Bryan Caplan  

Pacifism Defended

Reply to Aeon Skoble... Big-Picture Finance Questions...
It's time to reply to my critics.  Here goes:

Mike DC writes:

Suppose some jerk burns a Koran, and devout Muslims respond with by killing every American they come across. Should I, and other Americans not organize for our collective defense?

Whether we agree with Koran burning or not would be irrelevant once people that violently oppose Koran burning (but support all sorts of other violence against me) decide to target me.

If you've got a plan for "collective defense" that doesn't involve reckless endangerment of large numbers of innocent bystanders, I'd like to hear it.  But modern warfare sadly doesn't qualify.

Steve_0 writes:

You're normally an excellent philosopher, but here you seem too close to straw-man arguments, unjustified equivocation, and the fallacy of the excluded middle. The right to defense in the face of initiated violence does not mean unleashing WWIII. But the answer doesn't have to mean pacifism to the extent of losing ones own right to life.

There's an enormous middle ground between those two...

Defensive response may be messy, and I agree with the urge to minimize the collateral damage. But we don't become pacifist martyrs simply because the outcome isn't perfect.
I deny that pacifism makes us into martyrs.  The long-run consequences of war are sufficiently unpredictable that pacifism could easily be in our narrow self-interest.  Consider: If any of the main players in World War I - billed as "the war to end all wars" - had simply surrendered, even the "martyr" nation would likely have been better off than it was by the war's end - and World War II would have been avoided.

Lester Hunt writes:
The reason for the "not quite" is the doctrine of Double Effect. I see a difference between and evil effect which is intentionally brought about and a evil effect that is brought about as a foreseen but unavoidable side effect of pursuing a legitimate goal.
Is there a difference?  Sure.  But we greatly exaggerate the moral difference when foreigners are the ones who suffer the "unavoidable side effects."  If the police firebombed a domestic apartment complex to pursue the legitimate goal of killing Charles Manson, few people would consider the doctrine of Double Effect a strong defense.  Would you?

Kevin asks:
Would you say that the Jews of the Warsaw Ghetto were morally in the wrong? That wasn't literally a *war* of self defense - it would be more accurately classified as a *battle* of self defense, but does that create a substantial moral distinction?
The Warsaw Ghetto uprising was about as close to a true battle of self-defense as you're likely to find in modern warfare.  But I'd still say that it recklessly endangered large numbers of civilians.  And it's a perfect example of my point that the consequences of war are hard to predict.  The uprising (a) led to another 150,000 to 200,000 civilian deaths (most at German hands, of course), (b) didn't free Poland from the Germans, and (c) allowed Stalin to indirectly eliminate most of the Polish resistance while the Red Army sat on the sidelines.  Not worth dying for, not worth endangering bystanders for.

One-Eyed Man writes:
Recognizing that you have the right to something is very different from saying such conduct is obligatory or desirable. It might be morally permissible to kill thieves but still better to avoid doing so when possible.
Again, I'm not objecting to killing thieves.  I'm objecting to killing innocent bystanders while thief-hunting.

hsearles writes:

And what would a pacifist do when there is an army storming through one's homeland? Is it wrong to unite and join an army to oust the invaders? Is he just going to hope for the best and believe that the invaders have no ill will against him?

"Unite and join an army" always sounds good.  But what exactly is this army going to do?  Judging from virtually every army around, it's going to recklessly endanger large numbers of innocent bystanders.  And what are the odds its actions actually improve matters, rather than provoking reprisals and worse?  My complaint is that proponents of war "hope for the best" rather than facing these hard questions.

It is very easy to consider pacifism in an age where there is no threat of invasion, but one ought not forget that wars have had to be fought to reach this peace.

And I say that most of these wars were themselves caused by earlier rejection of pacifism.  To repeat, consider World War I.  Any major power that swallowed its pride could have averted not just the horrors of World War I, but the subsequent rise of Communism, Nazism, World War II, and more. 

Aeon Skoble writes:

The sad moral compromise we're forced into when we choose option 2 is both the lesser of two evils, and an evil the responsibility for which lies with the aggressor.

Question: What's the furthest you'll take this argument when the innocent bystanders aren't foreigners?  And if detailed historical study revealed that our government was the aggressor, do you think the other side would be justified in doing horrible things to us?  Consider: Suppose two countries are run by adherents of your position.  But one of them mistakenly believes the other country "started it."  The bizarre result: Both sides can keep escalating their level of brutality in good conscience.  Note: This isn't just a weird trolley problem; in the real world, both sides usually sincerely think "the other side started it."

Mark Brady writes:

This is akin to bait and switch. Although all the examples you have in mind concern states fighting states, you are a libertarian anarchist and as such reject statism in all its forms. What then is the relevance of these examples for your own political philosophy?

Plenty of relevance.  I also oppose guerrilla warfare and violent revolution on the same grounds.

Randy writes:
It seems to me that we're debating something that most of us learned in the schoolyard, i.e., it is possible for large numbers of semi-rational human beings to coexist in a limited space, but bullies do exist. And because a bully recognizes no limits, he or she must be taken out, by any means necessary.
But we patently don't take out bullies on the schoolyard "by any means necessary."  We don't throw a grenade into a room of kids to make sure the bully dies.  And the reason isn't just that an easier way to remove bullies exists.  If the only solution to the bully problem were throwing a grenade in a crowded schoolroom, we'd just learn to live with the bully's abuse.  Why?  Because the innocent bystanders aren't foreigners.

Doug MacKenzie writes:
To send a signal that one wont fight back against agression does more than allow thugs to walk all over others. Cowards who would normally remain at bay will take advantage of those who play the Amish strategy of pacifism. Individual rights have no practical meaning in such a context.
As I've explained before, this microeconomic analysis is woefully inadequate.  Yes, fighting does raise the cost of attacking you; but it also increases the demand for attacking you by making others angry.  The net effect is theoretically ambiguous and empirically unclear.

Matt writes:
 I think that if I were about to be attacked in an alley and I pulled out a handgun, in self-defense, and started shooting, there is a small chance that a stray bullet will kill an innocent bystander. I want to minimize that chance, but it won't prevent me from pulling the trigger.
I agree with you.  I'm not objecting to responsible risk-taking.  I drive.  My claim, rather, is that modern warfare is almost always irresponsible.  What militaries do isn't like taking a shot with a 1% chance of accidentally hitting a bystander.  It's more like throwing a grenade at a crowd because the gunman's somewhere in the middle.

A challenge to my critics: I've carefully stated my argument here.  The argument has three premises.  Please succinctly tell me what premise(s) you reject and why.

Comments and Sharing

TRACKBACKS (1 to date)
TrackBack URL:
The author at Eli Dourado in a related article titled Peace through Political Assassination? writes:
    Bryan Caplan writes three compelling posts on the common-sense case for pacifism. The short version of his argument is that it’s wrong to kill innocent foreigners (“collateral damage”), especially when the gain in doing so is not clea... [Tracked on April 25, 2011 4:55 PM]
COMMENTS (64 to date)
Kunal writes:

I think you're confusing the Warsaw Ghetto Uprising with the Warsaw Uprising. They were two different events, and I think the different circumstances make the case for the Warsaw Ghetto Uprising significantly stronger, even in the context of your argument for pacifism.

Kevin Dick writes:

I think the big problem with your theory is that it assumes the probability that you experience a threat of violence, and hence the expected amount of harm from capitulating, is exogenous.

However, if your pacifist position were widely known, you could expect every bully around to take advantage of you.

So it is disingenuous to calculate the cost-benefit of capitulating in a single conflict like WWI. Rather, you have to compare long term effect of different strategies.

Of course, there's been a lot of work on such evolutionarily stable strategies. And as far as I know, unconditional capitulation isn't one of them.

If you have an repeated game model that showed the superiority of your strategy, I would love a pointer.

Marc A Cohen writes:

You confuse the "Warsaw Ghetto Uprising" with the "Warsaw Uprising". In the Warsaw Ghetto Uprising, Jews in the Jewish ghetto of Warsaw organised a resistance movement, not even realistically to defend them from the Germans, but merely to kill as many Germans as possible and delay the German army as much as possible as it liquidated the Ghetto and sent the populace to death camps.

