Pro-war bias

by John Quiggin on January 11, 2007

The fact that people are so willing to support war is a puzzle that requires an explanation. After all, war is a negative-sum activity, so war between rational parties doesn’t make sense – there’s always a potential settlement that would leave both sides better off*. And empirically, it’s usually the case that both sides end up worse off relative to both the status quo ante or to a possible peace settlement they could have secured at a point well before the end of the war. Even the observation that rulers start wars and ordinary people bear the costs doesn’t help much – leaders who start losing wars usually lose their jobs and sometimes more, while winning a war is by no means a guarantee of continued political success (ask Bush I). All of this suggests that looking for rational explanations of war, as in the ‘realist’ tradition (scare quotes indicate that this self-ascribed title has little to with a reality-based focus on the real world) is not a good starting point.

So it makes sense to look at irrational sources of support for war. In this pice in Foreign Policy Daniel Kahneman (winner of the economics Nobel a couple of years back) and Jonathan Renshon start looking at some well-known cognitive biases and find that they tend systematically to favor hawkish rather than dovish behavior. The most important, in the context of today’s news is “double or nothing” bias, which is well-known in studies of choice under uncertainty as risk-seeking in the domain of losses (something first observed by Kahneman and Amos Tversky in their classic paper on prospect theory).

The basic point is that people tend to cast problems like whether to continue a war that is going badly in win-lose terms and to be prepared to accept a high probability of greater losses in return for a small probability of winning or breaking even (terms which are somewhat elastic in this context). So we get the Big Push, the Surge, the last throw of the dice and so on.

There are other biases that are based more in the way we manage things as a society than in individual psychology. The most important is the failure to treat decisions about war in terms of opportunity cost, by contrast with the way in which the budgeting process of governments (admittedly imperfectly) brings home the cost of other government activities. More on this soon, I hope.

  • This is not necessarily the case if your opponent is irrationally bent on your destruction, but one of the problems noted by Kahneman and Renshon is that people are overly willing to impute such motives to others, while perceiving themselves as peaceful and reasonable.

{ 1 trackback }

Easily Distracted » Blog Archive »
01.15.07 at 1:28 pm

{ 78 comments }

1

Gdr 01.11.07 at 8:21 am

The Kahneman/Renshon article is Why Hawks Win, yes?.

2

John Quiggin 01.11.07 at 8:34 am

D’oh! Fixed now, I hope.

3

abb1 01.11.07 at 8:49 am

This is not ‘war’ in the traditional sense; it appears to be an attempt to resolve internal political standoff in Iraq. The US troops will assume the task of ethnic cleansing of the Sunni Arabs (currently performed by Shia militias) and in response the Shia organizations will dissolve their militias and accept the rule of the puppet government. This seems to be the plan – last month Mr. Bush met Mr. al-Hakim and apparently they made a deal. Mr. Sadr, of course, still appears to be a wild card there, unless there’s something we don’t know, some kind of a deal there as well.

4

abb1 01.11.07 at 8:53 am

5

Matt Kuzma 01.11.07 at 9:06 am

Don’t forget the general failure of people to properly handle sunk costs. If we quit now, all that money and all those lives we spent were wasted! When we consider that victory means justifying past expenses, then it’s worth investing roughly as much effort as has already been spent to try to recover a win.

6

gr 01.11.07 at 9:35 am

Doesn’t all this generalize from what is, from a global perspective, a rather isolated phenomenon? It doesn’t seem to me that people in places other than the US or Britain (and some of its cultural cousins) harbour strong enthusiasm for war.

7

Cranky Observer 01.11.07 at 9:36 am

> After all, war is a negative-sum activity, so
> war between rational parties doesn’t make sense –
> there’s always a potential settlement that would
> leave both sides better off*.

As far as this specific war goes, IMHO the old “follow the money” adage works quite well.

But for war in general, the older I get the closer I come to this conclusion: male humans between the ages of 15-35 (more or less) like to fight. And they like to form small heirarchical groups to extend the fighting to the next village. All in good fun, although the weak and the unluckly are of course culled in the process along with a few bystanders.

Unfortunately, civilization and mass industrial culture have provided a ground for organizing groups and fighting vastly more fertile than anything Gaia envisioned when she crafted homo sapians. Sort of like what happens when you toss tons of sewage sludge into a small eco-balanced lake: the lake becomes totally clogged with algae and puetrifies within a few months.

Cranky

8

P O'Neill 01.11.07 at 9:38 am

I don’t think it’s a coincidence that Bush is pursuing a futile war and piling up huge levels of public debt at the same time. January 21st, 2009, they’re both someone else’s problem.

9

SF 01.11.07 at 9:41 am

The shift from looking at rational motives to irrational ones is a bit awkward. Couldn’t a rational motive consist in massing wealth in a military-industrial complex? In other words, one need not look at the state as a whole as the benefactor.

