The theory which asserts that human beings are genetically pre-programmed to learn language has been popularized most effectively by the American linguist Noam Chomsky. The assertion is argued on several counts. We will discuss four of these.
The structure of human speech organs
Humans are most often compared with apes and, in fact, our vocal apparatus is extraordinarily similar to that found in all mammals. The uniqueness of the human vocal apparatus, however, has more to do with the location of the organs and, especially, the larynx. In humans the larynx is located lower down in the neck. The consequence of this is that it increases the area available for us to modify sounds.
The ability to produce certain sounds is unique to humans. For example, only humans can produce the consonant sounds ‘k’ as in kite and ‘g’ as in gun, and the vowel sounds ‘o’ as in hot, ‘ee’ as in beet, and ‘oo’ as in boo. Moreover, apes do not have the necessary fine control over the diaphragm and other muscles involved in breathing which allow humans to voluntarily use air from the lungs for speech.
With respect to vocal tract anatomy, Tecumseh Fitch (2009) has video X-rayed several mammals as they are vocalizing and notes that the mammalian vocal tract is, in fact, extremely flexible. The anatomy dynamically rearranges itself while the animal is vocalizing, with the larynx being moved downwards. Fitch, therefore, claims that if it were just down to anatomy, all mammals would have the ability to talk:
“…the constraint that’s keeping a chimpanzee from speaking or indeed keeping a dog from speaking is not the peripheral vocal anatomy. Because any of these animals that we’ve looked at are able to lower the larynx and reconfigure the peripheral vocal tract into a human-like configuration, basically anytime they vocalize…So, it must be in the brain, by a process of elimination.”
So, to communicate orally we need both an appropriate vocal apparatus and a brain capable of processing large amounts of symbolic information. In humans we can identify a language centre in both hemispheres of the brain, within what is known as the planum temporale. In 94% of people the part of the brain known as Wernicke’s area (see Anatomy of the Brain), situated within the planum temporale, is larger in the left hemisphere than in the right. For these people, the left hemisphere is the dominant side for language.
In addition to this language processing part of the brain we also know that humans are adept at processing auditory signals. At about one month of age a human infant is capable of distinguishing between certain speech sounds, such as ‘p’ as in the word pin and ‘b’ as in the word bin, and ‘t’ as in the word tin and ‘d’ as in the word din. In addition, at just six months of age the infant is capable of detecting sounds as quiet as 1 decibel (dB) (for comparison, a whisper is around 10 dB). This suggests that humans’ auditory perception is genetically pre-programmed and indicates an innate preparedness to learn language.
Until recently it was thought that no language centre was located in the brains of the great apes. However, in 1998 a research team led by Patrick Gannon made measurements of the size of the planum temporale in 18 chimpanzee brains. Their findings showed that in 17 out of the 18 chimp brains the area was larger in the left hemisphere than in the right, i.e. 94% – just as in humans. This finding is, obviously, not conclusive but it does suggest the possibility that humans and chimps may share a common neurological substrate for language.
The speed of acquisition of language
Innatists argue that the staggering rate at which children acquire language skills can only be explained if one supposes that children are genetically pre-programmed to learn language. They claim that the child does not come to the language learning task with a blank mind but has an innate disposition to learn language.
This claim is put into perspective when we note that the average five-year-old has an expressive vocabulary of around 2000 words and that by the age of seven years this will have doubled. In addition, the size of the comprehension vocabulary (i.e. the words understood by the child but not necessarily used expressively) will be much higher than this. Further, a five-year-old will understand most grammatical structures and express their self clearly in most situations, using so-called wh-questions such as how, when and why. Their speech will be intelligible to familiar and unfamiliar listeners, as they will use all but the most difficult speech sounds. Moreover, the child will be able to use language for a variety of reasons: to express feelings, to make requests, to disagree, and so on. This is a remarkable feat to which chimps can never aspire.
Language is unique to humans
Whilst the human compulsion to communicate is realized by their use of language, some have challenged the claim that language can only be used by humans.
Attempts have been made to teach language skills to a variety of animals but the most success has been achieved through working with chimpanzees. The two most famous chimps are Washoe and Sarah who began their training in the 1960s. Now, because chimpanzees do not have the requisite vocal apparatus to be able to speak, Washoe was taught a version of American Sign Language and Sarah was taught to manipulate plastic tokens on a magnetic board.
Whilst the two apes were able to learn some aspects of language, the process was extremely slow and laborious and Washoe had a vocabulary of only 85 signs and 294 two-sign combinations after about five years training. It is quite an understatement to say that this does not compare favorably with the extensive language development achieved by the average five-year-old human. However, both Washoe and Sarah did demonstrate some key properties of language:
- A remarkable development occurred when Washoe adopted an infant chimp named Loulis. No sign language was used by humans in Loulis’ presence for five years and yet Loulis was able to learn over 50 signs through her association with Washoe. This demonstrates the property of cultural transmission, i.e. the chimps continued to use sign language and pass this on to other chimps without any input from humans.
- Sarah was presented with the construction ‘Brown (is the) color of chocolate’ when there was no chocolate in sight. Later, she was shown several colored objects and presented with the phrase, ‘Take brown’ and she successfully selected a brown object. Sarah was, therefore, able to use language to think of something that was not immediately present and to apply that learning at a later time. This is an example of the key language property known as displacement.
