Developing Careful Reading of Difficult Texts

Gary Iseminger

The second assignment we read in my intro course, after having read Descartes's Meditation I, is a
commentary raising the question whether Descartes's arguments that various of his beliefs are doubtful are "self-refuting." The commentator's answer is that some of them may be, but the most crucial one, involving the thought that one may, for all we know, be in the power of a malicious and powerful demon who devotes all his power to deceiving people, is not.

"Self-refuting" is a technical term, which the author formally defines (without, however, putting it in a box in the margin surrounded by flashing lights). The definition is that a self-refuting argument is one such that, if the conclusion is true, one of the premises can't be known. An example would be any argument with the conclusion "Nothing can be known." Understanding the whole assignment crucially depends on understanding this notion.

At the beginning of class I ask them to write down (without looking) what a self-refuting argument is and I collect the answers (unidentified). I ask them how many think they got it right. Usually about 16 or 18 of 25 will say yes. Then I read them the answer and ask them how many still think they got it right. The number usually drops to, say 10 or 12. Then I take them home and read them and tell them at the next class how many did get it right. The number is usually around 2 or 3. (I do not necessarily expect a memorized repetition of the formula, but one can't express the essential idea without mobilizing and minimally understanding some distinction in the neighborhood of premise/conclusion and some distinction in the neighborhood of true/known.)

I discuss common mistakes people have made and invite each student to find his or her own written answer and come to talk to me about it, which a few usually do. The message is one about careful reading and, especially, careful reading of argumentative prose where the author is using words precisely (e.g., philosophical prose). The results are usually sobering.

I do something like this a couple of more times during the term. The results don't generally improve much, if at all, which is even more sobering. But some, I think, at least realize that in situations like this they have understood the reading not at all, and perhaps recognize at least one crucial thing what they would have to do to understand it.

If you have any more questions, or would just like to talk this idea over, feel free to email giseming@carleton.edu or call at x4222.