10

Barry 01.11.07 at 9:46 am

“…and in response the Shia organizations will dissolve their militias and accept the rule of the puppet government.”

Last I heard, having a militia was almost a requirement for holding office in the Iraqi government. I submit that the idea is to ethnically ‘cleanse’ the Sunni Arabs, so that things are more peaceful, so to speak, in 2008. Then (if things go far, far better than likely), the GOP can declare victory and conduct a very token de-escalation in time for the summer presidential campaign season.

11

eweininger 01.11.07 at 10:05 am

Every beginner’s guide to poker-type manual has a section on factors you should consider when deciding whether to stay in a hand and factors you shouldn’t consider. The most important factors to consider are, of course, your probability of winning and the potential cost relative to benefit (size of the pot). (Also important is the size of your bankroll relative to the opponents’.) The most important factor to ignore is, of course, the amount of money you’ve already sunk into the pot.

Wouldn’t it be fun if policy-makers were forced to play on one of the ever-proliferating cable TV amateur poker shows?

OK it would be tedious. But in theory, it would be fun.

12

EnlightenedDuck 01.11.07 at 10:17 am

“Doesn’t all this generalize from what is, from a global perspective, a rather isolated phenomenon? It doesn’t seem to me that people in places other than the US or Britain (and some of its cultural cousins) harbour strong enthusiasm for war.”

To properly argue this point, we’d need to agree on a time-frame over which to evaluate this statement, and definitions for who is looking for wars. For instance, does the US get credit for warmongering for its actions in Somalia (1993 or present), Kuwait (1990-1991), or Kosovo (1998)? Does Lebannon get credited with a war for 2006? Does the PA get credited for an ongoing war with the continuing rocket attacks on Israel from Gaza? Was Afghanistan a war the US looked for, or was it a response to an attack? How do we evaluate Chinese saber-rattling with regards to Taiwan, and do we credit the US deterrent there? How about countries which work through proxies (e.g. Iran)? How are civil wars counted?

Do we want to look at sizes of populations, and consider wars per capita? Since 1990, how many wars has Ethiopia fought?

13

Ragout 01.11.07 at 10:38 am

The claim that war is a negative-sum activity only makes sense when the fight is over resources or something tangible. When one side is fighting for some kind of ideal, like freedom or independence or the triumph of true religion, war can be be positive sum if the ideal is valued highly enough. I think this kind of war is more common than a simple fight over resources.

14

Greg Hunter 01.11.07 at 10:41 am

War is not a zero sum game if the goal is to win. Killing on a grand scale is required to subdue any resistance and if done properly it could come to Americas advantage. Michael Schurer in Imperial Hubris called it out and it worked for King Leopold in the Congo. Let the military do its thing correctly and Moqtada and Iraq will bend to our will.

Get out or Get it Done. No luke warm response.

15

K R Hasan 01.11.07 at 11:28 am

I recall Bertrand Russell pointing out in one of his books that uptil the Boer War every war waged by the British government was unpopular among the common people. The Boer War was fought at a time when mass circulation of newspapers had begun,and they were able to change public opinion.
The Iraq war is no exception. The mainstream press in the US went along with the viewpoint projected by the government, and over 70% of the population supported the invasion.

16

marcel 01.11.07 at 11:34 am

JQ wrote: After all, war is a negative-sum activity, so war between rational parties doesn’t make sense – there’s always a potential settlement that would leave both sides better off*. And empirically, it’s usually the case that both sides end up worse off relative to both the status quo ante or to a possible peace settlement they could have secured at a point well before the end of the war.

I’ve heard much the same said about strikes. But it’s my impression that once you add reputation, signalling and committement into the mix, you can come up with good game theoretic justifications for strikes. I would imagine that it is relatively easy to translate these arguments over to war, especially when, perhaps unlike strikes, the beneficiaries differ from the burden bearers (if I may coin a Dubya-ism).

17

ajay 01.11.07 at 11:40 am

It doesn’t seem to me that people in places other than the US or Britain (and some of its cultural cousins) harbour strong enthusiasm for war.

Any sentence that leads to the exclamation “Ah, those peace-loving Afghans!” should probably be re-examined. Or, for that matter, the Vietnamese, who spent basically the entire time from 1940 to 1980 fighting someone or other. (NB: justifiably in many cases, true; doesn’t alter the fact that they fought a lot.)

As for popularity – notably, none of the 19th century wars were unpopular enough that the army was unable to find enough volunteer soldiers to fight them. The British Army was a volunteer army until 1916; the colonial armies remained so until independence.

18

H. E. Baber 01.11.07 at 11:54 am

Cranky’s quasi-right–human beings by their nature like to fight. Young males do most of the fighting because they can win–women and older men do less fighting because they’re not physically strong enough or energetic enough to survive in a melee that includes young, fit males.

