The following is part of an exchange between Just Security’s David Luban and Guest Author John Fabian Witt regarding Witt’s new book Lincoln’s Code: The Laws of War in American History. Luban’s original book review was published yesterday, which was followed by Witt’s reply.

John Fabian Witt’s response to my review of Lincoln’s Code concerns two points: whether I got Lieber right on the subject of military necessity, and whether I got military necessity right.

The first point is of historical and scholarly interest, and I willingly defer to Professor Witt’s verdict: Lieber’s restriction of military necessity to lawful actions “indispensable for securing the ends of war” (article 14 of the Lieber Code) was not something Lieber really meant. Rather, it was the sloppy product of hasty drafting without thorough outside review. Professor Witt knows this history far better than I ever will. Even so, article 14 has precedential value whether Lieber meant it or not. Certainly the second clause of the article, which says that only lawful actions can count as military necessities, is crucial to today’s understanding. Pleas of military necessity cannot excuse war crimes or other law of war violations.

Professor Witt’s second point is worth pursuing, because the tension between military necessity and humanitarianism remains a very live issue today. How broadly should we understand military necessity, and how much should appeals to military necessity be allowed to justify or excuse conduct that harms civilians?

I contrasted Lieber’s language with the leading post-war precedent, the List judgment in the second round of Nuremberg trials. List permits any lawful act that yields military advantage to count as military necessity—even if the advantage is nothing more than a small saving in time or money. Clearly, this doctrine confuses military necessity with military convenience.

Take an example involving a tank operating in an urban environment. In combat operations, where time is of the essence, the tank’s driver may decide to take the quickest route to his destination, even if it means rolling over, and destroying, cars parked on a narrow street. But suppose time is not of the essence, and the driver still chooses to take the quickest route, even though he could spare the cars by going a block out of his way. Clearly, the first case is one of genuine military necessity; the second, mere military convenience. Yet the List formula regards both of them as military necessity.

So it seems to me that whether Lieber meant it or not, there is something important in limiting harmful actions taken out of military necessity to actions that really are “indispensable” to achieving some military end. Otherwise, the civilian harm is, literally, unnecessary.

That’s why I think the Israeli Supreme Court’s 2004 decision in Beit Sourik made real progress. The case concerned the separation barrier Israel built during the second Intifada, and it ordered the Israeli Defense Forces to relocate part of the barrier to reduce hardship to Palestinian villagers. Minor military advantage isn’t enough.

Where does military necessity come in? The IDF commander asserted that building the barrier was a matter of military necessity, and the Court agreed, finding that the commander was making a security judgment, not a political one. (¶¶29, 32) As for the route, the commander asserted that no alternative route would work as well as the chosen one, and here too the Court deferred to the military commander’s judgment, even though Israeli military experts testifying for the Palestinians disagreed. (¶¶46-47, 56).

Under the List doctrine, that would have been the end of the story. The military showed that the chosen route yielded the most military advantage, and that clearly makes it military necessity under List’s definition. In fact, that finding might be the end of the story even the language of indispensability. If the chosen route confers marginal military advantage over all the alternatives, it is indispensable to gaining that marginal military advantage. (Admittedly, that reading waters down indispensability in a way that makes it nearly identical with mere convenience. But Professor Witt thinks that is what Lieber actually had in mind.)

But military advantage wasn’t the end of the story for the Israeli court, and that’s what makes Beit Sourik so important for the doctrine of military necessity. The Court insisted that it matters how much marginal military advantage the chosen route yielded, compared with the marginal harm to Palestinian civilians when compared with the militarily next-best alternative. If the chosen route provides only a tad more security, but inflicts a lot of additional hardship, it is illegal. (¶¶59-62)

Weighing the IDF’s chosen route against alternatives, and weighing military advantage against civilian harm, both sound like proportionality judgments, and that is indeed what the Israeli Court calls them. For that reason, Professor Witt concludes that Beit Sourik is a proportionality case, not a military necessity case.

I disagree, because I don’t think this is an “either-or.” By insisting that military advantage isn’t the end of the story, but must be weighed against civilian harm, the Court’s proportionality tests transform the doctrine of necessity away from List and Lieber.

One way to think about Beit Sourik is that the Israeli court left the concept of military necessity unchanged, but balanced it against humanitarian considerations—a reading that the Court itself suggests (¶32). But I think a better description it is that the Court recognized that necessity judgments are themselves disguised proportionality judgments.

A claim of military necessity is supposed to answer the question “Why did you do this injurious thing?” It answers: doing it was necessary. To the natural follow-up “Necessary for what?”, the answer is “necessary for the marginal gain in military advantage over the next-best alternative.” But that answer only has weight if the marginal gain matters. If the marginal gain is very slight, it doesn’t justify inflicting extra suffering on civilians. So, if claims of military necessity are going do any justificatory work—and what else are they supposed to do?—they always must answer to questions of proportionality.