Journal of Criminal Law and CriminologyVolume 56Issue 4 December Article 7

Winter 1965

Delinquency and Middle Class GoalsBertram Spiller

Follow this and additional works at: https://scholarlycommons.law.northwestern.edu/jclc

Part of the Criminal Law Commons, Criminology Commons, and the Criminology and CriminalJustice Commons

This Article is brought to you for free and open access by Northwestern University School of Law Scholarly Commons. It has been accepted forinclusion in Journal of Criminal Law and Criminology by an authorized editor of Northwestern University School of Law Scholarly Commons.

Recommended CitationBertram Spiller, Delinquency and Middle Class Goals, 56 J. Crim. L. Criminology & Police Sci. 463 (1965)

THE JoUnN.AL or CRImINAL LAW, CRnUNOLOGY AN-fD POLICE SCIENCE Vol. 56, No. 4Copyright @ 1965 by Northwestern University School of Law Printed in U.S.A.

DELINQUENCY AND MIDDLE CLASS GOALS

BERTRAM SPILLER*

The author is an Associate Professor of Sociology and Anthropology and Acting Head of theDepartment of Sociology and Anthropology at Washburn University of Topeka, Kansas. He is also aResearch Associate at the Menninger Foundation and was formerly a Fellow at the Judge BakerGuidance Center, Boston.

In this article, Dr. Spiller tests the common assumption that much delinquency can be accountedfor by the rejection of or inability to attain middle class goals. In a study of two urban lower classgangs, little evidence was found to support this hypothesis. Instead, greater support was given tobehavior supporting lower class culture, -but there were striking differences between gangs, andeven within segments of a gang, both quantitatively and qualitatively. Prestige in terms of middleclass values was unimportant. However, evidence was found for a continuum among gangs and forbases of prestige to vary with that continuum. Violence and violative behavior, as avenues of prestige,were most important in the lowest class gang, while culturally neutral adolescent behavior wasmore predominant in the less delinquent gang.-Editor.

Recent research in delinquency has stressed thedissociation between culturally approved goals ofsuccess and status-striving and structurally-linked lack of opportunity. Cohen, modelling histheory on the Merton-Durkheim position, statesthat status-discontent, motivated by class-linked opportunities for achievement, which aregenerally closed to the working class population,induces frustration in the working class boy,when he is unable to realize the cultural goals;he reacts by displacing his frustration throughaggressive, often delinquent behavior.' He is aptto flaunt the middle class norms, merely becausethey are middle class, and, for the sake of being"ornery," to espouse those lower class male adoles-cent goals which are opposed to the dominantcode. Cohen's argument rests mainly on the as-sumption that the working class boy is socializedunder the aegis of middle class norms, whichmean something to him, and he must "come toterms" with these norms.2 When he encountersmiddle class institutions, such as schools, and isrejected by them because he cannot measure upto their standards of cleanliness, punctuality,and achievement orientation, he rejects thesenorms, turns them upside down, and engages in"nonutilitarian", "negativistic", and "maliciousbehavior."

* The writer is grateful to National Institute ofMental Health Grant M-4010(A) for assisting in thesupport of the research into this subject. The researchwas the basis for the writer's doctoral dissertation,Bases of Prestige Among High and Lowe DelinquentStreet-Corner Groups (Boston University, 1961).

1 COHEN, DELmNQUENT Boys 186 (1955).2 Ibid. 87.

The assumption that all segments of societyinternalize middle class strivings is just as crucialto Block and Niederhoffer's theory as to Cohen's.Block and Niederhoffer maintain that delinquencyis a result of the frustrated striving for adultstatus. Because our society does not make ade-quate preparation for the induction of adolescentsto adult status, young people spontaneously de-vise their own rituals and symbolic behavior toassist themselves in orienting to adulthood. Theadolescent retreats from the barriers to adult sta-tus, such as compulsory school laws and work cer-tificates, by utilizing the gang as a "structuredoutlet for hostility to the adult social world". 3

While Block and Niederhoffer do not maintain assteadfastly as Cohen that lower class boys internal-ize the American Creed, it nevertheless is crucialto their theory, inasmuch as they maintain thatgang delinquency is the attempt to solve a prob-lem of adolescent role disjunction cnmmn to allsocial classes.

Cloward and Oblin extend the Merton-Durk-heim "illicit means" theory by formulating whatthey call the "illegitimate opportunity structure".4

Since the legitimate avenues to the most prestige-ful and income-producing occupations are almostclosed to the lower classes, the residents of theseslum areas select illegitimate means to achievethe status symbols. If youngsters live in areaswhere there is an integration of adult carriers ofcriminal and conventional values, and limited

3BLocK & NIEDERHOFFER, TE GANG: A STUDY

oF AnOLESCENT BEHAVIOR 138 (1958).4 CLOwARD & 0=aN, DELINQUENy AND OppoR-

TuNY (1960).

BERTRAM SPILLER

access to success-goals by legitimate means, de-linquency will be more or less rational, disciplined,and oriented toward a criminal career. Wherethe illegitimate means are not available, the re-sponse will be that of the typical fighting gang.When individuals cannot meet the criteria of thefighting gang and do not have illegitimate oppor-tunities available to them, they adhere to the"retreatist" or drug addict subculture.5 These for-mulations axe predicated upon two basic assump-tions: 1) the success-goal is universal and is inter-nalized by all strata of American society; 2)residents of "integrated" and "unintegrated"areas do not fully internalize the institutionalnorms which would insulate them from deviantsolutions to their status-discontent.

These theories have been subjected to somecriticism since their inception. Cohen and Shortresponded to the criticism of Cohen's theory bymodifying it slightly. Instead of speaking of ageneric "delinquent subculture", they delineatedseveral subspecies: the "parent male subculture",the "conflict-oriented subculture", the "drug ad-dict subculture", "semi-professional theft", andthe "middle class delinquent subculture." 6 Thisstill left the question of universal internalizationof middle class values unsettled.

Wilensky and Lebeaux noticed the gap in sub-stantive data which Cohen's assumptions raisedand pointedly asked where was the evidence ofnegativism, maliciousness, and nonutilitarianstealingY They specifically noted that "middleclass values and criteria of status are not inter-nalized by all with the same intensity".8 They goon to ask for research relative to the degree towhich lower class delinquents internalize the suc-cess goals, the resources available to them forstatus achievement, and who experiences howmuch status-discontent.

9

Kitsuse and Dietrick also questioned Cohen'sassumptions. They raised similar objections asWilensky and Lebeaux and pointed to certaincontradictory statements in Cohen's work. 10

Then, however, they say that despite the "logicalambiguities and inconsistencies in Cohen's state-ments, it may be maintained nevertheless that

5 bid. 161-186.6 Cohen & Short, Research in Delinquent Subcdtures,

14 J. Soc. IssuEs, 20-37 (1958).7 NVIIENSKY & LEBEAUX, INDusTRIAL SocIETY AND

Soc.AL WErxAR 185-203 (1958).8 Ibid. 202.9 Ibid.10 Kitsuse & Dietrick, Delinquent Boys: A Critique,

24 Am. Soc. Rxv. 208-215 (1959).

empirical research demonstrates the validity of hismajor thesis"." They do not mention this "em-pirical research", nor is the writer aware of whatthis could be.

