THE QUARTERLY JOURNAL OP EXPERIMi!NTAL PSYCflOLOOY. 1991. 4JA 1l)70I- 7J1

The Interlingual Identification of Spanish and English Vowels:

Orthographic Evidence

James Emil Flegc Uniller3ity of Alabama at BirmingluJm , Alabama, U.S.A.

When IOmeone who is learning a ICOOnd ian,u.sc (ll) produces. lOund in the L2 using. r.mili .... D.tive-Ianguage: (LI) calc,Ory. the L2 lOund i. uid 10 havc been "identified with" an Llsound. Although interlin,uaJ identification cxerts a powerful inOuence on II pronunciation. it is still poorly undentood. Orthographic classification wu used here to UleSS the interlinaual identifica­tion or Spanish and English vowels. Sixty native speaken or Spanish in three experiments judged the vowels fif. /I/. /r. /. and leI in multipk: tokens of Enalish words ("beal", "bil", "bet". "bal") spohn by teo native speaken or American Enalish. The subjects labelled each Enalish vowel by cirelina one or the five lellen uJed to spell the vowel phonemes of Spanish (viz. < i>. < e > . < a >. <0>. < u » or by cireli na"none" irthey thought they had beard a vowel not found in Spanish. Subjects wbo spoke Enalish as.n L2 IHCd the "none" label more often than did Spanish monolinguals •• ugesling Ib.1 L2 learning heightens bilinauals' awareness of cross-hinauagc: phonetic differences. Exper­ienced Spanish speakers or English did use the "none" label more orten than did inexperienced .ubjects (42-;. VI. 18·/.). A few .ubjects used the "noDe" label consistently for lei and /1/. suggestina that they may have (Caarded these vowels IS "new" (i.e., non-Spanish). However, the aroup dal. provided lillie .upport for the hypothesis that the adult Spanish learnen of English treated either /e l or /1/ as new. The areat majority of .ubjects, even those highly experienced io English, identi6cd Enali.h Ie/ with thei .. Spanisb I. '.

It is widely believed that adults will inevitably speak. a language learned after childhood with a foreign accent (Lenneberg 1967). Scovel (1969, 1988) claimed that foreign accents occur because or lost "plasticity" in "neuropsy-

fUquesu for reprints should be sent to J.mes Emil Fkae. Department of BiocoDUnUoica· lion, Univ-emty of Alabama aI Binninpam. Bo .. 50) University Station, 8.inninJham, AI.ba~ )5294-00111 USA.

Thi, JlOOyw •• lupportcd by NIH 11"01 DC00251.1be . utbor IMob 0 ,-$. Bobn. L.Cueva, C. Menl. and 1.. Skelton ror their conlributiolU 10 the IIOOy. Than .. Ire .110 ex~dcd 10 S. Fletcher .nd M. Munro ror comments on In earlier venion or lhis Irticle.

C 1991 The Experimentsl PJ:ychololY Society

702 FlEGE

chological mechanisms" deriving from the completion of cerebrallateraliza­tion at around the age of 12 (see also Patkowski, 1990; but cf. Flege, 1987&, Flege & Fletcher, submitted). Precisely which &spect(s) of speech learning changes with age has not been identified by advocates of the critical period hypothesis, but it is onen implied that processes that underlie sensorimotor learning may become less effective as the result of neural maturation (Sapon 1952; Scovel, 1988, p.62). Some L2 production errors do, in fact, seem to have a motoric basis (Flege, 1988c; Flege, McCutcheon, &:. Smith, 1987),1 but many others may arise from the inaccurate perception of L2 sounds.! The results of certain speech perception experiments suggest that adult learners of an L2 do not have representations for some sounds in an L2, at least not at first (e.g. Miyawaki et al ., 1975. Mochizuki, 1981; Sheldon &:. Strange 1982). For other L2 sounds, adult learners have been observed to perform consis­tently, albeit not precisely like native speakers of Engli sh. In these instances, the adultleamers appear to make use of native language (LI) representations that differ from those of native speakers (e.g. Flege & Hillenbrand, 1985, 1987; Flege &:. Ecfung, 1986; Flege, 1989b; Flege & Wang, 1990).

It is generally agreed that speech perception is an obligatory process. l2 learners attempt to make perceptual sense of the sounds making up L2 words, regardless of whether or not they have an accurate (i.e. native-like) perceptual representation for the L2 sounds. Unless they have managed to establish phonetic categories for L2 sounds (see Flege, 1990; in press, at b), adult learners will tend to iden tify unfamiliar L2 sounds in terms of an LI category (Weinreich, 1953; Wade, 1977. 1978; see also Butcher, 1976). This process, called "in terlingual identification", often amounts to fitting square pegs into round holes. Adult learners' perceptual "errors" may not be evident, however. because they often comprehend the L2 in the absence of accurate phonetic perception by exploiting semantic context and their knowledge of the world.

The occurrence of perceptual errors or mismatches in the processing of L2 words may be inferred. howev~r, from how adult learners pro"ou"c~ the l2 . When an unfamiliar L2 sound has been identified with an acoustically

, Fkae (1911k) aDd FIe. el al. (1987) eummed lhe: produc:don orthe word·8MI Ea&lw.. ,p/­fbI contrast by nati~ speaken or Chi-.. A ~anlfieant dilrerence in peak onll air pressure: between EoaJilh /pl and fbI IUJ,&elted that the Chinese .ubject. produced tbete IOUnd.J with different larynaeal artic:ulationt. However, they failed to 'Ultlin closure voic:in. in fbI (or IS kma al native .peakers, apparently because they had not learned to cnlarlE the .upnalotlll cavity actively durinllabU.1 clQ5ure ror fbI.

'The tcnn "sound", as u.ed here. refers 10 a cia .. of phones Ihal can be uied 10 contrast muninl. II il a terminological COnVenlc:1ICe Ihlt obvialet tM need 10 draw I diltinc:l}oo constantly beIWe-eO Ihe phonetic and phonemic levell. For In expotidon or Ih ill theoretically

importaoJ d,lIincli6n. W F1l~ (in prell. I).

INTERLINGUAllDENTIFICATION 703

different sound in the L1. the LI sound may be used in place orthe L2 sound (Valdman, 1976). Table I summarizes six studies that have reported on vowel substitution errors in Spanish-accented English. The target vowels in this table are those examined in the present study (viz. Iif. /II. f&l. and lref). The six studies differed in many ways, including the L2 experience or the Spanish subjecu examined, speech materials. and elicitation techniques. II is unclear in many instances whether the "substitutes" reported were English vowels, Spanish vowels, or a phonetic approximation or the intended English target vowel. Several conclusions can nevertheless be drawn rrom Table I: (I) Spanish speakers usually produce English !if as an [i1-quality vowel; (2) ir they mispronounce /II, it is usually as an [iJ-quality vowel; and (3) ir they mispronounce /&1. it is as an [e]-quality vowel.

One might also conclude rrom the data in Table I that Spanish learners or English arc more variable in their mispronunciation or !re/ than that or other rront. monophthongal English vowels. In producing /r./, they manirest both mid-vowel ([e], fe']. [&)) and back-vowel ([aJ) substitutes. The usc or multiple LI substitutes ror an unramiliar L2 sound, especially in ea rly stages or learning (Hammarberg, in press). may mean Ihat an unramiliar L2 sound has been judged to be perceptually equidislant rrom two or more LI sounds. It might also be taken to mean that a particular L2 vowel docs not have a readily identifiable counterpart in the LI.

None or the five Spanish vowels Ui, e. a, 0, ul) seems to match any or the 15 vowels or English. F1ege (1989a) round that the tongue is lower ror English IiI and lu/ than ror the corresponding vowels or Spanish. and the tongue is lower ror English IO/ lhan ror Spanish la/ (suggesting a narrower pharyngeal cavity ror the English vowel). English !oUI and especially I~I are more diphthongized than Spanish leI and 101 (Stockwell &. Bowen, 1965;

TABLE 1 Vowel Sut.tilUtion ErrOfl In S~nllh·Acc.nted

Englilh

TQ,.,~I

~Iisll VQ~I Mfnl Fr~Ilr'"

SuJmitul~ R~JHNI~d

M

'" /. / I- I

(ii' (iJ'.J.I~""

['F [aJ' ...... (8r

' Brennan &. Brennan (1981); • Hammond (1986): I FlcJC (1990) • Amaltae (1918); • Ornstein (1914); • r.hcDonaki (1989).

Ol~r &porf~d

SubJlilu/t1

Icl'

(81' [eJ' [e'r

704 FlEGE

Oalbor. 1980; Quiles, 1981; Flege, 1989a). Also, English vowels are generally longer than Spanish vowels because they are lengthened more under stress and in pre-pausal position (Delallre, 1966).

Wode (1977, 1978) suggested that there is a threshold beyond which an L2 sound will be recognized as distinct from any sound in the LI. A distinction between "new" L2 sounds and those that are merely "sinlilar" 10 sounds in the LI has been drawn by researchers (or several decades (De/allre. 1964, 1969; Briere, 1968), but no satisfactory metric now exists for determining whether an L2 vowel will be treated as new or similar (see below).l Unlike consonantal sounds. vowel sounds are discriminated readily even if they are not classified differently.· One obstacle to cla55ifying L2 vowels as new \'S.

similar has been uncertainty as to whether the distance between vowels in two languages is judged in terms of an auditory,' phonetic, or phonemic melric (see below). The qUC5tion raised in this study was whether Spanish speakers would come to recognize that any of the front , monophthongal vowels of American English are new (i.e. non-Spanish).

Orthographic Classification

The present study examined patterns of interlingual identification that might be responsible for vowel production errors like those in Table 1. Native Spanish subjects were asked to label English vowels with one of the letters used to write the five vowel phonemes of Spanish «i>, <e>. <0>, <a>, < u». The vowels they were asked 10 label (viz. /i/. /I/. /E/./Zl) occurred in evc words spoken by native speakers of American English.

I F1cje (in preu, a) ditcus.scd method.tl Ihat milht be uted to operallonalh:e the: dillinction belwc:c:n DCW and similar L2 soundJ. A similar U sound may be defined a. a aound that ill rcprelCntc:d by the: same: IPA symbola. a sound In the LI, provided it C8n be .hown to differ auditorily from the correspondinl LI sound. New JOund •• on the: other hand, may be defined as L2 sounds that a re rcprexnted by an IP" symbol not uKd for Iny aound in the: LI (and, of courtc:, which differ auditorily from the: nearest LI sound). Problems with this approach do nillt. hmona-. especially with the "symbol" lett (tee, e .l .. Footnote: .) .

