Hasker on skeptical threats, best explanations, and transcendental arguments
A redated post.
William Hasker, however, while previously endorsing the gist of my claim that the argument should be a best explanation argument rather than a skeptical threat argument, offered another suggestion in his mostly friendly response to me in Philosophia Christi. He wrote:
However, if the Skeptical Threat strategy claims too much for the Argument from Reason, there is a danger that the Best Explanation strategy may claim too little. On the face of it, this strategy seems to invite the following kind of response: “It may be true that we naturalists have not, so far, produced a satisfying explanation for the process of rational inference. But there is nothing especially surprising or alarming about this fact. Finding good scientific explanations is hard work and often takes considerable time, and the relevant sciences are still in their infancy. We must simply be prepared to wait a bit longer, until we reach the stage where the desired explanations can be developed.
He then makes the following recommendation:
The objection is not merely that naturalism has not yet produced an explanation of rational inference and the like, as though this were a deficiency that could be remedied by another decade or so of scientific research. The problem is that the naturalist is committed to certain assumptions that preclude in principle any explanation of the sort required. The key assumptions are three in number: mechanism (the view that fundamental physical explanations are nonteleological), the causal closure of the physical domain, and the supervenience of the mental on the physical. So long as these assumptions remain, no amount of ingenious computer modeling can possibly fill the explanatory gap. In order to bring out this feature of the situation, I propose that the first two stages of the Argument from Reason are best viewed as a transcendental argument in roughly the Kantian sense: They specify the conditions which are required for experience of a certain sort to be possible—in this case the kind of experience found in the performance of rational inference. .
I have already discussed the transcendental impact of the arguments from reason, and I think Hasker’s suggestion is a good one.