A study by Gordon, et al. also attacked the samegeneral problem raised here. Their results weresomewhat ambivalent. On the one hand, theyfound that their six groups of gang, non-gang lowerclass and non-gang middle class boys "evaluatedimages representing salient features of a middleclass style of life equally highly". 2 This wouldsupport Cohen, Cloward and Ohlin, and Blockand Niederhoffer. But, on the other hand, therewere sufficient differences between the groups towarrant their saying that Cohen's reaction-formation assumption was not supported. It issignificant that these researchers used the seman-tic differential test. It may be that the lack of ex-pected differences between their study groups wasdue to the tendency of paper and pencil tests tocancel out the differences between verbal and ob-served behavior. The authors implied as much.

Reiss and Rhodes found that there is no simplerelationship between ascribed status and delin-quency. The social structure of the school, thecultural tradition of the neighborhood, and thehomogeneity of the area were variables which hadto be considered.3 Delinquency varied accordingto whether a boy lived in a high status or lowstatus school. Low status boys attending highstatus schools had significantly fewer chances ofbecoming delinquent than low status boys attend-ing low status schools. Thus, there was little evi-dence that low status boys were more prone todelinquency the more they were subjected topressures from middle class norms. This definitelycontradicts Cohen's assumption.

In a recent review of delinquency, Bordua sum-marizes and criticizes the theories of Thrasher,Cohen, Cloward and Ohlin, and Miller. 4 How-ever, he makes little of the basic criticism of Cohen,and Cloward and Ohlin except to indicate thatCohen's presentation of what lower class boys

have or have not internalized is confused.1 5 Bordua

n Ibid. 213.2Gordon, Short, Cartwright, and Strodtbeck, Values

and Gang Delinquency: A Study of Street-Corner Groups,69 Amr. J. Soc. 117 (1963).

13 Reiss & Rhodes, The Distribution of Juvenile De-linquency in the Social Class Structure, 26 Amr. Soc.Xxv. 729 (1961).

14 Bordua, Delinquent Subcultures: Sociological In-

terpretations of Gang Delinquency, Tnm ANNAMS 338(Nov. 1961).

"r Ibid. 130.

[Vol. 56

DELINQUENCY AND MIDDLE CLASS GOALS

apparently recognizes that this is a critical issue inCohen's theory but does not seem to perceive it asespecially damaging to Cohen's position.

Thus, we have seen that the three theoriesbeing considered here pivot around the internal-ization of the American Creed by all social classes.Some critiques of this assumption have beenpresented which question this basic premise. Ifit can be demonstrated, then, that there is a dif-ferential response to this creed depending uponsocial class, all three theories will need revision.The research with which this paper is concernedis aimed at testing whether lower class adolescentstreet-corner groups do use the middle class normsas a "measuring rod" of behavior related to ac-quisition of status.

TAE METHOD

The data which were used to test the middleclass values assumption were those of the files ofSpecial Youth Project Research. These consistedof the "process records" of trained social workerswho interacted with street-corner groups in theirown milieu. Three male social workers maintainedcontinuous contact with four age segments of thetwo gangs being studied here for periods rangingfrom nine to twenty nine months. These processrecords constitute a very rich body of ethnographictype of data of the daily activities of these boysand total 3654 pages of material for the fourgroups.16 Most of this consisted of observed be-havior dictated by the workers, and smalleramounts consisted of interviews with the projectdirector and the research staff, and taped tran-scriptions of group interaction.

This body of material was converted into "be-havior sequences" or "acts" and subjected to astandardized content-analysis system modeledafter the Yale Cross Cultural Survey, but modi-fied for use with the material to be analyzed.Each "act" was based upon the "object-oriented-behavior-sequence ... an observed reported seriesof events in which an identifiable actor or set ofactors orient to an identifiable object of orienta-tion in an identifiable way". 7 Each act was ex-tracted from the records by trained coders (allPh.D. candidates) who typed up each sequence on

16 For more details on the nature of the data, thedistribution of pages per group, dates of collection,etc., see Spiller, Bases of .Prestige Anong High and LowDelinquent Street-Corner Groups (unpublished Ph.D.dissertation, Boston University, 1961) 35-39.

- Walter B. Miller, Records Coding Manual, onfile, Special Youth Project Research, P. 2.

separate data cards and coded the behavior.There were 65 major categories of behavior, suchas sexual behavior, theft, and vandalism, whichcould be utilized. Each set of records was thenchecked independently by another coder for re-liability. In all there were 14,549 typed data cardsfor the four groups being studied here. Each ofthese cards was then duplicated one or more times,according to the number of actors and objects,and filed chronologically in appropriate drawers."

The concept which was used to extract status-oriented behavior from the records was that of"prestige". Prestige is a more fluid version ofstatus and refers to some acquired command orpreeminence of one individual over another andfluctuates continuously. 19 It has the advantage ofbeing readily observed and recorded. It would bevisible in the "deference shown to an individual,in the readiness of others to support him in vary-ing ways, to take his advice, imitate his example,or merely express their admiration or approval"".2

The alternative concepts, status and esteem, wererejected as the major conceptual tool becausestatus is too stable over time, and esteem involvesan invidious value judgment.

In denoting the universe of status-oriented acts,each original typed card was read and evaluated asto whether it contained prestige-oriented materialin the light of Nadel's criteria. It was also de-cided that, for the sake of reliability and impar-tiality, only those acts which were obviouslyprestige-oriented would be denoted as such.It was felt that "reading in" too much might havebiased the data. The total number of prestige-oriented acts during the "effective contact"period of the workers was 5471. Out of this uni-verse of available acts, a 19.5% sample wasrandomly selected. This consisted of 1068 be-havior sequences.

Ideally it would have been desirable to have hadtwo disinterested people code the data-one as areliability check on the other. This was not pos-sible for reasons which cannot be discussed here.One of the authors, who was also a coder, per-formed the coding of the prestige-oriented material.In order to establish some measure of reliabilityof coding, it was decided that the test-retestmethod would be used. This consisted of coding

I's For more details on the coding system, see Spiller,op. cit. supra note 16, at pp. 39-44, 172-176.

9 There is fair-sized literature on this concept, butthe best description is in NADEL, Tan FOUNDATION OrSOCIAL ANTHROPOLOGY 171-172 (1953).

20 Ibid. 171-172.

BERTRAM SPILLER

the first 50 prestige-oriented cards for one of thegroups, then repeating the process after a lapse oftime and comparing the results. The comparisonindicated there was a high degree of correspond-ence between the first and second codings. In 82%of the codings there was complete correspondence,with two categories accounting for most of thediscrepancies. These categories were redefinedmore rigorously so as to eliminate most of theconfusion.