I would like to have you comment on this, as it seems to me that the participants had no moral obligation at all to allow themselves to be herded onto trains like cattle and sent to death camps without doing whatever was in their power to defend themselves. I contend that this is an instance that perfectly illustrates a situation where the people involved would have been justified to use ANY weapon at their disposal (including, for example, a nuclear weapon, if one had existed and they could get their hands on it).

Wallace Forman writes:

I think this chain demonstrates the difference between ex post and ex ante reasoning.

Obviously Brian is right that ex post, there will be less suffering in the particular instance if you surrender immediately. But "always surrender" is not a very inspiring ex ante rule for discouraging future aggression.

The best way to win a game of chicken is to tear off your steering wheel and throw it out the window. Nations don't resist invaders because they are trying to minimize losses ex post. They resist because they had already committed to resisting ex ante, in order to deter aggression in the first place.

Tom West writes:

I think the danger in examining questions like this is that if you push too hard, at its basic roots, it comes down to "I prefer to have any number of the 'other' die, than any of my own". The little lies that we tell ourselves to ensure we don't feel too bad stop us from fully exercising what lies at the bottom of our logic, but if you strip away all of that veneer, then we lose any reason for restraint whatsoever.

If the only choice is to be a murderer or a victim, then I'm a murderer. And since I now realize I'm a murderer, I have the freedom to...

Dangerous stuff. I think I'll stick with a healthy dose of hypocrisy.

George writes:
However, if your pacifist position were widely known, you could expect every bully around to take advantage of you.

Exactly. If a nation embraced Caplan's pacifist stance it should expect to be overrun by it's enemies. Caplan's strategy may be the "morally superior" one (I'll just accept this as true rather than argue against it), but it's also the strategy that will get those who follow it at best the subject of severe abuse, or everyone killed at worse.

So it is disingenuous to calculate the cost-benefit of capitulating in a single conflict like WWI. Rather, you have to compare long term effect of different strategies.

Why not use WWII as an example? If the Caplan strategy were used, then most if not all the Jews in Europe would be dead. Actually Gandhi advocated Jewish pacifism in the face of genocide IIRC. Which, again, may be the "morally superior" position, but it is also the position of a group which has allowed itself to be wiped out.

Is Caplan maybe just testing us so we all come to the conclusion that Nature does not support the human sense of right and wrong? What I mean is, pacifism may be "morally correct" but obviously Nature does not reward such a strategy.

Or maybe this is a test so we all accept that working within the confines of Nature is how we must actually live our lives. We may want pacifism to be morally superior, but if we are to survive and flourish then we must confine our moral judgments to within the boundaries nature has set.

MikeDC writes:

Well, any plan for collective defense (without harming innocent bystanders) has to be subject to the threat at hand. Abject surrender to Imperial Germany, for instance, sounds preferable to abject surrender to al Qaeda.

Simply put, I wouldn't argue against any of your premises that lead toward pacifism, but I can still imagine a variety of situations in which I think the ratio of innocents saved to killed (especially over time) would exceed that five to one ratio.

While we face extreme uncertainty over the long-run (your second point), this doesn't negate the fact that we should be more willing to fight when
* The enemy appears especially aggressive
* The enemy is especially incompatible with our continued existence and the ways in which we currently exist.
* We can, in a variety of ways, attempt to mitigate the loss of innocent life.

Doc Merlin writes:

'Actually Gandhi advocated Jewish pacifism in the face of genocide IIRC. Which, again, may be the "morally superior" position, but it is also the position of a group which has allowed itself to be wiped out.'

Then its not morally superior but is the position of death and destruction. The evil would like nothing better than for their victims to not fight back.

Evan writes:


I contend that this is an instance that perfectly illustrates a situation where the people involved would have been justified to use ANY weapon at their disposal (including, for example, a nuclear weapon, if one had existed and they could get their hands on it).

I agree with you that the Warsaw Ghetto Uprising was a just battle. However, it seems to me that using a nuclear weapon would be likely to fry or poison a lot of totally innocent Poles who lived near the Ghetto, and would therefore be unjustified.

Also, remember that just because instances where fighting is totally justified exist in real life, doesn't mean that they're common.


The best way to win a game of chicken is to tear off your steering wheel and throw it out the window. Nations don't resist invaders because they are trying to minimize losses ex post. They resist because they had already committed to resisting ex ante, in order to deter aggression in the first place.

This is the best argument I've heard against Bryan's position so far. He's addressed it to some extent here, but I don't entirely buy his theory. It seems like really obnoxious and aggressive policies would make people want to attack more the way Bryan suggests. I believe Germany's big naval buildup at the end of the 19th century was what made Britain into its enemy, for instance. But more muted defense policies would probably deter without inciting. Of course, I don't think Bryan advocates never defending yourself, I don't see anything in his position that would bar us from, say, bombing an invasion fleet in the middle of the ocean.

That being said, I think Bryan's case is strong enough to advocate a significantly smaller and less interventionist military than we have today.

Scott Wentland writes:

All of three of your premises are off base, because you refer to the "war" in total costs, rather than from an individual country or group's standpoint. Some wars have highly certain costs, particularly the ones where the enemy's primary objective is to kill you. And, I think it's correct to say the "war" has little long run benefit (in fact, virtually all are a net cost), but one side may have a net benefit if they win (or, at least less of a net cost than if they lose). Therefore, war is probably best framed as a prisoner's dilemma, as economists often do.

The cooperative solution is clearly the "best" solution (i.e. everyone is a peaceable pacifist). But "cooperating" or being a pacifist when the others are "not cooperating" and attacking you will cost you. With pacifism, you face this cost with certainty (e.g. when the Germans vowed to kill all Jews) rather than fighting back and having some probability that you live. Both sides often act on what they see as their self interest (which is often off-base, as you point out, so we have way too many wars because of miscalculation), and arrive at negative sum outcome.

Properly framed, if a war is a prisoner's dilemma, and you vow to be a "cooperative" pacifist, then that may simply spell your death. Sometimes, it doesn't mean your death (like when conquerers just want to take over pieces of land or pacifists will have the make that judgment case by case), but in many cases it does (as the other commenters point out, that was the case for the Jews during WWII). But simply saying that the "optimal" solution lies in the "cooperative" pacifist square is not a good argument for being a pacifist! Would you tell the prisoner not to confess?

So, I am on board with you when you say peace is the best solution. And, war/pacifism will probably even be less costly than war/war. But you lose me when you seem to suggest pacifism as a dominant strategy. That is, in order for you to fully justify pacifism to an economist, you have to figure out a way to justify to the prisoner in the prisoner's dilemma not to confess.

CBrinton writes:

I disagree with Caplan's point 2, or rather his interpretation of it; I think some wars do have noticeably better than even payoff ratios. In addition to WWII I'd name the Korean war and US Civil war (from the US/northern perspective) as examples, recognizing that not everybody will agree.

Has Caplan ever explained what the cutoff for "modern" war is? Was collective self-defense by violent means a sensible strategy at some point in the past, in his opinion? If so, approximately when did that change?

And Marc Cohen is right; Caplan seems to have misinterpreted the question about the Warsaw Ghetto Uprising.

Tracy W writes:

I don't know if WWI is a good example. Visiting France and Belgium is giving me a different perspective on it to reading predominantly British and NZ history books. If the local museums are right, the Germans were indeed committing a whole bunch of atrocities against civilians in occupied areas. What is the basis for the belief that it would have been better for a country to unilaterally surrender?

And Japan became fascist even though it was on the winning side of WWI so I'm not so confident that things would have worked out for the better in the long run either.

Gian writes:

Caplan bandies the "innocent bystanders" a lot without explanation. A lot of his thesis is based upon this concept.

But is this concept always meaningful? Who were these "innocent bystanders" when Mongols invaded?. In Warsaw Ghetto? How much the bystanding Poles or Germans "innocent"?.

Wayne writes:

We have to remember that Bryan is specifically talking about situations where the response to organized injustice is county vs. country warfare. For analytical/ blogging purposes, he is intentionally leaving a lot of response options off the table. Whenever extremes are used, there is a risk of coming across as absurd.

If he added responses such as a focused police action or a bribing scheme, then the analysis would come across sounder. We could then talk about the moral agency problems inherent in country vs. country warfare. That would be a more analytically rigorous path to take. But, I suspect it's not blog friendly. (Two swipes down on the iPad is my limit.)

