Bryan Caplan  

Robin's Wishful Thinking

Why Do Politicians Break Their... The Problem with Democracy...
I usually think of my colleague Robin Hanson as a hard-headed, "just-the-facts-ma'am" kind of guy.  But here's just the latest example I've found of Robin's wishful thinking:
In our culture we are supposed to oppose ordinary bloody war, preferring peace when possible there. But we do not generalize this lesson much to other sorts of  conflicts.  We celebrate those who take sides and win far more than we do peacemakers and compromisers.  But the principle is the same; every side can expect to get more of what it wants from compromise deals than from all out conflict.
My question: What makes Robin think that "every side can expect to get more" from compromise than conflict?  Doesn't anyone have a comparative advantage in conflict?  And all it takes to get a conflict is one willing combatant, no?

Maybe Robin is hiding behind the word "expect."  But I'll reply, "Doesn't anyone expect to have a comparative advantage in conflict?"  Maybe he's hiding behind the word "side" - individuals may gain, but not entire factions.  But can't we just view trouble-makers as being "on their own side"?

Robin's problem, in my view, is that he is needlessly trying to be "all things to all men."  But it's not possible.  Even the promotion of peace and compromise invites conflict with the practioners of war and total victory.  Despite Robin's hopes, he can't be neutral.  No one can.  Instead of vainly trying to be neutral, why not just try to be right?

Comments and Sharing

COMMENTS (12 to date)
needle writes:

yeah, been thinking for some time now that we should just drop the pretense of being a bunch of peace loving "nice guys "and embrace our warrior mentality.we americans have always been a warlike people. we should accept that fact and make the best of it. this "namby-pamby" "treat others well and they'll return our love" stuff is a bunch of new age psycho-babble for only gets you laughed at and treated like a punk.lets not only flex our muscle, lets crack some heads and not say we are sorry. it would be nice to see somebody else for once ask for forgiveness and pay dearly to have us stop bringing the pain.that's my wishful thinking :)

Giedrius writes:


you expect to die and Robin expects to live in the society of uploads. The things tailless apes consider right might not be very popular in his future world. So he has a reason at least to try to be neutral.

Stephen S writes:

Re: Giedrius,

Does he really expect to live in the society of uploads? Isn't an upload just a copy? It would do little to assuage my fear of death to know an eigen-self would live on :

JP writes:

Sounds like the prisoner's dilemma. A and B will both get more if they both act cooperatively than if they both act adversarially. But one of them will get more if he acts adversarially while the other acts cooperatively, while the cooperative one will now get less than if both acted adversarially. So unless they know each other or can make credible commitments to cooperation, it's safer for them to act adversarially.

Rob writes:

If we assume:
1) resources are consumed (without contributing to happiness) in a conflict.
2) no participants are overconfident regarding their expected gain from conflict.

then I would think there is always some compromise where everyone gets exactly their expected gain from conflict, plus a share of the resources that would have been used in the conflict. This seems to make Robin's claim true, regardless of what comparative advantage participants have in conflict.

I would guess that most conflicts are caused by 2 being false (that is, one or more participants being overconfident).

The Snob writes:


I think there is another piece, which is that knowledge of each others' motives is very important. If you don't have a decent sense of what it is that the other side is fighting for, then it can be difficult to figure out how and to whom to deliver your surrender.

The one instance in the past ten years that I've had a business dispute go all the way to the courthouse happened largely because we both thought the other side wanted to go after our crown jewels. In fact our goals were almost entirely separate and non-conflicting. Had we figured that out sooner we both would have saved a lot of money.

Rob writes:

The Snob: Good point, I guess I was also supposing all participants are willing to be honest about their goals in diplomacy.

Also, conflict may be necessary if two participants are fighting over one resource that cannot be divided up, so no situation can be mutually beneficial.

Greg writes:

Doesn't this just go back to game theory 101? Of course conflict is a dominant strategy in some situations.

Apart from that, sometimes the spoils really do go to the winner, far more than they would under a compromise. What would a compromise between the settlers and native Americans in North America have looked like? For the settlers to get the same share of the pie, it would have had to be one very lopsided deal.

Rob writes:

Greg: Do you mean conflict is dominant in something like the prisoner's dilemma? Robin says "compromise deals" so I assumed we are looking at situations where diplomacy is possible and deal contracts enforceable. Do you have examples where these hold and conflict is still dominant?

Being on the losing side of a very lopsided compromise should still be preferable to being on the losing side of a lopsided conflict. The lopsided compromise may seem unfair, but in general so will conflict. The native Americans would have been better off agreeing to give up most of their land to the settlers, if they were guaranteed they could keep the remainder without conflict.

peco writes:

No, this is different. One group may have a higher expected return from (D, D) than from (C, C) if it has a large advantage in the conflict and the return from winning a conflict is large. In this case, (C, C) would not be the optimal result for both groups, and the groups would not try to achieve cooperation.

Many other things can also cause conflict. If the resources required from to reach a compromise are greater than the resources that the same side would use up in a conflict, the conflict would happen (but then the conflict should really be called the compromise and vice versa because compromise should be defined to be the less costly situation). If one side is a (large?) group that faces a very high expected loss, it may not be politically possible for them to avoid conflict. If all the costs of the conflict are inflicted on one side, the other side has no incentive to compromise because a conflict would cost them nothing (this is not very common, but if one side only incurs a small cost from conflict then the small cost from negotiating a compromise may be equal to it, giving effectively the same result). This is really just a special case of the first situation, though.

Robin Hanson writes:

I added a response to my post.

Steve Roth writes:

Bryan, I think the problem arises in the absolute nature of the statement. It should read:

"*In many situations,* every side can expect to get more of what it wants from compromise deals than from all out conflict."

That is undeniably true--just ask any divorce lawyer.

They'll also tell you that humans quite frequently treat win-win situations--even stunningly obvious ones--as if they were zero-sum. (My explanation: foolish pride.)

Some experiments in the 50s and 60s demonstrated this in spades:

Comments for this entry have been closed
Return to top