Several other apes have been taught language skills and various claims that they have been able to demonstrate other key properties of language have been made.
Innatists noted that whilst different languages (e.g. English, Welsh, Chinese) have different rules or grammars, they also have many things in common. These language similarities are known as linguistic universals. There are many cited examples of linguistic universals but we will illustrate them here with just one example – the use of negatives.
It is claimed that children follow a remarkably similar pattern when developing the use of so-called negation. Typically developing children appear to follow a predetermined pattern of language learning. For example, at around 18 months of age children learning English form utterances made up of two-word combinations such as the following.
At first, when they attempt to make the utterances negative, children simply put no or not in front of the two-word combination:
no daddy gone
not Sarah play
no doggy bark
At a later stage the child appears to realize that the negative should be contained within the utterance and the no or not is then inserted between the words:
daddy no gone
Sarah not play
doggy no bark
Eventually the adult form is used:
daddy didn’t go
Sarah isn’t playing
the doggy is not barking
The important point to note here is that, with the obvious exception of the final adult forms, the child could not possibly have been imitating adult utterances. This is because the child’s utterances do not represent grammatical English constructions. In sum, adults do not talk like this! This is often referred to as the poverty of stimulus argument, i.e. such-and-such a grammatical feature must be universal because it would be impossible for children to learn it on the basis of the evidence they are provided with.
Linguistic universals, then, are seen as generalizations abstracted from some universal grammar that is shared by all languages, and which is thought to be innate to humans. A linguistic universal is, therefore, a statement that is true for all natural languages.
However, the claim that such absolute universals exist is controversial. It appears that absolute universals, i.e. statements that are unequivocally true for all natural languages (e.g. all spoken languages possess consonants and vowels; all natural languages use pronouns; all spoken language have at least three vowels) are actually quite rare. Typically, for the majority of so-called linguistic universals, counter-examples that refute the claim can be found. It seems, therefore, that most so-called linguistic universals are actually tendencies, i.e. they are statements about linguistic features that may not hold true for all languages but the features are, nonetheless, too widespread to be the result of chance.
Language acquisition device (LAD)
Chomsky referred to the child’s innate general language learning ability as the Language Acquisition Device (LAD). He claims that children have a blueprint in the brain that allows them to recognize the structure-dependence of language and to manipulate these structures. Consider the following utterance.
Graham kissed Margaret
The units of this utterance can be manipulated to produce the following:
Margaret was kissed by Graham
When we consider these two utterances it is apparent that they both have the same essential meaning. However, their form, or structure, is different. We would say, therefore, that they both have the same deep structure but that they have a different surface structure. The original utterance Graham kissed Margaret could be further manipulated to form the following utterances:
Who Graham kissed was Margaret
The person kissed by Graham was Margaret
It was Margaret who Graham kissed
Again, we must conclude that each of these utterances has the same essential meaning, or deep structure (i.e. Graham kissed Margaret), but that their surface structures are quite different. Now consider the following utterance.
the chicken was ready to eat
What is the deep structure of this utterance? This is much harder to determine because the utterance is ambiguous. The possible deep structures may be represented diagrammatically as follows.
the chicken was ready for the chicken to eat
the chicken was ready for someone to eat the chicken
On the basis of the utterance alone, we are uncertain as to whether or not the chicken is ready to eat for itself or if the chicken is ready to be eaten by someone else. In this instance, therefore, we see that two different deep structures give rise to just the one surface structure. Now consider the following utterances.
the chicken was anxious to eat
the chicken was delicious to eat
The syntax of these two utterances is alike. Structurally, therefore, they are akin and they are again said to have a similar surface structure. However, their meanings are quite different. In the case of the first utterance it is the chicken that is doing the eating. In the second utterance it is the chicken that is being eaten! Consequently, whilst these two utterances demonstrate a similar, but not exactly the same, surface structure they have different deep structures.
Chomsky argued that children ‘know’ about deep structures and that they are able to apply rules that allow them to manipulate these structures, giving rise to a variety of surface structures. He calls these grammatical rules transformations.
Chomsky’s ideas are persuasive and his theories have gained ground over the years. In sum, his proposals seem to imply that if a child has a properly functioning LAD then language will develop, regardless of the kinds of language experience the child is exposed to, as long as they are raised in an otherwise nurturing environment. Alternatively, if the LAD is damaged in some way it would seem that no amount of environmental support or teaching would make a difference.
However, two main criticisms may be made:
- The innatist view tends to focus on the internal, mental structures and thinking processes of the child. It is unlikely, therefore, that what research evidence we might be able to gather would enable us to fully understand exactly what is going on inside a child’s mind. Researchers have, however, developed techniques for making informed inferences from what they observe.
- The role of other people in assisting the child to learn language tends to be overlooked. As noted adult speech is fraught with hesitations, repetitions, slips of the tongue, and so on, and it therefore provides an imperfect model. However, research has shown that adults do in fact make considerable modifications to their speech when talking to children. These modifications are designed to assist the child with language learning and this type of modified talk is known as child-directed speech.
Fitch, W.T.S. (2009) Why Do We Talk? BBC Horizon 10 Nov 2009 [WWW] http://www.bbc.co.uk/programmes/b00nx7n4 Accessed 16.11.2009.