liability for innocent owners, raises substantial constitutional concerns.19 Because plaintiffs in

Page 394: Sotomayor Cases

likewise provide a defense for innocent owners. See N.Y. C.P.L.R. 1311(4)(d)(ii)-(iii) (innocentpersons subject to “adverse impact of a forfeiture”; defendants acquitted of the underlyingcrime), 1311(3)(b)(iv) (non-criminal defendants who either did not know “that theinstrumentality was or would be used in the commission of a crime” or did not knowingly obtainan “interest in the instrumentality to avoid forfeiture”). In Property Clerk v. Ferris, 77 N.Y.2d428, 431, 568 N.Y.S.2d 577, 579 (1991), the New York Court of Appeals stated that the“interests of justice” exception in N.Y. C.P.L.R. 1311 “is ‘unique’ . . . and nothing in the articlesuggests that it applies in the limited forfeiture proceedings available pursuant to AdministrativeCode § 14-140”; see also Property Clerk v. Deans Overseas Shippers, Inc., 275 A.D.2d 204,205-06, 712 N.Y.S.2d 492, 493 (1st Dep’t 2000) (same).

Moreover, in contrast to § 14-140, the primary federal civil drug forfeiture statute, 21U.S.C. § 881, contained an express “innocent owner” defense, which stated that “no conveyanceshall be forfeited . . . to the extent of an interest of an owner, by reason of any act or omissionestablished by that owner to have been committed or omitted without the knowledge, consent, orwillful blindness of the owner.” 21 U.S.C. § 881(a)(4)(C). Replacing this provision, the CivilAsset Forfeiture Reform Act of 2000, Pub. L. No. 106-185, 114 Stat. 202 (codified as amendedat 18 U.S.C. § 983), creates a uniform innocent owner defense whereby a claimant mayaffirmatively prove innocent ownership by a preponderance of the evidence. 18 U.S.C. § 983(d). The statute specifically mentions owners who “did not know of the conduct giving rise toforfeiture”; bona fide purchasers or sellers for value; claimants who acquired a property interestthrough marriage, divorce, or legal separation; spouses or legal dependents who acquiredproperty by inheritance or probate; and joint tenants and others with a partial interest in property.

20 In contrast, for orders of attachment granted without notice under New York state civilforfeiture law, contained at N.Y. C.P.L.R. art. 13-A, the claiming authority must move for ahearing within five days of the seizure. N.Y. C.P.L.R. 1317, 1329 (providing for ex parteattachment of assets for five days, after which the attaching authority must move for an order

26

this action seek only a prompt and effective means to test the legitimacy of and justification for

the City’s retention of their vehicles following the seizure but prior to the forfeiture proceeding,

we have no occasion to rule on the constitutionality of the seizure provision itself. Nevertheless,

the scope of the police seizure authority granted under § 14-140(b), together with the City’s

“direct pecuniary interest in the outcome of the proceeding,” James Daniel Good Real Prop., 510

U.S. at 55-56, further convinces us of the need to provide a prompt opportunity for independent

and neutral judicial review of the probable validity of the City’s retention of vehicles pendente

lite.20

Page 395: Sotomayor Cases

confirming the attachment and permitting the defendant to challenge the attachment, with theburden of proof on the attaching authority to establish the grounds for the attachment, its need,and the probability of success on the merits). In upholding these provisions against doubts as tothe constitutionality under the Fourteenth Amendment of ex parte pre-judgment attachment, theNew York Court of Appeals relied, inter alia, on available means to challenge promptly the basisfor the attachment. Morgenthau v. Citisource, Inc., 68 N.Y.2d 211, 222, 508 N.Y.S.2d 152, 157(1986). N.Y. C.P.L.R. art. 13-A “does not limit or supersede” N.Y.C. Code § 14-140, however,and applies only to felonies and therefore has been held not to apply to forfeitures under § 14-140. Grinberg, 181 Misc. 2d at 449, 694 N.Y.S.2d at 320-21 (quotation marks omitted).

27

In sum, there is a heightened potential for erroneous retention where an arrestee,

whether for DWI or some other suspected criminal conduct, is not the owner of the seized

vehicle. The plight of innocent owners, as exemplified by the experience of plaintiff Sandra

Jones, persuades us that an early retention hearing following seizure under N.Y.C. Code § 14-

140 is constitutionally required.

3. Other Suggested Remedies Do Not Provide Prompt Post-SeizureReview.

In prosecuting vehicle forfeiture actions under N.Y.C. Code § 14-140, the City has

consistently opposed motions for interim relief in the form of a retention hearing. For example,

in its Memorandum of Law opposing a motion for an “immediate retention hearing” in the case

of Property Clerk v. Ali, the City stated that

under the CPLR there is no basis upon which defendant can evenmake this motion. Likewise, the CPLR does not provide for such ahearing. Defendant has circumvented the traditional rules of civilprocedure by asking this court to entertain, and plaintiff to defend,against a motion that has no legal basis and a hearing that wouldclearly be improper under the rules.

Memorandum of Law in Support of Plaintiff’s Opposition to an Immediate Retention Hearing, at

12, Property Clerk v. Ali, No. 413408/99 (Sup. Ct. N.Y. Co.). In Ali, as here, the City

maintained that due process was satisfied by a resolution of the merits at the eventual civil

Page 396: Sotomayor Cases

21 The New York court rules include among “matters to be considered” at the conferencethe simplification of issues, setting a schedule for discovery, addition of necessary parties, andsettlement, 22 N.Y. Comp. Codes R. & Regs. § 202.12(c)(1)-(4), and “any other matters that thecourt may deem relevant,” id. § 202.12(c)(5).

28

forfeiture hearing.

Nevertheless, defendants here suggest that plaintiffs may assert their

constitutional rights and challenge the City’s continued retention of their vehicles through the

procedural means of a Request for Judicial Intervention (“RJI”) or an Article 78 proceeding

brought under New York state law. We disagree. Under current law, any review of the

legitimacy of the City’s continued retention of a vehicle would likely come, at the earliest,

months after its seizure. This delay is a result, first, of the City’s need to initiate forfeiture

proceedings. Assuming that a claimant requests the return of the property immediately upon

seizure by the police, the City has twenty-five days in which to initiate proceedings. Only after a

civil forfeiture proceeding is commenced and process is served can an owner submit a RJI. See

22 N.Y. Comp. Codes R. & Regs. § 202.6(a). Along with the RJI, a claimant files a request for a

preliminary conference. See id. § 202.12(a) (“If the action has not been assigned to a judge, the

party shall file a request for judicial intervention together with the request for a preliminary

conference.”). Under the New York rules, a “preliminary conference” is held no later than forty-

five days from the request “unless the court orders otherwise.” Id. § 202.6(b). The rules do not

explicitly permit a determination of probable cause or the legitimacy of continued retention at the

preliminary conference, or even provide for the taking of evidence, indicating that, at most, the

preliminary conference may serve (as plaintiffs suggest) to set a future date for a probable cause

hearing.21 Under the current rules, therefore, any determination of probable cause for the initial

Page 397: Sotomayor Cases

22 Plaintiffs have represented that, in the New York Supreme Court, First JudicialDepartment, Civil Branch, all forfeiture actions under N.Y.C. Code § 14-140 are assigned to thesame judge, making further delays likely.

29

seizure or the legitimacy of continued deprivation might come sometime within three months

after the seizure, or perhaps much later.22 The Fourth and Fourteenth Amendments demand a

more expeditious determination of a vehicle owner’s rights.

The City also suggests that an Article 78 proceeding under New York state law is

available to claimants to redress any constitutional grievances. Requiring plaintiffs to resort to

an Article 78 proceeding, however, would place the onus on each plaintiff to bring a separate

civil action in order to force the City to justify its seizure and retention of a vehicle. This civil

action provides the “[r]elief previously obtained by writs of certiorari to review, mandamus or

prohibition.” N.Y. C.P.L.R. 7801. To petition for mandamus, a claimant would have the burden

of showing a clear legal right to the release of his or her vehicle. See Ass’n of Surrogate & Sup.

Ct. Reporters v. Bartlett, 40 N.Y.2d 571, 574, 388 N.Y.S.2d 882, 884 (1976) (“[P]etitioners’

success in this proceeding in the nature of mandamus requires a showing of a clear legal right to

the relief sought.”) (quotation marks omitted).

The City cites Grinberg v. Safir as proof that relief is currently available in an

Article 78 proceeding. To challenge the City’s forfeiture proceeding and retention of his vehicle,

the petitioner in Grinberg filed an Article 78 action and, with it, a request for a temporary

restraining order. Although the court held argument only two days after the action was filed, it

denied the temporary restraining order, observing that “[l]ikely inconvenience is not proof of

immediate and irreparable injury,” Grinberg, 181 Misc. 2d at 447 n.1, 694 N.Y.S.2d at 320 n.1,

and eventually decided the case in favor of the City more than two months later.

Page 398: Sotomayor Cases

30

In sum, we conclude that the suggested remedy of an Article 78 proceeding does

not provide a prompt and effective means for claimants to challenge the legitimacy of the City’s

retention of their vehicles pendente lite. Cf. Fuentes, 407 U.S. at 80 (finding unconstitutional a

Pennsylvania statute that “allows a post-seizure hearing if the aggrieved party shoulders the

burden of initiating one”). Furthermore, inasmuch as plaintiffs claim that the federal

Constitution requires the state court to offer a remedy that is currently not available under state or

local law, this constitutional challenge need not proceed through the state court before it reaches

the federal courts. See Kraebel v. N.Y. City Dep’t of Hous. Pres. & Dev., 959 F.2d 395, 404-06

(2d Cir. 1992) (addressing the claim that an Article 78 proceeding provided all the process

plaintiff was due, and finding that “[i]t is well-established that [42 U.S.C.] § 1983 generally

allows plaintiffs with federal or constitutional claims the right to sue in federal court without first

resorting to state judicial remedies”); cf. Logan, 455 U.S. at 432 (“Each of our due process cases

has recognized, either explicitly or implicitly, that because minimum procedural requirements are

a matter of federal law, they are not diminished by the fact that the State may have specified its

own procedures that it may deem adequate for determining the preconditions to adverse official

action.”) (quotation marks and alterations omitted).

IV. The Mathews v. Eldridge Inquiry

The Supreme Court has set forth three factors to weigh in deciding whether the

demands of the Due Process Clause are satisfied where the government seeks to maintain

possession of property before a final judgment is rendered. See Mathews, 424 U.S. at 335.

These factors should be used to evaluate the adequacy of process offered in post-seizure, pre-

judgment deprivations of property in civil forfeiture proceedings. Cf. James Daniel Good Real

Page 399: Sotomayor Cases

31

Prop., 510 U.S. at 53 (finding that the Mathews inquiry “provides guidance” in determining

whether to “tolerate” an exception to the rule requiring pre-deprivation notice and hearing). The

factors include (1) the private interest affected; (2) the risk of erroneous deprivation through the

procedures used and the value of other safeguards; and (3) the government’s interest.

A. The Private Interest Affected

The first factor to be considered in the Mathews inquiry is “the private interest

affected by the official action.” Mathews, 424 U.S. at 335. The deprivation of real or personal

property involves substantial due process interests. See James Daniel Good Real Prop., 510 U.S.

at 53-54 (recognizing that “Good’s right to maintain control over his home, and to be free from

governmental interference, is a private interest of historic and continuing importance”); id. at 81

(Thomas, J., concurring in part and dissenting in part) (agreeing with the majority that

“[i]ndividual freedom finds tangible expression in property rights”); Doehr, 501 U.S. at 11

(“[T]he property interests that attachment affects are significant.”); Fuentes, 407 U.S. at 70-71

(holding that loss of household furniture and appliances warrants a pre-deprivation hearing).

The particular importance of motor vehicles derives from their use as a mode of

transportation and, for some, the means to earn a livelihood. An “individual has an important

interest in the possession of his [or her] motor vehicle,” which is “often his [or her] most

valuable possession.” Lee, 538 F.2d at 31; see also Perry v. McDonald, 280 F.3d 159, 174 (2d

Cir. 2001) (noting that an individual’s interest in driving a vehicle represents a due process

concern); Coleman v. Watt, 40 F.3d 255, 260-61 (8th Cir. 1994) (stating, in the course of

applying the Mathews factors to impoundment of a car under state law, that “[a]utomobiles

occupy a central place in the lives of most Americans, providing access to jobs, schools, and

Page 400: Sotomayor Cases

23 In contrast, a provision of the federal civil forfeiture statute allows release of propertypendente lite under certain circumstances, including when “continued possession by theGovernment . . . will cause substantial hardship to the claimant, such as preventing thefunctioning of a business, preventing an individual from working, or leaving an individualhomeless; [and] the claimant’s likely hardship . . . outweighs the risk that the property will bedestroyed, damaged, lost, concealed, or transferred if it is returned to the claimant during thependency of the proceeding.” 18 U.S.C. § 983(f)(1)(C)-(D). Similarly, N.Y. C.P.L.R.1311(4)(d) permits the dismissal or mitigation of a forfeiture action “in the interests ofjustice”—for example, in light of “the seriousness and circumstances of the crime to which theproperty is connected relative to the impact of forfeiture of property upon the person whocommitted the crime.” As noted earlier, however, N.Y. C.P.L.R. 1311 does not apply to “thelimited forfeiture proceedings available pursuant to Administrative Code § 14-140.” Ferris, 77N.Y.2d at 431, 568 N.Y.S.2d at 579.

32

recreation as well as to the daily necessities of life”); Stypmann v. City & County of San

Francisco, 557 F.2d 1338, 1342-43 (9th Cir. 1977) (finding a “substantial” interest in the

“uninterrupted use of an automobile,” upon which the owner’s “ability to make a living” may

depend); cf. Justice Robert H. Jackson, The Task of Maintaining Our Liberties: The Role of the

Judiciary, 39 A.B.A.J. 961, 963 (1953) (“My equal right to drive an automobile may be only a

claim to use of property, but it concerns my personal freedom as well.”).

Other considerations as well bear on the importance of the private interest at

stake. One is the availability of hardship relief under the applicable law. Cf. Dixon v. Love, 431

U.S. 105, 113 (1977) (noting the availability, under an Illinois statute, of provisions for “hardship

and for holders of commercial licenses, who are those most likely to be affected by the deprival

of driving privileges”). Under the New York City Civil Administrative Code, no provision is

made for situations in which the seizure and retention of a vehicle would cause particular

hardship. See N.Y.C. Code § 14-140 (authorizing seizure of all property used as an

instrumentality of crime).23 Another consideration is the length of deprivation, which increases

the weight of an owner’s interest in possessing the vehicle. See Logan, 455 U.S. at 434 (noting

Page 401: Sotomayor Cases

33

the Court’s concern under Mathews for “the importance of the private interest and the length or

finality of the deprivation”). As noted above, the City retains seized vehicles for months or

sometimes years before the merits of a forfeiture action are addressed. Finally, the importance of

the claimant’s possessory interest post-seizure and pre-judgment is not diminished by the

likelihood that the government will eventually prevail in forfeiture proceedings. See James

Daniel Good Real Prop., 510 U.S. at 62 (“Fair procedures are not confined to the innocent. The

question before us is the legality of the seizure, not the strength of the Government’s case.”);

Fuentes, 407 U.S. at 87 (“The right to be heard does not depend upon an advance showing that

one will surely prevail at the hearing.”).

For these reasons, we cannot agree with the district court’s cursory assessment of

the interest at stake based solely on its observation that the seizure of the vehicles occurred “in a

jurisdiction that abounds in mass transit facilities.” Krimstock, 2000 WL 1702035, at *6. The

seizure authority under the statute extends not only to cars registered in New York City, but to

any found there; it also encompasses commercial as well as noncommercial vehicles. If the

named class members serve as any indication, motor vehicle use is often found among those for

whom mass transportation is inadequate. Valerie Krimstock, for example, states that the seizure

of her vehicle hindered her from traveling from her residence in the Bronx to her job in North

Tarrytown and from visiting her daughter who suffers from mental illness and lives in

Pennsylvania. The seizure and retention of Clarence Walters’ vehicle made it difficult, he

reports, to reach his construction job sites—some located in areas of Long Island or New Jersey

inaccessible by mass transit—and as a consequence he lost a certain amount of work. James

Webb, a 77-year-old retiree, states that the seizure and retention of his vehicle made it difficult

Page 402: Sotomayor Cases

34

for him and his wife to see their doctors and to visit friends, and prevented him from driving his

granddaughter to school.

B. The Risk of Erroneous Deprivation Through the Procedures Used and theProbable Value of Other Safeguards

The second factor to be considered under the Mathews test is “the risk of an

erroneous deprivation of such interest through the procedures used, and the probable value, if

any, of additional or substitute procedural safeguards.” Mathews, 424 U.S. at 335. The

particular deprivation with which we are concerned here is the City’s post-seizure, pre-judgment

retention of plaintiffs’ vehicles. The district court concluded that the procedures used by the

City—a warrantless arrest and the ultimate forfeiture proceeding—adequately protect plaintiffs

against erroneous deprivation of their vehicles. Krimstock, 2000 WL 1702035, at *6-*7. We are

troubled by this conclusion. Neither the arresting officer’s unreviewed probable cause

determination nor a court’s ruling in the distant future on the merits of the City’s forfeiture claim

can fully protect against an erroneous deprivation of a claimant’s possessory interest as his or her

vehicle stands idle in a police lot for months or years.

Nevertheless, we conclude that, on balance, the second Mathews factor weighs in

favor of the City. We acknowledge that the risk of erroneous seizure and retention of a vehicle is

reduced in the case of a DWI owner-arrestee, because a trained police officer’s assessment of the

owner-driver’s state of intoxication can typically be expected to be accurate. See People v.

Bennett, 238 A.D.2d 898, 899, 660 N.Y.S.2d 772, 774 (4th Dep’t 1997) (holding that the court

properly “instruct[ed] jurors that the police officers were experts in determining a person’s state

of intoxication”).

Yet the City’s victory on the second Mathews factor is a narrow one. As noted

Page 403: Sotomayor Cases

24 On several occasions, the Second Circuit has stressed that “[f]orfeiture is a harsh andoppressive procedure which is not favored by the courts.” United States v. $31,990 in U.S.Currency, 982 F.2d 851, 856 (2d Cir. 1993) (quotation marks omitted); cf. United States v.Lasanta, 978 F.2d 1300, 1305 (2d Cir. 1992) (“While congress may have intended civil forfeitureto be a powerful weapon in the war on drugs, it would, indeed, be a Pyrrhic victory for thecountry, if the government’s relentless and imaginative use of that weapon were to leave theconstitution itself a casualty.”) (internal quotation marks and citations omitted), abrogated onother grounds, Florida v. White, 526 U.S. 559 (1999).

25 Some state statutes expressly admonish against governmental opportunism in seizingand forfeiting property. See Cal. Health & Safety Code § 11469(a) (“Law enforcement is theprincipal objective of forfeiture. Potential revenue must not be allowed to jeopardize theeffective investigation and prosecution of criminal offenses, officer safety, the integrity of

35

earlier, the risk of erroneous deprivation that is posed to innocent owners is a substantial one.

Moreover, our inquiry into the risk of error is partly informed by the City’s pecuniary interest in

the outcome of § 14-140 proceedings. As the Supreme Court has observed, greater procedural

safeguards are “of particular importance . . . where the Government has a direct pecuniary

interest in the outcome of the proceeding.” James Daniel Good Real Prop., 510 U.S. at 55-56;

cf. United States v. Funds Held in the Name or for the Benefit of Wetterer, 210 F.3d 96, 110 (2d

Cir. 2000) (“We have previously observed the government’s ‘virtually unchecked use of the civil

forfeiture statutes and the disregard for due process that is buried in those statutes.’”) (quoting

United States v. Statewide Auto Parts, Inc., 971 F.2d 896, 905 (2d Cir. 1992)).24 Under the

City’s Administrative Code, property found to have “been used as a means of committing crime

or employed in aid or in furtherance of crime” may, at the discretion of the police commissioner,

“be used or converted to use for the purpose of the [police] department or any city, state or

federal agency.” N.Y.C. Code § 14-140(e)(2); see also Hyne, 147 Misc. 2d at 780, 557 N.Y.S.2d

at 248 (noting that the forfeiture law’s “remedial purposes” include the fact that “revenue is

generated and applied toward the cost of law enforcement”).25

Page 404: Sotomayor Cases

ongoing investigations, or the due process rights of citizens.”); id. § 11469(f) (“Seizing agenciesshall avoid any appearance of impropriety in the sale or acquisition of forfeited property.”); Fla.Stat. § 932.704(1) (“It is the policy of this state that law enforcement agencies shall utilize theprovisions of the Florida Contraband Forfeiture Act to deter and prevent the continued use ofcontraband articles for criminal purposes while protecting the proprietary interests of innocentowners and lienholders and to authorize such law enforcement agencies to use the proceedscollected . . . as supplemental funding for authorized purposes. The potential for obtainingrevenues from forfeitures must not override fundamental considerations . . . . It is also the policyof this state that law enforcement agencies ensure that, in all seizures made under [the Act], theirofficers adhere to federal and state constitutional limitations regarding an individual’s right to befree from unreasonable searches and seizures . . . .”).

36

The Supreme Court has expressed additional concern when, as here, the

“erroneous deprivation” cannot be recompensed by the claimant’s prevailing in later proceedings:

[T]he availability of a postseizure hearing may be no recompensefor losses caused by erroneous seizure. Given the congested civildockets in federal courts, a claimant may not receive an adversaryhearing until many months after the seizure. And even if theultimate judicial decision is that the claimant was an innocentowner, or that the Government lacked probable cause, thisdetermination, coming months after the seizure, “would not curethe temporary deprivation that an earlier hearing might haveprevented.”

James Daniel Good Real Prop., 510 U.S. at 56 (quoting Doehr, 501 U.S. at 15); cf. Shapiro, 424

U.S. at 629 (noting that where irreparable injury may result from a deprivation of property

pendente lite, “the Due Process Clause requires . . . an opportunity for some kind of

predeprivation or prompt post-deprivation hearing at which some showing of the probable

validity of the deprivation must be made”). In contrast, for example, to benefits for which full

retroactive relief is awarded if a plaintiff ultimately prevails, see, e.g., Mathews, 424 U.S. at 340,

an owner cannot recover the lost use of a vehicle by prevailing in a forfeiture proceeding. The

loss is felt in the owner’s inability to use a vehicle that continues to depreciate in value as it

stands idle in the police lot. Cf. Property Clerk v. Duck Jae Lee, 183 Misc. 2d 360, 363, 702

Page 405: Sotomayor Cases

37

N.Y.S.2d 792, 795 (Sup. Ct. N.Y. Co. 2000) (permitting a secured lender to intervene in a

vehicle forfeiture proceeding brought against a DWI arrestee who had defaulted on monthly car

payments, so that “the value of the subject car will not continue to depreciate if plaintiff lets the

action languish”).

In sum, because we recognize that the risk of erroneous deprivation in the context

of DWI owner-arrestees is in many cases a reduced one, we conclude that the second Mathews

factor weighs in favor of the City. The scales are very nearly in equipoise, however, in light of

the comparably greater risk of error that is posed to innocent owners, the City’s direct pecuniary

interest in the outcome of forfeiture proceedings, and the lack of adequate recompense for losses

occasioned by erroneous seizures of vehicles.

C. The Government’s Interest

The third Mathews factor examines “the Government’s interest, including the

function involved and the fiscal and administrative burdens that the additional or substitute

procedural requirement would entail.” Mathews, 424 U.S. at 335. The City argues that it has

three principal interests in continuing to retain the vehicles post-seizure and pre-judgment.

The first, and the most compelling among those the City has adduced, is to

prevent a vehicle from being sold or destroyed before a court can render judgment in future

forfeiture proceedings. The City cites Calero-Toledo v. Pearson Yacht Leasing Co., 416 U.S.

663 (1974), for the proposition that when property is easily transportable to another jurisdiction,

the City must retain custody pending resolution of the forfeiture proceedings. The Supreme

Court, indeed, recognized that immediate seizure was necessary in Calero-Toledo because

otherwise the yacht at issue “might have disappeared had the Government given advance warning

Page 406: Sotomayor Cases

38

of the forfeiture action.” James Daniel Good Real Prop., 510 U.S. at 57. We note initially that

the Puerto Rican statute under which the yacht in Calero-Toledo was seized and forfeited

provided that notice of the seizure must be served upon interested parties within ten days

following the seizure and that those parties have fifteen days following service within which to

challenge the seizure by serving a complaint on the confiscating officer. The complaint, filed in

the Superior Court, “shall be heard without subjection to docket.” 34 P.R. Laws Ann.

§ 1722(a) (repealed 1988). In Calero-Toledo, the yacht was automatically forfeited when no

challenge was made to the seizure within fifteen days after service of the notice of seizure.

Calero-Toledo, 416 U.S. at 668.

The critical difference between Calero-Toledo and the present case is that

plaintiffs’ vehicles have already been seized and are in the hands of the police. Just as with real

property seized by the government in forfeiture proceedings, there is no danger that these

vehicles will abscond. See James Daniel Good Real Prop., 510 U.S. at 56-57 (discussing the

need for seizure of movable property). Plaintiffs seek a determination only of whether continued

retention of their vehicles by the City is valid and justified. Continued retention may be

unjustified when other means of restraint would accomplish the City’s goals. See id. at 59 (“In

the usual case, the Government . . . has various means, short of seizure, to protect its legitimate

interests” in forfeitable property). To ensure that the City’s interest in forfeitable vehicles is

protected, claimants could post bonds, or a court could issue a restraining order to prohibit the

sale or destruction of the vehicle. See id. at 58-59 (suggesting judicial means to ensure that real

Page 407: Sotomayor Cases

26 We note that in contrast to the forfeiture statute at issue in the present case, the PuertoRican statute in Calero-Toledo required that seized motor vehicles be appraised for theirmonetary value by the Office of Transportation and that plaintiffs “have the right to give bond . . . to answer for the seizure if the lawfulness of the latter is upheld.” 34 P.R. Laws Ann. § 1722(b) (repealed 1988).