Daniel Kuehn writes:

On your premises - I don't think it's so much that any of those premises are wrong as it is that they are incomplete. War ought to be justified, but so must failure to respond. Just as the net benefits of war must be considered, so must the net benefits of not waging war. The net benefits of not waging war ought to be compared to the net benefits of waging war.

And the net benefits of not waging war are also highly uncertain, and often quite negative.

You worry about killing innocent bystanders, for good reason. But it's not clear to me why you're under the impression that war will kill more innocent bystanders than not waging war. We need look no farther back than Rwanda for an example of that, and one wonders how many killing of innocents could have been prevented if we entered WWII earlier. I'm not proposing neocon wars all across the globe, but yours doesn't seem to be a case for minimizing harm to innocents. The world is messier than you seem to want to make it, and it seems like you're trying to pretend its not that messy so that you can preserve easy decision rules like pacifism. I think there ought to be less war, but I'm not buying your argument.

ajb writes:

For those who think Bryan is just against big state wars, remember that he favors individuals running away when attacked. Always.

As others have noted, all the literature on the Prisoner's Dilemma indicates that unilateral pacifism is not a stable scenario for pacifists in a world with even a few bullies and predators.

Moreover, to the extent that Caplan does not succeed in creating universal pacifism but merely weakening "our side's" resolve in a threatening situation (again see WW2 or bin Laden's reaction to the US running post Mogadishu) I blame him for some of the harm caused by the subsequent conflict.

Since he seems to deny such responsibility and thinks of morality only in terms of active harms while attributing secondary costs to warmakers but not secondary harms to peaceniks I'm content to say that his position is neither tenable, nor moral, nor even intellectually fair. And if he insists on his odd definition of morality, so be it. I sign up for the immorals and the Might Makes Right folks.

Randy writes:

True, Bryan, that the average schoolyard bully does not have a grenade, but the average soldier or terrorist does. "Any means necessary" incorporates a broad range of possible responses - from isolation and monitoring to mutually assured destruction.

For example, what if the bully is a political organization that claims to be my "government"? I'm thinking that subversion is an appropriate response - and subversion is certainly a form of warfare.

Telnar writes:

One question worth thinking about is the evolutionary stability of various strategies. If we focus only at the margin in our current world, it seems clearly better for a country like Luxembourg in 2011 to be pacifist, but how much of that is free riding?

The ability of states with aggressive goals and means like Nazi Gremany and the Soviet Union to gain strength over time is directly related to the way peer countries respond to them.

History is full of leaders who sought to advance through conquest. The fact that in 2011 such leaders have to consider that sometimes the US intervenes (as it did, for example, in the 1991 Gulf War) almost certainly diminishes the frequency with which such conquerers gain enough power to pose a global threat.

Unlike operations against small groups within a nation (as anti-terror campaigns usually are), security guarantees aren't likely to shift the demand curve for crossing the guarantor. They aren't free. I would model them as concentrating the losses from aggression into tail risk from a major conflict. Unlike minor wars, though, they don't create reasons for 3rd parties to oppose the guarantor (e.g. Brazil didn't arm against the US because it deployed troops to protect West Germany against the Soviets). Still, security guarantees prevent the scenario where an aggressive power grows over time because of its aggression.

Sonic Charmer writes:


I reject premise #2. The uncertainty of the benefits of war is overstated, in dimensions that matter.

If your metric is 'number of dead' than certainly this can easily conceivably be on the order of 1k or 10k or 100k or more, and optically that is a form of uncertainty. However, this is not the outcome-space nations are concerned with when considering war, i.e. war is (obviously) not largely an attempt to minimize # of deaths to the exclusion of all else.

In the dimension that matters, the outcome is actually rather binary: either the invaders take over your country and reshape its governance and culture, or they do not. Failure to act in defense means they probably will. Acting successfully in defense means changing that 'will' to 'will not'. Nobody in the actual situation thinks the choice between these two is all that fuzzy or uncertain.

The exception that proves the rule can happen if your nation is clearly, materially weaker than the enemy nation. In those cases, perhaps I agree that surrender is in fact the better option. But only because the 'success' option has such low probability as to be negligible. For example, I think that Saddam Hussein's government should have just surrendered to the U.S. This would clearly have been better for all concerned.

Nobody calls me 'anti-war' for this position, however. And it's a function of the relative strengths of the two countries. War is what happens when countries disagree about their relative strength, as the saying goes.

J Storrs Hall writes:

Let's imagine that we could have war according to the rules of medieval chivalry (even though we know that in practice these were usually honored in the breach): soldiers only attacked other soldiers, women and children were strictly to be left alone, peasants stayed with the land, etc, etc.

It's still a pretty grim business. But I think one can make a case that Western civilization could never have developed without it, as it formed a major part of the evolutionary pressure that forced European governments into competence over the period of say 800 - 1800.

As a consequentialist libertarian, I don't see a problem with a system that allows for a certain amount of war as long as the overall effect on civilization is positive. In fact, I would see prohibiting it as immoral.

Let me hasten to add that there is nothing in the foregoing that would justify current-day American adventurism.

Shane writes:

It's fascinating and I appreciate the arguments here.

One observation to make is that war seems to be much less common and much less bloody today than in the past. Huge areas that in the past were regularly engulfed by war are now free of it: most of Europe and North America for example.

Why is this? I know some theories point to "pax americana", the military domination of these regions by the single American superpower. Other theories look at free trade or democracy.

One way or another, peace has broken out under the shadow of nuclear weapons and other high tech military. If the US and other countries abandoned their militaries, would the peace remain?

darjen writes:

I can't really find anything that I disagree with in Bryan's post. I don't have the right to decide if innocent people should die in my defense. Nobody has the right to steal my money to defend themselves. The world would be a better place if citizens rejected government and rejected their governments foolish military adventures.

@J Storrs Hall:
As a consequentialist libertarian, I don't see a problem with a system that allows for a certain amount of war as long as the overall effect on civilization is positive.

I have a hard time coming up with any examples of war where the overall effect on civilization is positive.

Roger writes:


I agree with comments such as ajb's. You are counting the secondary and unintended consequences of defense, but ignoring the secondary and unintended effects of pacifism.

The result of extreme pacifism will result in exploitation of the pacifist. If most of us become pacifists, the world will be real easy to describe and the results are simple to quantify. The world will much look like it did prior to the modern era. 80 to 90% of us (and virtually all the pacifists) will scrape out a meager existence on land owned by the non-pacifist exploiter. Any surplus will be taken by the exploiter.

This world will support, oh... about a billion or so folks. The rest of us will starve or die of disease from our weakened condition. In other words, most of us will be dead, all of us will be poor and even the exploiters (longer term) will be worse off. Our life span will be about 30 years or so, our health terrible.

The path you suggest leads to feudalism at best, if not outright slavery. It leads to 6 billion deaths.

Yes, war is terrible and has ghastly unintended consequences. It should be an absolute last resort. But the consequences of extreme pacifism will be much worse and much more certain.

Sonic Charmer writes:

I have a hard time coming up with any examples of war where the overall effect on civilization is positive.

Part of the problem here is in thinking of 'war' as something you can either choose to do or not-do. Then you do indeed get results like: 'WWII was bad for civilization! Golly, why did we do it!'

But that's not the actual choice usually faced by anyone. Or, at most, it's a statement one needs to direct towards Hitler. (Which, good luck with that.) In the WWII example, you can't gauge the 'effect on civilization' of 'doing WWII' vs 'not doing WWII', to be doing anything meaningful you have to consider fighting back against and stopping Hitler vs. not doing so.

Note, in both options you had 'war' - there's no 'not-war' option. It's just that in the latter case you would have surrendered the war sooner. In any event, the idea that resisting and stopping Hitler was salutary for civilization is not even all that controversial or hard to defend, even if you think it's not a slam-dunk.

Aeon J. Skoble writes:

"in the real world, both sides usually sincerely think "the other side started it.""

