Vivian Cook  Online Papers  SLA Topics

Monolingual Bias in Second Language Acquisition Research
Vivian Cook
Revista Canaria de Estudios Ingleses, 34, 35-50, 1997

This paper questions whether second language acquisition research lives up to its claim to deal with the learner’s language as an independent system. First the paper describes the original assumption that learners have independent grammars and sees whether the terminology of second language acquisition research fits with this. Then it measures some typical L2 research methodologies against this assumption. It argues that the independence of the L2 user is a necessary tenet for second language acquisition research that has to be properly accommodated within the field, adopting the practice of bilingualism research.

1. The independent grammars assumption and Second Language Acquisition research

In the 1960s the child came to be seen as a speaker of a language of his or her own rather than as a defective speaker of adult language (McNeil, 1966). One way of analysing a child’s sentence such as Baby eat cookie is to take it as a defective version of the adult sentence I am eating a cookie, in other words what the child might say if he or she were an adult. The child has an insufficient knowledge of subject pronouns (baby), of the continuous tense (eat), and of articles (cookie). But, if the child has a grammar of its own, the sentence shows it contains no case distinction or grammatical inflections, a noun phrase does not need an article; the child’s sentence is a simple sequence of noun, verb and noun and so on. In its own right the child’s grammar is quite different from the adult’s set of rules. The child’s apparent ‘mistakes’ are only wrong when measured against adult speech, never wrong in terms of its own grammar.

This approach led L1 researchers not only to look more closely at actual children’s speech but also literally to write independent grammars for stages of children’s languages, for example the pivot-open grammar of Braine (1963), the negation grammars of Klima and Bellugi (1966), and the stages of acquisition of Brown (1973). The lone challenger was Smith (1973), who argued that the phonology of a young child could be described better through ‘deformation rules’ added to an adult grammar than through a grammar of its own. The overall approach can be summed up as the independent grammars assumption: language learners have language systems of their own that are independent of the systems of other speakers.

In some ways this independence is a spin-off from the recognition that the object of study of linguistics is linguistic competence—the ideal native speaker's knowledge of language (Chomsky, 1965a). All normal adult human beings have linguistic competence in their first language; competence is indeed defined as whatever it is that native speakers possess; any native speaker can then represent all native speakers. If the crucial aspect of language is the linguistic competence in the human mind, competence in a particular person is independent of other people’s minds. The competence of any individual is not more or less complete than that of any other individual: it is whatever it is. A three-year old has linguistic competence, a dialect speaker has linguistic competence, an adult 20 year-old has linguistic competence, each of them appropriate to its speaker. While the child’s competence grows into the adult’s, just as a caterpillar grows into a butterfly, it would be as foolish to describe the child’s grammar as that of a defective adult as to describe the caterpillar’s legs as defective wings. Any claim of deficiency could only be based on a comparison with their peers, whether fellow dialect speakers, fellow children or fellow butterflies.

The relevance of the independent grammars assumption to Second Language Acquisition research was realised by several people at roughly the same time. Pit Corder saw it as a ‘transitional idiosyncratic dialect’ (Corder, 1971); ‘Only by treating language learners’ language as a phenomenon to be studied in its own right can we hope to develop an understanding of the processes of second language acquisition’ (Corder, 1978, p.71). Nemser (1971) talked of an ‘approximative system’: ‘Learner speech at a given time is the patterned product of a linguistic system, La [approximative language], distinct from Ls [source language] and Lt [target language] and internally structured’. Most notably the independent grammars assumption featured prominently in the complex concept of ‘interlanguage’ (Selinker, 1972)—‘the existence of a separate linguistic system based on the observable output which results from a learner’s attempted production of a TL norm.’ The concept of interlanguage was in many ways the basis of modern SLA research; ‘what gave SLA its excitement was the concept of interlanguage’ (Davies, Criper, & Howatt, 1984, p.xii).

Typical introductions to the field of second language acquisition research continue to put forward the independent grammars assumption as a crucial component of interlanguage. Ellis (1994, p.354) sees interlanguage as the ‘almost theory-neutral’ term for the ‘system of implicit L2 knowledge that the learner develops and systematically amends over time’. Towell and Hawkins (1994) talk of ‘a grammatical system with its own internal organising principles which may or may not be related to the L1 and the L2’. Condition 1 in Spolsky’s synthesis of L2 learning (1989, p.16) is ‘A second language learner’s knowledge of a second language forms a systematic whole.’ Selinker (1992, p.260) claims that ‘the data studied in this volume reinforces the IL notion of the existence of a partially separate linguistic system united by what Weinreich has called interlingual identifications’.

One of the background ideas of second language acquisition research, perhaps the background idea, is therefore the independent grammars assumption, seldom discussed but never denied. Learners have a knowledge of language that is neither L1 nor L2 but something of its own, a true independent grammar. The fascination of the study of second language acquisition has indeed proved to be the differences from native knowledge of language.

