Thursday, July 9, 2009

More on the supposed paradox of identity

In my last post, I suggested there might be a paradox regarding personal identity. Specifically, it seems that our ability to refer to ourselves is undeniable, and yet we cannot specify what, exactly, we are referring to. Not even a purely abstract mind-stuff would seem to get the job done. So, what are we talking about, when we talk about personal identity?

If there is a paradox here, it is probably the same as the classic Ship of Theseus paradox. (There are other related paradoxes and philosophical arguments covered in the Stanford Encyclopedia entry on relative identity. I haven't reviewed that entry yet, so I'm not sure how similar or dissimilar my points here will be to anything on that site. I'm only offering the link as a suggestion--to myself as well as to others--for further research.)

The issue comes down to this: that we use different methods or standards for deciding on questions of identity depending on the circumstances. In some cases, the identity of an entity is determined by the identity of its component parts. This is called the Mereological Theory of Identity (MTI). For example, a bicycle that has been disassembled and then later reassembled is still the same bicycle, plausibly because enough of the parts are the same. Yet, in other cases, MTI does not seem to apply. Consider: Every part of the Ship of Theseus is gradually replaced. Eventually, every part of the ship is different, but the identity of the whole remains intact.

We are forced to look beyond MTI for some other criterion of identity. Spatio-Temporal Continuity (STC) is one candidate. The Ship of Theseus is changed gradually, so that its existence is not interrupted in space or time. The STC theory of identity works, but not in all cases. It doesn't work for the bicycle that was disassembled and later reassembled. So we are left with at least two distinct and incompatible theories of identity, two mutually exclusive notions of what it means to be something.

We might try to find a unifying principle here, something capable of uniting these two criteria into a single criterion of identity. This, I think, is related to what Kripke and Putnam were up to in their theorizing about natural kinds. Basically, they argued that identity is a matter of causality. The bicycle which has been disassembled and reassembled is the same bicycle, not because it has the same parts, but because the reassembled bicycle stands in the right sort of causal relationship to the original bicycle. Similarly, the new Ship of Theseus stands in the right sort of causal relationship to the old one.

This approach makes intuitive sense, but it only glosses over the problem. We are left wondering what "the right sort of causal relationship" means, and why two remarkably different criteria can both be "right." What determines the rightness here, if not the two divergent criteria we were trying to unify in the first place?

It would seem that Kripke/Putnam do not offer a unifying principle, but only a means of ignoring the divergence. We still want to know why the notion of identity does not reduce to a single criterion when our understanding of identity seems to be singular.

Instead of trying to unify the criteria, we could respond with a resounding "so what?!" We might acknowledge that there are divergent criteria for identity, and that we have been misled into thinking that identity is a singular phenomenon.

There is nothing obviously wrong with this response, though it does force us to wonder why the notion of identity seems to be singular, when it really isn't. That is, it leaves us wanting to know how the notion of identity has come about and why we are so easily misled about it.

As some of my earlier posts have indicated (see here, here, and here), my response to this issue is decidedly Wittgensteinian. We can understand the notion of identity by understanding how the language of identity is used. To understand why we think of identity as a singular phenomenon, we must understand what is happening when we regard entities as such. This goes for everyday objects as well as people. Personal identity is not a special case. To understand what it means to be a person, we must understand what is happening when people are regarded as such.

The notion of personal identity is not established by introspection. I do not first know that I exist, and then wonder about whether or not there are other minds. Rather, I learn that I exist by learning that I am not other people. Identity (including personal identity) is a social phenomenon. I can refer to myself according to any number of criteria of identity (there need not be only the two mentioned earlier in this post), in so far as there is a community in which the rules for reference make sense. The meaning of identity is not necessarily a matter of causality, though having "the right sort of causal connection" might be part of any number of rules for identity.

We are so misled by the notion of identity because we misunderstand the nature of language. Rather than worrying about a mind/body problem or any paradoxes of identity, we should instead look at how language works.