Noam Chomsky on Language Acquisition: How Do We Learn To Speak?

How do we learn to speak, write and think in our native language? This question is far easier to answer when applied to a second language rather than a first. For one thing, we tend to learn a new language in a systematic, intentional way. In contrast, learning our native language is not planned in this way. Indeed, figuring out how we actually come to learn our native language has perplexed linguists for a long time. It is an intriguing question in part because it may have implications for how we conceptualize language in general.

One of the most influential views of language acquisition comes from Noam Chomsky, and this article begins by setting out some basic assumptions about language that he and those who follow him make. It then proceeds to focus on the independence of language, and how this theory interacts with the existence of pseudo-linguistic communication systems. It finally focuses on disagreements about whether the basic linguistic information we are exposed to when we are young is sufficient to explain the acquisition of language.

It is important to be clear at the outset about certain central assumptions which Chomsky makes about language, and the bearing they have on how we should conceive of language acquisition. There is a strong sense in Chomsky’s work that he thinks of language as something special – to the point of being miraculous.

He is famously disinterested in any attempt to explain how language emerges in evolutionary terms, or by relating it to animal communication systems as a potential precursor. At one point, he says something to the effect that language must have come into existence as a kind of totally random mutation. He stops short of calling it an act of God, but makes more or less the secular version of this claim.

The Independence of Language 

The doctrine that language should be conceived of as independent from other elements of human capacity or activity is central to Chomsky’s theory of language acquisition.

There are several arguments for the independence of language, which Chomsky and those who follow him believe provide a strong basis for understanding language without reference to other cognitive functions (or at least as its own class of such functions).

First, there is the unique nature of language knowledge. Chomsky conceives of language as containing certain unchanging principles that hold across all languages. Such principles do not necessarily apply to all aspects of human thinking; it is not clear how these principles (the principle of Universal Grammar, or UG) could operate in areas of the mind other than language.

There is a degree of mutual reinforcement between arguments for the independence of language and arguments for there being some element of language which is innate or intrinsic to human beings.  Here is one argument for the independence of language which comes from language acquisition: the most general principles of language do not appear to be learnable in the same way that children learn to do other things, including learning other cognitive skills.

Language learning appears to happen in an entirely unique way.  And, as Chomsky and his followers would have it, the uniqueness of how language is learnt demonstrates a uniqueness of function, which points to a separate area of the mind focused on language: the so-called ‘language faculty’. This should be understood in contrast with cognitive theories that assume the mind is a single, unified system.

As Chomsky has it, “We may usefully think of the language faculty, the number faculty, and others as “mental organs”, analogous to the heart or the visual system or the system of motor coordination and planning.” The question, simply put, is this: why not characterize a seemingly self-contained, qualitatively distinct set of processes in the same way we normally do – be referring to an organ?

Another reason to think of language in these terms is to do with the universality of language. As Chomsky puts it: “this language organ, or “faculty of language” as we may call it, is a common human possession, varying little across the species as far as we know, apart from very serious pathology’”.

This is a helpful clarification of what the linguist is ultimately interested in when they study language. They are not interested in knowledge of any particular language (not of French or Arabic or English), but in the language faculty of the human species.

Of course, there is a very simple reason why we might not wish to refer to the language organ, or at least to heavily qualify our use of the term organ. We know where the heart is, we know where the kidneys are, but we do not have any such precise spatial location for a language organ.

Is Language Uniquely Human?

One criticism of the uniqueness of language follows a common pattern of critique for anything that human beings do which is supposedly totally unique: that is, a critique based on the observation of animals.

It has been pointed out that animals have, in certain laboratory contexts, been able to learn even more complicated communication systems than they do in the wild and it is postulated that these constitute either languages or genuinely precursors to language.

The Chomskyan rejoinder is generally to question whether the languages used in these experiments are fully human-like in incorporating certain principles that Chomskyans take to be ubiquitous of human language. It may, therefore, be possible to learn them via other faculties than language at the animal’s disposal. In a human being some aspects of language may also be learnable by some other means than the language faculty.

Quite clearly, some interesting question begging is going on here, at the point where Chomsky wishes to claim both that languages all have such a certain feature, and that any other communication systems aren’t languages because they don’t have a certain feature. This points to a deeper strangeness around the conception of language at work here, and the claim that it is qualitatively distinct from communication. It seems really quite natural to think of language as having grown out of the kinds of communication systems we notice that animals (especially apes and whales) seem to use.

Language acquisition is, on the Chomskyan conception, essentially a black box problem – that is, we are trying to make sense of a system wherein we know the inputs, we know the outputs, but we do not have a precise understanding of the system by which the inputs are processed to become outputs. As Chomsky puts it,

“Having some knowledge of the characteristics of the acquired grammars and the limitations on the available data, we can formulate quite reasonable and fairly strong empirical hypotheses regarding the internal structure of the language acquisition device that constructs the postulated grammars from the given data”.

This seems reasonable, and quite elegant. However, there are other, slightly strange things going on in the Chomskyan worldview.

Consider the following quote from Chomsky: “The Martian scientist might reasonably conclude that there is a single human language, with differences only at the margins”. To a non-linguist (or at least, to this non-linguist) that seems like a very odd thing to say, given that even those speaking languages really rather similar to one’s own might as well be making random noises with no particular structure if one cannot understand them.

However, it is clear that this claim (or one which is similarly bold) is required to defend the most ambitious claims which Chomsky makes, that is, that “the language organ is the faculty of language (FL); the theory of the initial state of FL, an expression of the genes, is universal grammar”.

The Linguistic Acquisition Device

Let us return to the idea of the black box. Imagine we are studying a child who is learning a language. We can analyze the primary linguistic data they have access to (e.g, the words that are spoken around them), and we can analyze the way in which they themselves come to speak. What we cannot easily access is how the first becomes the second. We are seeking explanations.

One way of conceptualizing this black box follows the ‘Linguistic Acquisition Device’ (LAD) model. There are three levels of explanation our theories must aspire to on this model. Observational adequacy is the first level that a linguistic theory has to meet: a theory is observationally adequate if it can predict grammaticality in samples of language, that is to say in the primary linguistic data of adult speech as heard by the child, otherwise called the input to the LAD. Descriptive adequacy is the second level: a theory achieves descriptive adequacy if it deals properly with the linguistic competence of the native speaker, i.e. the generative grammar output from LAD. Explanatory adequacy is the third level: a theory is explanatorily adequate if the linguistics theory can provide a principled reason why linguistic competence takes the form that it does, i.e. if it can explain the links between linguistic competence and primary linguistic data that are concealed within the LAD itself.

The crux of the problem called ‘Plato’s problem’, which is at the heart of Chomskyan ideas of language acquisition, is this. What if something comes out of the LAD which didn’t actually go in? What if the outputs do not seem to match the inputs?

“How do we come to have such rich and specific knowledge, or such intricate systems of belief and understanding, when the evidence available to us is so meagre?”
(Chomsky, 1987).

The answer provided by the Chomskyans is that much of our linguistic knowledge must come from the internal structure of the mind itself.  This argument is often called the “poverty of the stimulus” argument. There are various alternative ways of conceiving of this. One strategy we could call the empiricist approach, in contrast to Chomsky’s apparent rationalism. That is, we could say that although nothing is hidden ‘in’ the LAD as such, we are able to use the linguistic experience we acquire far more extensively than one might assume. We can generate better linguistic competence from less data than it would seem, perhaps because of some underlying structure in language or some underlying receptive structure in our brains.