• Phoneticians have Ionldcbatcd whether tranxriptKIn. ofvo_1t Ihould be detailed eoouah to represent any mxJjttNiJy Ikt«tDhh difference between vowels or just differenc:a that under-lie word contrasts In I particular taniUaae. It i. uncertain hOW' many different distinct vo_l "typtt" there are. Ladefop (\9119) estimated that an upc:rt phonetician mi&ht be capable of identiryinl u many u 200 vowels in absolute tr:md.

'The ute of an auditory-based meuic to tvalu.te the reiatiolUhip bet1ll'Otn vowtls in Iwo Ianlu.JCI cannot beditcOunled. Spanish and Enalilh listeners in the: F1ece etal. (in preparation) lIudy pve much the: AInt rcsponJc:t to pairs of vo_I, that were libelled differentially. For uamplc:, both Iroupl rated Spilni.h /i/-/al pili ... to be morc: dissimilar than 1i/-/c:J pain; and both aroupl rated the li/-/e/ PIli ... as more dissimIlar than /e/-/a/ pIIirs Such _ir«ment betwc:c:n aroupl implicsthe: uiltence of a univerulauditory-baaed dl.llnce metric. Interliniual compan­IOn. baled on luditory representation. would be elJlC'Cted to differ at timc:l from thOle: balled on

ttpreuntltionl of phonetic quality (see, C.I ., Klal!. 1987. Fi&UfC 31).

INTERLINGUAL IDENTIFICATION 705

Orthographic classification was a plausible technique for investigating inler­lingual identification because Spanish vowels (unlike those in English) stand in a one-to-one relationship with letters. It has been used successfully by Wiik (1965, cited by Flege, 1988b) to examine the classification of English vowels by native speakers of another language with "phonemic" spelling (i.e. Finnish).

Subjects in the present study were also pennitted 10 use the label "none". This label provided a means for detennining whether any of the English vowels were judged to fall outside the bounds of the Spanish vowel system. Results obtained by Butcher (1976) suggested that adult subjects can use "none" appropriately. In Ihat study, French and English subjects were asked to label a set of cardinal vowels that differed to varying degrees from Ll vowels. The English but not the French subjects used "none" frequently to label front rounded (fY/. 10/. ler./) and nasalized vowels (/!/. /d/, /3f). These vowels differed substantially from any English vowel but were similar 10

vowels in French.

Level of Analysis

Perception research suggests that speech sounds are processed at three successive levels: auditory, phonetic, and phonemic (see Elman, Diehl, &. Buchwald, 1977; Repp, 1981 ; Pisoni, Aslin, Perey, &. Hennessey, 1982; Werker & Logan. 1985; Werker, Klein. & Logan, 1986; Mann. 1986). It is uncertain at which leveJ(s) interlingual identification occurs. Bilingualsjudge interlingual distances by comparing incoming L2 phones to perceptual prototypes in long-tenn memory (Best, McRoberts, & Sithole, 1988; Flege. 1984, 1988a. 1988b).' The orthographic classification technique employed in the present study probably favoured the use of pho nemic representations (Mann, 1986), but it was also possible that th~ subjects would be influenced by representations at other processing levels.

Catford (1968) slated thai the basis for interlingual identification is "substantial rather than fonnal" (p. 164). This implies that the L1 prototype against which L2 phones are gauged may be phonetic in nature (Briere, 1966), not phonemic, as usually supposed (e.g. Trubetzkoy 1939; Scholes, 1968a, 1968b; Best et al., 1988). The distinction between phonetic and phonemic levels can be illustrated with reference to Spanish leI, a vowel phoneme with two allophones.7 Spanish leI is realized as an (e)-quality vowel

• Bilinluals miplIQeJI11K: diSlancc belween Lt and L2 sound, by implicitly complrin,liM: 6UIIH"U needed 10 produce Ihem (tee Scholet, l%Bbl, bul lhere it I. yel no evideDIZ ror lhis

'lbe lilophona or Spflnish leI .re In tompk:menuu)' dislributiOfl - I""1 iI., lhey oc:c:ur in mUlually ClIclusi~ phOnel1t tonlcalS. 8c:CIIUIC or Ihis, lei Ind III cannol be used 10 tonlratl meanin! in Spanish.

706 FLEGE

in most closed syllables. These variants are said 10 "match" the vowel in Engli sh "bet" (Dalbor, 1980), a claim supported by a recent L2 vowel intelligibility study (Flege, 1990).1 As suggested by the symbol used to represent it, Spanish leI may be implemented as [e) in other contexts (most open syllables except those preceding Irl, and in closed syllables ending in lsI or IzJ). American English I~/ , on the other hand, is implemented with a single allophone (viz. [e')).'

Spanish leI shows some formant movement, but not nearly as much as the diphthongal English le'I (Flege, Munro, & Fox, in preparation). In a study using g1ossometry, Flege (1989a) found that the average tongue position used for the (e) allophone of Spanish leI was intermediate between the tongue positions observed in the "onglide" and "offglide" of English Ie'/. It is uncertain whether bilinguals will identify a diphthongal L2 vowel with a monophthongal L1 vowel. It is also uncertain how they will deal with differing dittributions of allophones for phonemes in their two languages. If interlingual identification occurs at a phonemic level, then Spanish speakers of English should consistently identify the realizations of English 1&1 in terms of the Spanish leI phoneme. The error data in Table I indirectly supports this view.

There is reason to suppose, however, that interlingual identification does not occur exclusivtly at an abstract, phonemic level. If Spanish speakers compare English vowels to the Spanish leI phoneme, then their judgements should be influenced by the properties of both the [e) and [e] allophones of that phoneme. A recent study by Flege et al. (in preparation) suggests that this does not happen, at least not in certain tasks. Spanish and English subjects used a 9-point scale to rate pairs of CV syllables for degree of dissimilarity. Some pairs contained realizations of English I~I and leI. Had the Spanish subjects based their judgements on a phonemic representation, they should have judged the [~HeJ pairs 10 be more dissimilar than the English subjects because, for them, [e] phones would be associated via phonemic representations with [e] phones. However, the Spanish subjects judged the [ItHe) pairs to be less dissimilar than did the native English subjects (3.1 vs. 4.2, p - O.OOJ).ID

'In thlt . tudy, Spanilb spc:aken produced Enali'h "beat", "bit", "bet", and "bat". lbdr (t:1 auemptt were identified colTtlClJy by nali~ Enalish Jillmnw in mOlt inltancea, but not their attempts It Enalilb {il,/t!, and lrel. II remainalo be ICen whelher SpaniJh learnen ofEnSIiJh will be .utUllfuJ in producin, /' 1 in phonetic oonlexll in whkh lhe reJ allophone of Spanish leI would be expected, for eXimple in words like "Bess", "meg", "AYS".

'The qualilY of Ibis vowel varies, of oounc- ror eXimple, the otr&lKk of le'I may nol be reached &I a fUI .peakinJ nle.

" ThiJ .. me kind of reuonin, can be applied 10 CV ptlin conlainin. realiu tiolUl of En.lish I&/.nd Spanilh leI. If . ubjecll used phonemic· leve! repr_nlalions exclusively for loch paitt. one would expecl/ell peroeived dissimilarily for Sp!lnilh Iban English lubject. bea.\ae Ihe [II)

INTERlINGUAllDENTIFICATION 707

Changes in Interlingual Identification

In addition to unrcsolved qucstions pertaining to the processing Icvcl(s) at which interlingual identification occurs, it is uncertain whether patterns of interlingual identification will change as L2 learners grow familiar with the L2 sound system. A change in interlingual identification may be more likcly for certain L2 sounds than others.

Somc L2 sounds arc virtually identical to those in the Ll (Bohn & Flege, submitted), but most differ from any Ll sound, if only in the details of timing, amplitude, or placement of articulatory constrictions. It was once believed that cross-language differences were the primary source of speech lcarning difficulty (e.g. Stockwell & Bowen, 1965; Briere, 1968). This leads one to expect that L2 sounds that match an II sound closely will be produced authentically whereas those that do not match an LI sound closely will be produced poorly. It may be, howcver, that a "U"-shaped rather than a Iincar function best describes Ihe effect of varying diffcrences between LI and L2 sounds. That is L2 sounds that match an Ll sound closely, or else diffcr cotl3ilJerably from any sound in thc Ll may be produced authcntically, whereas L2 sounds that partially rcsemblc an LI sound may be pronounced poorly.

I have hypothesized that " ncw" l2 sounds will stop being identified with LI sounds at somc point in L2 learning, whcreas "similar" L2 sounds will continue being idcntified with an LI sound (Flege, 1988b; 1990; in press, a, b; see also Ollcr & Ziahosseiny, 1970; Wade, 1977, 1978). As a result, adult learners may be ablc to eSlablish phonctic catcgorics for ncw L2 sounds and cvcntually produce thcm authentically. Similar L2 sounds may prove 10 be the most difficult kind of sound for adult learners because, by hypothesis, they continue to be identified with Ll sounds (which blocks the phonetic category formation needed for authentic production). In support of this, several studies have shown that vowels not found in thc LI invcntory are produced more authcntically than are L2 vowels that differ acoustically from a counterpart in the LI (Mucller & Niedzielski 1963; Flcgc, 1987b; Major,

allophone o( Spanilh leI i. uid to match En,li,h /el realizations. IJutead. the Spaniu. li,tene ... rlted (e)-lel pai ... • 1 more diHimilar th.n Enalish lislenel'l ( • . 1 v •. ] .9. p * O.OII). One olher recentlludy point. to the opera lion of phonetk:-Icvel procuses. In I II00Y by flege (ubmitled). Spanish and Ena1i,h lubjecll atimaled dellree o( rhyme in Enatilh word •. Both aroup' nlled pain o( Enalilh word. with /t!/ aDd /el (e., . "mate"- "mel") 10 rhyme equally, bUI lhey pve diifen:ntrltil'p (or other word (>IIi .... II w .. wumed IlIat the Spanilh lubjecu would idenlif'y English J~/ with their Spanish /ct. and thai vowels c1.uilied u beiDJIthe same "will rhyme more th.n those clauilied u diKerent". Ilad the rhymin,judaemenlJ been made at a phOMmlC 1e~1, therefon: , the Spanish lubjccu .houk! have pellZived • ,Rller dearec or rhyme lhan did the

En&li~h lub~U,

708 FLEGE

1987; Bohn & Flege, 1990; but cf. Flege, 1990}. There is also some preliminary evidence that L2 consonants thai are unlike any in the LI may be produced more authentically than L2 consonants that differ acoustically from an LI counterpart. II

Figure 1 shows the distribution of formant values for six English vowels (/i,I,E, t.e,O,Uf) spoken by 30 male native speakers of American English (Peterson & Barney, 1952) and the five Spanish vowels spoken by 16 native Spanish males (Godinez, 1978). In agreement with articulatory data (Flege, 1989a), English Ii! and luI are lower in the F,- F) space than Spanish iii and /u/, and English 10/ is posterior to Spanish la/. The data in Figure 1 largely agree with a study of Argentine Spanish (Guirao & de Manrique, 1972, 1975), which showed that the distribution for Spanish iii fell within the ellipse for American English Ii/ (Peterson & Barney, 1952). Data reported by Skelton (1969; sec also Quiles, 1981 ; Navarro Tomas, 1950) suggest that this would not be true if the relatively open allophone of /if in closed syllables had been included. The data in Figure I agree with the Argentine Spanish data in showing that realizations of leI fell entirely within the space occupied by American English It/ (see also Pa~un, 1976; Ladefoged, 1975). The Argen­tine data for leI and that shown in Figure I may be misleading, however, because only the (e) allophone of /el was examined. Had [E) allophone tokens been included, the Spanish leI ellipse in Figure I would probably have encompassed much of the F,- F l space occupied by English lEI (see Skehon, 1969).