39

property is not sold or destroyed pendente lite).26 The need to prevent forfeitable property from

being sold or destroyed during the pendency of proceedings does not necessarily justify

continued retention of all vehicles when other means of accomplishing those goals are available.

A bond is in some respects a superior form of security because it entails no storage costs or costs

of sale.

A second reason offered by the City for maintaining custody of vehicles prior to

judgment in forfeiture proceedings is that the City’s in rem jurisdiction over the vehicles depends

upon its unbroken possession from seizure to judgment. The Supreme Court has held, however,

that possession of a res during the entire course of the proceedings is unnecessary to preserve

jurisdiction. See Republic Nat’l Bank of Miami v. United States, 506 U.S. 80, 88-89 (1992) (“We

hold that, in an in rem forfeiture action, the Court of Appeals is not divested of jurisdiction by the

prevailing party’s transfer of the res from the district.”). Noting that the in rem rules had their

origin in admiralty law, where a court established its jurisdiction by seizure, the Supreme Court

found that “the court must have actual or constructive control of the res when an in rem forfeiture

is initiated.” Id. at 87 (emphasis added); cf. The Palmyra, 25 U.S. (12 Wheat.) 1, 10 (1827)

(Story, J.) (“Whenever a stipulation [bond] is taken in an admiralty suit, for the property

subjected to legal process and condemnation, the stipulation is deemed a mere substitute for the

thing itself, and the stipulators liable to the exercise of all those authorities on the part of the

Page 408: Sotomayor Cases

27 In addition, we note that the City does not caption these civil forfeiture actions asproceedings in rem. Instead, they are styled as in personam actions—for example, PropertyClerk v. Ismael Delapaz.

40

Court, which it could properly exercise if the thing itself were still in its custody.”).27

The final interest adduced by the City is the need to prevent the offending

res—here, the seized vehicle—from being used as an instrumentality in future acts of driving

while intoxicated. Of course, at the time of initial seizure and retention the “offending res” is

only an allegedly offending res, inasmuch as the owner’s or owner-arrestee’s misconduct in

connection with the instrumentality has yet to be established in either a criminal or a civil

proceeding. Moreover, although the Supreme Court has found that certain situations of

“executive urgency,” James Daniel Good Real Prop., 510 U.S. at 60, call for action that

dispenses with normal due process guarantees, this case does not fall within that narrow category.

In defining situations of “executive urgency,” the Court has provided the examples of summary

seizures during wartime, seizures of contaminated food, and, formerly, the collection of taxes

when the very existence of the government depended upon them. See id. at 59-60. To take one

example, the Court allowed the seizure, without prior judicial process, of forty-seven barrels of

poultry from a Chicago food storage warehouse after city inspectors determined they were

“putrid, decayed, poisonous, or infected in such a manner as to render it unsafe or unwholesome

for human food.” N. Am. Cold Storage Co v. City of Chicago, 211 U.S. 306, 308 (1908). The

threat to the public was immediate, and the spoiled poultry, like contraband, was unlikely to be

used for some other legitimate purpose. Motor vehicles, in contrast, present no such threat and

maintain their usefulness. Cf. Austin v. United States, 509 U.S. 602, 621 (1993) (“‘There is

nothing even remotely criminal in possessing an automobile.’”) (quoting One 1958 Plymouth

Page 409: Sotomayor Cases

28 A claimant’s proven history of persistent drunkenness or repeated DWI violations,however, might justify a fact-finder in denying release of the vehicle pendente lite.

41

Sedan v. Pennsylvania, 380 U.S. 693 (1965)). In James Daniel Good Real Property, for

example, the Supreme Court found that enforcement of the drug forfeiture laws did not present

“a plausible claim of urgency” strong enough to dispense with normal due process guarantees.

James Daniel Good Real Prop., 510 U.S. at 61.

Even if driving while intoxicated were considered a matter of “executive

urgency,” the response the City has chosen, requiring the impoundment of vehicles until

forfeiture proceedings are terminated, is ill-suited to address the urgency. While initial seizure of

a vehicle serves the constructive purpose of keeping an individual from driving in an inebriated

condition, that purpose often loses its basis in urgency once the individual has regained sobriety

on the morrow.28 Furthermore, the remedy of continued impoundment leaves the alleged

offender free to drive while intoxicated in any other vehicle when the opportunity presents itself,

while depriving some potentially innocent owners of the often indispensable benefits of daily

access to their vehicles.

Finally, the City’s asserted interest in removing dangerous drivers from the road is

weakened if it extends not to all vehicles seized in connection with DWI arrests, but only to those

that might yield an attractive price at auction. The November 1988 Forfeiture Guide produced by

the Legal Bureau of the City of New York Police Department instructs that “[c]ertain categories

of property do not warrant forfeiture litigation due to their small value or the near impossibility

of a successful outcome,” including “[n]on-owner operated vehicles ten years old or older,”

unless, inter alia, “the vehicle has a special value, e.g., an expensive import.” 1988 Forfeiture

Page 410: Sotomayor Cases

29 In Statewide Auto Parts, this Court considered a due process challenge to the federalgovernment’s seizure of real and personal commercial properties pursuant to an ex parte warrantand the federal civil forfeiture statute. Statewide Auto Parts, 971 F.2d at 898-99. In other cases

42

Guide, at 24-25. We do not know whether this passage reflects current policy, but we note that

the City’s interest in safety cannot be paramount if it seeks to remove from the road only a

lucrative subset of the vehicles seized from intoxicated drivers.

D. Balancing the Mathews v. Eldridge Factors

Balancing the Mathews factors, we find that the Fourteenth Amendment guarantee

that deprivations of property be accomplished only with due process of law requires that

plaintiffs be afforded a prompt post-seizure, pre-judgment hearing before a neutral judicial or

administrative officer to determine whether the City is likely to succeed on the merits of the

forfeiture action and whether means short of retention of the vehicle can satisfy the City’s need to

preserve it from destruction or sale during the pendency of proceedings.

In James Daniel Good Real Property, the Supreme Court concluded that to seize

real property without notice and hearing, the “Government must show that less restrictive

measures—i.e., a lis pendens, restraining order, or bond—would not suffice to protect the

Government’s interests in preventing the sale, destruction, or continued unlawful use of the real

property.” James Daniel Good Real Prop., 510 U.S. at 62; cf. Statewide Auto Parts, 971 F.2d at

905 (urging district courts “whenever possible . . . [to] favor less drastic measures, such as

occupancy agreements, bonds, receiverships, lis pendens, or other means for preserving the status

quo ante seizure until the criminality underlying the claimed forfeiture can be established in the

context of a proper criminal proceeding with its attendant constitutional protections to the

accused”).29 Here, once the vehicles have been seized, and concerns for establishing jurisdiction

Page 411: Sotomayor Cases

involving seizure of real property, this Court has been careful to emphasize the value of lessdrastic means than seizure for protecting the government’s interest in forfeitable property. See,e.g., United States v. Premises & Real Property at 4492 South Livonia Road, Livonia, N.Y., 889F.2d 1258, 1265 (2d Cir. 1989) (“Any exigency that might be posed by the threat of anencumbrance on, or transfer of, the property may be met by less restrictive means than seizure,for example, by the filing of a lis pendens, as was done in this case, along with a restraining orderor bond requirement.”); see also United States v. Puello, 814 F. Supp. 1155, 1164 (E.D.N.Y.1993) (stating that, “[m]indful of [the Second Circuit’s] warning [to use less restrictive meansthan outright seizure], the Court has explored the possibility of these less drastic means with theparties.”); United States v. Joseph Health & Beauty Supply, 807 F. Supp. 323, 325 (S.D.N.Y.1992) (permitting a claimant to occupy seized business premises for the conduct of lawfulbusiness pending trial, as a less restrictive measure than outright seizure of the leasehold underfederal forfeiture law); cf. 18 U.S.C. § 983(f)(7), (j) (providing for restraining orders, injunctions,bonds and other means to preserve property during litigation); Supplemental Rules for CertainAdmiralty and Maritime Claims, Rule E(4)(f) (providing, except where the United States isseeking forfeiture, for a “prompt hearing at which the plaintiff shall be required to show why thearrest or attachment [of property] should not be vacated or other relief granted consistent withthese rules”); Fla. Stat. § 932.703(d) (providing that, if the court determines that probable causeexists to believe that property was used in violation of the state contraband forfeiture statute, “thecourt shall order the property restrained by the least restrictive means to protect against disposal,waste, or continued illegal use,” including such means as a “bond or other adequate securityequivalent to the value of the property”).

43

and immediate prophylactic custody are satisfied, we find that the Due Process Clause requires

that claimants be given an early opportunity to test the probable validity of further deprivation,

including probable cause for the initial seizure, and to ask whether other measures, short of

continued impoundment, would satisfy the legitimate interests of the City in protecting the

vehicles from sale or destruction pendente lite. Whether the vehicle is in the hands of the police

the morning after it has been seized, as in this case, or whether James Daniel Good’s property is

still in his hands the morning before the marshals arrive with a warrant, the question is what

reason the government has for refusing to exercise some means short of continued retention after

seizure to guarantee that property will be available to satisfy a civil forfeiture judgment.

Page 412: Sotomayor Cases

30 We also note that this Court has observed, in the context of the federal forfeiture laws,that “under the Barker [speedy trial] test, which we think applies to the holding of the forfeituretrial as well as to the filing of the action, there is a due process violation at some point.” UnitedStates v. Banco Cafetero Panama, 797 F.2d 1154, 1163 (2d Cir. 1986). “The Governmentargues that, once an action is filed, delays of any length may be granted to allow completion ofrelated criminal proceedings. To require prompt filing of a forfeiture action but allow indefinitepostponement of the trial would reduce the filing requirement to a nullity.” Id.

44

E. Inapplicability of United States v. $8,850 and the Speedy Trial Test

The City argues that the Mathews v. Eldridge balancing test is displaced by the

Supreme Court’s decision to apply the speedy trial test, and not the Mathews inquiry, in

examining the constitutionality of any delay in the return of property subject to future civil

forfeiture proceedings. See United States v. $8,850, 461 U.S. 555 (1983) (applying the speedy

trial test set forth in Barker v. Wingo, 407 U.S. 514 (1972), in finding that an eighteen-month

delay in filing a customs forfeiture action did not violate constitutional due process guarantees).

We disagree. As discussed in Section III.A above, plaintiffs’ claim does not

concern the speed with which civil forfeiture proceedings themselves are instituted or conducted.

Instead, plaintiffs seek a prompt post-seizure opportunity to challenge the legitimacy of the City’s

retention of the vehicles while those proceedings are conducted. The application of the speedy

trial test presumes prior resolution of any issues involving probable cause to commence

proceedings and the government’s custody of the property or persons pendente lite, leaving only

the issue of delay in the proceedings. The impoundment of property—or the incarceration of a

criminal defendant—certainly increases the hardship worked by any delay. The Constitution,

however, distinguishes between the need for prompt review of the propriety of continued

government custody, on the one hand, and delays in rendering final judgment, on the other.30

Page 413: Sotomayor Cases

31 Pursuant to 38-A R.C.N.Y. § 12-32, the voucher issued to an arrestee by the policeupon arrest and seizure must notify the arrestee of procedures for demanding the return ofproperty and the possible actions that the Property Clerk may take after the demand has beenmade. See also Butler v. Castro, 896 F.2d 698, 702 (2d Cir. 1990) (“[A] voucher must be givento an arrestee for non-contraband property seized [and] must also give notice of the procedures tobe followed to recover such property.”). Adequate notice of the right to a post-seizure retentionhearing could readily be added to this information. This procedure may not be adequate,however, where the arrestee and the vehicle owner are not the same person.

32 We note that both parties appear to agree that plaintiffs’ vehicles were not seized as “arrest evidence” pursuant to N.Y.C. Code § 14-140(b) but rather as instrumentalities of crime. According to the City: “The seven named plaintiffs [have had their vehicles] impounded and heldby the Property Clerk for forfeiture, because they were used in furtherance of a crime.” In anyevent, it is hard to imagine how an arrestee’s vehicle could serve as evidence in the ordinaryDWI case.

45

V. The Prompt Vehicle Retention Hearing

As a remedy, we order that claimants be given a prompt post-seizure retention

hearing, with adequate notice,31 for motor vehicles seized as instrumentalities of crime pursuant

to N.Y.C. Code § 14-140(b).32 There is no universal approach to satisfying the requirements of

meaningful notice and opportunity to be heard in a situation such as this. See Fuentes, 407 U.S.

at 96 (acknowledging that the “nature and form of [fair prior hearings] are legitimately open to

many potential variations”); cf. Green v. County Sch. Bd., 391 U.S. 430, 439 (1968) (“There is no

universal answer to the complex problems of desegregation; there is obviously no one plan that

will do the job in every case.”); Int’l Salt Co. v. United States, 332 U.S. 392, 400-01 (1947)

(“The framing of decrees should take place in the District rather than in Appellate Courts. They

are invested with large discretion to model their judgments to fit the exigencies of the particular

case.”). Moreover, the cost of additional procedures and the details of their implementation are

matters peculiarly suited to the experience of the district court and the knowledge of the parties.

Therefore, as we did in McClendon, 460 F.2d at 116, we leave it to the district court, in

Page 414: Sotomayor Cases

33 For example, we do not expect that these limited hearings will resemble the sometimesprolonged pretrial hearings that district courts in this Circuit have conducted to address probablecause for the government’s seizure of property under federal forfeiture laws. See, e.g., NamesRegistry Publ’g, 68 F.3d at 579 (noting that the district court below had held a two-day pretrialhearing to examine probable cause for an ex parte seizure of bank funds under federal forfeiturelaw); Daccarett, 6 F.3d at 45 (three-day pretrial evidentiary hearing to examine probable causefor seizure of electronic transfer funds under federal forfeiture law); Statewide Auto Parts, 971F.2d at 900 (oral argument followed by further written evidentiary submissions on the issue ofprobable cause to seize real and personal commercial properties under federal forfeiture law); seealso In re Seizure of All Funds in Accounts in Names Registry Publ’g, Inc., 887 F. Supp. 435,449 (E.D.N.Y.) (“[I]f a claimant challenges the validity of a seizure, . . . then the merits of theforfeiture trial are expedited and the government must establish probable cause for the forfeitureprior to the forfeiture trial.”), vacated on other grounds, 68 F.3d 577 (2d Cir. 1995). Unlikefederal forfeiture actions, which often involve complex evidentiary issues, multiple parties andwitnesses, and elaborate underlying criminal allegations, actions instituted against vehicle owners

46

consultation with the parties, to fashion appropriate procedural relief consistent with this opinion.

See Fuentes, 407 U.S. at 97 n.33 (“Leeway remains to develop a form of hearing that will

minimize unnecessary cost and delay while preserving the fairness and effectiveness of the

hearing . . . .”); cf. United States v. City of Yonkers, 197 F.3d 41, 57 (2d Cir. 1999) (noting that a

district court has “broad equitable discretion to apportion remedial costs” in desegregation cases).

Although we decline to dictate a specific form for the prompt retention hearing,

we hold that, at a minimum, the hearing must enable claimants to test the probable validity of

continued deprivation of their vehicles, including the City’s probable cause for the initial

warrantless seizure. In the absence of either probable cause for the seizure or post-seizure

evidence supporting the probable validity of continued deprivation, an owner’s vehicle would

have to be released during the pendency of the criminal and civil proceedings.

We hasten to point out that we do not envision the retention hearing as a forum

for exhaustive evidentiary battles that might threaten to duplicate the eventual forfeiture

hearing.33 Inasmuch as the purpose of the hearing is the limited one of determining whether the

Page 415: Sotomayor Cases

under the New York City forfeiture statute typically present relatively discrete, straightforwardissues.

34 At the time of oral argument before this Court, three of the seven namedplaintiffs—Jones, Krimstock, and Walters—had recovered their vehicles. Inasmuch as plaintiffsprimarily seek relief in the form of a preliminary injunction and class certification, the questionarises whether the claims of some plaintiffs have been rendered moot, and this is a question that acourt must address sua sponte. The Supreme Court has held in some cases, however, that “thetermination of a class representative’s claim does not moot the claims of the unnamed membersof the class. . . . That the class was not certified until after the named plaintiffs’ claims hadbecome moot does not deprive us of jurisdiction. . . . In such cases, the ‘relation back’ doctrine isproperly invoked to preserve the merits of the case for judicial resolution.” County of Riversidev. McLaughlin, 500 U.S. 44, 51-52 (1991) (internal quotation marks and citations omitted); seealso Sosna v. Iowa, 419 U.S. 393, 402 n.11 (1975) (same); cf. United States Parole Comm’n v.Geraghty, 445 U.S. 388, 399 (1980) (“Some claims are so inherently transitory that the trial courtwill not have even enough time to rule on a motion for class certification before the proposedrepresentative’s individual interest expires.”) (citing Gerstein v. Pugh, 420 U.S. 103, 110 n.11(1975)); Comer v. Cisneros, 37 F.3d 775, 798-99 (2d Cir. 1994) (noting that in some cases inwhich the claims of named plaintiffs have become moot prior to class certification, “the courtspermit the class certification to relate back to the filing of the complaint and hold that theplaintiffs have properly preserved the merits of the case for judicial resolution”); Robidoux v.Celani, 987 F.2d 931, 938-39 (2d Cir. 1993) (same).

47

vehicle should be returned to its owner during the pendency of proceedings, due process should

be satisfied by an initial testing of the merits of the City’s case. In addition, the retention hearing

will allow the court to consider whether less drastic measures than continued impoundment, such

as a bond or a restraining order, would protect the City’s interest in the allegedly forfeitable

vehicle during the pendency of proceedings.

On remand, in addition to fashioning appropriate relief, the district court should

ascertain how many of the named plaintiffs have recovered their vehicles from the Property Clerk

and determine whether their claims have thereby been rendered moot.34 To the extent that

mootness may affect the claims of any of the named plaintiffs, the court, in addressing the issue

of class certification, should consider whether exceptions to the mootness doctrine preserve the

Page 416: Sotomayor Cases

48

merits of the case for judicial resolution of the unnamed class members’ claims.

CONCLUSION

In conclusion, we hold that promptly after their vehicles are seized under N.Y.C.

Code § 14-140 as alleged instrumentalities of crime, plaintiffs must be given an opportunity to

test the probable validity of the City’s deprivation of their vehicles pendente lite, including

probable cause for the initial warrantless seizure. We remand to the district court to rule on

plaintiffs’ request to certify their class pursuant to Fed. R. Civ. P. 23, and to formulate, in

consultation with the parties, the appropriate injunctive relief needed to redress the constitutional

violations examined in this opinion.

Vacated and remanded.

Page 417: Sotomayor Cases

UNITED STATES COURT OF APPEALS1FOR THE SECOND CIRCUIT2

3August Term, 20014

5(Argued: March 13, 2002 Decided: September 13, 2002)6

7Docket No. 01-61688

9_____________________________________________ 10

11THE CENTER FOR REPRODUCTIVE LAW AND POLICY, JANET BENSHOOF, ANIKA12

RAHMAN, KATHERINE HALL MARTINEZ, JULIA ERNST, LAURA KATZIVE, MELISSA13UPRETI, CHRISTINA ZAMPAS,14

15Plaintiffs-Appellants,16

17v.18

19GEORGE W. BUSH, in his official capacity as President of the United States, COLIN20

POWELL, in his official capacity as Secretary of State, ANDREW NATSIOS, in his official21capacity as Administrator of the United States Agency for International Development, 22

23Defendants-Appellees.24

_____________________________________________2526

Before: McLAUGHLIN, LEVAL, and SOTOMAYOR, Circuit Judges. 2728

The United States District Court for the Southern District of New York (Preska,29

J.) dismissed for lack of Article III standing plaintiffs’ constitutional challenge to the federal30

government’s “Mexico City Policy” restricting the disbursement of funds to foreign non-31

governmental organizations. Finding that this case falls under an exception to the Supreme32

Court’s rule against assuming the existence of jurisdiction, we dismiss plaintiffs’ First33

Amendment claim on the merits without reaching the question of constitutional standing. We34

dismiss plaintiffs’ due process claim for lack of prudential standing, and we dismiss plaintiffs’35

equal protection claim on the merits. 36

Dismissal affirmed on different grounds.37

Page 418: Sotomayor Cases

2

1

_______________________________________________23

SIMON HELLER, The Center for Reproductive Law &4Policy, New York, NY (Janet Benshoof, on the brief), for5Plaintiffs-Appellants. 6

7GREGORY G. KATSAS, Deputy Assistant Attorney8General, Washington, D.C. (Robert D. McCallum, Jr.,9Assistant Attorney General; Robert M. Loeb and Sharon10Swingle, Attorneys, Department of Justice Civil Division;11James B. Comey, United States Attorney; Gideon A. Schor,12Chief Appellate Attorney, on the brief), for Defendants-13Appellees.14

______________________________________________1516

SOTOMAYOR, Circuit Judge:17

This suit was brought by a domestic organization that advocates reproductive18

rights and by attorneys employed by the organization. Plaintiffs challenge the so-called “Mexico19

City Policy,” pursuant to which the United States government requires foreign organizations, as a20

condition of receiving government funds, to agree neither to perform abortions nor to promote21

abortion generally. Plaintiffs maintain that these restrictions violate their First Amendment rights22

to freedom of speech and association. The district court dismissed the case for lack of subject23

matter jurisdiction, finding that plaintiffs lack standing under Article III of the Constitution. The24

district court was following the general rule, set forth by the Supreme Court in Steel Co. v.25

Citizens for a Better Environment, 523 U.S. 83, 94-95 (1998), that a federal court may not26

assume it has jurisdiction over a matter and proceed directly to the merits. The instant case is27

exceptional, however. Some twelve years ago we entertained and rejected, on the merits, the28

same constitutional challenge to the provision at issue here. We therefore find that this case falls29

Page 419: Sotomayor Cases

3

within an exception recognized by the Supreme Court in Steel Co., and we dismiss the First1

Amendment claim on the merits without deciding the standing question. Plaintiffs also bring2

claims under the Due Process Clause and the equal protection component of the Fifth3

Amendment. We dismiss the due process claim under the doctrine of prudential standing, as4

plaintiffs’ alleged harm does not fall within the zone of interests protected by the Due Process5

Clause. We dismiss the equal protection claim as without merit; while plaintiffs do have6

standing for this claim under the concept we have dubbed “competitive advocate standing,” the7

classification they challenge does not constitute an equal protection violation. 8

BACKGROUND9

We accept the allegations in the complaint as true on this motion to dismiss. The10

facts of this case, which are set forth in greater detail by the district court, see Center for11

Reproductive Law & Policy v. Bush, No. 01 CIV. 4986, 2001 WL 868007 (S.D.N.Y. July 31,12

2001) (“CRLP”), are as follows. Plaintiff The Center for Reproductive Law & Policy (“CRLP”)13

is a nonprofit advocacy organization devoted to the promotion of reproductive rights. Individual14

plaintiffs Janet Benshoof, Anika Rahman, Katherine Hall Martinez, Julia Ernst, Laura Katzive,15

Melissa Upreti and Christina Zampas are CRLP staff attorneys engaged in the organization’s16

global mission of reproductive law reform. Defendant George W. Bush is the President of the17

United States. Defendant Colin Powell is the U.S. Secretary of State and is thus responsible for18

“ensuring program and policy coordination among agencies of the United States Government in19

carrying out the policies set forth in the Foreign Assistance Act. . . .” 22 U.S.C. § 6593(b)(2). 20

Defendant Andrew Natsios is the Administrator of the United States Agency for International21

Page 420: Sotomayor Cases

1 The term derives from a United Nations conference held in Mexico City in 1984, atwhich the United States delegation presented a policy statement outlining the type of abortion-related restrictions at issue in this case. CRLP, 2001 WL 868007, at *2 n.1.

4

Development (“USAID”). At issue in this case is the so-called “Mexico City Policy”1 of the1

United States government, whereby foreign non-governmental organizations (“NGOs”) receiving2

U.S. government funds must agree to a provision called the “Standard Clause,” which prohibits3

the organizations from engaging in activities that promote abortion (also referred to as the4

“challenged restrictions”). 5

The Foreign Assistance Act of 1961 (“FAA”) authorizes the President “to furnish6

assistance, on such terms and conditions as he may determine, for voluntary population7

planning.” 22 U.S.C. § 2151b(b). The President’s authority to allocate FAA funding has been8

delegated to the Secretary of State and, in turn, to the Administrator of USAID. See Exec. Order9

No. 13,118, 64 Fed. Reg. 16,595 (Mar. 31, 1999); State Department Delegation of Authority No.10

145-1, 45 Fed. Reg. 51,974 (Aug. 5 1980); International Development Cooperation Agency11

Delegation of Authority No. 7, 45 Fed. Reg. 52,470 (Aug. 7, 1980). In 1973, Congress enacted12

the Helms Amendment, which prohibits the use of foreign assistance funds to pay for, among13

other things, “the performance of abortions as a method of family planning or to motivate or14

coerce any person to practice abortions.” 22 U.S.C. § 2151b(f)(1). This restriction applies only15

to the use of U.S. government funds; foreign NGOs receiving assistance may still promote16

abortion with non-U.S. government funds without violating the terms of the statute. The17

executive branch, however, has attached additional conditions to the granting of foreign18

assistance funds, as it is authorized to do by the FAA. See 22 U.S.C. § 2151b(b). These19

additional conditions are the subject of this suit.20

Page 421: Sotomayor Cases

5

The challenged restrictions originated in August 1984, when President Ronald1

Reagan announced the Mexico City Policy (“the Policy”). The Policy expressed the2

government’s disapproval of abortion as an element of family planning programs and set forth3

various ways in which the government would prohibit its funds from being used to support4

abortion overseas. Among these, it was announced that “the United States will no longer5

contribute to separate nongovernmental organizations which perform or actively promote6

abortion as a method of family planning in other nations.” CRLP, 2001 WL 868007, at *47

(citations omitted). 8

Pursuant to the Mexico City Policy, USAID incorporated the “Standard Clause”9

into its family planning assistance agreements and contracts. The Standard Clause provides that10

in order to be eligible for USAID funding, a foreign NGO must certify in writing that it “will not,11

while receiving assistance under the grant, perform or actively promote abortion as a method of12

family planning in AID-recipient countries or provide financial support to other foreign13

nongovernmental organizations that conduct such activities.” Id. at *5 (quotation marks14

omitted). The restrictions established in the Standard Clause extend to all activities of recipient15

NGOs, not merely to projects funded by USAID. Thus, in order to receive U.S. government16

funds, a foreign NGO may not engage in any activities that promote abortion. These restrictions17

do not apply to domestic NGOs such as plaintiff CRLP. 18

The Mexico City Policy was rescinded by President Bill Clinton in January 1993,19

but was reinstated by President George W. Bush in March 2001. President Bush issued an20

official memorandum that restored the abortion-related restrictions discussed above, including21

the Standard Clause. See Memorandum, Restoration of the Mexico City Policy, 66 Fed. Reg.22

Page 422: Sotomayor Cases

2 “Abortion as a method of family planning” does not include “abortions performed if thelife of the mother would be endangered if the fetus were carried to term or abortions performedfollowing rape or incest (since abortion under these circumstances is not a family planning act).” Restoration Memorandum, 66 Fed. Reg. at 17,306.

