Prospective Duties in Theory and Practice

We should start by recognizing that several unstated assumptions underpin our intuitions in the drowning child example. We are told that there is a “high chance” of us encountering a drowning child in the future, but we are also assuming that if we learn to swim, then we will certainly save the child, while she will certainly die if we do not. In other words, if we take the swimming lessons there is a 100 percent chance that the child will survive, otherwise there is a zero percent chance, and all this is foreseeable. If we tinker with these details, the intuition becomes considerably weaker. Suppose that instead of a high probability of encountering a drowning child, there is a very low probability, as is actually the case for most ordinary people. Suppose further that even if we did learn to swim, the child might die anyway. This happens all too often; attempted aquatic rescues regularly fail, occasionally resulting in the drowning of both parties on account of the so-called death grip that drowning victims instinctively apply to anything they can latch onto. Finally, factor in some probability that an intervening event will save the child's life even if we should fail to come to her rescue, so that our refusal of swimming lessons does not foreclose the possibility of her survival.

Are we still under a prospective duty to learn to swim? I very much doubt it, and here is why. It might be reasonable to demand that a person bears the costs associated with swimming lessons (in terms of time and money) if we can foresee that a child will certainly die if he or she declines, and if the lessons will guarantee the survival of the child, and so on. But bearing these same costs is too much to ask if we do away with these assumptions—if the chances are that the person will never encounter a drowning child in the first place, and it is acknowledged that he or she might fail to rescue her (or even make matters worse for her) if that person did learn to swim. Morality demands too much if it says that you have a prospective duty to bear the costs of swimming lessons just to possibly increase your chances of saving a drowning child by some indeterminate amount in the unlikely event that you encounter one in the future. This is precisely why we would not normally think to condemn someone for not learning how to swim; that decision does not normally constitute a moral failure.

With this in the background, let us turn now to the question of whether states are under a prospective duty to militarize.

The first thing to note is that unlike swimming lessons, militaries are obscenely expensive by any measure. The money spent on them in 2019 (nearly two trillion dollars) could have ended world hunger six times over, according to a recent estimate of what that would require.Footnote ³ And, of course, the “opportunity costs” of military spending are highest in the poorest countries. In the mid-1990s, the United Nations Development Programme calculated that if the developing world were to redirect just one-quarter of its military expenditures, by the year 2000 it could achieve primary healthcare for all, the immunization of all children, the elimination of severe malnutrition, the provision of safe drinking water for all, universal primary education, the reduction of illiteracy, and funding for family planning.Footnote ⁴

The fact that a state's military strength is a function of the comparative strength of its adversaries goes a long way toward explaining why it is that armed forces tend to become, as one Honduran anti-militarist memorably put it, an “all-devouring octopus of the national budget.”Footnote ⁵ William Graham Sumner summarizes how it tends to play out:

The problem, essentially, is that military power is a positional good, and with respect to such goods, one moves backward by standing still. The consequence is that militaries are not just obscenely expensive; they are inexorably expensive.

The costs of militarization are not just financial. In Ethics, Security, and the War-Machine, I draw attention to some of the underappreciated moral, social, and political costs that all societies bear, to a greater or lesser extent, simply by virtue of being permanently prepared for war.Footnote ⁷ My discussion there is largely abstracted from the features of particular societies, but as Cheyney Ryan points out in his contribution to this symposium, the kind of polity we are talking about, and its mode of war preparation, makes a significant difference in this regard.Footnote ⁸ Ryan emphasizes the distinction between “nationalized” and “postnationalized” war preparation, the defining feature of the latter being that its burdens are largely “externalized”; that is, not directly borne by the citizens of the polity. We see this, for instance, when all of the actual fighting is assigned to a small group of paid professionals while most ordinary citizens do not participate at all. This gives rise to the condition that Ryan elsewhere diagnoses as “the chickenhawk syndrome.”Footnote ⁹

The costs of militarization will also vary, I think, depending on whether we are talking about an “introverted” or an “extroverted” defense establishment. Kirk Bowman asks why it is that militaries have been such a cancer on Latin American countries in particular over the last century.Footnote ¹⁰ As Erich Weede notes, this part of the world “seems to have suffered from parasitic praetorians more specialized in bossing, exploiting, killing, and torturing civilians than in fighting other nations’ armed forces.”Footnote ¹¹ Part of the answer, according to Bowman, is that most countries in this region face no credible foreign threats, and armed forces with no external enemy to absorb their attention will tend to become introverted (read: meddlesome in domestic affairs). This suggests that military organizations are likely to impose distinctive costs on societies with no foreign enemies, compared to those that do face credible external threats.

