One of the ideas I have been juggling in my head which I thought might provide the basis for a future thesis if well enough developed was the idea of proposing a rationalist epistemology (as opposed to, say, anything like an empiricist epistemology). As a basic commitment, a rationalist epistemology would look to modality as though it were a constraint on epistemology. In other words, one’s epistemology would methodologically exclude suggestions which are not logically possible (i.e., suggestions for which no logically possible world exists such that they are true). One might thus be able to provide deductive closure if one can dissolve cases of Cartesian scepticism by suggesting that such scepticism is founded on what is not logically possible.
If something is not logically possible, then it ought not be considered as a constraint on our epistemology. For instance, if Descartes is to say that an evil demon could be making me think that “2+2=6” is true every time I think it, it still isn’t the case that the sentence “2+2=6” is logically possible. There is no sense in which two and two make six, or there is no logically possible world in which the sentence expresses a true proposition. Therefore, Descartes suggestion that it is still possible that we are being made to think that 2+2=4 is false, is simply not significant. It may be that “2+2=4” is false, but it is at least not logically possible that “2+2=4” is false. There is no logically possible world in which the sentence expresses truth. Therefore, there is no warrant for scepticism when considering that which is not logically possible. To follow this train of thought, this also means that some of the things which we know by intuition, such as “A v ~A” are necessary truths. There are, therefore, some self-justifying or at least basic foundational beliefs which cannot be doubted except on pain of (literally) irrationality. One can always object that the world may not reflect logic, and thus perhaps what seems to the mind to be logically impossible may still be the case – however, it is at least not logically possible for it to be the case, and one might wonder what kind of odd realism would actually entail an irrational world. If ‘the world’ is just a model we construct based on our experiences, then there just is no such thing as a logically impossible world. Moreover, if we are realists and rationalists then we will believe that logic is something like a governing principle (logos). Only a strict realist who is also an irrationalist can believe in a world which is logically impossible, and I suggest that these simply be dismissed out of hand. If one cannot disqualify a view from being legitimate by arguing that it isn’t logically possible, then how ever could anything be disqualified? Surely this principle should be compelling.
This isn’t entirely worked out, but there are two points to make. One is with respect to Natural Theology, and the second is with respect to the glaring weakness of this thesis. First, with respect to the existence of God, it seems clear that if certain things qualify as foundational once one accepts modality as a constraint on epistemology, then the existence of God follows as a logically necessary feature of ‘the world’ (understood broadly). For instance, if the Principle of Sufficient Reason is logically necessary, and it seems it may be, then it just is the case that God’s existence seems to follow along with it. Additionally, if one accepts that language is necessary for properly human thinking, and that language presupposes community, then one might be able to get to ‘other minds’ which is further than Descartes gets. Alternatively, one might accept something like an Ontological argument, which would provide God’s existence. Finally, if God exists then one can argue that a form of Axiarchism is logically necessary, and therefore that some form of theistic reliabilism is true. This would grant methodological confidence in properly basic beliefs like the existence of the past or the existence of other minds.
The great weakness of making modality a constraint on epistemology is that it isn’t clear that the human mind cannot confuse self-evident beliefs for beliefs which aren’t necessary. For instance, although somebody who suggests to me that “2+2=4” is possibly false deserves the response “well, that isn’t logically possible, and I’m only interested in what is logically possible”, it is still the case that I may in fact lapse and confuse “3+2=5” for a belief which isn’t true. Or worse, I may lapse and confuse “2+2=5” for a belief which is self-evidently true. The fact that logically possible worlds are cognitively constructed means that all talk about modality (what is logically possible, etc) is only ever as sure as the cognizer is reliable. There may be ways to overcome this difficulty, however. For instance, if one could argue their way to God’s existence and tac-on Axiarchism as logically necessary, then a reliabilism at least strong enough to justify this “rational epistemology” would also follow.