It’s far from clear to me that that’s true; soldiers, more or less by definition, are people who are more than usually willing to be killed in warfare. They are also typically engaged in (actual or attempted) killing of their own.
On the other hand, they commonly get killed in much greater numbers than civilians do, even when international law is being disregarded.
So the principle in question is really something like “If something doesn’t merit killing a small number of innocent third parties, then it doesn’t merit killing a large number of people who are actively trying to kill your people and who are willing to risk death in the process”.
Of course, that doesn’t make such a good soundbite. But I think the details that make it less of a soundbite have to be considered when evaluating it.
(This comment has been significantly edited since its original posting.)
Well, my thinking was more along the lines of “If it’s okay to kill soldiers, then it’s also okay to kill the civilian farmer growing food to feed the soldiers, without which the soldiers would be less able to fight.” In other words, I am indirectly arguing for an end to civilian immunity in war.
There is a well-known and proven technique that allows a conventional fighting force to defeat a guerrilla insurgency. It is known as collective punishment. All that is required is that the civilian population be more afraid of the conventional army than it is of the guerrillas; if they are sufficiently terrorized, they will turn over the guerrillas to the conventional forces. If that fails, the conventional forces can simply attack the population itself. Level a city or two, as Syria did, and soon you’ll have much less of a problem with insurgents. ;)
In a symmetric war, not targeting civilians is cooperating in an iterated prisoner’s dilemma; you don’t want to switch from C/C, except in the (seemingly very unlikely) event that the war will end so much more quickly that overall suffering is reduced.
What you say seems correct as a matter of pure terminal values, but this also seems like a great example of a situation where common-sense values that claim to be terminal* are implausible as such but contain real instrumental wisdom.
* to the incomplete extent that common-sense morality makes this distinction
I’ll agree that, a lot of the time, going out of one’s way to kill non-combatants just isn’t very useful. They’re non-combatants, after all, so they’re much, much less of a threat. It’s generally more efficient to kill only the people shooting back at you. But if your cause isn’t worth killing civilians over, should it become expedient to do so, then maybe it’s better not to resort to violence in the first place. Indeed, you can’t send a foreign army anywhere and expect to be considered liberators, or, at least, you can’t expect to be considered liberators for very long. If you’re going to invade some other country and expect to have a lasting influence on it, you’re a conqueror—so if you want to be successful, you have to admit that conquering is what you are doing, and do the job right!
(The U.S. really sucks at conquering these days; we haven’t conquered anything successfully since World War II ended, and I’m not sure that really counted. I suspect the last time we really conquered something was during the Philippine-American War...)
If killing the children of suicide bombers deters future suicide attacks, then, by all means, let’s all go out and kill us some children! (Although I suspect that killing their parents might be more effective, as most suicide bombers are unmarried.)
I also endorse a general policy of shooting through human shields; hostages won’t be taken if they are of no value, and intermingling civilian and military facilities, as Hezbollah does whenever it builds a school on top of an ammo dump, is already recognized as a war crime under international law.
Crono: If it is ok to kill soldiers it is ok to kill civilians. And vice versa.
gjm: But wait there are important differences between soldiers and civilians. Soldiers kill and are often willing to be killed. We should consider these distinctions in evaluating this issue.
Crono: Civilians are part of the war effort too and are willing to help in killing. (Adding after my comment: Collective punishment is a good tactic!)
Me: Some people aren’t part of the war effort. Protesters, children.
Crono: If killing children prevents suicide bombers, do it. Also, look at these other cases where we should kill civilians.
You’ve given some arguments for why you think killing civilians is good strategy in some circumstances. But I don’t see how this answers the issue gjm brought up. Even if there are circumstances in which killing civilians in justified it doesn’t follow that reasons for killing soldiers are good enough reasons to kill civilians. It seems like the fact that soldiers are more willing to die and that they are likely to kill if they aren’t killed first are very good reasons for requiring weaker justifications for soldier killing than civilian killing. Why is gjm’s point wrong?
Also, as this is a discussion about international law the question is about setting norms for war fighting. As such, do you think collective punishment and killing the children of soldiers to deter them make sense as norms governing conduct during wars?
I would also note that while you’ve given evidence that breaking international rules regarding killing civilians can be useful, the question is whether or not the benefits outweigh the costs. How much shorter will the war be if you kill all the farmers? How many fewer men will die in total because you nuked that city? How much less will future generations value human life because rape and torture are considered acceptable means to an end?
My knee-jerk reaction is that having these standards may indeed reduce overall casualties and that they are important for international image and future cooperation. I am open to evidence to the contrary though.