In order to facilitate analysis, the coding wasprogrammed for IBM. All of the categories usedin the larger research project were incorporated.In addition, several categories were added whichwere pertinent to this study. These included fourmodes of prestige-orientation; granting, denying,seeking, and accepting. When prestige is beinggranted, group members either support the be-havior or suggestions of another, follow his ex-ample, admire or approve him, take his advice,or generally defer to him. Denying is the oppositeof granting. Seeking consists of behavior which isdesigned to gain the support, adherence, admira-tion, or generally further or maintain the status ofthe individual within the group. Affirming thatthe support or admiration which has been givenby others to ego was well-deserved is the acceptingmode.

In addition to the machine tabulation of results,the writer perused the original cards to get a "feel"for the qualitative aspects of the material. Thiswas done by arranging the data cards according tothe group rubric, the four modes of prestige-orientation, and the seven major target-roles. Inthis way much of the qualitative "flavor" wasretained and many insights gained.

Finally, all of the behavior categories, such asdrinking, vandalism, and fighting, were classifiedas to whether they were essentially lower class,middle class, or culturally neutral (adolescent).2

1

Each category was conceptualized as dual be-havior which can be engaged in by either lower

21 Lower class adolescent behavior is characterizedby free expression of feelings with intense loyalty to acorporate group (specifically to those displaying ag-gressive masculinity), and pronounced suspicion of out-siders, especially authority figures, resulting in refine-ment of conning techniques, and use-abuse and discardof property. Middle class adolescent behavior is charac-terized by an acceptance of law-supportive institutions,a tendency to defer gratification, and an orientation toachievement by directed work effort, which results in atendency to atomistic behavior. Adolescent, or cul-turally neutral, behavior is defined as pleasure-orientedbehavior such as sports, club activities, outings, danc-ing, dating, and the like.

class or middle class boys. No category of behavioris the exclusive property of either social class,but the overall patterning in each class will differ."Hanging around" is an age-graded activity, butit is relatively unimportant to middle class boys,while it will always exert a high priority in thetypical lower class individual's life. Indices oflower or middle classness for each of these cate-gories were formulated. For example, the lowerclass version of drinking is typically characterizedby open, conspicuous, even flamboyant, heavyto moderate drinking in a group or individualsituation, accompanied by expressed or covertapproval of drunkenness, violative behavior, in-hibition release, and aggression. The middle classversion of drinking is characterized by light tomoderate consumption of alcohol in individual orgroup situations accompanied by a concern forconcealment, disapproval of excessive, conspicu-ous drinking, and condemnation of violative be-havior resulting from drinking.

CHAP4crTERSTICS OF Tu? GRoups

The study consisted of the Junior age segments(ages, 14-16), of the "Bandits" and "Outlaws",and the Intermediate age segments (ages, 16-18)of the same two groups.Y The aggregates of bothgroups have been known to "hang out" at thesame corners in Roxbury, Massachusetts for overthirty years, and are well-known to each other,the police, local inhabitants, and others. Eachpossesses a similar four age-graded aggregate ofabout 100 members each. Although less is knownof the Midget and Senior groups of each aggre-gate, the two groups being studied here were wellmatched for ethnicity (predominantly Irish),religion (predominantly Catholic), and age (anaverage difference of only four months between themean ages of the comparable age segments).n

They differed, however, on some other impor-tant variables. As can be seen in Table 1, the Inter-mediate Bandits were most delinquent, while thethe Junior Outlaws were least delinquent. Thissame gradient of Intermediate Bandits to JuniorOutlaws held true for social class origins (measuredby father's occupation). The Intermediate Banditshad no fathers in the entrepreneur or supervisorycategory, while the ratio increased from 3.8% forthe Junior Bandits, to 10.0% for the Intermediate

22 These are pseudonyms.2For more details on the history, and matching of

these groups, see Spiller, op. cit. supra note 16, at pp.54-58.

[Vol. 56

DELINQUENCY AND MIDDLE CLASS GOALS

TABLE 1GROI' MEMBERS winTE ON.eicAL REco ns: AGEs 10-16

Court Suspended Commit-Appear- Commit- ments

ances ments

Intermediate Bandits 84.4% 46.9% 50.0%N = 32

Junior Bandits N = 34 20.6 14.7 14.7Intermediate Outlaws 15.4 7.7 7.7

N = 26Junior Outlaws N = 24 8.3 0.0 0.0

Outlaws, to 14.3% for the Junior Outlaws. Theratios of skilled and semi-skilled tended to fluctuatebetween 53% and 62%, with no pattern beingdiscernible. However, a similar gradient appearedin the unskilled category: Intermediate Bandits46.5%, Junior Bandits 34.6%, Intermediate Out-laws 35.0%, and Junior Outlaws 23.7%. Thegradient in educational attainment by the fourgroups was less obvious and tended more to dif-ferentiate between the aggregates. There were nocollege entrants among either Bandit group,while the Intermediate Outlaws had 19.2%, andthe Junior Outlaws had 4.2%. On the oppositeside of the ledger, 15.6% of the Intermediate Ban-dits did not complete grammar school, 2.9% of theJunior Bandits did not complete grammar school,none of the Intermediate Outlaws failed to com-plete grammar school, and 8.3% of the JuniorOutlaws failed to do so.n

Thus, it can be seen that there was an associa-tion among higher delinquency, lower classorigins, and early school drop-outs; and con-versely, lower delinquency was associated withhigher social class origins and more education.What is striking about these findings is that suchminor differences in social class origins and educa-tional attainment should be so clearly related tovarying rates of delinquency. There was not a par-ticularly large social class gap between the Inter-mediate Bandits and Outlaws, yet, their delin-quency rates differed greatly. Not only did ratesof delinquency differ between the matched agegroups, but between brother groups as well.Small differences in social class origins wereassociated with, sometimes, drastically differentdelinquency rates. In a comparison of the Banditand Outlaw neighborhoods, it was also found that,although both neighborhoods would be described

24 See Spiller, ibid. 59-65, for more details on socialclass and education.

TABLE 2ORIENTATIONS TO LOWER CLASS, MIDDLE CLASS, AND

ADoLEscENT BEHAVIOR

Behavior Intermediate Junior Intermediate juniorAreas Bandits Bandits Outlaws Outlaws

N = 473 N = 472 N = 514 N = 435Lower Class 56.0% 32.2% 24.7% 24.8%Adolescent 41.0 64.4 70.5 70.8Middle Class 3.0 3.4 4.8 4.4

as "depressed areas," the Bandit area had morecrowded conditions, less home ownership, greaterincidence of dilapidated dwellings, and lowervalued homes.25 Thus, the conditions tradition-ally associated with lower social classes were alsomore strongly evident in the ecological realm aswell. Social class origins, then, would appear tobe the most cogent variable as far as delinquencyamong these groups is concerned.

