The topic of this essay is the difference between the law of war and just war theory as a branch of moral philosophy, but the place to begin is not with those differences but with the similarities.Footnote ¹ The categories basic to just war theory and those basic to the law of war have overlaps almost too obvious to note, starting with the foundational distinction between jus ad bellum and jus in bello. These are just war categories, but they are also legal categories, and they structure the entire architecture of the law of war—as they do the architecture of contemporary just war theory, even though some contemporary theorists reject the distinction as unsound. Jus ad bellum appears within law mostly in the UN Charter, together with interpretations of the Charter by UN organs.Footnote ² Article 2(4) of the Charter prohibits the threat or use of force against another state's territorial integrity or political independence, “or in any other manner inconsistent with the Purposes of the United Nations”; Article 51 carves out an exception for defense of a state or its allies against armed attack. To oversimplify somewhat, aggressive wars are unjust while defensive wars are just (in the sense of “permissible,” not “mandatory”). The Rome Statute of the International Criminal Court defines the ad bellum crime of aggression more precisely, but it adopts the same basic theme of condemning aggression (its definition comes into force in December 2017). Jus in bello appears in entirely different bodies of law, such as the Hague and Geneva regimes and various weapons treaties (for example, the Chemical Weapons Convention and the Protocol on Blinding Laser Weapons). The Rome Statute places the crime of aggression in a different article from the in bello war crimes. In other words, the Statute's very structure reflects the distinction between jus ad bellum and jus in bello.

In addition to adopting the same foundational distinction between ad bellum and in bello, there are other similarities between just war theory and the laws of war. For example, the in bello principles of distinction and proportionality overlap significantly within the law of war and the morality of war—so much so that the discussion of one often guides discussion of the other. On the ad bellum side, the International Court of Justice (ICJ) has incorporated the requirements of proportionality and necessity—drawn from just war theory—into the law of war, even though they appear nowhere in the UN Charter. Although in its Nicaragua v. the United States decision (1986) the ICJ declared that these principles belong to customary international law,Footnote ³ that declaration was in reality a sign of positivist embarrassment—that is, an unwillingness to admit that a legal decision might be grounded in a moral theory rather than positive law. In actual fact, there was no agreed-upon customary international law of ad bellum proportionality and necessity at the time of the Nicaragua decision, and the Court cited none. Instead, it alluded to “natural right”—a philosophical category—and harrumphed that “it is hard to see how this can be other than of customary nature.”

This example illustrates two interesting facts. The first is what I have already alluded to: the back-and-forth blurring of philosophical and legal just war categories. The second is the urge by lawyers to deny the philosophical impulse and cram just war concepts into legal pigeonholes. And vice versa: philosophers often take the law as their starting point or guide, either explicitly or implicitly, even when they purport to be following reason alone. This should come as no surprise. The early modern sources of international law—Grotius, Pufendorf, and Wolff—drew no distinction in their legal textbooks between the methods of law and philosophy, and they were and are studied by both historically minded philosophers and historically minded lawyers. There are contemporary counterparts to the disciplinary straddle of the classics. Adil Haque's newly published Law and Morality at War, for example, offers a superb effort to merge the modern law of war and just war theory in a seamless treatment.

So much for the similarities. There are also fundamental differences between the law of war and just war theory that go beyond their content, and those are my principal topic. Furthermore, these differences turn out to be important to the way law is used in current conflicts. I will discuss three general characteristics of law with no counterparts in moral philosophy: its need for binary, on-off distinctions; its package character, whereby individual legal rules arise within entire legal regimes; and the detachment of legal propositions from the reasoning that justifies them.

As to the first, consider as an example a currently debated issue: How do we tell the difference between war and nonwar? Or, in the more technical language of the law: How do we tell the difference between armed conflict and lesser forms of violence?Footnote ⁴

The first point I want to make is that law needs to draw lines; and if the lines are not there in the world, the law must make them up. In a tort case we ask: Was the defendant negligent or not? In life, negligence and blameworthiness come in degrees, but the legal question requires a binary: negligence or no negligence?Footnote ⁵ This matters enormously when it comes to armed conflict, because if we determine that violence has risen to the level of armed conflict, international humanitarian law (IHL) kicks in; if not, it does not. As in tort, the law demands a binary, yes-no or on-off answer.

