On April 6 President Trump ordered missile strikes against a Syrian airbase ostensibly in retaliation for Al Assad’s lethal gassing of civilians three days earlier.

Bas van der Vossen and I have recently finished writing our book Debating Humanitarian Intervention, appearing with Oxford University Press later this year. There we address the foundational issues pertaining to the justification of wars ostensibly aimed at rescuing persons from severe tyranny, civil war, and other forms of oppression. We take opposite sides in this debate. I think that sometimes these military engagements are justified; Bas thinks they virtually never are. With the Syrian incident fresh in our minds, we thought we would write some posts anticipating our respective positions on this very topical issue. This is the inaugural post.

Literally minutes after the strikes, the blogosphere became awash with legal analyses of the strikes under both constitutional and international law. Some of those analyses are interesting, but for jurisprudential reasons I detail here and here, I regard them as inconclusive at best, and as disguised advocacy at worst, so I will not pursue them here. Rather, I examine the incident under the lens of Just War Theory. JWT addresses the question whether the strikes are morally justified.

Under JWT a military action is justified if it meets (at least) two conditions: the cause must be just, and the action must be proportionate, that is, it must not cause excessive damage.

Just Cause

A war is just if, and only if, it is in defense of persons. If it is a war in defense of my compatriots it is called national self-defense. If it is a war in defense of others it is called humanitarian intervention. The United States government did not conduct the Syria strikes to defend Americans, so it was not a war in self-defense. Ostensibly, President Trump ordered the strikes as retaliation for the massacres perpetrated by the Syrian regime with chemical weapons against civilians. This action by Al Assad qualifies as a crime against humanity  -a new one, as on this he is a notorious recidivist. Was this, then, a humanitarian intervention? A possible skeptical answer is that the massacres had already occurred, so the strikes are objectionable because they are simply punitive, and not aimed at stopping an ongoing atrocity. Some believe that punitive wars are not permissible. Without resolving that issue, it is reasonable to interpret the strikes as aiming to deter the Syrian regime from committing future crimes against humanity, given that the perpetrator now knows that dire consequences would follow. The just cause, then, is not to stop ongoing atrocities but to prevent future ones. In that sense, I think the strikes qualify as humanitarian intervention. A problem remains, however: in order for the United States to have a just cause, the strikes must effectively deter the Syrian regime. If Al Assad ignores the threat and persists in his criminal ways with no consequence, then the strikes would have been for naught, an unjustified use of force.

Proportionality

However, having a just cause is not enough, as those of us who thought the United States had a just cause in Iraq in 2003 painfully learned. In addition, the act of war must not cause excessive damage. This simple notion conceals great complexities. What is excessive damage? At the risk of oversimplifying, war damage can be classified into (at least) two categories:

—Collateral damage comprises the foreseen (but not directly willed) deaths and maiming of innocent persons during the war.

—Supervening damage comprises the foreseen (but not directly willed) bad consequences , including killing and maiming of innocent persons, that occur after the war but can be reasonably traced to it.

To take an example: the collateral damage of the Iraq war comprised the deaths of Iraqi civilians as a result of the hostilities. The supervening damage comprised the harm done by the emergence, after the war, of jihadist insurgency ; by the vacuum of power that, it is thought, allowed ISIS and other terrorists to thrive; and, perhaps, by the Syrian civil war. These bad effects must, of course, be causally connected to the war in Iraq for them to count in the proportionality calculus.

Both collateral and supervening damage must be proportionate to the realization of the just cause. If a military engagement kills more innocents than the expected number of innocents that the engagement seeks to save, then it is collaterally disproportionate. If a military engagement produces bad ulterior consequences that result in the deaths of more persons than were actually saved, then it is superveningly disproportionate. This is not a simple count of lives lost and saved: sometimes the moral urgency of the cause allows for greater damage.

Applying these metrics to the Syrian airstrikes, we can see that there was little collateral damage. The airstrikes destroyed an airport, but were otherwise quite surgical. It is likely, then, that the airstrikes complied with the first prong of the test.

Estimating supervening proportionality is much harder. It requires an accurate prediction by the commander that his decision will not trigger catastrophic events. In the case of the Syria airstrikes, the truth is that we don’t know. If the military action induces Russia to adopt a global hostile attitude and subsequent wars ensue, then it is possible that the airstrikes will fail the test of supervening proportionality. Imagine that as a result of the airstrikes Russia hardens its support for Al Assad and enables him to perpetrate new atrocities. Or imagine that the airstrikes precipitate a proxy war between Syria and Iraq, each counting Russia and the United States as their respective protectors. Or image that Russia, in retaliation, invades Estonia.

So, unfortunately, judging the Syria airstrikes under JWT is not possible until these consequences are known. It is too early to tell. Yet this uncertainty raises a new and perplexing problem. Donald Trump ordered the strikes. Is he culpable of recklessness if he didn’t pause to calculate the consequences? What if he did such estimation as best he could? In that case, if things go well, we will tend to say he was right. But what if things go wrong? Will he be retroactively culpable, even if the estimation of risk was sound? These are difficult issues, because they pose a dilemma. If the probability of (justified) expected damage is less than 1, then the commander can never be sure that his action is right. Bas’ view, as I understand it, is that the ex ante odds of an intervention working out well – insofar as we can know them – are too poor. But this means commanders almost never are permitted to act. No war or revolution (since the same uncertainty arises in violent revolution) is ever justified. More: defensive force is never justified either, since we cannot be sure that defending our homeland will not cause a nuclear holocaust.

I choose to follow the intuition that, despite the unavoidable uncertainty, some wars, and hence some humanitarian interventions, are justified. Commanders must make calls on the spot. Sometimes they will be vindicated, sometimes they won’t. But the pacifist alternative according to which we should never act is, for me, impossible to swallow.

So, with trepidation, I suggest that given the information available and subject to the two conditions I specified (that the strikes must be an effective deterrent and that they must not make thing worse in the medium and long term), Trump did the right thing.