My interest in this puzzle was sparked by the Israeli Supreme Court’s decision to strike down the “Early Warning Procedure” employed by the Israel Defense Forces (IDF) in the West Bank. Under the Procedure, the IDF would approach a neighbor of a suspected Palestinian militant and request the neighbor to urge the suspect to surrender quietly to the security forces. If the suspect refused, the neighbor would then attempt to clear the residence from its other inhabitants. The stated goal of the Procedure was to reduce potential casualties, both among IDF and local civilians, in case the arrest turned violent. Despite some evidence that the Procedure was effective in reducing civilian casualties, the Court ruled that it violated strict prohibitions on the reliance on local civilians by an Occupying Power for security operations, and was therefore unlawful.
The prohibition cited by the Israeli court is but one instance of IHL’s absolutist stance. Others include the prohibitions on mercy killings, the assassination of rogue leaders, the use of non-lethal chemical weapons, or the intentional killing of any civilian – even where such actions are taken with the attempt to minimize humanitarian harm. The claim that certain prohibited acts might actually lead to the saving of innocent lives, even many thousands of lives, is categorically rejected by the laws of war. Put bluntly, in many cases IHL demands an excessive sacrifice of lives for preserving the integrity of the law.
What is particularly odd is that IHL does take into account military necessity in crafting its rules. But it has no comparable procedure for "humanitarian necessity". In other words, international humanitarian law can be bent for military ends, but not humanitarian ends.
My first thought on reading this thesis was skepticism, simply because everyone argues, or could argue, that their military actions ultimately saved lives (by bringing the war to a close sooner, for instance). Professor Blum, though, anticipates this objection and crafts her definitions accordingly.
The blueprint definition I ultimately suggest is designed to work in a way that would allow us to distinguish the “right” case from all the wrong ones. It is as follows:A person shall not be criminally responsible if, at the time of that person’s conduct: . . . The conduct which is alleged to constitute a crime was designed to minimize harm to individuals other than the defendant’s compatriots, the person could reasonably expect that his action would be effective as the direct cause of minimizing the harm, and there were no less harmful alternatives under the circumstances to produce a similar humanitarian outcome.
Three elements of this definition are worth emphasizing (and all elements are open to further debate and examination). The first is that to benefit from a humanitarian necessity justification, the actor must show that the violation of the law was designed to benefit not – or not only – his own fellow soldiers or civilians, but enemy nationals. The rationale behind this condition is that IHL was designed to curb the aggressive tendencies of parties in war and offer protections to the most vulnerable. It is therefore preoccupied primarily with how parties treat their enemies, not how they treat their own people. For the humanitarian necessity justification to be compatible with the project of IHL, it must follow a similar logic. Consequently, the paradigmatic case of interrogational torture, most commonly used to avert an attack on one’s own people, cannot be justified on the basis of a humanitarian necessity.
A second, straightforward element of the justification is that the actor used the least egregious means possible in choosing between two evils. Following this condition, whatever one’s judgment is of the atomic bombing of Hiroshima, the bombing of Nagasaki – just three days after Hiroshima and without testing alternative means of securing Japanese surrender – could not be justified under a humanitarian necessity.
A third crucial element is causation: The justification depends on a direct causal relation between the breach of law and the aversion of harm. This condition follows from the internal logic of IHL, which does not allow for unbounded cruelty in the name of bringing wars closer to end.
Anyway, it's a provocative and interesting idea. The folks at OpJur have this article as part of their online symposium, with a response from Matthew Waxman and reply from Professor Blum. Check it out.