TsE FINDINnGs

If the middle class values hypothesis is to besupported, we would expect that the study groupswould exhibit a great interest in things middleclass. As Table 2 indicates, none of the groupsexhibited anything but passing interest in prestige-oriented middle class behavior. What is strikingis that as social class declines, the proportion oflower class oriented status behavior increases,and as social class rises, the ratio of adolescentbehavior increases. There is also a slight tendencyfor middle class behavior to increase with socialclass, but the N's are too small to be accepted withany confidence.

Since these figures contain all modes of orienta-tion in all types of social interaction, it could beobjected that the almost exclusively adolescentmilieu unduly influenced the results. Therefore,the percentages of interaction towards adulttargets only were calculated. Table 3 demonstratesthat the same general relationships between socialclass, delinquency, and commitment to behaviorremained essentially the same. There was a gen-eral increase in orientation to middle class statusorientations, but it was hardly striking and didnot indicate any basic shift in outlook. In fact,the orientation to lower class behavior was evenmore supportive of the gradient from Intermediate

,5 It was also found that 70-80% of the members ofboth sets of groups lived within 12 contiguous blocks.For more details on the demographical data, see Spiller,ibid. 45-54.

19651

BERTRAM SPILLER

TABLE 3

ORIENTATIONS TO LOWER CLASS, MIDDLE CLASS, AND

ADOLESCENT BEHAVIOR: ADULT TARGETS

Behavior Intermediate Junior Intermediate JuniorAreas Bandits Bandits Outlaws Outlaws

N= 156N= 173 N=91 N= 121Lower Class 50.0% 31.2% 30.8% 24.0%Adolescent 45.5 64.2 59.3 69.4Middle Class 4.5 4.6 9.9 6.6

100.0% 100.0% 100.0% 100.0%

Bandits to Junior Outlaws. Another fact worth

noting is that both Bandit groups were more in-volved with adults in all types of situations thanthe Outlaws so that theoretically they had greateropportunity to demonstrate any middle classstatus problems which they might have had.

To further substantiate this contention, theorientations when the groups were not interactingwith their social workers and other local, presum-ably lower class, adults were determined. Table 4demonstrates that even when they were interact-ing with middle class adults (teachers, social

workers, businessmen, etc.), the results still con-formed to the gradient. The fact that the Banditsdid not utilize middle class avenues of prestigewhen interacting with middle class adults atteststo the entrenchment of lower class values. It alsotends to vitiate the middle class values hypothesis.If the Cohen, Cloward and Ohlin assumptions weretrue, these boys should have increased their middleclass status-oriented behavior, when in the pres-

ence of such individuals.26 The fact that theywere more adolescent and lower class orientedindicates their cultural foci.

26 Spiller also computed the orientations of the groupto the three behavior areas under each of the fourmodes of prestige-orientation. The same gradientpattern appeared in the Conferring and Denying modes,but in the Seeking mode the junior Outlaws were morelower class oriented than the Intermediate Outlaws, butless so than the Junior and Intermediate Bandits. Thiswas not due to any "status dilemma" but to thetemporary "desertion" by their worker. He transferredhis major attention to the Intermediates for severalmonths, and during this period, a more delinquentclique of Juniors attempted to demonstrate how "bad"they could be so that the worker would return to them.53.2% of their behavior was adolescent-oriented evenin this mode and this was still greater than eitherBandit group. Numerically, the seeking mode was notas important as the conferring and denying modes, soit was not felt that this single exception nullified thegradient pattern. The accepting mode was numericallyunimportant, and is therefore not included in thesediscussions. See Spiller, ibid. 108-111 for more details.

TABLE 4

ORIENTATIONS TO LOWER CLASS, MIDDLE CLASS, AND

ADOLESCENT BEHAVIOR: NON-LOCAL ADULTS ONLY

Behavior Intermediate Junior Intermediate juniorAreas Bandits Bandits Outlaws Outlaws

N=31 N=54 N=23 N=30Lower Class 58.0% 29.6% 17.40 16.7%Adolescent 42.0 70.4 73.9 76.7Mfiddle Class 0.0 0.0 8.7 6.6

100.0% 100.0% 100.0% 100.0%

DIFFERENTIAL AvENuES TO PRESTIGE AMONGTHE INTERMEDIATES: LOWER CLASS

BEHAVIOR

Although it has been shown that street-corner

groups, even within the same general milieu,can vary considerably in the degree to which theyadhere to lower class or adolescent bases of pres-tige, the exact nature of the avenues has not beendelineated. This study was directed not only attesting the middle class values assumption, butalso towards casting more light on differentialavailable avenues to status.

As can be seen from Tables 5, 6, and 7, there isvery little correspondence between the Interme-diate groups in rank order of selection of alterna-tive avenues. The Bandits tended to select themore law-violating avenues (such as theft, fighting,and trespassing) more frequently, and consist-ently, and to rank them higher. In each table thefirst four or five categories among the Bandits hada high law-violating potential. This was also trueamong the Outlaws, but the latter rated Belong-ingness (group loyalty) among the first four in eachinstance. There was little or no law violation in

this category. In a number of instances the Out-laws ranked potentially law-violating behaviorhigher, but utilized it less frequently than theBandits, who ranked it lower. Fighting, for ex-ample, in Table 5 was ranked fourth by the Ban-dits and 2.5 by the Outlaws but both oriented toit seven times. In Table 6, Drinking was rankedfirst by the Outlaws when seeking prestige butreceived fewer orientations than among the Ban-dits, who ranked it fourth. The reverse was nevertrue; the Bandits never ranked an area higher andutilized it less frequently. This indicates that the

two groups had different bases of prestige. TheBandits relied heavily upon the whole gamut oflower class behavior in all situations, whereas the

[Vol. 56

DELINQUENCY AND MIDDLE CLASS GOALS

TABLE 5

RANEGs OF LOWER CLAss BEHAVIOR FOR ALL TARGETS: CONFERRING iMOPE*

Intermediate Bandits Intermediate Outlaws

Behavior Number of Rank Behavior Number ofOrientations Orientations

Jobs 16 1.5 Drinking 12 1Theft 16 1.5 Belongingness 7 2.5General Violative Behavior 11 3 Fighting 7 2.5Fighting 7 4 General Violative Behavior 4 4Drinking 6 5.5 Hanging 2 5.5Absenting 6 5.5 Profanity 2 5.5Hanging 4 7.5Creating a Disturbance 4 7.5Sex 2 10Informing 2 10Trespassing 2 10Vandalism 1 12Totals 77 34

* Rank difference correlation was not computed for these rankings because of the small N of four.