My knee-jerk reaction is that having these standards may indeed reduce overall casualties
I think that’s the argument that by default requires proof, not the other way around. Intuitively, having external independent standards can only prevent me from best accomplishing my goal, whatever my goal may be. If my goal is to kill while also killing few civilians, I’ll go for that, but I’ll do it more efficiently in most cases than if I have to follow laws I don’t believe in.
My knee-jerk reaction is that having these standards may indeed reduce overall casualties
I think that’s the argument that by default requires proof, not the other way around.
I would say the opposite, out of conservatism, but I don’t expect to be able to argue the point as anything more than headbutting intuitions.
If my goal is to kill while also killing few civilians, I’ll go for that, but I’ll do it more efficiently in most cases than if I have to follow laws I don’t believe in.
I absolutely agree with you if my own actions are the only ones I am taking into account, however I expect that the actions of others will better align with my goals if the standards are in place.
Given the high cost of implementing these standards, some level of proof (or at least a calculation of expected future utility) should be given as justification for their existence. I can think of two ways of examining this:
1) Compare the casualties (or other desired metrics) of similar conflicts before and after the implementation of standards and in situations in which standards were adhered to or ignored.
2) Try to quantify metrics of interest in a hypothetical war with or without adherence to the standards. This of course is very difficult, but I’m not willing to say impossible.
(1) doesn’t really help us predict the effect of proposed new standards that have never been tried before, and that’s what we really want to do. I hope we can find a way to achieve (2) :-)
Yes, well then it hardly seems right to kill the farmers and other civilians who don’t protest for fear of their lives and the lives of their family.
The soldiers are often fighting on pain of death too. Pretty much the only people who it is ‘just’ to kill are the leaders who are throwing men at each other for their own personal gain. Assassination should be considered the most honourable form of combat in war.
Agreed. Though with soldiers it is a collective action problem. If enough of them were willing to disobey orders they would have little to fear. This makes the soldiers somewhat more culpable than, say, children. The point about assassination is a good one.
On the other hand, many conflicts have a self-perpetuating nature independent of the specific leaders involved. Assassinating Alexander the Great may very well have saved Persia from conquest, but assassinating FDR or Stalin would have been of little benefit to the Axis powers. Assassinating Hitler may or may not have helped the Allied powers, and I have no idea what effect assassinating Napoleon would have had. If an assassinated leader’s successor simply continues their policies, then assassination does little good.
Also, an assassination was the trigger for World War I. :(
It’s far from clear to me that that’s true; soldiers, more or less by definition, are people who are more than usually willing to be killed in warfare. They are also typically engaged in (actual or attempted) killing of their own.
On the other hand, they commonly get killed in much greater numbers than civilians do, even when international law is being disregarded.
So the principle in question is really something like “If something doesn’t merit killing a small number of innocent third parties, then it doesn’t merit killing a large number of people who are actively trying to kill your people and who are willing to risk death in the process”.
Of course, that doesn’t make such a good soundbite. But I think the details that make it less of a soundbite have to be considered when evaluating it.
Volunteer soldiers, who may comprise a minority of soldiers in the world.
(This comment has been significantly edited since its original posting.)
Well, my thinking was more along the lines of “If it’s okay to kill soldiers, then it’s also okay to kill the civilian farmer growing food to feed the soldiers, without which the soldiers would be less able to fight.” In other words, I am indirectly arguing for an end to civilian immunity in war.
There is a well-known and proven technique that allows a conventional fighting force to defeat a guerrilla insurgency. It is known as collective punishment. All that is required is that the civilian population be more afraid of the conventional army than it is of the guerrillas; if they are sufficiently terrorized, they will turn over the guerrillas to the conventional forces. If that fails, the conventional forces can simply attack the population itself. Level a city or two, as Syria did, and soon you’ll have much less of a problem with insurgents. ;)
In a symmetric war, not targeting civilians is cooperating in an iterated prisoner’s dilemma; you don’t want to switch from C/C, except in the (seemingly very unlikely) event that the war will end so much more quickly that overall suffering is reduced.
What you say seems correct as a matter of pure terminal values, but this also seems like a great example of a situation where common-sense values that claim to be terminal* are implausible as such but contain real instrumental wisdom.
* to the incomplete extent that common-sense morality makes this distinction
I’ll agree that, a lot of the time, going out of one’s way to kill non-combatants just isn’t very useful. They’re non-combatants, after all, so they’re much, much less of a threat. It’s generally more efficient to kill only the people shooting back at you. But if your cause isn’t worth killing civilians over, should it become expedient to do so, then maybe it’s better not to resort to violence in the first place. Indeed, you can’t send a foreign army anywhere and expect to be considered liberators, or, at least, you can’t expect to be considered liberators for very long. If you’re going to invade some other country and expect to have a lasting influence on it, you’re a conqueror—so if you want to be successful, you have to admit that conquering is what you are doing, and do the job right!
