SSI (test)

Strategic Insights: Proxy War Norms

Dr. C. Anthony Pfaff

Current trends in international relations suggest the United States will place a greater reliance on international partners in securing vital national interests. Growing assertiveness by regional state actors, increasingly capable nonstate actors, and a “war-weary” American public suggest the emergence of a “polyarchic” world order that will strain the United States’ ability to maintain sufficient forces overseas, where it currently exchanges defense commitments for access and basing.1 Rather, the United States may have to commit to a strategy broadly described as “off-shore balancing” that would rely on regional partners to uphold the balance of power in their own neighborhood, exchanging indirect U.S. support for the partner’s willingness to act in the interests of the United States.2 Even if it does not commit to such a strategy, current events suggest working through others to achieve strategic ends will be a feature in any future approach to international relations.

Such a strategy will not only encourage proxy relationships, but as these state and nonstate challenges arise, they encourage proxy wars as well. In fact, there are a number of proxy wars underway in places like Yemen, where the United States supports Saudi Arabia’s efforts to contain Iranian influence; Syria, where the United States, Iran, and Russia support different factions to achieve a variety of foreign policy goals; Iraq, where the Government of Iraq relies on militias to confront the Islamic State on its behalf; and Ukraine, where Russia backs a separatist movement ostensibly to protect Russian citizens, but more likely to keep Kiev off balance and prevent Ukraine’s drift toward the West.

Surprisingly, there is little written on the norms of proxy wars. While there is some international law that governs state sponsorship of foreign nonstate actors,3 the default position is if the proxy war is just, then so is the proxy relationship. However, entering into such relationships creates massive opportunities for moral failure. These opportunities arise because the introduction of the benefactor complicates already complex and somewhat subjective decisions made with regard to resorting to war and introduces a corrupting influence that risks distorting the reasons that drive those decisions. Because benefactors bring these moral complications, they bear the greater burden to address them. This point does not entail proxies have no responsibilities. What it does entail is that most moral decisions regarding proxy wars are often in the hands of the benefactor, without whom there would be no proxy relationship to judge.

Since proxy wars are wars, moral analysis should start with the traditional provisions of jus ad bellum, but also reflect the potentially corrupting influence a benefactor brings. In what follows, I will first discuss how the character of proxy war impacts its ethics, and then I describe that impact through the application of jus ad bellum conditions, as well as examine the kinds of moral hazards proxy wars give rise to even when those jus ad bellum conditions are met.


Andrew Mumford defines proxy war as “indirect engagement in a conflict by third parties wishing to influence its strategic outcome.”4 It is the indirect nature of the benefactor’s involvement that distinguishes a proxy relationship from other supportive relationships, such as, for example, an alliance.5 This point does not suggest that direct action by the benefactor is incompatible with proxy relationships. In Libya, for example, the international coalition provided support to rebel forces while at the same time directly attacked Gaddafi’s forces from the air. So, while air strikes did contribute to Gaddafi’s defeat, the coalition limited its risk—as well as its costs—by supporting proxies on the ground that acted as a surrogate for ground forces it would have otherwise had to commit. It is this use of surrogates to replace, rather than augment, benefactor assets or capabilities that characterizes the proxy relationship.


Proxy wars, like any other kind of war, must meet the requirements of jus ad bellum. These conditions include just cause, proportionality, legitimate authority, public declaration, just intent, last resort, and reasonable chance of success.6 For some criteria, the proxy must fulfill them, for the most part, by itself. The intervention of a benefactor will not make an unjust cause just, an illegitimate authority legitimate, or a wrong intention right. On the other hand, the benefactor’s involvement can potentially make the disproportionate proportionate, which can make alternatives to fighting less appealing and thus impact what counts as a last resort. Moreover, it can certainly affect a proxy’s calculations regarding its chances for success. This dynamic presents three morally relevant situations:

  • Proxy meets all elements of jus ad bellum and engages in conflict, but benefactor support is not causally related to hostilities.
  • Proxy meets all elements of jus ad bellum but does not initiate conflict without benefactor support.
  • Proxy cannot meet all elements of jus ad bellum without benefactor support (e.g., proportionality and reasonable chance of success), benefactor support is causally related to hostilities.