6

17,303, 17,309 (Mar. 28, 2001) (“Restoration Memorandum”). Accordingly, as a condition of1

receiving U.S. government funds, foreign NGOs again are required to agree not to perform or2

actively promote abortion as a method of family planning.2 3

Plaintiffs bring this suit for injunctive and declaratory relief. Plaintiffs’ primary4

claim, and the one with which the district court appears exclusively to have concerned itself, is5

based on the First Amendment. The thrust of this claim is that, as a result of the challenged6

restrictions, foreign NGOs are chilled from interacting and communicating with domestic7

abortion rights groups such as plaintiff CRLP, thus depriving plaintiffs of their rights to freedom8

of speech and association in carrying out the mission of the organization. Plaintiffs also allege9

that the restrictions violate the Equal Protection Clause of the Fifth Amendment by preventing10

plaintiffs from competing on “equal footing” with domestic anti-abortion groups, and that they11

violate the Due Process Clause by failing to give clear notice of what speech and activities they12

prohibit and by encouraging arbitrary and discriminatory enforcement. Finally, plaintiffs attempt13

to bring a claim under customary international law, the substance of which appears to be identical14

to their First Amendment claim.15

The district court dismissed the action in its entirety on the ground that plaintiffs16

lack standing under Article III of the Constitution. The court first noted that because the17

challenged restrictions apply only to foreign NGOs, not to domestic organizations such as CRLP,18

the Mexico City Policy does not affect plaintiffs directly. CRLP, 2001 WL 868007, at *7. The19

Page 423: Sotomayor Cases

7

court then applied the three-pronged standing test set out by the Supreme Court in Lujan v.1

Defenders of Wildlife, 504 U.S. 555 (1992), and concluded that plaintiffs had failed to2

demonstrate that (1) concrete injury in fact, (2) a causal connection between the alleged injury3

and the government’s conduct, and (3) that the alleged injury is sufficiently redressable by a4

favorable decision. CRLP, 2001 WL 868007, at *8-*12. 5

Our review is de novo. See Connecticut v. Physicians Health Servs. of Conn.,6

Inc., 287 F.3d 110, 114 (2d Cir. 2002). “The reviewing court may, of course, affirm on any7

ground appearing in the record below.” MFS Sec. Corp. v. New York Stock Exch., Inc., 277 F.3d8

613, 617 (2d Cir. 2002). 9

DISCUSSION10

I. First Amendment Claim 11

A. Plaintiffs’ Allegations12

The crux of plaintiffs’ First Amendment claim is their contention that the13

restrictions chill foreign NGOs from collaborating with domestic NGOs like CRLP because such14

collaboration may be viewed as promoting abortion and thus would jeopardize the foreign15

NGOs’ receipt of U.S. government funds. Plaintiffs argue that such collaboration is essential to16

their ability to carry out their mission as advocates of reproductive rights and that depriving them17

of this ability violates their freedom of speech and association. 18

Specifically, plaintiffs allege that they depend on collaboration with foreign19

NGOs in order to advocate abortion law reform in foreign countries; to gather reliable20

information regarding abortion laws; to disseminate publications and reports; to reach audiences21

worldwide in order to promote abortion law reform; to access victims and witnesses of human22

Page 424: Sotomayor Cases

8

rights abuses; to lobby the United States government to rescind the Restoration Memorandum; to1

influence international conferences, international legal tribunals, and world public opinion; to2

increase protection for the right to abortion in the United States; and to engage in open and free3

discussion about abortion. See Am. Compl. ¶¶ 7, 85, 88, 90, 91, 105-107. 4

Plaintiffs list several countries in which they currently have projects involving5

these activities and where foreign NGOs have agreed to the Standard Clause, id. ¶ 71, and they6

allege that all of these activities are significantly hindered in those countries. The use of the7

Standard Clause, according to plaintiffs, “prevents Plaintiffs from forming alliances with8

potential partner organizations in order to increase their abortion-related advocacy efforts’9

effectiveness.” Id. ¶ 100. One of the ways in which this problem manifests itself is by depriving10

plaintiffs of their audience for reproductive rights advocacy. Plaintiffs allege that the use of the11

Standard Clause “interferes with Plaintiffs’ conveyance of their ideas and political speech about12

abortion by chilling or prohibiting [foreign NGOs] from attending presentations given by13

Plaintiffs and from listening to Plaintiffs’ political advocacy.” Id. ¶ 106. These hindrances,14

according to plaintiffs, violate their right to freedom of speech and association. Similarly,15

plaintiffs allege that the challenged restrictions impede their ability to disseminate publications16

and reports “because [foreign NGOs] that would otherwise distribute the publications in foreign17

countries are prohibited or chilled from doing so.” Id. ¶ 103. Plaintiffs argue that this harm is18

actionable under Supreme Court precedent holding that “[t]he First Amendment protects19

[individuals’] right not only to advocate their cause but also to select what they believe to be the20

most effective means for so doing.” Meyer v. Grant, 486 U.S. 414, 424 (1988). Plaintiffs also21

invoke their right to receive information, claiming that the Standard Clause “interferes with22

Page 425: Sotomayor Cases

9

Plaintiffs’ ability to obtain information necessary to accomplish their abortion law reform efforts1

from USAID recipient [foreign NGOs],” and impedes plaintiffs’ access to victims and witnesses2

of human rights abuses related to reproductive issues. Am. Compl. ¶¶ 101-102. Plaintiffs3

explain that foreign NGOs are often the only vehicle to provide access to both general4

information and first-hand accounts regarding conditions in foreign countries, id., and that5

obtaining such information is necessary for domestic NGOs to fulfill their mission of advocating6

reproductive rights—including their ability to lobby the United States government, id. ¶ 108. 7

B. The Planned Parenthood Case8

We have been over this ground before. In Planned Parenthood Federation of9

America, Inc. v. Agency for International Development, 915 F.2d 59 (2d Cir. 1990), this Court10

entertained a constitutional challenge to the same Standard Clause incorporated by the USAID11

into financial assistance agreements with foreign NGOs. Like the instant case, Planned12

Parenthood involved a First Amendment challenge, based on freedom of speech and association,13

brought by domestic NGOs. As in the instant case, the plaintiffs argued before this Court that the14

Mexico City Policy effectively prevented them from associating and collaborating with foreign15

NGOs, which in turn prevented them from fulfilling their mission regarding reproductive rights16

advocacy. Planned Parenthood, 915 F.2d at 62-63. 17

This Court rejected the challenge on the merits, finding “no constitutional rights18

implicated” by the Policy and the Standard Clause. Planned Parenthood, 915 F.2d at 66. The19

Court reasoned that the domestic NGOs remained free to use their own funds to pursue abortion-20

related activities in foreign countries and that “[t]he harm alleged in the complaint is the result of21

choices made by foreign NGOs to take AID’s money rather than engage in non-AID funded22

Page 426: Sotomayor Cases

3 The Standard Clause as restored under President George W. Bush contains minoralterations from the original version challenged in the Planned Parenthood case. They are not ofsignificance here. The only substantive difference in the restored Standard Clause is that“treatment of injuries or illnesses caused by legal or illegal abortions” is now excluded from thedefinition of prohibited abortion-related activities. Restoration Memorandum, 66 Fed. Reg. at17,311. Other minor alternations include the change from “AID” to “USAID,” “grant” to“award,” and “birth spacing” to “child spacing.” CRLP, 2001 WL 868007, at *6 n.6.

10

cooperative efforts with plaintiffs-appellants.” Id. at 64. “Such an incidental effect” on the1

activities of the domestic NGOs, the Court held, did not rise to the level of a constitutional2

violation. Id. The Court concluded that “the Standard Clause does not prohibit plaintiffs-3

appellants from exercising their first amendment rights.” Id. Moreover, the Court explained that4

whatever one might think of the Mexico City Policy, “the wisdom of, and motivation behind, this5

policy are not justiciable issues,” and the Court found the restrictions to be rationally related to6

the “otherwise nonjusticiable decision limiting the class of beneficiaries of foreign aid.” Id. at7

64-65. Having rejected plaintiffs’ claims on the merits, this Court declined to address the8

question of whether plaintiffs had standing under Article III. Id. at 66. 9

Planned Parenthood not only controls this case conceptually; it presented the10

same issue. Planned Parenthood rejected the same First Amendment challenge to the same11

provision—the Standard Clause that was first instituted by President Reagan in the 1980s and12

was reinstated by President George W. Bush in 20013—and no intervening Supreme Court case13

law alters its precedential value.14

Plaintiffs’ attempts to distinguish Planned Parenthood are unavailing. First,15

plaintiffs argue that Planned Parenthood did not involve an equal protection challenge. This is16

true, but does not affect the First Amendment question. Second, plaintiffs argue that Planned17

Parenthood “mischaracterizes the [restrictions’] effect as ‘incidental.’” This argument does not18

Page 427: Sotomayor Cases

11

distinguish Planned Parenthood at all, but simply disagrees with its holding. Third, plaintiffs1

argue that the effect on their speech is more substantial than in Planned Parenthood because the2

provision “impedes Plaintiffs’ entire mission, not just one component of that mission.” The3

significance of this point is not clear to us as a legal matter and, in any event, the allegations4

made in the two cases are far too similar to support this distinction as a factual matter. 5

Finally, plaintiffs argue that Planned Parenthood “did not assess the right to6

obtain and impart information,” and that the litigants in Planned Parenthood “did not claim that7

their law reform advocacy in the United States and the United Nations was impeded.” By8

rejecting plaintiffs’ claim that the Mexico City Policy prevented them from associating and9

collaborating with foreign NGOs, however, this Court’s opinion in Planned Parenthood did, in10

fact, assess and reject the claim that plaintiffs’ right to obtain and impart information was11

impeded. See Planned Parenthood, 915 F.2d at 63-64 (noting and rejecting plaintiffs’ argument12

that “it is impractical for United States citizens or organizations to engage in abortion-related13

activities abroad without the cooperation of foreign organizations and that the Standard Clause14

deters ‘many of the most logical and effective foreign partners’”); Planned Parenthood Fed’n of15

Am., Inc. v. Agency for Int’l Dev., No. 87 CIV. 0248, 1990 WL 26306, at *5 (S.D.N.Y. Mar. 7,16

1990) (“[P]laintiffs argue that [the Standard Clause] has the effect of preventing foreign NGOs17

that receive AID funds and domestic NGOs from associating with each other for purposes of18

receiving or disseminating abortion information using non-U.S. government money . . . .”19

(emphasis added)). Likewise, although this Court’s opinion in Planned Parenthood did not20

explicitly describe the scope of plaintiffs’ claim regarding the restrictions on their law reform21

advocacy to include advocacy in the U.S. and in international tribunals, our holding clearly22

Page 428: Sotomayor Cases

12

contemplated and rejected that claim. See Planned Parenthood, 915 F.2d at 62 (noting that the1

Standard Clause does not hinder plaintiffs’ use of non-AID funds “in the United States or2

abroad”); Planned Parenthood, 1990 WL 26306, at *7 (“[Plaintiffs] also allege that the ‘reason3

for the promulgation of the policy and the Standard Clause was to advance the Reagan4

Administration’s effort to suppress pro-choice views and activities in the United States . . . and5

not for any purported concern with foreign policy . . . .’” (quoting complaint)).6

C. The Standing Issue7

The district court dismissed the instant case, not on the merits as we did in8

Planned Parenthood, but for lack of constitutional standing. A federal court has jurisdiction only9

if a claim presents a “case” or “controversy” under Article III of the U.S. Constitution. This10

“irreducible constitutional minimum” of standing requires (1) that the plaintiff has suffered an11

“injury in fact,” i.e., an invasion of a judicially cognizable interest which is concrete and12

particularized as well as actual or imminent, rather than conjectural or hypothetical; (2) that there13

is a causal connection such that the injury is fairly traceable to the challenged conduct; and (3)14

that it is likely, as opposed to merely speculative, that the injury will be redressed by a favorable15

decision. Lujan, 504 U.S. at 560-61. “Since this case remains at the pleading stage, all facts16

averred by the plaintiffs must be taken as true for purposes of the standing inquiry—as they must17

be for any other issue presented.” Lerman v. Bd. of Elections, 232 F.3d 135, 142 (2d Cir. 2000). 18

The district court held that plaintiffs failed to show these elements of standing. 19

See CRLP, 2001 WL 868007, at *8-*12. In reaching its conclusion, the court relied heavily on20

our analysis in Planned Parenthood. See, e.g., id. at *10 (“The Court of Appeals has already21

held that the government is within its constitutional authority in imposing restrictions or22

Page 429: Sotomayor Cases

13

conditions on the receipt of USAID funding by [foreign NGOs].”). In particular, the district1

court placed great weight on our statement in Planned Parenthood that the harm alleged by2

domestic NGOs is the result not of the Mexico City Policy itself, but of “choices made by foreign3

NGOs to take AID’s money rather than engage in non-AID funded cooperative efforts with4

plaintiffs-appellants.” Id. at *10, *11 (quoting Planned Parenthood, 915 F.2d at 64). Based on5

this language from Planned Parenthood, the district court found that “plaintiffs have failed to6

show that their alleged harms are caused by the challenged government policies.” Id. at *12. 7

It is not clear, however, that the district court’s reliance on Planned Parenthood is8

entirely justified in this context. We found in Planned Parenthood that the alleged harm suffered9

by domestic NGOs is attributable to independent decisions of foreign NGOs, but only for10

purposes of the merits of plaintiffs’ First Amendment claims. It does not necessarily follow that11

Planned Parenthood answers the question of causation with respect to constitutional standing. 12

One reason why Planned Parenthood might be deemed to resolve the standing13

question is that Planned Parenthood, though adjudicated on the merits, was decided on the14

pleadings. Thus, one could argue that this Court decided as a matter of law that the Mexico City15

Policy could not be deemed the legal “cause” of the alleged harm to domestic NGOs. Although16

this finding was used to form a different conclusion in Planned Parenthood—that plaintiffs’17

claims failed on the merits—it arguably could be employed in our standing analysis here. On the18

other hand, it could be argued that Planned Parenthood is not dispositive, particularly in light of19

an intervening Supreme Court case that clarified the causation aspect of the standing inquiry. In20

Bennett v. Spear, 520 U.S. 154, 167 (1997), plaintiffs argued that a Biological Opinion by the21

Fish and Wildlife Service influenced the Bureau of Reclamation to reduce the quantity of22

Page 430: Sotomayor Cases

14

irrigation water available to plaintiffs. Rejecting the government’s contention that plaintiffs1

lacked standing because the Bureau’s conduct constituted an “independent” act breaking the2

chain of causation under Lujan, the Supreme Court explained that “[t]his wrongly equates injury3

‘fairly traceable’ to the defendant with injury as to which the defendant’s actions are the very last4

step in the chain of causation.” Id. at 168-69. The Court stated that while “it does not suffice if5

the injury complained of is the result of the independent action of some third party not before the6

court . . . that does not exclude injury produced by determinative or coercive effect upon the7

action of someone else.” Id. at 169 (quotation marks and alterations omitted). Bennett can be8

read to support plaintiffs’ standing argument in the instant case. 9

We are thus faced with a situation of a sui generis nature, inasmuch as our10

conclusion depends in large part on how much weight one places on our language in Planned11

Parenthood—a case that analyzed essentially the same factual allegations as the instant case but12

in a somewhat different context. As Planned Parenthood does not, as the district court implied,13

resolve the standing issue conclusively, we are confronted with a novel question of Article III14

standing.15

D. The Steel Co. Case and Our Authority to Proceed to the Merits16

Because we believe that our decision in Planned Parenthood dooms plaintiffs’17

First Amendment claims on the merits, we must decide whether we should first address18

plaintiffs’ novel theory of constitutional standing with respect to these claims. 19

Between the time that we decided Planned Parenthood and the filing of the20

instant action, the Supreme Court issued a decision in which it criticized the practice whereby a21

court proceeds directly to the merits of a case while assuming arguendo that the plaintiffs have22

Page 431: Sotomayor Cases

15

constitutional standing to bring the suit. See Steel Co. v. Citizens for a Better Env’t, 523 U.S. 83,1

94-101 (1998). This practice, referred to by some courts as “hypothetical jurisdiction,” United2

States v. Troescher, 99 F.3d 933, 934 n.1 (9th Cir. 1996), was often used by federal courts3

seeking to avoid a difficult or novel issue of standing in favor of a relatively easy merits question. 4

In Steel Co., however, Justice Scalia explained that the determination of standing is a question of5

subject matter jurisdiction, and that a court lacks the authority to rule on a case unless it6

determines that jurisdiction exists. Steel Co., 523 U.S. at 94. “For a court to pronounce upon the7

meaning or the constitutionality of a state or federal law when it has no jurisdiction to do so is,8

by very definition, for a court to act ultra vires.” Id. at 101-02. Justice Scalia’s decision in Steel9

Co. commanded a five-Justice majority, although two of the five issued a concurring opinion,10

which expressed a more permissive view toward the practice of assumed jurisdiction. See id. at11

110-11 (O’Connor, J., concurring, joined by Kennedy, J.) (stating that “the Court’s opinion12

should not be read as cataloging an exhaustive list of circumstances under which federal courts13

may exercise judgment in reserving difficult questions of jurisdiction when the case alternatively14

could be resolved on the merits in favor of the same party”) (quotation marks and alterations15

omitted). This Court has heeded the admonitions of Steel Co., acknowledging that ordinarily we16

are not to assume the existence of jurisdiction in favor of reaching an “easier” merits issue. 17

Fidelity Partners, Inc. v. First Trust Co. of N.Y., 142 F.3d 560, 565 (2d Cir. 1998); see also In re18

Rationis Enters., Inc. of Panama, 261 F.3d 264, 267-68 (2d Cir. 2001) (citing Steel Co. rule).19

The Steel Co. majority opinion, however, discussed several previous Supreme20

Court decisions which, according to the Court, “must be acknowledged to have diluted the21

absolute purity of the rule that Article III jurisdiction is always an antecedent question.” Steel22

Page 432: Sotomayor Cases

4 See Joan Steinman, After Steel Co.: “Hypothetical Jurisdiction” in the FederalAppellate Courts, 58 Wash. & Lee L. Rev. 855, 862 (2001) (noting the Steel Co. Court’s“embrace, rather than disavowal,” of cases such as Norton and Avrech, both discussed infra).

16

Co., 523 U.S. at 101. Moreover, the Court chose not to state simply that, to the extent that1

previous cases might be read to permit assumed jurisdiction, those cases are overruled. Instead,2

the Court distinguished the cases on various grounds, thus leaving their precedential value3

intact.4 Two such cases are of particular relevance here.4

The first is Norton v. Matthews, 427 U.S. 524 (1976), in which the Court declined5

to address a jurisdictional issue and answered the merits question regarding whether certain6

aspects of the Social Security Act were unconstitutional. The Steel Co. Court distinguished7

Norton on the ground that, in Norton, “[w]e declined to decide th[e] jurisdictional question,8

because the merits question was decided in a companion case, Mathews v. Lucas, with the9

consequence that the jurisdictional question could have no effect on the outcome.” Steel Co.,10

523 U.S. at 98 (internal citation omitted). The Steel Co. Court explained that the outcome in11

Norton was “foreordained by Lucas” and thus “Norton did not use the pretermission of the12

jurisdictional question as a device for reaching a question of law that otherwise would have gone13

unaddressed.” Id.14

The Steel Co. Court also distinguished and did not overrule Secretary of Navy v.15

Avrech, 418 U.S. 676 (1974). The Court explained, “Avrech also involved an instance in which16

an intervening Supreme Court decision definitively answered the merits question.” Steel Co.,17

523 U.S. at 98-99. Avrech involved a constitutional challenge to a provision of the Code of18

Military Justice. When another case, Parker v. Levy, 417 U.S. 733 (1974), rejected a similar19

Page 433: Sotomayor Cases

5 The Steel Co. Court seems to acknowledge this when, after recognizing that cases suchas Norton and Avrech “have diluted the absolute purity of the rule that Article III jurisdiction isalways an antecedent question,” the Court urges that these cases do not support a rule that“enables a court to resolve contested questions of law when its jurisdiction is in doubt.” SteelCo., 523 U.S. at 101 (emphasis added). Moreover, a majority of the Justices in Steel Co. citedNorton approvingly for the proposition that a court may assume the existence of jurisdiction incertain circumstances. See id. at 110-11 (O’Connor, J., concurring, joined by Kennedy, J.); id. at111 (Breyer, J., concurring in part and concurring in the judgment); id. at 122 n.15 (Stevens, J.,concurring in the judgment, joined in relevant part by Souter, J.).

17

constitutional challenge to the same provision, the Court decided to dispose of Avrech on the1

merits, stating that it was “unwilling to decide the difficult jurisdictional issue which the parties2

have briefed.” Avrech, 418 U.S. at 677. The Avrech Court explained its rationale: “We believe3

that even the most diligent and zealous advocate could find his ardor somewhat dampened in4

arguing a jurisdictional issue where the decision on the merits is thus foreordained.” Id. at 678. 5

The Steel Co. Court thus distinguished Avrech, finding that the “peculiar circumstances” of6

Avrech did not permit the case to be cited for the more general proposition that any “easy” merits7

question may be decided on the assumption of jurisdiction. Steel Co., 523 U.S. at 99. 8

Thus, the majority opinion in Steel Co. appears to allow an exception to the rule9

against assuming the existence of standing in those “peculiar circumstances” where the outcome10

on the merits has been “foreordained” by another case such that “the jurisdictional question could11

have no effect on the outcome,” provided the court “d[oes] not use the pretermission of the12

jurisdictional question as a device for reaching a question of law that otherwise would have gone13

unaddressed.” Id. at 98;5 cf. Nippon Steel Corp. v. United States, 219 F.3d 1348, 1352-53 (Fed.14

Cir. 2000) (using the Steel Co. Court’s approval of Norton as authority to bypass a jurisdictional15

question and decide the merits in an “unusual situation” where the two issues are intertwined). 16

We find ourselves in largely the same situation as the Supreme Court found itself in Norton and17

Page 434: Sotomayor Cases

18

Avrech: plaintiffs in this case challenge a governmental provision (the use of the Standard1

Clause) as unconstitutional, and there is a controlling case in which this Court entertained and2

rejected the same constitutional challenge to the same provision. Our outcome on the merits is3

thus “foreordained” by Planned Parenthood. Under the Norton/Avrech exception acknowledged4

in Steel Co., we need not reach the academic question of Article III standing in this case.5

Our approach not only comports with the language of the Steel Co. majority6

opinion, but also advances the underlying rationale of Steel Co. and makes good sense as a7

constitutional matter. The concern of the Steel Co. majority was that deciding a case on the mere8

assumption of jurisdiction can lead to the rendering of advisory opinions in violation of Article9

III: “Hypothetical jurisdiction produces nothing more than a hypothetical judgment—which10

comes to the same thing as an advisory opinion, disapproved by this Court from the beginning.” 11

Steel Co., 523 U.S. at 101 (citations omitted). Turning to the instant case, we note that where the12

precise merits question has already been decided in another case by the same court, it is the13

adjudication of the standing issue that resembles an advisory opinion—the very concern that14

animates the Steel Co. rule. It would be ironic if, in our desire to avoid rendering an advisory15

opinion, we were to address a novel standing question in a case where the result is foreordained16

by another decision of this Court. See id. at 123-24 (Stevens, J., concurring in the judgment)17

(noting that by addressing a standing issue unnecessarily “the Court is engaged in a version of the18

‘hypothetical jurisdiction’ that it has taken pains to condemn”). We further note that the question19

of Article III standing is itself of constitutional dimensions, see id. at 124, and “the Supreme20

Court has for generations warned against reaching out to adjudicate constitutional matters21

unnecessarily,” Horne v. Coughlin, 191 F.3d 244, 246 (2d Cir. 1999). 22

Page 435: Sotomayor Cases

6 As plaintiffs’ claims based on customary international law are substantivelyindistinguishable from their First Amendment claims, they are dismissed on the same ground. We express no view as to whether those claims are otherwise viable.