If interlingual identification were based on proximity in a psychological space resembling the F,- FJ space in Figure I. we might expect the following:

1. English Ii/ will be identified with Spanish li/; 2. English /II will be identified with Spanish le/, IiI; 3. English let will be identified with Spanish leI; 4. English /IEI will not be identified consistently with a Spanish vowel.

" Best et.1. (1988) provided pcrceptu.l evidence that Bantu clkks may be sufficiently distinct from .ny En&lish ronson.nt that they are not identified in lenns or.n English category. No L2 production "udy, 10 my Imowled,e, hal uamined dick production. Such • study would be

proble:m"ic:al, however, becalUCclickiare likely 10 occur Kldom in human lanpuget because Ihey ue c:omplu .rtkul.torily. (Thi. would work againn the hypothesis th.! new L2 soundli will be IlUIJlered.) Seveml studies examined the production or Enlli.h /r/ . ad 11/ by J.panese speaken. These sounds might be n:prdcd u new, for tlic:re is 00 /II in J.panex .nd thc: J'pan_ /f/ is n:.Jized u a voiced tip-alveot.r n.p (_ F1e,e, 198&b, for a review). Speech production and pcroeption rex.o::h hu sugested that at aJl lOme native 'l'C'kerJ or Japanese can mlJter the !r/-fl i oonlra" ofEnglilh (M.cKain CI al., 1981 ; Mochizuki, 198t ; Sheldon &: Stnmgc, 1912; IlOl:

.110 flc:JC. 1988b). RQearc:h examinin, simil.r L2 stopt, on the other hand, indic8.le Ihat lhey will.eldom if ever be mutered by .dult arnen (F1ete, 1991).

1500

.. • \ • . , . . .... '

S/-/ ', :' i-, 'or : ... :' .. _.' . • \ . . , , , .. ,

"01

INTERLINGUAl.lDENTIFICATlON 709

1000 '" '"

'" .00

, '00 ,

" 00 • • , , .. •

S/ a/ .00 , , '.

.00

FIG. I. Fonnlnt frequency vllues for $II. Ena.li.h vowell (/i /. /tI.lcl./rt/.lo/./uf) lpoken by JO mile native apelken of American and En,li.h (Peteflon & Blmey. 1952) Ind five Spanish vowel, (IiI. le/./I/./o/. luf) spokcn by 16 male nltive 'peskcn ofSpilni.h (Godincl., 1918), in mell . The elliptel define 95~. confidence limits; they IrC baxd on Ih.c two principle componcnlt of vlriltion in F, Ind F, values.

Prediclion (I) agrees with mults obtained in a cross-language perceptual study by Scholes (1967). summarized in Table 2. In that study, English and Spanish subjects used keywords from their L1, or the label "none", to identify a matrix of 69 isolated synthetic vowels in which F. and F) frequencies were varied. The vowels typically heard as IiI by native English subjects were also heard as Ii! by the Spanish subjects. In agreement with the error analyses in Table I, but in apparent disagreement with Prediction (2) above, [I)-quality vowels were always heard as IiI by Scholes' Spanish subjccts, never as lel- Il Prediction (3), as stated , would be true only if interlingual judgements are made exclusively at a phonemic level. The finding that [£J-quality vowels were sometimes identified as " none" by

"One re .. o n for caution in reprds 10 Scholes' (1961) '" dltl il that just one of Inc 69 .ynlhcl;c vowcl stimuli was heard consiltently by En,lilh . ubjecu IS /tI, problbly ba:ausc Inc .Iimuli were 400mJeC lana.

710

TABLE 2 Tabulation of PerceplUll Olta Aeported by

Scholn (1967).

Labdlinl by nalive Spini.h listenerl or vowell typically claSIified .. fi l. le/. Jat./o"',. or /UI by nalive ipelker. or

EnaJish

SwNIs TyplaJlly Huud by Spmdslr Swbju/J CkuJ/fid by Eitf/Wt Swbjrcu s-d P~UflI

III III 100 N M 100 /<'/ ~, 100

'" 1<, 72 "OM " ,-, "OM .. ,., " ,., /. /-100% 100

lolil /01-100% 100 lui lui ..

"OM .. u.bc:Uin, by native EnaJish listeners or VQwel. Iypkatly

clauilied as IiI. /e/. /_/. /0/. or luI by nuive SP'lnish .pc.ken

SOlUltU Typkt3l1y C/IW/fitd by SpanisJr Subjwl

~,

'" 10/

.so-d

M N

'" ,-, '" ,-, "OM

'01 N /01 "OM

/01 lui ,U,

!ltQrd by &.11,", SwbjHu

Perunl

'" 20 10

" 29

" 12 .. 12 ,. J]

11 ,. lO

INTfRLlNGUAL IDENTifiCATION 711

Scholes' Spanish subjects suggcsted Ihat Iheir judgements were based, at least in part. on phonetic-level representations.IJ

Perhaps the most interesting observation to be drawn from Scholes' (1967) data is that the Spanish subjccts seemed to usc the "none" response predominantly to label (Ie]-quality vowels. This agrees with the acoustic data in Figure 1 and is consistent with the observation that Spanish speakers of English mispronounce English Iz/ in a variety of ways (Table I). This may mean that /z / emerges as a new vowel for native Spanish learners of English. Such a conclusion is be no means certain, however, because Scholes made usc of isolated synthetic vowels. The subjects might have used the "none" response because some incidental property of the stimuli (e.g. duration) diverged from values specified in their central representation of the closest Spanish vowel.

Decisions as to whether a vowel is "different" from any L1 vowel may be influenced by responsc biascs and perhaps subtle alpec:l5 of experimental design, such as the wording of instructions. The criterion used here to determine whether an English vowel emerges as "new" was the pattern of behaviour observed for groups of native Spanish subjects differing in English-language experience. Experiment I examined Spanish subjects who had little or no prior exposure to English (Le. Spanish monolinguals). Experiments 2 and 3 examined Spanish speak:ers of English who differed in L2 experience. Experiment 3 differed from the first two experiments in using instructions that encouraged subjects to use the " none" label. If English 11£/ emerges as a new vowel, we would expect Spanish speakers of English to use the "none" label for /1£/ more often than Spanish monolinguals but not to differ from the monolinguals in the frequency with which "none" is applied to other English vowels. If L2 learners recognize only gradually that certain L2 vowels are new, we might see no difference between inexperienced Spanish speakers of English and Spanish monolinguals, but observe the kind of difference just described between relatively experienced and inexperienced Spanish speakers of English .

EXPERIMENT 1

Method

Ten native speakers of English (5 males, 5 females) living in Birmingham, Alabama, read a randomized list containing multiple repetitions of English "beat", " bit", "bet", and "bat". The first three repetitions of each word were

"Thi, i, e.peci.lJ)' true: alven thai the Spa.nllb .ub)OC11 were offered 110'0 Spa.nidl keyword, wilh which 10 I.beltbe I),nlhelic vowell - one lluil contained the '11.lIophone (vR. ,,,,.) .nd one with the lei .lIophone (viz. tk).

712 FI..EGE

lowpass-liltered at 4.8 kHz and digitized at 10 kHz with 12-bit amplitude resolution. The formant frequencies of all 120 words (10 talkers)( 4 vowels)( 3 repetitions) were estimated using LPC analysis (14 coefficients for males, 12 for females) by placing a 25.6-msec Hamming window at the acoustic midpoint. Fundamental frequency (FO) measurements were based on the duration of three suocessive glottal periods al the vowel midpoint. "Vowel" duration was measured (in msec) from the onset to the offsct of periodicity in each waveform.

Table 3 presents the mean acoustic values for me five female talkers who produced the English IbVtl words used as stimuli in Experiments 1- 3. Table 4 presents the mean values for the five male talkers. As expected, Iii was longer than /II, and Izl was longer than lEI. Also as expected, FO was slightly higher for the two high vowels (IiI, I II) than for the two non-higb vowels (lEI. Izj). For the most part, the formant frequency values reported in Tables 3 and 4 correspond to values published for American English by Peterson Bnd Barney (1952).

In figure 2 the l!i tokens of each vowel obtained from the male talkers are plotted in an F.- F] space along with ellipses defining native Spanish males' productions of the Spanish vowels l i/ . leI. la/. 10/. and luI (Godinez. 1978). All but one English Iii token fell within the ellipse for Spanish Iii. Most of me English /II tokens fell within the Spanish leI ellipse. One English /e/ loken fell within the leI ellipse, but most fell outside any Spanish ellipse (usually closer to Spanish leI than la/). None of the English Iz / tokens fell within a Spanish vowel ellipse, but most were slightly closer to Spanish la/ than le/.

Tbe 120 English IbVtl words were normalized for peak intensity. then recorded on audio tape (Marantz PMD 420) for later playback to subjects. Two randomizations of the 12 words spoken by each talker (4 words x 3 repetitions) were recorded . The five male talkers were recorded on Side A of the tape. the five females on Side B. The interval between each word was fixed at 2.5 sec. The interval between talkers was 10 sec. Sides A and B were presented binaurally over headphones in a sound-booth to 20 monolingual speakers of Spanish. The paid subjects (11 males, 9 females) had a mean age of27 years (S.D.-9). Each reported having normal hearing. Most subjects had been in the United States for only several weeks at the time of testing. Fourteen were from Mexico. five were from EI Salvador, and onc was rrom Argentina. It is not known what influence. if any. the variations in native dialect of Spanish may have had on the results.