The concern here is the role the benefactor plays in making the war come about. To the extent the war would have occurred without the intervention of a benefactor, as in the first situation, then there are fewer moral barriers to benefactor participation. If the proxy meets the just war criteria and decides on its own to fight, then the benefactor is simply joining an already just enterprise. If it is just to come to another’s defense, it is just to come to that defense by proxy. The moral opportunity for the benefactor in such cases is to ensure that the just side wins with the least harm (to either side) caused. This point suggests that benefactor involvement that prolongs, widens, or increases a war’s destructiveness is impermissible, regardless of the justice of the proxy’s cause or the interests at stake for the benefactor.

More concerning are the next two situations, one where the proxy does meet the criteria but does not fight absent the support of the benefactor and one where the proxy cannot meet all the just war criteria without benefactor support. In both of these cases, the participation of the benefactor seems causally related to the proxy’s decision to go to war in ways it is not in the first situation.

In the former, the proxy meets all the criteria, including ones related to resources like proportionality and reasonable chance of success; however, chooses not to fight without benefactor support. We can imagine a number of reasons why this might be the case: the proxy wants to lower its cost, it is culturally risk averse, or it values the legitimacy an outside party might bring. In this case, it is not the reason it withholds a decision to fight absent external support; rather, the concern is that whatever the situation, the proxy did not feel that resorting to war was worth it, absent whatever benefit the benefactor would bring. In such a situation, the benefactor plays a more causal, though still indirect, role in bringing the war about. Given that it is usually better to avoid harm rather than to cause it, even in the service of justice, benefactors have to establish clearly that their involvement serves a greater good that overrides the proxy’s reluctance to go it alone. War, while it might not be the worst thing, should always be a necessary thing. Thus, in this situation, there is an extra burden on the benefactor to find alternatives to war. It may still end up being the case that the injustice suffered by the proxy warrants the intervention, but the fact the proxy would not fight, despite being permitted and able, suggests a benefactor should proceed with caution.

The latter is a different case. The proxy has met the conditions it can, but not all the conditions it needs. In this situation, the proxy suffers from an injustice or aggression that would justify war, but simply cannot morally do anything about it because it cannot fight proportionally or discriminately or it does not have a reasonable chance of success on its own. In this situation, the proxy’s choices are to suffer the injustice or to violate one (or more) of these other jus ad bellum or jus in bello conditions. In this case, the benefactor’s intervention makes it not only possible for the proxy to address the injustice, but to do so justly. Doing so, of course, would be a condition of the benefactor’s participation, regardless of the benefactor’s larger interest.


The application of jus ad bellum raises some additional concerns for the benefactor. First, to what extent do the benefactor’s cause, interest, and intent have to align with the proxy’s? If the proxy’s war is just, does it matter why the benefactor provides support?

The short answer is “probably,” but it also matters what kind of situation the proxy is in. In the first situation, where the war happens regardless of benefactor support, as long as the benefactor’s motivating reason is at least morally permissible, its support is likely as well. Moral hazards can arise when interests diverge, which is a point I will take up separately.

In the other two situations, where the benefactor’s support is a necessary condition for fighting, then it seems the benefactor’s motivating reason to intervene must relate to a cause of justice its intervention is supposed to address. For example, according to the National Security Strategy, the United States considers a “strong, innovative, and growing U.S. economy in an open international economic system that promotes opportunity and prosperity” as a vital national interest.7 It seems wrong, however, to enable a war to meet that interest. There is nothing wrong with pursuing this interest, there is also nothing wrong with it being realized collaterally as a side effect of a war. It is wrong to pursue it by force, even indirectly. Therefore, it cannot stand as a reason to enable a war.

On the other hand, this point does not entail that the benefactor must limit its concerns to confronting a particular act of aggression. James Pattison, for example, argues that causes of justice, such as preventing or mitigating gross human rights violations or humanitarian disasters or even ensuring a better peace, can justify intervention even if the proxy’s cause is unjust.8 Nevertheless, one does not always have to set the bar so high.

To see where the bar should be set, it is useful to distinguish between sufficient and contributing causes when determining what type of reasons justify the intervention of a benefactor. Sufficient causes are those that fulfill the just cause condition associated with responding to an act of aggression. Contributing causes, on the other hand, do not fulfill the just cause condition, but are themselves causes of justice, which, given a sufficient cause (such as the proxy’s), should count as justification for a proxy relationship.9 For example, reasons such as deterring future aggression, upholding a just international order, as well as preventing human rights violations could be valid reasons, even if by themselves they would not justify war.