If you mean that government propaganda typically aims for this effect, of course. But I wonder. Did average-joe Germans in the 30s really believe that the Poles and Czechs "started it"? Did average-joe Iraqis really believe that Kuwait initiated aggression? In any case, my arguments aren't about what people believe, they're about what's true. The Germans were in the wrong invading their neighbors and attempting to conquer Europe and exterminate Jews. The Iraqis were in the wrong trying to annex Kuawit. Going back farther, the Persians were in the wrong trying to conquer Greece, and where darjen asks upthread for "examples of war where the overall effect on civilization is positive," this is my favorite answer.
My position is that it's a category mistake to say of a war either that is just _or_ unjust, in the way we say "Super Bowl 23 was a good game." If A is unjustified in attacking B, B is justified in defending itself agaisnt A. So when we wonder whether the Greco-Persian War was just, we're at best being using a kind of shorthand. The Persian invasion was unjust, but the Hellenic defense was perfectly justified.

darjen writes:

@Sonic Charmer:

WWII was almost a direct result of the outcome of WWI and the treaty of Versailles. There's a pretty good chance Hitler could not have rose to power without the crushing, desperate economic conditions imposed on Germany by the Allies and their demand for reparations. WWI was a stalemate before the United States decided to join. Seems to me like WWII is one of the best historical arguments for non-intervention there is.

Kevin writes:

My thanks to Kunal and Marc A. Cohen for pointing out to Byran the difference between the Warsaw Uprising and the Warsaw Ghetto Uprising (Bryan's link states the latter but actually leads to the former). It was the latter I intended to use as a reference, when I mentioned "the Jews of the Warsaw Ghetto." I suppose I could have phrased that more carefully; my apologies if my ambiguity confused anyone else.

I agree with Evan that Marc is overstating things in advocating for a nuke, but the rest of Marc's comment seems basically a reflection of my own view.

In any case, Bryan asked for a response to his three point case, and I shan't demur.

Premise one I find largely agreeable.

Premise two, less so. When Bryan says "its long-run benefits have to be substantially larger than its short-run costs," I think Daniel Kuehn is correct to point out that the costs of not waging war are often huge and negative. Rwanda is a good example.

Hypothetical: in the mid 1930s, England decides it's high time to put a stop to this Hitler chap, before he fully rebuilds Germany's military. They mobilize their forces and attack full out, and attain victory, but at the cost of many German and British soldiers, plus 150,000 civilians. I'm quite certain Bryan would be arguing today (we're assuming the rest of history hasn't changed such that we were all born - it's just a hypothetical, people!) the same thing he's saying now, and he would be so epically wrong it approaches black comedy.

In this light, his "I'm arguing from ignorance, while you claim knowledge" line seems like little more than a rhetorical device to say "assume I'm right by default." He uses this device to set up bets with odds. I'd like to know how he would have structured a bet in the hypothetical I propose. In this hypothetical, the war ended a decade early, the Holocaust never came to be, and eastern Europe never ended up as part of the Soviet Union. I'd like him to state how, according to his theory, I would have been able to prove my case in the bet and collected my winnings.

Bryan has set up a system of pacifism that is unfalsifiable. I say this because part of having a falsifiable theory is that you can, at least in principle, explain a specific set of circumstances would disprove your views. This does not seem possible in Caplan's system. For example, how are we to determine if his third premise is met? I argue that had Germany been derailed before Hitler managed to rebuild his war machine, it would have been met. But how could I have proven this to a 1950s Caplan?

Caplan has slightly addressed the issue of preemptively attacking Germany pre WWII in an earlier post. His response is to say "But two can play at the hindsight game: Pacifism could easily have prevented World War I, leaving no room for the likes of Hitler to rise to power." And no doubt he could push that back further. But that's simply an infinite regress bait and switch. Yes, if pacifism had always been practiced throughout human history, then we wouldn't have to worry about it now. But that doesn't tell us anything about what we should do in the world today, as it actually exists. Would General Caplan have said at the time, "Yes, this Hitler seems a right unpleasant fellow, but if we hadn't have been involved in all that Great War tomfoolery, we wouldn't be talking about him today. Ergo, let's not get too worked up on trying to put a stop to him." My submission, ladies and gentlemen of the jury, is that Caplan has fallen into his own model of rational irrationality. It's easy for him to hold these views because the cost of them is zero, and the psychological gratification is above zero.

Further, Caplan's argument that deterrence is just as likely to lead to provocation is weak sauce. It doesn't mesh with intuition, and he's not even come close to building a case which would overcome this intuition.

Sonic Charmer writes:


You're not saying - at least, not convincingly - that WWII should not have been fought (from Allies' side; remember, whether there was going to be some war had already been decided in the affirmative, by people like Hitler). Instead, you, like Bryan, are saying that it 'could have been avoided', if some other historical contingencies (decades prior), i.e. 'WWI', had not obtained.

That's all well and good as a counterfactual historical thought experiment but has little bearing on an actual person actually living and holding authority at the time of Hitler and charged with deciding what if anything to do. He wasn't faced with some other contingency, he was faced with that one. And even if the 'could have been avoided' claim is true (I am sympathetic to it), it's not an argument against Allies waging WWII in the first place (it's an argument against waging WWI).

Also, it's a game that has no beginning. History did not begin with World War I. Undoubtedly one can find plenty of analysis tracing WWI's roots to something else one chooses to cite/blame - misguided Victorian chivalry, or whatever. It really only depends on how far back you wish to trace the line (Napoleonic wars? Enlightenment? fall of Roman Empire?). All of history is like this so I don't quite understand this sleight of hand attempt in choosing to draw the line at WWI and saying that it somehow proves all wars hence have bad outcomes. Although obviously such analysis has a long tradition tracing back almost to the close of WWI itself, 'lost generation' writing, etc.

Aeon J. Skoble writes:

It is entirely irrlevant to whether it was just to fight against Nazis in WWII to note that WWI set the stage etc. That reduces pacifism to an "I toldja so" game. I'm sure it's true that WWI and Versailles set the stage for Nazism etc., but when the Nazis are at your door, the question isn't "gee, if we had handled WWI better, would this still be happening?" It is "is it right to combat this threat?"

darjen writes:

It's not irrelevant. Actually, there is nothing that could be more relevant. The unintended consequences of entering WWI have everything to do with the cause of WWII. You can't tell me that WWII should have been fought on its own merits. WWII is inseparably connected to WWI.

Presumably humans are capable of learning from their past mistakes. And WWI was one of the biggest pointless mistakes of the past century. The sooner we learn this, the better. Nationalistic government must eventually be wiped out if we want to prevent any more world wars from happening. I'm not saying there would no longer be any conflict or war. But there certainly would be fewer global wars.

Tracy W writes:

Darjen, I agree that Hitler himself was likely a result of the Versailles Treaty at the end of WWI. But how can we know that pacifism in WWI wouldn't have resulted in an equally bad outcome? The Japanese in WWII committed way too many atrocities and they were on the winning side in WWI. I can imagine a different outcome where say France surrendered in WWI and the moral the German generals took from this was "hey war is a great way to get more power", combine that with a Great Depression, perhaps a Communist revolution in an undemocratic Kaiser regime, and a blood thirsty Communist dictatorship with expansionary plans might have emerged. Or a fascist one under a different leader.

Josh writes:

Ok, I'll take a shot. Regarding your 3 points:

1. I don't value all lives at $5M. I value "good people's" lives at $5M. But "evil" people are valued at substantially less, and usually negative. [From your comments on this post, it sounds as if you agree]. Furthermore, I view an enemy's friends as the approximately the enemy. For example, I don't value German civilians killed during WWII at anywhere near $5M. Those are people who gave Hitler power and then did not try to stop him. They are essentially the enemy. The fact that they don't wear uniforms does not mean they are my friends. Ditto German (private) property.

To summarize my objection to #1, the value of the damage done in war is substantially lower than you claim once you account for the fact that the damage I do to the enemy and those who support him are valued (by me) at very little.

2. The long-run benefits of war are substantial! What would have been the amount of lost output if Germany and Japan had won WWII? Even if we value each of the 4m axis civilian deaths at your $5m per (which I've already stated, I disagree with), that's only $20 trillion. Wouldn't a world run by German and Japanese dictators have experienced far more than that in lost growth, even now (which is "only" 70 years later; there's still a lot more time to come)?

Or given the state of China (especially up until ~10 years ago), wouldn't it have been worth more lives to try to stop communism there?

