An Essay on Different Theories of Language Acquisition

This essay is going to discuss what language is defined as and critically discuss different theories of language acquisition, which are the behaviourist approach, the biological approach and the Interactionalist approach. It will also discuss what criteria is needed for something to be considered a language for the evaluation purposes of animals studies, which attempt to teach animals language to determine whether or not language is a unique human ability.

According to Psycholinguist Roger Brown (1965) language can be defined as a set of arbitrary symbols. Which, taken together make it possible for a creature of limited powers of discrimination and a limited memory to transmit and understand an infinite variety of messages and to do this in spite of noise and distraction. (Brown, 1965 cited in Gross, 2009: 320)

Whilst other non-human species can communicate with each other to a certain degree of limitation it could be argued that this limitation and the fact that humans can create an infinite variety of messages that sets human communication apart from animal communication. Brown (1973) later devised a new definition of language in which he emphases that humans not only learn a repertoire of sentences but also acquire a rule system that allows the generation of an infinite variety of sentences. Psycholinguists refer to this rule system as grammar, which is much more than just parts of speech but rules that govern our language and patterns of speech (Jackendoff, 1993). The major components of grammar according to Gross (1996) are phonology, which is the knowledge of the sound system for language containing rules for combining phonemes (sounds), semantics, which is the knowledge of meaning that governs the process of combining morphemes (words). Lastly syntax, which is the knowledge of relationships between sound and meaning consisting of rules for combining morphemes to form meaningful and grammatically correct sentences.
But how does the language develop? Skinner (1957) attempted to apply the principles of operant conditioning in explaining how language developed. He accepted that vocalisations such as cooing and babbling were most likely inborn and argued that adults shape the babies sounds (Phonemes) into words (Morphemes) by reinforcing the sounds, which are closest to the actual pronunciation of that word. Through selective reinforcement these words are eventually shaped into sentences with reinforcement correct grammar and incorrect grammar being ignored. An example of positive reinforcement would be when a child asks for a glass of water this produces a drink that reinforces that form of words it can also come from parents when they become overly excited when their child is trying to vocalise. Skinner also believed in imitation, which is when children try to imitate verbal labels they receive immediate approval from the parents, which reinforces the act. As children begin to improve and learn new words through imitation their language begins to resemble that of adults (Moerk and Moerk, 1979).

Whilst imitation must be involved in the development of vocabulary its role in complex aspects such as syntax and semantics is not so clear. As stated earlier children imitate adult sentences but tend to convey them according to their current level of operating grammar. However a child is more likely to imitate correct grammar after an adult has recast the child’s sentence as opposed to when the same grammatical form occurs in normal conversation (Farrar, 1992; Nelson, 1977). Recasting is a rarity and sometimes non-existent in conversation between children and parents yet children still acquire a complex knowledge of grammar (Bee, 2000). Bandura (1977) expanded on the concept of imitation and argued that children imitate the general form of sentences and add various words to fill in these general forms known as deferred imitation, which is when language structures are stored in the child’s memory long before they are used. By storing these examples of adult language children have a delayed replay facility that allows them to produce different forms of language long after they have been acquired. Skinner’s view has been relatively criticised in particular by Noam Chomsky who was heavily critical of his theory and as a result developed his own language acquisition theory.

Chomsky (1959) disagreed with Skinner’s approach in several aspects and argued that his theory was based solely on the fact children learn from trial and error if they gained approval of a word they would repeat it if not then they would abandon it. Chomsky argued that children could acquire language much more quickly than such a device would suggest, within a short period of two years children already have a vast vocabulary and a set of complex grammatical rules. This would not be possible if children have to try every possible combination through trial and error as a result Chomsky proposed that every child has an innate language acquisition device (LAD).
This LAD is an inherited biological mechanism that allows a child to decode the language it hears around them revealing the basic rules and principles of language. But due to the fact that all human language is different this device would only work if all underlying principles of language were the same. Chomsky argued that human language only differs in what he termed ‘surface structure’, which is the individual rules of grammar and vocabulary, underneath these is a ‘deep structure’ in which the LAD of the child was aimed at decoding. The LAD would identify key distinctions and significant words and then use these to work out the complexities of language and how it should be used. Chomsky claimed that these deep language structures are associated with the fundamental meanings and actions of all languages and developed a grammatical system to identify the deep structure of utterances, which he called transformational grammar (TG).

