How well does the linguistic theory of Chomsky explain language acquisition?

In contrast to Skinner’s (1957) theory of language acquisition, which held that children are ‘blank slates’ with regard to language, and which explained language acquisition in terms of reinforcement and conditioning, Chomsky (1957) argued that language acquisition is subserved by an innate, dedicated ‘language faculty’, which he termed the Language Acquisition Device (LAD). Chomsky (1957) held that the language to which children are exposed is ‘degenerate’ and full of stops and starts. This is known as the ‘poverty of the stimulus’ hypothesis, which infers that children develop language despite the poor language (or ‘primary linguistic data’) to which they are exposed. Therefore, according to Chomsky (1957), only the existence of an innate device for language learning could account for the speed and relative ease with which children learn the language around them, from a very limited and grammatically incorrect sample of language and without formal teaching. Chomsky (1965) also proposed that children are guided in their acquisition of language by an in-built knowledge of ‘language universals’, i.e. knowledge of properties of language that are common to all languages. He subsequently added to his linguistic theory the concepts of Universal Grammar (UG) (Chomsky, 1986) – which is a more structured version of ‘language universals’ – and Principles and Parameters Theory (Chomsky, 1981), which states that the properties (parameters) of language vary from language to language and therefore language acquisition is a process of setting the right parameters based on experience. For example, some languages (English and French) require a subject, whereas others (Spanish and Italian) do not. Therefore, the null subject parameter is set to ‘off’ in English and ‘on ‘ in Spanish (Chomsky, 1981).

In order to evaluate how well Chomsky’s linguistic theory explains language acquisition, the essay will begin by examining some of the evidence that supports Chomsky’s account of language acquisition. Specifically, it will examine evidence suggesting that children can extract rules about language from the language they hear around them. It will then explore research suggesting that children can develop language in the absence linguistic input and in the absence of feedback about grammar. Then it will look at evidence for language universals. The essay will then explore some of the evidence that contradicts Chomsky’s theory of language acquisition. First, it will examine research suggesting that the properties of language may be less universal than Chomsky suggests, and that children therefore may not develop language on the basis of UG and parameters. It will then examine evidence suggesting that children’s linguistic competence is less advanced than some research suggests. Finally it will explore research findings concerning the ‘primary linguistic data’ to which children are exposed and look at evidence suggesting that children receive more feedback about grammar than Chomsky proposes.

Unlike Skinner’s (1957) theory of language acquisition, which holds that children learn language through reinforcement and conditioning – and which therefore does not account for the phenomenon of overgeneralisations (e.g. foot-foots) – Chomsky’s theory of language acquisition holds that children’s language is creative and that they can make novel utterances because they possess an awareness of the rules of language, e.g. the frequently quoted “allgone sticky” (Braine, 1971), which suggests that children do not simply imitate the speech of adults. Berko (1958), for example, found that children could add appropriate noun endings to a non-existent word, and thereby to a word they had heard before (e.g., wug → wugs). Furthermore, Brown and Fraser (1963) proposed that children’s telegraphic speech could be regarded as grammatical in that it was correctly formed, omitting only a verb or inflection (e.g., chair broken) and McNeill (1966) proposed that children’s telegraphic utterances reflected their innate knowledge of grammatical relations. For example, it was said that a child who said “drink milk” showed knowledge of verb-object relationships. Children’s two-word combinations have been shown to be highly similar across cultures, and 95 per cent of the time they show correct word order (Brown, 1973; Pinker, 1984).

In support of Chomsky’s theory, it has also been suggested that in addition to being exposed to poor primary linguistic data, a child’s poorly formed sentences are rarely corrected by adults, which infers that children develop language despite the absence of feedback as to which strings of words are not properly formed. This phenomenon is known as the ‘no negative evidence problem’ or the ‘logical problem of language acquisition’. For example, using transcripts of naturalistic dialogue between three parents and their children, Brown and Hanlon (1970) found that the parents’ responses to their child’s incorrect grammar (whether utterances were well-formed or not) was contingent upon the actual meaning of what was being said rather than the correct or incorrect grammatical structure. Children may therefore have some in-built knowledge about language, which provides them with information about what utterances are correct or incorrect.

