Page 197 - Encyclopedia of Philosophy of Language

formulations. The word elephant can be explained (partially, at least) by pointing at one, saying that that is an elephant. The point is that in doing so it is not just giving one example of how the word is applied, not just giving a hint for guessing at how to apply it in other cases. It is an explanation in its own right, a way of specifying the standards governing the use of the word, a way of expressing a rule. The tendency to think otherwise, that is, to think that no number of mere examples of using a word, or applying a rule, could determine future applications, presupposes that understanding of an expression, grasp of a rule, is something essentially private, something which can- not be fully conveyed to others. But, so Wittgenstein argues, nothing mental could determine the cor- rectness or incorrectness of future applications, since anything mental could at most be contingently related to them. Hence, understanding cannot be essentially private, a hidden mental phenomenon; if a word is understood, then it can also be explained to others.
Wittgenstein's point of departure is the practice of speaking a common language. Such a practice stands out as normative, involving training, explaining, and correcting. Accordingly, a speaker taking part in the practice is pictured as one who has acquired knowl- edge of the practice, a great number of interrelated abilities. Such an ability is not only in conformance with communal standards, but is an ability to specify and independently apply those standards.
Only within such a practice does a rule-formulation have meaning. The rule, however, is no more precise than the interpretation of the expression of the rule. Therefore, what is determined by, or is a consequence of, a rule, is nothing other than that which is deter- mined within the practice of applying that rule (and related rules). This has consequences within the phil- osophy of mathematics (Wittgenstein 1978). It is a necessary outcome of rules of mathematics that 2+3=5onlyinsofarasthisisregardedasanecessary outcome within mathematical practice. Saying that it is a necessary outcome is a legitimate way of express- ing a normative attitude, the attitude of treating 2+3=5asunshakable,immunetorevision,asarule of grammar, in Wittgenstein's sense, but humans are inclined to misconceive themselves as having observed a logical or metaphysical fact, independent of any human activity, as if the rules could grind out conse- quences on their own.
On Wittgenstein's view there must be something intermediary between finding out the consequences of a rule and just adding to the explanation, or definition, of the rule itself. The notion of such an intermediary is, however, problematic.
Following the publication of Kripke's seminal interpretation of Wittgenstein in Kripke (1982), much interest has been devoted to questions of rule following, normativity of language and the role of
speech communities. For an overview, see Boghossian (1989).
3.3 Deep Linguistic Competence
In Chomsky's view (1976, 1980, 1986), if there are two grammars for a given language, both of which correctly specify the class of sentences of that language, then there is a basis for claiming that only one of them is the correct one, that it specifies the class of sentences in the right way.
What the right way is depends on the linguistic knowledge of the speakers. That knowledge is knowl- edge of the rules of the language. It is not knowledge of any ordinary kind, but a special kind of competence, consisting in having the rules of the language intern- ally represented, in the mind, or in the brain. That grammar is correct which provides the rules which are so represented. This conception is shared by many other linguists. Linguistics is, accordingly, regarded as a branch of psychology.
According to Chomsky, the correct grammar is the one which conforms to general grammatical prin- ciples, together making out the so-called 'universal grammar.' These principles are common to all humanly possible languages. On the one hand these principles specify grammatical categories and cat- egory structures, like the noun phrase-verb phrase structure, which are common to all possible languages, and on the other hand they impose restric- tions on further rules; some transformations, for instance, are acceptable, while others are not.
The universal grammar is thought by Chomsky to play the decisive role in the explanation of language acquisition. The problem is to explain how the child, being exposed to only a comparatively small number of grammatically well-formed sentences, can develop the competence to produce an indefinite number of well-formed sentences himself. The reason why this is a problem is that the fragment of sentence examples whichthechildhasencounteredcanbedescribedby infinitely many different grammars, most of which do not correctly describe the whole language. Somehow the child learns to conform to rules that are correct, notonlyfortheinitialfragment,butfortheentire language.
This is explained as follows. Given a sufficiently large and diverse finite set of sentences of a language, the universal grammar, in virtue of the restrictions it imposes on acceptable rules, selects the correct gram- mar of the language. Assuming that the child's devel- opment is somehow guided, or determined, by the principles of the universal grammar, the set of sen- tence examples which the child encounters will yield internal representations of the rules of the correct grammar. Thus, the acquisition of linguistic com- petence can be explained by assuming that the uni- versal grammar, or. knowledge of it, is innate, perhaps in virtue of the structure of the brain.
Rules
175