2. ‘Failure’ of L2 learners to become native speakers

Another familiar theme of second language acquisition research is that the L2 learner’s language system is deficient compared to that of native speakers. Let us display a small gallery of quotations about L2 learners taken from well-known and well-respected SLA textbooks and articles of recent years to show the commonness of this assumption:

1. ‘failure to acquire the target language grammar is typical’. (Birdsong, 1992, p.706)

2. ‘learners often failed initially to produce correct sentences and instead displayed language that was markedly deviant from target language norms’. (Ellis, 1994, p.15)

3. ‘In L2 acquisition, on the other hand, it is common for the learner to fail to acquire the target language fully’. (White, 1989, p.41)

4. ‘children generally achieve full competence (in any language they are exposed to) whereas adults usually fail to become native speakers’. (Felix, 1987, p.140)

The common factor to all of these quotations is the word ‘fail’: L2 learners are failures. Other quotations continue the same theme:

5. ‘Very few L2 learners appear to be fully successful in the way that native speakers are’. (Towell & Hawkins, 1994, p.14)

6. ‘The lack of general guaranteed success is the most striking characteristic of adult foreign language learning’. (Bley-Vroman, 1989, p.43)

7. ‘the universal success of FLA and ... the relatively poor outcome of instructed SLA’. (Doughty, 1991, p.433)

These three quotations either paraphrase failure as ‘lack of success’ or contrast success by children with ‘relatively poor outcome’; L2 learners once again are failures.

8. ‘Unfortunately, language mastery is not often the outcome of SLA’. (Larsen-Freeman & Long, 1991, p.153)

Lack of “language mastery” is also presumably another paraphrase of failure.

These quotations could be duplicated many times. It seems that L2 learners are a sad group, more or less write-offs, never coming up to scratch. An early version is Selinker’s observation that 95% of L2 learners fail to achieve ‘absolute success in a second language’ (Selinker, 1972, p.33):

9. ‘... those adults who seem to achieve native speaker “competence”, i.e. those who learn a second language so that their “performance” is indistinguishable from that of native speakers (perhaps a mere 5% of all learners).’ (Selinker, 1972, p.49)

But what is the measure of success for these writers? Quotations (4), (5) and (9) explicitly mention the ‘native speaker’ as the touchstone; (1), (2), and (3) compare L2 learners with the ‘target language’, i.e. the knowledge of native speakers; (7) talks of the success of first language acquisition, i.e. the knowledge of native speakers; (4) refers to ‘full competence’, presumably that of native speakers; (8) uses the concept ‘language mastery’, presumably the knowledge of native speakers. Further quotations can illustrate the specific comparison to the native speaker:

10. ‘Condition 2. Native speaker target condition (typical, graded): second learner language approximates native speaker language’. (Spolsky, 1989, p.35).

11. ‘there have been a number of proposals to explain why, in general, adults fail to achieve full native-speaker competence’. (Felix, 1987, 144)

12. ‘Ideally, the end state represents a perfect command of the language’. (Klein, 1986, p.50).

The L2 learner’s goal is then ‘full native speaker competence’, paraphrased as ‘perfect command’ or ‘language mastery’. L2 learners are failures because they do not attain the same competence as native speakers. While these quotations are admittedly taken out of context, the cumulative effect in the SLA literature is clear and almost unanimous: L2 learners are failures compared to native speakers. Writers assume that the L2 learner needs to be compared with the native speaker, with only (10) hedging slightly with approximation to ‘native speaker language’.

On the face of it, it is paradoxical to assert simultaneously that L2 users have independent grammars of their own and that their grammars should be measured by those of native speakers: the two assumptions are incompatible. Yet the sets of quotations on the two assumptions in several cases come from the same people, for example Ellis, Towell and Hawkins, and Selinker. Clearly researchers see no contradiction between the independent grammars assumption and the vocabulary of ‘failure’. This could be a case of Orwellian double-think where someone believes two contradictory ideas at the same time or it might be that the Independent Grammars Assumption has been tacitly abandoned.

3. Methodology of L2 research

Perhaps actual L2 research methodology respects L2 users as independent people with independent grammars despite this negative terminology. A methodology for L2 research based on the independent grammars assumption would not spurn the native speaker as such, since much of the language the learner hears is derived from native speaker models. It would not, however, assume that the learner is a deficient native speaker but would describe the learner in ways that allow idiosyncracies of the learner’s own system to emerge. It could not measure L2 learners solely against a checklist of whether they have, or haven’t, mastered the competences of native speakers. Let us then see the extent to which some classic L2 methodologies live up to this assumption.

a) samples of learners’ productions

One basic research method is to collect examples of the learners’ speech or writing and to analyse them. Hardly surprisingly, the original intention was indeed to study the learners’ speech to establish the learner’s own language system through so-called Error Analysis; ‘everything the learner utters is by definition a grammatical utterance in his dialect’ (Corder, 1971, p.163). Only after this did the researcher go on to see if the causes for the learner’s system came from the L1, the L2, the situation, or some other factor in the learner’s mind. In some ways the only contradiction between Error Analysis and the independent grammars assumption was its name: learners do not make ‘errors’, they construct sentences that happen to be different from the target. As Corder himself recognises: ‘If, then, we call his sentences deviant or erroneous, we have implied an explanation before we have ever made a description’ (Corder, 1971, p.163). Hence the name ‘performance analysis’ came to be preferred. Though in practice Error Analysis has often meant finding out the ways in which the L2 learner has ‘failed’ to learn the L2, this is not inherent to the technique.

