Language Sciences 32 (2010) 589-593 Contents lists available at ScienceDirect Language Sciences journal homepage: www.elsevier.com/locate/langsci The native speaker and the mother tongue Nigel Love**, Umberto Ansaldo” {Department of Linguistics, University of Cape Town, 7701 Rondebosch, South Africa "Department of Linguistics, University of Hong Kong, Pokfulam Road, long Kong ARTICLE INFO ABSTRACT Keywords: ‘This article presents a historical account of the role and function in linguistic theorising of Language and identity the concepts “native speaker" and “mother tongue”, and serves to introduce a number of popolingtatien! articles (Language Sciences vol. 32 no. 6) raising questions about various aspects of the aes idealised monolingualism that underlies much modern linguistics. Native speaker ‘© 2010 Published by Elsevier Ltd, Ownership of languages If you were born and brought up from birth to speak a given language, for preference in a family where the parents or other adults had the same experience with the same language, then you are a native speaker of that language, which is your mother tongue. The term ‘native speaker’, and this simple account of what a native speaker is, seem to have been introduced into modern linguistics by Bloomfield (1935 {1933}, p. 43).' Since Bloomfield’s time, whether all or even most linguistic communities, con- temporary o historical, are structured in such a way as to leave room for the concepts ‘native speaker’ and ‘mother tongue’ to play a useful role; indeed, whether there is any such thing as a native speaker or a mother tongue, in this or any other proposed senses of these terms, have been hotly debated questions.” ‘A number of the articles in this issue of Language Sciences are contributions to the debate. Rajend Mesthrie questions “whether the traditional concept of the native speaker of English has any relevance in the globalized world”, and contends that “native speaker’ and ‘mother tongue’ are no longer transparent terms in the sociology of English”. Chaise LaDousa ob- Serves that students of language variation in India have been especially alive to the hewildering range of meanings and uses 1 the notion “mother tongue’, and quotes Pattanayak (1981, pp. 47-48) on the changing meaning of “mother tongue” in questions asked in successive Indian government censuses, where it went from being defined as “the language spoken by the individual from the cradle”, to “parent tongue” to “language ordinarily spoken in the household” to “language ordinarily used”. This strongly suggests that “mother tongue” was not a term well adapted for use in the Indian language situation in the first place. Umberto Ansaldo offers reasons for thinking that “we might want to start considering notions such as native Speaker, mother tongue as well as ‘normal’ transmission as peculiar products of monolingual acquisition under normative Pressure typical of the modern era", and maintains that these “are rather exotic communicative ecologies in the history of human language evolution, and the lessons derived from their study, albeit significant, could well end up being potentially exceptional, maybe even peripheral to the construction of general theories of language” That may be putting it mildly. For the fact is that this exotic communicative ecology was only ever a normative ideal even in those societies that tacitly upheld or uphold it (see Love, 2009). Singh (this issue) quotes Chomsky (2000, p, 43), as admit- * Corresponding author, E-mail address: [email protected] (N, Love) "This suggestion is made by Davies (2003, p. i). * For recent discussion, see eg. Paikedlay, 1985; Rampton, 1990; Cook, 1999: Ricento, 2000; Davies, 2003, (0388-0001/5 - see front matter © 2010 Published by Elsevier Led. 0i:10.1016)i.'angsei2010,09,003590 1 Love, U, Ansldo Language Sciences 32 (2010) 589-593 ting that, in the real world, monolingualism (i.e, the consistent and exclusive use of one determinate linguistic system) is almost unimaginable” ~ presumably even in those nation states of post-Kenaissance western Lurope that have done theit best to impose it. Nonetheless, the monolingual native speaker of a mother tongue has played a key role in the linguistic theories that have flourished in such societies. As Davies (2003) quotes C. A. Ferguson as putting it: “Linguists have long gi- ven a special place to the native speaker as the only true and reliable source of language dats Why should that be? The first question is what is meant here by “language data”. Language data are utterances taken as instantiating abstract linguistic units belonging to an identifiable language (variety, lect) that have been put together according to that language's rules for grammatical (phonological, morphological, syntactic, semantic) well-formedness, So [Sok"tsactondomaet) iS an in- stance of the Standard English (Standard British English, Standard Southern British English...) sentence the cat sat on the ‘mat and conforms to the grammatical rutes of that language. [akztsitondomeet} although there may well be (deemed to be) a language to which the underlying sentence belongs, does not exemplify a sentence of the language in question and thus not a datum relevant to establishing its grammatical rules. While [Ioktsetasisyrlomat] is not referable to any recognised language at all. Note that what is or is not “true or reliable” language data has nothing to do with actual or potential com- municative efficacy. Is this not just a laborious, fancy way of stating what could readily be inferred from any traditional grammar of English anda dictionary? What is wrong with *the cat sit on the mat is that it has what is presumably to be taken as a3 sg form of the verb sit that is not sanctioned by the grammar book, while “le cat s'est assis sur le mat contains words absent from any English dictionary. Yes, it is. That this should be so is a matter of the history and politics of linguistics as an academic discipline.’ Western Linguistics is heir to a two-millennia-old tradition of grammatical study. It inherited the whole complex panoply of grammatical concepts and categories that had been developed in that tradition. For Saussure (1922 [1916], p. 153) it was a matter of com- plaine that “la linguistique travaille sans cesse sur des concepts forgés par les grammairiens”. Chomsky, in contrast, considers (1965, pp. 63-64) “what a traditional grammar has to say about a simple English sentence”, and finds it “without question, sub- stantially correct and ... essential to any account of how the language is used or acquired’, Either way, whatever synchronic descriptive linguistics was to take as its subject matter, or however it was to differ from traditional grammatical inquiry, it was committed to using the descriptive and analytic tools inherited from the grammatical tradition. Because they were the only tools linguists had, and if they were simply abandoned at the outset it was hard to see how synchronic descriptive linguistics could proceed. ‘At the same time, modern linguistics aspired to be a science (Love 2009), But the grammarian's descriptive framework was devised for purposes that had nothing to do with science, They mostly had to do with pedagogy. If you want to learn to write Latin like Cicero, or as a Dutchman to read Swahili, then you need an analytic handle on how Cicero or the Swahili writer would or would not express himself. Providing that handle is the chief function of traditional grammar. But learning a language in this way is essentially a cultural accomplishment, and does not on the face of it open the way to any sort of sci- entific inquiry. If language is to be the subject matter of a science, a naturalistic perspective is called for. What is the objectively given phenomenon, where language is concerned? What are we confronted with, pretheoretically, as part of the world's furniture, that we can set about investigating scientifically? Human verbal behaviour, which is ultimately to be explained in terms of human biological attributes and endowments and human social and cultural structures, But elaborating grammars of lan- ‘guages does not in and of itself have any necessary or straightforward connection with anything anyone does. For one thing, there cannot be grammars before the advent of writing,“ and yet human beings had language long before the advent of writing The problem, therefore, was to find a way of making grammatical analysis part of a naturalistic account of verbal behaviour. ‘One important move was to attempt to draw a clear distinction between description and prescription in grammatical study, But this distinction has always been problematic: a linguistic prescription is not necessarily something other than a description, but may merely be something one does or refrains from doing with a description; while on the other hand to describe a language or any part of a language demands the imposition of limits on what falls within its scope, which ‘may be seen as an implicitly prescriptive procedure. As C.E. Bazeli pointed out, “[d]escribing a language, like mapping a coun- tty, is an affair of cutting and smoothing” (Bazell 1953, p. 93), and what is smoothed away is by implication downgraded, ‘marginalised, or simply rejected. ‘The insistence on description rather than prescription was reinforced by “internalising” languages and the grammars of languages; that is, locating them in speakers’ heads. This was first proposed by Saussure (1922 {1916)), and the proposal constitutes Saussure’s main claim to be the founding father of modern linguistics. Saussure took languages to be systems of cognitive structures, actualised in the form of associations between acoustic images and concepts inside the heads of indi vidual members of the language community. In fact Saussure’s langue is both a cognitive system in the individual and a social institution in the community, Specifically, the foundation for this lies in the Saussurean doctrine of the linguistic sign, which explains the sense in which the rules constraining the language-user come from within the language itself, and not from some external source, such as a grammarian. The linguistic sign, according to Saussure, unites a signifier and a signified. * On this topic see especially Haris (1981, eh 3). “ Whether there is any good reason to suppose that there might be mentally represented grammars independent of writing is a question hat will not be pursued here1. Love, U, Ansaldo/ Language Sciences 32 (2010) 589-593 591 But neither signifier nor signified are independently given, The one doesn't exist without the