TABLE 6

RANKINGS OF LOWER CLAss BEHAVIOR WmN SEEKING PRESTIGE: ALL TARGETS

Intermediate Bandits Intermediate Outlaws

Number of Raul Behavior Number of RankBehavior Orientations Orientations

Theft 16 - Drinking 8 1General Violative Behavior 13 1 Fighting 5 2Drinking 12 2 Jobs 4 3

Jobs 10 3 Belongingness 2 4.5Trespassing 8 - Sex 2 4.5Fighting 7 4 Vandalism 1 7Creating a Disturbance 4 5.5 General Violative Behavior 1 7Sex 4 5.5 Creating a Disturbance 1 7Vandalism 3 7Gaming 3 -

Absenting 2 -

Belongingness 2 8Tattoos 2 -

Hanging 1 -

Informing 1 -

Profanity 1 -

Totals 89 24

ra = +.30; N 8; P > .05.X2 + - i)

The formula utilized for the correction of ties was: r. 2 -2 -\/T--&--2

See SrEGEL, NONPARAmETBiC STATISTICS 206-210 (1956).

Outlaws were much more selective. They dis- unaware of the panorama of lower class behavior

approved of more aspects of lower class behavior as they utilized all but two areas among the three

but utilized fewer areas when seeking and con- modes. Table 7 is also somewhat deceptive in that

ferring prestige. One could not say that they were it appears that the Bandits disapprove of more

1951

BERTRAM SPILLER

TABLE 7

RANIoNGS OF LOWER CLASS BEHAvIOR WHEN DENYING PRESTIGE: ALL TARGETS

Intermediate Bandits Intermediate Outlaws

Behavior Number of Number of RankOrientations Rank Behavior Orientations

Jobs 18 1 Fighting 13 1Theft 13 2 Drinking 10 2General Violative Behavior 12 3 Belongingness 8 3Drinking 10 4.5 Absenting 6 4.5Fighting 10 4.5 Creating a Disturbance 6 4.5Sex 6 - Theft 4 6

Vandalism 5 6 Jobs 3 7Belongingness 4 8 Informing 1 10Absenting 4 8 Vandalism 1 10Creating a Disturbance 4 8 Gaming 1 10Gaming 3 10.5 General Violative Behavior 1 10Hanging 3 - Profanity 1 10Informing 3 10.5Profanity 2 12Trespassing 2 -

Totals 99 55

r. = +.31; N = 12; P > .05.

aspects of lower class behavior than the Outlaws.Such is not the case, however. Because of a coding

artifact Table 7 includes both those acts whichwere disapproved because boys engaged in drink-ing, fighting, vandalism, etc., and also those whereboys disapproved because they did not participatein theft, drinking, profanity, etc. In each caseapproximately half were disapproved acts andhalf disapproved for not engaging in specifiedbehavior. Thus, Table 7 actually indicates addi-tional support of violative behavior by the Ban-dits. They very actively supported violative be-havior of all kinds. They can appropriately becalled a "band of thieves" because of the fre-quency of their raids on local and downtownstores. They were expert shoplifters, and fre-quently returned to the neighborhood after aday's depredation laden with goods, which theywould then ritually display to each other.

What is more important than these frequencyrankings is the qualitative patterning of prestige-ful behavior. It will be noted that "Jobs" rankedfirst or third among the Bandits' approved anddisapproved acts. They supported a pattern ofworking which has been called "voluntary unem-ployment" by the writer. This consisted of quittingjobs frequently for unimportant reasons, quittingen masse, behaving so hostilely towards prospec-tive employers that they were not hired, or acting

in such a way as to invite firing. It was sometimesmore important to the group to be on band towelcome back a boy from reform school or to go onan outing than to work. Many would work onlyif they could work with four or five other groupmembers in the same factory. Sometimes theywould seek work in groups of four or five and giveup if the pay was too low or if they could not allwork together. In many ways the group deter-mined the appropriate time and pay for workand the appropriate time to quit. Group memberswere extremely sensitive to work and wouldmanipulate the prestige of the worker by acceptingor rejecting his job-finding activities.

The Bandits were also sensitive to their fam-ilies' pressures to obtain work. They were some-times ordered out of their homes early in the morn-ing and told to find work. Just as frequently theywould treat the matter laughingly and spend theday hanging with the gang at the corner. How-ever, they did occasionally accede to intense par-ental job pressures. Another figure to whom theyoften deferred was the local parole officer. Mostof the Bandits were parolees and were requiredto work or go to school. Much of the time theymanaged to "con" the parole officer by promisingor feigning to work or attend school. One boy,who was ordered by the parole officer to find workand given money for traveling to the employment

[Vol. 56

DELINQUENCY AND MIDDLE CLASS GOALS

office, took the money, called the office by tele-phone, and pocketed the extra change. This tookplace in the presence of the group so the boy wasable to fulfill his legal obligation, "con" the paroleofficer, and raise his group prestige simultane-ously.

The comparable situation among the Outlawswas almost completely lacking. It was generallyaccepted that all who could should work and nospecial prestige accrued from working. Some Out-laws even worked after school or during summervacation; this would have been unheard of amongthe Bandits, even if they had been in school.The only Outlaw who received any praise forworking was a boy with a deformed leg. His case,however, was exceptional.

The Bandits' well-established patterns of workbehavior, well-developed work attitudes, and ra-tionalizations led the writer to conclude that oneof the latent functions of their street-cornergroup was to prepare them for the exigencies oftheir expected masculine adult occupationalstatus. By receiving daily training in the patternof voluntary unemployment they were becomingconditioned to what was to become a highlyprobable adult employment situation-periodicunemployment. Since it was unlikely that manyBandits would acquire complex labor skills, thestatistical probability of their being unemployedduring future recessions was high. Most of themcame from such families. Thus it was essentialfor them to have a set of rationalizations for cop-ing with these almost certain economic and ego-deflating experiences. They acquired these in thegang. These attitudes and rationalizations in-cluded such things as feeling that money is im-portant for such things as dressing "sharply" butnot important enough to motivate full-time em-ployment. One boy expressed the attitude thatthe ideal situation was to work eight or nine weeksand loaf two or three. He also felt it was better towork in the winter, when it was too cold to "hangon the corner", and to loaf during the summer.Attitudes such as these supported voluntaryunemployment and enabled them to develop

psychological defenses against feelings of worth-lessness for being unemployed. This helps to

explain why so many lower class adult males canloaf for long periods without suffering severe egodamage. Their egos are already preconditioned.

From the almost total absence of the patternof voluntary unemployment among the Outlaws,

it seems safe to conclude that their group did notperform this function.