Aitchison (1983) agrees with Chomsky in that children are wired with the knowledge that language is rule-governed but disagrees with the fact that the LAD also consists of TG. Aitchson emphasises a process approach in which children have an in-built puzzle-solving device that allows them to analyse linguistic data. In contrast Chomsky (1979) argued that an innate language ability exists independently from other innate abilities due to the fact that the mind is constructed of mental organs that are differentiated and specialised. According to Chapman (2000) despite evidence that language is acquired through stages and develops over a period of years, the belief in some type of LAD has still persisted. The fact that language in-put of young children is well-formed and well-adaptive to their current focus of attention and understanding, it suggests that language development should be understood from the context of their social interactions instead of focusing on what the child possesses i.e. a LAD.
As cited in Gross (2009) there has been a growing acceptance that neither the behaviourist approach nor the biological approach can explain language development completely. Instead an integrated view argues that language development is both biological and social in that children cannot acquire language until they have reached a certain level of biological maturity and that language development is more closely associated with environmental input and cognitive development than Chomsky proposed. This approach is known as the language and social interaction approach and offers an alternative explanation to Skinner and Chomsky.
Smith, Cowie & Blader’s (1998) language and social interaction approach views language as being used to communicate needs and intentions. Several studies have shown how babies experience a social world and later develop language as Snow (1977) noted that adults give meaning to sounds that babies make such as burps and giggles viewing them as expressions of intent as are non-verbal communication such as smiling and eye contact. Snow interpreted this as a form of primitive communication that is significantly biased as it requires more input from the adult as they implement meaning to the babies verbal communication and non-verbal behaviour, from this perspective the child is an inadequate conversation partner.

Collins and Schaffer (1975) suggested a two-way exchange process that they called visual co-ordination and formats, which refers to when the two individuals focus on some common object. Putting the infants exploration of the environment into a social context in which an infant-object scenario is change to an infant-object-mother scenario (Schaffer, 1989). Bruner (1975; 1978) elaborated on Collins and Schaffer’s formats and referred to them as rule bound activities in which the infant is provided with many opportunities to associate language with familiar play. Brunner (1983) believed these formats consisted of a language acquisition support system (LASS) concerning himself with the pragmatics and function of language. Bruner viewed that learning a language required the learning of grammar and realising one’s intentions with the appropriate use of grammar, emphasising the intent that requires a more active role from the adult in helping the child’s language acquisition as opposed to just being input for the child’s LAD. Moerk (1989) argued that the LAD of the child consisted of the mother or primary caregiver in that their role was to simplify language breaking it down into more manageable segments of information that can be developed and built upon. As Durkin (1995) interprates this language development as a sophisticated extension of meaningful social interactions that the child and caregiver have constructed allowing langauge acquisition.

To establish whether or not language is unique to humans a distinction between animal communication and actual animal language is required also a definition of what language consists of is needed to fairly evaluate animal studies that attempt to teach them language. Many non-human species can communicate to each other for example the honeybee can communicate vast distances and locations to other members of its species through the means of a ‘waggle dance’ (Von Frisch, 1974). Other species have the ability to communicate through vocalisation, bodily odours or bodily posture; although this is a system of communication it is not a language. As human language can be modified by experience and is adaptable whereas non-human communication is quite inflexible and at times can be inaccurate if there are changes to the environment. For example when the honeybee has to fly upwind to its hive it will supply inaccurate information of distances and locations to other members of its species (Moffet, 1990).

A definition of language proposed by Charles Hockett (1960) devised a set of criteria consisting of thirteen design features of language. From this list of thirteen Aitchson (1983) claimed that ten should be efficient for evaluating purposes. But with further analysis of human language and non-human animal language he proposed that four were unique to humans, which are semanticity the use of symbols to refer to objects and actions. Displacement, the ability to make reference to objects and events in another time or place, for example talking about someone we have met or are going to meet. Creativity also known as productivity, the capacity to combine symbols through a means of production and understanding of novel combinations to create language that has never been pronounced before. Finally structure dependence, the understanding of the nature of language and structured chunks such as word order that allows the individual to be creative.

