Why should there be a science called ‘linguistics’?

Dinha T. Gorgis
Hashemite University

Since linguistics is concerned with the study of human language(s) , and since ‘language’ is a mirror of the mind , then ‘linguistics’ has the ultimate aim in uncovering the nature of human mind by attempting to devise (i.e. construct) a ‘ universal’ grammar through the study of language structure, acquisition, and use . Unlike other approaches to language-study , e.g. structural , this approach is ‘mentalistic’ or ‘rational’, i.e. it employs reasoning which requires a formal language , e.g. that of computer or logic , to account for natural language(s) . This is why this approach, called TGG, is often called ‘formal linguistics’. For Chomsky, the linguist who expounded this theory in 1957, we begin our work by constructing ‘particular’ grammars for world languages. A linguistic theory that does this must be: (1) observationally; (2) descriptively; and (3) explanatory adequate. These three criteria are interrelated ; each level has something to tell us : (1) the theory (or grammar ) must be able to specify (or say ) which sentences are well-formed and which are ill-formed at all levels , i.e. phonological , morphological , syntactical , and semantical ; (2) the theory in addition to (1) , must properly describe sentence structure , again at all levels, in such a way that the description fits in well with the intentions ( or judgments ) of the speaker of hypothetical ( ideal) native speakers ; (3) in addition to (2) , the theory must be able to impose a set of maximal constraints on the description , be psychologically real , and universally valid . That is without constraints we cannot develop a theory of language acquisition in children, and without knowledge of how language is acquired, there is no point in talking about ‘particular’ (and then ‘universal’) grammar. And in order to understand what we mean by ‘ particular’ and ‘ universal’ grammar , we need to relate them to ‘ core’ (unmarked) grammar , i.e. grammar of the regularities , and ‘peripheral’ (marked ) grammar , i.e. grammar of the irregularities or exceptions .[ The two make up ‘particular’ grammar ]. UG is NOT the total sum of ‘particular’ grammars, nor is the total sum of ‘core’ (or unmarked) grammars. UG is NOT a grammar of similarities between languages; this would be a grave (big) mistake. It might be safe to talk about ‘potential’ similarities, not the ‘surface’ ones. But what does this mean? It means that the child is born with a language faculty (which has these ‘potential’ universals). On the basis of experience, say in a Jordanian world, the child will make use of a finite set of rules, i.e. it will activate the rules in the mind, to generate Arabic sentences, but not English or Russian, etc. In other words, it will discard (dispense with) the rules that generate other languages and select what is only appropriate for Arabic in terms of ‘core’ rules relevant to Arabic. Later, the Arabic child will learn ‘peripheral’ rules and, hence, ‘particular’ grammar as used by all Jordanian speakers. At this stage, linguists will be interested in talking about ‘language-in-use’, the third main theory (or study as underlined at the outset). Here we are concerned with ‘ pragmatic’ or ‘communicative’ competence , whereas with ‘structure’ and ‘acquisition’ we are interested in ‘linguistic’ or ‘grammatical’ competence’. These naturally constitute two different, but related, set of abilities (or knowledge) .

Below are some questions that will enable you to understand the workings of TGG. They are meant to remove misconceptions about this unique approach to language study.


1 Do the intuitions of native speakers make up the totality of grammar? Certainly not. But why is that? Simply because speakers of the same language vary in their judgments of ‘acceptability’ or ‘grammaticalness’. If this is so, then, how come that the linguist relies on intuitions as the main source of data? Very simple; s/he abstracts principles that cohere with the theory .Does s/he throw away exceptions and counterexamples? No, s/he only puts them aside till new facts emerge. Why doesn’t the linguist depend on the actual utterances said by speakers, i.e. performance? Because of mistakes, slips of the tongue, regional and individual variations, etc.
2 How do we constrain grammars? Simply by imposing conditions on the principle that organize grammar. For example, many languages make use of the principles of inversion to make questions, e.g. the English He is a boy will become: Is he a boy? But if we follow this principle blindly, a sequence of words such as: He did it will wrongly give you *Did he it? Therefore, a condition such as NPــــــAux =>Aux ـــــNP will prevent the generation of *Did he it? Because ‘did’ is a lexical verb here. And so on, and so forth.
3.What is the significance (or importance) of constraining grammars?
Without constraints, our grammars will work freely and, hence, generating ungrammatical sentences. Constraining is a method that enables the linguist to attain generalizations step by step. And without such generalizations, the child’s task in acquiring any language would be extremely difficult, if not impossible.

4. So what is LAD? A device that has the power of converting the child’s experience into a system of knowledge, i.e. any particular grammar.
5. How many types of ‘universals’ are recognized?
Absolute ( formal or non-statistical ) ones are available to all natural languages without exceptions , e.g. structure-dependence principle , and relative ( substantive or statistical ) universals are those that have exceptions , e.g. Languages modify forwards or backwards .
6. What kinds of evidence are there for the accumulation of experience which enables the child to acquire its language?

7. What evidence is there for the existence of ‘competence?
Different kinds of linguistic abilities: the ability to say what is/is not the native speaker’s language, to paraphrase (lexically or syntactically), to detect ambiguities, to pass judgments on acceptability and grammaticality, etc.