The concept of obligatory occurrences has often been applied to learner language. Brown (1973) defined obligatory occurrences as occasions on which an adult native speaker is obliged to use particular morphemes in a sentence. One conspicuous SLA use of the technique was in the grammatical morphemes research from Dulay & Burt (1972) onwards. The essential method was to elicit language though a task such as picture description and then to evaluate the success which the L2 users supplied the grammatical morphemes in linguistic ‘obligatory occurrences’.

But for whom are these occurrences obligatory? A revealing quotation comes from Dulay & Burt (1973, 253-254); ‘For example in the utterance “she is dancing” a mature native speaker of English would never omit the functor -ing because it is obligatory that -ing be attached to any verb in English when expressing a present progressive action’. The obligatory occurrences technique measures the extent to which learners conform to the grammars of native speakers for whom the occurrences are actually obligatory. The percentages of success are in a sense percentages of native speakerness. The technique restricts L2 learners’ sentences to deviations from those of native speakers. It cannot in itself reveal if, say, they have a totally different system of their own; no-one knows if the occurrences are obligatory for their grammars. The description of the learner is being forced into the mould of the native speaker. The research perceives the learners through a particular window of obligatory occurrences that cuts out all their differences from the native speaker. Klein and Perdue (1992, p.333) warn in particular of the ‘closeness fallacy’ where learner utterances have a false resemblance to those of the L1 speaker. However good the intention to show the independence of L2 learners, this cannot be achieved through a technique that only admits natives to be correct, that crucially depends on native speaker comparison.

More recently the obligatory occurrences technique has been applied within the Multidimensional Model put forward by Manfred Pienemann. In the original ZISA project (Meisel, Clahsen and Pienemann, 1981) the learners’ sentences were scored for the extent to which they used a grammatical rule in obligatory contexts. Pienemann sees interlanguage as ‘the sum of all the rules a learner has acquired so far’ (Pienemann, 1989, p.54). On the one hand then the methodology is restricted to native speaker comparison through obligatory occurrences. On the other the model treats the learners, not as developing a system of their own, but as adding bits of some ideal grammar of the language, presumably the native speaker’s. Again the Independent Grammars Assumption is apparently broken.

So the obligatory contexts research technique reveals nothing about the nature of the interlanguage per se, only about its deviations from the native target. A researcher could hope to build up a picture of the interlanguage indirectly from such data and could try to see what unique interlanguage system could explain these patterns of deviancy; describing an apple in terms of a pear may capture some of its characteristics compared to an orange But this second stage deduction constrains the extent to which the peculiarities of the interlanguage itself can be established to its deviations from native speech; the essence of an apple could never be pinned down in terms of pears. If interlanguage is indeed an independent language, scoring learner speech for obligatory native contexts is as absurd as scoring English for presence of Italian morphemes.

b) grammaticality judgements

Grammaticality judgement tasks require people to say whether particular sentences are ‘acceptable’, ‘OK’, or ‘grammatical’, depending on the researcher’s preference. Linguists have seldom employed full-scale grammaticality judgement tasks when studying native speakers: instead the linguist invents a single sentence which is so obviously part of the language that it is up to someone to challenge this. Nor have grammaticality judgements often been used with L1 children, since it is unclear how meaningful the task may be to them, though one group has tried to reinstate the technique (McDaniel & Cairns, 1990).

In second language acquisition research, grammaticality judgement tasks are common in recent years, though not without substantial criticisms, by, for example, Birdsong (1989), and Winitz (1996). Typical examples are Bley-Vroman, Felix, and Ioup (1988), who used them to test L2 learners' knowledge of subjacency as did Johnson and Newport (1991) and White (1985), White (1986) who used them to test pro-drop, and Gass (1979) to test relative clauses. Their link to the native speaker has mostly gone unnoticed. For, whose is the judgement of which sentences are grammatical or ungrammatical if not that of the native speaker? Implicitly the purpose of the grammaticality judgements task is to establish whether the L2 learner lives up to the native norm. Sometimes this link may be implicit; natives are assumed to score close to perfection on such tasks without actual investigation, thus often resulting in the comparison of the L2 learner’s performance with the native speaker’s competence, naturally to the disfavour of the L2 learner. On other occasions this link to the native speaker may be explicit; a control group of native speakers carry out the same task, thus at least ensuring that legitimate comparisons are made between the performance of both groups. Take the well-known research by Bley-Vroman, Felix, & Ioup (1988) on subjacency, which showed inter alia ‘slightly over half of the non-native speakers typically exhibit the correct UG-based judgements on any given UG effect’ (p.24). What is a correct UG judgement here but that of a native speaker?