So is it possible to teach animals language? As it is possible that they could have the necessary biological and cognitive capabilities to acquire language but have not needed to do so, but on the other hand it could be that humans are the only species that posses the necessary capabilities for language acquisition.
Early attempts to teach animals language by Herman, Richards and Wolz (1984) have involved trying to teach two-bottled nosed dolphins Phoenix and Akeakamai artificial language. One language was visually based in which the trainers used their arms and legs to form gestures and the second was acoustically based using computer-generated sounds. But this research only ever tested their comprehensive abilities of language and not their ability to produce it. From this research it is clear that to establish whether or not animals can learn language that it is important to examine both the animals comprehensive abilities and their ability to produce language.

Most studies involved in attempting to teach animals language have used other primates such as chimpanzees as they are highly intelligent and social creatures. There have been numerous attempts to teach chimpanzee’s human-like language but due to the fact that an essential part of human language is word meaning and syntax. It is vital that animals learn the meaning of words and the ability to combine them into an infinite number of sentences to conclude that they have in fact learnt a language.

Kellogg and Kellogg (1933) were the first to try and teach chimpanzee’s a language they raised a female chimpanzee named Gua alongside their son a type of cross-foster rearing, overall Gua was never able to vocalise any recognisable words and only ever appeared to understand a few. A similar attempt by Hayes (1951) in which he attempted to raise a chimpanzee named Viki as a human child and teach her to speak. Hayes was also unsuccessful as after six years Viki was only able to produce four poorly articulated words, which were ‘mama’, ‘papa’, ‘up’ and ‘cup’. But on a more successful note with additional training she was able to understand more words and more combinations of words. These early attempts to teach chimpanzees to speak and learn language have a fundamental limitation in that chimpanzees are physiologically unable to produce speech using their vocal tracts and was probably the main hindrance to progress. From these failures nothing can be concluded about primates abilities to learn language but due to the fact that chimpanzees have good dexterity later attempts to teach them language was based on systems of a type of sign language and using artificially created symbols.

But the most famous case of attempting to teach a chimpanzee language is that of Washoe who is a female chimpanzee that was capture in the wild when she was one year old. Washoe was then raised as a child in whom she did things such as toilet training, eating, playing and other social activities (Gardner and Gardner, 1969, 1975). She was taught American Sign Language (ASL) that is used by people with hearing impairment and as like spoken language it involves words and syntax. By the age of four Washoe was able to produce eighty-five signs and was able to comprehend even more, years later her vocabulary eventually increased to one hundred and fifty to two hundred signs (Fouts, Shapiro, & O’Neil, 1978). It was claimed that like young children she would make over generalisation errors for example the sign ‘Flower’ to represent ‘Flower-like smells’. It was also argued that when she did not know a sign that she had the ability to create a new one. For example the first time Washoe saw a duck she did not know a sign that represented it so she combined two signs she already knew calling it a ‘water bird’, she was able to correctly combine up to five signs long. She was also able to answer questions that started with WH-words, which are questions that start with what, where, when or who.

An area of interest whilst investigating whether or not chimpanzees can learn language is that if they successfully learn a language will they then go on to teach their offspring or will the offspring learn language through observing their parents. More prominently due to the fact that there is little evidence that human children learn language completely from their parents, it was noted that Washoe’s adopted son Loulis had spontaneously acquired signs from Washoe and was observed as being taught by Washoe. From this it is unclear whether or not language has been transferred or if it is a sophisticated communication system (Fouts, Fouts & Van, 1989; Fouts, Hirsh & Fouts, 1982).

Premack (1971) attempted a different approach in which he trained a chimpanzee named Sarah in a laboratory setting, manipulating small plastic symbols. The symbols could be ordered according to certain rules to create a language called premackese, the advantage of this is that it required much less memory load as the symbols were always in front of Sarah. Sarah could produce lexical concepts in which she could create novel strings of symbols this however was only at a low level of substituting one word for another. But she did however create syntactically complex sentences and also showed a level of metalinguistic awareness as she could talk about the language system itself. This research however has showed that there was little evidence to show that Sarah was actually forming syntactic units by grouping strings of symbols together.