I do not mean that the law draws a bright line between war and nonwar. Just the opposite: the line is notoriously murky. Additional Protocol II to the Geneva Conventions does “not apply to situations of internal disturbances and tensions, such as riots, isolated and sporadic acts of violence and other acts of a similar nature, as not being armed conflicts.”Footnote ⁶ The official interpretation by the International Committee of the Red Cross of where to draw this line between armed conflict and non-armed conflict is that “the armed confrontation must reach a minimum level of intensity and the parties involved in the conflict must show a minimum of organization.”Footnote ⁷ No specification exists of what that “minimum level of intensity” or “minimum of organization” might be. Obviously, this vague standard is less than helpful in answering the yes-no question in any case where genuine doubt might arise.

Nevertheless, vague or not, and unhelpful or not, jurists and the decision-makers they advise must use it to provide a “yes” or “no” answer to the question, “Is this an armed conflict?” The law requires a yes-no or on-off answer because of what turns on the answer: whether the legal regime in place is the law of war or the law of peace. Some lawyers and publicists identify the law of peace with each state's domestic law, constrained by international human rights law. But whether or not they are right about the human rights constraint, there is a difference between the two regimes: In general, the laws of war expose civilians and their property to more risk and more violence than do the laws of peace in any society.

Once we answer “yes” to the question “Is this an armed conflict?” the rules of IHL kick in as lex specialis—a “special law” that displaces or modifies other rules, including international human rights law, in cases where the two yield different answers. Importantly, that lex specialis includes all the IHL rules. This observation highlights the second feature that distinguishes law from morality: laws come in packages, as entire legal regimes. In particular, the modern laws of war grow out of treaties that were negotiated as packages. So, once one determines that the “Is it an armed conflict?” switch is in the “on” position, an entire panoply of rights and obligations follows all together.

This can lead to odd consequences. For example, some lawyers have argued that even after the Israeli disengagement from Gaza in 2005, Gaza remains under Israeli occupation because of Israel's continued control over Gaza's borders.Footnote ⁸ This argument turns on a controversial conception of occupation at a distance, about which I am quite skeptical.Footnote ⁹ Apart from the merits of the argument, one might wonder whether anything beyond terminology turns on it. The answer is yes, because the law of belligerent occupation is an entire legal regime, which comes as a package. Part of this legal regime is the requirement that the occupier take on governance responsibilities over the occupied territory, which would make Israel responsible for the wellbeing of the Gazan population, notwithstanding that the real-life governing authority in Gaza is Hamas. In just war theory, we might well consider the merits of the occupation-at-a-distance conception without reaching the odd conclusion (equally disagreeable to Hamas and the Israelis) that Israel is responsible for governing Gaza. In the law of war—specifically, the law of belligerent occupation—the latter follows from the former.

It is perhaps superfluous to say that neither of the features of the laws of war that I have mentioned—binary, yes-no, or on-off distinctions and normative propositions that come in packages—is essential to philosophical just war theory. For revisionist just war theorists, the question “Is it an armed conflict?” is hardly even relevant, because the basic questions about the moral justifiability of individual acts of lethal violence do not turn on the legal context in which the violence occurs. Even for nonrevisionists, there is no need to suppose that the yes-no question needs to be answered before determining rights and responsibilities. As for principles coming in packages, philosophers make their living by showing that principles that seem interconnected are actually logically independent of each other, so that you can hang onto one principle while discarding the other.

The fact that (unlike in philosophy) legal regimes come as packages can be very annoying if we think the package has inadequate rules in it. Why can't we refine and revise the rules? The answer is that to a certain extent we can and do: courts interpreting legal standards do so all the time, and treaties can be amended, as the Rome Statute was amended to define the crime of aggression. But the practical need for an off-the-shelf package of norms makes revision a perilous affair. Commentators and scholars often lament that Additional Protocol II to the Geneva Conventions—governing non-international armed conflicts—contains shockingly few in bello restrictions on states fighting against rebels; and I have heard some of them suggest that AP II ought to be brought back to the negotiating table for improvement. My own view is that renegotiation is a terrible idea. Trying to get states to reopen a compromised legal regime like AP II is not likely to improve it. What is more likely is replacing bad rules with no rules at all, as states seize the opportunity to wriggle out of their treaty commitments. States cannot be counted on to tinker with parts of a treaty package; they are just as likely to amend the wrong part, or even to throw the entire package in the pond because they no longer agree.

A third feature of the law of war that makes it essentially different from just war theory is that its rules and standards must be detachable from the reasoning that justifies them. Lawyers going forward use the “holding” or “rule” of a precedent, not the ratio decidendi. This can lead to some unpleasant results.

Consider another much-discussed question about U.S. drone attacks: Where is the armed conflict in which targeted killings can be considered lawful acts of war rather than extrajudicial executions? Nobody denies that the United States is engaged in an armed conflict in Afghanistan. But how about Somalia and Yemen? In these countries the proclaimed adversaries of the United States—”associated forces” of al-Qaeda—reside in only a small part of the territory. Does the armed conflict exist only where there is a “hot” battlefield?