This is a preview of the whole essay

Furthermore, Chomksy’s position is supported by evidence showing that children acquire language – and even invent it – if exposure to language is minimal or absent (e.g. groups of deaf children who develop gestural systems to communicate with one another. Goldin-Meadow and Feldman (1977) studied such a group of congenitally deaf children aged around three and found that the gestural system they developed shared many similarities with spoken language. Firstly, gestures were used in combinations that seemed to possess a grammatical structure, and secondly, the number of gestures the children were able to use seemed to increase at a similar rate as words in spoken language in hearing children. Similarly, studies of creolisation, whereby a pidgin language is transformed into a Creole by virtue of being learned by children as a first language, strongly suggest that children impose universal properties on the language they learn. Pidgin is a rudimentary language with very little grammar and is developed by different language communities living together in order that they can communicate. While researching two groups of emigrants to Hawaii around the beginning of the 20th century, the first comprising of people who emigrated during adulthood and learned the Hawaiian pidgin (HPE), and the second group comprising of those who had emigrated as children and learned HPE while growing up, Bickerton (1981) observed that the children’s language – now known as Hawaiian Creole English (HCE) – possessed a grammar and was fluid, whereas the adult’s language remained inconsistent. Similar findings have emerged from studies into Nicaraguan Sign Language (NSL), which was also created initially as a pidgin language (Pinker 1994). When comparing the signing of children who started learning NSL before 1983 and children who started learning NSL from 1983 onwards, Senghas (1995) found that the language had become more structurally complex over time (i.e. it had evolved systems of verbal inflections and complex syntax) and the differences in structural complexity were observed primarily in the signing of children who started learning language at an early age. Therefore, it would appear, as Chomsky suggests, that language may possess certain universal properties. Having examined the evidence that children can acquire language in the absence of linguistic input, the essay will now explore the notion that language acquisition may be guided by innate knowledge of the properties of language.

All of the world’s languages distinguish between nouns and verbs (Chomsky, 1986), and this distinction is so fundamental that it can even arise in the absence of linguistic input. For example, David, a child with no exposure to sign language, invented his own system of gestures in which one type of gesture was used for nouns and another used for verbs (Goldin-Meadow, 1993). Similarly, certain formal universals seem to exist in many of the world’s languages. In a study of word order in 30 different languages, Greenberg (1963) found that although the order of subject, object and verb varies across languages, not all possible orders are found. Specifically, in 44 per cent of languages the dominant word order was subject-object-verb (SOV), whereas no language used the order object-subject-verb (OSV). Furthermore, the fact that English-speaking children initially omit subjects from their speech (e.g. eat biscuit) (Brown, 1973) has been interpreted as evidence that children at this stage have yet to set the null-subject parameter (i.e. they think they are learning a null-subject language like Spanish) (Hyams, 1994). Evidence for linguistic universals may therefore indicate that children are born equipped with an in-built awareness of the properties of language. However, the existence of linguistic universals does not necessarily constitute evidence of a distinct language-learning facility. It is equally possible that linguistic universals could arise from the universal properties of cognition, such as memory, attention and perception (Bates, 1997), and the distinction between nouns and verbs could reflect the cognitive distinction between objects and relations. Language universals could also be a by-product of more general constraints on language processing, e.g. some word orders may be easier to process than others (Sampson, 1989). Having examined the evidence in support of Chomsky’s theory of language acquisition, the essay will now explore evidence that contradicts this theory.