19

We hold that where, as here, a governmental provision is challenged as1

unconstitutional, and a controlling decision of this Court has already entertained and rejected the2

same constitutional challenge to the same provision, the Court may dispose of the case on the3

merits without addressing a novel question of jurisdiction. The Supreme Court followed this4

approach in Norton and Avrech, and approved of those cases in Steel Co. Plaintiffs’ First5

Amendment claims are therefore dismissed for failure to state a claim.66

II. Due Process Claim: Lack of Prudential Standing 7

Because Planned Parenthood did not address due process claims brought by8

domestic NGOs in this context, we address the due process claim separately and dismiss it on the9

alternative ground of prudential standing. 10

“The doctrine of standing, which addresses the question of whether the plaintiff is11

entitled to have the court decide the merits of the dispute or of particular issues, embraces both12

‘constitutional’ and ‘prudential’ requirements.” Sullivan v. Syracuse Hous. Auth., 962 F.2d13

1101, 1106 (2d Cir. 1992) (quotation marks and brackets omitted). The constitutional14

requirements, derived from Article III, are the injury in fact, causation, and redressability15

elements set out by the Supreme Court in Lujan. On the other hand, “[t]he prudential16

requirements of standing have been developed by the Supreme Court on its own accord and17

applied in a more discretionary fashion as rules of judicial ‘self-restraint’ further to protect, to the18

extent necessary under the circumstances, the purpose of Article III.” Id. (internal citations19

omitted). Pursuant to the doctrine of prudential standing, a court must ask whether a plaintiff’s20

Page 436: Sotomayor Cases

20

claim rests on the legal rights of a third party, asserts only a generalized grievance, or asserts a1

claim that falls outside the zone of interests protected by the legal provision invoked. See Valley2

Forge Christian Coll. v. Ams. United, 454 U.S. 464, 474-75 (1982); In re Appointment of Indep.3

Counsel, 766 F.2d 70, 74 (2d Cir. 1985). Of particular concern in the instant case is “the4

requirement that a plaintiff’s complaint fall within the zone of interests protected by the law5

invoked,” Crist v. Comm’n on Presidential Debates, 262 F.3d 193, 195 (2d Cir. 2001) (quoting6

Allen v. Wright, 468 U.S. 737, 750-51 (1984)), coupled with the rule against asserting the rights7

of a third party. Plaintiffs’ claims do not fall within the “zone of interests” protected by the Due8

Process Clause. 9

Plaintiffs’ due process claim is based on their allegation that the challenged10

restrictions fail to give clear notice of what political speech, public education, and law reform11

activities they prohibit and that they encourage arbitrary and discriminatory enforcement. Am.12

Compl. ¶ 140. It is not the plaintiffs, however, who are allegedly left uncertain of their rights by13

unconstitutionally vague language in a government provision; it is the foreign NGOs who are14

allegedly left in this position. Plaintiffs’ harm is derivative of this due process-type harm, and15

their alleged injury (albeit an unactionable one) concerns First Amendment interests. Plaintiffs’16

allegation, simply put, is that the vague language of the Standard Clause causes the foreign17

NGOs to be overly cautious in avoiding interaction with plaintiffs, which in turn harms18

plaintiffs’ speech and association interests. On appeal, plaintiffs expressly acknowledge that19

“[t]his vagueness claim is premised on the [restrictions’] chilling effect on protected speech and20

association.” As plaintiffs do not assert a harm to their own interest in receiving due process of21

law, this is precisely the sort of claim that the prudential standing doctrine is designed to22

Page 437: Sotomayor Cases

21

foreclose. Plaintiffs cannot make their First Amendment claims actionable merely by attaching1

them to a third party’s due process interests. See Haitian Refugee Ctr. v. Gracey, 809 F.2d 794,2

809 (D.C. Cir. 1987) (explaining that because due process rights “do not protect a relationship”3

between a third party and a litigant, a plaintiff “could never have standing to challenge a statute4

solely on the ground that it failed to provide due process to third parties not before the court”). 5

Plaintiffs’ due process claim is therefore dismissed for lack of prudential standing. 6

III. Equal Protection Claim7

A. Plaintiffs Have “Competitive Advocate Standing”8

Plaintiffs argue that the district court failed to undertake a separate analysis of9

their Article III standing to bring an equal protection claim. Because we agree with plaintiffs that10

the case law regarding constitutional standing for equal protection claims is distinct, and because11

Planned Parenthood does not foreclose this claim on the merits, we address the question of12

Article III standing with respect to this claim. As the case law and the legal theories involved are13

quite different, this constitutional standing analysis does not inform the question on which we14

reserved judgment above regarding constitutional standing to bring the First Amendment claims. 15

We find that plaintiffs do have constitutional standing to bring an equal protection claim. 16

With respect to the equal protection claim, the relevant portion of the complaint17

reads:18

The [use of the Standard Clause] violates the Equal Protection component of the Fifth19Amendment to the United States Constitution because it prohibits plaintiffs from20associating with USAID-recipient [foreign NGOs] for the purpose of promoting abortion21law reform, but permits other United States citizens and residents to associate with22USAID-recipient [foreign NGOs] for the purpose of opposing abortion law reform, and,23more generally, permits association with USAID-recipient [foreign NGOs] for the24purpose of rendering speech opposed to abortion more effective.25

Page 438: Sotomayor Cases

22

Am. Compl. ¶ 138. On appeal, plaintiffs flesh out the equal protection claim by explaining that1

the use of the Standard Clause, “by prohibiting [foreign NGOs] from collaborating with2

Plaintiffs, denies Plaintiffs the opportunity to compete on an equal footing with opponents of3

abortion law reform.”4

Though plaintiffs do not employ the term, this argument is essentially a theory5

that this Court has dubbed “competitive advocate standing.” We have acknowledged the6

possibility that a plaintiff may have standing to bring an equal protection claim where the7

government’s allocation of a particular benefit “creates an uneven playing field” for8

organizations advocating their views in the public arena. In re United States Catholic9

Conference, 885 F.2d 1020, 1029 (2d Cir. 1989). In order to “satisfy the rule that he was10

personally disadvantaged,” a plaintiff must “show that he personally competes in the same arena11

with the party to whom the government has bestowed the assertedly illegal benefit.” Id.12

Plaintiffs have standing under this theory. CRLP is an advocacy organization that13

communicates its viewpoint regarding issues of abortion and reproductive rights, and it competes14

with anti-abortion groups engaged in advocacy around the very same issues. The Standard15

Clause has bestowed a benefit on plaintiffs’ competitive adversaries by rewarding their suppliers16

of information, the foreign NGOs, with government grants, while withholding those grants from17

suppliers of information who would deal with CRLP. This is precisely the type of situation that18

the doctrine of competitive advocate standing contemplates. See id.; cf. Adarand Constructors,19

Inc. v. Pena, 515 U.S. 200, 211 (1995) (finding, under the test for standing articulated in Lujan,20

504 U.S. at 560, that a non-minority subcontractor had standing to contest a government policy21

that gave a financial incentive to general contractors to give preference to minority22

Page 439: Sotomayor Cases

23

subcontractors in awarding subcontracts).1

B. The Equal Protection Claim is Without Merit2

Because this classification “neither proceeds along suspect lines nor infringes3

fundamental constitutional rights,” it must “be upheld against equal protection challenge if there4

is any reasonable state of facts that could provide a rational basis for the classification.” F.C.C.5

v. Beach Communications, Inc., 508 U.S. 307, 313 (1993); see also Weinstein v. Albright, 2616

F.3d 127, 140 (2d Cir. 2001). Here there can be no question that the classification survives7

rational basis review. The Supreme Court has made clear that the government is free to favor the8

anti-abortion position over the pro-choice position, and can do so with public funds. See Rust v.9

Sullivan, 500 U.S. 173, 192-94 (1991). Plaintiffs’ equal protection challenge is thus without10

merit.11

CONCLUSION12

For the reasons stated, we affirm the district court’s dismissal of this action,13

though on different grounds.14

Page 440: Sotomayor Cases

SOTOMAYOR, Circuit Judge, dissenting:

The central issue in this case is whether the ne exeat provision in the Hong Kong

custody order confers on either Mr. Croll or the Hong Kong court “rights of custody” within the

meaning of the Hague Convention on the Civil Aspects of International Child Abduction (“Hague

Convention” or “Convention”). The majority concludes that it does not and, therefore, that the district

court lacked jurisdiction to order Christina’s removal to Hong Kong. Interpreting the text of the

Convention in light of its object and purpose, and taking into account the relevant case law in this area,

I reach the opposite conclusion. In my view, the majority seriously misconceives the legal import of the

ne exeat clause and, in so doing, undermines the Convention’s goal of “ensur[ing] that rights of custody

. . . under the law of one Contracting State are effectively respected in the other Contracting States.”

Hague Convention, art. 1, done Oct. 25, 1980, T.I.A.S. No. 11670 at 4, 1343 U.N.T.S. 89, 98,

reprinted in 51 Fed. Reg. 10,494, 10,498 (1986), implemented by the International Child Abduction

Remedies Act (“ICARA”), 42 U.S.C. § 11601 et seq. I therefore respectfully dissent.

The Hague Convention seeks “to protect children internationally from the harmful

effects of their wrongful removal or retention and to establish procedures to ensure their prompt return

to the State of their habitual residence, as well as to secure protection for rights of access.” Hague

Convention, Preamble, 51 Fed. Reg. at 10,498. Significantly, the Convention draws a clear line

between “rights of custody” and “rights of access,” reserving the remedy of return solely for breaches of

the former. Compare Hague Convention, arts. 1, 3, id. (providing for the return of children removed

or retained in violation of custody rights), with Hague Convention, art. 21, id. at 10,500 (providing that

a party may petition for arrangements, short of the child’s return, to secure the effective exercise of

Page 441: Sotomayor Cases

2

access rights). In this regard, the majority correctly observes that “an order of return is available only

for wrongful removals or retentions, and removals or retentions are wrongful only if they are ‘in breach

of rights of custody.’” Ante at [11] (quoting Hague Convention, art. 3, 51 Fed. Reg. at 10,498)

(emphasis in original).

Article 3 of the Convention provides that the removal or retention of a child is

“wrongful” where:

(a) it is in breach of rights of custody attributed to a person, an institution or any otherbody, either jointly or alone, under the law of the State in which the child was habituallyresident immediately before the removal or retention; and

(b) at the time of removal or retention those rights were actually exercised, either jointlyor alone, or would have been so exercised but for the removal or retention.

Hague Convention, art. 3, 51 Fed. Reg. at 10,498 (emphasis added). Thus, Mr. Croll cannot succeed

in securing Christina’s return to Hong Kong unless he can demonstrate that her removal was “in breach

of rights of custody” and, furthermore, that at the time of Christina’s removal from Hong Kong, those

rights of custody “were actually exercised, either jointly or alone, or would have been so exercised but

for the removal.” Hague Convention, art 3, id. For the reasons discussed below, I conclude that

Christina’s removal to the United States was “wrongful” under the Convention because (1) it

constituted a “breach of rights of custody” jointly held by Mr. Croll and the Hong Kong court, and (2)

Mr. Croll would have exercised his custody rights under the ne exeat clause in the custody order but

for Christina’s removal from Hong Kong.

Page 442: Sotomayor Cases

1 The majority states that the custody order “confers the sole ‘custody, care andcontrol’” upon Ms. Croll. Ante at [2] (emphasis added); see also id. at [18] (stating that “the custodyorder awards custody solely to the mother”) (emphasis added). However, nowhere does the HongKong court use the word “sole” or “solely” in connection with Ms. Croll’s custody rights.

2 The Hong Kong custody order also provides that “[e]ither parent may request theImmigration Department not to issue passports allowing the said child to go abroad without his/herknowledge.”

3

I. Was Christina Removed from Hong Kong “in Breach of Rights of Custody”?

Under the terms of the Hong Kong custody order, Ms. Croll is vested with “[t]he

custody, care and control” of Christina,1 and Mr. Croll is vested with rights of “reasonable access.”

Particularly relevant to this case, however, is the order’s further grant of rights to Mr. Croll under the ne

exeat clause. The parties agree that under this provision, Ms. Croll may not remove Christina from

Hong Kong without the consent of either Mr. Croll or the Hong Kong court.2 In other words, the ne

exeat clause confers a veto power on Mr. Croll to block Christina’s international relocation, unless the

Hong Kong court explicitly approves such removal. In essence, the ne exeat clause endows Mr. Croll

with significant decisionmaking power: absent an order of the Hong Kong court to the contrary, he can

require that Christina remain in Hong Kong or, alternatively, he can use his veto power as leverage to

influence Ms. Croll’s selection of the destination country. Because Mr. Croll may not invoke the

Convention’s return remedy based on his “reasonable access” rights, the issue in this case is whether he

may secure Christina’s return under the Convention by virtue of his rights under the ne exeat clause.

Page 443: Sotomayor Cases

4

The majority mischaracterizes the issue as being a question of whether the ne exeat

clause “transmute[s] access rights into custody rights under the Convention.” Ante at [28]. Clearly, the

ne exeat clause works no such magic. In my view, the question presented is whether

the ne exeat clause—wholly independent of Mr. Croll’s access rights—confers “rights

of custody” under the Convention. The Convention’s text, object and purpose, as well as the relevant

case law in this area, convincingly direct an answer in the affirmative.

A. The Text, Object, and Purpose of the Convention

The critical interpretive challenge in this case involves the definition of “rights of

custody” as used in the Convention. The majority begins this undertaking by surveying a host of

American dictionaries to support its “intuition that custody is something other and more than a negative

right or veto.” Ante at [14]. Relying on these sources, the majority finds that the “custody of a child

entails the primary duty and ability to choose and give sustenance, shelter, clothing, moral and spiritual

guidance, medical attention, education, etc., or the (revocable) selection of other people or institutions

to give these things.” Ante at [15]. While traditional American notions of custody rights are certainly

relevant to our interpretation of the Convention, the construction of an international treaty also requires

that we look beyond parochial definitions to the broader meaning of the Convention, and assess the

“ordinary meaning to be given to the terms of the treaty in their context and in the light of [the

Convention’s] object and purpose.” Vienna Convention on the Law of Treaties, May 23, 1969, art.

31.1, 1155 U.N.T.S. 331, 340 (stating general rule on the interpretation of treaties); see

Page 444: Sotomayor Cases

5

also Restatement (Third) of Foreign Relations Law § 325 (1987) (same).

Contrary to the majority’s position that “[n]othing in the Hague Convention suggests

that the drafters intended anything other than this ordinary understanding of custody,” ante at [15], the

Convention and its official history reflect a notably more expansive conception of custody rights. The

report containing the official history and commentary on the Convention clarifies that “the intention [of

the Convention] is to protect all the ways in which custody of children can be exercised.” Elisa Pérez-

Vera, Explanatory Report to the Hague Conference on Private International Law, in 3 Acts and

Documents of the Fourteenth Session (Child Abduction) 426, para. 71 (1980) (emphasis in original)

(“Pérez-Vera Report”). This broad notion of custody rights is also consistent with Article 3, which

provides that “rights of custody” may arise from a variety of sources, including by “operation of law or

by reason of a judicial or administrative decision, or by reason of an agreement having legal effect under

the law of [the child’s country of habitual residence].” Hague Convention, art. 3, 51 Fed. Reg. at

10,498. In this way, the Convention plainly favors “a flexible interpretation of the terms used, which

allows the greatest possible number of cases to be brought into consideration.” Pérez-Vera Report,

para. 67. Consequently, in determining whether the rights arising under a ne exeat clause constitute

“rights of custody” under the Convention, I discern an intent of inclusion rather than exclusion, so as to

effectuate the drafters’ goal of making the treaty applicable to all possible cases of wrongful removal.

Page 445: Sotomayor Cases

6

Although the treaty does not generally define its legal terms, see Pérez-Vera Report,

para. 83, the risk that “an incorrect interpretation of [custody and access rights] would

. . . compromis[e] the Convention’s objects” led the drafters to include Article 5, which offers further

guidance on the meaning of the term “rights of custody.” See Pérez-Vera Report, para. 83. I note,

however, that the provision was left deliberately vague due to the drafters’ failure to agree on a more

precise definition. See Pérez-Vera Report, para. 84 (“[S]ince all efforts to define custody rights in

regard to [particular situations] failed, one has to rest content with the general description given [in the

text].”). Article 5 provides that:

For the purposes of this Convention –

(a) “rights of custody” shall include rights relating to the care of the person ofthe child and, in particular, the right to determine the child’s place ofresidence; . . .

Hague Convention, art. 5, 51 Fed. Reg. 10,498 (emphasis added); see also Pérez-Vera Report, para.

84 (noting that under Article 5, “rights of custody” include those rights relating to the care of the child,

and that the Convention seeks to clarify this otherwise general definition “by emphasizing, as an

example of the ‘care’ referred to, the right to determine the child’s place of residence.”). As I interpret

the Convention, rights arising under a ne exeat clause include the “right to determine the child’s place of

residence” because the clause provides a parent with decisionmaking authority regarding a child’s

international relocation. Thus the ne exeat clause vests both Mr. Croll and the Hong Kong court with

“rights of custody” for the purposes of the Convention. See Hague Convention, art. 5, 51 Fed. Reg. at

10,498.

Page 446: Sotomayor Cases

7

A parent’s ne exeat rights fit comfortably within the category of custody rights the

Convention seeks to protect. The Convention states at its outset that its object is, along with returning

children wrongfully removed from their habitual residence, “to ensure that rights of custody . . . under

the law of one Contracting State are effectively respected in the other Contracting States.” Hague

Convention, art. 1, 51 Fed. Reg. at 10,498. The Pérez-Vera report explains that

the problem with which the Convention deals . . . derives all of its legalimportance from the possibility of individuals establishing legal and jurisdictionallinks [in the new country] which are more or less artificial. In fact, resorting tothis expedient, an individual can change the applicable law and obtain a judicialdecision favourable to him. [Such a decision] bears a legal title sufficient to‘legalize’ a factual situation which none of the legal systems involved wished tosee brought about.

Pérez-Vera Report, para. 15. At its core, therefore, the Convention’s return remedy targets those

individuals who cross international borders, presumably in search of a friendlier forum, flouting the

custody law of the child’s home country in the process. See Blondin v. Dubois, 189 F.3d 240, 245-

46 (2d Cir. 1999) (describing the Convention’s purpose as “‘preserv[ing] the status quo and . . .

deter[ring] parents from crossing international boundaries in search of a more sympathetic court.’”)

(quoting Friedrich v. Friedrich, 983 F.2d 1396, 1400 (6th Cir. 1993)).

In light of the Convention’s broad purpose, the concept of “wrongful removal” clearly

must encompass violations of ne exeat rights. When a parent takes a child abroad in violation of ne

exeat rights granted to the other parent by an order from the country of habitual residence, she nullifies

that country’s custody law as effectively as does the parent who kidnaps a

child in violation of the rights of the parent with physical custody of that child.

Page 447: Sotomayor Cases

8

Moreover, where, as here, the parent seeks a custody order in the new country, she seeks to legitimize

the very action—removal of the child—that the home country, through its custody order, sought to

prevent. To read the Convention so narrowly as to exclude the return remedy in such a situation would

allow such parents to undermine the very purpose of the Convention.

B. The Majority’s Approach

In reaching the opposite conclusion, the majority contends that “rights of custody,” as

used in the Convention, refers to a “bundle of rights” of which a parent must possess a certain portion in

order to be protected by the Convention, and that possession of only one of those rights — in this case,

the “right to determine the child’s place of residence” by exercising and leveraging a veto power over

the child’s international relocation —is insufficient to confer custody on the party possessing that power.

See ante at [16]. In my view, however, the Convention’s definition of “rights of custody” contemplates

a bundle of rights that are protected regardless of whether a parent holds one, several or all such

custody rights, and whether the right or rights are held singly or jointly with the other parent. In fact, the

Convention expressly protects joint custody rights, see Hague Convention, art. 3, 51 Fed. Reg. at

10,498, which may assume a number of forms, including situations in which one parent possesses sole

physical custody of the child but shares certain decisionmaking authority with the other parent. The

Convention contains no indication that in such an arrangement, a parent must possess some minimum

number of rights of custody in order to qualify for protection.

The majority also maintains that a parent’s ne exeat right does not equate with Article

Page 448: Sotomayor Cases

3 To be sure, the right to prevent a child’s removal from her home country does notconstitute an absolute right “to determine the child’s place of residence.” That a right is limited,however, does not render it meaningless for purposes of the Hague Convention. See Pérez-VeraReport, para. 71 (characterizing “joint custody” as “dividing the responsibilities inherent in custody

9

5’s “right to determine the child’s place of residence” because the latter right necessarily entails

“specific choices” regarding the child’s living situation rather than simply decisions regarding the country

in which she lives. See ante at [16-18]. Like the majority’s definition of “custody,” however, this

conclusion ignores the basic international character of the Hague Convention. While such “specific

choices” certainly constitute facets of custody, the broader decision as to whether a child will live in

England or Cuba, Hong Kong or the United States, is precisely the kind of choice the Convention is

designed to protect. See Pérez-Vera Report, para. 56 (“Although the Convention does not contain any

provision which expressly states the international nature of the situations envisaged, such a conclusion

derives as much from its title as from its various articles. . . . [T]he international nature of the

Convention arises out of a factual situation, that is to say the dispersal of members of a family among

different countries.”). The Hague Convention provides a remedy not when a parent moves the child

from city to suburb or from home to boarding school, but when he or she transports the child across

national borders. In light of this international context, the term “place of residence,” as used in the

Convention, logically contemplates decisions regarding international relocation. Accordingly, the right

to choose the country in which a child lives, like the authority over the child’s more specific living

arrangements, constitutes a “right to determine the child’s place of residence” under Article 5,

and thus a “right of custody” under the Convention.3

Page 449: Sotomayor Cases

rights between both parents”). Furthermore, that a right is a veto or “negative right” does not diminishits status as a right. See, e.g., Cruzan v. Director, Missouri Dep’t of Health, 497 U.S. 261, 281(1990) (recognizing the due process right to refuse life-sustaining medical treatment).

10

The majority avoids this conclusion by asserting that the power to determine a child’s

country of residence “protects rights of custody and access alike, and [gives] no clue as to who has

custody.” Ante at [17]. But while such a power may have the effect of ensuring a parent’s reasonable

access, and in fact may be included in a custody order for precisely that purpose, ne exeat rights

circumscribe the choices of the parent with physical custody of the child in a way that “reasonable

access” rights do not. Absent a ne exeat clause, the international relocation of a child does not

necessarily violate the other parent’s access rights; the parents still may work out an arrangement that

satisfies the rights of “reasonable access” even across international borders. On the other hand, when a

parent expatriates her child without securing the necessary consent, she has, by definition, violated the

other parent’s ne exeat rights.

The majority also posits that the Convention would be “unworkable” if it provided the

return remedy for violations of a parent’s ne exeat rights. See ante at [20-21]. Because an order of

return can require only Christina, and not Ms. Croll, to return to Hong Kong, the majority claims that

“we cannot plausibly read the Convention to compel the removal of a child from a parent who exercises

all rights of care to a country in which no one has that affirmative power or duty.” Ante at [21]. The

majority mistakenly assumes that the custody order in a given case is the sole source of a parent’s rights

and duties vis-a-vis his or her child. To the contrary, a parent’s duty to care for a child, like his or her

Page 450: Sotomayor Cases

4 The majority faults the dissent for its “assumption” that a court will “alter custody rights”upon the child’s return, and raises fears that absent such alteration, the child will be uncared for uponher return to Hong Kong. See ante at [23]. Christina’s care upon her return is neither premised onassumptions nor relevant to the issue before us. First, it strains credulity to suggest that a father who, ashere, searches the world for his child to get her back and files a petition in a foreign forum in order todo so, would, upon succeeding in his efforts, simply permit his child to stand abandoned in the airportupon her return. The majority confuses physical care of a child with legally-ordered custody. Furthermore, if Christina’s care upon her return to Hong Kong were really a concern in the instant case,the appropriate remedy would not be reversal but a remand to the district court to assess the parties’intentions. See Feder v. Evans-Feder, 63 F.3d 217, 226 (3d Cir. 1995) (“[I]n order to ameliorateany short-term harm to the child, courts in the appropriate circumstances have made return contingentupon ‘undertakings’ from the petitioning parent.”); In re Walsh, 31 F. Supp. 2d 200, 207 (D. Mass.1998) (“Numerous courts granting petitions under the Convention have recognized the legitimacy ofexacting appropriate undertakings from the parents designed to ensure that the children will be caredfor properly during transit and that no harm will come to the children pending disposition in the countryof habitual residence.”) (citations omitted), aff’d in part, rev’d in part on other grounds sub nom. Walsh v. Walsh, __ F.3d __, 2000 WL 1015863 (1st Cir. July 25, 2000). Second, the issuesconcerning Christina’s custody upon her return are beyond the scope of the Convention, which dealssolely with returning a child to the country of habitual residence so that its courts, and not a foreigncourt in the country to which the child has been wrongfully removed, can adjudicate custody rights withrespect to that child. This concept underlying the Convention — that the child is best served byentrusting decisions regarding his or her custody to the courts of the child’s country of habitualresidence — stands in direct contradiction to the majority’s parochial view that foreign courts cannot betrusted in the same manner as American courts to competently make necessary decisions regarding thechild. See ante at [23] (“on this point the dissent is generalizing from local American law”).

11

rights of custody, may arise from many sources, including the law of the country of habitual residence.

That the custody order in this case granted “custody, care and control” of Christina to Ms. Croll,

therefore, does not direct the conclusion that Mr. Croll will have no responsibility to care for Christina

upon her return to Hong Kong.4 I therefore reject the majority’s dire forecast that ordering Christina’s

return, without Ms. Croll at her side, risks leaving Christina helpless in Hong Kong without parental

care.

Page 451: Sotomayor Cases

5 The majority quotes A.E. Anton, the former chairman of the Hague ConferenceCommission, who opines that “breach of a right simply to give or to withhold consent to changes in a

12

Moreover, the majority’s characterization of a return remedy for violations of ne exeat

rights as unworkable fails to account for the Convention’s protection of any number of joint custody

arrangements in which the parents trade physical custody or in which one parent possesses physical

custody and the other parent contributes to decisions about the child’s upbringing. By the majority’s

reasoning, were the parent with physical custody to remove the child from the country of habitual

residence, the court would have no power to return the child, because no adult would be required to

care for him or her upon return. Such a conclusion, however, would largely eviscerate the

Convention’s protection of joint custody rights.