Terrance, Petitto, Sanders & Bever (1979) who were far more pessimistic about the language capabilities of chimpanzees taught a chimpanzee named Nim Chimsky (As a pun of Noam Chompsky) a language based on ASL. Overall Nim learned one hundred and twenty five signs and made over twenty thousand utterances in two years, some of which consisted of two or more signs in combination. In two-word utterances it was noted that there seemed to be a regularity in that the second thing mention was usually a place and that with longer utterances this broke down. Nim’s longer utterances tended to be quite repetitious for example ‘banana me eat banana eat’ rather than conveying any real structural syntax. Nim was rarely spontaneous with his signing, as his utterances tended to be in reply to his trainers or to do with some type of immediate activity while the rest of his utterances were merely repetitions of signs made by his trainers. However O’sullivan and Yeager (1989) noted that the methods of training Nim could have limited his linguistic skills as they found that when Nim was put in a conversational setting instead of a formal training setting that he performed better.

These early attempts to teach chimpanzees language are limited in value, as they do not tell us whether or not chimpanzees are capable of learning language, if anything they tell us more about their limitations that hinder the learning of language. But strong claims have been made about the performance of a pygmy chimpanzee named Kanzi, as earlier research used the common chimpanzee other studies have suggested that bonobo and pygmy chimpanzees are more intelligent, sociable and are much more natural at communication. It has been claimed that Kanzi has made a vital step of acquiring the understanding that symbols are associated with things in the world. Unlike other chimpanzee studies that used formal training by reinforcement with food, provided the correct symbol was presented Kanzi observed his mothers training on the Yerkish system of lexigrams and as a result acquired an understanding of these symbols. With the understanding of these symbols Kanzi then started to interact with people in their daily routines and was exposed to English, his comprehension of yerkish and English was then compared to that of young children (Savage-Rumbaugh et al, 1993). Kanzi’s performance was just as good sometimes better than a two-year-old, by the age of forty-six months he could understand fifty symbols and could create up to eight hundred combinations. He could understand word order and the meaning of verbs, for example he knew the difference between ‘get the rock’ and ‘take the rock’, as up to 80% of his output was spontaneous as opposed to either being prompted or merely just imitations.

Seidenberg and Petitto (1987) questioned Kanzi’s semantic and syntactic abilities and argued that he did not have an understanding of names as humans do. For example Kanzi used the word ‘strawberry’ as a name as when he would request to eat strawberries he would request to go to where the strawberries grow. Furthermore, Kanzi’s acquisition of grammatical skills was much slower than humans and the complexity of his sentences was not equivalent to that of a three-year-old child. Savage-Rumbaugh (1987) argued that critics over estimate the linguistic abilities of children and under estimate the abilities of chimpanzees. But Kako (1999) argued that Kanzi showed no sign that he possessed word function or the ability to use morphology as he does not seem to modify his language to accommodate for number as we do to form plurals. But despite the arguments for and against Kanzi he is by far the best case for language like-abilities in chimpanzees. Although Kanzi is a pygmy chimpanzee and may possess far better linguistic skills than the common chimpanzee his success could have originated from the fact that he was exposed to language at such a young age (Deacon, 1997). Proposing the question that is early exposure to language as important to chimpanzee’s for language acquisition, as it appears to be in humans?

From these studies it is clear to see that chimpanzees have learnt something but it is not necessarily language although they have learnt to associate names with actions and objects there is more to language than just this. Word meaning for example raises many issues as to how we use names and how they are different from association. Due to the fact pigeons can be conditioned so that they respond differently to two different pictures (Herrnstein, Loveland & Cable, 1997) it’s easy to imagine how they could be taught to respond one way (one peck) to one word and another (two pecks) to a different word, suggesting that they are even naming the words. But this naming is very different to humans as humans also know the meaning to what they are naming, they know that a tree has leaves and roots and what a tree looks like. If the word tree was presented to the trained pigeon by observing it’s pecking behaviour the best it could do is indicate that the word ‘tree’ looks more like ‘tee’ than some other words. With this in mind the use of signs by chimpanzees could be equated more to pigeons than that of humans.