And then there's the fact that generally violent dictatorships are not necessarily just going to stop trying to conquer people once they've conquered you. If you let them conquer you because the cost of fighting them is too high, then once they do, why are you so sure you won't get forced to kill others (or killed resisting)?

So to summarize my objection to #2, the benefits of war are generally far higher than you assume.

I have no problem with your #3. I agree that the benefits must greatly exceed the costs. We simply disagree what those numbers are.

David N writes:

Robert Nozick posed an interesting dilemma: Your enemy throws you down a well. You survive the fall, and at the bottom find a "ray gun" capable of vaporizing people. Your enemy then throws your best friend down the well, surely to crush you when he lands. What would Caplan do?

As to the three premises, I think #1 does not consider opportunity cost. The immediate costs of war are clearly awful; the ultimate cost of pacifism may be much worse. I wrote "may be," but I feel "would be," and feel confident there is plenty of historical support.

#2 is framed very favorably to Caplan. Let's stipulate there are no (net) benefits to just wars, only costs to both sides. You may estimate the cost of pacifism (e.g. enslavement or genocide of your people and destruction of your culture) as greater than the cost of war (e.g. casualties on your side and to innocents). Can you estimate with certainty? Of course not. And if "benefits" can be uncertain so can costs. The argument from uncertainty works both ways, which is maybe why many of us don't find it very compelling.

Considering the above I think #3 is just plain wrong. A just war need not have any long-run benefits, only minimization of costs. If you kill my brother and aim for my sister, then I kill you and accidently kill an innocent bystander, my sister's life is not a "benefit." It sucks for the innocent bystander but I didn't value his life less than my sister's.

ajb writes:

Even Bryan's attempts to link more "just" wars to unjust wars (e.g. WWII and WWI) miss the standard economic point that decisions occur at the margin.

Who cares if WWII wouldn't have come about if there had been no WWI? At the margin, weakness in the face of Hitler (and arguably disarming the US) encouraged the attacks in the 1930s.

If you want to play consequentialist, we can argue that civilization and trade itself led to wars.

After all if all humans were content to live as isolated pacifist hunter gatherer families then no conflicts would ever have started! There, see? I've used the Caplan arguments to show commerce led to wars. (Insert exploration of the new world, the desire for settled agriculture, etc.)

Bah. This is my last post on this. It's too silly for words. I only hope Caplan never faces a world where his choices affect his immediate family in ways more terrifying than running away from ghetto thugs.

darjen writes:

@Tracy W
I can imagine a different outcome where say France surrendered in WWI and the moral the German generals took from this was "hey war is a great way to get more power"

If France surrendered to Germany in WWI, there would have been a lot less innocent people killed, and a lot less property destroyed. Even if France had to pay some taxes to Germany, at least for a while, they probably would have been better off economically.

As some Arab states have recently shown, it certainly is possible to resist or throw off abusive government in a non-violent (or mostly non-violent) fashion. Even abusive governments ultimately rely on consent of the people...

hsearles writes:

Thanks for taking the time to reply and now for my rebuttal:

"'Unite and join an army' always sounds good. But what exactly is this army going to do? Judging from virtually every army around, it's going to recklessly endanger large numbers of innocent bystanders. And what are the odds its actions actually improve matters, rather than provoking reprisals and worse? My complaint is that proponents of war 'hope for the best' rather than facing these hard questions."
I would rather risk large numbers of innocent civilians to defend myself against an invading army than to simply yield and hope for the best. Just because John thinks that the best solution to an oncoming army is to simply yield does not mean I will respect that decision. Sometimes fire must be fought with fire.

"And I say that most of these wars were themselves caused by earlier rejection of pacifism. To repeat, consider World War I. Any major power that swallowed its pride could have averted not just the horrors of World War I, but the subsequent rise of Communism, Nazism, World War II, and more."
However, the past cannot be undone and we cannot live in the past wishing that had someone only done x then war could have been prevented. The initial conditions must be taken for granted so speaking how a different course of action could have changed them is nothing less than wishful thinking. Even if previous instances of war could have been preventable does not mean that instances now or in the future can be.

Even if WWI was folly and a necessary condition for the rise of both Bolshevism and Nazism, it does not follow that future wars involving both are illegitimate. How is that a relevant fact when the Red Army is invading Warsaw in 1920? Even the fact that the Polish-Soviet War was mostly the result of Polish aggression is still not something that will influence the morality of standing against the Bolsheviks at that point in time. The same is true in WWII, was the only moral decision available for the British government after Germany declared war on Poland was to go “This could have all been prevented if the Great War were avoided?”

Now I shall move to your premises:

“1. The immediate costs of war are clearly awful.”

“2. The long-run benefits of war are highly uncertain.”
Both for the good and for the bad. In the explication of your premise:
“ Some wars - most obviously the Napoleonic Wars and World War II - at least arguably deserve credit for decades of subsequent peace.  But many other wars - like the French Revolution and World War I - just sowed the seeds for new and greater horrors.  You could say, "Fine, let's only fight wars with big long-run benefits."  In practice, however, it's very difficult to predict a war's long-run consequences.  One of the great lessons of Tetlock's Expert Political Judgment is that foreign policy experts are much more certain of their predictions than they have any right to be.”
you are one-sided since you do not mention that somethings the long-run benefits may be better than expected. I would say that the long-run benefits of WW2 were better than expected since the war stopped what would have been the eventual extermination of much of the non-German Eastern European population. Here, I must answer a potential critique that Nazi extermination policies were the result of WW2 and would not have to occur otherwise with the fact that there was a logic to Nazi racial policies that slowly built-up to the death camps (supporting this is Saul Friedländer's research in Nazi racial policies) and Nazi plans made before the war to populate Eastern Europe with German colonists implying that the native population would have not been either removed or exterminated. Yes, the policies for extermination would have surely been different, but they would have had largely the same effect.

“3.For a war to be morally justified, its long-run benefits have to be substantially larger than its short-run costs.”
To be blunt, I do not think there is a sound concept here. The reason is not only is the entire idea that morality is nothing but a calculus or profit-and-cost extremely problematic (the utility monster being one troubling instance), but it begs the question who is the accountant that determines the moral worth of a war? Like it or not there is only one possible accountant, God, and I don't see any reports about divine tabulations being given to the world, so God cannot be your judge.

Another problem, though not a necessary part of your argument even though it does display problems with it, is your suggestion that a just war would have a 5:1 profit-to-loss ratio. I think that this is completely arbitrary, but I will use it to get either an unacceptable position or a contradiction (I'll let the readers choose which they prefer):

To start off, I will take as self-evident that the individual will be the only judge of the potential profit-and-loss of his actions. Ergo, using Caplan's suggestion, in order for an individual to justly participate in a war there must believe that the war will have a 5:1 profit-to-loss ratio.

Let us assume that an invading army, the Black Horde, invades the free lands of Caplania. Before their invasion, the Black Horde has promised that they will only kill blonde-haired members of the Caplanian population and, using census-data gathered there, showed that they are only 1:10 of the population and own only 1:20 of the nation's wealth. All non-blonde-haired citizens of Caplania will remain unmolested.

Now we have two individuals, one a black-haired citizen of this land and another a blonde-haired. For the black-haired citizen it is clear that not fighting the Black Horde would result in a loss ration definitely less than 1:5 since only 1:10 of the population will die and 1:20 of wealth destroyed while more would probably be killed and destroyed if he were to join an army to fight against the Black Horde. However, the blonde-haired citizen is in quite a conundrum and there are two ways he could go about deciding what to do: he could either decided that his life is the most precious thing to him and decide that therefore he should fight or he could decide that his life is not worth more than other people's and decide that he should just submit to the Black Horde and die. If the former is true, then we have a contradiction since how could it be just for two individuals to take two diametrically opposed, and even incompatible, courses of actions (yes, it is possible to escape this contradiction, but I think that the metaphysical view that results is unpalatable). If the latter is true, then we must accept that one's life is not the supreme moral good and I do not believe it should ever be morally necessary to not defend oneself for the sake of other individual's lives.

Two final notes: 1. my moral intuitions inform me that giving up 1:10 of one's neighbors to certain death without a fight is repugnant and 2. emigration is not always an option available to the vast majority of people so arguing that the blonde-haired citizens could have simply emigrated in wake of the Black Horde is not a sound reply.

