Wednesday, November 29, 2006

"Self to Self"

On November 1st, the workshop met in Cobb 103 to discuss David Velleman's paper, "Self to Self".


Jason introduced the paper by pointing out how Velleman distinguishes the psychological sense of self from the metaphysical sense of personal identity. Velleman is concerned with conditions of selfhood through time. Roughly, what makes a future or a past self mine is my ability to anticipate certain experiences or have certain intentions or memories without needing to identify whose experiences, intentions, or memories they are. That ability is distinguished from the activity of imaginitive identification with someone else, e.g. Napoleon.

I can imagine that I am Napoleon by imagining certain experiences as being had by Napoleon from a first-personal point of view: how the smoke and noise of Austerlitz must have seemed to him, for example. This kind of imaginitive identification requires imagining having his experiences first-personally. But in so doing, I have to additionally think of these experiences as being had by Napoleon. My imagining that I am Napoleon would therefore involve thoughts of the following kind:

I see that the allied center is weak.
I am going to ask Marshal Soult how long it will take the troops to reach the Pratzen Heights.
I am ordering the attack on the Austrian & Russian forces.
"I" here refers to Napoleon.

The final thought ("I" here refers to Napoleon) is what distinguishes "imagined seeing" from memory or anticipation, which are more intimate relations I have to my own selves. What accounts for the special kind of relation we have to memories or anticipated experiences or intentions that is missing in the case of mere imaginings? It is some kind of (special?) causal relationship:

"I don't have to specify a person from whose point of view I am trying to frame my intention, because the future point of view is fixed by the future causal history of the intention itself" (71).

"I do not center the memory on any past subject--it is just presented to me as having been copied from a visual impression, and it consequently represents things as seen by the subject of the impression from which it was, in fact, copied. Who he was is then determined by the image's causal history" (59-60).

Velleman thinks that with the distinction between self and personal identity in place, he can make better sense of what we are interested in in splitting cases. We care about being able to anticipate experiences and frame intentions without having to have the additional thought that the self having the experiences is, e.g., David Velleman. If I know that my brain is going to be split and implanted in two different bodies, I can't anticipate experiences without also specifying which of the recipients is going to have the experience, the thought goes.

PART II: Discussion was wide-ranging. The following is a small selection of the topics we covered:

1. Jason asked whether it's correct to say that a memory "purports to be a copy of a visual impression". What does this copy purport amount to? Is it part of the content of the memory? It seems that usually, I can just remember an event, or an object, without there being any sense that my memory is a "copy" of a visual impression.

2. It struck the workshop as odd that Velleman is happy to say that given a suitable causal history, I can have Napoleon's memories in the same sense in which I can have my own, as presented to me as being mine without any additional thought of identification. So, for example, if Napoleon's memories were somehow transferred to me, it would be true to say that I remember being at Austerlitz and giving the order to attack the Allied forces. [Would this be any consolation to Rachel, the replicant in Bladerunner who finds out that her childhood memories are implants, supplied by the niece of the person who programmed her? It seems that Deckard, equipped with Velleman's account of selfhood, could say that those memories are Rachel's, even though the experiences that are their source were had by a different person.]

3. I asked about Velleman's account of intention. Velleman says "I don't have to specify a person from whose point of view I am trying to frame my intention, because the future point of view is fixed by the future causal history of the intention itself" (71). But what about a surprise splitting case, where on November 29th I frame an intention to write and drop off a check at the housing office tomorrow, but while I'm sleeping, my brain is split and implanted in two different bodies. The intention survives the split, and on November 30th, both recipients of my brain halves write a check and drop it off at the housing office. If the future point of view is fixed by the causal history of the intention, then which of these two points of view was the one that I had in mind in framing my intention? It doesn't look as if there is an obvious answer to that question. So it appears that the "future causal history" of the intention is not enough to fix the person from whose future point of view the intention will be carried out.