Adolescent Behavior

When we examine the Bandit-Outlaw activitiesin the adolescent sphere, we find that there is morecorrespondence of avenues of prestige. As Tables8, 9, and 10 indicate, the correlations were quitehigh in the conferring and seeking modes but lowin the denying. Since behavior relative to sports,dancing, music, club jackets, and the like is cul-turally neutral, it is not surprising that the Inter-mediates should agree for the most part in suchareas. It will also be noted that both groupsutilized virtually the same categories, with theexception that the Outlaws actually utilized threefewer than the Bandits (Dancing, Touring, MIu-tual Affection). However, this does not mean thatthe Bandits were more adolescent oriented. TheOutlaws utilized these avenues much more fre-quently. They developed a well-organized clubin which meetings, sports, uniforms, distinctiveclub jackets, and fund raising activities werevery important. These became the major avenuesof increasing personal prestige. Among the Ban-dits, on the other hand, there was only an abortiveclub. One clique of sports-minded boys attemptedto channel the group into typical adolescent ac-tivities, but two leaders of the more delinquentcliques scotched these attempts. These latter twoboys possessed more influence and tended todominate the group as a whole. They counteractedthe influence of both the few sports-minded boysand the social worker. The latter was also rela-tively helpless in diverting the group towardsthe law abiding avenues of prestige. It should benoted that he was not a particularly effectivesocial worker in this milieu, but it is doubtful ifany other social worker would have been muchmore effective. These boys had too well-establishedavenues of prestige and were reached at too latean age.

As in the case of lower class behavior, the quali-tative differences in adolescent categories wasalso differentiating. At parties, for example, theBandits tended to become drunker and moreboisterous. They frequently became so boisterousand noisy that the police were called. At one of

their parties, several boys threw empty liquorbottles from the roof of the building. The Outlawsseldom created so much disturbance that theirhouse parties were broken up by police. They

BERTRAM SPILLER

TABLE 8RANEII'IGS op ADOLEScENT SuPPoRTED BEHAvIOR: CONFERRING MODE

Intermediate Bandits Intermediate Outlaws

Behavior Number of Rank Behavior Number of RankOrientations Orientations

Club and Athletics 16 1 Club and Athletics 55 1Money 6 2.5 Mutual Hostility 14 2Mutual Hostility 6 2.5 Purchases 11 3Purchases 5 4 Money 5 4Area Office 4 5.5 Ceremonies 4 -Mating 4 5.5 Gift Giving 3 -Automobiles 4 - Mating 3 5.5Parties 3 7 Parties 3 5.5Dancing 2 - Music 2 7.5Music 2 8.5 Entertainment 2 7.5Entertainment 2 8.5 Area Office 1 9Mutual Affection 2 -Smoking 2 -

Totals 48 103

r. = +.83; N = 9; P < .01.

drank quite freely, but derogated members whoexceeded their known capacities.

There was also a difference in their acceptedstyles of mating. The Bandits were very roughwith their girl friends, conceptualized them as sexobjects only, and "conned" them wheneverthe opportunity arose. The Outlaws also rough-housed with local girls but were not as sadistic.They tended to see them more as potential mar-riage partners, although they were not averseto taking advantage of them sexually if circum-stances allowed. They also were less inclined to"'con" girls for other ulterior motives. To give anexample of Bandit behavior toward girls there isthe instance of one boy who, while drunk, pushedhis girl down a flight of stairs, deliberately spilledice cream down her dress, and suggested shecopulate with other gang members, and evencomplete strangers.

Another important difference was in the orien-tation to Mutual Hostility.u The Bandits had adecided tendency to use hostility less frequently,but to explode when they did. The Outlaws reactedtowards each other with more frequent but lowlevel hostility. These differences were statisticallysignificant at the .01 level (chi square). For ex-

27 Mutual Hostility refers to both overt and covertexpression of aggression. Overt hostility could take averbal or actional form, while the "joking relationship"consisted of patterned, more institutionalized verbalforms of aggression, such as ribbing, joking, "ranking","playing house", etc.

ample, the Bandits beat one boy for going swim-ming alone with the worker. Another boy brokehis wrist in the process of badly beating anotherboy for stealing his girl. Informers were usuallygiven severe beatings. Intragroup violence of thisorder was never reported among the Outlaws. TheOutlaws generally invoked threats, ostracism, andother verbal sanctions, rather than physical ag-gression. Both groups used the "joking relation-ship" for similar purposes (to give the group orindividual an opportunity to enjoy someone else'sdiscomfiture), but the Outlaws employed thisdevice at a statistically less frequent level.2

Thus, although the groups utilize virtually thesame adolescent avenues of prestige, they differgreatly both in frequency and quality of use.

COMPARATIVE PATTERNS AMONG THE JUNIOR

BANDITS AND OUTLAWS: LOWER CLASSBEHAVIOR

The juniors' utilization of lower class avenuesof status was different in some ways from theIntermediates'. As Tables 11, 12, and 13 indicatethere was a high correlation only in conferring.This contrasts with the Intermediates, where theIntermediate Outlaws conferred prestige so in-frequently in lower class terms that it was notpossible to compute a correlation. In the other

23See Spiller, op. cit. supra 16, at pp. 129-132 formore details on hostility and the joking relationship.

[Vol. 56

DELINQUENCY AND MIDDLE CLASS GOALS

TABLE 9

RANKINGS OF ADOLESCENT SUPPORTED BEHAVIOR: DENYING MODE

Intermediate Bandits Intermediate Outlaws

Behavior Number of Rank Behavior Number ofOrenatonaviorhaio RnOrientations Orientations

Mutual Hostility 53 1 Club and Athletics 86 1Club and Athletics 13 2 Mutual Hostility 66 2Automobiles 6 3 Mating 9 3Ceremonies 4 4 Purchases 8 4Area Office 3 6.5 Money 7 5Money 3 6.5 Area Office 3 6Mating 3 6.5 Music 2 -Purchases 3 6.5 Automobiles 2 7Entertainment 2 9 Entertainment 1 9Parties 1 - Smoking 1 -Touring 1 - Ceremonies 1 9Gift Giving 1 10 Gift Giving 1 9Totals 93 187

r. = +.59; N = 10; P < .05.

TABLE 10

RumINGs or ADOLEScENT SupPORTED BEHAVIOR: SEEKING MODE

Intermediate Bandits Intermediate Outlaws

Number of Number of RankBehavior Orientations Rank Behavior Orientations

Club and Athletics 10 1 Club and Athletics 28 1Mutual Hostility 9 2 Mutual Hostility 12 2Money 7 3 Purchases 8 3Mating 6 4 Mating 5 4Parties 5 5.5 Entertainment 4 5Purchases 5 5.5 Money 3 6.5Entertainment 2 7 Parties 3 6.5Area Office 1 9 Automobiles 2 8.5Automobiles 1 9 Ceremonies 2 8.5Ceremonies 1 9 Music 2 -

Area Office 1 10Totals 47 70

r. = +.84; N = 10; P < .01.

two modes, however, there were low correlations.Although there were somewhat higher correlationsthan between the Intermediates, rank order cor-relation is not discriminating enough with such asmall N to warrant the conclusion that the Juniorsagreed more on lower class avenues of prestige.It is apparent that a similar divergence in statusavenues existed between the Juniors. The JuniorBandits chose lower class avenues more frequentlythan the Outlaws except in the seeking mode.However, this does not necessarily mean that theJunior Outlaws were more lower class oriented

when seeking prestige. This result was due to theaforementioned "sibling rivalry" between theJunior and Intermediate Outlaws for the servicesof their social worker. Despite the delinquentactivities of one clique among the Junior Outlaws,the group as a whole remained essentially adoles-cent oriented.

