Suppose I went back in time and tried to shoot my grandfather before my father was conceived. Then either I would hit or I would miss. If I hit, absurdity results. What is less discussed in the literature is that if I miss, absurdity also results. Suppose that I miss due to sloppy aiming. (This is the case most favorable to my argument. But I think a similar story can be told for other causes of missing.) Then, my sloppy aiming is explanatorily prior to my grandfather's survival. But my grandfather's survival is explanatorily prior to my existence, and hence to my sloppy aiming. Hence, we get an explanatory circle, which is absurd.
Just some thoughts. But is this really "explanatorily prior"? I suppose it is. But if your having missed your grandfather is explanatorily prior to your existing, then why wouldn't, say, your choosing not to go back in time and shoot your grandfather? Perhaps the absurdity of time travel either way is a point against the B-theory?
ReplyDeleteWhat makes an explanatory circle absurd? Does it violate any first principles of logic?
ReplyDeleteJames:
ReplyDeleteAh, so in fact, the mere causal possibility of time travel is enough to generate an explanatory circularity. Nice. So time travel can't be possible, except with pretty strong restrictions (e.g., you can travel to a spacelike separated point, but not to a point in a past light cone of anything you've visited).
The B-theory doesn't by itself say that time travel is possible.
Drew:
Well, I think it violates a first principle of reason. :-)
I agree that circular causation is unintelligible. I had others explain to me that there is nothing wrong with it, and was wondering if you had any good polemics against it.
ReplyDeleteDrew,
ReplyDeleteI think the primary issue here can be put in a more obvious way than the idea of causal circularity. What Alex is saying implies a violation of the principle of non-contradiction (a first principle if ever there was one). In the Grandfather Paradox, you both are, and are not, explanitorily prior to your own existence (hence the defiance). As Aristotle says in his "Metaphysics", "For the same characteristic simultaneously to belong and not belong to the same object in the same way is impossible."
Some naive thoughts, formed over the years from SciFi (e.g. Many Dimensions, and 12 Monkeys): If you had been in the past before you were born then you would have to go back in time to be there. What would be impossible would be not going back. If you had not been there, then it would be impossible to go back and be there. That is, time travel is only impossible if it is not necessary. If it is possible, then you may think that you have a choice about whether to go back or not, but really you don't. But it seems relatively plausible to me that we are sometimes wrong about the freedom of our decisions. And if there is an ultimate end to things which determined that their creation was good, and if earlier things often cause later things, then we do have explanatory circles (with different sorts of explanations on the different arcs).
ReplyDeleteThe B-theory doesn't by itself say that time travel is possible.
ReplyDeleteBut whereas the A-theory is incompatible with time-travel, the B-theory allows physicists to think of a particle-antiparticle generation as an antiparticle reversing direction in time by absorbing background energy, which makes it akin to normal particle interactions via exchange of energy. Given particle physics, the B-theory does effectively say that time-travel is possible, because of how natural such a description would then be.
Hi Alex,
ReplyDeleteNice post. Not to disagree, but just to note that in the case of the "future formation of closed time-like curves" in GR (which is not really dynamical, because these things are supposed to exist beyond the Cauchy horizon, so one has to put them in by hand), the objects already exist so one only has causal circularity; not a contradiction. It's interesting that in the physical case that most resembles the Grandfather Paradox, i.e. where the topology of space forces a causal curve to loop back and intersect the initial data hypersurface, no one ever brings up the fact that the initial data is both taken to be somehow "created" or "posited" and yet self-causing!
N