I’m grateful to Michael Krepon for this invitation to contribute to Arms Control Wonk. I have long respected Michael’s work here.

In this piece, I’d like to offer a response of sorts to Justin Anderson’s fine recent piece here at ACW, though I don’t mean it to target his piece alone, but rather the implications of some of the arguments that he and others have made concerning the legality of the use of nuclear weapons.

Gro Nystuen and Kjølv Egeland recently published an excellent piece in Arms Control Today providing a concise yet thorough review of jus ad bellum and jus in bello principles of international law relevant to the possession and use of nuclear weapons. These are principles of law related to the international use of force by states, and to the conduct of forces during armed conflict.

I recommend Nystuen and Egeland’s piece to readers, and I agree with their legal assessment that there does indeed remain a “legal gap,” and not just a compliance gap, identifiable as the absence in general customary law of a complete prohibition on the use and possession of nuclear weapons, as distinguishable from general customary rules that completely prohibit the use and possession of chemical and biological weapons. It is to this legal gap that the currently trending “Humanitarian Pledge,” and the related “Humanitarian Impact of Nuclear Weapons” multinational movement are addressed.

In his piece, Justin Anderson takes issue with one statement in a UN General Assembly resolution that is related to this movement. Essentially, his argument is in agreement with Nystuen and Egeland’s conclusion that there are at least conceivable uses of nuclear weapons that would not manifestly violate the principles of jus ad bellum or jus in bello. This indeed was the realization that prevented the International Court of Justice from being able to decide, in its 1996 advisory opinion, that the use or threat of nuclear weapons was in all cases unlawful. Anderson provides a hypothetical case in which, he contends, the use of nuclear weapons would be lawful:

Should the United States or an ally, for example, face an imminent nuclear attack, the U.S. military might advise the president that preventing the attack would require a rapid strike, launched at a distance, using munitions that would completely disable or destroy – rather than merely degrade – the belligerent forces preparing the attack. These requirements might rule out available conventional options; in this scenario, a U.S. nuclear strike would be a legitimate response due to the military necessity of completely neutralizing the target in order to prevent a catastrophic, mass-casualty attack against the United States or an ally.

Neither I nor, I think, Nystuen and Egeland would disagree with Anderson’s assessment of this hypothetical, so long as the principles of necessity and proportionality in jus ad bellum, and the principles of discrimination and proportionality in jus in bello, were in fact observed in the resulting nuclear strike.

However, while not disagreeing with the idea that there exist theoretical possibilities for the lawful use of nuclear weapons, I do think one has to bear in mind the extremely narrow sets of circumstances where such lawful uses could take place. Take Anderson’s hypothetical above. Yes, this could happen. The chances of its happening are quite remote, however. And what of the other often mentioned hypothetical cases for the lawful use of nuclear weapons, in which all of the principles of jus ad bellum and jus in bello would be satisfied, and in which conventional weapons might under the prevailing circumstances be ineffective – e.g., the enemy submarine out at sea, or the hard and deeply buried enemy bunker in the middle of the desert? Again, while similarly theoretically possible, these scenarios are also similarly unlikely to present themselves in the course of reasonably foreseeable armed conflicts.

The concern that arises is in trying to harmonize the theoretical usefulness of nuclear weapons in these very circumscribed and unlikely cases, with the existing stockpiles of nuclear weapons maintained by nuclear weapon states, and argued by them to be justified on the basis of these potential uses. The disparity is of course most acutely discernible in the cases of the United States and Russia, which each possess over 7,000 nuclear weapons, by far most of which are equipped to deliver strikes of a destructive power that could only conceivably be legally justifiable in the extremely unlikely case that Justin Anderson has described.

Is it reasonable, therefore, to argue that the United States needs 7,000 nuclear weapons, by far most of which could only legally be used in one highly unlikely situation – that of an imminent nuclear weapons launch by another state? This is to say nothing of justifying the cost of maintaining and upgrading these weapons, as Joe Cirincione often usefully reminds us.

One can’t shake the suspicion that when military and other national-security types talk about these narrow hypotheticals in order to defeat arguments that the use of nuclear weapons is always unlawful – and thereby also provide at least political justification for the possession of nuclear weapons – they aren’t really thinking that these are the only occasions when nuclear weapons use might be desirable. Rather, these are just the only uses they want to talk about. And that when push comes to shove, if allowed to maintain and upgrade such excessive nuclear arsenals, and create new platforms for their delivery – including cruise missiles, for example – considerably more situations than these might start to look like nails to a man holding a hammer – a shit-ton of hammers, in fact.

If nuclear weapon states were genuine in their representations that they need their nuclear weapons only for cases where international law would be satisfied by their use, and bearing in mind the cost of maintaining nuclear weapons stockpiles of the size maintained by the U.S. and Russia, surely we would be looking at an empirical reality of nuclear weapons possession much more in line with the “low numbers” that James Acton has compellingly written about.

The fact that we are not faced with a low-numbers reality appears, therefore, to belie arguments by nuclear weapon states that they intend to abide by existing international law in their planned, or at least conceived-of, uses of nuclear weapons.

So while it’s true that current international law does not provide for a general prohibition on the possession and use of nuclear weapons, it does contain obligations for states that significantly limit their options. I’ve written about the disarmament obligation in the NPT, and current nuclear weapon state noncompliance with it, at length elsewhere. These disarmament obligations are currently being pressed, if incompetently, by the Marshall Islands against nuclear weapon states at the International Court of Justice.

With regard to the use of nuclear weapons, we fortunately have not witnessed their use in armed conflict since 1945. So no state could be said to currently be in violation of the rules of jus ad bellum or jus in bello because of such use. However, while legal gaps remain in international law that could conceivably allow for the lawful use of nuclear weapons, the limited scope and likelihood of occurrence of circumstances in which such use would be lawful, when compared to the size and cost of efforts to maintain nuclear weapon state nuclear arsenals, makes nuclear weapon state arguments concerning their intention to abide by those obligations in the future ring a bit hollow.