Page 136 - Encyclopedia of Philosophy of Language

Truth and Meaning
family resemblance or that he intended to introduce a new technical term into philosophy or linguistics, nevertheless his discussion—along with the expression 'family resemblance' itself—has generated a great deal of interest in the philosophy of language and else- where.
1. Family Resemblance,Definition, and Counter- examples
Wittgenstein's discussion of family resemblance (1958: Sects. 65-ca. 92) begins with captivating imagery, in response to the question of whether games have some- thing in common in virtue of which they are games:
Don't say: 'there must be something in common, or they would not be called "games"'—but look and see whether there is anything common to all.—For if you look at them you will not see something that is common to all, but similarities, relationships, and a whole series of them at that. To repeat: don't think, but look!... And the result of this examination is: we see a complicated net- work of similarities overlapping and criss-crossing: some- times overall similarities, sometimes similarities of detail. (Sect. 66)
I can characterize these similarities in no better way than by the expression 'family resemblance'; for the various resemblances between members of a family overlap and criss-cross in just that way: build, features, color of hair, gait, temperament, etc— (Sect.67)
Some take the imagery to exhaust the point, which they then see as tied to either or both of these theses: first, there is no feature (other than family resem- blance) shared by all items to which concept X (e.g., the concept 'game') applies; second, a weakening of the first, no feature (other than family resemblance) shared by all items to which concept X applies is absent from all items to which X fails to apply. Either thesis might be taken to apply only to certain concepts, which would then be of a special type, viz. family resemblance concepts; or to apply to all concepts, in which case family resemblance is a feature essential to concepts, at least human ones. In either case, the the- ses are generally taken to rule out defining a concept, at least by stating 'necessary and sufficient' con- ditions—that is, definitions of the kind 'an item fits the concept "chair" just in case it is F ,... Fn.' For, according to this view, it is always possible to find a counterexample to any purported definition of this sort, at least where family resemblance holds.
It should be noted, however, that a sufficiently nar- row notion of feature might rule out a definition of the above form while there were still statable necessary and sufficient conditions for the concept to apply, e.g., disjunctive ones. Wittgenstein considers and rejects this as a possible way round his point. This rules out any reading of the point on which it is specifiable which features make for relevant resemblance.
If this is the idea of family resemblance, it seems questionable. In fact, many have questioned it. Any
philosopher nowadays is reasonably skillful at the counterexample game. ('Suppose Martians had crys- talline brains and wheels. Couldn't one of them, though conscious, actually function as a tram? And then couldn't a tram intentionally run over some- one? ') However, suppose one wanted to construct an adequate semantic theory of English. In dealing with the lexicon, should just any such counterexample be taken into account? If not, then it might be argued— and some have done—that Wittgenstein is in fact wrong about most of the lexicon. (Incidentally,Witt- genstein places no weight in the discussion on a dis- tinction between, say, concepts and predicates. This is in line with his general eschewing of technical dis- tinctions between types of items, such as sentences and propositions. Here it is worth keeping track of his use of 'Satz' and of its standard translations into English.) Moreover, there are those who hold that, while 'tin,' for example, might have turned out to be something other than it did, it could not have been anything else, which seems to make the concept 'tin' definable by necessary and sufficient conditions.
2. RulesandtheApplicationofConcepts
Perhaps, though, family resemblance concerns a deeper point. Grant that family resemblance is, at least in part, a principle about all cases of what a concept would apply to. There is still the problem of what it is for a concept to apply to something, and of what sort of case would be a case of that; indeed there is still the problem of just which phenomenon Wittgenstein has in mind. Some features of the text may point to the answer.
2.1 Rules
One remarkable fact is that Wittgenstein's famous discussion of rules is entirely contained within the family resemblance discussion—insofar as it is purely concerned with problems of how a rule could, in fact, require such-and-such in a specific case. (The dis- cussion occurs in Sects. 84-7.) One point of the dis- cussion is: for any rule, and any occasion for applying it, there are mutually inconsistent courses of action (applications of it) such that, for each, one can con- ceive of it being in fact what that rule requires. The rule might be: place a marble in the left basket just in case it is blue, otherwise place it in the right basket. Now choose your marble. There is a way of conceiving (understanding) the rule in which following it means putting that marble in the left basket; and another way in which following it means placing that marble in the right basket. For each candidate way of fol- lowing the rule in this particular case, there is a way of understanding the rule which is a possible under- standing of that rule, on which that is what following the rule requires. For each such understanding, it is conceivable that that should be the right under- standing of the rule.
114