A second distinction worth making here is that between states that are “prepared for war” in a narrow sense and states that are “prepared for war” in a broader sense. A state that concentrates on maintaining the means to rebuff foreign conquerors and plunderers is prepared in the narrow sense. A state that is ready to repel aggressors, but also ready to engage in expeditionary wars to further its economic interests abroad, ready to engage in humanitarian operations against oppressive regimes, and so on, is prepared for war in a much broader sense. We should naturally expect there to be more significant costs associated with the latter kind of war preparation. In their recent book Tyranny Comes Home, Christopher Coyne and Abigail Hall discuss what they call the “boomerang effect.”Footnote ¹² In the course of preparing itself for certain kinds of armed conflict abroad—forcible regime change, for example—a state must develop means of social control for use against the target population. The trouble is that after the conflict ends, these methods of social control that have been fine-tuned abroad are now innovations that might be—and often are—used in domestic governance. The boomerang might take some time to return, but Coyne and Hall suggest that the effect is difficult to avoid entirely:

Following on from this, it is worth stressing that militarization inflicts unique costs on democratic societies. In her contribution to this symposium, Neta Crawford highlights the various ways in which war building erodes democratic norms, processes, and institutions.Footnote ¹⁴ We see this, for example, when the courts and legislature completely defer to the defense establishment on all matters related to national security, the parameters of which are defined by the defense establishment itself. The result is a reduction in transparency and accountability. But, according to Crawford, the tension between war and democracy runs much deeper than this: “Because war is the assertion of might makes right, the negation of the rule of law, and because the values of war and militarism are antithetical to the normative values of democracy, it is hard to hold these conflicting norms at the same time.”Footnote ¹⁵

There is another dimension to this antithetical relationship worth highlighting. It becomes apparent once we appreciate that the core identity-grounding commitment of liberal democracy is that all citizens should enjoy equal protection under the law. That is to say, no political community can properly call itself a liberal democracy unless it is committed to the realization of the equal rights of all citizens.Footnote ¹⁶ And yet, we are constantly told that military institutions cannot give their members the same rights and liberties that the rest of us take for granted, as this would make them ineffective in battle. If this is right, then maintaining an effective army requires a kind of moral stratification; the consignment of certain citizens to a separate class that must remain effectively walled off from the rights and freedoms extended to everyone else.Footnote ¹⁷ Insofar as such stratification is incompatible with the defining normative commitments of liberal democracy, this is yet another kind of democratic erosion that we can add to those identified by Crawford.Footnote ¹⁸

Whatever the costs of war preparation, let us accept for the sake of argument that a state can be obligated to bear them under the conditions assumed by Fabre: where the state can foresee with a high degree of certainty that it will need to use military force to rescue its citizens from some grave threat in the future; where the rescue attempt will surely succeed; and so on. What about under different conditions, though? Suppose that a state is unlikely to ever encounter any such scenario, that it might fail to rebuff the threat even if it did respond with military force, and that the use of military force under such a circumstance might actually backfire and make matters worse for its intended beneficiaries. It would be very implausible, I think, to say that none of this makes any moral difference, and that the state is still under a prospective duty to bear the very high cost of maintaining a military.

Putting all this together, the plausibility of the claim that states are under a prospective duty to militarize seems to depend on three premises: (1) states are highly likely to encounter grave threats to their citizens that only military force can neutralize; (2) the use of military force to overcome such threats is highly likely to succeed; and (3) the use of military force to overcome such threats is not at all likely to backfire and make things worse for its intended beneficiaries.

The second premise is obviously false. Research by Erica Chenoweth and Maria Stephan reveals that between 1900 and 2006, only around one-quarter of all major armed campaigns achieved their goals.Footnote ¹⁹ And it would be a mistake to suppose that the failures are confined to nonprofessional militias or small, poorly resourced national militaries. A study by Taylor Seybolt gives us a useful snapshot in this connection. After a careful analysis of seventeen military interventions across six countries between 1991 and 1999, most of which involved major Western powers sending troops into developing countries for humanitarian reasons, Seybolt concludes that “of the 17 interventions, nine succeeded in saving lives; four failed to save lives and two of these made life worse for at least a short time; and four had a mixed record, meaning that they saved lives but in the context of failing to save many more.”Footnote ²⁰

So, the use of military force is not highly likely to succeed. At best, it delivers mixed results: we are not guaranteed to win the wars that we fight. What about the other two premises stated above? These initially seem more plausible, but on closer inspection they, too, begin to buckle.