A native Spanish research assislant read instructions in Spanish 10 the SUbjects. They were lold to identiry the vowel in each word by circling < i >. < e >, < a >. < 0 >, or < u > on a specially prepared answer sheet, or by circling " ninguno" (none) if they heard a vowel "nol found in Spanish". The subjects were told to guess jf they were uncertain . The words spoken by one talker (24 :!limuli) WC:[G prC5c:nlc:d for practice: before: the: e:Jlpc:riment begltn ,

INTERLlNGUALIDENTIFICATION 113

TABLE 3 Duration' and FrequenCV- in tha Vowel. Found in

" Beet", "Bit", "Bet", and " Bat" .. Spoken by Five Adult Famala Nail..,. Spnke •• 01 American

Engll. h

Yowtl To/ktr Ow. '" FI F1 F3

fiI , '" ' 43 329 26" 322. , '" '79 H. )107 3620 , 19' 213 H2 '.82 368'

• '36 '93 37. 2728 ,m , '46 19' , .. ,>" "'58 M 171 , .. J60 "'" '41'

N .. '44 '" 2096 "" , '22 '87 '" "'" "" , 14' 204 ... 223' .,., • "' 20' '" 21S2 "''' , 21. 19. '16 208. ,m

M '" , .. .,. 2182 , ... I-I '45 137 "" 204] 2856 , '" '" 731 lOll ]1)10

] 206 '86 ... 1928 " .. • '47 '84 ... 2100 "" , '" ' 94 '" 208' 276)

M II. '" "" 2037 292 1

I- I , '" '" 65. "'" "" , '16 15' m '900 '86] ] '" '92 '" 182) "'41 • 209 '61 '" 1915 278' , '85 '68 ,., 2041 2611

M '" 16] '66 , .. , '84' • Duration in 01IeC.

• Frequency in Hz. Nor,.. £a,ch lalkh" mean .1 bued on three tokenl .

Half of the subjects heard a male talker on Side A for practise, the other half a female talker on Side B. This may have b«:n insufticient in a few instances, A total 5.5';' of the data, all from three subjects, had to be discarded because ofmiHing responses. I'

.. Each liltener muked 24 answer. in telponIC to the vowels or the 10 talken., a. required, but the three liJlenrr.jusl mentioned probllbty did not rCfpond 10 every word a.r If w/Uprtunltd

(continued overlear)

714 FLEGE

TABLE 4 Duration' and Frequency" In the Vowel. Found In

"Baat'", "Blf', " BeC, and "e.f' .. Spoken by Five Adull Male Native Speakers of ArTl8fican

English

$10_1 TQJk~, Dw. f1} FI F1 FJ

II 1 III 111 26' 2n) )111 2 164 II> 322 "" 2'" , 202 104 312 2044 " .. • '" ... '" 2512 2876 , 234 '" 279 2260 3007

AI '" 109 '" 2]65 2962

'" 1 108 102 .,. "" 2139 2 120 111 '" "" "46 , 1" 117 461 1S55 2451 , 146 .. ." "68 "" ,

'" I" '" Illl] "48 M 1>1 107 ." 183.5 ""

'" "' " '21 1821 264' 2 161 "' '''' 11Il7 "09 , 21' 10' '" 1430 2288 , 160 80 ". "" 2694 , 247 "' m 11125 "79

AI 180 101 '" "34 2582

/. / '" " 711 1926 2604 2 '" " 766 168) 2615 ) 273 10) 69' 1>30 22" • 223 81 701 17)9 2'" , 300 "' 661 1 ... 2738

AI 231 " 707 "44 2'"

• Duration in msec . • Frequency in Hz. NOIr: Each talker mean is hued on three loiF:erlJ.

in the few bkJcb. II appears thaI these listener. missed a response but attempted to conceal this fael by markin, an . nswcr(l ) at the end of the answer sheet .Ocr the last word had been presented, raulting in rc:sPOIlKl that were out of sequence with tbe stimuli . All dala obtained in response to. JHorlN;:ul.ar Ullker were discarded if two or more obviously inappropriate responllCS were obtained for thallalter. "Inappropriatc" was defined here as an 0::: a > response for eithcr "bear' (/bill) or "bil" ({bill). IIICrolI hi&hly unlikely that such a response could be anythinJ other than a miSlike,

INTERLINGUAllDENTIFICATlON 715

F2 In • • h

r--'---r---r--,-_"C'T'"'~r--'---r---r-c" ','~'~r--' __ -r---r--l"r'=-,'" , I

... --- -- -... , --, , -\_-- - : :---"," "',

" ' , , , , , , \. ...... " .. _----

,

--. .... ni .... / 01

'"

'"

60' • • 1

-o /-- ... ,

/ ' / ' 100 S ,

~.?- t.,1 I

.. " I " • J , I / \ /

, -S .... nl.1I 1. 1 ... _ --

' " _ r III _ ~ I 'h l - 0 I ,e l _+

o Cn911.11 / 1/ t oll: ....

II 1"'i!1 1. 1I / 1/ t ol< .. n.

o 1t"'i!1I . 11 III t ot.n.

.00

1"911.11 I ",! t~~ - 900

FlO, 2. The F,- F, values (in mels) in vowel tokenl spoken by S male native lpea-ken: or American En,liah thai wen: IUICd al Itimuli in &perimcmtl 1- 1. The e.lIipICI define the. Spanilh vowell fil./e/./I/./o/. fuI) u spoken by 16 mile native lpeakers or Spanish (Godinez. (971).

The frequency with which <i>. < e > . < a > . < 0 > , < u > , and "none" were used by each subject to label the four English vowels was computed. For all but the three subjects just mentioned, the frequencies for each vowel were based on 60 responses (10 native English talkers )( 3 replicate tokens)( 2 randomizations). T·tcsts were performed to determine whether a particular response was used to label each vowel at a rate significantly exceeding zero. Since 24 t-lests were performed (4 vowels )( 6 possible response labels), an alpha level of 0.004 was used to give an experiment-wise error ratc of 0.10 (Snedecor & Cochran, 1980).

Results and Discussion

Figurc 3 shows the mean frequency with which < i> . < e > , < a > , < 0 >, < u > , or "nonc" was used to labellhc English vowels. An asterisk in this figure and those to follow indicates that a response label was used at a significantly above-chance rate (p < O. iO). The Spanish subjccts' most fre­quent response ror both Englifih fiI lind {If vowels was < i > (94~. , 68%),

716 FLEGE

CJ ; ","0 -. """ u =. c::sJ none 01 the obove

100

80 • • c 0

~ 60 ." :s c • '0 :E .. •

20

a

English Vowel

FIG. 3. The meIIn percentage of inillnea in which the monolio,ual Spanish ,ubjects in ~perimmt I UICd tbe < i>. <e>, <'>, <0>, < u > •• nd " oone" labels for the vowels in "beat", "bi!". "bet",.Dd "bat" ({I/./If./tI. IE/). Responses thai were Jiven significantly more oficn lhan UfO Ire m.rked by an ulcrbk; the error ban enclole :t I S.E.

suggesting that they identified both English vowels in terms of the Spanish IiI phoneme. The subjects also gave a few < e > and "nonc" responses for /1/ (190/.,12%), but not at abovc-<:hance ratcs. The most frequent response for English /£/ was < c > (81 %). The subjects also used < a> to label English /£I in a small (13%) but significant number of instances. The most frequent response for 11£/ was <a> (71 %), but <e> was also used at an above­chance rate (17';').

The results for Iii and lEI agree with the acoustic and perceptual results presented in the Introduction. The results for /II agree better with Scholes' (1967) perceptual data than an inference drawn from acoustic data, viz. that /1/ would be identified frequently with Spanish leI. Perhaps the relatively short duration of the English III. or its centring offglide (F1ege et a!.. in preparation), distinguished it from Spanish Ie/. The most surprising result was the one oblained for Iz/. Previo usly published acoustic data showed that this English vowel is about midway between Spanish lei and lal in an F1- F1

space (Figure I) - that is, Izl does not seem to occupy a portion of the acoustic vowel space exploited by a Spanish vowel . The particular 11£1 tokens used here (at least the males' ) fell outside any Spanish vowel ellipse. They were only slightly closer, perhaps, to the Spanish la/ than Ie/. This, taken IOgelhee wilh Ihe finding Ihal Ihe Spani,h ,ubj,,!! in Sohal,,' (1967) 'Iudy onen applied the "none" label to [1£]-quality vowels, suggests that Spanish

INTERUNGUAL IDENTIFICATION 717

monolinguals would not identify Enalish ltel in tenns of a Spanish vowel calegory.

The Spanish monolingual subjccll nevertheless identified lrel consiSiently whh the Spanish /a/. Thus the labelling data obtained in this experiment suggestlhat Spanish learners of English do flor lreal English /ICI as new. This conclusion is consistent with the results of a recent intelligibility study. Flege (1990) found that Spanish speaken' attempts at lrel in the word bat were identified by American English listeners as fI or 101 in over one· fourth of inslances. This relatively bigh frequency of misperccption. does not prove that Spanish leamen: fail to establish a category for ltel. but it is at least consistent with that interpretation.

EXPERIMENT 2

Experiment 2 further explored the extent to which Sp/lni.h speakers of English will note a difference between English Iii. /II, le/, Iz / and vowels found in their Ll. More specifically. the purpose of Experiment 2 was to determine whether native Spanish speaken: of English would use the "none" label.ignificantly more often Ihan the monolingual subjccts in Experimenl I, and significantly more often for lrel than for the other three English vowels examined.

Method

The stimuli from Experiment I were presented to 20 native Spanish speakers of English living in Birmingham, Alabama. The mean age of the subjccts (10 male., 10 female) was 29 years (S.D. - 6). They had arrived in the United Stales at an average age of 2S years (S.D . - 6) and estimated using English aboul 6S ~. of the time on a daily basis (S.D.- 19). The subjecu differed considerably in terms of l2 experience. Lenglh of residence in Ihe United Stales ranged from 4 months 10 10 yean (M - S years, S.D.- S). Ten " inexperienced" subjects had lived in the United States for only 0.8 years, on the average, and 10 "experienced" subjects had lived in the United Stales for 6.8 years. A lotal of 4% of the data was declared missing (see Footnote 14). The frequency with which the six response categories «i> , < e > , < a > , <0>, < u >, "none" ) were used to label the four English vowels was calculaled as before.