Explanations of why there is war and violence fail because they ask the wrong question. The question is why DON’T people fight, or even better, why do some people fight less than others. E.g. ask not “why do underclass youths in urban ghettos form gangs and fight” but “why DON’T upper middle class youths in expensive suburbs form gangs and fight.”

Once the question is put in this way the answers are easy: women don’t fight because most aren’t physically strong enough to compete; middle class youths don’t engage in gang violence because there are opportunity costs–and because they other outlets for the fight impulse, including contact sports.

Peace is a hard sell because it’s fundamentally contrary to our nature. That’s why the left, trying to sell the idea of non-aggression is always on the defensive and will always lose. Instead we should start with the assumption that people of course prefer to fight and, as policy to stop them from doing so, crank up the opportunity costs and provide harmless avenues for aggression–like philosophy. That’s why I’m in the game.

19

roger 01.11.07 at 11:55 am

I usually like Kahneman’s work, but this article points to the problems with disembedding a sequence of actions from a more general context. War as an active thing – a fight – exists, at least since the 30s, in a more general war culture. This is Hitler’s most successful legacy, actually – the incorporation of military structures into every part of the political economy. That was taken up by both the Soviets and the U.S. In that sense, we still live in Hitler’s world. From the interstate highway system the U.S. built in the fifties (which was officially funded, it should be remembered, in a bill that carried the title ‘security’) to the scattering of suburbs away from target cities to the reliance of the Clinton administration on arms deals to keep at least one small part of the American manufacturing sector working, war is ‘prehended’ in the entire system. Individual and real ‘wars’ are simply accidents of that larger system. The war is all the time, 24/7.

20

abb1 01.11.07 at 12:04 pm

H. E., how do you explain the Swiss, the Swedes, the Norwegians – societies that in the last couple hundred years have been more democratic than most and decisively preferred NOT to fight?

21

David Moles 01.11.07 at 12:46 pm

I wrote a paper (The Decision for War and the Limits of Rationality — HTML; PDF here) about this a few years ago in grad school. It doesn’t break any new ground but it’s a half-decent start on the literature, at least as of late 1999.

After all, war is a negative-sum activity, so war between rational parties doesn’t make sense – there’s always a potential settlement that would leave both sides better off.

The main problem with this is that with imperfect information, in the absence of an enforcement mechanism or a reasonable balance of power, it’s hard to determine what that settlement is and hard to credibly commit to it. If you were North Korea, would you trust G.W.’s assertion that the US had no plans to invade? If you were G.W., would you trust Musharraf’s assertion that al-Qaeda had been expelled from Pakistan and wouldn’t be allowed to return?

22

David Moles 01.11.07 at 12:57 pm

P.S. For the scientifictionally minded, the main thing I’m proud of in that paper is that I managed to slip Iain Banks, Ken Macleod and Terry Pratchett in among the Keegans and Van Crevelds.

23

marcel 01.11.07 at 1:37 pm

H. E. Baber (#18) writes:

Cranky’s quasi-right—human beings by their nature like to fight. Young males do most of the fighting because they can win—women and older men do less fighting because they’re not physically strong enough or energetic enough to survive in a melee that includes young, fit males.

I suspect that I am at most a few years younger than Cranky, and probably considerably older than Baber. I don’t think that young men fight just because they are the ones who can win. After a certain age – probably 18 or 19, but it only becomes noticeable decades later – men’s testosterone levels decline.

I recall a passage in one of the Platonic dialogues (probably the Symposium, but I cannot find it now) where Socrates (I think) mentions this – about how as you get older, the decline in sex drive is like being relieved of a burden, of freeing oneself from a harsh slave driver, about how much more pleasant life is as a result.

There is much to this view. I recall not being able to understand how any self-respcting male could hold this opinion. Having reached that age, and discussed this with a few friends, it makes sense. There is a reason for the minimum age restriction on being president, and it is not only to ensure a certain degree of experience and the resulting maturity. Young men are a bit insane from what is, in society as opposed to states of nature, a chemical overdose.

Young women may have their own hormonal issues, but I am not so intimately familiar with them, so I restrict my comments to the male of the species.

24

Justin 01.11.07 at 1:55 pm

Sounds like something Cephalus says in the first book of the Republic.

25

radek 01.11.07 at 2:11 pm

When one side is fighting for some kind of ideal, like freedom or independence or the triumph of true religion, war can be be positive sum if the ideal is valued highly enough.

Not really. The belligerent who is expected to loose should just agree to have a certain portion of the population convert or to accept the new beliefs but with old dressings or to agree to some kind of quasi-independent status. That way both avoid the inevitable costs of conflict.

Of course this kind of approach may contradict the very notion of a “strongly held ideal”. But in some sense “strongly held ideal” is very akin to being not-rational (though not necessarily “irrational”)

26

novakant 01.11.07 at 2:38 pm

I think it’s much simpler: generally speaking both life and politics are pretty drab and boring – war changes that, provides a rallying point, an enemy, a goal, endless excitement, meaning.