Far from being unworkable, the application of the return remedy in the context of ne

exeat violations directly and fully advances the Convention’s goal of preventing parents from unilaterally

circumventing the home country’s custody law. In contrast to access right violation cases where

returning the child to her country of habitual residence would not itself guarantee the effective exercise

of such rights, ordering the return of a child based on a ne exeat violation will, in and of itself, give full

effect to a parent’s ne exeat rights.

Finally, the majority cites to a string of authorities under the caption “Intent of the

Drafters” to support its narrow reading of the Convention. See ante at [21-26]. With one exception,

these authorities stand only for the unremarkable proposition that under the Convention, the return

remedy is unavailable for breaches of parents’ access rights.5 In sum, those authorities shed no light on

Page 452: Sotomayor Cases

child’s place of residence is not to be construed as a breach of rights of custody in the sense of Article3.” Ante at [22] (quoting A.E. Anton, The Hague Convention on International Child Abduction, 30Int’l & Comp. L.Q. 537, 546 (1981)). Although Mr. Anton’s views support the majority’sinterpretation of the Convention, the majority neglects to emphasize that his article represents only hispersonal views and not the official legislative history of the Convention. See Anton, supra at 537 (“Thispaper, however, must not be taken to reflect any views other than those of the author.”). His article istherefore appropriately viewed as simply the opinion of one scholar. Other scholars, in contrast, haveconcluded that ne exeat rights do constitute “rights of custody” under the Convention. See, e.g., PaulR. Beaumont & Peter E. McEleavy, The Hague Convention on International Child Abduction 72-73 (1999) (“If an individual is a child’s sole custodian there are, prima facie, no restrictions on himrelocating with that child. If there are such restrictions, that implies that the custody right must in someway be limited. Where this is so it must be that another body or individual holds a corresponding rightin relation to the child. Therefore, should the primary carer remove the child abroad, . . . he wouldhave breached the custody rights of the other party, if their consent had not been sought.”) (emphasis inoriginal); Linda Silberman, “Custody Orders Under the Hague Abduction Convention,” in A NewVision for a Non-Violent World: Justice for Each Child (Proceedings of the 4th Biennial InternationalConference of the International Association of Women Judges, 1999), at 236 (arguing that in cases ofnon-removal order violations, “a failure to recognize the Convention remedy of return would beinconsistent with the careful compromise [between custody and access rights] that the Conventiondefinition [of custody rights] has put in place.”).

13

the issue relevant here, i.e., whether ne exeat rights constitute “rights of custody” for the purposes of

the Convention.

C. International Case Law

While not essential to my conclusion that ne exeat rights constitute “rights of custody”

under the Convention, I note that my analysis is consistent with the decisions of most foreign courts to

consider the issue. See generally Air France v. Saks, 470 U.S. 392, 404 (1985) (in construing the

terms of a treaty, “the opinions of our sister signatories [are] entitled to considerable weight.”) (quoting

Benjamins v. British European Airways, 572 F.2d 913, 919 (2d Cir. 1978)). Given the desirability

Page 453: Sotomayor Cases

14

of uniformity in treaty interpretation, see Denby v. Seaboard World Airlines, Inc., 737 F.2d 172, 176

n.5 (2d Cir. 1984), these cases lend support to my understanding of the Convention.

Most foreign courts addressing this question have interpreted the notion of “rights of

custody” broadly in light of the Convention’s purpose and structure. The Family Court of Australia, for

example, has characterized the “spirit of the Convention” as ensuring “that children who are taken from

one country to another wrongfully, in the sense of in breach of court orders or understood legal rights,

are promptly returned to their country so that their future can properly be determined within that

society.” In the Marriage of: Jose Garcia Resina Appellant/Husband and Muriel Ghislaine

Henriette Resina Respondent/Wife, Appeal No. 52, 1991 (Fam.) (Austl.), para. 26. Accordingly,

the court held that the custody order at issue—which provided reciprocal ne exeat rights for each

parent—created “rights of custody” in the otherwise non-custodial father.

The English Court of Appeal has employed a similarly broad reading of the Convention,

holding that Article 5 “may in certain circumstances extend the concept of custody beyond the

ordinarily understood domestic approach” so as to ensure “that within its scope [the Convention] is to

be effective.” C. v. C., [1989] 1 W.L.R. 654, 658 (C.A.) (Eng.). In C. v. C., the court ordered the

return of a child where an Australian order granted custody to the child’s mother, but also provided that

the father and mother would remain “joint guardians” and that neither parent could remove the child

from Australia without the consent of the other. See id. at 656. Interpreting the language of Article 5,

the court found that because the Australian custody order allowed the father to exercise a measure of

control over the child’s place of residence, the father possessed “custody rights” within the meaning of

Page 454: Sotomayor Cases

6 The majority attempts to distinguish C. v. C. as involving an order of joint guardianship,see ante at [27], but the C. v. C. court explicitly relied on the language of the ne exeat provision andnot the joint guardianship clause in determining that the father possessed “rights of custody” under theConvention. See [1989] 1 W.L.R. at 657-58 (“[The lower court judge] heard argument as to theeffect of . . . joint guardianship. . . . [Accordingly, t]he judge’s attention does not appear to have beensufficiently drawn to the effect on the definition in article 5 of the Convention of clause 2 of theNovember 1986 order, that neither parent should remove the child from Australia without the consentof the other.”).

15

the Convention:6

[T]he father had, in my judgment, the right to determine that the child shouldreside in Australia or outside the jurisdiction at the request of the mother. . . .[He has] some control over not only the child leaving the jurisdiction, but alsoas a place to which the child was going, and not only the country; for instance,to live in London under suitable circumstances. . . . The father does not havethe right to determine the child’s place of residence within Australia but has theright to ensure that the child remains in Australia or lives anywhere outsideAustralia only with his approval.

Id. at 658. The Israeli High Court of Justice, when presented with facts nearly identical to those in the

instant case, similarly interpreted “rights of custody” to encompass a parent’s rights under a non-

removal order. See C.A. 5271/92, Foxman v. Foxman (H.C. 1992) (Isr.) (finding that the

Convention’s definition of “custodial rights” should be “broadly construed,” so as to cover cases in

which parental consent is required before a child is taken out of the country); cf. C.A. 1648/92,

Tourna v. Meshulem (H.C. 1992) (Isr.) (finding “rights of custody” in a father who, by virtue of a joint

custody order, had authority to refuse consent to the child’s change in residence).

In addition to these cases, which address custody rights held by a parent with ne exeat

rights, the English Court of Appeal has also held that a court entering the custody order in the child’s

Page 455: Sotomayor Cases

7 It is not apparent to me how the majority turns this case — involving a veto power overinternational relocation possessed by a court and a parent — into a decision that requires return of thechild “whenever a court enters a custody order” that is violated, regardless of the terms of that order. See ante at [19]. This dissent has never suggested that Mr. Croll or the court would be entitled to anorder of return if they did not possess ne exeat rights, nor does B v. B or this dissent suggest that returnwould be required in the scenario posed by the majority of “expatriation in derogation solely of parentalrights of access.” See ante at [19].

8 Although the majority states that “we and the district court are the only courts in theUnited States,” ante at [9], to consider the issue before us, I note that two American courts have alsorecognized “rights of custody” in connection with custody orders containing a ne exeat clause. See

16

place of habitual residence may itself possess “rights of custody” under the Convention in certain

circumstances. See B. v. B., [1993] 2 All E.R. 144 (C.A.) (Eng.). The court in B. v. B. noted that

under Article 3 an “institution or other body” as well as an individual may hold custody rights and thus

concluded that the removal of the child by the parent with physical custody in breach of an interim

custody order conditioned on the child remaining within the jurisdiction violated the rights of both the

other parent and the court issuing the interim custody decree. See id. at 148-49. Echoing the

reasoning of C v. C, the court found that because the restriction required the parent with physical

custody to remain in the court’s jurisdiction, and thus impliedly gave the court and the parent without

physical custody the right to veto an international move, it vested both with the power to determine the

child’s residence. See id. at 148-49. The court therefore affirmed the order of return on the ground

that the child’s removal had been wrongful within the meaning of the Convention. See id. at 153.7

These cases reflect strong support among our sister signatories for the proposition that

“rights of custody” are implicated where a custody order vests either a parent or the court with the

power to block a parent with physical custody from deciding to expatriate her child.8 While there are

Page 456: Sotomayor Cases

David S. v. Zamira S., 151 Misc.2d 630, 574 N.Y.S.2d 429 (N.Y. Fam. Ct. 1991); Janakakis-Kostun v. Janakakis, 6 S.W.3d 843 (Ky. Ct. App. 1999), review denied (Dec. 9, 1999), petitionfor cert. filed, 68 U.S.L.W. 3595 (U.S. Mar. 8, 2000) (No. 99-1496). In my view, however, thesecases are of limited utility because they fail precisely to define “custodial rights” or to differentiate themfrom access rights.

17

several cases in other jurisdictions that are certainly in tension with this view, I find the reasoning in

those cases unpersuasive.

At least one French court has determined that a custody order requiring the mother to

raise her children in England and Wales did not create custodial rights in the father because such a

reading would infringe on the mother’s right to expatriate. See T.G.I. Periguex, Mar. 17, 1992,

Ministere Public v. Mme Y., D.S. Jur. 1992 (Fr.). However, the court in Mme. Y. did not address the

meaning of Article 5’s “right to determine the child’s place of residence” provision and instead focused

on the mother’s expatriation rights under the European Convention for the Protection of Human Rights

and Fundamental Freedoms. See id. at 315-16. But deciding a Hague Convention case on the ground

that the custodial parent must remain free to expatriate her child begs the crucial interpretive question of

who, for purposes of the Convention, are “custodial parents” in the first place. Nothing in the

Convention suggests that one parent’s right to expatriate overrides another parent’s rights of custody.

On the contrary, the paramount importance the Convention places on custodial rights suggests that

where custodial rights and expatriation rights conflict, the latter must yield to the former. To adopt

another reading would, among other difficulties, make the Convention’s protection of joint custody

impossible, as joint custody by its very nature limits each parent’s unilateral decisionmaking power,

Page 457: Sotomayor Cases

9 I also note that courts in France appear divided on this issue. See Martha Bailey,“Rights of Custody” Under the Hague Convention, 11 B.Y.U. J. Pub. L. 33, 40 (1997) (discussingFrench cases).

18

including his or her power to relocate to another country with the child. In my view, therefore, the legal

presumption against restrictions on expatriation answers little.9

Apart from the Mme. Y. decision, two other cases are in tension with the proposition

that ne exeat rights constitute “rights of custody” under the Convention. In two separate decisions, the

Canadian Supreme Court has suggested — in dicta — that the Convention’s “wrongful removal”

provision does not cover cases in which a parent acts in violation of an express provision in a custody

order granting ne exeat rights. In the first, Thomson v. Thomson, [1994] 119 D.L.R. 4th 253 (Can.),

the court ordered a child’s return based on an interim non-removal order in order to “preserve

jurisdiction in the Scottish court to decide the issue of custody on its merits in a full hearing at a later

date,” but noted in dicta that such a remedy would be unavailable for violation of a final non-removal

order because the purpose of such an order was simply to “ensure permanent access to the non-

custodial parent.” Id. at 281. In the second case, D.S. v. V.W. [1996] 134 D.L.R. 4th 481 (Can.), the

court held that a return remedy was not available under the Convention for violation of an implicit

removal restriction in a custody order, and — relying in part on the dicta in Thomson relating to

express provisions in permanent custody orders — stated that a violation of such an implicit restriction

would concern only access rights, not custodial rights. Id. at 501-06. However, the court nevertheless

ultimately upheld the lower court’s order of return on the alternate ground that such return was in the

Page 458: Sotomayor Cases

19

best interests of the child under Quebec domestic legislation. Id. at 516-17.

For the reasons explained above, supra I.A., I am unpersuaded by the argument that

ne exeat clauses in permanent non-removal orders relate solely to access rights, the view endorsed by

the Canadian Supreme Court. Nor do I consider significant the Canadian Supreme Court’s emphasis

on the distinction between interim and permanent custody orders. To be sure, a court issuing an interim

custody order has a strong interest in preventing a child’s removal before it has the opportunity to make

its final custody determination. But nothing in the Convention’s language or official history supports the

notion that this interest is any more important than the court’s interest in enforcing the final custody

order once issued. The dichotomy between an interim and permanent custody order is, therefore, for

the purposes of the Convention, a distinction without a difference.

I note also that while the D.S. decision to uphold the order of return was unanimous, six

(out of nine) justices expressed reservations regarding the opinion’s analysis of custodial rights and

obligations, see 134 D.L.R. 4th at 484, 518; see also Bailey, supra, at 49, thereby raising serious

doubts as to whether the opinion’s conception of ne exeat clauses in relation to the Convention truly

represents the rule in Canada. Scholars have also strongly criticized the Canadian interpretation of

custody rights under the Convention. See, e.g., Bailey, supra, at 42-50; Linda Silberman, “Custody

Orders Under the Hague Abduction Convention,” in A New Vision for a Non-Violent World: Justice

for Each Child (Proceedings of the 4th Biennial International Conference of the International

Association of Women Judges, 1999), at 235-240. Therefore, following what I consider to be the

more compelling reasoning of the English, Australian, and Israeli cases, I would join the courts of those

Page 459: Sotomayor Cases

10 Ms. Croll does not argue on appeal that Mr. Croll’s petition was defective underArticle 3(b) of the Convention. The majority reaches this issue sua sponte.

20

countries in finding that rights arising under a ne exeat clause constitute “rights of custody” for the

purposes of the Hague Convention.

II. Did Mr. Croll Or The Hong Kong Court “Actually Exercise” Ne Exeat Rights?

Apart from the central issue of whether ne exeat rights constitute “rights of custody”

under the Convention, the majority also holds that Mr. Croll’s petition fails to satisfy the requirement of

Article 3(b) of the Convention, which provides:

The removal or retention of a child is to be considered wrongful where . . . at the timeof removal or retention those rights [of custody] were actually exercised, either jointlyor alone, or would have been so exercised but for the removal or retention.

Hague Convention, art. 3(b), 51 Fed. Reg. at 10,498 (emphasis added). According to the majority,

“[t]he right conferred by the ne exeat clause is not one that Mr. Croll ‘actually exercised,’ and it is

circular to say that he would have exercised it but for Christina’s removal, because the right itself

concerns nothing but removal itself, and would never have been exercised had Mrs. Croll been content

to stay in Hong Kong during Christina’s minority.” Ante at [19] (emphasis in original).10 This

description mischaracterizes the right that a ne exeat clause creates.

The right given to Mr. Croll and the Hong Kong court by the ne exeat clause was the

authority to withhold or grant consent to removing Christina from Hong Kong. Had they refused to

grant Ms. Croll permission to take Christina to the United States, or even had they agreed to grant

Page 460: Sotomayor Cases

21

permission, they would have “actually exercised” the custody rights granted by the custody order.

While I agree that neither Mr. Croll nor the court did, in fact, “actually exercise” this right, it seems clear

to me that Ms. Croll’s conduct in removing Christina without the necessary consent was precisely what

prevented them from doing so. Because Ms. Croll deprived Mr. Croll and the court of the opportunity

to exercise their veto power by surreptitiously removing Christina from Hong Kong without first seeking

consent, the ne exeat right is one that “would have been so exercised” but for Christina’s unlawful

removal. Article 3(b) therefore poses no barrier to finding that Christina’s removal was wrongful under

the Convention.

For the foregoing reasons, I conclude that Christina’s removal from Hong Kong to the

United States was “wrongful” under the Convention because her removal (1) constituted a “breach of

rights of custody” jointly held by Mr. Croll and the Hong Kong court, and (2) Mr. Croll or the court —

or both — would have exercised their veto rights under the ne exeat clause but for Christina’s removal

from Hong Kong. Accordingly, I would affirm the district court’s decision to grant Mr. Croll’s petition

for an order of return.

Page 461: Sotomayor Cases

United States Court of AppealsFOR THE

SECOND CIRCUIT__________________

At a stated Term of the United States Court of Appeals for the SecondCircuit, held at the United States Courthouse, Foley Square, in the City of New York,on the day of two thousand,

______________________________________________

Lee N. Koehler,Plaintiff-Appellant,

v. 98-9624

Bank of Bermuda (New York) Ltd.Defendant-Appellee,

______________________________________________

A request for a vote as to whether the panel decision should be reconsidered sua sponte by the Court inbanc having been made by a judge of the Court, and a poll of the judges in regular active service havingbeen taken, a majority of the Court has voted not to reconsider the decision in banc. The mandate shalltherefore issue. Judges Leval, Calabresi and Sotomayor dissent.

FOR THE COURT: Roseann B. MacKechnie, Clerk

By:____________________________ Beth J. Meador,

Administrative Attorney

Page 462: Sotomayor Cases

USCA Order 4

UNITED STATES COURT OF APPEALS1FOR THE SECOND CIRCUIT2

3August Term, 19994

5(Argued: August 30, 1999 Decided: April 10, 2000)6

(On Reconsideration by the Court In Banc Dissent: September 28, 2000)78

Docket No. 98-9624910

____________________________________________________________________________1112

LEE N. KOEHLER, 1314

Petitioner-Appellant,1516

v.1718

THE BANK OF BERMUDA (NEW YORK) LIMITED, a New York Corporation, THE BANK 19OF BERMUDA LIMITED, a Bermuda Corporation, REEFS BEACH CLUB LIMITED, a 20Bermuda Corporation, and A. DAVID DODWELL, a Bermuda citizen, 21

22Defendants-Appellees.23

____________________________________________________________________________24252627

SOTOMAYOR, Circuit Judge, with whom Judge LEVAL concurs, dissenting from the denial of28rehearing in banc: 29

30Judge Calabresi dissents in a separate opinion.31

32Federal courts may, under their alienage jurisdiction, hear controversies between 33

“citizens of a State and citizens or subjects of a foreign state.” 28 U.S.C. § 1332(a)(2) (1994). Based34

upon a prior holding of this Court in Matimak Trading Co. v. Khalily, 118 F.3d 76 (2d Cir. 1997),35

cert. denied, 522 U.S. 1091 (1998), the panel in this case concluded that Bermuda corporations and a36

Bermuda citizen were not “citizens or subjects of a foreign state,” and, therefore, that a controversy37

involving such parties was not within the alienage jurisdiction of the federal courts. Koehler v. Bank of38

Bermuda (New York) Ltd., 209 F.3d 130, 139 (2d Cir. 2000). Because a rehearing in banc would39

Page 463: Sotomayor Cases

1 The British Overseas Territories (also referred to as “Dependent Territories”) include Anguilla,Bermuda, British Indian Ocean Territory, the British Virgin Islands, the Cayman Islands, the FalklandIslands, Gibralter, Montserrat, the Pitcairn Islands, Saint Helena and dependencies, South Georgia andthe South Sandwich Islands, and the Turks and Caicos Islands. See Brief Amicus Curiae of theGovernment of the United Kingdom of Great Britain and Northern Ireland in Support of MatimakTrading Co. as Petitioner for Writ of Certiorari at 6 n.5, Matimak Trading Co. v. Khalily (97-893)(hereinafter U.K. Matimak Brief).

Some of the British Overseas Territories have become important commercial centers. As of

3USCA Order 4

provide a much-needed opportunity for the full Court to reexamine the flawed and internationally1

troublesome position that corporations and individuals from territories of the United Kingdom do not fall2

within the alienage jurisdiction of the federal courts, I dissent from the denial of the petition for rehearing3

in banc. 4

5

I. 6

This is a question of “exceptional importance.” Fed. R. App. P. 35(a)(2). Its import7

reaches well beyond our government, to our relations with foreign nations, and the access of foreign8

entities and individuals to the federal courts. Both the Executive Branch and the government of the9

United Kingdom of Great Britain and Northern Ireland have asked that we reconsider the reasoning we10

employed in Matimak. This Circuit’s understanding of the scope of alienage jurisdiction is squarely in11

conflict with that of the other circuit courts that have addressed this question. When issues of such12

enduring significance are presented, I believe that the Court in banc should reexamine the merits of its13

conclusion to ensure that substantial numbers of individuals and corporations are not erroneously14

deprived of access to our federal courts. 15

The defendants in this case include Bermuda corporations and a Bermuda citizen. 16

Bermuda is not recognized by our State Department as an independent state. It is, rather, a “British17

Overseas Territory.”1 Essential to this case is the fact that despite the myriad ways in 18

Page 464: Sotomayor Cases

1997, 563 banks and 34,169 other companies were incorporated in the Cayman Islands, at least 8,224businesses were incorporated in Bermuda, at least 100,000 companies were incorporated in the BritishVirgin Islands, and 12,911 companies were incorporated in the Turks and Caicos. See id. at 10-11.

Several of these territories, including the Cayman Islands and Bermuda, are consideredsignificant tax havens. See Mark Baker, Lost in the Judicial Wilderness: The Stateless CorporationAfter Matimak Trading, 19 Nw. J. Int'l L. & Bus. 130, 132 n.8 (1998) (noting that the holding inMatimak adds an “element of unpredictability” to the world of tax structuring).

2 The Matimak decision has been extensively criticized by commentators. See III Finance Ltd. v.Aegis Consumer Funding Group, Inc., No. 99 Civ. 2579, 1999 WL 1080371, at *2 (S.D.N.Y.Nov. 30, 1999) (collecting sources).

4USCA Order 4

which the United Kingdom exercises dominion over Bermuda, British law terms Bermudan citizens and1

corporations “nationals,” but not “subjects,” of the United Kingdom. See United Kingdom2

Government’s Diplomatic Service Procedure Manual, Vol. 7, Annex 1, Rules 1(b), 2(a) (1996). 3

Previously, this Court in Matimak held that a corporation organized under the laws of Hong Kong4

could not sue New York defendants in federal court because Hong Kong was, at the time, a5

Dependent Territory of the United Kingdom, and therefore the plaintiff corporation was not a “citizen6

or subject” of a foreign “state.”2 Relying on Matimak, the panel here concluded that Bermuda7

corporations and a Bermuda citizen were not “citizens or subjects of a foreign state,” 28 U.S.C. §8

1332(a)(2) (1994), and therefore not within our alienage jurisdiction. 9

The people of Bermuda would undoubtedly be surprised to learn that they are10

“stateless.” But this is precisely the conclusion upon which these decisions rest. See Matimak, 11811

F.3d at 86 (“[Plaintiff-Hong Kong corporation] is thus stateless. And a stateless person–the proverbial12

man without a country–cannot sue a United States citizen under alienage jurisdiction.”). Having found13

such entities or individuals “stateless,” the panels in this case and in Matimak had no difficulty denying14

these litigants access to the federal courts because “[t]he raison d’etre of alienage jurisdiction is to avoid15

entanglements with other sovereigns that might ensue from failure to treat the legal controversies of16

Page 465: Sotomayor Cases

3 See U.K. Matimak Brief at 9 (“The United Kingdom is keenly concerned that the citizens andcorporations of its Dependent Territories be able to bring and defend suits in neutral foreign foraconcerning their global commerce.”); Brief Amicus Curiae of the Government of the United Kingdomof Great Britain and Northern Ireland in Support of Appellant at 2, III Finance Ltd. v. AegisConsumer Funding Group (No. 00-7016) (hereinafter U.K. Aegis Brief) (“The United KingdomGovernment submits that it would not be in the interest of its trading relationship with the United Statesfor the corporations of the United Kingdom Overseas Territories to be excluded from United Statesfederal courts.”); Diplomatic Note No. 13/2000 from the British Embassy in Washington, D.C. to theUnited States Department of State, Feb. 2, 2000 at 1 (“The United Kingdom [] views with greatconcern the potential application of the Matimak rationale to individual Overseas Territories residents,as well as to commercial enterprises.”) .