Once again, though, it was in the qualitativeaspects that the Juniors differed more dramati-cally. The Bandits were more committed to dis-approved aspects of the same lower class behavior.The Bandits actively supported and rewarded

19651

BERTRAM SPILLER

TABLE 11

RANKINGS OF LOWER CLASS BEHAVIOR BY THE JUNIORS WHEN CONFERRING PRESTIGE: ALL TARGETS

Junior Bandits Junior Outlaws

Behavior Number of Rank Behavior Number of RankOrientations Orientations

Fighting 18 1 Fighting 6 1Drinking 6 - Jobs 3 2Vandalism 6 2.5 Theft 2 4Jobs 6 2.5 Vandalism 2 4Creating a Disturbance 5 4.5 Creating a Disturbance 2 4Theft 5 4.5 Gaming 2 4Belongingness 4 6 Profanity 2 -Trouble 4 - Absenting 1 7.5Hanging 2 - Belongingness 1 7.5Gaming 2 7Absenting 1 8Informing 1 -Totals 60 21

r. = +.81; N = 8; P < .05.

TABLE 12

RANKINGS OF LOWER CLASS BEHAVIOR BY THE JUNIORS WhEN DENYING PuEsTIGE: ALL TARGETS

Junior Bandits Junior Outlaws

Number of Number ofBehavior Orientations Rank BehaviorOrientations Bank

Fighting 23 1 Fighting 12 1Creating a Disturbance 10 2 Trouble 8 2Drinking 7 3 Drinking 5 3Trouble 6 4 Belongingness 4 4Theft 5 5 Absenting 3 6Absenting 3 6 Profanity 3 6Sex 2 7.5 Theft 3 6Vandalism 2 7.5 Hanging 2 9Gaming 1 11 Sex 2 9Belongingness 1 11 Vandalism 2 9Hanging 1 11 Creating a Disturbance 1 12Jobs 1 11 Gaming 1 12Profanity 1 11 Jobs 1 12Totals 63 47

I. = +.45; N 13; P > .05.

participation in gang fights, vandalism, boastedof or admired theft, eagerly supported drinkingbouts, etc. The Outlaws tended to engage in ver-bal support of these activities, but engaged lessfrequently in actional support. Where the Banditsactually participated in and fomented gang fights,the Outlaws threatened violence but seldom im-plemented their threats. They did not participatein one full scale gang fight during their worker'stenure, whereas the Bandits engaged in several.

The high rank which fighting held in the Out-laws' orientation to lower class behavior wasmainly a coding artifact-actions and verbaliza-tions were combined. The situation was similarin other categories as well. Where the Banditscarelessly or wantonly damaged property, anddefended the behavior, the Outlaws heeded theirworker's advice not to. Where the Bandits stolebeer, gasoline, golf balls, etc., and boasted of it,the Outlaws jokingly spoke of attempting hold-

[Vol. 56

DELINQUENCY AND MIDDLE CLASS GOALS

TABLE 13

RANxINGS OF LOWER CLASS BEHAVIOR BY THE JUNIORS WBEN SEEKING PREsTIGE: ALL TARGETS

Junior Bandits Junior Outlaws

Number of Rank Behavior Number ofBehavior Orientations Orientations Rank

Fighting 8 1 Fighting 9 1Theft 4 2.5 Jobs 8 2Trouble 4 2.5 Drinking 5 3.5Creating a Disturbance 3 4 Theft 5 3.5Vandalism 2 - Belongingness 2 5.5Absenting 1 7.5 Creating a Disturbance 2 5.5Belongingness 1 7.5 Profanity 2 -Drinking 1 7.5 Absenting 1 8.5Gaming 1 7.5 Gaming 1 8.5Hanging 1 7.5 Hanging 1 8.5Jobs 1 7.5 Informing 1 -Trespassing 1 - Trouble 1 8.5

Sex 1 -

Totals 28 39

r. = +.34; N 10; P > .05.

ups. The Outlaws found less lower class behaviorto criticize and generally were concerned withless serious forms of the same conceptualizedbehavior. Both ranked fighting first as a form ofdenying prestige; but the Outlaws playfullypunched each other, pelted each other with wadsof paper, or threatened assault, while the Banditsthrew bottles or lighted matches, or threatenedbodily harm with a knife. This order of differenceoccurred throughout the records; these exampleshappened to appear in the sample population.

This is not intended to deny that the Outlawsdid not engage in any violative behavior. Theyalso stole autos, drank, and damaged property,but it occurred much less frequently. They ex-erted more stringent controls and regulation ofbehavior which disrupted orderly group activityand continuously appealed to the worker to exertmore control. They were reluctant, however, toenforce the norms regarding law-violating be-havior. Their major concern was with club andathletic activities rather than with illegal be-havior. The Bandits were much more prone tooverlook minor activity-disrupting behavior aswell as minor and major law-violations.

In the matter of orientation to work, the Out-laws were more likely to boast of having a job or ofdefending their ability to get jobs. Having a jobinferred adultness, independence, and masculin-ity. It was also an easy way of gaining the ap-proval of their worker. One of the principal

functions of a job among Bandits appeared to beits tendency to keep them out of trouble. The mostdelinquent Junior Bandit expressed this equationseveral times. There was little indication thateither group actively supported the voluntaryunemployment pattern. There were hints, though,that the more delinquent element of Bandits wouldhave given this support if the social worker hadnot been present. Considering that they were notas delinquent, nor as lower class, at the samestage of development as their brother segment,it is unlikely that they would have supportedthis pattern as fervently as the IntermediateBandits. The Junior Outlaws gave no indicationat all that they supported voluntary unemploy-ment; they were eager to make money and becomeself-sufficient.

The other major avenue of prestige-drinking-tended to rank higher among the Bandits, but tobe emotionally more important to the Outlaws.This was probably because theft, trespassing,vandalism, etc. were less important to them. Since

they did not give group-wide support to behaviorwith a high potential of law-violation, drinking

loomed larger because it was exciting and of lowlaw-violating potential. Where the first fiveranked behaviors among the Bandits had a highlaw-violating potential, the Outlaws selected only

two-fighting and theft-among their first five.For both groups drinking ostentatiously was a

19651

BERTRAM SPILLER

symbol of adulthood and masculinity and a major three modes indicated there were more similar-means of prestige. ities than differences. They were obviously more

nearly alike than the Intermediates. Where theAdolescent Behavior latter had only two high correlations, the Juniors

There were even fewer differences between the exhibited high correlations in all three modes.Juniors when orienting to adolescent supported This is consistent with their more equidistantbehavior. Tables 14, 15, and 16 show that they class origins.supported nearly the same behavior in nearly the It is obvious that both groups gave primarysame rank order. The high correlations in the importance to athletics and club activities. They

TABLE 14RANKING or ADoLEscENT SUPPORTED BEHAVIOR BY THE JUNIORS: CONFERRING MODE

Junior Bandits Junior Outlaws

Number of Rank Behavior Number of RankBehavior Orientations Orientations

Club and Athletics 78 1 Club and Athletics 46 1Purchases 15 2 Mutual Hostility 11 2Mutual Hostility 11 3 Purchases 7 3Money 6 4 Area Office 5 4.5Automobiles 3 - Mating 5 4.5Entertainment 2 5.5 Entertainment 4 6Playing Facilities 2 5.5 Dancing 3 -Area Office 1 8 Mutual Affection 3 -Gift Giving 1 8 Money 2 7.5Mating 1 8 Music 2 -Parties 1 - Playing Facilities 2 7.5Touring 1 - Ceremonies 1 -

Gift Giving 1 9Totals 122 92

r. = +.62; N = 9; P < .05.

