On the Parfit point: I think he is correct, but than we simply are forced to make comparisons between potentially existing persons. We get into the question of what people we want to exist. Peter Hammond argues that interpersonal comparisons of well-being are basically preferences for diferent kinds of people, so this may not be that troubling in a utilitarian framework.

I’ve always thought utilitarianism (and any policy based on treating everyone equally) makes sense as a basis for public policy, and not as a plausible theory of individual ethics. If I read the early utilitarians correctly, they took the same view.

Charlie, you’re spot-on.

… since generations overlap, if, at all times, we treat all people now alive as equal then we must treat all people now and in the future as equal.

this could be rephrased as:

… since generations overlap, if, at all times, we treat all people now alive as equal this will have the attendant effect of treating unborn future generations as equal

I’m not sure it’s so much personal obsessions as subcultural ones; this sounds like a collection of biases specifically associated with a certain type of science-fiction fandom. Forty or fifty years ago, it’d have been General Semantics, the Dean Drive and the Hieronymus Machine.

Speaking of science fiction, though, I’d be more willing to go along with John’s proposal if I didn’t worry that it might obligate me to completely disregard the welfare of people today, or for that matter the whole human species in 2100, in favor of the putative welfare of ten quadrillion people living in the year 29,465,783, because there are potentially so many more of them. Maybe the fact that I don’t have the foggiest idea how to help them helps ethics from devolving into complete madness. But I can’t help thinking that I’d be pretty miserable going around with the knowledge that the only reason I ought to give a damn about my daughter is that I don’t know enough to fulfill my far truer moral obligation to generations millions of years in the future who may or may not exist regardless.

Maybe this is just another manifestation of fundamental biological limits on utilitiarianism. But it does seem to me sometimes that we need some consequentialist analog of the regularization/renormalization that keeps divergent integrals from blowing up results in particle physics.

Considering diminishing birth rates in the developed world, it seems that there is some kind of a mechanism on the procreation side, but the drive to consume in the dominant strain seems virtually unrestricted, which is indeed the cause of its dominance.

Environment-minded balanced civilizations (like North-American indians) get wiped out and disappear, the splendid blonde beast prospers.

“Our obligation to other people is contingent on their actually existing – note that the overlapping generations point makes the idea of separate current and future populations untenbale.”

If we are only concerned with the future consumption (as economist suggest) of people that are alive today then we have a clear moral duty NOT TO DO ANYTHING ABOUT CLIMATE CHANGE. The time lag between current efforts and future benefits (measured in benefits to consumption compared with the do nothing scenario) is too far off. Those making policy today best secure the future consumption of people alive today by not taking costly measures to address climate change.

Yet we can see the obvious implications of this choice structure. This means that we are valuing the future consumption of people alive today as opposed to the future consumption of people that are not alive now but will be alive in the future. Already here we are making a moral choice, and this choice is not dissolved by the fact of overlapping generations. We can see this easily when we consider the fact that each generation will, if they are only concerned with the future consumption of people alive at that time, make the same choice not to do anything about climate change. They may wish that past decision makers had chosen to address climate change but the best strategy for securing the future consumption of people alive at any given time will be to continue to do nothing. This is because any efforts they make will at best only give marginal improvements on climate (i.e. marginal improvements on the effects of climate change on consumption), and as a result avoiding the costs of mitigation will give much more in terms of consumption to the young of any given generation.

Again we are well aware on this feature of the climate change problem, and when we implement a policy based on the principle that “obligation to other people is contingent on their actually existing” we make a clear moral choice that the predictable pollution of the atmosphere and the effects on people over the long term is not morally important. So the fact of overlapping generations in no way allows us to avoid questions about what we owe to people in the middle to distant future.

(For an example of a study on the impacts on consumption levels from mitigating climate that simply compares consumption levels at various points in time between the mitigation and do nothing scenarios (i.e. a study that avoids controversial claims about how to weigh welfare between generations and the appropriate discount rate) see, Y. Okan Kavuncu and Shawn D. Knabb, “Stabilizing Greenhouse Gas Emissions: Assessing the Intergenerational Costs and Benefits of the Kyoto Protocol,” Energy Economics 27, no. 3 (2005).)

End of story on this one for me is that if you think the ethically proper discount rate is zero the fact that individuals have positive time preferance over their lives doesn’t play. Even if you think that we, as the present generation, have the right to discount future generation’s welfare, it (mostly) still doesn’t change anything. At most you’re still gonna get the same rate of time preferance you started out with. So while I may disagree about what the proper social discount rate is, this particular line of argument is a dead end. I spend a day thinking about it just to come to the conclusion that it doesn’t matter. Sometimes we come to know more when we learn of what we don’t know.

Of course social welfare comparisons when you have a (potentially) infinite number of agents are a mess and so something’s gotta give. You always run into the Diamond paradox Michael referred to above. But those setups are general enough so that you’re always between the Scylla and Harybdis. The Scylla eats you by not letting you say anything definite without making lots of crazy assumptions. The Harybdis eases you in with seeming plausibility and then makes you choke up crazy conclusions.

The only value in the analysis is to realize that “Hey! If I swim faster I will move closer to the other side where…the other monster is waiting…wait…swimming faster or slower has no impact on the final outcome. How did I get here in the first place again?”

And I have to say that re-reading some papers from the 1950’s and 1960’s Econometrica (which I did on this joyous occasion) is a very humbling experience. Even folks I have no agreement with like Lange, Tinbergen and Ronescu … let’s just say I’d hate to disagree with them and have them find out about it. Perhaps the present generation needs to suck more surplus out of the previous ones before we start worrying about how to screw over the next 7.

On the other hand Odysseus did make a choice and, though costly, he made it through. Take your chances with Scylla friends.

Oddly, some of the transhumanist types I run into combine several seemingly-contradictory attitudes: 1.) there’s no real environmental problem, technological progress will take care of it; 2.) the world is hopelessly fucked up by politics, and there’s really no home; 3.) bulding homes in space to escape to really shouldn’t be that hard.

What I find odd is tech optimism combined with utter political fatalism, which is usually also combined with global warming denial and an intense hatred of environmentalists. This mix comes from smart, hyper-rational people, but it sometimes seems like a Rorschach test of personal obsessions.

Yes, but the year 2100, when people who we know and care about today may still be alive, is close enough to worry about.

I’m not trying to be snarky; I’m just trying to use an extreme example as an illustration. Overlapping or not, there’s some point far enough in the future that even you don’t care about it yet, isn’t there?

Ken, I don’t think I ever had any substantive disagreement with John E. It’s just that he keeps claiming (as in #8 above) that economists can’t get questions like this right.