5USCA Order 4

aliens on a national level.” Matimak, 118 F.3d at 82 (internal quotation omitted). These panels1

implicitly reason that absent a “state,” there is no sovereign to offend and therefore no cause to provide2

federal alienage jurisdiction. 3

This assurance is undermined by the strong reaction to our decisions by the United4

Kingdom.3 Whatever other intention the panels here and in Matimak may have had, there can be no5

doubt that the fundamental purpose of alienage jurisdiction--to void offense to foreign nations--is6

frustrated by the Matimak decision and its further application by this panel. Paradoxically, the country7

we offend by these holdings is not only a strong ally, but the very country the drafters of the alienage8

jurisdiction provision had in mind more than two hundred years ago when they sought to open the9

federal courts to foreign litigants. See Kevin R. Johnson, Why Alienage Jurisdiction? Historical10

Foundations and Modern Justifications over Disputes Involving Noncitizens, 21 Yale J. Int’l L. 1,11

7-8 (1996) (noting the failure of state courts to enforce debts owed to British creditors following the12

Revolutionary War). 13

This Court, in Matimak, attempted to shift responsibility for the disturbing14

consequences of its reasoning to the Executive Branch. Because the Department of State maintains that15

British Overseas Territories are not independent “states,” the Matimak court reasoned that it was16

Page 466: Sotomayor Cases

6USCA Order 4

forced to conclude that Bermuda corporations were stateless. See Matimak, 118 F.3d at 831

(commenting that “it is for the Executive Branch, not the courts, to anticipate where potential2

‘entanglements’ with such entities are appreciable enough to recognize sovereign status”). The3

Executive Branch, however, has urged us not to use the definition of “statehood” taken from the4

context of diplomatic recognition as a basis for denying British Overseas Territories the benefit of5

federal alienage jurisdiction. The Executive Branch has emphasized that to do so may cause the United6

States to “face an international controversy with British authorities for failure to provide a neutral forum”7

for individuals or corporations of a British Overseas Territory in federal court. Brief Amicus Curiae for8

the United States at 8, Matimak Trading Co. v. Khalily (96-9117). 9

Our Circuit is alone in concluding that federal alienage jurisdiction does not extend to10

citizens and corporations of British Overseas Territories. The Third Circuit, largely out of deference to11

the Executive Branch’s position that Hong Kong corporations were considered, at the time, “subject to12

British sovereignty,” found that they fell within the federal courts’ alienage jurisdiction. Southern Cross13

Overseas Agencies, Inc. v. Wah Kwong Shipping Group Ltd., 181 F.3d 410, 413 (3rd Cir.1999). 14

The Seventh Circuit has held that a Cayman Islands corporation could be sued in federal court under15

alienage jurisdiction, explaining that, “[c]ertainly, the exercise of American judicial authority over the16

citizens of a British Dependent Territory implicates this country’s relationship with the United17

Kingdom–precisely the raison d’etre for applying alienage jurisdiction.” Wilson v. Humphrys18

(Cayman) Ltd., 916 F.2d 1239, 1243 (7th Cir.1990), cert. denied, 499 U.S. 947 (1991). The19

Fourth Circuit, without discussion of the issue, has found that a Bermuda resident–apparently the same20

individual defendant sued in this case--was a “citizen” or “subject” of a foreign state for alienage21

jurisdiction purposes. Koehler v. Dodwell, 152 F.3d 304, 308 (4th Cir. 1998). 22

Owing to the fact that our characterization of corporations and citizens of British23

Page 467: Sotomayor Cases

7USCA Order 4

Overseas Territories as “stateless” has given rise to precisely the sort of damage to foreign relations the1

statute was meant to avoid, it is questionable that this distinction has its origins in the statute. Nor is this2

dubious characterization imposed upon us by the Executive Branch, which has advocated a contrary3

rule. Neither has this distinction been accepted by our sister circuits. Moreover, this Circuit has4

previously concluded, albeit without discussion, that “[t]here is no question” that alienage jurisdiction5

existed between citizens of the United States and a Bermuda corporation. Netherlands Shipmortgage6

Corp. v. Madias, 717 F.2d 731, 735 (2d Cir. 1983). Finally, two respected senior circuit judges from7

the panel in this case, Judge Jon O. Newman and Judge Richard J. Cardamone, have expressed8

disagreement with the merits of our precedent in Matimak. See Koehler v. Bank of Bermuda (New9

York) Ltd., __ F.3d __, __ n. _ (2d Cir. 2000). All this being the case, it seems incumbent upon us, as10

a full Court, to reexamine the basis upon which our panels both here and in Matimak reached their11

conclusions. 12

II.13

An examination of the merits leads to the conclusion that Matimak misapplied the terms14

“citizens or subjects of a foreign state” in a fashion inconsistent with both the historical understanding of15

these terms and a contemporary understanding of the relationship between the United Kingdom and its16

Overseas Territories.17

The panel in Matimak began its analysis with the unremarkable proposition that “a18

foreign state is entitled to define who are its citizens or subjects.” Matimak, 118 F.3d at 85 (citing,19

inter alia, United States v. Wong Kin Ark, 169 U.S. 649 (1898)). The court then concluded that a20

British Overseas Territory corporation did not fall within the scope of alienage jurisdiction because21

British law did not designate the corporation a “citizen” or “subject” of the United Kingdom or indicate22

Page 468: Sotomayor Cases

4 Aside from the substantial authority cited for the proposition that a foreign state determines its owncitizenship and nationality law, and for the relationship between the terms “citizen” and “subject,” theMatimak opinion contains little authority to support its analysis of the “citizenship” or “subjecthood” ofcorporations in British Overseas Territories. See Matimak, 118 F.3d at 85-86. One unpublisheddistrict court opinion is cited to support the suggestion that the corporate law of the Cayman Islands,another British Overseas Territory, is “clearly independent from the United Kingdom’s [law].” See id.at 86 (citing St. Germain v. West Bay Leasing, Ltd., No. 81-CV-3945 (E.D.N.Y. Sept. 30, 1982)). Another district court opinion from 1979 is cited to argue that corporations formed in Hong Kong werenot given the benefit of British nationality. See id. (citing Windert Watch Co. v. Remex Elecs. Ltd.,468 F. Supp. 1242 (S.D.N.Y. 1979)). Although the Matimak opinion cites also to a leading treatise,the same treatise currently reaches the opposite conclusion from the panel. See 15 James Wm. Moore,et al., Moore’s Federal Practice § 102.76 (3d ed., 1999) (“A citizen of a British dependent territory isdeemed to be a citizen of the United Kingdom and its Overseas Territory. Consequently, federalcourts may properly invoke diversity jurisdiction over suits in which a citizen of the Cayman Islands orBermuda is a party.”).

The cases cited by the Matimak court in support of the proposition that a stateless personcannot sue a United States citizen in federal court regard an individual whose citizenship has beenrevoked by a sovereign and nowhere suggest that a British Overseas Territory’s people or corporationscould exist in a condition of perpetual statelessness. See Matimak, 118 F.3d at 86.

8USCA Order 4

that the corporation was under the control of the United Kingdom. Matimak, 118 F.3d at 85-6.4 1

None would argue with the notion that a foreign state is entitled to define what persons2

or entities fall into its categories of “citizen “or “subject,” or any other of a variety of legal forms that3

exist under its own domestic immigration, nationality, and commercial law. The domestic meaning that4

any particular country may give to the terms “citizen” or “subject” does not, however, bind our courts in5

determining whether an individual or entity falls within the statutory meaning of such terms as provided6

by our law of alienage jurisdiction. The wide disparity in meaning that exists among countries7

concerning such terms requires that our alienage jurisdiction be determined not according to the8

appearance of the words “citizen” or “subject” (or translation thereof) in the pages of a country’s9

domestic code, but according to whether United States law deems such persons or entities to be10

“citizens or subjects” under our Constitution and statutes for the purpose of alienage jurisdiction. To11

proceed otherwise would be to “allow foreign law to deny privileges afforded under the Constitution . .12

Page 469: Sotomayor Cases

5 This task is analogous to that of deciding the state of domicile of a party in a diversity action infederal court. See 28 U.S.C. § 1332(a)(1) (1994) (providing federal jurisdiction for suits between“citizens of different States”). While a court may look to state law definitions of domicile and statecitizenship for guidance, “[d]etermination of a litigant’s state of domicile for purposes of diversity iscontrolled by federal common law, not by the law of any state.” 15 James Wm. Moore, et. al.,Moore’s Federal Practice § 102.34[3][a] (3d ed., 1997).

6 The legislative debates concerning the Judiciary Act of 1789 referred to the alienage jurisdictionprovision as providing access to the federal courts for “foreigners” or “aliens.” See 1 Annals ofCongress (1st Cong.) 810, 814, 825 (Joseph Gales ed., 1834) (House debates).

9USCA Order 4

. [and perhaps] unintentionally promote discrimination against certain classes of people or entities.” 1

Matimak, 118 F.3d at 89-90 (Altimari, J., dissenting).5 2

As an historical matter, the drafters of the Constitution chose the words “citizens” or3

“subjects” to refer to the broad category of those under the authority of a foreign power. See Bank of4

the United States v. Deveaux, 9 U.S. (5 Cranch) 61, 87 (1809) (Marshall, C.J.) (recognizing that the5

Constitution “established national tribunals for the decision of controversies between aliens and a citizen6

[of the United States]”), overruled in part on other grounds by Louisville, Cincinnati & Charleston7

R.R. Co. v. Letson, 43 U.S. (2 How.) 497 (1844). The Judiciary Act of 1789 used the word “alien”8

apparently as an equivalent term to “citizens” or “subjects” in the first rendering of the statutory grant of9

authority to exercise federal alienage jurisdiction. Compare U.S. Const. art. III, sec. 2, cl. 110

(extending jurisdiction to controversies “between a State, or the Citizens thereof, and foreign States,11

Citizens or Subjects”) with Judiciary Act of 1789, ch. 20, § 11, 1 Stat. 73, 78 (extending jurisdiction12

to suits in which “an alien is a party”).6 Oliver Ellsworth, the principal architect of the Judiciary Act of13

1789 that contained the alienage jurisdiction provision, referred to the need to provide a federal forum14

for controversies between United States citizens and “foreigners.” See Charles Warren, New Light on15

the History of the Federal Judiciary Act of 1789, 37 Harv. L. Rev. 49, 60 (1932) (quoting Letter of16

Oliver Ellsworth to Judge Richard Law, Apr. 30, 1789). “[T]he Framers often referred to [non-U.S.]17

Page 470: Sotomayor Cases

7 At the time the Constitution was written and the first alienage jurisdiction statute was enacted, theterm “subject” referred to a person who lived under the control of another. See Samuel Johnson, ADictionary of the English Language (1755) (defining a “subject” as “[o]ne who lives under the dominionof another”). See also 2 Noah Webster, American Dictionary of the English Language at 84 (1st ed.,1828; facsimile ed. Foundation for American Christian Education 1985) (defining a “subject” as “[o]nethat owes allegiance to a sovereign and is governed by his laws. The natives of Great Britain aresubjects of the British government. The natives of the United States, and naturalized foreigners, aresubjects of the federal government. Men in free governments are subjects as well as citizens; ascitizens, they enjoy rights and franchises; as subjects, they are bound to obey the laws.”) (emphasis inoriginal); 2 James Kent, Commentaries on American Law 258 n.b (6th ed., 1848) (“Subject andcitizen are, in a degree, convertible terms as applied to natives; and though the term citizen seems to beappropriate to republican freemen, yet we are equally with the inhabitants of all other countries,subjects, for we are equally bound by allegiance and subjection to the government and law of theland.”) (emphasis in original).

10USCA Order 4

citizens, subjects and foreigners interchangeably,” and “while foreign modes of government are hardly1

‘technicalities’ in any other sense, the Framers apparently did not consider them relevant to the exercise2

of federal jurisdiction.” Southern Cross Overseas, 181 F.3d at 416 (internal quotation marks and3

citations omitted).7 4

In 1875, the alienage jurisdiction provision was amended, replacing the term “alien”5

with the current reference to “citizens” or “subjects.” Act of Mar. 3, 1875, 18 Stat. 470, 470. This6

change, causing the statute to mirror the language of the Constitution, was motivated by the need to7

clarify that an alien could not sue another alien in federal court, and not from dissatisfaction with the8

original statutory term “alien” as impermissibly broader than the terms “citizens” or “subjects” found in9

the Constitution. See Johnson, 21 Yale J. Int’l L. at 21. 10

Although early cases did not explore the precise boundaries of the terms “citizen” and11

“subject” as used in alienage jurisdiction, the Supreme Court did have the opportunity to interpret these12

same terms in other contexts. Their general use confirmed that these terms referred to a range of13

relationships characterized by the acceptance of the authority and protection of a sovereign and an offer14

of allegiance. In 1830, Justice Story, addressing the issue of United States citizenship for expatriates15

Page 471: Sotomayor Cases

8 Justice Story continued, “ [t]wo things usually concur to create citizenship; first, birth locally withinthe dominions of the sovereign; and secondly, birth within the protection and obedience, or in otherwords, within the ligenance of the sovereign. That is, the party must be born within a place where thesovereign is at the time in full possession and exercise of his power, and the party must also at his birthderive protection from, and consequently owe obedience or allegiance to the sovereign, as such, defacto.” Inglis, 28 (3 Pet.) U.S. at 155.

9 The characterization of such corporations as “stateless” by the Matimak court is particularlyjarring considering that corporations are creations purely of law, and, unlike individuals, exist onlythrough an exercise of sovereignty. See Matimak, 118 F.3d at 89 (Altimari, J., dissenting) (“Astateless corporation is an oxymoron.”).

11USCA Order 4

noted that “[t]he rule commonly laid down in the books is, that every person who is born within the1

ligeance of a sovereign is a subject; and, e converso, that every person born without such allegiance is2

an alien.” Inglis v. Trustees of the Sailors' Snug Harbour, 28 U.S. (3 Pet.) 99, 155 (1830) (Story,3

J).8 In construing the terms of the Spanish Treaty of 1795, the Supreme Court in The Pizzaro, 154

U.S. (2 Wheat) 227 (1817), rejected the claim that the term “subject” in the treaty applied “only to5

persons who, by birth or naturalization owe a permanent allegiance to the Spanish government,” holding6

more simply that, “in the language of the law of nations . . . a person domiciled in a country, and7

enjoying the protection of its sovereign, is deemed a subject of that country.” Id. 245-46. 8

9

It has long been established that “a corporation created by the laws of a foreign state10

may, for the purposes of suing and being sued in the courts of the Union, be treated as a ‘citizen’ or11

‘subject’ of such a foreign state.” National Steam-Ship Co. v. Tugman, 106 U.S. 118, 121 (1882). 12

The defendant-Bermuda corporations in this suit were created under the laws of two different13

countries–Bermuda and the United Kingdom–but under the laws of only one recognized “state,” the14

United Kingdom.9 While the Bermuda’s Companies Act of 1981 provides procedures for15

incorporating companies in Bermuda, 6 Revised Laws of Bermuda, Title 17, Item 5, Part II (1989 &16

Page 472: Sotomayor Cases

10 Similarly, a corporation formed under the local Company law of Northern Ireland is not a“British” company in the sense of being formed under the British Companies Act 1985, but isnevertheless regarded by the British government as a national of the United Kingdom and thereforewithin the scope of § 1332(a)(2). See U.K. Aegis Brief at 11.

11 Bermuda was permitted to draft a constitution by the British Parliament pursuant to the BermudaConstitution Act 1967, 7 Halsbury’s Statutes of England and Wales, Bermuda Constitution Act 1967(4th ed., 1999 reissue), which can be revoked by an Act of the British Parliament. 6 Halsbury’s Lawsof England, para. 1042 (4th ed. reissue, 1992). Bermuda’s government is administered by a governorappointed by the Crown, id. at para. 994, who has the power to adjourn or dissolve Bermuda’slegislative assembly. Id. at para. 1000. An act dissolving the legislative assembly is deemed to be anexecutive act of the Queen. Id. at para. 1024 & n.1. The United Kingdom maintains supreme controlover Bermuda’s external relations and national defense, id. at para. 983, and has the power to alterBermuda’s boundaries. Id. at para. 992.

12 This conclusion corresponds to the position taken by the Department of State, see, e.g., Letter ofLinda Jacobson, Assistant Legal Adviser of the Department of State to Alan W. Dunch (submitted inthe Koehler litigation) (“[I]t is the position of the United States . . . that Bermuda residents andcorporations are subjects of a foreign state, i.e., Great Britain, for purposes of the federal diversitystatute, 28 U.S.C. § 1332.”); Southern Cross, 181 F.3d at 417 (citing Department of State’s view

12USCA Order 4

Update 1996), Bermuda and its government exist “under the sovereignty of the Crown.” 6 Halsbury’s1

Laws of England, para. 803 (4th ed. reissue, 1992).10 All authority to make laws for the “peace, order,2

welfare and good government” of Bermuda is granted to the Bermuda legislature solely by the United3

Kingdom, which regulates Bermuda lawmaking. Id. at para. 1027.11 4

Despite this, the panel in this case stated simply that “[b]ecause Bermuda is also a5

British Dependent Territory, Matimak governs . . . [and t]he district court therefore lacked subject6

matter jurisdiction over the state law claims against the Bermuda defendants.” Koehler, 209 F.3d at7

139. In such cases, when our Department of State determines that a country is not a sovereign state,8

the more reasonable conclusion is not that its corporations are “stateless,” but rather that they are9

subject to some other sovereign. Dependent upon the law of the United Kingdom, Bermuda10

corporations exist under the sovereignty of the United Kingdom. They are, for purposes of 28 U.S.C.11

§ 1332(a)(2), “subjects” of the United Kingdom.1212

Page 473: Sotomayor Cases

that “since the ultimate sovereign authority over [a Hong Kong corporation was] the British Crown, [it]should be treated as a subject of United Kingdom sovereignty for purposes of alienage diversityjurisdiction.”), the Department of Justice, see, e.g., Matimak, 118 F.3d at 86 (“The JusticeDepartment concludes that because the ultimate sovereign authority over the plaintiff is the BritishCrown, Matimak should be treated as a subject of United Kingdom sovereignty for purposes of §1332.”), and the British government, see, e.g., U.K. Matimak Brief at 7 (“Corporations of the BritishDependent Territories should be considered ‘subjects’ of the United Kingdom for purposes of thealienage jurisdiction of 28 U.S.C. § 1332.”); U.K. Aegis Brief at 4 (“The position of the UnitedKingdom Government is that entities incorporated in any territory for which the United Kingdom isinternationally responsible are regarded by the United Kingdom Government as United Kingdomnationals and, therefore, are “citizens of subjects” of the United Kingdom for purposes of alienagejurisdiction.”).

13 The reasoning of Matimak applied to all foreign corporations would produce an absurd result. The term “national” is often used instead of “citizen” or “subject” to describe the identify of a foreigncorporation. See Restatement (Third) of Foreign Relations Law § 213 (1987) (“For purposes ofinternational law, a corporation has the nationality of the state under the laws of which the corporation isorganized.”). If indeed courts must rely solely on the words found in the domestic codes of othercountries and there we discover that corporations are referred to only as “nationals” and not as“subjects” or “citizens” of a particular country, the Matimak analysis would force us to deny themaccess to the federal courts.

13USCA Order 4

The people of Bermuda, because they live under the sovereignty of the United1

Kingdom, are “citizens or subjects” of the United Kingdom for purposes of alienage jurisdiction. The2

individual defendant in this case, like other Bermudians, is a national of the United Kingdom for3

purposes of its own laws. The status of being a “national” of the United Kingdom, conferred by virtue4

of birth in a British Overseas Territory, fits comfortably within the original meaning of “citizen” or5

“subject” for the purposes of alienage jurisdiction. The United Kingdom continues to function as6

sovereign over Bermuda, ruling over its affairs, protecting it, and enjoying the allegiance of its citizens. 7

The narrowness with which the panels in this case and in Matimak attempted to apply the terms8

“citizen” and “subject” is uncharacteristic of the history of their use and the principles underlying their9

adoption.13 10

CONCLUSION11

Page 474: Sotomayor Cases

14USCA Order 4

Alienage jurisdiction was established by our Constitution and early statutes to1

strengthen our relations–particularly our commercial relations--with foreign nations. The 2

importance of these goals has only increased with time as both international relations and global trade3

have become more complex and our nation has assumed a central role in both. Having deprived a4

considerable number of foreign entities and individuals of an opportunity to adjudicate their claims in a5

federal forum, the full Court should consider whether the reasoning of the panels here and in Matimak6

is sound. Because these panel decisions have caused a clear split in authority with the other circuit7

courts, and in light of the potential damage to relations between the United States and the United8

Kingdom and other nations, it can only be hoped that the Supreme Court chooses to address the9

resolution of this issue expeditiously10

11

12

13

14

15

16

17

18

19

20

21

Page 475: Sotomayor Cases

15USCA Order 4

1

2

UNITED STATES COURT OF APPEALS3FOR THE SECOND CIRCUIT4

56789

10LEE N. KOEHLER,11

12Plaintiff-Appellant,13

14v. No. 98-962415

16THE BANK OF BERMUDA (NEW YORK) LIMITED, a17New York Corporation, THE BANK OF BERMUDA18LIMITED, a Bermuda Corporation, REEFS BEACH CLUB19LIMITED, a Bermuda Corporation, and A. DAVID20DODWELL, a Bermuda citizen,21

22Defendants-Appellees.23

24252627

CALABRESI, Circuit Judge, dissenting from the denial of a rehearing in banc:28

For the reasons ably stated by Judge Sotomayor in her opinion dissenting from a denial of29

rehearing in banc, this case involves an issue of exceptional importance. The underlying question30

has divided any number of federal judges. On that basis, if no other, review of the panel opinion is31

warranted. Accordingly, I join Judges Leval and Sotomayor in dissenting from the denial of32

rehearing in banc.33

34

Page 476: Sotomayor Cases

* The Honorable Ellsworth A. Van Graafeiland was originally assigned as a member of thepanel but, due to illness, did not attend oral argument or participate in the disposition of thisappeal. The appeal is being determined by the remaining members of the panel, who are inagreement. See 2d Cir. R. § 0.14(b); Murray v. National Broad. Co., 35 F.3d 45, 46 (2d Cir.1994).

UNITED STATES COURT OF APPEALSFOR THE SECOND CIRCUIT

August Term, 1999

(Argued: April 19, 2000 Decided: October 06, 2000 )

Docket No. 99-7995

_____________________________________________

JOHN E. MALESKO,

Plaintiff-Appellant,

v.

CORRECTIONAL SERVICES CORPORATION, formerly known asEsmor Correctional Services Inc. and “JOHN DOES #1 TO JOHN DOES #10,”inclusive, the names of said John Doe Defendants are presently unknown butintended to indicate officers and managers and guards of the corporate defendant,

Defendants-Appellees._____________________________________________

Before: POOLER and SOTOMAYOR, Circuit Judges.*

Appeal from a judgment of the United States District Court for the Southern District of

New York (Martin, Judge). The district court dismissed Malesko’s complaint against the Correctional

Services Corporation (“CSC”) on the grounds that: (1) Malesko could not assert against CSC, a

private corporation acting on behalf of the federal government, the cause of action

Page 477: Sotomayor Cases

2

set forth in Bivens v. Six Unknown Named Agents of Federal Bureau of Narcotics, 403 U.S. 388,

397 (1971), which provides a damage remedy for violations of constitutional rights by federal agents,

and (2) CSC could invoke government contractor immunity to such a claim in any event. The district

court also dismissed Malesko’s Bivens claims against an individual CSC employee as untimely and, for

the same reason, denied Malesko’s motion for leave to amend the complaint to substitute additional

CSC employees for “DOE” defendants. We agree with the district court’s decisions as to the

individual defendants, but we find that the court erred in dismissing the complaint as to CSC.

Affirmed in part; vacated and remanded in part.

STEVEN PASTERNAK, Pasternak, Feldman &Plutnick, P.A., Livingston, NJ, for plaintiff-appellant.

GEORGE P. STASIUK, Clifton Budd & DeMaria,LLP, New York, NY (George F. Brenlla, onthe brief), for defendants-appellees.

SOTOMAYOR, Circuit Judge:

Plaintiff-appellant John E. Malesko appeals from the July 28, 1999 judgment of the

United States District Court for the Southern District of New York (Martin, Judge) dismissing his

complaint against defendant-appellee Correctional Services Corporation (“CSC”) and a specifically

named CSC employee, and denying him leave to file a second amended complaint. For the reasons

that follow, we vacate the district court’s dismissal of the claims against CSC. We affirm, however, the

district court’s dismissal of Malesko’s claim against the individual CSC employee as time-barred and

Page 478: Sotomayor Cases

3

the denial of leave to file the second amended complaint on the same ground.

BACKGROUND

On December 3, 1992, following his conviction for federal securities fraud, Malesko

was sentenced to eighteen months imprisonment under the supervision of the Federal Bureau of Prisons

(“BOP”). While in the custody and care of BOP, Malesko was diagnosed with a heart condition,

which was treated with prescription medication. On February 2, 1994, Malesko was transferred to Le

Marquis Community Corrections Center, a halfway house where he was to serve out the balance of his

sentence. The halfway house is operated on behalf of BOP by CSC, a private corporation.

Malesko was assigned to living quarters on the fifth floor of the halfway house and was

permitted to use the elevator to travel from the lobby to his room. On or about March 1, 1994,

however, CSC allegedly instituted a policy requiring inmates residing below the sixth floor to use only

the staircase to travel from the first-floor lobby to their rooms. Despite CSC’s policy, Malesko claims

that he was permitted to use the elevator because CSC staff knew of his medical condition. According

to Malesko, however, on March 28, 1994, a CSC employee prevented him from using the elevator to

go from the lobby to his room on the fifth floor. The employee instead directed Malesko to climb the

staircase, even though Malesko reminded the employee of his heart condition. While climbing the

stairs, Malesko suffered a heart attack, fell, and injured himself. Malesko also claims that

approximately ten days prior to this incident, he had run out of the medication prescribed for his heart

condition, and that CSC had failed to replenish his medication as of that date.

Page 479: Sotomayor Cases

1 Defendants-Appellees do not dispute that the CSC employee in question was Jorge Urena.

4

On March 27, 1997, Malesko filed a pro se action against CSC in the United States

District Court for the Southern District of New York, claiming violations of his rights in connection with

the foregoing. Malesko did not name any specific individuals as defendants; instead, he named ten

“unknown” “DOE” defendants. Malesko’s complaint bore the following caption:

JOHN E. MALESKO, Plaintiff -against- CORRECTIONAL SERVICESCORPORATION FORMERLY KNOWN AS ESMOR CORRECTIONALSERVICES INC., “JOHN DOE #1 TO JOHN DOE #10" INCLUSIVE, THENAMES OF SAID JOHN DOE DEFENDANTS ARE PRESENTLY UNKNOWNBUT INTENDED TO INDICATE OFFICERS AND MANAGERS AND GUARDSOF THE CORPORATE DEFENDANT

On February 2, 1999, Malesko, by counsel, filed an Amended Complaint, which was identical to the

initial complaint in all material respects except that it substituted Jorge Urena as “JOHN DOE

DEFENDANT #1" and alleged that Urena was the CSC employee who prevented Malesko’s use of

the elevator on March 28, 1994 and directed Malesko to climb the stairs.1 On February 10, 1999,

CSC moved to dismiss the Amended Complaint. On February 17, 1999, Malesko cross-moved

seeking to file a second amended complaint to name as “DOE” defendants additional CSC employees

allegedly responsible for Malesko’s injuries.