This point is made at slightly greater length in The Man without Qualities, if I recall correctly.

27

Sebastian Holsclaw 01.11.07 at 3:16 pm

The answer is that people don’t intuitively deal well with the problem of opportunity costs nor do they deal well with sunk costs.

Those two issues show up in all sorts of areas of life–including war.

28

Joel Turnipseed 01.11.07 at 3:37 pm

Abb1,

The Sweden, Norway, and Switzerland are the Lake Forest, Bloomfield Hills, and Wayzata of the previous analogy… also: Norway’s special forces are very happy to be fighting in Afghanistan (at least according to the couple of Norwegian special forces officers I met last year).

Else, I essentially agree with H.E. Baber’s point.

Also, whoever made the point about signalling is on to something fundamental that’s not made by JQ in his post or in the linked article: signalling can be FUNDAMENTAL to long-term success, even when it’s FOLLY in the short-term. For instance? The Marine Corps’ policy of “No Marines Left on the Battlefield.” This policy, upheld more often than not, may lead to more casualties than necessary… BUT: it also means that Marines are more willing to get into the fight, and more aggressively–and thus, to win the battle. Also, cf., those wiry little fuckers no one ever messes with… just. not. worth. it (the guy, say, in Trainspotting).

29

Peter Melia 01.11.07 at 3:42 pm

Think of Churchill !

30

airth10 01.11.07 at 3:43 pm

Today war and violence are signs of unsophistication. With Bush and his war we are seeing two unsophisticated parties going at each other. The process of war has had a sophisticating influence in the past, like between Britain and France. They learned in time there is no percentage in it. That is something Bush and his opposite have to learn and get out of their systems.

31

jason laning 01.11.07 at 3:52 pm

Speaking of Socrates and the Republic, perhaps the dialog with Glaucon where Socrates discusses how injustice originates in the luxurious state (as opposed to his initial description of the ideal, non-luxurious state) is the more pertinent passage. Since greater resources will be required in order to provide wants beyond needs:

“S: …Then we must enlarge our borders; for the original healthy State is no longer sufficient. Now will the city have to fill and swell with a multitude of callings which are not required by any natural want…

S: …Then a slice of our neighbours’ land will be wanted by us for pasture and tillage, and they will want a slice of ours, if, like ourselves, they exceed the limit of necessity, and give themselves up to the unlimited accumulation of wealth?

G: That, Socrates, will be inevitable.

S: And so we shall go to war, Glaucon. Shall we not?

G: Most certainly, he replied.

S: Then without determining as yet whether war does good or harm, this much we may affirm, that now we have discovered war to be derived from causes which are also the causes of almost all the evils in States, private as well as public.”

It’s a very old equation: the pathological obsession with the pursuit of wealth requires vigilant militarism. Security of one’s possessions (or wealthy lifestyle) becomes confused with the security of one’s existence.

32

aaron 01.11.07 at 3:57 pm

I don’t buy the assumption that war is a net negative, see my comment in Hansen’s OverComingBias thread. It’s just that the benefits aren’t easily traced and also are often intentionally overlooked due to our anti-violence bias.

33

raj 01.11.07 at 4:04 pm

The fact that people are so willing to support war is a puzzle that requires an explanation.

~sigh~

John, do you really need to be reminded of Herman Goering’s quotation, reproduced at Snopes.com?

The quotation that was paraded around the Internet was bad enough. The actual quotation, at the bottom of the Snopes.com page, was far worse

We got around to the subject of war again and I said that, contrary to his attitude, I did not think that the common people are very thankful for leaders who bring them war and destruction.

“Why, of course, the people don’t want war,” Goering shrugged. “Why would some poor slob on a farm want to risk his life in a war when the best that he can get out of it is to come back to his farm in one piece. Naturally, the common people don’t want war; neither in Russia nor in England nor in America, nor for that matter in Germany. That is understood. But, after all, it is the leaders of the country who determine the policy and it is always a simple matter to drag the people along, whether it is a democracy or a fascist dictatorship or a Parliament or a Communist dictatorship.”

“There is one difference,” I pointed out. “In a democracy the people have some say in the matter through their elected representatives, and in the United States only Congress can declare wars.”

“Oh, that is all well and good, but, voice or no voice, the people can always be brought to the bidding of the leaders. That is easy. All you have to do is tell them they are being attacked and denounce the pacifists for lack of patriotism and exposing the country to danger. It works the same way in any country.”

Read it and weep.

34

John I 01.11.07 at 4:05 pm

No. 7 (Cranky) nailed it. This war is all about the benjamins. The well-connected military industrial corporations make money hand over fist regardless of nominal “winning” or “losing.” Staying the course, “surging”, “the next six months are critical”, yadda yadda yadda, all prolong the draining of US taxpayers wallets into the coffers of Halliburton. That part of the equation is pretty simple.

