Time discounting, altruism and identity by degree
Inside baseball alert: This post may be a bit technical/dry for some readers taste. On the other hand, this might be an interesting introduction for non-philosophers to how philosophers think about things like this (not that I am an especially good exemplar of a philosopher).
A lot of people think that pure time discounting- (regarding your own future interests as less important than your present interests is simply because they are in the future), as, even if not strictly irrational, at least unreasonable in some sense(1). The parable of the grasshopper and the ants, which suggests that the grasshopper is a fool, can be read as a parable against pure time discounting.
I want to argue that if pure time discounting is indeed unreasonable, we might have some grounds for thinking that anything less than perfect altruism to every other person alive -weighing their interests as heavily as our own- is also unreasonable. The argument that leads us to this conclusion is, to be honest, not especially strong, but it is still interesting regardless. Sometimes what is interesting about an argument is not its conclusion alone, but picking which premise or bit of reasoning you disagree with.
First of all, consider the question of what makes you tomorrow the same person as you today. Greatly simplifying many results from studies of the metaphysics of personal identity, there are good reasons to think that personal identity is a matter of degree, and that you tomorrow is only partially the same person as you today.
But the people around you are quite a bit like you in many respects. They share biology, experiences and culture. You are both interpenetrated by the same memeplexes. You have a high degree of similarity with them.
Consider Trayvon, Trayvon is reflecting on the needs and interests of himself tomorrow (we will call Trayvon-tomorrow Trayvon2). He is also reflecting on the needs and interests of Julie. It seems that the gap between Trayvon1 and Trayvon2 is smaller than the gap between Trayvon1 and Julie, but it is difficult to identify a difference of kind, rather than quantity, between these gaps.
Thus if being less than 100% similar to Trayvon2 is not a reason for Trayvon1 to discount Trayvon2’s interests, why should being (much) less than 100% similar to Julie be a reason to discount her interests?
Suppose that Trayvon1 is 60% similar to Julie and 99% similar to Trayvon2, if this doesn’t give Trayvon1 a reason to even slightly discount Trayvon2’s interests, why does it give a reason to discount Julie’s interests? Where is the magic crossing point? 95%? 90%? The supposition here is that if warranted altruism is to decline as similarity decreases, it should decline smoothly, not stay at 100% then suddenly start declining past some arbitrary threshold.
What I love about this argument is that it can be read in so many ways, depending on whether we reject one of its many (implicit) premises and bits of reasoning or accept its conclusion. Among many other ways we could interpret it:
A defense of the idea that pure time discounting is reasonable based on reductio ad absurdum of the opposite. OR
A reductio ad absurdum of the idea that personal identity is a matter of degree. OR
A reductio ad absurdum of the idea that our obligations to our future self are based on our degree of similarity with them OR
A defense of using an arbitrary similarity threshold, past which we fully weight the interests of the future person as important as our own, based, again, on reductio. OR
A defense of perfect altruism.
Footnotes: (1)Note for philosophers: Doing a bit of Googling before I hit publish, I would note that Shane Fredrick (2003) presents similar considerations about the relationship between time discounting and personal identity, though he develops things in a different direction. Fredrick notes that Parfit developed something like the view that pure time preference is made rational by diminishing similarity. Kieran reminds me to comment that Hare, Harsyani and many others have developed the argument that there is a parallel between interpersonal and intertemporal concern and comparison. The small unique contribution I am trying to make here is simply that there is something of an inconsistent triad between A) the popular view that intertemporal pure discounting should be 0. B) the popular view that we should weigh our own interests at least somewhat more than those of others and C) the popular, broadly Parfitian view of personal identity as a matter of degree etc. etc.
As always, I’m just playing around with ideas here, and make no claim to be an expert on this area of philosophy.
Doesn't an obligation to your future self arise naturally, precisely because of the high degree of similarity to them (i.e. you expect them to have similar values, so you have instrumental reasons to value their existence/wellbeing/power)? This seems consistent with all of A, B and C at once.
A test case for similarity as a basis for time discounting would be if you expect yourself to become less like current you, then go back to being more like current you.
Dreaming may be a case like this! The me that is asleep is much less like current waking me than other waking mes are, and I do feel that I probably care about dreaming me less. Not zero, but even granting that dream-harms aren't real, and that I'm less likely to remember them, I'd rather my dreaming self experience a harmless pain than my waking self. But there are other confounds there, of course.