Results and Discussion

A series of Mann---Whitney lests were carried out on the frequency data to detenmne whelher the 10 experienced and the 10 inexperienced subjects responded differently. In many instances a response label was not used at all, or only mllIgiolllly, Tcna wcr, clfri,d out only if a label was used in at least S% of instances overall. This criterion was mel in 12 instances. A 0.004 alpha

718 FLEGE

level was used for the 12 tests to give an experiment-wise error rate of 0.05 (Snedecor & Cochran , 1980).

No difference between the experienced and inexperienced speakers of English was significant. Only two between-group differences approached significance. The experienced subjects gave more < e > responses to lEI than did the inexperienced subjects (S7~. vs. 31 %); conversely, they gave fewer < a > responses to /EI than did the inexperienced subjects (26% vs. 531/1) [U - 21 .5, p - O.028; U - 18.0, p - 0.015). These tendencies suggest that Span­ish subjects who are relatively inexperienced in English may have difficuhy in relating English lEI to the (E) allophone of Spanish lei if it occurs in an unfamiliar phonetic context (i.e. preceding a word-final stop).

The responses obtained for the experienced and inexperienced speakers of English have been combined in Figure 4 because none of the between-group differences was Slalistica.lly significant. English /iI was labelled as < i > in 840/, of instances. English /II received the following responses at above­chance rates: < e > (39~1), < i> (36 110), and "none" (21°/1). An inference drawn from acoustic analyses of Spanish and English vowels in the Introduc­tion was that English /II might be identified more closely with Spanish lei than with Spanish I i/. The response pattern obtained here in Experiment 2 is more consistent with this inference than is the response: pattern obtained in

100 •

80

60

. 0

20

0

= ; -e

"'" • fZ2l 0

""" " CSJ none

English Vowel

FIG. 4. The mel n percentasc or instaOCCI in which the Spanilh spuken of Enalish in Experiment 2 uJc:d < i >, < c >, < a > . <0>, < u >. and " none" al labell for the \lowell in "beal", " bil". "bel", and "bit" (/i/.N./I/./a/). Data for experienced and ineJlpcric:nccd native Spanisb speakers of EnJlish bave been pooled beau. IhcIc l ubl'"ouJ» did nOI d iffer Ji&nificaDUy. fn:qucnciCI that 5i&nificanlly ucccdcd mo Irc: marked by an &lteriski the error ban cndOlC :t I S. E.

INTEALINGUAL IDENTIFICATION 719

Experiment I. Thus, the Experiment 2 data for N suggest indirectly that Spanish speakers of EngJish may be beller able to appraise the phonetic characteristics of English vowels than subjects without such familiarity. Three responses were obtained at above-chancc ratcs for English lsI: < e> (44%), < 8> (39~.). and "none" (13~.) . One might expect English /e/ lo be identified with Spanish leI because this phoneme has an le] allophone. "None" responses might be expected because the lsI realizations in "bet" occurred in an unaccustomed phonetic contat (i .e .• preceding a word-final stop). The only explanation that comes to mind for why English lEI was sometimes labelled <a> is the acoustic similarity of some of the English /EI realizations to Spanish /a/ (sec Figure 2). Two responses at above-chancc rates were obtained for English Iz /: < a > (S2~.) and "none" (12~.). This finding does not support a conclusion that might be drawn from Scholcs' (1967) perceptual study, viz. thaI Spanish speakers judge EngliSh 12/ to be a new vowel.

"None" responses were obtained morcofien from bilinSuaJs in the present aperiment than from the monolingual subjects in Experiment I. The frequencies of none responses obtained in ExperimenLS I and 2 were: IiI (0% vs. 8%), /II (12% VS. 21~.). leI (I~. vs. 13%). Ie/ (10% VS. 12%). The differences between experiments were all significant at the 0.05 level (U - 107.0. 119.0, 65.0, 118.0). The same procedures were used for the bilinguals and monolinguals, 50 these l'Ciults suggest that the Spanish speakers of English had begun to note some of the acoustic differences between English and Spanish vowels. However, allhough the Spanishl English bilinguals used the "none" label more often than did the Spanish monolinguals. few bilinguals used it as their predominant response for the English vowels_ The rale of "nonc" never exceeded an average 20~. for any vowel in Experiment 2; and only 12 of the 20 individual subjects used " none" in more than 10% of instances in responding to the four vowels. If we consider all 40 subjects who parlicipated in Experiments I and 2, the rate of "nonc" responses exceeded 50% in only 10 of the 160 possible instances (40 subjects x 4 vowels). Six subjects used "none" as (heir predominant label for /II. and four used it predominantly for 1r.1 (sec Figure 7 below).

Of the four American English vowels examined. N and IE/ are probably the least like a Spanish vowel phoneme in acoustic phonetic terms. We reexamined the data obtained for those subjects from Experiments I and 2 who were not reluctant to use the "none" response. The "non-reluctant" subjects were defined as aU those who responded "none" in more than 10% of instances overall. The frequencies with which the "none" response was used in labelling N and IIItI by these 18 subjecLS were compared (0 the frequencies for IiI and lEI. The combined frequency of "none" response for the non-reluctant subjects was significantly grealer for /1/-/1£1 than li/-/el (28Y, vs. 13 ~.) [H= 12.1, df-3, p - 0.OO7}. Still, had the Spanish subjccts established phonetic categories for English /1/ and leel. we would have

720 flEGE

expected "none" responses to be used morc consistcntJy. Thus the results obtained in the first two experiments do not strongly support the hypothesis that any or the English vowels eumincd here will be treated as "new" by adult Spanish learners or English.

EXPERIMENT 3

The inter·subject variation in the use or " none" suggests that it may be difficult to detennine, in an absolute sense, whether bilinguaJs judge an English vowel to be phonetically or phonemically distinct rrom vowels in the LI inventory. The orthographic classification or L2 vowels in Experiments I and 2 resembles an experiment described recently by Johnson (1990), where subjects were asked to detennine whether words synthesized with different FO values had been spoken by the same talker or by different talkers. One listener identified only those word!: produced with an ithnlica/ FO values as having been produced by the same talker. Other listeners had a more IiberaJ standard, Some, in fact, seemed willing to allribute any detectable FO difference to a difference in speaker identity.

It is possible that some subjects in the Experiments I and 2 were more willing to we the "none" label than other subjects as the result of a general response bias. Ir so, it should be possible to manipulate the frequency of "none" responses. We attempted 10 do so in Experiment 3 by using instructions that were quite different rrom those used in the two earlier experiments. The new instructions encouraged subjects to usc the "none" label. Under the new instructions, " none" might be used often enough to warrant concluding that an English vowel emerges as " new" ror Spanish speakers of English.

Method

The stimuli used in Experiments I and 2 were again used here. The on ly imponant change in proceduTCS was in the instructions read in Spanish to the subjects. The subjects in Experiments I and 2 were told that the purpose or those experiments was to "compare" Spanish and English vowels. Here, they were told that the purpose of the experiment was to identiry differences between Spanish and English vowels. The ExperimeOl I and 2 subjects were told to circle one or five vowellet!ers, but to use "none" ir they did not hear a vowel from Spanish. Included in the instructions ror Experiment 3 was the observation that "many English vowels are unlike any vowel in Spanish". The subjecl5 were told to use "none" if they heard one or these "different" English vowels, but to use one of the five letters if they thought they had heard 8 Spanish vowel.

Twenty additional native Spanish speakers of English were drawn from the same population as in Experiment 2. The average age of the subjects flO

tNTERLINGUAL tDENTtFtCATtON 721

male, 10 females) was 28 years (S.D. =6). These subjects had arrived in the United States at an average age of 25 years (S.D." 7), had studied English for an average of 5 years (S.D. - 5), and estimated that they used English about 56'1. of the time on a daily basis (S.D.-J l). A subgroup of ten inexperienced subjects had lived in the United Stales for only 0.4 years on the average at the time of the study; 10 relatively experienced subjects had lived in the Uniled States for an average of 6.4 years.

Results and Discussion

A series of Mann-Whitney tests was carried out to test for between-group differences. Tests were not carried OUI when a label was used in less than 5%

of instances. The alpha level was set at 0.0045 for the II tests that were perfonned to ensure an experiment-wise error rate of 0.05. No between­group difference reached significance. The largest difference observed was for the " none" response as applied to /II. The experienced subjects used "none" more ofien than did the inexperienced subject! (51';' vs. 25'11) (U-2J.0. p-O.04J). A non-significant trend in the opposite direction was observed for "none" responses to /z/ (13 1

;. vs. 23%) [U= 49 .0, p = 0.971]. The data for the experienced and inexperienced subjects have been

combined in Figure 5 because the differences between the two subgroups. were non-significant. The "none" response was used more often here than in

100

80

60

.0

20

o

0;

• • =0

~ 1,1

=0 "'" u CSJ none

/II

I'

1·1 Icel FIG. S. TIM: mun pen:cnl.ge or inll.1IOe5 10 which tIM: Spanilh speakers o( Enatisb in Experiment l, whcreilUlnKtiolUcncourflJCd "none" ""pcm_. uoed < i >, <C>, < ' >. <0>, < u >, Ind ""onc" 10 labclthc yowcls in "beat", "bit", "bet", Ind "bill". D~II (o r experienced a"d inellperienced 5peakers or Enalilh have been pooled bccaUIC IhcIC lublVoup6 did nol differ li.nitieantly. An asterisk indicates lIbove<hllocc rrcqutnciGlj the error ban enclo" % I S.E.

722 FLEGE

Experiment 2 (23% vs. 14%), butlhe difference was smaller than expected, given the major changc in instructions (sec above). Mann- Whitncy tests wcre used to test for differences in the frequency of "nonc" responses in Experi­ments 2 and 3. The difference between Experiment 2 and Experiment 3 was non-significant for IiI (8% vs. 9~.) and for Iz l (12% vs. 18%) lV- 190.0, 180.0] . The difference for /II (21~, vs. 38%) was marginally significant, V-133.S, p-O.072; the difference for leI (13% VI. 27%) reached significance at the O.OS level, V-IlLS. p -0.0 16. Recall that the Spanish speakers of English in Experiment 2 used "none" more onen than did the Spanish monolinguals in Experiment I (14'1, vs. 6'/0 overall). It thus appean that manipulating instructions had about as much inHuencc: on subjects' likeli­hood to use "none" for an English vo ..... el as did learning to speak EngJish.

