The MWI and Distributive Justice

Round 1

Reviewer 1 Report

See attached file

Comments for author File: Comments.pdf

Author Response

Thanks to this reviewer for asking us to clarify our understanding of Everettianism. We have added a paragraph at the end of our first section that explains which version of Everettianism we are assuming.

We have also corrected the error in our reference to Price 2010, and added a reference to Saunders 2010.

Reviewer 2 Report

This is a super interesting paper! It definitely should be published in this Quantum Reports issue. The paper presents an interesting set of cases which, taken together, suggest that Everettians should make surprisingly unusual choices certain sorts of goods should be distributed across certain sorts of agents.

My only comment concerns the set-up in the Everettian case. Basically, the comment is just a request for more details about what, exactly, the Everettian case -- as presented in Section 5 -- is. But let me give a bit more information about why I'm making this request.

Who, in the Everettian case, are Ann and Bob? The text refers to "Ann's successor" (p. 3) and to "Bob's" successor (p. 3). So it sounds like in the case which the authors have in mind, Ann and Bob are two distinct people, each of which has a -- unique, it seems -- successor who gets the drug. And so it sounds like there's a branching event, which happens in Ann and Bob's futures, where the drug is given to those agents' successors after that branching event.

I'm having trouble understanding that. Suppose that Ann and Bob really are two distinct people who exist prior to some branching event, and who have unique successors who get the drug. Then of course, after branching, each of Ann and Bob will have many successors, some of whom get the drug and some of whom don't. There are many ways to continue developing the details of the case, but for the purposes of illustration, let me pick this one: since everything physically possible happens on some branch or other, there will be a branch where (i) both Ann and Bob get the drug, (ii) Ann gets the drug and Bob doesn't, (iii) Bob gets the drug and Ann doesn't, and (iv) neither Ann nor Bob get the drug. Now I'm wondering: when the authors write "a 60% future on which Ann's successor gets the drug" (p. 3), do they mean that the sum of the amplitudes assigned to branches (i) and (ii) is 60%? And when the authors write "a 40% future on which Bob's [successor gets the drug]" (p. 3), do they mean that the sum of the amplitudes assigned to branches (i) and (iii) is 40%? That'd be weird because that's assigning an amplitude of 0% to branch (iv), which is physically unrealistic (given non-zero branch weights on the wavefunction, prior to measurement, for the possibilities (i)-(iv), it'll follow that branch (iv) gets some nonzero weight). That'd also be weird because, note, the authors are `double-counting' the weight being assigned to branch (i).

Anyway, I can't tell if this is the sort of situation that the authors have in mind, in Section 5, or if the authors have in mind another situation instead. Perhaps, for instance, the authors had in mind a situation where Ann and Bob are both themselves successors of some person Carol: so Ann and Bob are on two different branches in Carol's future, where the two branches are weighted by 60% and 40%, and the branches are generated by some measurement that e.g. Carol performs and that leads to her having two successors. Is that the idea instead?

At first, this is the case that I attributed to the author. But then, in this case, choosing the option which the authors call "Lottery-Mixing" (Section 5, pp. 3-4) seems a lot like making the orthodox choice. That is, choosing Lottery-Mixing_{Everett} (the subscript is there to distinguish this version of Lottery-Mixing from the version discussed in Section 4 pp. 2-3) seems entirely analogous to choosing Mixed-Weeks, and entirely unanalogous to choosing Lottery-Mixing in the Section 4 case: whereas only one person gets the drug in Lottery-Mixing (Section 4), two people get drugs in both Lottery-Mixing_{Everett} and Mixed-Weeks. So I'd have called Lottery-Mixing_{Everett} the orthodox choice; the choice, that is, recommended by the standard, orthodox view that considerations of fairness should lead us to assign extra value to options which divide drugs up between people who are suffering, even if one person is suffering more than the other.

So to summarize: it could be helpful if the authors went into detail about the case in Section 5 a bit more, so as to clarify why the above issues aren't really issues at all. As Section 5 currently stands, I'm not sure exactly what the case is. Are Ann and Bob two distinct people, prior to a branching event, who have successors? If so, then some of the authors' remarks -- about assignments of weights -- are a bit confusing (see the points about branches (i)-(iv) above). Or are Ann and Bob two distinct people, after a branching event, who are both the successors of some one person? If so, then other of the authors' remarks -- about which choices are/aren't orthodox -- are a bit confusing (see the paragraph immediately above this one).

Author Response

Much thanks to this reviewer for asking us to clarify how we are thinking of Ann and Bill in our quantum lottery. We have split the first paragraph of section 5 into two. The new first paragraph is designed to make it clear who and where Ann's and Bill's "successors" are.