Wednesday, May 13, 2009

It seems I could later think I earlier thought but not know if I did, therefore I don't know now I exist?

I judge that I'm thinking and that if I think I am. Also, I know I'm CL. So, among other things, I judge that CL now thinks and thus CL now exists. That's what I think to myself now. Does that belief constitute knowledge?

Between now and later, I'll be kidnapped and my brain split into two. Each half is placed into a new body. Upon waking, I'll remember everything I now know but I won't then be able to say knowingly 'Earlier I judged that CL thought and thus must have existed' because I wouldn't know this to be true. I won't later know it because I won't then believe it. I won't then believe it because I don't know whether I can say truthfully that I, CL, existed earlier given that I don't know how to describe splitting cases. Thus, because I won't later know whether I, CL, existed previously when someone named 'CL' formulated the cogito I don't now know that it's true that I judge that I, CL, think and thus must exist.

The crucial assumption appears to be this one: if S knows that p at t1, and if (at some later time) t2, S remembers everything S knew at t1, then S knows p at t2. It seems at t1 that I know I exist at this time. It seems I later remember everything I knew prior to the surgery but it seems that after the surgery I do not know whether I existed before that surgery. It seems that either my cogito judgment does not constitute knowledge now or the principle is false. I tend not to side with Descartes, but this time I might.


Richard said...

It seems like the principle should only apply to objective, non-indexical claims. This is most obvious for tensed claims: at t1 I know it's raining, at t2 I remember all but no longer know it's raining (because now it isn't). But your case brings out that 'I' can be problematic in the same way as 'now'. Replace all such 'centering' terms and we have no problem: later I can still know (just as I did earlier) that RC thought and thus existed at t1. I just don't know whether RC is me.

Clayton said...


Good point. I was thinking over dinner that there was a problem along those lines even if we stuck with the indexical thoughts. My problem might be solved by going second-order (e.g., switching from 'I think, therefore I am' to 'I judge that I'm thinking that I think and thus that I must exist'). I'll have to find a way to restate the problem.

Fwiw, I think there are other problems with the principle since it seems that we should be able to construct Gettier type cases and cases of knowing less by knowing more that cause trouble for the principle but I had hoped to show that the principle implied that things that we all think should constitute apriori knowledge or self-knowledge didn't.