On July 28, 1999, the district court entered a judgment granting CSC’s motion to

dismiss the Amended Complaint, denying Malesko’s motion to file a second amended complaint,

dismissing the Amended Complaint as against Urena and instructing the Clerk of Court to close the

case. See Malesko v. Corrections Servs. Corp., No. 97 Civ. 4080 (JSM), 1999 WL 549003

Page 480: Sotomayor Cases

5

(S.D.N.Y. July 28, 1999). The district court treated Malesko’s Amended Complaint as raising claims

under Bivens v. Six Unknown Named Agents of Federal Bureau of Narcotics, 403 U.S. 388, 397

(1971), which provides for a cause of action for damages against federal agents who violate

constitutional rights.

The district court dismissed the Amended Complaint as to CSC on two grounds. First,

it held that Malesko could not bring a Bivens claim against a corporation such as CSC because,

according to the district court, “[a] Bivens action may only be maintained against an individual.” Id. at

*1. Second, the district court held that, even if Malesko could assert a Bivens claim against a private

corporation, CSC was nevertheless “shielded from liability” because CSC had “contracted with the

federal government to carry out a project on behalf of the government.” Id.

The district court also denied Malesko’s motion to file a second amended complaint

naming additional specific CSC employees as defendants. The district court found that because the

statute of limitations had run on Malesko’s Bivens claims as of March 28, 1997 (one day after he filed

his initial complaint), Malesko’s subsequent assertion of such claims against additional defendants was

time-barred and therefore a “futile amendment” under Fed R. Civ. P. 15. See id. at *2. The district

court rejected Malesko’s contention that his substitution of specifically named defendants for the

“DOE” defendants should “relate back” to the filing date of his initial complaint. See id. at *2-*3.

The district court dismissed the Amended Complaint as to Urena on similar grounds,

finding that the Amended Complaint substituting Urena as a defendant had been filed nearly two years

after the statute of limitations had run. See id. at *3. This appeal followed.

Page 481: Sotomayor Cases

6

DISCUSSION

I. Claims Against CSC

Because the district court dismissed Malesko’s claims against CSC on the pleadings,

we review that decision de novo, accepting as true all material factual allegations in the complaint. See

Jones v. New York Div. of Military and Naval Affairs, 166 F.3d 45, 48 (2d Cir. 1999).

A. Bivens Claims and Private Corporations

We note initially that the question of whether a Bivens claim may lie against a private

corporation is an issue of first impression in this Circuit. The district court dismissed Malesko’s Bivens

claim against CSC because it concluded that a Bivens claim may only be asserted against an individual

federal agent, not against private corporations such as CSC. The district court reached this result

through a purported application of the Supreme Court’s decision in FDIC v. Meyer, 510 U.S. 471

(1994), which held that Bivens claims may not be brought against agencies of the federal government.

See Malesko, 1999 WL 549003 at *1 (citing Meyer, 510 U.S. at 483-87). The district court did not

explain why Meyer’s holding regarding federal agencies precluded a Bivens claim against CSC, which

is not a federal agency. Reviewing this question de novo, we now hold that a private corporation

acting under color of federal law may be subject to a Bivens claim.

Although the issue is new in this Circuit, several circuit courts recognized prior to the

Meyer decision that Bivens-type claims could be asserted against private corporations so long as the

corporations engaged in “federal action,” i.e., they acted under color of federal law. See, e.g.,

Schowengerdt v. General Dynamics Corp., 823 F.2d 1328 (9th Cir. 1987) (holding that private

Page 482: Sotomayor Cases

7

corporation employed by Department of Navy to provide security services would be subject to Bivens

claims if, on remand, its actions were found to constitute federal action); Gerena v. Puerto Rico Legal

Serv., Inc., 697 F.2d 447 (1st Cir. 1983) (recognizing that private entities may be subject to Bivens

claims, but dismissing claims at issue against legal assistance corporation because it did not engage in

federal action); Dobyns v. E-Systems, Inc., 667 F.2d 1219 (5th Cir. 1982) (permitting Bivens claim

against private contractor that supplied personnel, materials, transportation and services to federal

government); Yiamouyiannis v. Chemical Abstracts Serv., 521 F.2d 1392 (6th Cir. 1975) (per

curiam) (remanding for factual determination of whether private corporation engaged in federal action

for purposes of Bivens). During this pre-Meyer time period, no circuit court ever held that private

entities were not subject to Bivens claims.

In 1994, the Supreme Court addressed in Meyer whether “to expand the category of

defendants against whom Bivens-type actions may be brought to include not only federal agents, but

federal agencies as well.” 510 U.S. at 484. Meyer involved a Bivens claim against the Federal

Savings and Loan Insurance Corporation based on an alleged violation of an individual’s due process

rights. See id. at 484-85. The Court, noting that no Court of Appeals decision other than the decision

on appeal had ever implied a Bivens cause of action directly against a federal agency, declined to do so

for several reasons. First, part of the rationale for creating the Bivens cause of action against federal

officials had been to compensate for the fact that “a direct action against the government was not

available.” Id. at 485. Second, allowing Bivens suits against agencies would contravene “the purpose

of Bivens [which] is to deter the officer.” Id. The Supreme Court explained that, because a plaintiff

Page 483: Sotomayor Cases

2 In Reuber, the D.C. Circuit had held that Bivens encompassed actions against private partiesacting under color of federal law, and had remanded the case for a determination of whether thedefendant, an operator of a government-owned facility pursuant to a contract with the National CancerInstitute, had so acted. Id.

8

bringing a Bivens claim against an agency could thereby avoid facing the qualified immunity defense

available to officers, “there would be no reason for aggrieved parties to bring damages actions against

individual officers” and “the deterrent effects of the Bivens remedy would be lost.” Id. Finally, the

Court, observing that “a direct action for damages against federal agencies . . . would be creating a

potentially enormous financial burden for the Federal Government,” refused to expand the

government’s direct liability without congressional legislation to that effect. Id. at 486.

Although Meyer addressed only whether Bivens claims could be brought against

federal agencies and thus, on its face, did not appear to implicate the question of whether a private

corporation could be sued under Bivens, other circuits have subsequently reached differing conclusions

regarding its impact on such claims. In the D.C. Circuit, the Court of Appeals has held that Meyer

precludes Bivens suits against private entities acting under color of federal law, thereby overruling an

earlier pre-Meyer D.C. Circuit decision allowing such claims. See Kauffman v. Anglo-American

School of Sofia, 28 F.3d 1223 (D.C. Cir. 1994), overruling

Reuber v. United States, 750 F.2d 1039 (D.C. Cir. 1985).2 Declaring in Kauffman that its decision

was “controlled by FDIC v. Meyer,” the D.C. Circuit, over a vigorous dissent by then-Chief Judge

Mikva, held that Bivens claims could not be asserted against the Anglo-American School of Sofia, a

“private and independent organisation” established by the U.S. Department of State “to provide

Page 484: Sotomayor Cases

9

elementary-level instruction to the children of American and British diplomats stationed in Sofia,

Bulgaria.” 28 F.3d at 1224 (internal quotation marks omitted). The Kauffman majority reasoned that

because a private entity must act under color of federal law in order to be subject to Bivens, such an

entity is “equivalent” to a federal agency and, under Meyer, must be treated as “if [it] were really a

federal agency.” Id. at 1227.

The Kauffman majority, echoing Meyer’s rationale that the deterrence purpose of

Bivens would be frustrated if claims could be asserted against federal agencies, found that employees

of private entities would be undeterred from engaging in unconstitutional conduct if claims could be

asserted against their employers. See id. at 1227. The court similarly adopted Meyer’s reasoning that

Bivens claims against federal agencies would create a “potentially large financial drain on the

government” and concluded that the “diversion of resources from a private entity created to advance

federal interests has effects similar to those of diversion of resources directly from the [federal]

Treasury” because such costs would be passed on to the government. Id. at 1227-28. The Kauffman

majority thus held that Meyer’s reasoning precluded Bivens claims against private entities acting under

color of federal law. Id. at 1228.

In contrast, the Sixth Circuit in Hammons v. Norfolk Southern Corp., 156 F.3d 701

(6th Cir. 1998), concluded that “[n]othing in Meyer prohibits a Bivens claim against a private

corporation that engages in federal action,” and held that such claims should be allowed. Id. at 705.

The Hammons court observed that while Meyer had focused on Bivens’s deterrence rationale in

concluding that claims should not be asserted against federal agencies, “the primary goal of Bivens was

Page 485: Sotomayor Cases

10

to provide a remedy for victims of constitutional violations by federal agents where no other remedy

exists, regardless of whether the official would be deterred in the future from engaging in such conduct.”

Id. at 706 (citing Bivens, 403 U.S. at 407-08 (Harlan, J., concurring)). The court therefore concluded

that “the importance of remedying constitutional violations” favored the allowance of Bivens claims

against private entities acting under color of federal law regardless of whether such allowance deterred

future violations. Id. The Hammons court also found, in contrast to the D.C. Circuit, that Meyer’s

concern about “the potentially enormous financial burden on the Government” was inapplicable with

respect to private corporations because claims against private corporations do not directly impact the

“federal purse” and thus “do[] not implicate federal fiscal policy.” Id. (citations and internal quotation

marks omitted).

Having determined that “Meyer [was] not dispositive” on the matter, id. at 706, the

Sixth Circuit proceeded to address whether private corporations should be subject to Bivens claims.

The court observed that Bivens claims against federal officers have consistently been treated in the

same manner as claims against state officers under 42 U.S.C. § 1983 (1994). See id. at 707 (“[T]he

standards of liability in Bivens actions are similar to the standards under

§ 1983. Both the Supreme Court and this Court have noted that actions brought under § 1983 raise

identical concerns as those raised in Bivens actions.” (citing Butz v. Economou, 438 U.S. 478, 500

(1978))). The court further noted that “[i]t is undisputed that corporations engaging in state action can

be sued under § 1983.” Id. (citations omitted). Accordingly, the Hammons court held that

corporations engaging in federal action should, by analogy, be subject to Bivens claims. See id. at 708.

Page 486: Sotomayor Cases

11

In so holding, the court observed that it joined the Courts of Appeals for the First, Fifth and Ninth

Circuits in their conclusions (albeit pre-Meyer) that a Bivens claim may be brought against private

corporations engaging in federal action. See id.

We find the Sixth Circuit’s analysis in Hammons persuasive and, substantially for the

reasons articulated by that court, we hold that a private corporation acting under color of federal law

may be sued under Bivens. As an initial matter, we do not believe that Meyer is dispositive here

because private entities acting on behalf of the federal government are not the equivalent of federal

agencies. In Cohen v. Empire Blue Cross, 176 F.3d 35 (2d Cir. 1999), we found that Empire Blue

Cross, acting as a fiscal intermediary on behalf of the government to process Medicare claims, was akin

to an agent of the government and not a federal agency. See id. at 42 (“[Defendant] is a private not-

for-profit corporation established under the laws of the State of New York. It is not funded by the

United States, and the United States has no proprietary interest in its operations . . . . [It] may act as

an agent for the government, [but] it is not an institutional arm of the government such as a department,

commission, board, or bureau.”). We see no reason to treat differently other private corporations

acting on behalf of the government.

Moreover, the reasons the Supreme Court articulated in Meyer for declining to extend

liability to federal agencies are not compelling with respect to the question of whether private

corporations should be subject to Bivens liability. Although deterring wrongdoing by individuals is an

important goal of Bivens liability, we find an extension of such liability to be warranted even absent a

substantial deterrent effect in order to accomplish the more important Bivens goal of providing a

Page 487: Sotomayor Cases

3 In our view, the non-unanimous Kauffman decision finding such impact is incorrect, or, at best,resulted from the fact that the entity at issue was arguably a federal government agency. Although theKauffman majority referred to the defendant school as a “private” entity, the school had been set up bythe U.S. Department of State, part of its governing board was appointed by a U.S. ambassador, and itwas directly funded, in part, by the U.S. government. See Kauffman, 28 F.3d at 1224.

12

remedy for constitutional violations. Justice Harlan, in his concurring opinion in Bivens, elaborated

upon the relationship between these two goals:

I agree with the Court that the appropriateness of according Bivens compensatory reliefdoes not turn simply on the deterrent effect liability will have on federal official conduct. Damages as a traditional form of compensation for invasion of a legally protectedinterest may be entirely appropriate even if no substantial deterrent effects on futureofficial lawlessness might be thought to result. Bivens, after all, has invoked judicialprocesses claiming entitlement to compensation for injuries resulting from allegedlylawless official behavior, if those injuries are properly compensable in money damages. I do not think a court of law–vested with the power to accord a remedy–should denyhim his relief simply because he cannot show that future lawless conduct will thereby bedeterred.

Bivens, 403 U.S. at 407-08 (Harlan, J., concurring). In any event, we are not convinced that allowing

Bivens claims against private corporations would undermine the deterrence purpose of Bivens. Even

assuming a plaintiff would decline to sue the offending employee and sue only the employer, we believe

that an employer facing exposure to such liability would be motivated to prevent unlawful acts by its

employees. Furthermore, although a private entity contracting to carry out governmental functions

might pass on the costs of Bivens liability to the government, we do not believe such liability has the

type of direct impact on federal fiscal policy that the Supreme Court in Meyer was concerned would

result from imposing Bivens liability directly upon federal agencies.3

Finally, as noted above, several Courts of Appeals before Meyer’s issuance recognized

Page 488: Sotomayor Cases

13

that private corporations may be subject to Bivens claims. We are not persuaded that the Supreme

Court implicitly overruled this long line of cases in the Meyer opinion without making reference to any

of them. We therefore find that Meyer does not control our analysis of whether Bivens claims may be

brought against private corporations.

In deciding that Bivens liability should extend to private corporations, we are influenced

strongly by the law governing § 1983 claims. We have consistently treated Bivens and § 1983 actions

as analogous for most purposes, and we “have typically incorporated § 1983 law into Bivens actions.”

Tavarez v. Reno, 54 F.3d 109, 110 (2d Cir. 1995) (per curiam) (citing Carlson v. Green, 446 U.S.

14, 18-20 (1980)); see also Ayeni v. Mottola, 35 F.3d 680, 688 n.10 (2d Cir. 1994); Ellis v. Blum,

643 F.2d 68, 84 (2d Cir. 1981). The Supreme Court has made it clear that private corporations

engaging in state action may be sued under § 1983. See Lugar v. Edmonson Oil Co., 457 U.S. 922,

936-37 (1982). We see no reason not to incorporate that law into the Bivens context and permit suits

against private corporations engaging in federal action. Moreover, we find that extending Bivens

liability to reach private corporations furthers Bivens’s overriding purpose: providing redress for

violations of constitutional rights. See Bivens, 403 U.S. at 407-08 (Harlan, J., concurring).

Accordingly, we join the Courts of Appeals for the First, Fifth, Sixth and Ninth Circuits

and hold that a private corporation acting under color of federal law may be subject to liability under

Bivens.

Page 489: Sotomayor Cases

14

B. Government Contractor Defense

In addition to finding that a Bivens action could not be maintained against CSC, the

district court held that CSC would have been immune from such a suit in any event by virtue of the

immunity for government contractors set forth in Boyle v. United Technologies, Corp., 487 U.S. 500

(1988). See Malesko, 1999 WL 549003 at *1 (citing Boyle, 487 U.S. at 505-09; Yearsley v. W.A.

Ross Const. Co., 309 U.S. 18, 19 (1940); Norwood v. Esmor, Inc., No. 95 Civ. 8281 (LAP), 1997

WL 65913 (S.D.N.Y. Feb. 13, 1997)). We disagree and hold that on the record presented to the

district court, CSC may not avail itself of the government contractor defense.

In Boyle, the Supreme Court articulated the standard for determining when federal law

shields government contractors from state tort liability arising from design defects in military equipment.

See Boyle, 487 U.S. at 512. The Supreme Court held:

Liability for design defects in military equipment cannot be imposed, pursuant to statelaw, when (1) the United States approved reasonably precise specifications; (2) theequipment conformed to those specifications; and (3) the supplier warned the UnitedStates about the dangers in the use of the equipment that were known to the supplierbut not to the United States.

Boyle, 487 U.S. at 512. The Supreme Court reasoned that “selection of the appropriate design for

military equipment” involved “judgment as to the balancing of technical, military, and even social

considerations” that should not be “second-guess[ed] . . . through state tort suits against contractors”

carrying out the design specifications of the federal government. Id. at 511 (citations and internal

quotation marks omitted). Thus, the Court concluded that “state law which holds Government

contractors liable for design defects in military equipment does in some circumstances present a

Page 490: Sotomayor Cases

4 We also question, although need not decide, whether the government contractor defense mayever be invoked as a shield against Bivens claims. As the Court in Boyle explained, the governmentcontractor defense exists to protect against situations where the “application of state law [claims]would frustrate specific objectives of federal legislation.” Id. at 507 (internal quotation marks omitted)(emphasis added). The Bivens cause of action, however, was created to advance federalconstitutional interests. Such claims are not state law claims but rather arise under the U.S.Constitution.

15

‘significant conflict’ with federal policy and must be displaced.” Id. at 512.

Although the government contractor defense has primarily developed in the context of

military contractors, the defense has been applied more broadly by some courts to protect contractors

in non-military contexts. See, e.g., Boruski v. United States, 803 F.2d 1421, 1430 (7th Cir. 1986)

(defense applies in “civilian relationships” where “a contractor has acted in the sovereign's stead and

can prove the elements of the defense”) (citation and internal quotation marks omitted); Burgess v.

Colorado Serum Co., 772 F.2d 844, 846 (11th Cir. 1985) (“Both the history of the defense and its

general rationale lead us to the conclusion that it would be illogical to limit the availability of the defense

solely to ‘military’ contractors”). This Circuit has never expressly addressed the issue.4 However, this

case does not present the appropriate forum in which to do so since, even assuming the potential

applicability of the defense outside the military context, the requirements necessary for its application

are not satisfied here.

The government contractor defense only shields a government contractor from claims

arising out of its actions where the government has exercised its discretion and judgment in approving

precise specifications to which the contractor must adhere. See Lewis v. Babcock Indus., Inc., 985

F.2d 83, 86-87 (2d Cir. 1993) (holding that government contractor defense requires “that the design

Page 491: Sotomayor Cases

16

feature in question was considered by Government officers, and not merely by the contractor”). In

other words, “[s]tripped to its essentials,” the government contractor defense is “to claim, ‘The

Government made me do it.’” In re Joint Eastern and Southern Dist. New York Asbestos Litg.,

897 F.2d 626, 632 (2d Cir. 1990). Here, Malesko alleges that various CSC policies or practices led

to his injury — assigning him to a room on the fifth floor, forbidding him from using the elevator, and

failing to refill his heart medication prescription — but neither he nor CSC claims that the government

played any role in formulating or approving these policies or practices. Because no allegation exists in

the record before the district court that the government played any role in these decisions, we conclude

that the district court erred in holding that CSC was protected from liability by the government

contractor defense. See, e.g., In re Hawaii Asbestos Cases, 960 F.2d 806, 813 (9th Cir. 1992)

(holding that suppliers of insulation products containing asbestos to the Navy could not invoke defense

where suppliers “simply failed to allege, let alone establish, that . . . they were acting in compliance with

‘reasonably precise specifications’ imposed on them by the United States”) (quoting Boyle, 487 U.S. at

512).

II. Claims Against Individual Defendants

The district court dismissed the Bivens claims against Jorge Urena on the ground that

the claims were time-barred. See Malesko, 1999 WL 549003 at *3. For the same reason, the district

court denied as “futile” Malesko’s motion to file a second amended complaint substituting additional

specific individuals for “DOE” defendants. Id. at *2-*3. We review de novo the district court’s

Page 492: Sotomayor Cases

17

dismissal of the Bivens claims against Urena on the pleadings. See Jones, 166 F.3d at 49. We review

the district court’s denial of leave to amend the complaint for abuse of discretion. Id. If that denial was

based on a legal interpretation, however, we review the decision de novo. Id.

In New York, the statute of limitations for bringing a Bivens action is three years from

the accrual of the claim. See Chin v. Bowen, 833 F.2d 21, 24 (2d Cir. 1987). The parties do not

dispute that Malesko’s claim accrued on March 28, 1994, the day on which he suffered his heart attack

while climbing the CSC stairs. Thus, the statute of limitations on Malesko’s Bivens claims expired on

March 28, 1997, one day after he filed his original complaint.

Malesko, however, did not identify any of the alleged “DOE” defendants listed on his

original complaint until February 2, 1999, when he submitted an Amended Complaint substituting Jorge

Urena as “JOHN DOE DEFENDANT #1.” On February 17, 1999, Malesko moved for leave to file

a second amended complaint that would purportedly have substituted additional CSC employees as

“DOE” defendants. It is therefore clear from the record that the statute of limitations expired on

Malesko’s Bivens claims before he added or sought to add any individual defendants. Malesko

attempts to avoid this conclusion by arguing that, pursuant to Fed. R. Civ. P. 15(c), the date of filing for

his amendments identifying individual defendants may “relate back” to the timely date on which the

original complaint was filed. See Aslanidis v. United States Lines, Inc., 7 F.3d 1067, 1075 (2d Cir.

1993) (stating that “‘John Doe’ pleadings” can “be used to circumvent statutes of limitations” where “all

the specifications of Fed. R. Civ. P. 15(c) are met”).

Under Rule 15(c), when an attempt is made to bring in a new party, the date of the

Page 493: Sotomayor Cases

18

amendment adding that party will “relate back” to the date of the original complaint only when (1) the

claim arises out of the same conduct originally pleaded and (2) within (ordinarily) 120 days of the

original filing date,

the party to be brought in by amendment (A) has received such notice of the institutionof the action that the party will not be prejudiced in maintaining a defense on the merits,and (B) knew or should have known that, but for a mistake concerning the identityof the proper party, the action would have been brought against the party.

Fed. R. Civ. P. 15(c)(3) (emphasis added). Thus, in order to have the filing dates for the Amended

Complaint and second amended complaint relate back to the date of the original Complaint, Malesko

must show, inter alia, that he failed to name Urena and other specific CSC employees due to a

“mistake concerning the identity of the proper party.” Fed. R. Civ. P. 15(c)(3)(B). A plaintiff is not

considered to have made such a “mistake,” however, if the plaintiff knew that he was required to name

an individual as a defendant but did not do so because he did not know the individual’s identity. See

Barrow v. Wethersfield Police Dep’t, 66 F.3d 466, 470 (2d Cir. 1996) (“Rule 15(c) does not allow

an amended complaint adding new defendants to relate back if the newly-added defendants were not

named originally because the plaintiff did not know their identities.”).

Here, it is clear that Malesko believed that there existed individual defendants who

were potentially liable for his injuries, but did not know their exact identities prior to the expiration of

the statute of limitations. As revealed by the caption from his original complaint, Malesko

“INTENDED TO INDICATE OFFICERS AND MANAGERS AND GUARDS OF THE

CORPORATE DEFENDANT [CSC],” but he did not do so because “THE NAMES OF SAID

Page 494: Sotomayor Cases

19

JOHN DOE DEFENDANTS ARE PRESENTLY UNKNOWN.” Under these circumstances,

Malesko cannot avail himself of Rule 15(c) such that his untimely substitution of specifically named

individuals for “DOE” defendants may “relate back” to the date he filed his original complaint. See

Tapia-Ortiz v. Doe, 171 F.3d 150, 151-52 (2d Cir. 1999) (holding that plaintiff, who, in his original

complaint, sued “Doe” defendant officers who allegedly applied excessive force, could not “relate

back” his amendment naming specific officers two years after the statute of limitations had run);

Barrow, 66 F.3d at 470 (holding that plaintiff named “John Does” because he did not know the

arresting officers’ names and that, “[s]ince the new names were [subsequently] added not to correct a

mistake but to correct a lack of knowledge, the requirements of Rule 15(c) for relation back are not

met”).

We therefore affirm the district court’s decision dismissing the Amended Complaint as

to Urena and denying Malesko’s motion to file the second amended complaint because these claims

were barred by the statute of limitations.

Page 495: Sotomayor Cases

5 In a discussion spanning less than one page of his appellate brief, Malesko argues for the firsttime that he is entitled to recover under the Americans With Disabilities Act, 42 U.S.C. § 12131, etseq., and the Rehabilitation Act of 1973, 29 U.S.C. § 794. Although CSC invites us to address themerits of these claims, we decline to do so. The district court may consider the arguments on remandafter the issue has been adequately briefed. See Able v. United States, 88 F.3d 1280, 1292 (2d Cir.1996) (declining to address the merits of a claim where the issue was not considered in the district courtand remanding for such consideration).

20

CONCLUSION

For the foregoing reasons, we affirm the district court’s dismissal of Bivens claims

against defendant Urena and its denial of leave to file the second amended complaint, but vacate the

district court’s dismissal of claims against CSC and remand for further proceedings consistent with this

opinion.5

Page 496: Sotomayor Cases

No.97-9162

IN THE UNITED STATES COURT OF APPEALSFOR THE SECOND CIRCUIT

MARILYN BARTLETT,

Plaintiff-Appellee

v.

