I have some moral uncertainty regarding totalism vs. averageism, and have at times vacillated between the two. However, one of the points that has led me to increasingly favor totalism is a rebuttal to the veil of ignorance argument in favor of averageism (I.e., “wouldn’t you prefer to be a random person in [averageist paradise] vs [totalist paradise]?”).
The rebuttal is partially just that averageism arbitrarily ignores people that don’t exist in its calculations. If you can get an averageist to agree that such non-people should be considered, then it becomes mathematically obvious that the repugnant conclusion is at least less repugnant than the alternative, which is “some number of people exist and live happy lives, but some orders-of-magnitude-larger number of people just have completely net-zero existences (whereas they could have had slightly net-positive lives), completely swamping the average.” In other words, whenever a supposed averageist paradise is described, you should basically just add in what I call “grey people”: a number of purely net-zero-experience people who would have existed in the totalist alternative world.
Thus, this seems to largely defeat the “repugnant conclusion” objection: you can’t call it more repugnant than the alternative even according to the average wellbeing.
Of course, getting averageists to accept “non-people should be considered” seems to be the far trickier part. I was already somewhat open to this idea so I was somewhat probably easier to persuade than most people. However, some of the points here that persuaded me were:
We are naturally biased in favor of caring about moral ideas which validate/benefit us, and we can only examine this question if we already exist. Thus, our intuitions and feelings will probably be biased in favor of frameworks that only care about existing people.
Why shouldn’t we care about potential people? (This is especially important to pair with the previous point.) It seems like the onus should be on averageists, since totalists can at least point out “you would prefer to be a person who exists with a net-positive life to having no existence at all.” (Admittedly, there may be a flaw/circularity in this assumption, but circularities are sometimes hard to avoid in ethics (and descriptive logic), and I’m not confident it is actually problematically circular: is it really unreasonable to assume that a rational person, presented with a “never exist” button and a “live millions of extremely happy years with no suffering” button would prefer the latter? If not, there doesn’t seem to be a reason to think that changes at any point before the experience in the latter option diminishes to “live a pure net-zero-happiness life”.) One rebuttal to this which I think I often returned to (perhaps subconsciously or indirectly) when I defended averageism was “but the repugnant conclusion!”—but, as I describe above, that does not seem to be a problem if you accept the idea that potential people matter.
It seems that averageism has its own extreme ends: according to it, a world with only 1 person suffering a torturous life is worse than a world with billions of people suffering slightly-less-torturous lives on average. Of course, if one doggedly insists that averages are all that matter, then this is all consistent. But it seems hard to reject the intuition that there is some moral relevance in the aggregate experience vs. the average experience.
I have some moral uncertainty regarding totalism vs. averageism, and have at times vacillated between the two. However, one of the points that has led me to increasingly favor totalism is a rebuttal to the veil of ignorance argument in favor of averageism (I.e., “wouldn’t you prefer to be a random person in [averageist paradise] vs [totalist paradise]?”).
The rebuttal is partially just that averageism arbitrarily ignores people that don’t exist in its calculations. If you can get an averageist to agree that such non-people should be considered, then it becomes mathematically obvious that the repugnant conclusion is at least less repugnant than the alternative, which is “some number of people exist and live happy lives, but some orders-of-magnitude-larger number of people just have completely net-zero existences (whereas they could have had slightly net-positive lives), completely swamping the average.” In other words, whenever a supposed averageist paradise is described, you should basically just add in what I call “grey people”: a number of purely net-zero-experience people who would have existed in the totalist alternative world.
Thus, this seems to largely defeat the “repugnant conclusion” objection: you can’t call it more repugnant than the alternative even according to the average wellbeing.
Of course, getting averageists to accept “non-people should be considered” seems to be the far trickier part. I was already somewhat open to this idea so I was somewhat probably easier to persuade than most people. However, some of the points here that persuaded me were:
We are naturally biased in favor of caring about moral ideas which validate/benefit us, and we can only examine this question if we already exist. Thus, our intuitions and feelings will probably be biased in favor of frameworks that only care about existing people.
Why shouldn’t we care about potential people? (This is especially important to pair with the previous point.) It seems like the onus should be on averageists, since totalists can at least point out “you would prefer to be a person who exists with a net-positive life to having no existence at all.” (Admittedly, there may be a flaw/circularity in this assumption, but circularities are sometimes hard to avoid in ethics (and descriptive logic), and I’m not confident it is actually problematically circular: is it really unreasonable to assume that a rational person, presented with a “never exist” button and a “live millions of extremely happy years with no suffering” button would prefer the latter? If not, there doesn’t seem to be a reason to think that changes at any point before the experience in the latter option diminishes to “live a pure net-zero-happiness life”.) One rebuttal to this which I think I often returned to (perhaps subconsciously or indirectly) when I defended averageism was “but the repugnant conclusion!”—but, as I describe above, that does not seem to be a problem if you accept the idea that potential people matter.
It seems that averageism has its own extreme ends: according to it, a world with only 1 person suffering a torturous life is worse than a world with billions of people suffering slightly-less-torturous lives on average. Of course, if one doggedly insists that averages are all that matter, then this is all consistent. But it seems hard to reject the intuition that there is some moral relevance in the aggregate experience vs. the average experience.