Many of those who accept the PSR do so unreflectively because they take the PSR to be self-evident. I do not think that there is any good argument against the propriety of doing so. We are perfectly within our epistemic rights to accept the Law of Excluded Middle (LEM), namely the claim that for all p we have p or not-p, because of the self-evidence of LEM, without needing any further argument for it. However, it will be of no use to opponents of the PSR or of the LEM to be told that the claim they deny is self-evident to us. Presumably, the claim is not self-evident to them, and we can all agree that there are many things that people have claimed to be self-evident that, in fact, are false, so the fact that the claim is said by us to be self-evident does not provide these opponents with much reason to accept it. There may be a presumption that what people take to be self-evident is, in fact, more likely true than not, but this presumption is often easily defeated.
One might think that philosophical disagreement about the PSR shows that the PSR is not self-evident, or at least that those of us who take it as self-evident should not see this as providing any reason to believe it to be true. Otherwise, how could competent philosophers such as David Hume or Graham Oppy fail to see it as self-evident? Or, worse, how is it that some of these philosophers take as self-evident claims incompatible with the PSR?
If we think we should accept the LEM because of its self-evidence despite some brilliant intuitionist mathematicians' denials of it, we will be unimpressed by this argument. And it is not clear on what grounds we could accept the LEM other than self-evidence. Is there some inductive argument like: "For many propositions p, we have concluded that the LEM holds. Hence, the LEM holds for all propositions p"? I doubt it. The problem is that an inductive argument of the form "Many Fs are Gs, thus all Fs are Gs" is epistemi-cally close to worthless by itself. Many dogs are spotted, thus all dogs are spotted? We would do slightly better if we could show that most Fs are Gs, although even that would be very weak ("Most humans are female, thus all humans are female"). But how would we check that the LEM holds for most propositions? To check that the LEM holds for a proposition is, presumably, to determine that this proposition is true or to determine that this proposition is false, since in either case, the truth of the LEM follows for the proposition. But most propositions are such that we cannot determine whether they are true or false.
In any case, the argument from philosophical disagreement is weak. It might be that our judgment as to what is or is not self-evident is fallible, and Hume and Oppy have simply judged wrongly. Or it might be that it is possible to be talked out of seeing something as self-evident, just as it is possible to be (rightly or wrongly) talked out of all sorts of commonsensical beliefs. Finally, it could be that the PSR's opponents have failed to grasp one or more of the concepts in it due to their substantive philosophical positions. Thus, Hume's equating constant conjunction with causation suggests that he does not have the same concept of causation as I do - that he is talking of something different - and the fact that he thinks causation thus understood yields explanations, as well as his belief that infinite regresses can be explanatory, show that his concept of explanation is different from mine. Differences in views of modality are also relevant. As a result, it is far from clear to me that Hume has even grasped the PSR in the sense that I assign to it. And if not, then his failure to see it as self-evident is irrelevant.
I can give a similar story about Hume's seeing as self-evident propositions that are incompatible with the PSR, such as that no being's existence is necessary.2 Hume's concept of the necessity of p is that a contradiction can be proved from the denial of p. If LEM is
2. This is incompatible with the PSR, given the other ingredients in the cosmological argument.
true, this is equivalent to equating necessity with provability. But defenders of the Leibniz-ian cosmological argument typically use a notion of broadly logical necessity when they claim that God is a necessary being, and broadly logical necessity is weaker than provability.
At this point, it may seem as if the defense of the self-evidence of the PSR destroys all possibility of philosophical communication. If philosophers all mean different things by the same terms, how can they even disagree with one another? Two points can be made here. The first is that in many cases, when philosophers use a word such as "cause," they both mean by it what ordinary language does and they have an account of what the word says which they think is faithful to the ordinary meaning. And if this is true, then when one philosopher says "A causes B" and the other says "A does not cause B" there is a genuine disagreement between them even if their analyses of causation are different, since the first philosopher holds that A causes B in the ordinary English sense of "causes" (which he rightly or wrongly thinks is identical with his analysis of the term) and the second denies this. Second, disagreement is possible because even though philosophers may use the term "causes" differently, they will tend to agree on some entail-ments, such as that if A causes B, then both A and B occurred and B's occurrence can be explained, at least in part, in terms of As occurrence. So differences in meaning do not undercut philosophical communication, but they seriously damage the argument against self-evidence.
Self-evidence might well give those of us to whom the PSR is self-evident a good reason to believe it. But if we want to convince others, we need arguments.
Was this article helpful?