((moved here from the suffocating depths of open thread part 2))

Back when I first heard of "timeless decision theory", I thought it must have been inspired by Barbour's timeless physics. Then I got the idea that it was about treating yourself as an instance of a set of structurally identical decision-making agents from across all possible worlds, and making your decision as if you had an equal chance of being any one of them (which might be psychologically presented to yourself as making the decision on behalf of all of them, though that threatens to become very confused causally). But if the motivation was to have a new theory of rationality which would produce the right answer for Newcomb's "paradox" (and maybe other problems? though I don't know what other problems there are), then it sounded like a good idea.

But the discussion in this thread and this thread makes it look as if people want this "new decision theory" to account for the supposed success of "superrationality", or of cooperative acts in general, such as voting in a bloc. There are statements in those threads which just bemuse me. E.g. at the start of the second thread where Vladimir Nesov says

since voters' decisions are correlated, your decision accounts for behavior of other people as well, and so you are not only casting one vote with your decision, but many votes simultaneously

I should know enough about the possibilities of smart people tripping up over the intricacies of their own thoughts not to boggle at this, but still, I boggle at it. The decision made by other people are caused by factors internal to their own brains. What goes on in your brain has nothing to do with it. Their guess or presumption of how you vote may affect their decision; your visible actions in the physical world may affect their decision; but the outcome of your decision process does not causally affect (or "acausally affect") other decision processes in the way that Vladimir seems to imply. At most, the outcome of your decision process provides you (not them) with very limited evidence about how similar agents may decide (Paul Almond may make this point in a forthcoming essay), but there is no way in which the particular decision-making process which you perform or instantiate is causally relevant to anyone else's in this magical way.

Then there are other dubious ideas in circulation, like "acausal trade" and its generalizations. I get the impression, therefore, that certain parties may be hoping for a grand synthesis which accommodates and justified timeless ontology, superrationality (and even democracy?!), acausal interaction between possible worlds, and one-boxing on Newcomb's problem. The last of these items is the only one I take seriously (democracy may or may not be worth it, but you certainly don't need a new fundamental decision theory to explain why people vote), and the grand synthesis looks more like a grand trainwreck to me. Maybe I'm wrong about what's happening in TDT-land, but I thought I'd better speak up.