Various researchers (e.g. Sampson, 1989; Putnam, 1975; Braine, 1994; Bates, 1997) have questioned the concept of linguistic universals and the notion that any such linguistic universals could be indicative of an underlying language-specific device. Slobin (1997) argues that there is too much variability across languages and too much historical change to posit an innate language acquisition device. For example, Bates (1997) points out that Chinese has not grammatical inflections, whereas Eskimos are confronted with a language in which “an entire sentence consists of a single word with more than a dozen inflections” (1997. p16). Similarly, Slobin (1982) and Akiyama (1984) argue that different types of languages require different processes of language acquisition. Slobin (1982), for example, found that children learning purely inflectional languages (Turkish) were able to identify the subject and object of a sentence (e.g. “the squirrel catches the dog”) on hearing one inflected noun without hearing the rest of the sentence, whereas children learning word-order languages (English and Italian) could not. Furthermore, Turkish-speaking children were able to understand agent-patient sentences such as the above at age 2, compared to English-speaking, Italian-speaking and Serbo-Croatian speaking children, who understood these sentences at age 3. These findings suggest that different languages pose different problems for the child, and they raise questions about the existence of linguistic universals. If language acquisition really were a process of setting parameters and languages really were as similar as Chomsky suggests, surely children would cope with similar sentences at around the same time.

In addition, Tomasello (2000) argues that young children’s linguistic competence has been grossly overestimated. He points out, for instance, that although children do show evidence of some syntactical knowledge (e.g. the ability to substitute nominals for each other – “allgone juice” -> “allgone paper”), their creativity is rather limited. For example, he argues that young children do not use a verb outwith the sentence frames in which they have heard it used. Lieven, Pine and Baldwin (1997), for example, studied 12 English-speaking children aged 1-3 and found that almost all used all of their verbs and predicative terms in only one construction type, i.e. verb-argument configuration (e.g. mommy break and daddy break constitute one construction type, whereas break cup mommy break cup and break with stick constitute 3 different construction types. In addition, various studies (Bowerman, 1976; MacWhinney, 1978; Braine, 1976; Tomasello, 1995; 1999 and 2000) suggest that many young children’s constructions are item-specific rather than systematic. Specifically, a child may use lexical items like ‘up’ in many different combinations but use similar items like ‘down’ or ‘on’ as single-word utterances (Tomasello, 2000). Therefore young children’s linguistic ability may be item-based rather than systematic, which could infer that young children do not possess abstract linguistic categories. This would therefore cast doubt on Chomsky’s assumptions about UG. Proponents of UG, however, would interpret such findings in terms of limitations on performance or in terms of the maturational hypothesis (Chomsky, 1986).

Furthermore, some research findings contradict Chomsky’s assumptions about the impoverished language-learning environment. Newport, Gleitman and Gleitman (1977) estimate that 99.93% of the speech that children hear is fluent and grammatically well-formed, and Labov (1969) argues that this is true of adult speech in general. Furthermore, the ‘motherese’ hypothesis states that adult speech to young children is generally slower, shorter, higher in pitch, exaggerated in intonation and grammatically well-formed (Snow & Ferguson, 1977). This suggests that the primary linguistic data to which children are exposed, especially the speech of their caregivers, may be less impoverished than Chomsky suggests. However, it is important to note that the use of ‘motherese’ (or ‘child-directed speech’), is culturally determined, though almost universal. The Samoan culture, for example, rejects the notion of child-centredness and parents within this culture do not alter their speech when talking to children (Ochs, 1985).

The assumption that children do not receive overt feedback about their grammar (the ‘no negative evidence problem’) has also been questioned. Hirsch-Pasek, Treiman and Schneidermann (1984) attempted to verify Brown and Hanlon’s (1970) findings that parents do not correct their children’s grammar, but found that although explicit parental approval and disapproval were unrelated to well-formedness, the “language learning environment does present subtle cues that distinguish between well-formed and ill-formed sentences” (1984 p81). Mothers of two-year-olds, for instance, repeated ill-formed sentences almost twice as often as they did well-formed sentences. Furthermore, Demetras, Post and Snow (1986) found that “implicit feedback” is provided differentially for well-formed and ill-formed utterances to the extent that “children could rely on it as a basis for adjusting or confirming their rules” (1986 p.286). A further study by Bohannon and Stanowicz (1988) reported that although only 34 per cent of phonological errors receive explicit feedback (in the form of recastings, expansions, non-repetitious clarification questions), 70 per cent of ill-formed utterances are re-casted or expanded by adults. Therefore, there may be reason to believe, as Bohannon and Stanowicz conclude, that “theories of language that use the no-negative-evidence issue as their raison d’être may be on shaky ground” (1988 p.688).