Thc subjects in Experiment 3 used <1> to labellhe English lit realiza­tions in most instances (88~.). For /II. three responses were used at above­chance rates: <I> (39'1.), Done (380/_), and < e> (23%). Taken togelher with tbe resUh5 of Experiment 2, this suggeslS that Scholes' (1967) 400-msec synthetic stimuli probably did not pennit an accurate assessment of how Spanish speaken identify English /II. Apparently, the acoustic overlap between English /II and Spanish leI realizations is sufficient to cause at least some Spanish/ English bilinguals to identify the two vowels with one another perceptually.

Three responses were obtained at above-chance rates for English /e/: < e > (48%), none (27%), and < a > (22~.). Taken together with the results of Experiment 2, this suggests that both phonetic and phonemic-level processing influenced the subjects' labelling. Had labelling been based entirely on phonemic-level decisions. one would have expected Ihe subjects 10 use the < e> response consistently. Apparently. they were influenced by the acoustic-phonetic similarity of certain English Ie) realizations to Spanish lal (see Figure 2). and perhaps also by the fact that the English (e] realizations occurred in an unfamiliar phonetic context. Two response!l were obtained at above-chancc: rates for Iz /: <a> (82'1.) and "none" (18~,). Taken together with the results of Experiment 2, this strongly suggests that, contrary to an inference drawn from Scholes' (1967) perceptual study. Spanish speakers of English do not regard llel as a new vowel.

EFFECTS OF ENGLISH-LANGUAGE EXPERIENCE

We saw little difference in the frequency of "none" responses between relatively experienced and inexperienced Spanish speakers of English in ' Experiments 2 and 3. The possibility existed that significant effects or L2 experience might be evident if the data from these two experiments were combined. That was the purpose of this section. Characteristics of the 20 experienced and 20 inexperienced subjects from Experiments 2 and 3 are

INTEALlNGUAllOENTIFICATION 723

TAeu 5 Chlracteristlcs 01 the NIIMI Splnllh Subfectl in

Experiments 1-3

Monolill,1Ill1 1/l1X~'lIncrd

Spanb' ill EII6/lsII

N, 20(11 M/9 F) 20 (10 MI lO F)

Ap 26.1 (U) 11.9 ( 6.1) 10.010. ' 11.1 ( 6.1) LOR' 0.6 ( Q.<t) YRS' 5.6 ( • . 1) PER' 59% (18)

'AJC or srrival in the Uniud Stales, ill yell ....

• i.en,lh or raidc:ncc in the United SIIIU, '" )'CIn. 'Yean of duJroom inslnIClion in EnJlll.h. in yun . • Sclr.atinated percentaae dilly ute or EnJlilh.

Not, : "M'· indka lCS male. "F" indkatcs rwuole.

£Jc~'lem:,d ill &,IWI

20 <10 MIlO F)

11.1 ( 6.S) 21.., ( 59) 6.6 < 2.5) U ( 4.6)

62% (24)

summarized in Table 5. The experienced subjects had arrived in the United Slates at an earlier average age than the inexperienced subjccts (22.5 vs. 28.1 years) and had lived there longer (6.6 vs. 0.6 years). The IwO groups were matched fairly closely, however, in terms of chronological age, amount of formal cducalion in English, and self-estimated daily usage of English.

One-half of Ihe experienced Spanish speakers of English received the "neutral" Experiment 2 instructions, whereas the other half received the instructions from Experiment 3, which encouraged the use of "none". The same proportion orinexperienced and experienced Spanish subjects received the two types of instruction, so it is possible to draw valid comparisons between the [wo groups. It would not be valid, on the other hand, to compare either group of bilinguals to the 20 Spanish monolinguals, whose characteris­tics are also summarized in Table 3. This is because all of the monolinguals received the "neutral" instructions. The monolinguals' results arc neverthe­less juxtaposed to those of the bilinguals in Figure 6 for the sake of completeness.

Inspection of Figure 6 reveals that the subjects in all Ihree groups used < i > predominantly to label English IiI, and < a > predominantly 10 label Iz /. There was nol a systematic increase in "none" responses for Iz /, suggesting that Iz l does not emerge as a new vowel. The greatest effects of L2 experience seemed to be for /1/ and lEI. Use of the <i> label for English /1/ decreased, and use of " none" responses for English /1/ increased, as function of L2 experience. This suggesls that /1/ may be treated as a new vowel by at least some experienced Spanish speakers of English.

Compared 10 the monolinguals, the Spanish speakers of English were more likely to judg(i the Bngli!ih yowel {f.{ in "bet" as non-Spanish. For some

9 •

9 •

VOWEL .. IV '00r-------~=-~~--~---

o Spano"" •• Ioo·K'''''~, _ Inup inE~ ~ hI*". in Engll""

N .. 2OIqroup 80

60 .. 20

0'-----'-

'00 o SpaNsI'! WOI'IOIjng _ Inexp '" EnqIi.1'! !5:S:9 E.ocper '" En9l1st. eo

N _ 2O/~roup

60 .. ~ *

20

o <,'

RESPONSE CATEGORY

<,'

<,>

VOWEl .. 101

o SpanosI'! Wonol;"9. _ lnexp on Enql'''' ISS!! E_ '" EniI~""

N "20/~

11 I ~ II VOWEL ... 1.,.1

Not a Span­

~-.. RESPONS£ CATEGORY

flO. 6. Comparison or the resul ts obWocd in Experiments 1-3.

INTERLINGUAL IDENTIFICATION 725

reason, the inexperienced subjocts who had just begun learning English seemed to identify English /E/ wilh Spanish /a/ al a relatively high rate. The experienced speakers of Spanish did so at a lower rate. thereby resembling the Spanish monolingual!.

Mann-Whitney tests were carried out to test for differences between the experienced and inexperienced Spanish subjects. The alpha level was sel al 0.0045 because a total of 11 lesls was perfonned. (Tests were nOI carried OUI in the 13 inslanccs where a label was used in fewer than 5~. of instances.) None of the between-group difference reached significance. It is worth notina. however, thai the difference in the frequency with which the experienced and inexperienced subjocts used "none" in labelling II/ (42~, vs. IS'!.) came very close to reaching significance (U - 99.S, p - O.OO56}, as did the between-group difference in <It> responses for leI (21~. v,. 40''''; U- 105.0, p - O.OO95].

GENERAL DISCUSSION

Spanish monolinguals and Spanish speakers of English who differed in English-language experience used one of the five vowel letters of "none" to classify the American English vowels in "beat", "bit", "bet", "bat" (i.e., fil, 11/, /1'./, and /Ee/) Their response patterns were. for the most part, predictable on the basis of the acoustic relationship between Spanish and English vowels as described in the Introduction .

The study yielded several important findings. Changing instructions so as to encourage the use of "none" resulted in a 9% increase in the frequency of "none" responses. The size of the increase was of about the same magnitude as the difference between Spanish monolinguals and Spanish speakers of English who received the same instructions (6% \/s. 14~,). Apparently, learning English heightens native Spanish speakers' awarcnes.s of acoU!ltic­phonetic differences between Spanish and English vowels. Directing subjects' allention to cross-language vowel differences via experimental instructions may result in a similar heightened awareness. This will need to be explored further, however, as the observed effect may simply have been due to a change in response bias (I.e. an overall greater propensity 10 respond " none").

There was little evidence that Spanish learners of English will judge IE/ in the context of word-final ", [0 be a "new" (non-Spanish) vowel. Even experienced Spanish speakers of English seem to continue identifying English /1£/ realizations in terms of their Spanish tal category. In this sense they differ from native speakers of Gennan, another language that has no /1£1 phoneme. Gennans seem 10 identify English lEI with Gennan leI or /e:/ (Bohn &. Flege, 1990). Native speakers of Mandarin Chinese and Korean also seem to identify English /z / with an /1'./ (or It /-like) category in their LI (Flege, 1990).

726 FLEGE

Some Spanish subjects in lIle present study seemed 10 be aware of the difference between English /1/ and vowels in the Spanish inventory. Inexperi­enced Spanish speak.ers of English applied Ihe "nonc" (i .c., non-Spanish) label to English /1/ more onen than did Spanish monolinguals. Also experienced Spanish speakers of English used "nonc" for /1/ more orten than did inexperienced Spanish speakers of English. The difference between the experienced and inexperienced subjects (42 'Y_ vs. IS-t.) just missed reaching significance, however (p - 0.(056).

The results obtained here for /II and Irtl provide insight into results obtained in a forced-choice identification experiment by FJege and Boho (1989). The mem~rs of synthetic continua ranging from "beat" to "bit" (/ii-III ) and "bet" to " bat" (fl /-/rt/) were presented to native speakers of English and to experienced and inexperienced Spanish speakers of English. Only a few Spanish subjects resembled the native English subjects in showing a clear crossover from {lI to /II responses as a function of spectral changes. (Those who did not use spectral changes divided the continuum on tbe basis of duration differences.) However, nearly all of the Spanish subjects re­sembled the native English subjects in showing a clear crossover from 1£1 to llel responses as a function of spectral changes in the stimuli . The results obtained here suggest that the Spanish subjects tested by Flege and Bohn (1989) probably did not succeed on the I£/-/rtl continuum as a result of having formed an Ir./ category. They may simply have appeared to resemble native English subjects as a result of identifying the Ir./ endpoint of the synthetic continuum in terms of their Spanish /a/ category. The few Spanish subjects who resembled the native English subjects for the /i/-M continuum, on the other hand, may have succeeded as the result of having formed an N category (unless, of course, they identified the /11 endpoint with Spanish Ie!).

An examination of responses to English 1£1 suggC$ted that the ortho­graphic responses may not have been based entirely on phonemic level classifications. Had this been so, we would have expected the < e> label to have been applied consistenUy to realizations of English Ie/. This is because Spanish leI has an [£] allophone said to match the EngJish 1£1 in " bet" (Dalbor, 1980). However, the subjects orten used < a > and "none" reo sponses for English 1£1. Of the 20 bilingual subjects in Experiment 2 (where use of the "none" response was not specifically encouraged), four used < e > in more than 7S t!. of instances to identify English leI. three used <a > in more than 75-/_ of instances. and the remaining 13 used both < e > and < a >. The < a > responses might be attributed to the acoustic phonetic similarity of certain realizations of English lEI and Spanish /al (see Figures 1 and 2). The "none" responses 10 En&lish tel might be attributed to the fact that the English /£/ realizations occurred in a phonetic context in which such vowels are not found in Spanish (i .e. preceding a word ·final / tJ).

