Theories of language acquisition should clearly be evaluated in terms of how adequately they predict and explain the behaviour that is observed. Despite the fact that a detailed description of language acquisition has not yet been obtained (Lucy, 1992), a lengthy and controversial discussion has taken place in the literature concerning theoretical explanations of children’s language acquisition and development. In this essay, only a brief summary of the three major models will be put forward, which will allow us to review some of the principal questions still to be empirically answered: why do children acquire language; how do they go about acquiring it; and, what is it that they acquire?
The systematic study of language acquisition began in the late 1950’s, around the time when cognitive science was in its developmental stage. Chomsky’s (1959) ‘Review of Skinner’s Verbal Behaviour’ ruthlessly attacked the prevailing consensus – the mind consists of sensorimotor abilities with a few general laws of learning which control gradual changes in the organisms behaviour. Therefore, language must be learned and since verbal behaviour is the only manifestation of thought that can be externally observed (Skinner, 1986), thinking must only be a form of internal verbal behaviour (Pinker, 2000). Chomsky, on the other hand claimed that children learn languages based on highly subtle and abstract principles, yet they acquire these principles without any explicit instruction. Hence, language acquisition depends on an innate, species-specific unit that is separate from general intelligence (Chomsky, 1957).
Besides these two opposing theses, there exists another model of language development: the evolutionary/biological model. The shape of the human vocal tract is clear evidence of evolution demanding speech. The larynx is situated low in the throat and the vocal tract has a sharp bend that creates resonant cavities (the mouth and the pharynx), both of which help produce vowel sounds. But these have come with a life threatening sacrifice in efficiency of breathing, swallowing and chewing (Lieberman, 1984 cited in Pinker, 2000). Choking has been a common cause of accidental death until the invention of the Heimlich manoeuvre (Pinker, 2000).
The evolutionary selective advantages for language must have been significantly large to prevail over such basic needs such as eating and swallowing efficiently. One is made to wonder that if language is truly derived from Darwinian natural selection, then finding it in our closest relative, i.e. the chimpanzee, must not be difficult. However, in several famous but controversial studies, chimpanzees have been taught to use basic hand-signs and to understand some spoken commands (Gardner & Gardner, 1969; Premack et al., 1983, 1991 cited in Pinker, 2000).
These have been successful, but only to a limited extent. This lack of homology does not cast any uncertainty on the Darwinian account of language evolution. Humans did not evolve from chimpanzees; they both had common ancestors perhaps 6-7 million years ago which leaves 300,000 generations in which language could have evolved in humans (Stein, 2003).
The aforementioned evolutionary model is a relatively novel concept. The following section will return to the more classical and empirically evidenced theories discussed earlier, beginning with the behaviourist view. The behaviourists argue that language development takes place even before birth and is part of the general learning process of a child. They only study behaviour that can be measured or observed, hence it is synonymous with the ’empiricist’ view. Skinner (1957) was the first to carry out this empirical study and claimed that the child basically learns through imitation of the linguistic environment around her.
According to Skinner’s (1957) operant conditioning principles, when care-takers reinforce a particular behaviour in a child, she learns which behaviour is correct and which is not. Moreover, Skinner boldly attempts to account for the acquisition of syntax. He claims that the structure of a sentence consists of a line of associations between the words in the sentence, so if a child knows the words “dog” and “run” and hears them used together, such as “the dog is running”, the child may imitate “dog run” and be rewarded by the care taker.
In an attempt to test the above ‘classical’ theories, new studies of heart deceleration (or reduction) have demonstrated that foetuses are capable of discrimination and some form of rudimentary cognition (Lecanuet et al., 1992). This particular study found that heart rate decelerated when the stimulus exposed to the foetus changed from the word ‘babi’ to ‘biba’. This evidence then suggests that the foetus is capable of discriminating between phonetic stimuli, as measured by heart rate. Another study by Kuhl et al. (1992) provides further evidence of the affect of children’s linguistic environment on their language development.
Native Swedish and American 6 month old infants, were made to hear vowels which were specific to only their native language to test if they noticed a difference when the language was not their native one. Their reactions were strikingly different. Swedish infants responded when the English vowel came along but not when the Swedish one was heard, and vice versa. In other words, the infants, even at the age of 6 months had learnt not to pay attention to slight changes in speech sounds of their own language, but they were still able to distinguish it in another language. Although the children were too young to understand the meaning of what they heard, the speech of people around them had affected the development of their perceptual mechanisms. These results support the native language recognition hypothesis put forward by Moon et al. (1993).