What’s more interesting is the question of why the taxpayers fell for it for so long. For that I blame the 18-35 males and the media that panders to that market segment. I’m probably nowhere near Cranky’s age, but the last four years have made me cynical far beyond my years. This administration has relly laid bare the true evils that Ike warned about.

35

aaron 01.11.07 at 4:10 pm

“When one side is fighting for some kind of ideal, like freedom or independence or the triumph of true religion, war can be be positive sum if the ideal is valued highly enough.

Not really. The belligerent who is expected to loose should just agree to have a certain portion of the population convert or to accept the new beliefs but with old dressings or to agree to some kind of quasi-independent status. That way both avoid the inevitable costs of conflict.

Of course this kind of approach may contradict the very notion of a “strongly held ideal”. But in some sense “strongly held ideal” is very akin to being not-rational (though not necessarily “irrational”)

Huh, accepting such a compromise would make the instigator stronger and more capable of demanding more in the future. Bad, bad idea.

36

airth10 01.11.07 at 4:20 pm

I agree, war is not a net negative. However, today anything positive that can come from war can be achieved at a lesser cost, like through politics and diplomacy. This cost differential, though, has had to be learned.

Just recently Belarus and Russia had a spat over the flow of oil through a mutual pipe line. In the past that difference between them might have provoked a war or use of force. Today, they negotiate because the parties involved have learned that war or force solve little at a huge expense.

37

aaron 01.11.07 at 4:22 pm

Cranky’s quasi-right—human beings by their nature like to fight. Young males do most of the fighting because they can win—women and older men do less fighting because they’re not physically strong enough or energetic enough to survive in a melee that includes young, fit males.

Actually, there is a more practical reason for keeping women from combat (when it is major part of your economy). This stereo type alone raises the question, “if women are so useless, why not use them as fodder?” The practical reason for perpetuating the stereo type is that your population growth rate is far more dependant on the number of women in the population than men.

38

John Emerson 01.11.07 at 4:38 pm

The Swiss, Swedes, and Norwegians were military thugs during their early history — the Swedes as recently as 1709. They’re pretty much the same people genetically, so the difference must be institutional. (Unless hundreds of years of war weeded out the aggressive genes.)

39

a 01.11.07 at 4:44 pm

“After all, war is a negative-sum activity, so war between rational parties doesn’t make sense – there’s always a potential settlement that would leave both sides better off.” Am I missing something, or do you mean rational parties with perfect foresight? If both sides believe they will win a war with little effort (see WWI), then both sides can be rational to want to fight. (Or is your claim that the sides cannot be rational if they believe contradictory things???)

40

abb1 01.11.07 at 5:12 pm

#35 – it’s probably institutional but it could also be environmental (I think this is what Joel was saying) – suppose they somehow became so rich and pampered that they lost all their stamina. But I think it’s pretty obvious that them being so satisfied is the result, not the cause.

41

John Quiggin 01.11.07 at 6:25 pm

I’m planning a big post on the strength of the war party in the US (and to a lesser extent UK, and also Australia, which is something of a special case) by contrast with Europe in particular, and other developed countries. Coming Real Soon Now.

42

H. E. Baber 01.11.07 at 7:28 pm

I suspect that I am at most a few years younger than Cranky, and probably considerably older than Baber. I don’t think that young men fight just because they are the ones who can win. After a certain age – probably 18 or 19, but it only becomes noticeable decades later – men’s testosterone levels decline.

Bad guess, Marcel. I’m female and, let’s say, over 40 (and heterosexual). You can shove your hormonal thesis.

Wanna make something of it?

43

radek 01.11.07 at 10:14 pm

Along the same lines, why do people litigate, rather then just settle out of court? Same principle.

And some people do have an antipathy type of utility; “I don’t care how much I suffer as long as my opponent suffers some!” But yeah again, that’s not really in the realm of the “rational”.

44

Eli 01.11.07 at 11:05 pm

You write that “war between rational parties doesn’t make sense – there’s always a potential settlement that would leave both sides better off.” The psychological causes of war that you draw our attention to are fascinating, but it does not follow from the above statement that the potential for a peaceful settlement always exists that all wars are hence irrational. Wars are often the result of information asymmetry; certainly a peaceful settlement could be reached if both sides knew exactly what the outcome of armed conflict would be, but since that information is not always or even often readily available, most wars are not easy to avoid in this fashion. This does not make the decision to go to war irrational, per se.

45

anonymous 01.11.07 at 11:06 pm

Cranky is 100% correct.

Men seem to love explosions.

The bigger the better.

And those physics geeks at Los Alamos
got to make the biggest human created
explosion(s) of all time!

46

Chris 01.11.07 at 11:28 pm

In defense of young men:

Even those of us who aren’t blinded by testosterone are faced with the reality that (as a generalization) women like powerful and dangerous men. And since the best way to get a reputation for being a badass is to kill people . . .