The study did not provide strong support ror the hypothesis that etrtain

INTER LINGUAL IDENTIFICATION 727

o 1.40nOlin~uoll (E11' 1) • Bilinguo, ~EXP. ~ 6. Bilinguoll Exp. J

100

'" • • 80

'" • • is c • 8. 60 '" 0 '" • • • 8 '" • '" " '. ft 0 , c 40 0 0 • c

0 ~ • '" " 20

~I '" • .. ~ I ,

0 ! • * "bit" " bot"

FlO. 7. The percenlage or times tbe lubjecll in E.J:perillKnll 1- 1 \lied the "none" (i.e. nOI SP'lni.h) to labellhe Englis.h vowelJ /II and lEI (in bU, bul). 1bc mean values ror mou .ubjecu i. bued on 60 ruponlCS (10 native English !.Ilken)( 1 token, )( 2 randomiZition.) in mOil jlUltallCU.

English vowels will be treated as "new" by adult native speakers of Spanish. There was, however, a high degree or inter-subject variabi lity in the use orthe "none" response, especially for III and lle/. Figure 7 shows the frequency with which the "none" label was applied 10 these vowels by the 60 subjects who participated in Experiments 1- 3, It appears Ihat English llel and especially English /1/ fell outside the bounds of the Spanish vowel system for at least some subjects. It has been hypothesized that the authentic production of L2 sounds that are not identical to sounds in the 1I requircs the fonnation of additional phonetic categories, and that L2 phonetic category fonnalion is blocked when interlingual identification persists (Flege, 1988b; 1990; in press, a, b). The possibility therefore exists that the subjects who frequently judged English llel and /II to be non-Spanish were better at producing these vowels than were the subjects who seldom used the "none" label. Production was not tested in this study. however. It would be userul to examine production and orthographic classification together in a ruture experiment.

REFERENCES Amlluc, J. (1978). 1bc lIC.quililion or EnJ.lilh VOWeR. P~rl In /l1I,lIlsl/U, II, 42l-4,7. Belt, C., McRobert. , G ., &: Sitholc, N. (t988). E .... mjnation o r perceptual reorpniution ror

nonnltive .pecc:h contralla: Zulu click discrimination by En,lish-JpelIkinl adults and

728 FlEGE

inranlS. JolU1tlJi of £X~r~nla/ P~JlCJtoIory: lIunwt PlrCt!pllotl (JIId P~rformmru, fl, 345-, ... Bobn, 0 .·5. 4: Fleae, J . E. (1990). Interlin,UlI identification and the role or roreign lanluage

expniel)(Z in U vowel perception. Applied P3ychollnlwullu, 11 ,30)",]28. Bohn, 0.·5. &. FleF. J.E. (IUbmitted). The production or new and .imilar vowc:lI b, adult

Qcrmau amen or EnJlbh. Slwd/;u fit S#CotId Ltm,IlllIC Acqwuilion. BttMan, E. 4: Brennan. J . (1981). Meuurcmcnt or KUnt and attitude to_cd M«ican

Americ:an IpHlCb. JOl/nwJ of P'ycllolu.,WUI/c RUNTc", 10,417- SOI Britre. E. (1966). An investiption or pbono&op:.l IDtcrfffence. LtmIllil6', 41, 769-796. Briere, E. (196Il). A p/ttHIoIorialllludy of Inlvf","" · The Hap: Mouton. Butcher, A. (1976). The ioflucl)(Z or the native IIn,UlF on lbe pem:plion or vowel quality.

ArbtiuMrk"",6 (lnJtitute or Pboneda. Univenity or Kiel). Cllron1, J.e. (1968). CoatruliveanalysiJ anet IanJUlF teachin&- In J. Alltis (Ed.), COIltrQJIIPf'

/fltruUlla and Iu ~tIDtorrcaJ unplicoliotu. Wuhinlton: OeorJCtown Unlver'lily Pla.I. Oalbot, J (1910). SpanWt ".OMIIIClaliotr. Ilwory _/Natlltt New YOfk: tlolt, Rinehart. 4:

WinJlon. Ikllttre, p, (1964). Comparin, the vocalk: reaturet or Enaluh, Gennan. Spanitb. and Frend!.

/nl"""liotItJI RtfW of AppliN Lilqllulk~, I, 11-91. I)r;latlfC, P. (1966). A oompan.oD or J)'llIble Icn,th conciltionin, amon, Ian,UIlJell. lnltr·

ftQli«uI/ Rt," of App/led Unpullu, 4. 11)",198 Delallre, P. (1969). An acouJ1k: and articulatory Itud)' or vowc:1 redUC1ion in rour IangUlges.

InlmwlioMl &,n tlf Applied LiIlIuu.ia. 7. 29S-121_ Elman, J., Diehl, R .• 4: Bucbwald, S. (1917) Perccptual.witchinl in biJin,uall. Jounwl of 1M

ACOfUIkil/ Socidy of Anwrico, 61, 871 - 974. FIeF, J.e. (1911). The phonoloaklll M,i. orrom,n aa:enl. TESOL QNamrly, 1J,443-455. RclJC, J .E. (19&4). The deteo::lion or French ICXlC nt by Amenc.n Iiltener •• JOllrnal of'M

Aemulfall Soclely of A~rico. 76, 692- 707. FkF. J.E. (19871). A critical period ror learninl 10 pronounoe (oR'ian Ian,uaFll7 Applltd

Un'lIulk~, 8. 162- 177. Fkae, J.E. (l917b). The production or "new" and "Jimilar" phollQ in a rOrn," languaae:

eviderwz rOt tbe effect or equivalence clauiblion. J_NIl of P/roMI/Q, IS. 47-65, f1e." J.E. (l98b). FactOfS aJfectin, de.,..ee orpucdved rordlD IC:CZnt in Enalish Knteoces.

JOWJWi of 1M AtOlUlictJi Soc"" Df A"."./ca. 84. 10-79 FkIl'.I.E. (l98Ib). The productionlnet pm:cplion ortpCtlCh sounds in roreiJu Ian,UlJeII. ln

U. Winitz (Ed.), HUIfUln tOmmwfiCfllioft and III dUwJeu. A "fir'" 1988 (pp. n4--WI). Norwood. NJ: AtHell:.

f1eae. J.E. (198&:). The dn-eiopmml or Ikill in producin& wOfd·final / p/ and fbI: Kinemalk: panmeten. JOfU1ItJI of 1M Aemulkill S«lely of A"",/ca, 84. 1639-1652. fie.,. J.E. (1989a). Differenca in inventory Ji~ a/feet the location but not the precision or IOn,~ poIitionin, in vo .. -d produdtoo. UmtIlllJ' OM Sp«elt, 11, 12.)...141.

f1ese, J.E. (l989b)_ CbillCllC IUbjccu' pMq)tion or the word·llnaf En,IWI JI/-/dl contrut: Performal)(Z be.ron and all~ tnurunl- JOt/nIDI of IIw AcooulktJI SDtW'y of A","Ia>, 86, 1684-1697.

F'lcJC. J.E. (1990). Specdliearnin,tn 1 second IanJUl.IC. ln C Fer,lI$On, L Menn A C. Stoel­Oammon (Eds.). PllctwlOfictJi ik",laptrmtl medtlJ, rt_cA. aNI app/ltollon. Parkton, MD: York Press.

f1eF. J.E. (1991). Age or k.a.mID,.IfCCIJ the authentialY or yoke onlCl time (VOT) in stop c:onSOD.lnu produced in a $eCOnd lanaUlIJC. JOWI"fIIlI oflhe ACOIIJ IINI Sodrty of A"",rica. 89. 39s-411.

Fiep. J E (in prCM a). The tnlelllJlbtlit), or English vowebt SJM1kcn by Bnlish and Dutch ulkm. In R. Kent (Ed.), InttlfltUlflf,)' in spttth dilordm: IllIo'f, ~QSII,tnlPl', "nd nIIlIIQ,rm~nt. Philadelphia: 10hn BenjaminJ.

INTER UNGUAL IDENTifiCATION 729

f'k&c. J.B. (in prea b). P~n;:eption aod production. The: relevance of phonetic inpul 10 U phonolop:alIc:aminJ.In C. FCf,I1IOI1 It T . HLlCbllCr (&b.). C'Os.tcN"OIfJ In IffONi kuttwagl! tuq,,/sil/otl (IIIl/ IUtI"lst/e tMOq. Philadelphia: John Iklljamhu.

FlcIC. J.E.. (.ubmiucd). The UK Dr rhyminljudlilleni. 10 ~Yaluale illtCflinlLial identification Dr Spanish and EnJlish vowels. Journal of PhOMliu.

FkjC. J.E.." Bohn. 0 -S (1989). The petCCptioll of I?nalish yowell by lIadve Spanish .peaken. Journal of lhe ACOlUlkai Soc~ty of Antl!rlco. 85, 585 (AWIrK1).

fkjC. J.E.. • Eeninl. W. ( 1986) . l.ill.uistic and developmental etreeu On the production and pcnzplion or Slop COfUOnt.ntJ. Ph_tll:,.. 4J, 15$--111

Fkge, J.E.. It Eeftil1&. w. (19811). Imitation Dr a VOT eonlinuum by IlIItiVC JpUk~n or EnJlilh and Spanish: Evidenc::e ror phonetic ClIe,ory formation. Jl1IlTNlI of ,''' AtlHUlko/ Soc.wly of AmYicll. 6J. ng..740.

f'kse. J.E. & Fletcher. K. (submitted). AI what •• ollc:arnln. (AOL) doa a fordan KOCnt first become pcn;:eptibM:1 JOIUtt4l of 1M ACOIUlkaJ Soc~', of Atrlnkll.

f'kae. J.E. • Hillcnbnod, J. (1985). Difl"crcntir.1 ute of Icmpor&l ~ 10 lhe ft-7-J t.Ontrul by IIJI!tYC aod nOG-lIJIuve speakers of En,hsh. J_naJ of 1M ACOfUlkIlI Soc~/y of Antl!'lca, 19, 5011-511.

Rtp. J.E.. Hillenbrand. J. (1987). Difl'erenliallllC ofdosurc YlMCln,and rcleax bunt CUC$IO lIop Yoidn, by IIJIlive lpeaken of French Ind En&luh JOImI4J of PhotI"iu. ". 2Ill-208.

RtF- J.E., McCutcMon. M ... Smith. S. (1987) The: develoPfJJCnt of.ktll in producin, word­final EnJlUh 1I0fl1- Jountill of 1M AtauJlic4I Sod,/y of Antl!f'ktl, 6}, 0)..447.

f'kae. J.E., Munro. M., • Fol., R. (in prqI.). 11w ullnlltlpa/ Idtttlf/ktllitltl of Spotwlt aNI &,/W! 'H)"Wtb ' Dwimilo,;/, fa/iII,J. OepIrtment of Biocommunkallon. UniytTJily of Alabama II Binnlllgham.

fkce, J.E . .t. Wins. C. (1990). NlliYo-linlUlac phonolaelk eon.trainll .!fect how wdl Otinese lubjecu petuive tbe word·1l1ll1 Enllilh / t/-/d/ conlml. JOlUnaJ of PhQfl~/ics, 17. 299--315.

