Unmiddled

I have recently been re-reading Thomas Aquinas's commentary on Aristotle's Posterior Analytics, in Richard Berquist's translation. One thing that has jumped out me more this time around than before is just how different the Aristotelian approach to first principles is, compared to what is usually found today. First principles are, for the Aristotelians, immediate principles -- but this term 'immediate' is potentially misleading, because by 'immediate' the Aristotelians mean, quite literally, unmiddled. Immediate propositions are those that require no middle term. And Aristotle's Posterior Analytics can be interpreted entirely as a reflection on the relationship between middled and unmiddled propositions: it clarifies what is meant by both, proves that the latter must exist, gives methods for moving from mediate propositions to immediate propositions, and so forth. This is very different from what you'd find in most kinds of modern foundationalism, which are based not on objective methods but (usually) on identifying some privileged sort of experience (the way to get at first principles after Descartes, I suppose). But most of the issues with this sort of foundationalism don't exist for the Aristotelian approach to first principles: there is an objective (even if, on occasion, difficult) method for identifying first principles, so there is no problem of whether we know them: after all, we do know them, or at least can: you just follow the method up through mediate principles and you'll eventually arrive at them. Epistemological questions end up being purely a matter of detail: given that you know them, you naturally want to know what's involved in knowing them, and that's all the work there is for epistemologists to do. This is not minor work, but it makes justification a relatively minor issue. There is no appeal to special experiences (self-referential, or direct, or what have you); the first principles are established by a method that is, at least in a broad sense, logical. That's a very different perspective than what usually seems to be assumed in modern discussions of foundationalism.