For psycholinguists to concur that primates use words like us then these primates would have to learn that names refer to objects that can be carried across contexts and that they have the same understanding of word meaning as we do. Despite the achievements of Kanzi there is no empirical research that can answer these questions as of yet. For example Nim could sign apple or banana if they were in front of him one at a time but when the fruits were placed together he could not respond correctly, suggesting that he did not understanding the meaning of the signs in the way that we do. In conclusion although chimpanzees have learnt the association of symbols to the world it is still debatable whether or not chimpanzees have learnt the meaning of symbols in the way that we know the meaning of words. Although research has shown that chimpanzees appear to demonstrate syntactic abilities by combining symbols in a rule-governed way to form sentences it has been argued that these sentences are nothing more than ‘frames’. In that it is nothing more than a sophisticated version of conditioning and does not show the creativity of word-ordering rules. It would be as if the previously mention pigeons had learnt the whole sentence instead of just an individual word, they would not be able to recognise that ‘the dog chased the cat’ is the same as ‘the cat was chased by the dog’. As humans we can have a finite number of grammatical rules and words but can combine them to form an infinite number of sentences (Chomsky, 1957).

The way in which we evaluate the performance of the chimpanzee language acquisition is with that of children but the problem with this is that it is unclear as to what age group of children the chimpanzees should be compared to. Even though the cognitive abilities of chimpanzees and young children are not so different their linguistic abilities are, suggesting that language processes are independent of cognitive process. Chomsky (1968) claims that human language is a special faculty independent of cognitive process containing an innate biological basis that has only evolved in humans as he argues that only humans possess an LAD that enable us to acquire language. Due to this at the very least it can be said that children acquire language whereas chimpanzees have to be taught it.

When further research depicts how linguistic chimpanzees raise their offspring a much clearer picture of language acquisition will be presented. But until then it can be concluded that it has not yet been proven whether or not chimpanzees can acquire language. But due to the biological implications of language acquisition in children it can be said that language is a unique human ability.

Bibliography

Aitchison, J. The Articulate Mammal (2nd Edition) London: Hutchinson, 1983.

Bandura, A. ‘Self-Efficacy: Toward a unifying Theory of Behaviour change’ Psychological Review, 84, 1977, pp.191-215.

Bee, H. The Developing Child (9th Edition), Boston: Allyn & Bacon, 2000.

Brown, R. A first Language: The Early Stages, Cambridge MA: Harvard University Press, 1973.

Brown, R. Social Psychology, New York: Free Press, 1965.

Bruner, J. S. ‘Acquiring the use of Language’ Canadian Journal of Psychology, 32, 1978, pp.204-218.

Brunner, J. S. Child’s Talk Using to Learn Langauge, Oxford: Oxford University Press, 1983.

Bruner, J. S. ‘ The Ontogenesis of Speech Acts’ Journal of Child Language, 2, 1975, pp.1-21.

Chapman, R. S. ‘Children’s Language Learning: An Interactionalist Perspective’ Journal of Child Psychology & Psychiatry, 41(1), 2000, pp.33-54.

Chompsky, N. Language and mind. New York: Harcourt Brace, 1968.

Chomsky, N. Language and Responsibility Sussex: Harvester Press, 1979.

Chomsky, N. ‘Review of Skinner’s Verbal Behaviour’ Language, 35, 1959, pp.26-58.

Chompsky, N. Syntactic structures. The Hague: Mouton, 1957.

Collins, G. M & Schaffer, H.R. ‘Synchronisation of Mother Infant-Pairs’ Journal of Child Psychology and Psychiatry, 16, 1975, pp.315-320.

Deacon, T. The symbolic Species. Harmondsworth: UK: Penguin books, 1997.

Durkin, K. Developmental Social Psychology, From Infancy to Old Age, Oxford: Blackwell, 1995.

Farrar, M. J. ‘Negative Evidence and Grammatical morpheme acquisition’ Developmental Psychology, 28, 1992, pp.90-98.

Fouts, R. S., Hirsch, A., & Fouts, D. H. ‘Cultural transmission of a human language in a chimpanzee mother/infant relationship’ In H. E. Fitzgerald, J. A. Mullins, & P. Gage (Eds.), Psychobiological perspectives: Child nurturance series, 3, pp.159-193. New York: Plenum Press, 1982.

Fouts, R. S. Shapiro, G. & O’Neil, C. ‘Studies on linguistic behaviour in apes and children’ In P. Siple (Ed.), Understanding language through sign language research, pp.163-185. London: Academic Press, 1978.