Godinu. Jr. M. (1978). A com~tltjYe lIudy of Mime Romance Yowell. UCU W.".."., Pa~rJ ill Phon"kJ, 41, l-19 (Dcpal1mcnt of Linguinics, Uni~enlty of C.lifornil at Lo5 Anlltlcl) .

Ollirao. M . .t. de M.nrique, B. (1\172). Identlfk:a tlon of ArIC-nlJne Sp!lnidi yowcl. In A. Ripull .t R. CharbonnCIIII (Ed • . ). Proc,ttllnlJ of Ih, 71h in/"1IIlIioMl toftl"» of ~Iic JC~naJ.

The UI,ue: Mouton. OutflO, M . .t de M'lIrique. B. (1975). IdentIfication of Aracnline Spanuh yowels. JDUrno/ of

PJyclw/UtIWJlic R,JIIllfch. 4, 11- 25. IIIJ1lDt1t'bc:rs. B. (in plUI). Conditions on transfer in phonoloJY. In J Leather & A. JImC:I

(Edt-). N,w IOUrId.. po. P'oatdill,J of 1M AmJ/,rdam Jympwllim "'" ucONJ.lwIIlIIIl~ JpHClI.

Dordreebl: Faris. lI.mmond, R. (1986). Error allllysil.nd lhe nllllt'II.ppro.ch 10 luchin, rorciJft IaO,I1IJeS.

U"IIUII MotknJIU. 11, 129---1l9. JohDlOn, K (1990) The: role of puceived apeaka- Mimhty In Fa normaltllUon of Yowell.

JDfUMI of 1M AeOll.llicaJ ~/, of A'fttricll. U, 642-6504 Klall, D. (1987). ReYiew of tul-ID-Ipeech convemon fOf Ea&lisb. JDNnIiJI of 1M AttNJllalI

Sock" of A".,.,ktl. 81. 137-793. Ladcfoaed. P. (1975). A eDUrJl! in ~/lcJ. New York: 1I'n:Dllrt, B.-.c:c JovanoYkh. Ladd'oscd. P (1989). Rt'prcseDtin. Phonelic Siructure. UCLA WOI"k.n, PapnJ in Ph_/icJ,

7J (Deparlmenl of Un(Llisties. UniYcnlly of Califomll '1 LoI: AnJdcs). Lennebera, E. (1967). ~ bioIogkaJfollllliilliotU of lan,,,"," New Yorli: : Wiley. MacDonald. M. (1989). The inllLIC.-w;:e ofSpt.nldi phooololY Olltbe En&1i'h lpaken by United

Sllles lii'pllDia:. In P. Bjarkman & R. Hlmmond (Ed •. ). Amnltwt Spmt/slr P'otIunciDllon . Wllhln.lon. DC: GeorJe Walhllliion University PIUI.

MacKlin, K .• Best. c.," SIr.nse. W (1981). CaltlOrII:a1 pcn:cpdon of En,hsb /r/ 'lid I1J by

730 FLEGE

Japanese blUnlual •. Appl~d Lin,uistics. 1. 369-390. Major, R.C. (1987). Pllonological.lmUarity, markedl\Clll, and rate or LI acquilitlon. SluditJ in

Suond Lon,~~ Acquisition. 9. 63-82. Mann, V. (1986). Distinluisltin& ullivertaland lansuage-dependentlevel. of.peech perception:

Evidence:: from J.~nese liJ\cncn' peroeplion ofEnglilh "I" and "r". Cogn/llon, U, 169-196. Miyawaki. K .. Slrange, W .. Verbrusse. R.. Libenn.n, A., Jcnlr:.illl. J ., &: Fujimur2, O. (1975).

An effect of linlUi.tic cxperlel'lCe: The di~rimin"ion of /rl and III by native apeaken or Japancsc.nd Enalilh. P~rupllon of Psychophysics. /8.3) 1- 340.

Mochizuki. M . (1981). The identification of Ir/ .nd III in n.tur.l and .ynthetie .peech. Journal a/ Phan~llcJ. 9, 28J-J03.

Mueller. T. AI. Niedzielski. H. (1963). The inllllc:TlCC or di~rimjnation training on pronunci­ltion. Modtrn umguogt JourM/, 52, 410-416.

Nav.rro Tomas, T. (19SO). Manrml tk pronwll:folion upoAola (6th cd.). Madrid: C.S.I.c. Oller. 1. '" Zi.houdny, S. (1970). The contrutive .nlly.i. hypothesi •• nd spellinB erron.

l..onguogt uom/n"10. 181- 189. Om.tein, J. (1974). Mexican- American JOciolin,ui.tiCi. In B. Hoffer. 1. Om.lein (Ecb.),

Sociolin,u"tic$ in Ih~ $o.uhWt!J/. San Antonio, TX: Trinity Univer.it), Prcu. Pap;un, O. (1976). How may vowel.ysu:nu dlfl'er1 UCLA WorkIn, Poptrl in I'hon~t/Q. J/,

38-4' (Department or Unlui.tics, Univetllily or C.Hfomia " Lo. Angeles). Patkowlki, M. (1990). Age and Iccent in a ICCOnd IlnBuage: A reply to Jlmes Emil Flege.

Appl/~d UllgulJllu, I J, 73-89. Petetllon,G. AI. Bamcy. H. (19'2). Control methods used in Istooy of the vowels. Joumalo/IM

Acoustical Sod~ly of Amu/co, U, 17S-184. Piaoni, 0 ., Aslin, R" Perey. A .. .t. Hennessey. B. ( 1982). Some efl'ectl ofl.bor2!ory lraininl on

idcntil\ca!ion and diJcrimination of vokinB conlfllStI. Journal 0/ £¥~r~"'al Psycholo,,: HUmiln Ptrupllon and I'~r/o'm(mu. 8. 297- 314.

Quiles, A. (1981). Fon~/lca Acus/lca tk 10 t..trrgrm £Spatiola. Madrid: Bibliotcca Romania HiJpanica.

Rcpp, B. (1981). Two strlliegies in fricative discrimination. p~ruP/{on (DId Psychophysfu, JO, 217-227.

Repp. B., Healy, A., &. Crowder, R. (1979). C.tcBories and conlext in the pe~ption of isolated steady-nale vowelt . JOUJ"1UJ1 0/ £¥~rilrwnIQI Psychology: Hum(HI P~rupllon (HId Ptr/orrn­anc~, j, 129-14'.

Sapon. S. (1952). An application of p5ycholopcalthcory to pronunciation problems in ICCOnd language learninl. Modtrn LDn,uast JourlUJ/, J6. 111- 114.

Sc:holeJ. R. ( 1967). Phoneme Cl tcgorUalion or synlhetic vocalic stimuli by speakers of Japanese. Spani.h. Perlian, and American En&lish. l..onguag~ and Spttch. 10,46-68.

Scholes, R. (l968a). Phonemic interrerence as. perceptual phenomena. lAIIguog~ and Spuch, 11,8&-103.

Sc:holea, R. (l968b). Peroeptual categoriution of eynthetK: vowels as. tool in dialectology and typolo8Y.l.ai!6ung~QlldSpnch, II. 194-207.

Scovel, T. (1969). FordlJl Iccentl, Ilngu.ge acquuition, and cerebral dominal'lCe. lAngungt uarniJr,,28, 129-142.

Scovel, T. (\988). A lime /0 Sptak. A plychofinguisllc: inquiry mlo 1M crllfcol period/or hlUl1Clll spt~ch . New York: Newbury HoU5C.

Sheldon, A. AI. Strange, W. (1982). The acquisition or Irl and Iii by Japanese learnen of English: Evidence thaI speech production am precede speech peroeption. Appli~d Psycho­IIII,islla. J, 243-261.

Skelton, R. (l969). The pe.ttern of Ss-nilh vowel sound •. In/~ma/lOffal Rc_kw 0/ Applifii Utrguisllcs, 1, 231- 237.

INTERLINGUAL IDENTIFICATION 731

Snedecllf, O.W . .t. Coclull.~, w.o. (1980). S'Il,utlatlrrw,hods. Amel,IA,: lowl Stal~ Unh'cr­si ly Prcu.

Siockwell, R . .t. Bowen, J. (196S). TM 1011mb 0/ Spa"uh and En,/uh, Chalo: Universily of <:halO Prw.

Trubell.koy, N (111)11) Gnllldr .. v dn plwtoforic. Berkcley. Univeuity or California Prcu (EnJlisb Iralll1llKlu by C. Baltue, 1\169).

Vlldman. A. (1976). 11I"odIKIIOII'o FUllch phONJio,y rmd mlWpJroIDlY. Rowley, MA: Newbury HOUM;.

Wanrrich, U. (I9j]). /..oJI,utlfCJ In cOfl'DCI, findlntS GIld "obIcrnJ. The llisue: Mouton. WHker. J. '" Lopn, J. (198S). Crou-lan,UlP evidence rOl" three faclors in spcClCb pcm:pdon.

PfflYP'ton..t PI~pIt)'siu,17, )s-44. Werker, J .. KJciu, R.,.t. Lopn, J. (1986). AJlditlW)', p4l1#1C,k, tIIId~"'k fHfK~iII, 0/"011'

..a,/.,. spud!; OplioNlJ (K obIwrllfN)'1 MlnlUenpl, Dalbousie Univcmly Wcrker, J . ... T-. R. (1\18]). Developmental wup aero.. cluidhood in lbe pcrcc:ption of

aon·ut1w: toundl. CllIuJdilm JDlrmQ/ 0/ Psyd!o/OIJi, 17, )5--44. Wetker, J . .t. T_, R. (1984). Pboannic and phonetic radon In adilit aou-Ian,uaF ~ ~ion. J_..aJ I1{IM AtofUlklll Sockl)' 0/ AnvrlN, 15. 186&-1811.

Wlik, K. (INS). Finruib and Ea&liab Vowda. ANIaIs 0/1"" U""crJlI)' oj Tiri:1I. P4 (Tuttu. Finland).

W<Xk, H. (l9TI). The L2 acquisilion of Ir/. PltiJMlica, ]4, 200-217. Wade. H, (1978), The beJinnin, of non-doal room L2 pbtNlolopc.1 acquisilion. illlff­

flQJioNI/ Rc.Ww oj Applin! U!wIlUlfu, 16. 109-12.5.