47

Ragout 01.12.07 at 2:02 am

Radek,

I don’t see your point. I claimed that war is typically a positive sum game, since for intangibles like freedom, one side’s gain doesn’t have to equal the other’s loss. You argue that there’s always a peaceful solution that dominates war (presumably on the assumption that contracts can be costlessly enforced).

In fact, your argument proves the opposite of what you claim. You acknowledge that war can have a positive payoff. You reply that peace will always have a bigger positive payoff. But increasing the payoffs to peace makes the game *more* positive-sum, not less!

48

Ragout 01.12.07 at 2:44 am

John Quiggin writes:
war is a negative-sum activity, so war between rational parties doesn’t make sense – there’s always a potential settlement that would leave both sides better off

But I think that this is demonstrably false. The prisoner’s dilemma is an example of a game that (1) is negative sum and where (2) the only equilbria is dominated by a potential settlement that would leave both sides better off. And yet the dominant strategy is the one where both sides lose. So this counterexample demonstrates that (1) and (2) do not imply that “war between rational parties doesn’t make sense.”

49

lurker 01.12.07 at 2:52 am

@25, the Finns, e.g., tend to feel that a war in which they managed to avoid becoming Soviet citizens was a positive sum game. But I guess they could have agreed to a certain percentage of the population being murdered, a certain percentage sent to Siberia and the rest enslaved. I’m not quite convinced that they would have come out ahead, in terms of people killed and property destroyed, though.

50

Andrew Brown 01.12.07 at 2:59 am

The alleged pacifism of Scandinavians is a complex question.

The Swedes are completely pacifist now (driving there this summer, I heard the minister of defence say, on the radio, that “we must stamp out the macho culture in the army”), but it was an important part of their national myth that the could have fought in WW2 if anyone had been foolish enough to invade; the Norwegians did actually fight, and resist Hitler very toughly, while the Finns stood off the Russians with extraordinary courage and cunning.

51

ajay 01.12.07 at 5:32 am

war is a negative-sum activity, so war between rational parties doesn’t make sense – there’s always a potential settlement that would leave both sides better off

Eli is right – this belief only holds if both sides agree on the probability of winning and the costs involved. War between rational, fully-informed parties doesn’t make sense, because at least one of them will realise either a) that they cannot win or
b) that their probability of winning is small enough to make the expected payoff very negative, or
c) that the payoff from winning doesn’t outweigh either the expected cost of fighting or even the cost of winning.

But this is not the case! War isn’t a dice game, it’s an extremely complicated and uncertain endeavour. Both sides can have wildly different estimates of the probability of victory, the cost of victory and the cost of defeat. Both sides will generally be wrong; and when, based on their own assumptions, the numbers work out for both sides, you get a war.

I don’t see why this is such a puzzle. Either JQ’s making some subtle point I don’t get, or he has a very odd idea of war and intelligence.

52

abb1 01.12.07 at 6:34 am

The Soviet-Finnish war and Swedes’ alleged willingness to defend themselves obviously don’t work as counterarguments: these are quintessential episodes of self-defense, while the original argument postulated natural aggressiveness. On the contrary, these episodes only substantiate the theory that indeed under certain circumstances people with normal instincts don’t exhibit any serious aggressiveness.

53

abb1 01.12.07 at 6:42 am

And some people do have an antipathy type of utility; “I don’t care how much I suffer as long as my opponent suffers some!” But yeah again, that’s not really in the realm of the “rational”.

It may be irrational, but I think there’s something here to say for the concept of justice. Many people won’t settle for the optimal utilitarian outcome simply because they feel it’s unjust, unharmonious. Just like one may starve to death even when there is perfectly good source of protein available (like human flesh or something).

54

John Quiggin 01.12.07 at 7:34 am

Ragout, the prisoner’s dilemma is a positive-sum (or co-operative game).

55

John Quiggin 01.12.07 at 7:38 am

Various people have made the obvious point that wars can arise if both sides see the odds as being in their favour (although obviously at least one has to be wrong, and mayve both). A couple of points here.

First, this is inconsistent with what economists call common knowledge of rationality. The fact that the other side is willing to go to war even though, on your assessment, this is a bad decision implies either that they are irrational or that they know something you don’t. In the latter case, you should revise your own judgements. If both parties share the sensible prior belief that war is a bad idea in general, they should work out that they are better off settling.

Second, the problem is that the kind of disagreement where both parties overestimate their chances appears to be more prevalent. This is the optimism bias of K&R.

56

SamChevre 01.12.07 at 9:28 am

I think abb1 nailed it: utilitarian considerations AREN’T the only ones. Many people won’t settle for the optimal utilitarian outcome simply because they feel it’s unjust.

I’d say that’s exactly true.

The arguanent JQ makes can be generalized to violent conflict generally–but in many cases, you end up with “the rational result is to negotiate a settlement where you only pay HALF your lunch money to the bully.” Most people, I believe, see that as a bad outcome.