other. Not only that, but neither the signifier, the signified nor the association between them exist except in the context of the particular system of such asso- ciations that constitutes the language. They are created by the language, not external to it. Saussure thus elevated a feature of everyday thinking about languages to the status of a theoretical postulate. We have a perfectly good and harmless use for expressions like “We know our native language”, as contrasted, for instance, with other languages that we know only imperfectly or don’t know at all. Of course, like much everyday talk “We know our native lan- guage” is extremely vague and not to be taken too seriously, in the sense that most people would find it hard to say exactly what propositions about languages and their relationships with speakers they thought such a statement committed them to. What Saussure did, in effect, was to show how you can take a particular interpretation of "we know our native language” and make it the cornerstone of a theory relating languages to speakers that would allow linguists to continue, as they had done for millennia, to carry on talking about grammar, while claiming to contribute to a scientific study of linguistic psychology. Although Saussure does not seem to have used (the French equivalent of) the term, the native speaker, so far as modern linguistic theory is concerned, starts life as the figment of Saussurean and post-Saussurean structuralism in whose head the internalised grammar is to be found. Specifically, he (or she) has come to be seen as the sole authoritative source of infor- mation about the language and its grammar. The main reason the native speaker is an authority is that he has naturalistically acquired (not learnt) the language by mere exposure to and interaction with other native speakers. This doesn’t mean that he is overtly acquainted with the con- cepts of grammatical theory, or able to tell you that such and such is ungrammatical because the noun needs to be in the dative, or because his is not a pro-drop language. What it means is that he has access to the ‘intuitions’ that lead him to accept or reject sentences of his language as grammatical or ungrammatical and thereby provide the linguist with his data base, A native English-speaker probably won't be able to say that English is a head-last language, but he will be able to say that “I have a dog big black is unacceptable as an English sentence. Of course, a competent non-native speaker of English may be able to report with equal accuracy that "I have a dog big black is unacceptable, but that is irrelevant because the fact that English doesn’t work that way is likely to have been explicitly taught (“In English adjectives come before their nouns”) to the non-native, If so, you know this as a fact about something external to yourself. It is on a par with knowing that Canberra is the capital of Australia. But the native speaker knows it without knowing that he knows it, by having unconsciously projected the rule from his linguistic experience. In his case the rule is a hypothesis that seeks to explain something about his verbal behaviour: the fact that he puts adiectives before their nouns and recognises deviations from that as unacceptable. Here is yet another idealisation. It is a moot point whether there are any human cultures in the real world in which small children are not to some extent, and from a very early age, overtly taught their language, and if that is so the distinction between acquiring a first language and learning a second one becomes blurred (cf. Singh's comment on “‘the somewhat dubi ous argument that so called Universal Grammar or UG is not available to second-language learners"). At any rate, the ideal subject for naturalistic inquiry would be a first-language speaker who had acquired his language without hearing or engag- ing in any analytic metalinguistic talk about it at all. But we see the ideal realised only in the case of apes, such as Kanzi, the famous bonobo who is said to have an ability in English equivalent to that of a 2.5 year old child, (See Savage-Rumbaugh et al, 1998.) But Kanzi, presumably, knows nothing about those abilities and cannot talk about them. In other word: although non-human apes can be taught language to some extent, it seems likely that they are incapable of engaging the kind of reflection on what they do linguistically that would allow them systematically to decontextualise linguistic signe and having done so, treat them as objects of contemplation and inquiry, thereby embarking on the process of constructing what human apes know as languages. So there is an irony here. The ideal, metalinguistically uncontaminated native speaker wouldn't be a speaker of a language at all, because metalinguistic contamination is the essential process by which language gives rise to languages, Another reason that native speakers are taken as authoritative is that a language is often seen as fundamentally ‘belong- ing to’ them, and some socially high-status subset of native speakers have the privilege of determining what the language ‘really’ is or should be - what the permissible words and constructions are, how they should properly be pronounced and spelt, etc. So the native speaker's pre-eminent role in linguistic theory