In conclusion, the evidence examined suggests, as Chomsky’s linguistic theory proposes, that there may be an innate component to language acquisition. In particular, evidence that children can develop language in the absence of linguistic input supports Chomsky’s contention that language is innate. In addition, Chomsky’s linguistic theory can account for the phenomenon of overgeneralisation in children and for the production of novel utterances. It can also explain the fact that children seem to show an awareness of syntactical rules. Furthermore, a vast number of languages show similar properties. However, the existence of linguistic universals need not be explained in terms of the LAD or UG. Linguistic universals could arise a consequence of the more general properties of cognition. Furthermore, Chomsky’s account of language acquisition focuses on the development of syntax to the exclusion of other aspects of language acquisition, particularly semantics and pragmatics and the social context in which language is learnt, which has been shown to be an extremely important aspect of language acquisition (e.g. Bruner, 1977). Furthermore, research findings contradict Chomsky’s contention that the language-learning environment is poor (Newport, Gleitman and Gleitman (1977) and that children receive little feedback about how well constructed their utterances are (Hirsch-Pasek, Treiman and Schneidermann, 1984). In conclusion, Chomsky’s account of language acquisition may be useful as a hypothetical construct, i.e. it may explain the features of language that need to be acquired. However, it does not account for all the factors that are thought to be involved in language acquisition (e.g. semantics, pragmatics and the social context of language acquisition).

References

Akiyama, M.M. (1984) Are language acquisition strategies universal? Developmental psychology, 20. 219-228

Bates, E (1997) On language savants and the structure of the mind: a review of Neil Smith and Ianthi-Maria Tsimpli, “The mind of a savant: language learning and modularity”. International journal of bilingualism, 1(2). 163-179

Berko, J. (1958) The Child’s learning of English Morphology. Word, 14 150-177

Bickerton, D (1981) Roots of Language Ann Arbor, Michigan: Karoma

Bohannon, J.N. and Stanowicz, L. (1988) The issue of negative evidence: adult responses to children’s language errors. Developmental psychology, 24. 684-89

Bowerman, M (1976) Semantic factors in the acquisition of rules for word use and sentence construction, in D Morehead and A Morehead, Normal and Deficient Child Language, Baltimore: University Park Press

Braine, M.D.S. (1971) The acquisition of language in infant and child in C.E. Reed (Ed)

The Learning of Language New York: Appleton-Century-Crofts

Braine, M.D.S. (1976) Children’s first word combinations. Monographs of the society for research in child development, 41

Braine, M.D.S. (1994) Is nativism sufficient. Journal of child language, 21. 9-31

Brown, R (1957) Linguistic determinism and parts of speech Journal of Abnormal and Social Psychology 55, p1-5

Brown, R (1973) A first language: The early stages. Cambridge, Mass.: Harvard University Press

Brown, R and Fraser, C (1963) The acquisition of syntax in CN Cofer and B Musgrave (eds)

Verbal Behavior and Learning: Problems and Processes p158-201 New York: McGraw Hill

Brown, R and Hanlon, C (1970) Derivational complexity and order of acquisition in child speech in J.R Hayes (Ed) Cognition and The Development of Language John Wiley & Sons, New York P11-53

Bruner, JS (1977) Early social interaction and language development in HR Schaffer (eds) Studies in Mother-Child Interaction London: Academic Press

Chomsky, N. (1957) Syntactic structures. The Hague: Mouton

Chomsky, N (1981) Lectures on Government and Binding. Dordrecht, Netherlands: Foris publications