57

Ragout 01.12.07 at 9:35 am

John Quiggin,

Sure the payoffs in the PD can be chosen to make the game positive-sum. But they can also be chosen to make it negative-sum. And one counterexample suffices to disprove an argument.

(I’m speaking of the 1-shot prisoner’s dilemma, while I think that in JQ’s reply, he’s thinking of the repeated PD).

In any event, the PD is a perfectly good model of a rational war. Peace may yield the highest payoff, but both sides know that that if they strike first, they will win big and the other guy will lose big. So they both decide to attack, getting the low payoff of war.

58

merkur 01.12.07 at 9:55 am

“The fact that people are so willing to support war is a puzzle that requires an explanation.” Only if you don’t share their view, surely? Reducing human behaviour primarily to its economic implications is the reason why you are puzzled in the first place.

59

a 01.12.07 at 11:05 am

“First, this is inconsistent with what economists call common knowledge of rationality. The fact that the other side is willing to go to war even though, on your assessment, this is a bad decision implies either that they are irrational or that they know something you don’t.”

No! The fact that the other side is willing to go to war implies either that they are irrational or that they *believe* something you don’t. If it were knowledge then it would have to be true (but a belief doesn’t have to be true…).

60

Andrew Brown 01.12.07 at 12:15 pm

abb1: I don’t think that the Finns thought of their war as defensive the whole time. They had dreams of recovering the whole of Karelia — there is nothing particularly traditional or natural about the present Russo-Finnish border. And there were certainly some Swedes who volunteered to fight alongside the Finns. I believe the Olof Palme’s older brother was one. In any case, the previous record of both countries suggests that they were extremely fond of war for many centuries.

61

aaron 01.12.07 at 12:54 pm

This discussion seems to assume that all parties in a war are irrational, but only one must be irrational. And that all motivations for war are selfish.

Two motivations for an agressive war are: The perception that the opponent is incompetant in managing his resources and the world would benefit from removing the governing insitution from a region/populus; Selfishness, oppressing a people or taking their resources without compensation for the benefit of a different population.

62

radek 01.12.07 at 1:14 pm

John’s right above about the common knowledge of rationality. Personally I think it’s more a question of lack of committment technology – i.e. the parties must come to a settlement that not only divies up the pie in an efficient (though not nec. or even probably equitable) way but that also includes provisos to ensure the extortionary situtation won’t repeat itself – in a lot of cases people do give half their lunch money to the bully and when they don’t it’s usually because they know the bully will bully them for the rest anyway. In the cases where the other side cannot committ not to attack anyway after a settlement you might as well “fight to the death”.

So it’s 1939 and the Poles give Hitler the Danzig Corridor, save themselves 5 years of a brutal occupation, having 1/5 of their pop killed, Soviet invasion and becoming a Soviet satellite. That’s a good deal even if you care about honor, justice and fatherland. Except that any fool knows that the war would’ve happened anyway.

Also I think it’s still better to think of litigation as an example since it avoids all these issues of “ideology” etc. mostly.

Oh and abb1, I wasn’t thinking of people fighting for justice. I was thinking of people fighting out of (irrational) hatred.

63

radek 01.12.07 at 1:24 pm

I guess a more nerdy way to put it would be that the set of renegotiation-proof contracts is empty.

64

abb1 01.12.07 at 2:37 pm

But Radek, most us have a skewed perception of justice to begin with; we tend to sympathize with the side we identify with. I agree that it’s not exactly rational, but this is just how our brains work.

65

Quo Vadis 01.12.07 at 2:55 pm

Many wars have been initiated on the assumption that war was inevitable and the relative strength of the potential combatants would change over time. The party with the early advantage then has an incentive to initiate the conflict. The “lessons learned” from previous wars are often used to reinforce this assumption.

Is this assumption ever valid? History provides us with plenty of opportunities to engage in “what if”s.

66

John Quiggin 01.12.07 at 2:57 pm

Radek, the critical case for Poland was the Warsaw rising in 1944. If the leaders knew that the Red Army would not come to their help, they were wrong to stage the rising, since it was almost certain to be defeated with heavy casualties, as actually took place.

By contrast, the Warsaw Ghetto uprising in 1943 was a case where it was better to die fighting. And as you say it was rational to resist Hitler, having confirmed by experiment that making deals with him wouldn’t work.

67

aaron 01.12.07 at 3:32 pm

Because wars occurs when both parties believe they can win:

In cases with two type II motivation parties, there will be a net negetive.

In a case with type I and type II, it depends on the victor.

In cases with two type I parties, there will be a net posetive, because with both parties believing they will win, the one with be best institutions will win, resulting in greater longrun growth and productivity.

