In practice, I think those with person-affecting views should refuse moves like trade 1 if they âexpectâ to subsequently make moves like trade 2, because World 1 â„ World 3*. This would depend on the particulars of the numbers, credences and views involved, though.
EDIT: Lukas discussed and illustrated this earlier here.
In practice, I think those with person-affecting views should refuse moves like trade 1 if they âexpectâ to subsequently make moves like trade 2, because World 1 > World 3.
You can either have a local decision rule that doesnât take into account future actions (and so excludes this sort of reasoning), or you can have a global decision rule that selects an entire policy at once. I was talking about the local kind.
You could have a global decision rule that compares worlds and ignores happy people who will only exist in some of the worlds. In that case Iâd refer you to Chapter 4 of On the Overwhelming Importance of Shaping the Far Future.
EDIT: Added as an FAQ.
(Nitpick: Under the view I laid out World 1 is not better than World 3? Youâre indifferent between the two.)
You can either have a local decision rule that doesnât take into account future actions (and so excludes this sort of reasoning), or you can have a global decision rule that selects an entire policy at once. I was talking about the local kind.
Thanks, itâs helpful to make this distinction explicit.
Arenât such local decision rules generally vulnerable to Dutch book arguments, though? I suppose PAVs with local decision rules are vulnerable to Dutch books even when the future options are fixed (or otherwise donât depend on past choices or outcomes), whereas EU maximization with a bounded utility function isnât.
I donât think anyone should aim towards a local decision rule as an ideal, though, so thereâs an important question of whether your Dutch book argument undermines person-affecting views much at all relative to alternatives. Local decision rules will undweight option value, value of information, investments for the future, and basic things we need to do survive. Weâd underinvest in research, and individuals would underinvest in their own education. Many people wouldnât work, since they only do it for their future purchases. Acquiring food and eating it are separate actions, too.
(Of course, this also cuts against the problems for unbounded utility functions I mentioned.)
You could have a global decision rule that compares worlds and ignores happy people who will only exist in some of the worlds. In that case Iâd refer you to Chapter 4 of On the Overwhelming Importance of Shaping the Far Future.
Iâm guessing you mean this is a bad decision rule, and Iâd agree. I discuss some alternatives (or directions) I find more promising here.
(Nitpick: Under the view I laid out World 1 is not better than World 3? Youâre indifferent between the two.)
I donât think anyone should aim towards a local decision rule as an ideal, though, so thereâs an important question of whether your Dutch book argument undermines person-affecting views much at all relative to alternatives. Local decision rules will undweight option value, value of information, investments for the future, and basic things we need to do survive.
I think itâs worth separating:
How to evaluate outcomes
How to make decisions under uncertainty
How to make decisions over time
The argument in this post is just about (1). Admittedly Iâve illustrated it with a sequence of trades (which seems more like (3)) but the underlying principle is just that of transitivity which is squarely within (1). When thinking about (1) Iâm often bracketing out (2) and (3), and similarly when I think about (2) or (3) I often ignore (1) by assuming thereâs some utility function that evaluates outcomes for me. So Iâm not saying âyou should make decisions using a local rule that ignores things like information valueâ; Iâm more saying âwhen thinking about (1) it is often a helpful simplifying assumption to consider local rules and see how they performâ.
Itâs plausible that an effective theory will actually need to think about these areas simultaneouslyâin particular, I feel somewhat compelled by arguments from (2) that you need to have a bounded mechanism for (1), which is mixing those two areas together. But I think weâre still at the stage where it makes sense to think about these things separately, especially for basic arguments when getting up to speed (which is the sort of post I was trying to write).
Do you think the Dutch book still has similar normative force if the person-affecting view is transitive within option sets, but violates IIA? I think such views are more plausible than intransitive ones, and any intransitive view can be turned into a transitive one that violates IIA using voting methods like beatpath/âSchulze. With an intransitive view, Iâd say you havenât finished evaluating the options if you only make the pairwise comparisons.