emerges as an amalgam of the requirement of ‘natural acquisition and the idea that native speakers own and control their mother tongue. In this collection Christopher Hutton discusses language ownership with particular reference to legal property rights, ask- ing whether there are juridical principles in terms of which one might obtain a legal right thar would canstrain non-native speakers from learning and speaking a language. Hutton distinguishes a liberal-individualist view of language ownership, according to which “language is the non-private property of non-individuals, not the collective property of a defined group” from the "the mother tongue~native speaker tradition”, in which "a language is the collective property of its native speakers, understood collectively as a Volk or ethnos (‘people’), The Volk is defined as a historically continuous descent group, which owns a distinct language and has a defined territory. Its language and culture are part of its collective property. Rather than an open, unstructured space, the language is a highly structured systematic object. It is stamped with the mark of its creator and owner, the community of native speakers". The idea that the native speaker functions to save for naturalistic science the analytic study of grammatical structure is borne out by Rajendra Singh’s contribution. Singh is exercised by the need to maintain a focus on “linguistic form and archi- tecture” in the face of multilingualism as the “unmarked case". As he puts it, this requires “dispensing with the idea that a592 Ni. Love, U. Ansoldo/ Language Sciences 32 (2010) 589-593 theory of linguistic form is a theory of the internalised linguistic knowledge of a native speaker in a completely homogenous speech community”. ‘but what else might it be? The challenge, says Singh, Is to find a theory of lingulstic form that embraces the reality of multilingualism. For example, vowel epenthesis “appears everywhere when a language has to deal with unacceptable con- sonantal clusters .. . The appearance of the epenthetic vowel to adapt loan words or to break up unpronounceable consonan- tal clusters shows that epenthesis is not a rule of Punjabi, Hindi, English, or Vulgar Latin but that itis available to all humans. To talk about it as a rule of this or that language is to impose a monolingual view on the multilingual potential of human speakers.” As for morphology, “the morphological system of a speaker who knows a few thousand words of Hindi, Punjabi, and English may in parts look like the morphological systems used in monolingual Hindi, Punjabi, and English communities, but itis her morphological system, presumably shared with other speakers, and it is built with the same bricks as any other morphological system and, of course, has the same architecture, albeit a minimal one, as any other morphological system. There is no need here to fix her morphological competence in one of these three languages and to talk about her having bor- rowed this or that bit of morphology from the other two languages” ‘The question here is whether Singh envisages that the formal architecture will differ depending on the degree or kind of multilingualism in different cases, and on the particular languages involved. If so, the demise of the monolingual native speaker brings in its train the demise of the idea of a language as a definite cognitive structure owned (‘cognised’) by anyone at all, even an idealised anyone. ‘Alternatively, Singh's challenge is simply a version of the quest for linguistic universals. In which case he may be intrigued by Jan Wawrzyniak’s paper in this collection, a thorough discussion of Anna Wierzbicka's claim to have identified a universal mother tongue, or at least a set of universal meanings attached to a variety of different forms. According to Wawrzyniak, Wierzbicka’s position is that “all healthy human beings are from conception equipped with one language, identical within each individual”. The universal mother tongue is manifest in what Wierzbicka calls “natural semantic metalanguages”, which are subparts of full (tacitly known) natural languages. Thus an NSM contains linguistic signs the forms of which are different from the forms of linguistic signs of other NSMs, but the meanings are identical across all the NSMs, thereby being identical across all languages, Singh's worry is that a focus on multilingualism will lead toa shift of linguists’ attention fram language structure to, in the sociology and psychology of language use. And, indeed, three papers in this collection address the concept of a speak- er’s identity’ ‘Ansaldo argues that “the evolution of Sri Lanka Malay .... can be explained as a process of interacting and of negotiating linguistic identities in a new environment, as can be seen in the restructured vernacular in which substantial features of Lan- kan grammar combine with a predominantly Malay-derived lexicon”. The linguistic identity of Sri Lankan Malays is defined by multilingualism: their identity is shaped by the plurality of linguistic codes to which they have access. ‘The outcome of identity realignment in this situation is the creation of a new language by recombination of selected existing features of languages into a new grammar. Thus a new linguistic code is created that captures the