TABLE 15RANKiNGS OF ADOLEscENT SUPPORTED BEHAVIOR BY THE JUNIORS: DENYING MODE

Junior Bandits junior Outlaws

Number of Raul Behavior Number of RankBehavior Orientations Orientations

Club and Athletics 59 1 Club and Athletics 78 1Mutual Hostility 49 2 Mutual Hostility 55 2Purchases 10 3 Mating 11 3Mating 5 4.5 Purchases 6 4Money 5 4.5 Money 3 5.5Automobiles 3 6 Parties 3 5.5Ceremonies 2 - Music 2 7Dancing 2 8 Racing 2 -Gift Giving 2 - Automobiles 1 10.5Playing Facilities 2 8 Dancing 1 10.5Smoking 2 8 Playing Facilities 1 10.5Music 1 10.5 Smoking 1 10.5Parties 1 10.5Totals 143 164

r. = +.62; N = 11;P < .05.

[Vol. 56

DELINQUENCY AND MIDDLE CLASS GOALS

TABLE 16RAxKINGS Or ADOLESCENT SUPPORTED BEHAVIOR BY THE JUNIORS: SEEKING MODE

Junior Bandits Junior Outlaws

Number of Raul Behavior Number ofBehavior Orientations Orientations Rank

Club and Athletics 12 1 Club and Athletics 16 1Mutual Hostility 6 2 Mutual Hostility 13 2Purchases 4 3 Mating 5 -Ceremonies 3 4.5 Purchases 5 3Money 3 4.5 Area Office 2 -Dancing 1 6 Dancing 2 5Entertainment 1 - Ceremonies 2 5

Money 2 5Music 2 -Automobiles 1 -

Smoking 1 -

Totals 30 51

ra = +.96; N = 6; P < .01.

either conferred or sought prestige most frequentlyor criticized members for poor performance inthose activities. Both groups had well-liked,effective social workers who aroused their en-thusiasm for dub-associated activities. The symbolwhich both groups fixed upon was the club jacket.Most of the references to Purchases refer to clubjackets. Much dub activity was focused aroundjacket fund raising. Most other adolescent ori-ented behavior was similar. They admired skillin athletics, ridiculed physical ineptness, enviedboys who had a way with the girls, rewardedgroup co-operation, derogated uncooperativeness,admired skill in dancing, teased awkward boys,and the like. There was remarkable uniformity inmost activities labeled adolescent.

One area in which they differed, as did theirolder brother segments, was Mutual Hostility.The Junior Outlaws were less inclined to use thejoking relationship. They were more inclined touse hostility rather than the joking relationship.This difference reached the .02 level of signifi-cance.u However, among the Intermediates itreached the .01 level. This lower significancelevel conforms to the generally fewer differencesbetween the Juniors, but it also substantiates theimportance of small social class differences.

Although the Outlaws used hostility more fre-quently, it was of a low level of intensity.30 They

19 Ibid. 153.30 Miller, Geertz, and Cutter, Aggression in a Boys'

Street-Corner Group, PSYCHIATRY 24 (Nov., 1961). Thisarticle discusses in great detail hostility among theJunior Outlaws.

seldom exploded or used sadistic tactics in enforc-ing norms. They were very concerned about thenorms but hesitated to endanger group solidaritywith highly charged reactions. There was muchbickering and carping but little violence. TheBandits, on the other hand, like their older seg-ment, did not use hostile gestures so frequently,but when they did, it was often explosive but didnot disrupt group solidarity. This was probablydue to greater agreement on norms among theBandits so that violent enforcement was not per-ceived as disruptive. Since they came from rela-tively homogeneous class backgrounds, there wasprobably greater social consensus. This seems tobe reflected in their lesser use of hostility. TheOutlaws derived from more heterogeneous classbackgrounds and probably did not possess thesame degree of values consensus. This was re-flected in their more frequent but low level bicker-ing. As Simmel noted, groups which fear to dis-rupt a tenuous social bond do not allow themselvesthe luxury of violent reactions.3'

CONCLUSIONS

We have seen that Cohen, Cloward and Ohlin,and Bloch and Niederhoffer assumed the uni-versality of the internalization of middle classsuccess goals. This hypothesis was subjected totesting and was not supported by observed inter-actional data. It was found that the more delin-quent groups were more committed to lower classvalues and focal concerns, while the less delinquent

a SnmL, CoHmcT 45-48 (1955).

19651

BERTRAM SPILLER

groups were more supportive of culturally neutraladolescent avenues of status. Middle class avenuesof prestige were numerically unimportant. More-over, this relationship did not change significantlyeven when the interactional arena was middleclass; all groups maintained, or even strengthened,their essential commitment to lower class oradolescent avenues. Contrary to those who main-tain the middle class values assumption, the dataindicated that lower class gang boys are involvedin their own little world relatively isolated fromthe middle classes. The general tendency of socialscientists to assume that their own middle classposition is the measuring rod of all classes is con-tradicted by these data. Psychoanalyticallyoriented people have been particularly prone totake this view, but it is also common amongsociologists. It is naturally more tempting to viewdelinquency as a "sickness" since then one avoidsexamining the cultural background and concen-trates on the "deviant" aspects. What is usuallyoverlooked is the specifying of what is being

"deviated" from, and in terms of whose definition.The tendency to be defensive about one's ownvalue position has probably prevented manyfrom perceiving anything but emotional sicknessor social disorganization in lower class delin-quency. Miller and Sutherland are exceptions,but their viewpoints are in the minority.

The data of this study are limited in generaliz-ability because of the small sample of four groupsfrom one city, but Hollingshead, Crawford, et al.,and Whyte have reported behavior which issupportive.u These varied sources, as well as thisstudy, indicate that lower class gang membersare motivated more by attempts to achievestandards and measure up to qualities valuedwithin their own subcultural milieu, than byefforts to achieve culturally distant and ill-understood values.

2 HoLLINGSHEAD, ELMTOWN'S YoUTH (1949); CRAw-FORD, MALAMUD, ANi) DutmsoN, WORKING WITHTEEN-AGE GANGS (1950); WHYTE, STREET CORNERSocIETY (1955).

[Vol. 56