Tuesday, July 13, 2004

Refuting the Alethic Relativist, Part Two

My strategy for refuting the alethic relativist is to drive him onto the prongs of a trilemma. A trilemmatic argument against a thesis is an argument possessing exactly three mutually exclusive ‘horns’ or lemmata each of which is unacceptable, but which are jointly exhaustive of the possibilities available to the defender of the thesis, one of which the defender must accept. The thesis in question is

R. All truths are relative

which is equivalent to

R*. All truths are true-for-X.

In Part One, I showed that if (R) applies to itself and is taken to be true simpliciter (true period, nonrelatively true), then (R) is self-refuting. Clearly, if all truths are relative, and (R) is true, then (R) cannot be true simpliciter, but must itself be true-for-X, or relatively true. The consistent alethic relativist must adopt a relativized relativism. But this too leads to difficulties explained in Part One.

There is, however, a second way to avoid the contradiction that ensues when (R) is taken to be both self-applying and true simpliciter, and that is to deny that it applies to itself. This yields

ER. All truths except (ER) are relative.

Accordingly, there is a class of relative truths and a class of absolute truths, where the latter contains (ER) and its immediate logical consequences. The problem that now arises is whether it makes any sense to suppose that there are two kinds of truth, absolute and relative. Since there is no difference among absolute truth, truth simpliciter, and truth, this is equivalent to the question whether or not there are two kinds of truth, truth and relative-truth.

Now this is absurd on the face of it. It is like saying that that there are two kinds of leather, leather and artificial leather. Obviously, leather is not a proper kind of leather, where a proper kind of F is a kind that admits of other kinds of F. And since artificial leather is not leather at all, but merely a material that resembles leather closely enough to be confused with it, artificial leather is not a kind of leather. ‘Artificial’ in ‘artificial leather’ is an alienans adjective (like ‘decoy’ in ‘decoy duck’) and not a specifying adjective. Now just as it is nonsense to say that leather and artificial leather are kinds of leather, it is also nonsense to say that truth and relative-truth are kinds of truth. ‘Relative’ in ‘relative truth’ is an alienans, rather than a specifying, adjective.

To sum up. The concept of truth is the concept of something absolute by its very nature. Talk of relative truth is just a confused way of referring to something quite distinct from truth, rational acceptability, for example. The proposition that water is an element was rationally acceptable to the ancient Greeks, in the sense that it was reasonable for them to believe this proposition, but it is not rationally acceptable to us. Rational acceptability is a relative concept. It depends on who is doing the accepting, how he is situated, what his background knowledge is, etc. Truth, however, cannot be identified with rational acceptability. To see this, consider that it was reasonable for the ancient Greeks to believe that water is an element, reasonable for Dalton to believe that it is the compound HO and reasonable for us to believe that it is H2O. If truth = rational acceptability, then, given the truism that the truth about X corresponds to the way X is, it follows that the chemical structure of water has changed – which is absurd. Therefore, truth cannot be identified with rational acceptability.

But what about identifying truth with rational acceptability (or warranted assertibility) at the ideal limit of inquiry? This doesn’t work either as a later post may demonstrate.