The lengthy debate on access to UG in second language acquisition has mostly phrased the question, not as ‘do L2 users have grammars that incorporate principles and parameters of Universal Grammar?’ but as ‘do L2 users have grammars that are the same as those of native speakers?’ It might be perfectly possible to have access to UG but to create a grammar that is different from that of the target native speaker, for example the learners studied by Finer and Broselow (1985), who have settings for the binding parameters found neither in the L1 nor the L2 but permitted by Universal Grammar; after all UG licenses all the languages that actually exist together with many more that have not yet existed. Testing L2 users for approximation to L1 grammars is no test of UG. Any L2 grammar might well be a possible language according to UG though different from the native’s. If the independent grammars assumption is adhered to, the best that a grammaticality judgements test can do is compare the judgements of L2 learners who have one grammar of a language with those of native speakers, who have another.

To meet the independent grammars assumption, the grammaticality judgements test could not be directly aimed at comparison with the native speaker but at discovery of the L2 user’s grammar. In this regard, the use with non-native speakers would parallel the acceptability experiments of Quirk and Svartvik (1966) which tried to establish the boundaries of what the English speaker knows, not comparing them prescriptively with a group to which they do not belong. A parallel use would then test the unique grammatical intuitions of L2 users, rather than whether these intuitions correspond to those of native speakers. It would establish what the learner’s grammar was, not whether it differs from the native speaker’s.

c) other methods

Other less popular methods also depend on the native speaker. The technique of elicited imitation was borrowed from first language acquisition ideas by Slobin and Welsh (1973) and has appeared sporadically in L2 research ever since, for example Cook (1973), and Flynn (1987). Elicited imitation requires L2 users to repeat a particular sentence and then evaluates any changes or omissions that are made; it makes a straightforward comparison with the native speaker. As Flynn (1987, p.88) puts it, ‘the linguistic behaviour elicited reflects a map between the target language grammar and the learner’s developing grammar’. The sentences for repetition are chosen because of their relevance to the native speaker; the deficiencies of the L2 user are measured in terms of what native speakers are supposed to do in the same circumstances. Again to quote Flynn, this ‘allows us to measure an L2 learner’s development vis-à-vis the native speaker’s model’. While Flynn did not test native controls, the prime motivation for her research being the comparison of different L2 groups, quotations like these make it clear that the standard of comparison is always what a native speaker would do. Or, to take more recent examples, the crosslinguistic test of vocabulary devised by Kempe and MacWhinney (1996) had to start from ‘a native speaker base-line of performance against which FL learners could be compared’; Birdsong (1993, p.717) describes his study as construing 'ultimate attainment in L2A in terms of whether nonnatives can display evidence of possessing native linguistic norms': in other words the usual native speaker basis is present in both.

It would nevertheless be misleading to give the impression that all L2 research is committed to the comparison with the native speaker, either by necessity or in practice. There are individual exceptions that conscientiously bear in mind the independent grammars assumption. Young (1991) for example examined variation in L2 users’ speech to listeners of different races; Adamson and Regan (1991) looked at the variation between the sexes in use of /i„/ or /in/, not in terms of successful native use but in terms of the contrast between the two sexes of L2 users. Vast areas of SLA research are inherently concerned with unique aspects of the L2 process itself rather than anything in the monolingual state. For example there is little attempt in individual variables research to compare L2 users with native speakers in the area of motivation; the learner needs a reason for learning a second language but not for learning a first; it would be as meaningful to ask the child’s motivations for learning to eat. Classroom learning of second languages is related to the classroom learning of other subjects, not to first language acquisition; cognitive style, learning strategies, extroversion/ introversion, aptitude, or even IQ have little relevance to first language acquisition, however crucial they may be to classroom second language learning.

An example of work that explicitly accepts the independent grammars assumption is the European Science Foundation research with European migrant workers, which claims: ‘The question to be answered is whether a learner variety is based on recognisable organisational principles, how these principles interact, and whether they also apply to fully-fledged languages’. (Klein and Perdue, 1992, p.1). By looking at learners’ productions in their own terms rather than in native-biased obligatory contexts, this project established a set of five common principles operating at the base stage of L2 acquisition using data from four different L2s involving five L1s. Learners from different backgrounds learning different L2s achieve a ‘basic variety’ which represents ‘a relatively stable and natural equilibrium between semantic, pragmatic and phrasal constraints.’ (p.315), a common interlanguage system in its own right.

4. Prejudice in second language acquisition

There seems then to be inconsistencies between, on the one hand, the assumption that learners have independent grammars and, on the other, the vocabulary of ‘failure’ and the native-biased methodology. Many SLA researchers would probably answer that the independent grammars assumption is not a core element of second language acquisition research; it belonged to an early stage of second language acquisition research and no-one has bothered to repudiate it; their research would not be substantially altered if the assumption were dropped and SLA research was explicitly native-oriented. This section argues that the independent grammars assumption is necessary if second language acquisition research is to be a reputable sub-discipline of linguistics.

One way in is to transform some of the quotations given earlier into remarks about the relationship between men’s and women’s language.

