Noam Chomsky and His Theory of Language Acquisition Device

In Noam Chomsky’s paper “On the Nature, Use, and Acquisition of Language”, he discusses the study of language. Chomsky is coming from a rationalist point of view on language acquisition. He believes that the initial state of the language faculty is an input-output system that takes data as the input, and produces the knowledge of some language as the output. So, Chomsky believes that the initial state of the language faculty is a language acquisition device that allows for humans to be able to speak in a language. His theory is also known as universal grammar. Universal grammar is an innate set of principles and adjustable parameters that are common to all learnable natural languages. As Chomsky is a rationalist, he believes that the language acquisition device is innate to all humans, in that the device is something we are born with. After reading this paper over and over again, I think I am inclined to side with the rationalists, in that there is a language acquisition device that humans have innately. I decided to side with Chomsky and the rationalists because Chomsky had so much evidence that he used to back up his theory of language acquisition device. He used a lot of convincing evidence to turn me away from being an empiricist.

Before Chomsky created his language acquisition theory, the study of language was trying to integrate into the natural sciences. It was believed that linguistics was a part of human psychology and by extension a part of human biology. He then discusses an empirical assumption, in which there is a specific faculty of the mind/brain that is responsible for the acquisition and use of language. This faculty is only unique to humans and is common among all humans. This idea was developed during the cognitive revolution, which was concerned with the states of the mind/brain that enter into thought, planning, perception, learning, and action. Chomsky rejects a behaviorist theory, which tried to establish a connection between stimulus situations, reinforcement, and behavior because it did not consider the states of the mind/brain. The states of the mind/brain are one of the most important parts of behavior and why people behave the way that they do. An animal with a mind/brain is in a certain state. The animal is shown new sensory inputs, leading to a change in state. With the new state, the organism can do certain actions that it could not do in its previous state. But, there is no direct relationship between the sensory inputs that led to a change of state and the actions done by the organism. A child would not learn Spanish if it never heard Spanish before. If the child had an appropriate amount of data, then it would learn Spanish. However, there is no direct relation between the data and what the child says. The effort to study the relationship between sensory data and behavior, without looking at the mind/brain and its changes, is one that is doomed for failure.

Chomsky’s main goal is this paper is to discuss the language acquisition device and universal grammar. He does this very well. The language faculty has an initial state that is genetically determined, and unique to humans. The initial state can mature into a multitude of steady states, based on the number of languages available. To go from the initial state to a steady, mature state is mostly based on data early in life. If an infant has only ever heard English, it will not begin to speak in Spanish. Learning a language is something that happens to a child without them even being aware of it. It is like puberty in that sense. Language is acquired by the initial state of the language faculty determines possible rules and modes of interaction. Language is acquired “by a process of selection of a rule system of an appropriate sort on the basis of direct evidence. Experience yields an inventory of rules, through the language-acquisition device of the language faculty”. The language acquisition device is also known as universal grammar. Universal grammar is an innate set of principles and adjustable parameters that are common to all learnable natural languages. Think of the sentence John is too clever to expect anyone to catch. This makes sense, even though it is a bit unclear. However, John is too clever to expect anyone to catch is a nonsense sentence that cannot be understood. This is not known through training, but through the inner resources of the mind/brain, from the genetically determined constitution of the language factory. The initial state of the language faculty consists of a collection of subsystems, or modules, each of which is based on certain general principles. Think of a system as a complex network associated with a switch box that has a finite number of switches. A slight change in any of the switches could result in an infinite amount of linguistic expressions. A child’s mind must determine how the switches are set and simple data must be sufficient to determine the switch settings, in order to acquire a language.

Chomsky uses multiple examples to back up some of his theories. He used the three principles of the mind/brain in order to further his argument about universal grammar. The three principles are the empty pronoun principle, the subject principle, and the inversion principle. The John example spawns out of these principles. This shows that, without training, people are able to tell if a sentence is correct or nonsensical. People do this innately, throughout all learnable languages. Another example that Chomsky used was Socrates with the slave boy. Socrates asks the slave boy a series of questions, in order to show that the slave boy understands the truths of geometry without any prior experience. This helps Chomsky show that the mind/brain is, at the very least, capable of having some innate features in it. However, I do not think this example helps or hurts Chomsky’s argument. This is because the questions Socrates asked the slave boy were very leading, and some could say that it takes away from the experiment, which I think is true. Chomsky also used Descartes to help him further his argument. Descartes rejected the idea that perception is a process in which the form of something is imprinted on your brain. Descartes used the example of a blind man with a stick can make a mental representation of a cube by touching certain points of the cube with the stick. However, the cube is not imprinted onto the blind man’s mind/brain. The mind uses its own resources and principles to create a mental representation of the cube. Descartes also used the example of a triangle. A person can be shown a triangle and perceive it as one, even if it is a much more complex figure, rather than a Euclidean triangle. This is because the Euclidean triangle is produced in our minds because the mechanism of the mind is based on the principles of Euclidean geometry and produces these figures as models for perception and learning. This is contrasted with the empiricist David Hume saying that we have no idea of a triangle or a straight line because we cannot differentiate perfect images of the objects from the defective ones of the real world.

I also think that universal grammar is backed up by Ann Senghas' research about the Nicaraguan sign language that was recently created in the 1970s. Before the 1970s, deaf children in Nicaragua had no system of language to communicate. This changed after a special school was opened up that allowed for deaf children to attend. The deaf children eventually were able to develop their own type of sign language. The NSL (Nicaraguan Sign Language) became a new language and developed very quickly over time. The changes in the sign language can reveal learning mechanisms in childhood. They observed two of the learning mechanisms developed. They are, “The first is a dissecting, segmental approach to bundles of information; this analytical approach appears to override other patterns of organization in the input, to the point of breaking apart previously unanalyzed wholes. The second is a predisposition for linear sequencing; sequential combinations appear even when it is physically possible to combine elements simultaneously, and despite the availability of a simultaneous model”. This just helps Chomsky show more evidence that language is not really a learned trait, but one that is developed through the inner workings of the mind/brain.

In conclusion, Chomsky just makes so much sense in his arguing that there is a language acquisition device that is innate in all humans. The language acquisition device is also known as universal grammar. Universal grammar is an innate set of principles and parameters that is common in all learnable languages. Chomsky’s examples were very good and very difficult to respond to. Sometimes, Chomsky would bring up a potential objection to one of his ideas, but he would just shut that objection down almost immediately. The Senghas paper definitely could serve as a support to Chomsky’s argument that language acquisition is innate. The deaf children, even though NSL is a relatively new language, were able to grow and develop the language more in each generation. The second cohort signers were able to develop the language more than the first cohorts, in ways that the first cohorts could not do as adults. I think that Chomsky did a great job with his theory of language acquisition and the evidence he has to defend it. While I understand the empiricist view on language acquisition, I think the way Chomsky defends his theory is a much better way.