It would be a mistake in general to suppose that one could derive all important truths from a few first principles, but one particular way to fall into this error would be to think that one can or should derive metaphysics or other branches of scientific or philosophical knowledge from epistemology. One reason for this is implied by the previous post: insofar as we come to know knowledge from a knowledge of the known or knowable, it is never possible to begin by thinking about knowledge, but one must begin by thinking about things.
Descartes’s discussion of the scenario of the evil demon appears to be an example of this error, insofar as he seems to suggest that his first knowledge is of the fact that he is thinking something. In fact, it is quite evident that one must know oneself in order to have the thought, “I am thinking,” and that one would never think about thinking without first thinking about something else.
Still, Descartes’s adoption of this error is only partial. He does realize that this is not the natural order of coming to know. Thus in the Synopsis of his Meditations he says:
Finally in the Sixth I distinguish the action of the understanding from that of the imagination; the marks by which this distinction is made are described. I here show that the mind of man is really distinct from the body, and at the same time that the two are so closely joined together that they form, so to speak, a single thing. All the errors which proceed from the senses are then surveyed, while the means of avoiding them are demonstrated, and finally all the reasons from which we may deduce the existence of material things are set forth. Not that I judge them to be very useful in establishing that which they prove, to wit, that there is in truth a world, that men possess bodies, and other such things which never have been doubted by anyone of sense; but because in considering these closely we come to see that they are neither so strong nor so evident as those arguments which lead us to the knowledge of our mind and of God; so that these last must be the most certain and most evident facts which can fall within the cognizance of the human mind. And this is the whole matter that I have tried to prove in these Meditations, for which reason I here omit to speak of many other questions which I dealt incidentally in this discussion.