13. ‘Failure to acquire men’s grammar is typical of women’. (=1)

14. ‘Women often failed initially to produce correct sentences and instead displayed language that was markedly deviant from men’s norms’. (=2)

15. ‘It is common for women to fail to acquire men’s language fully' (=3)

16. ‘Very few women appear to be fully successful in the way that men are’. (=5)

17. ‘The lack of general guaranteed success is the most striking characteristic of women’s language learning’. (=6)

18. ‘Unfortunately, language mastery is not often the outcome of women’s language learning’. (=8)

These sentences are unacceptable nowadays; it is no longer possible to make such blatantly prejudiced remarks. But why should it be acceptable to say that an L2 learner is a failure but unacceptable to say that a woman (or representative of any other minority group in a society – blacks, working class, or whatever) is a failure? Why should an L2 learner’s speech be ‘markedly deviant’ but not a woman’s?

The nub of Labov’s sociolinguistic argument about discrimination is that one group should not be measured against the norm of another group (Labov, 1969). English-speaking women pronounce the -ing ending of the present tense (looking) as /i/ rather than /in/ more often than men everywhere in the English-speaking world (see for example Chambers, 1992). Measure women’s speech against men’s and they are unsuccessful. But so are men when measured against women. Black Americans (and Singaporeans) use sentences without copula verbs, such as John happy, more than white Americans; measuring Black American English against White American English will naturally show the inferiority of Black English. Lower class English boys pronounce hat as ’at (see Milroy, 1982 for a survey). Clearly they score badly compared to the non-h-dropping middle-class.

Labov’s argument is that people who speak differently from some arbitrary group are not using grammars that are better or worse: they are just speaking differently. Women’s use of /i„/, black Americans’ zero copula, working class h-dropping are different from the other sex, race or class, but they are no worse and no better as language. Each of these groups have language systems of their own but none of them are wrong. A language comes in many shapes and colours as spoken by the young, the old, the women, the men, the lords in their palaces, the poorman in his hovel. Women have the right to be women, as blacks have the right to be black, or indeed men to be men. And they have the right to show which group they belong to through their language. Given that there are two groups of speakers, Group A and Group B, it is meaningless to say that Group A does not conform to the norm of Group B. All branches of linguistics accept the creed of ‘difference not deficit’ with rare exceptions.

Why then should L2 users alone be singled out as deficient for being what they are, bilinguals, and not what they are not, monolinguals? By definition a second language user is not a monolingual and will never be, just as women and men are incapable of changing places. There are exceptions – women who function as men, and vice versa, L2 users who pass as monolinguals in their L2. But it is no more sensible to base a study of L2 acquisition on the exceptional cases than it would be for a study of sexual characteristics. L2 users have to be looked at in their own right as genuine L2 users, not as imitation native speakers.

Yet both lay-people and researchers have taken it for granted that L2 learning is a special case. L2 users are not believed to have the rights of other groups; it is perfectly respectable to denigrate them as failures compared to a group that they are not by definition. When their grammar differs from native speakers, when their pronunciation betrays where they come from, when their vocabulary differs from native usage, these are taken as signs of their failure to master the second language. SLA research assumes that Group A, L2 users, should be measured in terms of Group B, native speakers. Just as it was once claimed that women are failed men, blacks are failed whites, working class children should be more like the middle-class, deaf people should not use sign language, so L2 users are failed specimens of native speakers. The only target of second language acquisition is native speaker competence. In any other context in language studies this would be treated as blatant prejudice.

However, the second language acquisition research literature shows scant realisation that this issue is relevant; only Bley-Vroman (1983) attempted to apply the logic of Labov’s argument to second language acquisition. In as much as there is an answer to the charge that is being brought, it presumably would be that second language acquisition is indeed a special case. One argument might be that L2 learners are different from other groups in that all the others are native speakers. A woman is a native speaker of woman’s language, a thirty-year-old Glaswegian taxi-driver is a native speaker of thirty-year-old Glaswegian, and so on. The reason for respecting the language of these groups is that their native speakerness is them; your group membership is proclaimed through your speeh whether you come from Yorkshire or Barbados. L2 learners are different since they are indeed not native speakers of the language by definition. But again this comes back to the hallowed status of the native speaker. Why does only the native speaker have language rights? A second language is as vital to the lives and occupations of many people as a first; why should their activities be denigrated by being seen as pale imitations of the ‘true’ lives and occupations of monolinguals? The reason for language discrimination between groups seems to be the assertion of power through language, whether men over women, whites over blacks, the healthy over the disabled. Since the 1970s it is no longer possible to casually assume that one sex or one race should take precedence in linguistics. With the exception of the power asserted over the mentally ill through the popular use of terms like ‘schizophrenic’, the only group that can be treated with such contumely are second language learners. As Sridhar and Sridhar (1986) point out, ‘Paradoxical as it may seem, Second Language Acquisition researchers seem to have neglected the fact that the goal of SLA is bilingualism’.