The Decline of Aggression

In earlier periods of human history, states regularly attacked one another and therefore regularly needed to call upon their armed forces to defend their citizens, but things have changed. Cognitive psychologist Steven Pinker notes that “after a 600-year stretch in which Western European countries started two new wars a year, they have not started one since 1945. Nor have the 40 or so richest nations anywhere in the world engaged each other in armed conflict.”Footnote ²¹ In the period immediately following the Peace of Westphalia, 1648–1714, the risk of becoming involved in war for any state was estimated to be one chance in fifty-nine. From the end of World War II to the end of the Cold War, it was one chance in 167, and it dwindled further to one chance in 250 in the decade immediately thereafter.Footnote ²² Let us not exaggerate. International aggression does still happen from time to time. But it has become extremely rare and the consensus among mainstream conflict researchers today is that it is likely to become rarer still.Footnote ²³

A number of factors are thought to have contributed to this decline. Economic interdependence is one. Global markets have altered the material cost-benefit equation of aggression, such that free trade, rather than violent conquest is almost always the economically optimal strategy now. The infrastructure of global commerce basically makes it cheaper to buy things than to steal them. Further, as people have become wealthier and better educated, self-realization has become more important. And as self-realization has become more important, self-sacrifice has become less appealing. In this way, improved material conditions have made people generally less willing to participate in, and less accepting of the personal costs associated with, war making.Footnote ²⁴ Multilateral institutions have also played a role in the decline of international conflict. Global bodies such as the UN, and regional ones such as the European Union and African Union, facilitate cooperation among member states and punish defectors politically and economically.Footnote ²⁵

Besides these material and institutional forces, ideational and normative developments have also contributed to the decline of international conflict. First, there is the triumph of the state system. The idea that the world ought to be organized into independent, self-determining sovereign countries is now widely accepted and deeply entrenched. Ideologies of global empire have been consigned to history and to the extreme fringes of present belief. Second, there is the moral subordination of the state. No longer does anybody think that states exist for their own glory or to serve God; we mostly think of the state as an instrument whose value derives from the benefits it confers upon its citizens. This conception of the state is less permissive than the one it supplanted, which allowed national leaders to self-aggrandize through war and territorial expansion.Footnote ²⁶

So much for premise (1). There may some be exceptions, of course, but for most states today, it is not highly likely that they will need to defend their citizens against foreign aggression at some point in the future. It is highly unlikely, and this is not purely or even predominantly a function of military deterrence or the balance of hard power.

Backfire

When military force is employed to protect civilians, it sometimes backfires. For example, NATO's humanitarian intervention in Kosovo was intended to put a stop to the ethnic cleansing of the province's Albanian population. On March 23, 1999, the day prior to the commencement of the NATO intervention, the number of people that had been forced out of their homes was estimated at 230,000. By the end of the war, 1.4 million were displaced, and of these 860,000 had fled Kosovo entirely. Prior to the intervention, 2,500 people had been killed in the civil war between the Serbs and the Kosovo Liberation Army. During the intervention, approximately 10,000 were killed, most of them Albanian civilians killed by the Serbs, but there was also some collateral damage caused directly by NATO.Footnote ²⁷ These statistics led Noam Chomsky to suggest that ethnic cleansing was in fact the consequence of NATO's intervention, not its cause.Footnote ²⁸

We can say something even stronger regarding the use of military force for national defense. To suggest that it can make matters worse for its intended beneficiaries (the citizens of the parent society) is something of an understatement. There is an important respect in which it almost invariably does make things worse for them.

To see this, recall David Rodin's distinction between “genocidal” (or “murderous”) aggression and “political” (or “lesser”) aggression. Murderous aggression is where one state invades another with the intention of exterminating, forcibly expelling, or enslaving its people. Political aggression is where one state invades another to gain control over its political institutions and/or its territory or resources, rather than to kill or expel the people that live under those institutions or that inhabit that territory. In the latter case, only if the target state puts up a fight will there be interpersonal violence; the threat of killing and maiming is conditional.Footnote ²⁹

Most international aggression is of the political rather than the murderous kind. “No man is so ferocious that he tries to win in order to kill,” writes Rousseau. “One kills in order to win.”Footnote ³⁰ So it is, usually, with aggressors; their aim is to conquer, rule, and extract resources, and recourse to physical violence is taken only because armed resistance makes it necessary. In the absence of resistance, political aggression is relatively “bloodless”; the people of the target country have their political interests undermined (they lose their sovereignty and self-determination, for example) but they remain alive and physically unharmed.Footnote ³¹

A state faced with political aggression thus has the following choice to make: It can surrender without putting up a fight, in which case the state loses its sovereignty and the community loses its self-determination, but the citizens are unlikely to be killed and injured on the same scale that they would be if a war of national defense were waged. Alternatively, the attacked state can wage a war of national defense to try to retain its sovereignty/self-determination, but by so doing it predicably brings about a state of affairs in which more of its people are killed and injured than would otherwise be the case. Whether the state's responsibility to protect its citizens demands that it fight or surrender in such cases is not a question that I intend to, or need to, settle here. I am simply pointing out that in most cases of national self-defense, where a state deploys its military to fend off a political aggressor, that state will make its citizens worse off in one very important respect: By turning bloody what would otherwise be a bloodless invasion, the state diminishes the average citizen's prospect of survival.

