Aristotle says that there are proper sensibles, like color or scent or whatever-it-is-that-only-touch feels; and there are common sensibles, like size or number (later Scholastics added a few more), but he said that things like substance are only accidentally given to sense. Thus, reality or objectivity is only given to sense accidentally, the way a rock might be hot or a surface might be
phusia fuschia fuchsia. Appealing to sensation to prove reality is like trying to explain the heat in the stone by appealing to the chemical properties of stones, or explaining why a surface is fuchsia by repeatedly returning to the Euclidian properties of a flat surface. The evil deceiver argument is therefore a specialized consideration of what is sensible per accidens. The benefit of Aristotle’s approach is that it situates the problem of the real within a larger context, and provides a new set of possible responses. If it is really true that denying reality and substantiality are in the same boat, then evidence for substantiality might count as evidence for reality as well. In other words, asking “how do I know if the world is really there?” is decided by the same sort of criterion or power by which we decide the question “how do I know that the green and the shape of this leaf belong to one thing? or “how do I know that the apple I am holding is both red and one in number? After all, why should we say that there is one reality that connects these two accidents or features?” Again, say I’m talking to my wife and I’m struck by the Cartesian thought “how do I know my wife is really real, and not just the product of some evil deceiver/ scientist / dream sequence?” I’d say that this is the same thing as to ask “Setting aside whether my wife is real, how do I know that this flesh-colored, smooth, Pantene scented, five-foot-eight tall quantity is one wife? Even if I can figure out if any of this is real, I’ll still need to figure out why I posit a unity of flesh-colored and five-foot-eight! After all, I know that in the past I have been deceived about two accidents forming one thing, in fact, I’ve been deceived about this more often than about what was real!”
The analysis can keep going on these lines – good and true, for example, are only sensible per accidens and so will fall under the same sort of analysis as real and substance.
Aristotle gave us the basic fact: we cannot use the sensible per se as evidence for the sensible per accidens. No amount of appeal to the one can establish the other, any more than hearing could be any sort of final arbiter about how something tastes. So what do we say now? Is the problem that we think things like real or good need to give some sort of evidence for themselves, or that, if they must give evidence, that for some reason this must come from what is properly sensible? Or is there simply a truth – that the sensible world cannot be known as real, good, true, substantial, etc. by anything given within it per se?