Why take the risk of this escalating into a seriously negative sum outcome?
Because if someone does that to you (walking up to you and insulting you to your face, apropos of nothing), then the value to them of the situation’s outcome no longer matters to you—or shouldn’t, anyway; this person does not deserve that, not by a long shot.
And so the “negative sum” consideration is irrelevant. There’s no “sum” to consider, only the value to you. And to you, the situation is already strongly negative.
Now, it is possible that you could make it more negative, for yourself. Possible… but not likely. And even if you do, if you simultaneously succeed at making it much more negative for your opponent, you have nonetheless improved your own outcome (for what I hope are obvious game-theoretic reasons).
Aren’t you, after all, simply asking “why retaliate against attacks, even when doing so is not required to stop that particular attack”? All I can say to that is “read Schelling”…
It would have very likely made it a lot more negative to myself. I expect that would have escalated the situation and would have prolonged the whole thing for at least twice as long.
Generally I am in favor of punishing things like this, but it seems much better for that punishment to not happen via conflict escalation, but via other more systematic ways of punishment.
Because if someone does that to you (walking up to you and insulting you to your face, apropos of nothing), then the value to them of the situation’s outcome no longer matters to you—or shouldn’t, anyway; this person does not deserve that, not by a long shot.
I disagree with the morality that is implied by this statement. (It’s a really small part of my moral parliament.)
And to you, the situation is already strongly negative.
There may be some value difference here between you and I, in that you may consider being insulted to have a strongly negative terminal value?
Now, it is possible that you could make it more negative, for yourself. Possible… but not likely.
If the situation escalated, I might get injured or arrested or be less likely to be invited to future parties, or that person might develop a vendetta against me and cause more substantial harm to me in the future, all of which seem potentially a lot more negative.
And even if you do, if you simultaneously succeed at making it much more negative for your opponent, you have nonetheless improved your own outcome (for what I hope are obvious game-theoretic reasons).
I already addressed this when I wrote “I can imagine the risk being worth it for someone who needs to constantly show others that they can “handle themselves” and won’t easily back down from perceived threats and slights. But presumably habryka is not in that kind of social circumstances, so I don’t understand your reasoning here.”
Aren’t you, after all, simply asking “why retaliate against attacks, even when doing so is not required to stop that particular attack”? All I can say to that is “read Schelling”…
No, I was saying that cost-benefit doesn’t seem to favor retaliating against that particular attack, in the particular way that cousin_it suggested. (Clearly it’s true that some attacks should not be retaliated against, right?)
>> Why take the risk of this escalating into a seriously negative sum outcome?
Because if someone does that to you (walking up to you and insulting you to your face, apropos of nothing), then the value to them of the situation’s outcome no longer matters to you—or shouldn’t, anyway; this person does not deserve that, not by a long shot.
I think I’m with Wei_Dai on this one—insulting me to my face, apropos of nothing, doesn’t change my valuation of them very much. I don’t know the reasons for such, but I presume it’s based on fear or pain and I deeply sympathize with those reasons for unpleasant, unreasoning actions. Part of my reaction is that it’s VERY DIFFICULT to insult me in any way that I won’t just laugh at the absurdity, unless you actually know me and are targeting my personal insecurities.
Only if it’s _NOT_ random and apropos of nothing am I likely to feel that there are strategic advantages to taking a risk now to prevent future occurrences (per your Schelling reference).
Because if someone does that to you (walking up to you and insulting you to your face, apropos of nothing), then the value to them of the situation’s outcome no longer matters to you—or shouldn’t, anyway; this person does not deserve that, not by a long shot.
And so the “negative sum” consideration is irrelevant. There’s no “sum” to consider, only the value to you. And to you, the situation is already strongly negative.
Now, it is possible that you could make it more negative, for yourself. Possible… but not likely. And even if you do, if you simultaneously succeed at making it much more negative for your opponent, you have nonetheless improved your own outcome (for what I hope are obvious game-theoretic reasons).
Aren’t you, after all, simply asking “why retaliate against attacks, even when doing so is not required to stop that particular attack”? All I can say to that is “read Schelling”…
It would have very likely made it a lot more negative to myself. I expect that would have escalated the situation and would have prolonged the whole thing for at least twice as long.
Generally I am in favor of punishing things like this, but it seems much better for that punishment to not happen via conflict escalation, but via other more systematic ways of punishment.
I disagree with the morality that is implied by this statement. (It’s a really small part of my moral parliament.)
There may be some value difference here between you and I, in that you may consider being insulted to have a strongly negative terminal value?
If the situation escalated, I might get injured or arrested or be less likely to be invited to future parties, or that person might develop a vendetta against me and cause more substantial harm to me in the future, all of which seem potentially a lot more negative.
I already addressed this when I wrote “I can imagine the risk being worth it for someone who needs to constantly show others that they can “handle themselves” and won’t easily back down from perceived threats and slights. But presumably habryka is not in that kind of social circumstances, so I don’t understand your reasoning here.”
No, I was saying that cost-benefit doesn’t seem to favor retaliating against that particular attack, in the particular way that cousin_it suggested. (Clearly it’s true that some attacks should not be retaliated against, right?)
>> Why take the risk of this escalating into a seriously negative sum outcome?
I think I’m with Wei_Dai on this one—insulting me to my face, apropos of nothing, doesn’t change my valuation of them very much. I don’t know the reasons for such, but I presume it’s based on fear or pain and I deeply sympathize with those reasons for unpleasant, unreasoning actions. Part of my reaction is that it’s VERY DIFFICULT to insult me in any way that I won’t just laugh at the absurdity, unless you actually know me and are targeting my personal insecurities.
Only if it’s _NOT_ random and apropos of nothing am I likely to feel that there are strategic advantages to taking a risk now to prevent future occurrences (per your Schelling reference).