Let us now take stock. At the beginning of this essay, I granted that a prospective duty to create and maintain a military might arise under certain circumstances. What the observations above suggest, however, is that these are not the circumstances that most states now find themselves in. If a state is unlikely to encounter the kind of threat that necessitates and justifies a military response, and a military response might fail to neutralize the threat anyway or even make matters worse for its intended beneficiaries, then said state is not under a prospective duty to bear the costs associated with militarization. This description applies to most states nowadays, and so most states are not under a prospective duty to be militarized.Footnote ³²

A Closer Analogy?

Up to this point, I have gone along with Fabre's analogy between learning to swim and building a military. But I would suggest there is probably a more apt analogy to guide our thinking in this space.

In 2019, mass shootings in Dayton, Ohio, and El Paso, Texas, led to renewed calls for tighter gun control in the United States. High-capacity assault weapons were at the center of the conversation, with many demanding that they be prohibited. In response, Twitter user Willie McNabb asked what alternative he had to an assault rifle if he ever needed to shoot “30–50 feral hogs that run into my yard within 3–5 mins while my small kids play.”Footnote ³³ McNabb was attempting to justify the private ownership of semiautomatic assault rifles by describing a scenario in which only some such weapon would suffice to neutralize a threat to his children. Essentially, he was trying to pass off assault-weapon ownership as a prospective duty. The Tweet quickly went viral and became the object of global ridicule.

Bernard Brodie once lamented that “worst case fantasies” peddled by a “cult of the ominous” dominated thinking about international security throughout the Cold War.Footnote ³⁴ McNabb's conclusion is what we get when we think about home security in the same way. Hardly anybody took McNabb's “argument” seriously, and I suspect most readers of this journal will agree that individuals are not under a prospective duty to keep assault weapons in their homes just in case feral hogs attack. In fact, most readers probably think that individuals have a defeasible duty not to bring these weapons into their homes. But why should we think that? If individuals have a right, and indeed a duty, to defend themselves and their loved ones against wild animals, why do they not also have a duty to acquire the means necessary to do so?

Any plausible explanation will appeal to some combination of the following reasons.

First, feral hog attacks on humans do happen occasionally, but they are exceedingly rare. In the most recent incident that we know of, a Texas woman was trampled to death walking between her car and the front door of her home in 2019. “Exsanguination due to feral hog assault” is the cause of death formally recorded in the medical examiner's report, but this was only the fifth documented case of it in the United States in nearly two hundred years.Footnote ³⁵

Second, using an assault rifle against a stampede of feral hogs might fail to neutralize the threat it poses, and could actually increase the danger. The sound of gunfire might spook the animals and make them even more aggressive, for example, and there is also a risk that the people one is trying to protect will be accidentally shot.

Third, firing multiple rounds in quick succession makes it very likely that some bullets will go astray, which means imposing risks of unintended harm on bystanders and neighbors. This is especially likely given that in desperate situations people with firearms have shown a tendency to keep on squeezing the trigger until the magazine is empty, whatever its size.Footnote ³⁶

And fourth, merely keeping a gun in the home imposes considerable risks on its occupants. There have been countless cases of minors gaining access to guns kept on the premises ostensibly for their own protection, and then killing themselves or another member of the household, accidentally or on purpose. These so-called child access shootings happen all the time, so it is fallacious to think that having a gun in the home enhances the safety of its inhabitants in every respect. The reality is that it protects them against some dangers, but simultaneously exposes them to others.

If these considerations, taken together, explain why individuals are not under a prospective duty to arm themselves with assault weapons, then I submit that similar considerations at the international level give us reason to reject the idea that states are under a prospective duty to militarize. After all, as we have seen, the kinds of threats that demand a military response are increasingly rare in this day and age. Further, using military force against these threats is not guaranteed to overcome them, and can even make matters worse—indeed it almost invariably does make matters worse for the intended beneficiaries in one important respect. To this we can add that the use of military force imposes risks of unintended harm on innocent bystanders. And just as a gun in the home endangers its occupants, so, too, soldiers in the barracks endanger their parent societies. The people of Myanmar are being reminded of this presently.