68

wandering wonderer 01.12.07 at 4:36 pm

Can irrational incentives be bought off? I’d argue evidence from some civil wars in Africa suggest the answer is yes – a willingness to sacrifice a measure of independnece for access to financial resources or gains from natural resources – rebel groups in diamond rich areas specifically noted in negotiations they’d require a higher price to be bought off than rebels in areas with less lucrative natural resources.

Makes me wonder about Amitai Etzioni’s pooh-poohing in the National Interest the job creation package Bush suggests as “Dollars for Peace in Iraq”. Maybe dollars for peace works.

69

radek 01.12.07 at 10:19 pm

You’re right John though I think in the case of the ’44 Rising the leaders were also banking somewhat on the Germans’ being “rational” – not wasting resources on defending a city which was bound to be captured anyway (and most of the German troops were being pulled out right before the rising). In both cases – reactions of Germans and Soviets – they got it wrong, with tragic consequences.

70

Anarch 01.12.07 at 11:55 pm

To piggyback off samchevre’s point, war is a perfectly rational course if you know that your opponent is going to compromise with instead of fighting back: asymptotically, you end up with everything and they with nothing. [Alternatively, in a less unrealistic format, asymptotically you end up with everything negotiable and they with nothing beyond the barest necessities.] Reminds me a little of the old computer game Balance of Power: threatening to nuke the opponent is a surefire means of success in the short, it’s only the (either short-term irrational or long-term rational) impulse to fight back which produces cataclysm.

71

abb1 01.13.07 at 9:27 am

Here: Analysis: Iraq PM’s silence telling. Similar to what I said in #3:

Al-Maliki aides have suggested the prime minister will attempt to avoid an all-out attack on the militia by attempting first to focus the new security drive on Sunni insurgent-held regions of the capital. If that were to prove successful, the theory goes, al-Maliki could then go the Mahdi Army and demand it disband because Shiites were no longer under threat.

And that appears to be the deal: Americans help SCIRI/Dawa kill some Sunnis – and then SCIRI/Dawa help Americans crack down on anti-American militias, Sadr. Then, I guess, the SCIRI/Dawa ayatollahs are somehow supposed to become American clients? Well, why not, if Saudi princes and Jordanian kings are…

72

abb1 01.13.07 at 9:48 am

What I don’t understand is why everyone thinks this plan can’t work. Is Mr. Sadr too strong? Maybe he is, but what if he has to meet an unfortunate end by a car-bomb or something? Things happen, you know.

73

Alex 01.13.07 at 3:52 pm

38: The Swiss, Swedes, and Norwegians were military thugs during their early history—the Swedes as recently as 1709. They’re pretty much the same people genetically, so the difference must be institutional. (Unless hundreds of years of war weeded out the aggressive genes.)

Indeed. Walking around the Royal Palace in Stockholm, a Swedish and peaceloving blogger explained to me with some relish that Charles XII’s invasion of the Polish-Lithuanian Commonwealth killed a percentage of the population that can only be compared with Hitler’s impact on Poland.

72: Middle Eastern warlords don’t go away when you kill them. (See Lebanon, Sadr’s dad, al-Hakim’s dad…)

74

Jon Kay 01.13.07 at 6:16 pm

> 72: Middle Eastern warlords don’t go away when you kill them. (See Lebanon, Sadr’s dad, al-Hakim’s dad…)

Neither al’Sadr, al’Sadr’s family, nor his successors will matter nearly so much if al’Sadr’s militia can’t get violent on the streets without getting shot at.

75

aaron 01.14.07 at 3:47 am

Even without Sadr, there’s that whole hezbollah thing Iran built for his father.

76

magyarman 01.14.07 at 7:24 am

Old men start wars and require young men to fight (and die) in them.

77

radek 01.14.07 at 12:12 pm

Charles XII’s invasion of the Polish-Lithuanian Commonwealth killed a percentage of the population that can only be compared with Hitler’s impact on Poland.

I think that peace loving Swedish blogger is taking too much credit. There was like, 8, different wars going on in Poland during this time. Swedes, Cossacks, Russians, Transylvanians, Prussians, not to mention a nobles’ rebellion and a peasant rebellion. So yeah in total it’s true but the credit should be properly divided.
And anyway, it was downhill for the Swedes since that time too, the big winners of all that mess being Russia and Prussia.

http://en.wikipedia.org/wiki/The_Deluge_%28Polish_history%29

and Sienkiewicz wrote some books about it;
http://en.wikipedia.org/wiki/The_Deluge_%28book%29

78

Phoenician in a time of Romans 01.14.07 at 4:13 pm

And empirically, it’s usually the case that both sides end up worse off relative to both the status quo ante or to a possible peace settlement they could have secured at a point well before the end of the war.

Depends on the methods of wealth creation in society. Almost certaily true for information-age societies (since human resources dedicated to war and the military are “wasted”). Not so true for industrial age societies (sweet, sweet oil) and almost certianly not true for agricultural age societies (capturing land WAS wealth).

Comments on this entry are closed.