Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

mitchell_porter2 comments on Identity Isn't In Specific Atoms - Less Wrong

24 Post author: Eliezer_Yudkowsky 19 April 2008 04:55AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (68)

Sort By: Old

You are viewing a single comment's thread.

Comment author: mitchell_porter2 19 April 2008 08:00:36AM 1 point [-]

Over already? I thought we'd hear about many worlds, measure theory, decoherence, and Julian Barbour before we came to the end.

If this is the end, then it's time to evaluate the picture we've been given. Basically, it's nonsense. This is not particularly Eliezer's fault. As a sketch of how quantum mechanics works, it is accurate, and since quantum mechanics is generally not held to be in need of explanation itself, to some degree it has the imprimatur of orthodoxy as a sketch of reality itself. But that just means it is officially sanctioned nonsense.

Let's consider some of what we've been told. Suppose a physical system starts in situation A, and ends in situation B. The probability of this happening can be found by summing the amplitudes for a number of "histories" which began with A and end with B. OK; but what do we think actually happened in between? If those amplitudes were probabilities, it would be reasonable to say that just one of those histories actually happened. Not so in quantum mechanics; in quantum mechanics, sometimes things never happen because all the ways that they can happen cancel each other out (the amplitudes sum to zero). That is nonsense, it is an obvious indication that we are conceptualizing things incorrectly.

In any case, Eliezer seems to be saying that between A and B, what happens is everything and nothing. We have amplitude flows in configuration space. Configurations themselves do not change, just the amplitudes associated with them. Well, in real life something definitely changed - A became B. The picture needs more detail, to expain what that involves. Eliezer clearly favors the many-worlds explanation, but I can't critique it unless he shows me the details. For now, we just have underspecified nonsense.

It also turns out that the point of this digression was to make an argument about personal identity and its continuity over time: whatever the reason is that I, now, am to be considered the same entity as I, five minutes ago, it can't have to do with persistence of my physical parts, because my physical parts don't "have identity". So, if I focus on the proton in a particular hydrogen atom in a particular nucleotide in a particular cell in my body, I don't have any grounds for thinking that the proton that's there now is the same proton that was there a millisecond ago; at least, it's not that proton to any greater degree than it is also any other proton in the universe.

This is clealy an odd view. There are at least two possible reactions to it. One is to say, well, it's odd, but it's what Authority tells me that Experiment is saying, so I'd better believe it. Or, one might want to look a little closer at the details, and see if the peculiar interpretation holds up. In this case, I think that proceeding to the perspective of field configurations might be called for. I'm not at all sure that this stuff about indistinguishability is anything more than an artefact of taking particle configurations to be fundamental, rather than field configurations. (And that, incidentally, is another aspect of the quantum problem that this exposition hasn't mentioned, the question of which "basis" to use.)

Just some preliminary thoughts.