Last Updated on May 6, 2015, by eNotes Editorial. Word Count: 1400
Chomsky sets out to answer three questions in Knowledge of Language. First, what constitutes knowledge of language? Second, how is knowledge of language acquired? Finally, how is knowledge of language put to use? In order to answer these questions, Chomsky first lists six sentences to demonstrate the inadequacy of the traditional ideas that language is a relationship between a string of words and meanings and that people learn language by example and analogy:
1. I wonder who the men expected to see them.2. The men expected to see them.3. John ate an apple.4. John ate.5. John is too stubborn to talk to Bill.6. John is too stubborn to talk to.
Example and analogy fail to explain how people know that “the men expected to see them” has two different meanings in the first two sentences. The sentences are almost identical in words and structure, but in the first an unknown person is expected to see the men, while in the second the men expect to see unknown persons. In the fourth sentence, if “apple” is omitted, the meaning that John ate something is still implied, but in the sixth sentence, the omission of “Bill” does not imply that Bill is too stubborn to talk to someone. Therefore, the omission in the sixth sentence cannot be understood by analogy with the omission in the fourth sentence. Chomsky points out that at an early age children understand sentences like these. Furthermore, although almost no one could explain these structures to foreign students of English, they can learn them.
Chomsky notes that similar problems of structure and comprehension exist in other languages. Therefore, there must be innate brain functions that allow people to produce and understand such sentences. Also, Chomsky makes the point that knowledge of a language and use of a language are not synonymous. People with similar knowledge differ in their ability to use it, and impairment in ability to use it does not necessarily mean impairment in knowledge. To answer his first and second principal questions, Chomsky limits himself to the nature of the brain functions, which he calls universal grammar, and to a person’s internal knowledge of a language, which is what he means by the word “language.” He does not attempt to answer the third question (except indirectly, inasmuch as internal language can be studied only as it is expressed in use) until the final chapter on “Orwell’s problem.” By limiting the definition of language to these mental concepts, Chomsky can define knowledge of language as states of mind, which can be studied as parts of psychology and, ultimately, biology. The purpose of Knowledge of Language is to identify these brain functions, or capacities, and their parameters in order to develop the production and comprehension of various languages. He calls the pre-language state “SO” and the language-knowing state “SL.”
The main problem of identifying the universal grammar capacities of the mind is that the principles of the SO state must be general, or inclusive, enough to account for the variety of languages but restricted enough to account for the fact that a specific language can be learned. What Chomsky calls the SO capacities are psychological principles, with parameters that control the forms a language takes as a person learns from his or her individual contacts with the world and society. These principles and parameters must be adequate to describe a specific language, as well as to explain the development of a language under the conditions of an individual’s experience.
Chapter 3 is devoted to identifying some of these principles and parameters. In this chapter, Chomsky draws heavily upon research by other linguists, as well as on his own research since the early formulation of his transformational rules. The result of those burgeoning linguistic studies is a partial withdrawal from the concepts of phrase structure and even transformational rules. The main criticism of Chomsky’s early transformational theory was that it relied too much on syntactic structures and not enough on the parameters imposed by semantic requirements of the words in the structures. The parameters described in Knowledge of Language have a much greater semantic element. That is, the rules that govern sentences such as the six listed above depend on elements of meaning in the words rather than on the rules of phrase structure.
Three of the universal grammatical principles set forth in chapter 3 explain the differences in meaning in those six sentences. The trace principle states that a moved or omitted element leaves behind a trace of its presence, indicated by the letter e. The co-occurrence principle states that a verb demands a subject; transitive verbs and prepositions demand objects. The binding principle states that a pronoun is bound to an antecedent outside its own domain. (A domain is the smallest grammatical unit that includes the pronoun.)
For example, in “John ate an apple,” the verb “ate,” being transitive, requires an object, “apple.” If the object is omitted, as in “John ate,” an object trace remains: “John ate e (something).” (The e, having no antecedent, can, but not must, be filled by an indefinite pronoun.) In the sentences “John is too stubborn to talk to Bill” and “John is too stubborn to talk to,” the phrase “talk to” demands a subject for “talk” and an object for “to.” Therefore, the two sentences have semantic traces left where the subject and object have been omitted: “John is too stubborn to (e) talk to Bill” and “John is too stubborn to (e) talk to (e).” In the first of these sentences, “(e) talk to Bill” is the domain of e, so e cannot refer to “Bill” but is bound to “John,” which is outside its domain. In the second sentence, “(e) talk to (e),” the second e cannot refer to the first e and is bound to “John.” That leaves the first e with no antecedent; since the verb “talk” demands a trace subject, however, e can be filled by an indefinite pronoun, such as “anyone,” just as the omitted object in “John ate” was indefinite. When the three inherent semantic principles are applied, the six sentences and their meanings can be analyzed.
Chomsky presents the parameters of these three universal grammatical principles, as well as those of several other hypothetical principles, for the English language. In many cases, other languages are also used, to demonstrate the universality of the principles. Some of the other principles under consideration in chapter 3 are called X-bar theory, case adjacency, projection, licensing, theta theory, and full interpretation.
Chapter 4 answers specific criticisms of Chomsky’s use of the word “rule” to describe the functions which the brain performs to produce and comprehend sentences in a language. The basic criticism is that a person’s behavior cannot be said to follow a rule under certain conditions: when the person does not know that a rule exists to be followed, when the person is merely following an innate psychological or biological stimulus, when the person merely believes he or she is following a rule, and, since the word “rule” implies a generality, when the person’s language is an internal, individual construction. Chomsky answers these criticisms in part by using the term “principles” instead of “rules.” He also argues that where all evidence points to regularity, it is not illogical to assume that a rule exists and then to test the validity of the rule against further evidence.
Other critics have objected to the use of the word “know” to describe the mental capacities of the SO state. Chomsky admits that the word may not be entirely satisfactory and has even suggested the word “cognize” for the type of inherent principles and parameters he has postulated. He admits that much of what is now called knowledge is at least subconscious, such as the ability to interpret space and perspective. Inasmuch as no one can adequately define knowledge, in any case, Chomsky chooses to stipulate his own definition.
Although Knowledge of Language defends the theories Chomsky developed after 1957, the tone of the book is not defensive. The author never suggests that he has the definitive answers to the mystery of human language, but only that his theories provide the best explanation for all the data known. He is aware that if language is indeed purely a function of the brain, new technology for studying the brain could lead to a corroboration or refutation of his theories.