Notes on Conflicts of Interest Scenarios (Tutoring Max 49)Date: 2020-10-18
Can all these be resolved?
- We both want the same diamond.
- We both want the same computer.
- We both want to marry the same woman.
- We both want the same slot on the manned mission to the moon.
- We both want to be President (of the same country).
- We both want to be the top commander of the army.
- I want to speak my mind but you don’t like what I have to say and would prefer I shut up.
- I want to kiss you but you don’t want to kiss me.
- I sell printers and you sell printers and we’re competing for customers.
Conflicts of interest (CoIs) seem to exist sometimes. When considering rational ppl or trying-to-be-rational ppl, those conflicts don’t actually exist–they’re illusions which can be resolved. They look like conflicts because we’re ignoring the bigger picture. Ppl involved in the CoI shouldn’t want to ‘win’ via a system which use force to get an outcome. They should want a system that’s fair and works generally. A system with universality.
Systems which use force or unwritten rules are not preferable to free-market situations b/c they have adverse consequences outside of one’s control (e.g. violence, ‘winners’ being decided by something like physical attractiveness or social status, etc). They outcomes – when decided with alternative systems – are worse for ppl involved. Reasons include: bad distribution of resources, outcomes being based on perceived problems that a person can’t solve (e.g. not handsome enough), harm being done (e.g. violence), etc.
Expansion of situation: we are both in a shop buying an engagement ring for our respective soon-to-be fiancées, and want the same diamond (diamond-A).
The initial ‘solution’ is that the shop sells diamond-A to whomever asks for it first. Person-A gets it. This is okay because both ppl can agree to a first-come-first-serve model (which is typical and expected).
Maybe person-B really wants the diamond. They can offer to buy it from person-A. This is okay because it’s consensual trade where both ppl are better off.
Say person-A says they want to buy it but hasn’t paid, but person-B has the cash now. The shop could work on a first-come-first-serve basis where the transaction is the important moment (who can pay first), so person-B gets it. this is an agreeable system.
Maybe there is another diamond (diamond-B) that one of the ppl is happy with, so person-A gets diamond-A, person-B gets diamond-B.
in each case an alternative system of distribution (based on attractive looks, or social status, or bribes, or whatever) is not preferable – it’s a worse society to live in.
Say it’s a rare old computer so there’s only one of them and it’s not fungible. We can agree on a system which is fair, like an auction, and proceed on that basis.
She should choose who she wants to be with (if either of us). We shouldn’t want to be with someone who doesn’t want to be with us (that would be bad for both me and her). We should both want her to be able to consider both of us. If I had an advantage (e.g. knew her earlier) and tried to stop her meeting you b/c I thought she’d prefer you, then it means I have to keep that effort up WRT you and any one else she might meet. So eventually I’d need to be coercive or forceful to do that. Hurting the person you want to marry is a shit thing to do (and a bad way to live long term), so I shouldn’t want to prevent her evaluating other potential partners. I should actually be in favour of that because it means problems are apparent sooner rather than later. Living in a relationship where big problems will occur and that can’t be resolved (e.g. she changes her mind about wanting to marry me) is bad for me, so if there will be problems I should want to know about them as soon as possible.
Say there are 3 crew slots and 2 crew members have been decided and are better candidates than us (at least for those slots, like the other crew have skills we don’t).
- We shouldn’t want to be chosen if that would jeopardise the mission – it being successful is more important. we can agree that the most qualified person should be chosen, or the person otherwise chosen s.t. the mission has the greatest chance of success. Maybe we’re equally qualified, though.
- We don’t want a system where one of us is harmed (e.g. I hurt your family to keep you out of the mission). If I wanted that it could mean my family (or me) is hurt, which I don’t want.
- We don’t want the mission to be jeopardised for political reasons (or other parochial stuff), so we should be in favour of a selection criteria which is publicly and politically defensible (and just).
- We don’t want a system where one of us is prevented from doing stuff in the future like other moon missions.
- We don’t want a system where NASA (or whomever) regrets their decision (e.g. because it was made via nepotism or whatever).
- We don’t want a system where we hate each-other because that could mean we can’t be on the same future mission or otherwise end up excluded from other stuff.
- We can agree on a system based on merit
- We can agree on a system where NASA maintain a suitable body of astronauts (like a minimum number of astronauts kept in reserve), so some rotation is necessary (maybe one of us went on the last mission so the other should go on this one) – We can also agree on a system which takes into account future rotations, e.g. flip a coin and one of us goes on this one, and the other goes on the next mission
- We can agree on a system that doesn’t bias one of us for external reasons like social status (if that happened, all missions would be worse off and have a lower chance of success)
operating under these sorts of systems is preferable to winning the slot under a different system. if it was some different system then how could we be confident that our crew is the best crew possible?
Note: curi and I sort of started discussing this at the end of Tutoring Max #49.
We should both be in favour of a good system for selecting a president. We can agree on important features such a system should have, like not favouring one of us. We should want a system where the victory conditions are clear and compatible with our values. We should want a system where we could lose b/c it’s possible the other person is a better choice regardless of what we believe.
The conflict only exists when we have bad, irrational systems for choosing a president. If the system is bad then we can both agree changing the system is more important (and subsequently find a system which satisfies both our goals).
If there are other candidates, we should prefer those candidates who will institute a better system to those who won’t. If there are perverse mechanics in the selection system (e.g. like those in first-past-the-post when you have 2 similar candidates running s.t. it decreases the chance of a favourable outcome) then we should both be in favour of cooperating to maximise the chance of one of us winning over bad candidates. We can find such a system.
We could also run a pre-election or something to decide which of us runs in the main election (similar to primaries in USA).
I worry that I’m missing something. Are these adequate answers? Do any of the apparent conflicts persist after what I’ve written?
I think these are hard problems to write about – in some ways – b/c there are always unknown and unspecified details which could be chosen to make the situation a ‘nightmare situation’ (as curi put it in TM#49).
Going to have a think and maybe come back to this later.
You can leave a comment anonymously. No sign up or login is required. Use a junk email if not your own; email is only for notifications—though, FYI, I will be able to see it.
Comments powered by Talkyard.