Dan Carroll writes:

I applaud Prof Caplan for taking a principled but unpopular stand on war. I agree that the deterrant principle does not, in itself, justify the use of military power - it is a sometimes optimal strategy after one has justified the use of military power (otherwise it would not be credible). Indeed, the deterrant for most nations is a "scorched earth" policy of exacting the maximum penalty and minimizing the benefits of would be invaders, even if victory itself is impossible.

While I agree that war is vastly overrated, and I am greatly concerned with the US policy of increasingly bombing foreigners when they inconvenience us, I think he takes it too far. Specifically, point #2 is incomplete.

While the long run benefits of fighting war are not always clear, in certain circumstances the long run costs of not fighting are more certain. It would have been cheaper for the Europeans to go to war against Hitler in 1937 than in 1939 (it would have been even cheaper to strike a more balanced treaty after WWI). More generally, war is always evil and immoral (there is no such think as a "just war"), but sometimes it is the lesser evil, such as allowing escalating genocide.

As implied, I agree with #1. The problem with #3 is that such a calcuation is impossible on any ground and therefore it can be used to disallow almost any policy or action if not limited to estimating the cost/benefit of war.

Evan writes:
For example, I don't value German civilians killed during WWII at anywhere near $5M. Those are people who gave Hitler power and then did not try to stop him. They are essentially the enemy. The fact that they don't wear uniforms does not mean they are my friends. Ditto German (private) property.
What about German babies? German toddlers? Germans who voted against the Nazis? Don't pretend you're not killing innocents in a war. You always are.

Question for all of Bryan's critics. Why aren't you trying to overthrow the government right now? I don't care what your ideology is, there must be something it's doing that you think is evil and bad. Why don't your arguments for war and against surrender apply to war with your own government? Why don't your fancy game-theoretic tit-for-tat reasonings apply to revolting against your government if it does something you really don't like (gun control, socialized medicine, war on drugs, etc.)? I'm not saying you don't have a good reason for supporting war, but not revolt. I just want to know what it is.

Once you've articulated you reason for surrendering to the government, please go back and make sure it's consistent with your reasons for not surrendering to foreign invaders. Again, I'm sure many of you do have consistent reasoning, I just want to know what it is.

Sonic Charmer writes:

Why aren't you trying to overthrow the government right now? I don't care what your ideology is, there must be something it's doing that you think is evil and bad. Why don't your arguments for war and against surrender apply to war with your own government?

Um. Because nobody here was making the blanket argument 'it's always right to engage in warfare, regardless of details, as long as at least one (>1) thing the opponent does is evil and bad', perhaps?

There's only one stance on offer here that is blanket and intended to be applied independent of details, and it's that of Prof. Caplan.

Evan writes:
There's only one stance on offer here that is blanket and intended to be applied independent of details, and it's that of Prof. Caplan.
That has not been my reading of his stance, based on his previous posts. My understanding is that he believes that war, even defensive war, is usually more trouble than it's worth because it almost always generates massive negative externalities against innocents. He doesn't have anything against fighting in a perfect world that had no externalities (I remember in one post Bryan even argued revenge could be justified), his problem is that we live in a world where it's nearly impossible not to generate externalities in the course of the fighting.

That's why I keep commenting that Bryan is seriously misusing the word "pacifism," it's usually associated with absolute, total nonviolence, even in a perfect situation with no externalities. Bryan's pacifism seems to be more of the Himura Kenshin/Vash the Stampede variety, where it's acceptable to fight, but only if you don't generate any negative externalities towards innocent people (which is, of course, really hard).

Philo writes:

You will probably fare better in the long run if you can credibly threaten people with violence, should they mistreat you. To make your threat credible, you will occasionally have actually to practice violence. This destructive behavior, though regrettable in itself, is justified: for what would happen to you if you presented no credible threat is even more regrettable.

Sonic Charmer writes:

Evan, what is 'a perfect world that [has] no externalities'? Whatever it is, we don't live there. The case you seem to want me to focus on as something Bryan would allow as an exception to his rule, has roughly zero probability of occurring. So no, I don't see that as refuting my claim that his stance is a blanket one.

keks writes:

Another item I have not seen yet in your defense (kudos for this, it is very good and a lot of work) is whether it is wise for an individual, in some circumstances, to remain a pacifist.
Imagine you're a Jew in the Warsaw Ghetto, or a Libyan military officer having risen up against Ghaddafi and you have no apparent or easy flight route. Your choice is to either 1) fight against your oppressor, with high likelihood of death in battle but also a chance of defeating your opponent, however small; or 2) remain pacifist, do nothing, with p approaching 1 that you'll be tortured to death a few days down the road.
In both cases you've obviously got small probabilities that you'll make it. And you might, say, assign a 4:1 chance you'll make it with the first option rather than the second one. To ensure your own survival, it is better to choose not to be a pacifist. In the long run, being pacifist, paradoxically, might wipe out pacifism?

scineram writes:
That's why I keep commenting that Bryan is seriously misusing the word "pacifism," it's usually associated with absolute, total nonviolence, even in a perfect situation with no externalities. Bryan's pacifism seems to be more of the Himura Kenshin/Vash the Stampede variety, where it's acceptable to fight, but only if you don't generate any negative externalities towards innocent people (which is, of course, really hard).

They were actually more pacifists than Bryan, somewhere between him and absolute pacifists. They refused to kill even enemies.

Warsaw Uprising in 1944 and Warsaw Ghetto Uprising in 1943 are completely unrelated historical events and you've mixed up both.

Don't worry, it's a common mistake.

PrometheeFeu writes:

I reject #3: "For a war to be morally justified, its long-run benefits have to be substantially larger than its short-run costs."

I replace it with: "For a war to be justified, its properly discounted benefits have to exceed its properly discounted costs."

If there are immediately apparent benefits to be gained, those should most definitely be taken into account.

PrometheeFeu writes:

The point regarding the reckless endangerment of innocent lives is actually more complex than you seem to think Bryan. If I am under attack, my life is being put recklessly in danger. I can if I organize to defend myself with others move around where the danger is placed. But ultimately, I have not necessarily created more danger. I may have even decreased the amount of danger.

Furthermore, I find the conclusions of your arguments to be ultimately absurd. Let's imagine a world in which we have bloody dictators bent on conquest and beautiful libertarian pacifist paradises. Crank that model through a couple iterations and the bloody dictators will have conquered the beautiful libertarian pacifist paradises encountering no resistance. We now have a world where the remaining bloody dictators fight it out. The conclusions of the actions you are proposing Bryan is the creation of a world where might makes right. I reject that as being a morally unacceptable outcome for any libertarian.

Evan writes:
Evan, what is 'a perfect world that [has] no externalities'? Whatever it is, we don't live there. The case you seem to want me to focus on as something Bryan would allow as an exception to his rule, has roughly zero probability of occurring. So no, I don't see that as refuting my claim that his stance is a blanket one.
Any situation where engaging the enemy without being likely to kill any bystanders is possible counts. So shooting at the enemy as they're crossing your border, attacking an invasion fleet, bushwacking a patrol, etc. all count. Rare circumstances in which the enemy is likely to kill all the bystanders fairly soon anyway (like the Warsaw Ghetto Uprising) probably also count. This is a moderately common scenario in defensive wars, but many modern wars that superpowers engaged in don't fit this model.

Also, some wars transition from a fairly just scenario with no externalities to a much less just scenario. The Korean War, for instance, was a perfectly moral war until MacArthur idiotically decided to follow the destroyed NK army over their border. If only he'd stopped there China wouldn't have gotten involved and there'd have been so much less bloodshed.

I'm not as extreme as Bryan is, I recognize that there are some instances where you have to risk killing bystanders. To protect from the anti-foreign bias, I'd say it's acceptable to attack an enemy city and risk killing civilians if you would, in an otherwise similar situation, attack an enemy-occupied city and risk killing your own civilians.

Tracy W writes:

Darjen, why do you say that so confidently? The Germans apparently were committing some atrocities in the occupied areas. I had always been told by my British and NZ history books that the atrocities were British propaganda, and always believed it, but now I'm not so sure. See for some of what I talked about. What the French faced from surrendering was not necessarily as mild as "some higher taxes".