(The U.S. really sucks at conquering these days; we haven’t conquered anything successfully since World War II ended, and I’m not sure that really counted. I suspect the last time we really conquered something was during the Philippine-American War...)
True, but these are pretty rare these days.
Then is it okay to kill peace protesters? Children?
If killing the children of suicide bombers deters future suicide attacks, then, by all means, let’s all go out and kill us some children! (Although I suspect that killing their parents might be more effective, as most suicide bombers are unmarried.)
I also endorse a general policy of shooting through human shields; hostages won’t be taken if they are of no value, and intermingling civilian and military facilities, as Hezbollah does whenever it builds a school on top of an ammo dump, is already recognized as a war crime under international law.
Summary of the thread until now:
Crono: If it is ok to kill soldiers it is ok to kill civilians. And vice versa. gjm: But wait there are important differences between soldiers and civilians. Soldiers kill and are often willing to be killed. We should consider these distinctions in evaluating this issue. Crono: Civilians are part of the war effort too and are willing to help in killing. (Adding after my comment: Collective punishment is a good tactic!) Me: Some people aren’t part of the war effort. Protesters, children. Crono: If killing children prevents suicide bombers, do it. Also, look at these other cases where we should kill civilians.
You’ve given some arguments for why you think killing civilians is good strategy in some circumstances. But I don’t see how this answers the issue gjm brought up. Even if there are circumstances in which killing civilians in justified it doesn’t follow that reasons for killing soldiers are good enough reasons to kill civilians. It seems like the fact that soldiers are more willing to die and that they are likely to kill if they aren’t killed first are very good reasons for requiring weaker justifications for soldier killing than civilian killing. Why is gjm’s point wrong?
Also, as this is a discussion about international law the question is about setting norms for war fighting. As such, do you think collective punishment and killing the children of soldiers to deter them make sense as norms governing conduct during wars?
I would also note that while you’ve given evidence that breaking international rules regarding killing civilians can be useful, the question is whether or not the benefits outweigh the costs. How much shorter will the war be if you kill all the farmers? How many fewer men will die in total because you nuked that city? How much less will future generations value human life because rape and torture are considered acceptable means to an end?
My knee-jerk reaction is that having these standards may indeed reduce overall casualties and that they are important for international image and future cooperation. I am open to evidence to the contrary though.
If you happen to be the weaker side then potentially quite a lot longer.
I think that’s the argument that by default requires proof, not the other way around. Intuitively, having external independent standards can only prevent me from best accomplishing my goal, whatever my goal may be. If my goal is to kill while also killing few civilians, I’ll go for that, but I’ll do it more efficiently in most cases than if I have to follow laws I don’t believe in.
I would say the opposite, out of conservatism, but I don’t expect to be able to argue the point as anything more than headbutting intuitions.
True of normative reasoners, not of humans. See Ethical Inhibitions.
I absolutely agree with you if my own actions are the only ones I am taking into account, however I expect that the actions of others will better align with my goals if the standards are in place.
Given the high cost of implementing these standards, some level of proof (or at least a calculation of expected future utility) should be given as justification for their existence. I can think of two ways of examining this:
1) Compare the casualties (or other desired metrics) of similar conflicts before and after the implementation of standards and in situations in which standards were adhered to or ignored.
2) Try to quantify metrics of interest in a hypothetical war with or without adherence to the standards. This of course is very difficult, but I’m not willing to say impossible.
(1) doesn’t really help us predict the effect of proposed new standards that have never been tried before, and that’s what we really want to do. I hope we can find a way to achieve (2) :-)
Their soldiers have usually done that for you.
Yes, well then it hardly seems right to kill the farmers and other civilians who don’t protest for fear of their lives and the lives of their family.
The soldiers are often fighting on pain of death too. Pretty much the only people who it is ‘just’ to kill are the leaders who are throwing men at each other for their own personal gain. Assassination should be considered the most honourable form of combat in war.
Agreed. Though with soldiers it is a collective action problem. If enough of them were willing to disobey orders they would have little to fear. This makes the soldiers somewhat more culpable than, say, children. The point about assassination is a good one.
Agreed. Especially since they have a gun and could at least KO their CO and leg it.
There are problems with a norm that says killing foreign leaders is OK, but wedrifid’s point also has merit.
For a start, paranoid leaders kill more of their own civilians than secure ones.
On the other hand, many conflicts have a self-perpetuating nature independent of the specific leaders involved. Assassinating Alexander the Great may very well have saved Persia from conquest, but assassinating FDR or Stalin would have been of little benefit to the Axis powers. Assassinating Hitler may or may not have helped the Allied powers, and I have no idea what effect assassinating Napoleon would have had. If an assassinated leader’s successor simply continues their policies, then assassination does little good.
Also, an assassination was the trigger for World War I. :(
Agreed.