The options involved might look the same, but now you have to really assume youâre changing which options are actually available over time, which, under one interpretation of an IIA-violating view, fails to respect the viewâs assumptions about how to evaluate options: the options or outcomes available will just be what they end up being, and their value will depend on which are available. Maybe this doesnât make sense, because counterfactuals arenât actual?
Against an intransitive view, itâs just not clear which option to choose, and we can imagine deliberating from World 1 to World 1 minus $0.98 following the Dutch book argument if weâre unlucky about the order in which we consider the options.
In practice, I think those with person-affecting views should refuse moves like trade 1 if they âexpectâ to subsequently make moves like trade 2, because World 1 â„ World 3*. This would depend on the particulars of the numbers, credences and views involved, though.
EDIT: Lukas discussed and illustrated this earlier here.
*EDIT2: replaced > with â„.
You can either have a local decision rule that doesnât take into account future actions (and so excludes this sort of reasoning), or you can have a global decision rule that selects an entire policy at once. I was talking about the local kind.
You could have a global decision rule that compares worlds and ignores happy people who will only exist in some of the worlds. In that case Iâd refer you to Chapter 4 of On the Overwhelming Importance of Shaping the Far Future.
EDIT: Added as an FAQ.
(Nitpick: Under the view I laid out World 1 is not better than World 3? Youâre indifferent between the two.)
Thanks, itâs helpful to make this distinction explicit.
Arenât such local decision rules generally vulnerable to Dutch book arguments, though? I suppose PAVs with local decision rules are vulnerable to Dutch books even when the future options are fixed (or otherwise donât depend on past choices or outcomes), whereas EU maximization with a bounded utility function isnât.
I donât think anyone should aim towards a local decision rule as an ideal, though, so thereâs an important question of whether your Dutch book argument undermines person-affecting views much at all relative to alternatives. Local decision rules will undweight option value, value of information, investments for the future, and basic things we need to do survive. Weâd underinvest in research, and individuals would underinvest in their own education. Many people wouldnât work, since they only do it for their future purchases. Acquiring food and eating it are separate actions, too.
(Of course, this also cuts against the problems for unbounded utility functions I mentioned.)
Iâm guessing you mean this is a bad decision rule, and Iâd agree. I discuss some alternatives (or directions) I find more promising here.
Woops, fixed.
I think itâs worth separating:
How to evaluate outcomes
How to make decisions under uncertainty
How to make decisions over time
The argument in this post is just about (1). Admittedly Iâve illustrated it with a sequence of trades (which seems more like (3)) but the underlying principle is just that of transitivity which is squarely within (1). When thinking about (1) Iâm often bracketing out (2) and (3), and similarly when I think about (2) or (3) I often ignore (1) by assuming thereâs some utility function that evaluates outcomes for me. So Iâm not saying âyou should make decisions using a local rule that ignores things like information valueâ; Iâm more saying âwhen thinking about (1) it is often a helpful simplifying assumption to consider local rules and see how they performâ.
Itâs plausible that an effective theory will actually need to think about these areas simultaneouslyâin particular, I feel somewhat compelled by arguments from (2) that you need to have a bounded mechanism for (1), which is mixing those two areas together. But I think weâre still at the stage where it makes sense to think about these things separately, especially for basic arguments when getting up to speed (which is the sort of post I was trying to write).
Do you think the Dutch book still has similar normative force if the person-affecting view is transitive within option sets, but violates IIA? I think such views are more plausible than intransitive ones, and any intransitive view can be turned into a transitive one that violates IIA using voting methods like beatpath/âSchulze. With an intransitive view, Iâd say you havenât finished evaluating the options if you only make the pairwise comparisons.
The options involved might look the same, but now you have to really assume youâre changing which options are actually available over time, which, under one interpretation of an IIA-violating view, fails to respect the viewâs assumptions about how to evaluate options: the options or outcomes available will just be what they end up being, and their value will depend on which are available. Maybe this doesnât make sense, because counterfactuals arenât actual?
Against an intransitive view, itâs just not clear which option to choose, and we can imagine deliberating from World 1 to World 1 minus $0.98 following the Dutch book argument if weâre unlucky about the order in which we consider the options.