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

Chomsky, N (2000) New Horizons in the Study of Language and Mind Massachussets: Cambridge University Press

Cowie, F (1999) What’s Within? Nativism Reconsidered. New York: Oxford University Press

Demetras, M.J. Post, K.N. and Snow, C.E. (1986) Feedback to first language learners: the role of repetitions and clarification questions. Journal of child language, 13. 275-92

Goldin-Meadow, S and Feldman, H (1977) The development of language-like communication without a language model. Science 197 p401-403

Goldin-Meadow, S (1993) When does gesture become language? A study of gesture used as a primary communication system by deaf children of hearing parents. In Gibson, K.R. and Ingold, T. (Eds) Tools, language and cognition in human evolution. pp. 63-85. New York: Cambridge University Press

Greenberg, J.H. (Ed) (1963) Universals of Language. Cambridge, MA: The MIT Press

Hirsch-Pasek, Treiman, Schneiderman (1984) Brown and Hanlon revisited: Mothers’ sensitivity to ungrammatical forms. Journal of child language, 11, 81-88

Hyams, (1994) Commentary: Null subjects in child language and the implications of cross-linguistic variation, in Lust, B and Hermon, G (Eds). Syntactic theory and first language acquisition: Cross-linguistic perspectives, Vol. 2: Binding, dependencies, and learnability. (pp. 287-299). Hillsdale, NJ: Lawrence Erlbaum Associates

Labov, W (1969) The logic of non-standard English. Georgetown monographs on language and linguistics, 22. 1-31

Lieven, E., Pine, J., Baldwin, G. (1997) Lexically-based learning and early grammatical development. Journal of Child Language, 24. 187-220

MacWhinney, B (1978) The acquisition of morphology. Monographs for the society for research in child development, 41

McNeill, D (1966) Developmental Psycholinguistics in F. Smith & G.A. Miller (Eds)

The Genesis of Language: A Psycholinguistic Approach Cambridge, Mass.: MIT Press

Newport, E.L, Gleitman, H, and Gleitman, L.R. (1977) “Mother I’d rather do it myself”: Some effects and non-effects of maternal speech style in C. Snow & CA Ferguson (eds) Talking to Children: Language Input and Acquisition Cambridge: Cambridge University Press

Ochs, E (1985) Variation and error: A sociolinguistic approach to language acquisition in Samoa, in Slobin, D.I. (Ed). The crosslinguistic study of language acquisition, Vol. 1: The data; Vol. 2: Theoretical issues. (pp. 783-838). Hillsdale, NJ: Lawrence Erlbaum Associates

Pinker, S (1984) Language Learnability and Language Development. Cambridge, MA: Harvard University Press

Pinker, S (1994) The Language Instinct London: Penguin Books

Putnam, H (1975) “The meaning of meaning”, Mind, Language and Reality: Philosophical Paper, 2. pp 33-69

Sampson, G (1989) Language Acquisition: Growth or Learning? Philosophical Paper,18, pp203-40

Skinner, B.F. (1957) Verbal Behavior. New York: Appleton Century Crofts

Slobin, D (1979) Psycholinguistics Glenview: Scott, Foresman

Slobin, D (1982) Universal and particular in the acquisition of language. In E. Wanner and L.R. Gleitman (Eds) Language Acquisition: State of the Art. Pp128-170. Cambridge, MA: Cambridge University Press

Slobin, D (Ed.) (1997) The crosslinguistic study of language acquisition. Mahwah, NJ: Lawrence Erlbaum Associates

Snow, C. E. (1977) Mother’s speech research: from input to interaction in C Snow and C Ferguson (eds) Talking to Children: Language Input and Acquisition Cambridge: Cambridge University Press

Tomasello, M (1995) Language is not an instinct. Cognitive development, 10. 131-156

Tomasello, M (1999) The cultural origins of human cognition. Harvard university press

Tomasello, M (2000) Do young children have adult syntactic competence? Cognition 74, 209-253