[Link] Truth-telling is aggression in zero-sum frames (Jessica Taylor)

post by ioannes (ioannes_shade) · 2019-09-11T05:53:59.188Z · LW · GW · 2 comments

https://unstableontology.com/2019/09/10/truth-telling-is-aggression-in-zero-sum-frames/ (archive)

Excerpt:


If we adopt a frame that says that unusual social plots are actions that are against someone (which is a zero-sum frame), this leads to the conclusion that truth-telling is aggression, as it is necessarily part of an unusual social plot.
Non-zero-sum frames, of course, usually interpret truth-telling positively: it contributes to a shared information commons, which helps just about everyone, with few exceptions. People are often capable of switching to non-zero-sum frames in natural emergency situations, but such situations are rare.
To transition from a zero-sum frame to a non-zero-sum frame, from normalized lying to normalized truth-telling, requires a special social plot involving unusual truth-telling. Because it almost never happens by default.

2 comments

Comments sorted by top scores.

comment by Wei Dai (Wei_Dai) · 2019-09-11T11:39:16.344Z · LW(p) · GW(p)

I think Jessica is either wrong or making a leap of logic that I can't follow. My perspective here is that there are usually good game-theoretic reasons for not telling the truth that don't seem to have anything to do with "zero-sum frames". "Zero-sum frame" seems to be a wrong diagnosis for why people don't tell the truth, and therefore Jessica's prescription of moving to a "non-zero-sum frame" isn't likely to work. To show this, I'll analyze the two examples that Jessica gives of possible truth telling not being done:

  1. A waiter at a restaurant saying "I'm embarrassed to work here." - Say there are two types of waiters, those who are embarrassed to work at a restaurant and those who are not. It reflects badly on a person to work a job that they're embarrassed about (others could infer that they can't find another job, whereas someone who isn't embarrassed might just like that particular job), so there's no incentive for the "embarrassed" type to distinguish themselves from the "non-embarrassed" type. The "non-embarrassed" type does have an incentive to distinguish themselves, but can't, because if they started saying "I'm not embarrassed to work here" then the "embarrassed" type could just lie and say the same thing. So the equilibrium is that nobody says anything.

  2. A person on a date telling the other person that they're unattractive. - This one is less obvious because if we model this as a one-shot game, there's no apparent incentive to avoid telling the other person that they're unattractive, so you can end the date early and avoid wasting time. I think the incentive comes from the non-one-shot nature of this (and virtually every other) game. If you end the date early, it hurts the other person's interests (not to mention feelings) because word might spread that you ended the date early which lets other people infer that you probably found them unattractive which reduces their social status. The problem for you is that it also lets others infer that you're willing to hurt the interests and feelings of people you interact with for your own (immediate) benefit, so they'd be less likely to want to interact with you in the future.

comment by cousin_it · 2019-09-11T09:03:23.604Z · LW(p) · GW(p)

The usual story is that couching truth in politeness is positive-sum - it helps everyone, compared to a world where everyone's truth thorns are out all the time. Jessica does take a passing shot at that story ("non-zero-sum frames, of course, usually interpret truth-telling positively"), but it doesn't seem conclusive to me.