Lawyers who face this question answer it by invoking the so-called “Tadić test” formulated by the International Criminal Tribunal for the Former Yugoslavia (ICTY).Footnote ¹⁰ It is worth recalling the context. Duško Tadić was accused of abusing Muslim captives in the notorious Omarška prison camp (among other crimes). Omarška was situated in territory controlled by the Bosnian Serb forces, and therefore outside the zone of hot combat. Tadić argued that there was no armed conflict at Omarška, and therefore IHL did not apply. According to his defense, the ICTY had no jurisdiction because Article 1 of its Statute limits its jurisdiction to “serious violations of international humanitarian law,” and IHL applies only in armed conflicts.

Unsurprisingly, the Tribunal disagreed. It makes no sense to suppose that war crimes are committed only where the bullets are flying, and not in prison camps. Furthermore, some IHL treaty provisions, such as those protecting prisoners of war, plainly apply outside the zone of hot conflict. The judges concluded that the armed conflict exists throughout the entire territory of the state where it is taking place. That finding constitutes the Tadić test for the geographical scope of the battlefield.

Clearly, the ICTY's aim was to ensure that the protections of IHL are spread as widely as possible. But lawyers have detached the test from this purpose and used it to argue that the IHL's permission to kill spreads across an entire country.Footnote ¹¹ It then follows that because the United States is fighting against al-Qaeda in Yemen (formally, “al-Qaeda in the Arabian Peninsula” or AQAP), the U.S. armed conflict against AQAP exists throughout Yemen. It further follows that adversaries can be targeted by drones anywhere in Yemen, and as a consequence the drone strikes are lawful acts of war and not extrajudicial executions without due process. In this way, a legal test designed to maximize civilian protection under the law of war morphs into a legal test that minimizes civilian protections under the law of peace.

I am not claiming that lawyers invoking the Tadić test to defend drone strikes in Yemen are cynically doing something erroneous or illicit. They got the test right. The problem is that once the law arrives at a holding like the Tadić test it floats free of its original context and the rationale that supports it. Propositions in just war theory, on the other hand, can never be detached from the reasoning that supports them: no reasoning, no philosophy. Indeed, my own experience is that nothing infuriates a philosopher in quite the same way as a lawyer who treats the categorical imperative as the “holding” of Kant's moral philosophy or treats the Difference Principle as the “holding” of Rawls's theory of justice. Doing so seems to reduce reasoned arguments to mere memes.

I have identified three features of the law of war that derive from the fact that it is law, not morality:

• • its requirement of binary, on-off answers to complex qualitative questions;

• • its bundled character, which means that legal conclusions such as “it is an armed conflict” carry with them an entire package of rules, some of which make sense in particular contexts, but some of which do not; and

• • its detachment of propositions of law from their original rationale.

Now, legal theorists might object that I am mistaken about the character of law. Legal realists in particular reject the kind of legal formalism I am talking about as “transcendental nonsense,” in the words of Felix Cohen.Footnote ¹² Oliver Wendell Holmes, the prophet of legal realism, insisted that the life of the law is not logic but experience, and he demanded that judges decide cases based not on formalism but on “considerations of social advantage.”Footnote ¹³ Today, a century after the realist revolution, Richard Posner argues for pragmatic decision-making that solves problems, not formalistic argument.Footnote ¹⁴

But even a realist or pragmatist might agree that the law of war should have the three anti-philosophical characteristics I have identified. IHL will often be applied in situations of high stress, and with little time for deliberation, by men and women in the middle or lower ranks of hierarchical bureaucratic organizations. It needs to be usable off the shelf, whereas moral decision-making is by its nature bespoke decision-making.

An officer friend once gave me a plastic wallet card titled “Boots on the Ground,” which is issued to soldiers in the U.S. Army. On one side is the military code of honor; on the other, it has ten Soldier's Rules. These “ten commandments” (I have no doubt the number ten was no coincidence) are bite-size summaries of the legal jus in bello. My friend was a bit dismissive of what he called “wallet card ethics,” but I recall thinking that wallet card IHL might be exactly what a nineteen-year-old recruit fighting in Afghanistan needs.

Dickens's Mr. Bumble complained that the law is an ass. That may be true, and it will seem especially true when we compare IHL's legalisms with the philosophical effort to sort out just and unjust uses of violence, without the necessity of popping them into legal boxes. But an ass might bear a load that would be hard for a man or woman to carry.