Regarding the fussiness of different population-ethical theories, I will propose a not-fully-worked-out alternative to your bounded views, which seems more plausible to me. (Though overall linear unbounded seems most likely to me.)
The motivating intuition behind bounded views, I believe, is to reject (extreme) scope sensitivity, and to say that beyond some point, just more of the same good things has diminishing value. (But probably more of the same bad things does not diminish in disvalue, so there is some asymmetry.) One natural way to operationalise this would be to define the total value of the world as the average welfare of an individual, multiplied by some concave function of the number of individuals (but where this function is linear below 0). This is a joint view,[1] but avoids scale tipping because adding one more happy galaxy on top of countless others will barely change either the average welfare or the number of people. If the concave function has a sufficiently high ceiling, this just reduces to totalism (as average times number equals total) but with lower ceilings, this could capture the common-sense intuition against scope-sensitivity.
What do you think? Iâm conscious there are impossibility theorems here, so no doubt my proposed theory would have some very counter-intuitive conclusions.
Aggregating goods and bads separately seems quite implausible to me, for starters it feels hard to say whether the unit of aggregation should be moments, or whole lives, or galaxies or something else.
This has been proposed in the philosophy literature! Itâs the simplest sort of âvariable-valueâ view, and was originally proposed by Yew-Kwang Ng. (Although you add linearity for negative worlds.)
I think youâre right that it avoids scale-tipping, which is neat.
Beyond that, Iâm not sure how your proposal differs much from joint-aggregation bounded views that we discuss in the paper?
Various issues with it: - Needs to be a âdifference-makingâ view, otherwise is linear in practice - Violates separability - EV of near-term extinction, on this view, probably becomes very positive
Good point, those seem like important weaknesses of the view (and this is partly why I favour totalism). And good to know re Yew-Kwang Ng. Yes, it is a version of your joint-aggregation bounded viewâmy main point was that it seemed like scale-tipping was one of your main objections and this circumvents that, but yes there are other problems with it as you note!
Regarding the fussiness of different population-ethical theories, I will propose a not-fully-worked-out alternative to your bounded views, which seems more plausible to me. (Though overall linear unbounded seems most likely to me.)
The motivating intuition behind bounded views, I believe, is to reject (extreme) scope sensitivity, and to say that beyond some point, just more of the same good things has diminishing value. (But probably more of the same bad things does not diminish in disvalue, so there is some asymmetry.) One natural way to operationalise this would be to define the total value of the world as the average welfare of an individual, multiplied by some concave function of the number of individuals (but where this function is linear below 0). This is a joint view,[1] but avoids scale tipping because adding one more happy galaxy on top of countless others will barely change either the average welfare or the number of people. If the concave function has a sufficiently high ceiling, this just reduces to totalism (as average times number equals total) but with lower ceilings, this could capture the common-sense intuition against scope-sensitivity.
What do you think? Iâm conscious there are impossibility theorems here, so no doubt my proposed theory would have some very counter-intuitive conclusions.
Aggregating goods and bads separately seems quite implausible to me, for starters it feels hard to say whether the unit of aggregation should be moments, or whole lives, or galaxies or something else.
This has been proposed in the philosophy literature! Itâs the simplest sort of âvariable-valueâ view, and was originally proposed by Yew-Kwang Ng. (Although you add linearity for negative worlds.)
I think youâre right that it avoids scale-tipping, which is neat.
Beyond that, Iâm not sure how your proposal differs much from joint-aggregation bounded views that we discuss in the paper?
Various issues with it:
- Needs to be a âdifference-makingâ view, otherwise is linear in practice
- Violates separability
- EV of near-term extinction, on this view, probably becomes very positive
Good point, those seem like important weaknesses of the view (and this is partly why I favour totalism). And good to know re Yew-Kwang Ng. Yes, it is a version of your joint-aggregation bounded viewâmy main point was that it seemed like scale-tipping was one of your main objections and this circumvents that, but yes there are other problems with it as you note!