Pandaemoni writes:

I posted this in a different thread, but for some reason it never made it past the filters...anyway:

Question: What's the furthest you'll take this argument when the innocent bystanders aren't foreigners?

Your argument on this point reminds me of Adam Smith's hypothetical about an earthquake that swallows China:

Let us suppose that the great empire of China, with all its myriads of inhabitants, was suddenly swallowed up by an earthquake, and let us consider how a man of humanity in Europe, who had no sort of connexion with that part of the world, would be affected upon receiving intelligence of this dreadful calamity. He would, I imagine, first of all, express very strongly his sorrow for the misfortune of that unhappy people . . . . And when all this fine philosophy was over, when all these humane sentiments had been once fairly expressed, he would pursue his business or his pleasure, take his repose or his diversion, with the same ease and tranquillity, as if no such accident had happened. . . . If he was to lose his little finger to-morrow, he would not sleep to-night; but, provided he never saw them, he will snore with the most profound security over the ruin of a hundred millions of his brethren, and the destruction of that immense multitude seems plainly an object less interesting to him, than this paltry misfortune of his own. To prevent, therefore, this paltry misfortune to himself, would a man of humanity be willing to sacrifice the lives of a hundred millions of his brethren, provided he had never seen them? Human nature startles with horror at the thought, and the world, in its greatest depravity and corruption, never produced such a villain as could be capable of entertaining it. But what makes this difference? When our passive feelings are almost always so sordid and so selfish, how comes it that our active principles should often be so generous and so noble?
Theory of Moral Sentiments (1759)

The hypothetical remains if it's not the loss of his own finger, but the death of a countryman. There are circles of increasing indifference as one takes the loss outward from one's self to close friends and family, then to more removed kith and kin, then to neighbors, then fellow citizens, then fellow westerners, etc. The more narrow the "ingroup" the more I will internalize their loss. By the time you get to innocent bystanders who happen to be citizens of an enemy nation, you are very clearly looking at people we would tend to identify as "outgroup."

Evolution bred into us a strong set of ingroup/outgroup biases. Who is in the outgroup is obviously a very subjective matter, and for anyone who believes in objective morality, then the treatment of outgroup innocent bystanders should be no different than the treatment of ingroup innocent bystanders. In that case, one should never apply different standards to the kind of actions that are moral than one would apply if one's own friends and family were likely to be in the area when that action was taken.

That said, the ingroup/outgroup distinction is hard to brush away, and we all struggle hard at times to see justify the differences. Taking a defensive action is perfectly moral, but if it is okay for a military to blow up a building with a rocket to kill an insurgent sniper, then it shouldn't matter whether the building is in Baghdad or Baltimore. In that case, police should arguably have the right to blow up buildings if armed hostiles are inside shooting at them.

In fact, that has happened in the U.S. Contrary to what Caplan suggested, no one went to jail when, in 1985, Philadelphia police dropped a bomb made of C-4 and dynamite on a row home in North Philly, even though they burned down multiple city blocks (and more than 60 homes) as a result and killed all but two of the men women and children living inside (six adults and five children died). The crime these people had committed? They were violating Philly noise laws in the middle of the night (shouting religious and political rhetoric with bullhorns), they may have had inappropriately large compost heaps. They also started shooting at police who came to investigate those crimes.

Move Bombing

I personally cannot completely shake my ingroup and outgroup biases. Like the man in Smith's hypothetical, if my brother lost his right hand on the same day all of Iran was swallowed by the Earth, my brother's injury would probably cause me the greater concern. The question I do not know the answer to is whether my instincts are correct, or whether my rationality should overrule my instincts in these cases. I have a suspicion that evolution honed my instincts, but it honed them to enhance my survivability, not my morality.

tracy w writes:

And Darjen, while we are at it, how do you know that the Germans wouldn't have gone on to do worse things? Countries, including democracies, often do bad things to their colonies. Eg famines in British India, slavery in the Spanish empire, stupid mercantalist policies that impoverish both colony and colonist. Plus the distinct possibility that the German empire might have become fascist even without Versailles, like the Japanese empire.
Your view of the choice the French faced in WW1 strikes me as way too optimistic.

Jon writes:

It's premise 2) that is the weakest, and I made (perhaps overlong) stab at attacking it here.

In a nutshell, I think you grossly overestimate your ability to determine the long-run consequences, and you may not even realize you are doing it.

Dave writes:

As someone who opposed the Iraq and Afghanistan wars on very similar utilitarian grounds (with a bend more towards "what's in it for us?"), I understand your points. But by painting all conflict with one broad brush, you're missing the other side of your premises:

1) In some, but not all cases, the immediate costs of inaction are clearly awful. The Rape of Nanking, for instance. The Holocaust. Slavery. There are more times in history than we'd like where innocent bystanders are killed, tortured, and abused and where pacifism does not mitigate further killing.

2) The long-run benefits of acquiescence are highly uncertain. You make several bold predictions about how any player in World War I which surrendered would have staved off a host of 20th century atrocities, along with Nazism, Communism, and the like. You cannot be certain of these benefits any more than a "hawk" can be about the long-run benefits of war. See The Trail of Tears, for instance.

3) For inaction to be morally justified, its long-run benefits need to be substantially larger than its short-run costs, too.

Aeon J. Skoble writes:

Dave- good points. Also Rwanda.

Lemmy caution writes:

I think the consequences of the united states saying that they would not get involved with another war would not be very drastic. Who will the bullies be. Canada? Mexico? Some "red dawn" scenario? The continental united states has not been at risk for a long time.

Actual bullies don't pick on the biggest kid in the class when he says he isn't going to get into fights anymore.

I am not sure that there is a more successful aggressor nation than the united states. Wars against the native Americans and Mexico have given the united states the vast defensible expanse of the continental united states. The united states should stop while it is ahead.

John writes:
Bryan's pacifism seems to be more of the Himura Kenshin/Vash the Stampede variety...
I can't believe you just made me look up an anime reference.
Tim Starr writes:

All your premises are misconceived:

1) High immediate costs of war: This assumes that the "costs of war" are caused by "war" without looking at the causes of war itself. Aggressors start wars without their victim's consent, by definition. Their victims don't get a choice about whether to go to war, they only get a choice about whether to fight back or surrender. Thus, the costs of war are all the fault of the aggressors, not the defenders. A better term for this would be "the cost of aggression," not "costs of war." Furthermore, you completely overlook the high immediate costs of not fighting back against aggression. Several of the countries occupied by Nazi Germany (e.g., Austria, Czechoslovakia, Denmark, etc.) didn't fight back. The result was that their people were subjected to totalitarian rule, enslavement, and millions of them were mass-murdered. The Soviets started out WWII at peace w/ the Nazis (the Hitler-Stalin Pact), but that hardly led to a low-cost war experience for the Soviets when the Nazis invaded.

2) The benefits of war: The greatest omission in the libertarian pacifist interpretation of history is that wars often increase freedom. Freedom was born in Ancient Greece out of the Persian Wars. The French Revolution ended serfdom and capitation taxes in continental Europe. Britain's participation in the Napoleonic Wars turned it from a protectionist slave trading naval empire with impressment where labor unions were illegal into one that banned the slave trade worldwide, abolished slavery in its colonies, abolished impressment, and legalized labor unions. The British navy turned away from privateering to suppressing piracy.

Your causal analysis of the long-term costs of war remains flawed. WWI didn't cause WWII. WWI & WWII were started by the undemocratic great powers (Russia, Austria, & Germany in WWI, Germany, Russia, & Japan in WWII). If Germany had been democratized after the Revolutions of 1848 instead of remaining an absolute monarchy, it wouldn't have started WWI. If Germany had been fully democratized after WWI, Hitler would've never come to power & started WWII.

3) Your cost-benefit analysis posits long-run benefits against short-run costs. This omits short-run benefits and long-run costs. It also omits the unseen costs, both long- and short-run, of inaction.

Vincent Cook writes:


Practically every category of human action, and not merely acts of self-defense or of retaliation, has some potential to cause harm to others.

