In an argument on the (supposed) logical distinction between jus ad bellum and jus in bello, Michael Walzer argues that war is different from criminality because while war is an activity governed by rules and conventions, criminality is not:

The crucial point is that there are rules of war, though there are no rules of robbery (or rape or murder). The moral equality of the battlefield distinguishes combat from domestic crime (Just and Unjust Wars, p. 128).

Even apart from questions about the logic of this argument–whether Walzer’s intended conclusion follows from the premises–his main premise strikes me as obviously false. There certainly are “rules of robbery.” The basic rule of a robbery is: if you hand over your valuables, you’ll be allowed to live; if not, not. This is an unjust rule, but it’s certainly a rule. The rule can of course be violated; it would be naive in a given case to expect strict adherence to it. But I think it’s unquestionably “the rule of robbery.”

Does murder have a comparable rule? I think so. The rule is: if you’re going to kill someone, then pro tanto, you should kill them as quickly and expeditiously as possible, getting it over with. The rule is incompatible with torturing someone as a prelude to killing them, or using an inefficient method of killing someone that causes great suffering or pain. The implication of the preceding rule is that if torture is to be employed, it requires a different rationale than mere murder. Again, all unsavory, but still rule-governed.

What about rape? Criminologists commonly distinguish three kinds of rapist: power-assertive, opportunistic, and sadistic. The first two kinds of rape are obviously rule-governed. Power-assertive rapists seek dominance, and so, avoid inflicting more harm than they judge “necessary” to assert dominance. “Only harm them if they struggle” is the rule. Opportunistic rapists regard themselves as engaging in legitimate sex acts; such rapists adopt rules that reinforce the illusion of doing so. For instance, some may avoid sexual contact with children; others will insist that the victim “wanted it.” There are too many variations here to list them all, but the point is, I think it’s plausible to think that rule-following is involved. Evil, but rule-governed.

Sadistic rapists, by contrast with the first two, aim specifically to inflict pain. Here, the only rule is: do what inflicts the most pain. Is that really a rule in the relevant sense? Maybe not. This might be a case where the act is too unstructured to be genuinely governed by a rule. But one exception doesn’t really help Walzer’s case. His aim is to distinguish criminality from warfare. There are sadistic crimes, and there are sadistic wars or acts of war. Both are exceptions to a general rule about rule-following. That fact establishes a parallel between warfare and criminality, not a principled distinction.

Take an example not on the list: burglary. Is there a rule here? I think so. The rule is: take what’s monetarily valuable; leave what isn’t. A burglar, qua burglar, doesn’t take wedding pictures or items of sentimental value simply because they’re there. He steals; he doesn’t desecrate. Desecration is its own crime, involving its own rules.

Does every crime have rules? I don’t know. I’d have to go through them all. But I think it’s clearly false to assert or imply, as Walzer does, that there are no rules of crime, whether we take “rules” to be “purely instrumental counsels of prudence,” or we take them to be “interpersonal limits on action that function as defeasible expectations on what’s to happen.”

So I think Walzer’s venture fails from the outset. War is more like criminality than he suggests; criminality is more rule-governed than he realizes. On top of that, the supposed rule-governedness of warfare doesn’t do much to support the supposed logical independence of jus ad bellum and jus in bello, whether war is fundamentally distinct from criminality or not. In any case, total warfare is no more (or less) rule governed than sadistic crime. It’s just the same thing, writ large. This doesn’t prove that jus in bello is logically parasitic on jus ad bellum, but it does prove that Walzer’s method of rendering them independent of each other is misconceived.