Nor is it just second language acquisition research that assigns god-like status to the native speaker. Virtually all linguistics assumes that bilingualism can be tacked on as an appendix to the study of monolingualism rather than being a topic of its own, deliberately ignoring the competences of most human beings in the world. The grammar that linguists describe is that of the ‘pure’ native speaker, unsullied by the knowledge of other languages. A ‘normal’ human being knows English or French, but not both. To take a typical quotation: ‘We exclude, for example, a speech community of uniform speakers, each of whom speaks a mixture of Russian and French (say, an idealised version of the nineteenth-century Russian aristocracy). The language of such a speech community would not be "pure" in the relevant sense, because it would not represent a single set of choices among the options permitted by UG but rather would include "contradictory" choices for certain of these options.’ (Chomsky, 1986, p.17). Competence belongs to the idealised native speaker in a homogenous community; it is an abstraction from the real person in the real situation. Dealing with people who know two languages would be too complex. Or, to use another metaphor of Chomsky’s, if you want to study water, you look at it in its purest possible form, not taking samples direct from the Hudson River: the competence of the monolingual must be the norm for purposes of description, regardless of the fact that there are probably more people in the real world who know more than one language than there are monolinguals. Illich and Sanders (1988) claim ‘From Saussure to Chomsky "homo monolinguis" is posited as the man who uses language - the man who speaks.’ To reply to Chomsky’s argument in an equivalent metaphor, you do not break water up into H2 and O to study it, you look at the molecule H2O; in other words, bilingualism is not just the concatenation of L1 and L2 but a state of its own.

A second counter-claim is that most L2 learners actually want to be taken for monolinguals, in other words if Group A (L2 learners) wants to become Group B native speakers, it is justifiable to measure them by the standards of Group B and consequently to claim that they are all failures. In former days some women adopted male styles of language or male pseudonyms as writers from Asher Bell to George Eliot to James Tiptree Jr; the jazz musician Mezz Mezzrow wanted to be a white negro; some people have operations to make their eyes appear more Western, and so on. Many people indeed want to be things that they are not and can never be, but this does not mean that these goals are necessary or indeed attainable. People’s wishes and wants are the products of their lives and experiences of the world, the attitudes and stereotypes of their society, and the pressures of business selling techniques. Much of the liberation politics of the 1960s onwards has established the rights of people not to want to be members of another group; blacks no longer want to pass for whites, gays for heterosexuals, etc. Indeed many L2 learners do not actually want to become native speakers; they need the language for the myriad of personal reasons that people have for learning another language—their religious beliefs, their international careers, their holidays, etc; they require the second language for the purposes of an efficient L2 user, not for those of an native speaker; many learners are inspired by instrumental motivations rather than integrative (Dornyei, 1990). François Grosjean (1989) points out that bilinguals have often been affected by monolingual prejudice; bilinguals have been made to feel guilty that they are not native speakers, however efficient they may be at a second language. Even if their motivation is avowedly to become native speakers, this is the product of the forces in society that regard L2 learners as failed native speakers, neatly exemplified by the SLA experts above.

But the one thing that the L2 learner cannot be by definition is a native speaker. L2 learners are not a group like children that automatically metamorphoses into another; they are not even a group like men that can occasionally change into women; they are not native speakers for now and for ever more. Native-speaker-hood is not a changeable condition, unless time-travel permitted the person to be born of different parents speaking another language. The ability to achieve ‘native speaker competence’ (9, 11) would be as miraculous as a duck turning into a swan.

A third answer might be that second language acquisition research is concerned with learners, by definition an interim state defined in terms of the target they are heading for. On the one hand this is simply a denial of the independent grammars assumption; the grammars of L2 learners and L1 children are defective versions of the final state of competence not systems in their own right. On the other hand it is a sleight of hand over the term ‘learner’. An adult native speaker who has completed their L1 acquisition is not referred to as an L1 learner. Yet in much of the literature L2 learners are learners for the whole of their lives; they never acquire another term, unless it is ‘bilingual’. Since only a small minority achieve an L2 system indistinguishable for the native’s L1 grammar, the vast majority are learners for evermore. The term ‘L2 user’ is far less prejudicial. Most L2 users who have finished their L2 learning can only be considered ‘learners’ if they are, naturally enough, defective native speakers.

5. Bilingualism and the independent speaker

Alongside the newcomer of SLA research, there has existed a long tradition of research into bilingualism, which to a large extent forms a separate academic field. Most texts on bilingualism debate at length the varying definitions of bilingualism. One parameter of variation is the amount of language that needs to be known, from maximal—the balanced bilingual ‘with native-like control of two languages’ (Bloomfield, 1933), otherwise called the ambilingual (Halliday, McIntosh and Strevens, 1964)—to minimal —’the point where a speaker can first produce complete meaningful utterances in the other language’ (Haugen, 1953, p.7).