If you interpret the non-aggression principle to mean a strict moral obligation to avoid all actions that pose the slightest risk of causing a rights violation, then that principle would obligate every moral agent to do the impossible, or at least obligate moral agents to withdraw from society and live in autarkic isolation from one another. In this first interpretation, the real moral objection is to posing any risk to others, not to state-sponsored warfare per se. One could be just as fearful of driving a car down the street for fear of running the neighbor's kid over as one would be of shooting at an outlaw and accidently hitting an innocent bystander.

On the other hand, if you interpret the non-aggression principle to mean that one must take responsibility for each instance where one violates the rights of another (presumably by submitting legal disputes to arbitration and by paying any arbitration awards to the victims of one's aggressive acts), then it is morally permissible to put others at risk as long as one is willing and able to pay for any harm one causes, and as long as one doesn't evade the legal processes by which the nature of one's actions can be impartially adjudicated. In this second interpretation, the right of retaliation likewise is limited to targeting those outlaws who refuse justify their actions in arbitration proceedings or who refuse to compensate those identified in such proceedings as their victims.

In the second interpretation, we no longer have pacifism in the strict sense you have defined it. It is still possible to use force as long as one is prepared to take responsibility for one's acts in the manner outlined above. Nevertheless, there still are ample grounds for criticizing the state's war-making under the second interpretation. It is not simply the case that the state harms innocents in its pursuit of alleged aggressors; the state also refuses to submit its cause to impartial adjudication and to compensate the innocents that it harms.

You might well believe that, under an obligation to compensate, retaliation would be generally impractical. Your personal cost-benefit calculus might conclude that the costs of compensating innocents for collateral damage would always overwhelm whatever benefits would accrue from destroying the outlaws who have harmed you.

However, it is important not to confuse matters of principle with your personal cost-benefit analysis. Another person could still comply with the second interpretation of the non-aggression principle if they were willing and able to pay whatever compensation was required of them.

There is nothing in the non-aggression principle itself that specifies what price is worth paying for deterring aggressors. All the non-aggression principle really demands is that one does not shirk the costs associated with the harm one causes.

Likewise, the crucial problem with the state's war-making is precisely that the politicians always strive to shift the costs of the state's use of force onto others without their consent, whether these others be taxpayers, draftees, or innocent bystanders. In that sense, anarcho-capitalists are anti-war even if they don't object to all uses of force as under your definition of pacifism.

Best regards,

Gil writes:

"Evil will always triumph over good because good is dumb".

By saying good people can't ever fight back because fighting is the preserve of an evildoers means evil people wins. Evil people have no qualms about "collateral damage" and therefore have the highest chance of achieving their goals.

Vincent Cook writes:


It has been awhile since we have discussed these matters, but reading your response to Bryan brings a couple of observations to mind.

First, innocent bystanders are under no moral obligation to pay for the costs of fighting your enemies. Most of the alleged benefits of the various wars you cite are simply irrelevant to the issue Bryan has raised--if such an obligation doesn't exist, then it doesn't matter how many Nazi dragons you slay. You still owe them compensation when you inflict collateral damage on them.

If you harm these bystanders, it is not rational for them to dismiss their injuries on the grounds that the devil made you do it, or that Hitler made you do it. The fact of the matter is that if you act, you are the cause of whatever consequences follow from your acts. To ask others to ignore your role as a causal agent, and to pretend that your enemies are automatically their enemies too, is to repudiate their status as independent rights-bearing moral agents.

Second, even if you could justify some sort of collective calculus for evaluating wars, the particular examples you cite leave you open to the charge that you are cherry-picking the outcomes you like. In each of the wars you cite above, one could also come up with compelling reasons for thinking that each war might have set back liberty in some ways.

There is a lot I disagree with in your interpretation of history, but let me stick to one war, namley World War II. In citing the Nazi threat, you neglect to mention that a couple of the enemies of the Axis Powers, namely Mao and Stalin, wound up killing and enslaving even more people than Hitler did. Chiang Kai-Shek was no slouch in the democide derby either, and the British, French, Dutch, and American Empires were not without their criminal features.

When someone goes around proclaiming that this was the "good" war, how does one go about deciding that sacrificing someone to Stalin is more than counterbalanced by saving someone else from Hitler?

Best regards,

Larry Ruane writes:

I agree with Bryan. Here is one of the most beautifully written articles (especially since it describes a personal story) that I have ever read on this topic, and it had a profound effect on my thinking (so much so that I still remember reading it in 2006):

I Don't Have to Fight You

(By David R. Henderson)

Tim Starr writes:

Vince: Good to hear from you, I've been meaning to write you privately about another matter. To respond to your points in turn:

1) Innocent bystanders: The subjects of an aggressor regime may or may not be innocent. If they are supporters of the regime, then they are legitimate targets to the extent that they contribute to the regime's military capacity. If they are not supporters of the regime, then they are obligated to oppose the regime as much as they can, and to stay out of the way of enemy attacks upon the regime as much as they can. As for compensation, Germany and Japan were compensated after the war with the democratization of their political systems, greatly increasing their freedom, direct economic aid to help them rebuild, and military protection against external enemies (e.g., the Soviets) ever since.

2) WWII: While it's true that the alliance w/ Stalin compromised the morality of the Allied cause in WWII, at the time Germany was the greater threat, as it had a First World economy and military at its disposal and was rapidly developing the capability to project power across the Atlantic via the Amerikabomber and the naval buildup that was cut short by the outbreak of war, and he already had a big enough U-boat fleet to lay siege to Britain. Britain could only win the Battle of the Atlantic with the addition of the naval capacity of the USA. Hitler's ultimate goal, as expressed in his Second Book, was war with the USA. Also, Stalin's occupation of central and Eastern Europe was not a necessary consequence of WWII, as Patton had the men, material, motive, and opportunity to invade and get rid of Stalinism once and for all right before his death. He may even have been kileld for it. Mao's triumph in China was due to the US withdrawal of support for Chiang Kai-Shek. Mao was unquestionably far worse than Chiang Kai-Shek.

3) Cherry-picking: I'm actually trying to address what I consider the strongest counter-examples against my position. There's a lot more that I've left out. For instance, one of the long-term effects of the French Revolution was the demise of slavery in much of the world. The French invasion of Spain weakened Spain's hold on most of its colonies in the Western hemisphere, thus enabling them to win their independence in the original "Wars of Liberation." Those new regimes all abolished slavery. After the American Civil War, the only slave states remaining in the Western hemisphere were Cuba (still a Spanish colony until the Spanish-American War) and Brazil (a Portugese colony).

It may be that on balance any particular war was more bad than good. However, I rarely see any of the good taken into account by self-avowed pacifists like Caplan. If pacifism is to be a respectable and attractive position, then it must be after a full accounting.

CBrinton writes:

Tim Starr: "After the American Civil War, the only slave states remaining in the Western hemisphere were Cuba (still a Spanish colony until the Spanish-American War) and Brazil (a Portugese colony)"

Not quite--slavery still persisted in Puerto Rico, also a Spanish colony. There were still some slaves (maybe a couple thousand; no good estimates seem to exist), remaining in Paraguay.

But your larger point, that Western Hemisphere slavery was abolished in large part by war, is of course correct.

Tim Starr writes:

CBrinton: Thanks for the correction. I didn't know about Puerto Rico (although it stands to reason that it should have the same status as Cuba), and forgot about Paraguay. As I recall, slavery was weakened in Paraguay when slaves were granted freedom in exchange for military service in a war Paraguay fought with one of its neighbors.

Brazil is Jeff Hummel's favorite example of peaceful abolition. However, military force played a big part there, too - specifically, the Royal Navy, which became supreme worldwide thanks to winning the Napoleonic Wars and was committed to enforcing the ban on the slave trade. Brazil refused to ban the slave trade until the 1850s, when Britain destroyed several of its slave ships in Brazilian waters. Then Brazil banned the slave trade, thus starting the inevitable demise of slavery in Brazil, as Brazil was like all slave societies in the Western hemisphere except the USA in that it had negative population growth and was thus dependent upon constant importation of new slaves. Within a generation, slavery was abolished in Brazil, too - the last slave society in the Western hemisphere.

Which is not to say that slavery was completely gone, of course. Mexico still had widespread illegal slavery before the Revolution, and in the United States we re-imposed slavery by another name after Reconstruction.

Comments for this entry have been closed
Return to top