Yes, my apologies for the length. Thanks for the response. I didn't even mean to post the topic; I was copying it into a note pad app and ended up submitting it by accident, but meant to go back later and edit it.
Parfit’s paper on the unimportance of identity
Thanks, I will have to check that out.
Now you’re putting an observer in there as well, and that adds more relations. The two are suddenly quite distinct. Your beacon serves the same purpose, adding more relations.
Yup, this is the point I wanted to get at. Probably should have deleted everything to this point to be concise. My basic argument would be this:
1. The prior papers on the universes only containing multiple copies of completely similar objects call into question if positing such a thing is coherent. The two spheres can be described entirely as one sphere. There would never be a way to discern between them.
2. Add in more relations, and suddenly you have a difference. Discernibility seems to spring forth from the creation of relationships.
3. However, many views posit that, depending on the outcome of the astronaut placing the beacon, we can determine the properties the example universe had
before any relationships generated discernibility. If the astronaut sees a beacon over every sphere, this implies only one sphere ever existed. If he sees it every other sphere, this implies two spheres. But the common viewpoint is that the multiple/singular sphere(s) must have existed prior to the relationship that discerns between them.
4. The problem here is that if you accept the perspective of point 3, it follows that there are potentially infinitely many differences that don't make a difference lurking behind apparent reality. You end up with a rapidly inflated ontology of unobservable differences.
I don't have a full model worked out. I merely have a hunch that this problem can be resolved formally by defining things in terms of their relationship to other things. This avoids the problem of substratum that universals/tropes have to attach to, the problem of infinite indiscernible differences, and Della Rocca's infinite number of identical objects on top of each other, in that the actual relationships that exist between things are finite.
The problem I have is that it seems like defining discrete systems or objects this way is going to have to be arbitrary, based on pragmatism. But maybe this isn't such a big deal if objects are already defined in arbitrary ways, which seems to be the case given all the problems with the concept of identity.
So here:
I’m wondering about your focus on indiscerniblity. If I create another ‘me’ in a room, facing me, they’re in theory indiscernible. No model is going to pick out a preferred one.
You can't discern which of you holds your "identity," but you can discern between there being one of you and there being two, because the relationships between your two selves are going to be different from the relationship of just your one self to your self.
To sum up: drop identity outside some pragmatic legal uses, and define discernability by relationships, rather than properties things have of themselves. Maybe this gets you a bit closer to a definition that gets at what identity tries to get at, without as much baggage. I figured someone has already made an argument like this with much more time to flesh it out, but I haven't had luck finding it.
For instance, most people presume that sort of pragmatic/memory identity you use above in all your examples, but there are quantum interpretations that destroy that sort of identity, leaving only event-identity: A thing is only identical to itself at one moment in time (and not even that since a moment in time is not unambiguously definable).
So under MWI, ‘world’ split off and in another world I have a broken leg and in this one I don’t. Both of us have an identical history of a day ago: We share the exact same person-state a day ago. So if that prior state is X, and Y is me now and Z is me with the leg issue, then if Y=X and Z=X, then Y=Z and I have and don’t have a broken leg, a contradiction. Therefore X and Y are different identities to satisfy Liebnitz’ Law. There is no persistent idenity of anything by this very non-pragmatic definition. This goes against most people’s personal intuition of having such a persistent identity, and hence considerable resistance to something like MWI, especially if you’re religious and the god needs an identity to judge in the afterlife.
Assuming MWI is wrong and this sort of splitting is fiction, how about your alien and the switched loved-one? Suppose the other person committed some crime before the abduction, and the law can prove it. Now the memories are switched. Which one do you throw in jail? The legal system isn’t set up to handle this case.
Interesting points; that's sort of how I've thought of the issues with identity vis-a-vis MWI as well. You might not need MWI to get to this sort of issue. Max Tegmark's book "Our Mathematical Universe," discusses how there is considerable support for cosmic inflation as the origin of our universe. But there is no reason to think cosmic inflation ever ends, which means we have an infinite space. However, quantum mechanics implies that there is a limit to how many ways an observable universe with our physics can exist in discernible states. There are an absolutely gigantic number of these possible states (10^10^123 is an estimate if I recall correctly), but with eternal inflation, there are guaranteed to be other identical versions of you, and some with only slight differences. These aren't in another part of the wave function, but rather just extremely far away.
But it seems that this definition you’re using is only a pragmatic one: It is useful to assign a sort of legal identity to the various states of dog, so that the various non-identical states combine into one pragmatic identity. This can be attacked, but not so easily with a complex mammal.
So for instance, suppose I ‘borrow’ a friend’s pet starfish. I cut the thing in half and both halves grow back the missing parts and now there’s two of them. I give one back to the friend, who has his pet returned. Or did I? Perhaps I returned the copy and kept the original.
This can be done with humans as well. Given a pair of identical twins., which is the original one that was first conceived, and which is the other one produced by the splitting and separation of the zygote? The pragmatic definition totally fails here, but the legal definition doesn’t much care. It might care which comes out first, but that has nothing to do with the above question.
Good examples. I wasn't trying to argue for a specific definition at this point, just trying to summarize some of the common paradoxes proposed. Is it generally taken that diachronic identity, through time, is pragmatic? The impression I've got from my reading is that people have attempted to develop
absolute solutions for identity over time.
That’s just the nature of dog. Do it with the starfish and the tail is a starfish (at least if it’s big enough). You question should not be if the tail is a dog, but if it is the same dog.
I've seen the question posed that way in various solutions. I just don't buy it. Intuitively, a tail isn't a dog, it's an incomplete part. The emergent whole is what matters. But the tipping point between a thing missing some of its parts and ceasing to exist seems like it has to be necessarily arbitrary.
Is there an easy answer here I'm missing? I took philosophy 101, but it was unfortunately just a chronological slog through "the great minds" with no topical organization.