Bilingualism researchers are more reluctant to subordinate their subjects to monolinguals than are SLA researchers. Romaine (1989, p.282) insists ‘it is clear that a reasonable account of bilingualism cannot be based on a theory which assumes monolingual competence as its frame of reference’. Appel and Muysken (1987, p.3) lament ‘All too often imposing Bloomfield's criteria on bilinguals has led to their stigmatisation as being somehow deficient in their language capacities.’ Hoffman (1991) states ‘For the vast majority of bilinguals, “bilingual competence” is not measurable in terms of monolingual standards’. These read very differently from the quotations from SLA researchers. Many bilingualism researchers furthermore describe the benefits of bilingualism over monolingualism, such as the cognitive and social advantages, and extra skills such as code-switching. In short, the ability to function through two languages cannot be measured by monolingual competence: a person who can juggle two balls cannot be limited by the standards of somebody who can juggle only one. The overall emphasis in bilingualism studies is an acceptance of the L2 user as an L2 user to be measured by the standards of L2 users, not by those of monolinguals. The L2 user is not an imitation monolingual in the L2 but a genuine bilingual—a type all of its own: bilingualism is not double monolingualism but a different state. The ‘ultimate attainment’ of second language is not, and could never be, monolingual competence.

So bilingualism researchers generally deny that the description of bilingualism is parasitic on the monolingual: L2 users are not failed monolinguals but successes in their own right. Most researchers nowadays concur with Mackey that the definition should be towards use rather than knowledge; ‘Bilingualism is not a phenomenon of language; it is a characteristic of its use’ (Mackey, 1970). What counts is whether the speaker can use two languages, not whether he or she ‘knows’ them. Bilingualism research allies itself more with a sociolinguistic notion of variability. Romaine (1989) explicitly rejects a competence-based approach in favour of use and blames the Saussurean structuralist tradition for ‘the belief that an entity, whether it is a society, language, or so forth, can be viewed as a structured self-contained whole, an autonomous entity, which is consistent with itself’ (p.286).

To conclude, the field of second language acquisition research needs to take heed both of the assumptions current in other fields of linguistics and of those in the closely related field of bilingualism studies. The aim is not to measure the L2 learner against a native standard but to discover ‘why ... adults attain the state they do’ (Klein and Perdue, 1992, p 334). Bley-Vroman (1983) pointed out that the development of systematicity within interlanguage was inherently native-biased, but no-one appears to have been listening. The current paper goes further by claiming that the vast majority of SLA research is tarred with the brush of native comparison. SLA research has viewed L2 learning from a monolingual standard, and concerned itself with the differences of L2 learners from natives; very few of its results stand when native comparison is removed. It is necessary on the one hand to develop the handful of research that does treat the L2 learners in their own right, on the other to attempt to salvage the parts of L2 research that can be interpreted without reference to a native monolingual standard. But will anyone take heed?

References

Adamson, H. & V. Regan. 1991. 'The acquisition of community norms by Asian immigrants learning English as a second language: a preliminary study'. Studies in Second Language Acquisition 1:1-22.

Appel, R. & Muysken, P. (1987), Language Contact and Bilingualism, London: Edward Arnold.

Birdsong, D. (1992). 'Ultimate attainment in second language acquisition'. Language 68: 706-5 5.

Birdsong, D. 1989. Metalinguistic Performance and Interlanguage Competence. New York: Springer.

Bley-Vroman, R (1983), ‘The comparative fallacy in interlanguage studies: the case of systematicity’, Language Learning, 33:1-17

Bley-Vroman, R.W. (1989), 'The logical problem of second language learning', in S. Gass & J. Schachter (eds.), Linguistic Perspectives on Second Language Acquisition, CUP, 41-68

Bley-Vroman, R.W., Felix, S., & Ioup, G.L. (1988), 'The accessibility of Universal Grammar in adult language learning', Second Language Research, 4 (1), 1-32

Bloomfield, L. (1933), Language, Holt, New York

Braine, M. (1963)), 'The ontogeny of English phrase structure: the first phase', Language, 39, 1-13

Brown, R. (1973), A First Language: The Early Stages, London, Allen and Unwin

Chambers, J.K. (1992), ‘Linguistic correlates of gender and sex’, English World-Wide, 13, 2, 173-218

Chomsky, N. (1965a), Aspects of the Theory of Syntax, MIT Press

Chomsky, N. (1965b), 'Formal discussion: the development of grammar in child language', in U. Bellugi and R. Brown (eds.), The Acquisition of Language, Indiana, Purdue University

Chomsky, N. (1986), Knowledge of Language: Its Nature, Origin and Use, Praeger, New York

Cook, V.J. (1973), ‘The comparison of language development in native children and foreign adults’, IRAL, XI/1, 13-28,

Corder, S.P. (1971), 'Idiosyncratic errors and Error Analysis', IRAL, 9, 2, 147-159. Reprinted in Richards (1974)

Corder, S.P. (1978), 'Language-learner language', in Richards, J.C. (ed.) Understanding Second and Foreign Language Learning: Issues and Approaches, Newbury House, 71-9

Davies, A., Criper, C., & Howatt, A. (ed.) (1984), Interlanguage, Edinburgh University Press

Dornyei, Z. (1990), ‘Conceptualising motivation in foreign-language learning’, Language Learning, 40, 1, 45-78

Doughty, C. (1991), ‘Second language instruction does make a difference’, Studies in Second Language Acquisition, 3, 431-469

Dulay, H.C. & Burt, M.K (1973), ‘Should we teach children syntax?’ Language Learning, 3: 245-57.