new cultural profile that emerges through identity alignment. So “language creation is the result of a hybridization process within which identity alignment in a community is fundamental. Identity alignment involves innovative construction of a new grammar based on multilingual resources", “Identity alignment ... suggests that language creation is not the result of some excep- tional situation in which speakers were denied the social, cultural or cognitive settings necessary for a ‘normal’ acquisition of a target; rather, it is a creative process in which speakers have agency in the transmission and transfer of linguistic features". On the other hand, Pablé et al. claim that ‘identity’ is not amenable to scientific description, as sign-making and sign- interpreting are ‘private’ and cannot be detached from how an individual integrates the actual situation with his or her ‘own personal experience, and that it doesn't follow from the fact that we have a word identity in our language that it must be possible to detect and to conceptually isolate such identities, “When one uses the word ‘identity’ one is not referring toa set of fixed meanings proposed by dictionaries, but rather one integrates (in often unpredictable ways) previous personal experiences with that word, as required by the present situation”. “It is precisely because of the impossibility of assigning identity fixed definitive boundaries that identity cannot be accounted for scientifically.” And Alex Kravchenko's paper might be seen as endorsing from personal experience the point that where ‘identity’ is concerned there may be precious few valid generalisations to be had. ‘The native speaker rose to his privileged position in modern linguistics in response to a scientisation of the Western ‘grammatical tradition whereby languages, once no more than a distillation of the linguistic practices of a literary elite, were Feconceptualised as cognitive structures existing inside the head of an idealised monolingual speaker. This scientisation has always been problematic in many ways, some of them illustrated in the papers in this collection. But the fundamental prob- Jem is the primary focus on languages themselves. Languages are not natural objects, and they do not become natural objects by theoretical fiat. There has never been agreement as to what a language is or even how we might set about deciding what a language is. What counts as a language is a matter of cultural and political context and purposes, and these can vary without limit, As every undergraduate introduction to linguistics points out, there is no justification for distinguishing, say, Swedish and Danish as separate languages while treating Mandarin and Cantonese as variants of the same language. There is no con- sensus as to where one language ends and another begins, either in time or in space or in terms of social differentiation. We tolerate competing discourses in terms of one of which Latin and Spanish are two different languages but in terms of another, chronologically separated versions of one language. A language, in short, is a second-order cultural construct, and as such itLove, U Ansold/ Language Sciences 32 (2010) 589-593 503 can be constructed in as many different ways as the constructors - including linguists - please. Languages, whether “exter- nal" or theoretically projected as “internal”, are products of verbal belavivus, Hut prerequisites for it, and perliaps dhe maint service to linguistics performed by this collection is its highlighting of the need to find an alternative to LR. Firth's program- matic statement (Firth, 1968 (1957]) that “the business of linguistics is to describe languages", Acknowledgements The articles in this special issue of Language Sciences arose out a conference on this topic held in Cape Town, December 11-13, 2008 (www.cariad.co.zajconference). The organisers are grateful to the University of Cape Town for financial assistance, References Bazell, CE, 1953. Linguistic Form. Istanbul Press, Istanbul Bloomficl, L, 1935 [1933]. Language. Allen & Unwin, London. Chomsky, N, 1965. Aspects of the Theory of Syntax. MIT Press, Cambridge, MA. ‘Chomsky, N, 2000, The Architecture of Language, Oxford University Press, New Delhi ‘Cook, V, 1999. Going beyond the native speaker in language teaching, TESOL Quartriy 33 (2), 185-209, Davies, A, 2003, The Native Speaker: Myth and Reality. Multilingual Matters, Clevedon, Firth, JR, 1968 (1957). A synopsis of linguistic theory 1930-1955. In; FR, Palmer (Ed), Selected Papers of JR. Firth. Longman, London, Harris, 1981, The Language Myth. Duckworth, London, Love, N, 2009, Science, language and linguistic culture. Language and Communication 29, 26-46 Paikeday. T.. 1985. The Native Speaker is Dead! Paikeday, Toronto. Pattanayak, D.?.. 1981. Multilingualism and Mother Tongue Education Oxford! Univesity Press, Delhi Rampton, M..1, 1980. Displaing the ‘native speaker’ expertise, affiliation, and inheritance, ELT Journal 44 (2) 97-101 Ricento, T, 2000, Historical and theoretical perspectives in language policy and planning. Journal of Sociolinguistics 4, 196-213 Saussure, F, de, 1922 [1916]. Cours de linguistique générale. Payot Pai. Savage-Rumbaugh,S, Shanker, 8G, Taylor, TJ. 1998. Apes, Language, and the Human Mind, Oxford Ur sity Press, Oxford