NEW YORK STATE BOARD OF LAW EXAMINERS, et al.,

Defendants-Appellants

ON APPEAL FROM THE UNITED STATES DISTRICTCOURT FOR THE SOUTHERN DISTRICT OF NEW YORK

SUPPLEMENTAL BRIEF FOR THE UNITED STATESAS AMICUS CURIAE

MARY JO WHITE BILL LANN LEE United States Attorney Acting Assistant Attorney for the Southern District General of New York

SARA L. SHUDOFSKY JESSICA DUNSAY SILVER Assistant United States MARIE K. McELDERRY Attorney Attorneys Department of Justice

P.O. Box 66078 Washington, D.C. 20035-6078 (202) 514-3068

______________________________________________________________________________________________________________________________________

Page 497: Sotomayor Cases

TABLE OF CONTENTS

PAGEARGUMENT:

THIS COURT CORRECTLY CONCLUDED THAT BARTLETT ISAN INDIVIDUAL WITH A DISABILITY UNDER TITLE II OF THE ADA . . . . . . . . . . . . . . . . . . . . . . . . . 2

A. The Record Is Clear That Bartlett LacksAutomaticity In Her Reading . . . . . . . . . . . . . . 4

B. Bartlett Is Substantially Limited In Reading Even When Taking Into AccountHer Self-Accommodation Techniques . . . . . . . . . . . 8

C. This Court’s Determination That Bartlett Has A DisabilityThat Causes Substantial Limitations In The Major Life Activity Of Reading Is Consistent With The Supreme Court’s Decisions In Sutton, Murphy, And Albertsons . . . . . . . . . . 11

CONCLUSION . . . . . . . . . . . . . . . . . . . . . . . . . . 16

CERTIFICATE OF SERVICE

- i -

Page 498: Sotomayor Cases

TABLE OF AUTHORITIES

CASES: PAGE

Albertsons, Inc. v. Kirkingburg, 119 S. Ct. 2162 (1999) . . passim

Bartlett v. New York State Bd. of Law Exam’rs, 970 F. Supp. 1094 (S.D.N.Y. 1997), reconsideration denied, 2 F. Supp. 2d 388 (S.D.N.Y. 1997), aff’d in part, vacated in part, 156 F.3d 321 (2d Cir. 1998), vacated and remanded, 119 S. Ct. 2388 (1999) . . . . . passim

Bragdon v. Abbott, 524 U.S. 624 (1998) . . . . . . . . . . . . 14

Murphy v. United Parcel Serv., Inc., 119 S. Ct. 2133 (1999) . . . . . . . . . . . . . . . . . . . . . passim

Sutton v. United Air Lines, Inc., 119 S. Ct. 2139 (1999) . . . . . . . . . . . . . . . . . . . . . passim

Taylor v. Phoenixville Sch. Dist., 184 F.3d 296 (3d Cir. 1999) . . . . . . . . . . . . . . . . . . . . 14

STATUTES:

Americans with Disabilities Act of 1990, 42 U.S.C. 12102(2)(A) . . . . . . . . . . . . . . . . . . 2Title II, 42 U.S.C. 12131 et seq. . . . . . . . . . . . . . 2

RULES AND REGULATIONS:

29 C.F.R. 1630.2(j)(3)(i) . . . . . . . . . . . . . . . . . . . 16

MISCELLANEOUS:

Jeanne S. Chall, Stages of Reading Development (1983) . . . . . . 4

135 Cong. Rec. 8519 (1989) . . . . . . . . . . . . . . . . . . 15

Patricia R. Dahl, A mastery based experimental programfor teaching high speed word recognition skills(abstract), 11 Reading Res. Q. 203 (1975-1976) . . . . . . . 4

Sally E. Shaywitz, Current Concepts: Dyslexia, 338 New Eng. J. Med. 307 (1998) . . . . . . . . . . . . . 15

- ii -

Page 499: Sotomayor Cases

IN THE UNITED STATES COURT OF APPEALSFOR THE SECOND CIRCUIT

No. 97-9162

MARILYN BARTLETT,

Plaintiff-Appellee

v.

NEW YORK STATE BOARD OF LAW EXAMINERS, et al.,

Defendants-Appellants

ON APPEAL FROM THE UNITED STATES DISTRICTCOURT FOR THE SOUTHERN DISTRICT OF NEW YORK

SUPPLEMENTAL BRIEF FOR THE UNITED STATES AS AMICUS CURIAE

On June 24, 1999, the Supreme Court granted the petition for a

writ of certiorari in New York State Board of Law Examiners v.

Bartlett, No. 98-1285, vacated this Court's September 14, 1998,

decision, and remanded the case to this Court for reconsideration

in light of Sutton v. United Air Lines, Inc., 119 S. Ct. 2139

(1999), Murphy v. United Parcel Service, Inc., 119 S. Ct. 2133

(1999), and Albertsons, Inc. v. Kirkingburg, 119 S. Ct. 2162

(1999). On July 30, 1999, this Court issued an order directing the

parties to file supplemental briefs to consider the effect of those

three decisions. Having previously filed a brief as amicus curiae

in this appeal, the United States hereby submits this supplemental

brief to address the issue presented by the Supreme Court's remand.

Page 500: Sotomayor Cases

- 2 -

1/ Plaintiffs in Sutton had severe myopia, but with the useof corrective lenses, their vision was 20/20 or better. 119 S.Ct. at 2143.

2/ As relevant to this case, the statutory definition ofdisability is "a physical or mental impairment that substantiallylimits one or more of the major life activities of [an]individual." 42 U.S.C. 12102(2)(A).

ARGUMENT

THIS COURT CORRECTLY CONCLUDED THAT BARTLETT IS ANINDIVIDUAL WITH A DISABILITY UNDER TITLE II OF THE ADA

The Supreme Court determined in Sutton that "if a person is

taking measures to correct for, or mitigate, a physical or mental

impairment, the effects of those measures -- both positive and

negative -- must be taken into account when judging whether that

person is 'substantially limited' in a major life activity and thus

'disabled' under the [Americans with Disabilities] Act." 119 S.

Ct. at 2146.1/ In so holding, the Court relied in part upon the

fact that the ADA requires an individualized inquiry into the

question whether an individual has a disability. Id. at 2147.

In its September 14, 1998, decision, this Court held that

plaintiff Marilyn Bartlett is an individual with a disability

protected by Title II of the Americans with Disabilities Act of

1990 (ADA), 42 U.S.C. 12131, et seq. Bartlett v. New York State

Bd. of Law Exam'rs, 156 F.3d 321 (2d Cir. 1998).2/ In reaching

that conclusion, this Court stated that a disability should be

assessed without regard to the availability of mitigating

measures. Id. at 329. As a result, the Supreme Court granted

the Board's petition, vacated this Court's decision, and remanded

Page 501: Sotomayor Cases

- 3 -

the case to this Court for reconsideration under the legal

standard announced in Sutton, Murphy, and Albertsons.

Although in reaching its decision in this case this Court

endorsed a principle that has subsequently been rejected by the

Supreme Court, the Court’s conclusion that Bartlett is an

individual with a disability remains correct. Following Sutton

and the related cases, the appropriate inquiry in determining

whether an individual has a disability within the meaning of the

ADA is whether, notwithstanding the use of a corrective device or

mitigating measures, the "limitations an individual with an

impairment actually faces are in fact substantially limiting."

119 S. Ct. at 2149. As discussed below, the record in this case

demonstrates that, despite her efforts at self-accommodation,

Bartlett is substantially limited in the major life activity of

reading. 156 F.3d at 329. The self-accommodation techniques

used by Bartlett do not mitigate the crucial element of her

dyslexia: her lack of automaticity in reading. Accordingly, even

when taking her attempts at self-accommodation into account,

Bartlett is substantially limited in the major life activity of

reading. Because that is the only conclusion that can be drawn

from the record, together with the district court’s findings, the

district court’s judgment should be affirmed on that basis.

Page 502: Sotomayor Cases

- 4 -

3/ See, e.g., Pl.'s Ex. 129, Jeanne S. Chall, Stages ofReading Development 119 (1983), citing Patricia R. Dahl, Amastery based experimental program for teaching high speed wordrecognition skills (abstract), 11 Reading Res. Q. 203, 209 (1975-1976).

A. The Record Is Clear That Bartlett Lacks Automaticity In Her Reading

Experts recognize that the skill of reading has at least two

major components3/: accuracy of word identification and

“automaticity”--the ability to "recognize[] a printed word and

[be] able to read it accurately, and immediately; in other words,

automatically and without [conscious effort]." Bartlett v. New

York State Bd. of Law Exam'rs, 970 F. Supp. 1094, 1107, 1113

(S.D.N.Y. 1997). The Board of Law Examiners took the position in

the district court that Bartlett's scores on the Word Attack and

Word Identification subtests of the Woodcock Reading Mastery Test

(Woodcock subtests) were alone sufficient to determine whether

Bartlett has a learning disability. The Woodcock subtests used

by the Board’s expert, however, measure only one of the

components of reading, i.e., the ability to identify words

accurately and not the major component underlying adult reading,

i.e., automaticity. Automaticity has to be assessed by a reading

measure that includes time; the scores on the Woodcock subtests

did not measure Bartlett’s lack of automaticity because those

tests are untimed and do not reflect the great difficulty she has

in deciphering each word. As the district court recognized, the

principal problem with using the scores on psychometric testing

as the sole determinant of whether an individual has a learning

Page 503: Sotomayor Cases

- 5 -

disability is the fact that "no test measures automaticity

directly." 970 F. Supp. at 1113. The Board’s complete reliance

on Bartlett’s scores on the Woodcock subtests to determine

whether she has a learning disability therefore presents an

incomplete and misleading picture. Accordingly, the district

court properly rejected the Board’s position that the Woodcock

subtests scores should be determinative, finding that “[b]y its

very nature, diagnosing a learning disability requires clinical

judgment,” and “is not quantifiable merely in test scores.” Id.

at 1114.

Recognizing the importance of clinical judgment, the

district court relied on the experts’ clinical observations of

Bartlett when she read aloud. The opinion of all three experts

who observed her noted her "stark lack of automaticity" under

those circumstances. 970 F. Supp. at 1113. In his trial

affidavit, Dr. Richard Heath testified that Bartlett "reads aloud

in a hesitant manner, slowly and without automaticity." Id. at

1107. He stated that, "[i]n particular, [Bartlett] had a great

deal of difficulty reading polysyllabic words, vowels (especially

diphthongs, digraphs and in ascertaining differences between long

and short vowels), consonant blends and silent consonant

conventions." Ibid. He reported (ibid.) that

on the more complex reading passages, Dr. Bartletttypically read the passages over two or three timesbefore she could respond to that test item. She usescontextual cues to facilitate her decoding. She readsvery slowly. She will reread a phrase or sentence tomake sure she gets it. You can often see her lips moveor hear her read quietly to herself and when she doesthis, you can hear the mispronunciations. When she is

Page 504: Sotomayor Cases

- 6 -

4/ The district court also credited the studies of adultdyslexics conducted by Dr. Maggie Bruck, on which the Board’sexperts relied to support their testimony. Dr. Bruck stated, and

(continued...)

faced with an unfamiliar polysyllabic word she is veryslow to break down the word to different parts and shewill mispronounce parts of the word. She is slow tosynthesize the morphemes into a word.

Dr. Heath administered the same Woodcock subtests used by

the Board, and his opinion was that the results of that testing

confirmed Dr. Phillip Massad’s earlier diagnosis of learning

disability. 970 F. Supp. at 1107. His clinical observation of

Bartlett revealed her difficulties in arriving at answers. Dr.

Heath described the fact that Bartlett "had to make several

attempts to sound out words which should have been second nature

to her,” and her "reading was full of hesitations, and self

corrections.” Ibid. (quoting Heath affidavit). As an example,

Dr. Heath stated (ibid.):

[P]laintiff will attempt to read a word such as"instigator" as "investigator." Since she will hearthat it sounds incorrect she will start over and oftencorrects her reading of the word after severalattempts. On the Woodcock, this would be credited as acorrect response, even though it took her threeattempts to get it right and took more time than itwould have taken a person who did not have to read inthis fashion.

Dr. Heath also stated that, although “[w]ord attack skills are

generally well formed by junior high school age,” Bartlett's

“pattern of word attack is indicative of someone whose decoding

skills are not fully formed,” and that she “decodes pseudo-words

at a fourth grade level.” Id. at 1107-1108.4/

Page 505: Sotomayor Cases

- 7 -

4/(...continued)the district court found, that the Woodcock subtests are "poordiscriminators" for measuring whether an adult has a learningdisability "unless the subject’s reaction time [i]s measured." 970 F. Supp. at 1113-1114. In addition, Dr. Rosa Hagin testifiedthat, because the Woodcock subtests do not test automaticity orreading rate, "they are poor indicators of a decoding problem inindividuals like plaintiff who function at higher cognitivelevels." Id. at 1110.

5/ The record also contains test data from the DiagnosticReading Test (DRT). 970 F. Supp. at 1108. Bartlett’s readingrate was compared with the highest grade norm for that test,which is college freshmen. Ibid. (table). The test results showthat Bartlett’s slow reading rate is comparable to the 4thpercentile of college freshmen when timed, while hercomprehension was at the 50th percentile. When she took the testuntimed, her comprehension was at the 98th percentile, but, atthe same time, that required her to read at an even slower rate,comparable to the 1st percentile of college freshmen. Ibid.

During the hearing, the district court also directly

observed the condition and manner used by Bartlett to read and

write, including using her fingers to keep her place in the text,

spelling errors, and mirror writing. She read aloud, "haltingly

and laboriously," at 40 words per minute and took approximately

ten minutes to write a 48-word passage that was dictated to her.

970 F. Supp. at 1110.5/

This Court agreed with the district court in rejecting the

Board’s argument that scores on the Woodcock subtests are the

"dispositive measure" of whether an adult has a learning

disability, 156 F.3d at 329, and nothing in the Supreme Court’s

decisions in Sutton and the related cases calls that conclusion

into question.

B. Bartlett Is Substantially Limited In Reading Even When Taking Into Account Her Self-Accommodation Techniques

Page 506: Sotomayor Cases

- 8 -

6/ Indeed, as the district court noted, the Board’s expert"acknowledge[d] the Woodcock’s weakness with regard todiscriminating for lack of automaticity." 970 F. Supp. at 1114.

The Board does not appear to dispute the finding that

Bartlett reads without automaticity.6/ Rather, the Board’s

principal argument here (Defendants-Appellants’ Supplemental Br.

6-7) is that Bartlett’s "self-accommodation permits her to read

at an average level compared to the average person in the general

population," and thus that she "does not have a reading or

learning disability that would entitle her to accommodations" for

taking the New York bar examination. That argument is based upon

the district court's finding that, when compared to the general

population, Bartlett has achieved "roughly average reading skills

(on some measures)." 970 F. Supp. at 1120.

The Board’s reliance on this aspect of the district court’s

finding is at odds with this Court’s clear rejection of the

Board’s argument that scores on the Woodcock subtests are the

"dispositive measure" of whether an adult has a learning

disability. 156 F.3d at 329. The district court’s finding was

specifically qualified by the court as being based on “some

measures.” Those measures were the Woodcock Word Attack and Word

Identification subtests. As noted above, this Court clearly

found that those measures were not adequate to judge whether

Bartlett has a learning disability. Since the Woodcock subtests

measure only her ability to identify words, without regard to the

time it takes or the mistakes she makes before arriving at the

correct answer, Bartlett's average scores on those subtests do

Page 507: Sotomayor Cases

- 9 -

not identify the substantial limitations she experiences in the

reading process. As discussed below, Bartlett’s self-

accommodation techniques provide a degree of mitigation with

respect to the word identification component of the reading

process; they do not, however, provide mitigation with respect to

Bartlett’s lack of automaticity in her reading.

Dr. Rosa Hagin, an expert who testified during the hearing

in the district court, described the "set of personal skills"

that Bartlett has "evolved * * * to compensate for her

disability." 970 F. Supp. at 1109. The "cues" Bartlett used to

assist her were "slowing down the rate of response, verbal

rehearsal of rote sequencing items, [and] pointing cues to assist

in keeping her place on visual text." Ibid. She “use[d] her

finger to keep her place,” and read the more complex passages

over several times as a means of obtaining “contextual cues to

facilitate her decoding.” Id. at 1107 (internal quotation marks

omitted). She "had to sound out the words repeatedly before

coming to an answer." Id. at 1113. Dr. Hagin credited

Bartlett’s "earlier work as a school teacher where phonics were

stressed" in allowing her to attempt to develop "self-

accommodations." Id. at 1109. Significantly, however, Dr. Hagin

noted that those self-accommodations, which permit her to decode

words if she has a sufficient amount of time, "account for her

ability to spell better and to perform better on [the untimed

Woodcock] word identity and word attack tests than would be

expected of a reading disabled person," ibid., because, as this

Page 508: Sotomayor Cases

- 10 -

Court noted, 156 F.3d at 329, both of those subtests allow

Bartlett unlimited time to identify a word. They do not measure

the fact that she reads without automaticity. Ibid. Thus,

although Bartlett has developed methods that permit her, with

additional time, to decipher the written words, the record shows

that the essential component of automaticity continues to be

absent in her reading.

Accepting the district court's subsidiary findings, this

Court found, in essence, that Bartlett's barely average scores on

the Woodcock subtests are only a part of the picture and that

lack of automaticity is the crucial element in her dyslexia.

This Court therefore rejected the district court’s conclusion

that Bartlett was not substantially limited in the major life

activity of reading, Bartlett v. New York State Bd. of Law

Exam'rs, 2 F. Supp. 2d 388, 392 (S.D.N.Y. 1997), making a legal

determination that the district court’s finding concerning

Bartlett’s average scores on the Woodcock subtests was not a

sufficient basis for that conclusion. Instead, this Court

properly relied upon the record and subsidiary findings made by

the district court in concluding that Bartlett was substantially

limited in the major life activity of reading, and her impairment

significantly restricts the condition and manner of her reading

“as compared to the manner and conditions under which the average

person in the general population can read or learn.” 156 F.3d at

329.

C. This Court’s Determination That Barlett Has A Disability That Causes Substantial

Page 509: Sotomayor Cases

- 11 -

Limitations In The Major Life Activity Of Reading Is Consistent With The Supreme Court’s Decisions In Sutton, Murphy, And Albertsons

Nothing in the Supreme Court’s decisions in Sutton, Murphy,

or Albertsons calls into question this Court’s conclusion that

Bartlett is a person with a disability. Bartlett's “history of

self-accommodations” does not foreclose a finding that she has a

disability. 156 F.3d at 329. In Sutton, the Supreme Court made

clear that the "use or nonuse of a corrective device does not

determine whether an individual is disabled; that determination

depends on whether the limitations an individual with an

impairment actually faces are in fact substantially limiting."

119 S. Ct. at 2149 (emphasis added). Because, with the use of

corrective measures, the plaintiffs in Sutton reached 20/20

visual acuity and could "function identically to individuals

without a similar impairment," ibid., the Court held that they

were not substantially limited in any major life activity.

Thereafter, in Murphy, the Supreme Court accepted the Tenth

Circuit’s conclusion that "when medicated, petitioner’s high

blood pressure does not substantially limit him in any major life

activity." 119 S. Ct. at 2137. In Murphy, the Court was

presented solely with the question whether mitigating measures

should be considered in determining whether an individual’s

impairment substantially limits a major life activity; the Court

was not presented with the question whether the Tenth Circuit’s

conclusion as to substantial limitation was correct.

Specifically, the Supreme Court in Murphy had "no occasion * * *

Page 510: Sotomayor Cases

- 12 -

to consider whether petitioner is ‘disabled’ due to limitations

that persist despite his medication." Ibid. (emphasis added).

Finally, in Albertsons, the Supreme Court amplified its

ruling in Sutton, holding that mitigating measures undertaken

within the body’s own systems, just as those undertaken with the

use of artificial aids like medications and devices, must be

considered in determining whether an individual is disabled under

the ADA. 119 S. Ct. at 2169. The Supreme Court did not consider

whether plaintiff, who had monocular vision, was disabled under

the ADA, but merely held that the statute requires "monocular

individuals, like others claiming the Act’s protection, to prove

a disability by offering evidence that the extent of the

limitation in terms of their own experience * * * is

substantial." Ibid.

The Supreme Court’s holdings in Sutton, Murphy, and

Albertsons do not, therefore, compel a different conclusion than

the conclusion reached by the Court in this case: that Bartlett

is substantially limited in the major life activity of reading.

The record in this case amply demonstrates that the limitations

Bartlett "actually faces are in fact substantially limiting."

Sutton, 119 S. Ct. at 2149. As discussed above, on the basis of

the district court’s findings concerning (1) the shortcomings of

reliance on the Woodcock subtest scores alone, and (2) the

extensive expert testimony, based upon clinical observation,

concerning the manner in which Bartlett reads, this Court

concluded (156 F.3d at 329):

Page 511: Sotomayor Cases

- 13 -

In this case, Dr. Bartlett suffers from alack of automaticity and a phonologicalprocessing defect that significantlyrestricts her ability to identify timely anddecode the written word, that is, to read ascompared to the manner and conditions underwhich the average person in the generalpopulation can read or learn.

Accordingly, the "extent of the limitation in terms of"

Bartlett’s "own experience * * * is substantial." Albertsons,

119 S. Ct. at 2169.

Nor do the attempted self-accommodation techniques employed

by Bartlett change this result: reading remains slow, effortful,

and extremely time-consuming. The record in this case is clear

that the self-accommodations that Bartlett has developed do not

mitigate the crucial element in her dyslexia: her lack of

automaticity in reading. Unlike the situation in Sutton, where

corrective lenses brought the plaintiffs' eyesight to 20/20,

there is no medication or corrective device that can permit

Bartlett to read with automaticity. As this Court found,

individuals with dyslexia suffer a persistent, chronic deficit in

their ability to “decode the written word.” 156 F.3d at 329. As

a result of that impairment, Bartlett always experiences a lack

of automaticity when she reads. Without automaticity, Bartlett

will never be able to read at a rate and in a manner that

approaches the norm, even with her attempts at using the self-

accommodation techniques she has learned. Deciphering words

without automaticity requires an enormous amount of conscious

effort. As the district court noted, Bartlett reads “slowly,

haltingly, and laboriously.” 970 F. Supp. at 1099. “She simply

Page 512: Sotomayor Cases

- 14 -

7/ See Taylor v. Phoenixville Sch. Dist., 184 F.3d 296, 309(3d Cir. 1999) (individual with bipolar disorder who takeslithium to control most severe aspects of disorder can still besubstantially limited in major life activity of thinking becauseof effect of uncontrolled symptoms and side effects of the drug).

does not read in the manner of an average person.” Ibid.

Accordingly, Bartlett experiences substantial "limitations that

persist despite [the mitigating measure]." Murphy, 119 S. Ct. at

2137.

Bartlett’s situation is analogous to the individuals

described by the Court in Sutton who use a prosthetic limb or a

wheelchair for mobility. The Court noted that such individuals

“may be mobile and capable of functioning in society but still be

disabled because of a substantial limitation on their ability to

walk or run.” 119 S. Ct. at 2149. The ADA “addresses

substantial limitations on major life activities, not utter

inabilities.” Bragdon v. Abbott, 524 U.S. 624, 641 (1998).7/

The fact that an individual such as Bartlett has succeeded

in obtaining advanced educational degrees in other fields and has

completed law school does not prevent her from being an

individual with a disability within the meaning of the ADA.

Although individuals with dyslexia such as Bartlett have a

deficit in phonological processing impairing the manner and ease

with which they are able to decipher words, the “higher-order

cognitive and linguistic functions involved in comprehension,

such as general intelligence and reasoning, vocabulary, and

syntax, are generally intact.” Sally E. Shaywitz, Current

Concepts: Dyslexia, 338 New Eng. J. Med. 307, 308 (1998)

Page 513: Sotomayor Cases

- 15 -

(footnotes omitted). This “pattern” helps to explain the

“paradox of otherwise intelligent people who experience great

difficulty in reading.” Ibid.

One of the chief purposes of the ADA is to remove barriers

that prevent persons with disabilities from reaching their full

potential and to allow them to participate fully in society. See

135 Cong. Rec. 8519 (1989) (remarks of Sen. Cranston). In order

for Bartlett to access her higher-order cognitive abilities, she

needs more time than an individual without a phonological

processing deficit to decode and identify the printed word and

she needs other accommodations that would help to compensate for

the effects of that deficit. Just as a person in a wheelchair

can use an above-ground entrance to gain access to a building if

a ramp is available, an individual with a learning disability can

draw meaning from high level text if she is allowed the time she

requires to slowly decipher each word. To such an individual,

time is her ramp. The record demonstrates that Bartlett’s

achievements thus far have come as a result of extraordinary

efforts not required by individuals without disabilities. She

should not be excluded from the protections of the Act because of

accomplishments made despite her disability.

As we have argued, the record is sufficient for this Court

to reaffirm its earlier decision. The fact that Bartlett’s lack

of automaticity is not susceptible to self-accommodation means

that this Court’s conclusion that she is substantially limited in

the major life activity of reading is correct, even when she is

Page 514: Sotomayor Cases

- 16 -

8/ Since the record demonstrates that Bartlett issubstantially limited in reading, we agree with this Court’sdetermination, see 156 F.3d at 329, that it is unnecessary todecide whether she is substantially limited in the major lifeactivity of working.

compared with the average person in the general population. 29

C.F.R. 1630.2(j)(3)(i). The Supreme Court’s remand does not

compel this Court to reach a different conclusion because

Bartlett’s lack of automaticity is not improved by any self-

accommodation or mitigation.8/

CONCLUSION

For the foregoing reasons, this Court should reinstate its

earlier determination that Bartlett is an individual with a

disability who is entitled to accommodations for taking the New

York bar examination. Alternatively, if this Court believes that

further findings by the district court are necessary, it can

Page 515: Sotomayor Cases

- 17 -

remand the case to the district court for a determination whether

Bartlett’s lack of automaticity results in a substantial

limitation in reading.

Respectfully submitted,

MARY JO WHITE BILL LANN LEE United States Attorney Acting Assistant Attorney for the Southern District General of New York

SARA L. SHUDOFSKY JESSICA DUNSAY SILVER Assistant United States MARIE K. McELDERRY Attorney Attorneys

Department of Justice P.O. Box 66078 Washington, D.C. 20035-6078 (202) 514-3068

Page 516: Sotomayor Cases

CERTIFICATE OF SERVICE

I hereby certify that I have served the foregoing

Supplemental Brief for the United States as Amicus Curiae by

mailing two copies, first-class, postage prepaid on counsel of

record at the following addresses:

Eliot SpitzerAttorney General of the State of New YorkPreeta D. Bansal Solicitor GeneralDeon J. Nossel Assistant Solicitor General120 BroadwayNew York, New York 10271

Jo Anne SimonPatricia Ballner257 Dean StreetBrooklyn, New York 11217

This day of November, 1999.

Marie K. McElderry Attorney