Dulay, H.C. & Burt, M.K. (1972), ‘Goofing: an indicator of children's second language learning strategies’, Language Learning, 22: 235-252.

Ellis, R. (1994), The Study of Second Language Acquisition, OUP

Felix, S. (1987), Cognition and Language Growth, Foris, Dordrecht

Finer, D., & Broselow, E. (1985), 'Second language acquisition of reflexive binding', NELS, 16

Flynn, S. (1987), A Parameter-setting Model of L2 Acquisition, Dordrecht, Reidel

Gass, S. (1979), 'Language transfer and universal grammatical relations', Language Learning, 29, 327-344

Grosjean, F. (1989), 'Neurolinguists, beware! The bilingual is not two monolinguals in one person', Brain and Language, 36, 3-15

Halliday, M.A.K., McIntosh, A., & Strevens, P. (1964), The Linguistic Sciences and Language Teaching, Longman, London

Haugen, E. (1953), The Norwegian Language in America, Philadelphia: University of Pennsylvania Press.

Hoffman, C. (1991), An Introduction to Bilingualism, Longman, London

Illitch, I. & Sanders, B. (1988), ABC: Alphabetisation of the Popular Mind, North Point Press, Berkeley

Johnson, J.S. & Newport, E.L. (1991), 'Critical period effects on universal properties of languages: the status of subjacency in the acquisition of a second language', Cognition, 39, 215-258

Kempe, V. & MacWhinney, B. (1996), ‘The crosslinguistic assessment of foreign language learning’ , Applied Psycholinguistics, 17, 149-183

Klein, W. & Perdue, C. (1992). Utterance structure: developing grammars again. Amsterdam: John Benjamins.

Klein, W. (1986): Second Language Acquisition. Cambridge: Cambridge University Press.

Klima, E.S., & Bellugi, U. (1966), 'Syntactic regularities in the speech of children', in Lyons, J., and Wales, R.J. (eds.), Psycholinguistics Papers, Edinburgh U.P.

Labov, W. (1969), ‘The logic of non-standard English’, Georgetown Monographs on Language and Linguistics, 22, 1-31

Larsen-Freeman, D. & Long, M. (1991). An Introduction to Second Language Acquisition Research. Longman, London & New York.

Mackey, W.F. (1970), 'The description of bilingualism', in J. Fishman (ed.), Readings in the Sociology of Language, The Hague, Mouton, 554-584

McDaniel, D. & Cairns, H.S. (1990), ‘The child as informant: eliciting linguistic inutuitions from young children’, Journal of Psycholinguistic Research, 19, 5, 331-44

McNeill, D. (1966), 'Developmental psycholinguistics', in Smith, F., & Miller, G.A. (eds.), The Genesis of Language: a Psycholinguistic Approach, MIT Press

Meisel, J.M., Clahsen, H. & Pienemann, M. (1981), 'On determining developmental stages in natural second language acquisition', SSLA, 3, 2, 109-135

Milroy, J. (1982) ‘ On the sociolinguistics of h-dropping in English’, Odense University Studies in English, 4, 37-53.

Nemser, W. (1971), ‘Approximative systems of foreign language learner’, International Review of Applied Linguistics, 9:115-24

Pienemann, M. (1989), 'Is language teachable?', Applied Linguistics, 10, 1, 52-79

Quirk, R. & Svartvik, J. (1966), Investigating Linguistic Acceptability, Mouton, The Hague

Richards, J (ed.) (1974), Error Analysis, Longman, Harlow

Romaine, S. (1989), Bilingualism, Blackwell, Oxford

Selinker, L. (1972), ‘Interlanguage’, International Review of Applied Linguistics, X(3). Reprinted in Richards (1974)

Selinker, L. (1992). Rediscovering Interlanguage. London: Longman.

Slobin, D.I. & Welsh, C. (1973). Elicited imitation as a research tool in developmental psycholinguistics. In Ferguson, C. & Slobin, D. (eds.), Studies of Child Language. New York: Holt Rinehart Winston

Smith, N. (1973) The acquisition of phonology, CUP

Spolsky, B. (1989). Conditions for Second Language Learning. Oxford: Oxford University Press.

Sridhar, K.K., & Sridhar, S.N. (1986), 'Bridging the paradigm gap; second language acquisition theory and indigenised varieties of English', World Englishes, 5, 1, 3-14

Towell, R., & Hawkins, R. (1994), Approaches to Second Language Acquisition, Multilingual Matters, Clevedon

White, L. (1985), 'The acquisition of parameterised grammar: subjacency in second language acquisition', Second Language Research, 1, 1, 1-17

White, L. (1986), 'Implications of parametric variation for adult second language acquisition: an investigation of the pro-drop parameter', in Cook, V.J. (ed.) (1986), Experimental Approaches to Second Language Acquisition, Oxford, Pergamon

White, L. (1989), Universal Grammar and Second Language Acquisition, Amsterdam, John Benjamins

Winitz, H. (1996), ‘Grammaticality judgements as a function of explicit and implicit instruction in Spanish’, Modern Language Journal, 80, 132-46

Young, R. (1991), Variation in Interlanguage Morphology, Peter Lang, New York