On the Discovery of Married Bachelors
Oct. 1st, 2011 06:42 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
apollianaAnalytic sentences are sentences that give uncontroversial and exhaustive definitions of words. Very few words are as easily defined as 'bachelor,' however; definitions sprawl, and diverge from each other. Does a divergent definition imply a lack of analyticity? If so, there is more to analyticity than 'what is said of a sentence giving a term's definition.' It would seem that some degree of simplicity is required; but I'm not sure there's any good reason for requiring it.
Perhaps, in addition to being the characteristic of a sentence stating a definition, what marks analyticity is epistemic: it seems to you as if the definition states no more than is contained already in the term named in the sentence's subject. If it does not seem that way, philosophers will hesitate to call the sentence analytic. A divergent definition may state more than is contained in the subject term in a sense--if we take it that any given understanding of the term is an understanding of it under one of that set of definitions, but not of all of them. (Dancing around analyticity is the metaphysical notion of essential qualities: an analytic statement will name the essential qualities of its subject, but not its inessential ones.)
There is still another criterion of analyticity alleged: that of what we have learned from definitions, v. what we have learned from the world: I do not look to the world to find out whether bachelors are married or not; I look to the definition of 'bachelor.' But the world seeps in everywhere. We might learn the term by definition; or we might learn it ostensively, in which cases our experiences clearly mediate learning. Perhaps the more important criterion is how we determine whether something is true ; but even then...the fact that things we experience fit the definitions contributes to our willingness to assent to those definitions. Even some of Euclid's definitions are better understood by drawing them.
It is alleged that the analytic/synthetic distinction is no good because we might discover married bachelors. But that's not the best description of what would have to happen: linguistic drift would have to make it such that 'bachelor' changes its meaning for other reasons. (Maybe 'bachelors' are all married men in disguise, as part of some larger deception.) Then the intrepid skeptic will say, "We've discovered that bachelors are really married!" (meaning, not really 'bachelors' at all). If the deception and the awareness of it persist for a long time, the word may change its meaning; but the original discovery requires that the original meaning still be operative. This possibility does not mean that our word could really mean something different than what it seems to right now; in fact, it requires it. So the possibility of future meaning-changes does not in itself stand in the way of our current level of certainty about our definition statements.
Still, few terms are like 'bachelor' in that we do not assume discoveries will be relevant to their meanings. (But here are some that do: 'Philosophers study philosophy,' 'Cordates have hearts,' 'Squares are two dimensional figures with four equal sides'.) It is legislated that the unmarried men are called 'bachelors.' It's a name for them. No other characteristics are relevant to their fitting that definition. ('Bachelors are messy' is not analytic, if mostly true.) Many things we have labels for are not defined before they're labeled--resulting in divergent definitions and family resemblance concepts. It's not clear whether it makes sense to call attempts to confirm definition sentences about such concepts 'empirical,' however: if we already know them, we know what they mean, even if they do not have 'essential' definitions of the kind given as examples of analyticity. Multi-feature definitions will be more easily amended due to linguistic or empirical change (all the dogs on earth become three-legged), without jarring the speaker.
In any case, the possibilities of (1) linguistic change over time due to changing facts about the things we call by certain names, or (2) multi-part definitions do not stand in the way of our knowing, here and now, what our words mean--and being able, should we be intellectually astute, to set it out. (I suspect that some people think we need widespread analyticity in order to do this, though I have no idea why they would.)
Perhaps, in addition to being the characteristic of a sentence stating a definition, what marks analyticity is epistemic: it seems to you as if the definition states no more than is contained already in the term named in the sentence's subject. If it does not seem that way, philosophers will hesitate to call the sentence analytic. A divergent definition may state more than is contained in the subject term in a sense--if we take it that any given understanding of the term is an understanding of it under one of that set of definitions, but not of all of them. (Dancing around analyticity is the metaphysical notion of essential qualities: an analytic statement will name the essential qualities of its subject, but not its inessential ones.)
There is still another criterion of analyticity alleged: that of what we have learned from definitions, v. what we have learned from the world: I do not look to the world to find out whether bachelors are married or not; I look to the definition of 'bachelor.' But the world seeps in everywhere. We might learn the term by definition; or we might learn it ostensively, in which cases our experiences clearly mediate learning. Perhaps the more important criterion is how we determine whether something is true ; but even then...the fact that things we experience fit the definitions contributes to our willingness to assent to those definitions. Even some of Euclid's definitions are better understood by drawing them.
It is alleged that the analytic/synthetic distinction is no good because we might discover married bachelors. But that's not the best description of what would have to happen: linguistic drift would have to make it such that 'bachelor' changes its meaning for other reasons. (Maybe 'bachelors' are all married men in disguise, as part of some larger deception.) Then the intrepid skeptic will say, "We've discovered that bachelors are really married!" (meaning, not really 'bachelors' at all). If the deception and the awareness of it persist for a long time, the word may change its meaning; but the original discovery requires that the original meaning still be operative. This possibility does not mean that our word could really mean something different than what it seems to right now; in fact, it requires it. So the possibility of future meaning-changes does not in itself stand in the way of our current level of certainty about our definition statements.
Still, few terms are like 'bachelor' in that we do not assume discoveries will be relevant to their meanings. (But here are some that do: 'Philosophers study philosophy,' 'Cordates have hearts,' 'Squares are two dimensional figures with four equal sides'.) It is legislated that the unmarried men are called 'bachelors.' It's a name for them. No other characteristics are relevant to their fitting that definition. ('Bachelors are messy' is not analytic, if mostly true.) Many things we have labels for are not defined before they're labeled--resulting in divergent definitions and family resemblance concepts. It's not clear whether it makes sense to call attempts to confirm definition sentences about such concepts 'empirical,' however: if we already know them, we know what they mean, even if they do not have 'essential' definitions of the kind given as examples of analyticity. Multi-feature definitions will be more easily amended due to linguistic or empirical change (all the dogs on earth become three-legged), without jarring the speaker.
In any case, the possibilities of (1) linguistic change over time due to changing facts about the things we call by certain names, or (2) multi-part definitions do not stand in the way of our knowing, here and now, what our words mean--and being able, should we be intellectually astute, to set it out. (I suspect that some people think we need widespread analyticity in order to do this, though I have no idea why they would.)