Fouts, R. S., Fouts, D. H. & Van Cantfort, T. ‘The infant Loulis learns signs from cross-fostered chimpanzees’ In R. A. Gardner, B. T. Gardner & T. Van Cantfort (Eds.), Teaching sign language to chimpanzees, pp. 280-292. Albany: State University of New York Press, 1989.

Gardner, R.A. & Gardner, B.T. ‘Evidence for sentence constituents in the early utterances of child chimpanzee’ Journal of Experimental Psychology: General, 104, 1975, pp.244-267.

Gardner, R.A. & Gardner, B.T. ‘Teaching sign language to a chimpanzee’ Science, 165, 1969, pp.664-674.

Gross, R. Psychology The Science Of Mind And Behaviour (5th edition), London: Hodder Education, 2009.

Hayes, C. The Ape in Our House, New York: Harper, 1951.

Herman, L.M. Richards, D. G. & Wolz, J.P. ‘Comprehension of sentences by bottled nosed dolphins’ Cognition, 16, 1984, pp.129-219.

Herrnstein, R. Loveland, D. & Cable, C. ‘Natural concept in pigeons’ Journal of Experimental Psychology: Animal learning and memory, 2, 1977, pp.285-302.

Hockett, C. D. ‘The Origins of Speech’ Scientific American, 203, 1960, pp.88-96.

Jackendoff, R. Patterns in the Mind: Language and Human Nature, Hemel Hempstead: Harvester-Wheatsheaf, 1993.

Kako, E. ‘Elements of syntax in the systems of three language-trained animals’ Animal learning and behaviour, 27, 1999, pp.1-14.

Kellogg, W. N. & Kellogg, L. A. The Ape and the Child, New York: McGraw-Hill, 1933.

Moerk, E.L ‘ The LAD was a Lady, and the Tasks were ill-defined’ Developmental Review, 9, 1989, pp.21-57.

Moerk, E. L. & Moerk, C. ‘Quotations, imitations and generalisations: Factual and methodological analyses’ International Journal of Behaviour Development, 2, 1979, pp.43-72.

Moffett, M. ‘Dance of the Electric Bee’ National Geographic, 177, 1990, pp.134-140.

Nelson, K. ‘Facilitating Children’s Syntax acquisition’ Development Psychology, 13, 1977, pp.101-107.

O’Sullivan, C. Yeager, C. P. Communicative context and Linguistic competence: The effects of social setting on a chimpanzee’s conversational skills. In R.A. Gardner, & T. E. van Cantford (Eds), Teaching sign language to chimpanzee’s (pp.269-279). Albany, NY: Suny Press.

Premack, D. ‘Language in chimpanzee?’ Science, 172, 1971, pp.808-822.

Savage-Rumbaugh, E. S. ‘Communication, symbolic communication and language: A reply to Seidenberg and Petitto’ Journal of Experimental Psychology: General, 116, 1987, pp.288-292.

Savage-Rumbaugh, E.S. Murphy, J. Sevcik, R.A. Brakke, K.E. Williams, S.L. & Rumbaugh, D.M ‘Language comprehension in ape and child’ Monographs of the Society for Research in Child Development, 58, 1993, pp. 3-4.

Schaffer, H. R. Early Social Development, Hove: Earlbaum, 1989.

Seidenberg, M. S. & Petitto, L. A. ‘Communication, symbolic communication and language: Comment on savage-Rumbaugh, Macdonald, Sevcik, Hopkis & Rubert (1986). Journal of Experimental Psychology: General, 116, 1987, pp.279-287.

Skinner, B. F. Verbal Behaviour New York: Appleton-Century- Crofts, 1957.

Smith, P.K. Cowie, H. & Blader, M. Understanding Children’s Development Oxford: Basil Blackwell, 1998.

Snow, C. E. Mothers Speech Research: From Input to Interaction New York: Cambridge University, 1977.

Terrance, H. S. Pettito, L.A. Sanders, R.J. Bever, T. G. ‘Can an ape create a sentence?’ Science, 206, 1979, pp. 891-902.

Von Frisch, K. ‘Decoding the Language of the Bee’ Science, 185, 1974, pp.663-668.