Wednesday, July 27, 2011

thesis (1.3)

We might want to say that, though a well-accepted definition of "cause" remains elusive, we can still talk about causes of various kinds, and even make general statements which are true regardless of the kinds of causes to which we might choose to apply them. This fact seems fairly well-evident; for even if we have some favorite linguistic or metaphysical account of causality, certainly not everyone agrees, and yet we should like to say that those on both sides of the disagreement are able to understand and affirm true statements about causality. Consider for instance a naive speaker of English who has never considered the question of what precisely constitutes a cause: why should we deny that he can wield the language, which includes such terms as "cause" and "effect"? Surely it is not necessary that a speaker must abstractly analyze his speech before he can successfully use it to communicate; otherwise what shall we say, for instance, of young children who are just beginning to learn to form meaningful sentences? So if we wish to deny that naive speakers can speak meaningfully about causal relations, it seems fair to insist on some other reason than simply that he hasn't bothered to lay out an adequate definition of causality.

On the other hand, let's consider what it means for a person to speak meaningfully about causality without being able to properly define his terms: Suppose our naive speaker forms and utters the sentence

(1) The heat of the oven caused the dough to bake into bread.

Not only does it seem clear that (1) expresses meaningful statement, but also it's easy to imagine circumstances under which this statement would be true. Let's further suppose that such circumstances obtain. How, then, has the speaker managed to construct a true statement without being able to define his terms? To answer, I suggest that the speaker has learned to do so; and if we assume that he is a native speaker who first began using the language of (1) in his childhood, then it seems plain that his learning came by example from observing other speakers. So the speaker knows something of how his fellow speakers tend to use the terms in (1), and he has drawn upon this knowledge to successfully construct a sentence which expresses a true statement.

With this in mind, I further suggest that

(1.3) We must use the best judgment of native speakers of a language regarding the linguistic tendencies of other speakers in order to investigate the truth versus falsity of statements expressed by sentences in that language whose terms are not well-defined.

Among such statements are all those which involve the concept of causation. Consider for example the sentence

(2) For every cause x, there exists an effect of x.

If a person utters (2), we might try identifying his statement as true or false by asking ourselves if we can imagine some cause c and a set of circumstances under which the sentence "there exists an effect of c" would express a falsehood. If we cannot bring to mind any such circumstances, then, assuming our imagination is typical for speakers of the language, this may indicate that (2) can never be false. Now, to rely on our limited imagination to provide counter-examples is to invite error. Even to pool the collective imagination of a community of speakers devoted to the task can hardly guarantee success. Yet the only viable alternative of which I am aware is to appeal to our linguistic intuition, which seems to afford even less protection from error.

A person might try to resist this fragile method by drawing upon a pre-existing stock of statements which we know to be true, and which either imply (2) directly or else from which we may infer that (2) is probably true. However, such an alternative seems to rest ultimately upon the appeals of intuition and imagination, and thus cannot offer us any more confidence. For suppose S is a set of statements known to be true. If no sentence of S references causality, then since (2) is a conclusion about causality, then the sentences of S cannot by themselves inform us that (2) is true, or probably true; and for any statement of S that references causality, we can inquire of it how we know it to be true without appealing to a definition of causality, and hence we enter into a regress of inquiry which must terminate in some other means of decision.

Suppose now we have a sentence of the following form:

(3) For every cause x, Px.

I propose that we are justified in (tentatively) taking (3) to express a true statement if both of the following conditions hold:

(i) We (either individually or as a community of speakers of the language) have attempted without success to imagine a particular entity c which we infer from experience with other speakers of the language is appropriate to be called a "cause," and for which it is not the case that Pc;

(ii) It seems intuitive to us that (3) should be the case.

If these are satisfied, then our confidence that (3) is true should be proportional to the strength of our intuition in (ii) and the amount of effort expended in (i). Also, I propose we ought to reject (3) as expressing a false statement if

(iii) We can imagine a particular entity c which we infer from experience with other speakers of the language is appropriate to be called a "cause," and for which it is not the case that Pc.

We can consult similar criteria for varying sentence forms.

No comments:

Post a Comment