One of the attractions of proxy relationships for the benefactor can be the ability to pursue security objectives while keeping the proxy relationship secret, as the United States attempted to do when it supported the Contras in Nicaragua.10 Given the requirement for public declaration, which not only allows the enemy an opportunity to address an injustice but also one’s own population to determine whether a military response is worth it,11 it is worth asking if such secrecy is ever permissible. On the surface, it would seem the answer is no. Given the cost of war as well as the potential for escalation, it makes moral sense that the benefactor should also publicly declare its relationship.

The concern here, however, is that sometimes a public declaration can make escalation harder to manage or undermine the proxy’s cause. For example, though not a proxy relationship, had the United States declared its support for Iran’s Green Movement, it would have allowed the Iranian Government to portray the movement as “foreign” and undermine its appeal.12 In cases like this, particularly when causes of justice are at stake, it may make moral sense to keep the relationship secret. The problem with the U.S. relationship with the Contras was not so much that public declaration would have undermined their cause; it just would have subjected it to unwanted oversight. That was the moral error. Therefore, even if one justifies keeping a proxy relationship secret, it still should be subject to some oversight by representatives who can act on behalf of the population.


In meeting proxy requirements, such as those necessary to meet the conditions of proportionality, reasonable chance of success, and last resort, one must make judgments about future costs and alternatives that are difficult, if not impossible, to anticipate. The introduction of the benefactor, however, can affect these judgments by appearing to reduce costs and provide additional capabilities that make the fight seem just and prudent. While doing so may appear to be to the advantage of the proxy, the reality is often more complex.

The difficulty is, of course, benefactors can convey these advantages while at the same time unintentionally offsetting them, complicating their calculation. Given the propensity for escalation, frozen conflict, and diffusion inherent in proxy wars, it is hard to know how costly a conflict will be, clearly demonstrated by events in Syria, Yemen, and elsewhere. Moreover, even without a benefactor, one can never know if one has tried all possible alternatives before resorting to war. Given the typical urgency of decisions to resort to war, the introduction of the benefactor is likely to make non-violent, but perhaps costly, alternatives seem unattractive, making last resort difficult to determine.

One thing to note regarding reasonable chance of success is that benefactors should intervene with the proxy’s success in mind. While there may be good operational and political reasons to limit assistance, as was done in Syria, it makes no moral sense to provide proxies with insufficient assistance to meet their political objectives. Doing so just prolongs the fighting and suffering. Thus, for any assistance, there must be an established causal connection to how it contributes to proxy success. These complications provide a nice segue into the next concern regarding moral hazards.


Moral hazards. Meeting just war conditions, of course, does not exhaust the moral concerns associated with the use of proxies. The proxy relationship also introduces moral hazards that must be managed if one is to meet the range of obligations associated with proxy warfare. Moral hazards, a term that originated in economic theory, arise when some persons assume greater risk because they know other persons will bear the burden of that risk.13 For example, the provision of medical insurance can increase the cost of medical care, because, among other things, the lower cost of care for the individual encourages greater use while at the same time discourages shopping around for the best prices for the best care. The resulting incentive structure thus places pricing in control of institutions and not to normal market forces making costs very difficult to control.14 While medical insurance is arguably a good thing, failure to manage its associated hazards can effectively render it, to some degree, self-defeating.

It is easy to see how such a dynamic can emerge in a proxy relationship. Benefactors and proxies enter into these relationships precisely because they help both parties mitigate costs and risks associated with achieving their relevant interests. Because the risks are lower, the incentives to avoid war are lowered or even removed. It is thus not without some irony that President Dwight D. Eisenhower called proxy wars “the cheapest insurance in the world.”15 Much like the case with medical insurance, the introduction of the proxy relationship can encourage the kind of risk taking that not only makes war more likely, but more costly as well.

The presence of moral hazards does not directly affect the permissibility of a particular proxy relationship. However, failure to manage these hazards can effectively transform an otherwise permissible intervention into one that is impermissible. These hazards arise because of variations in benefactor and proxy interests, willpower, and capabilities. These variations also lead to divergent interests and optimistic estimates about the true cost of war that can drag both parties into a conflict they might otherwise have avoided.

Divergent interests. Related to aligning interests is the concern regarding diverging interests. Proxy relationships do not just make judgments to go to war more likely, they risk corrupting those judgments as well. As noted previously, it can be permissible for benefactor and proxy interests to diverge. But even when permissible, the divergent interests inherent in any proxy relationship risk not only going to war to serve unjust ends, they also risk expanding the fighting as the parties involved attempt to realize multiple, and sometimes exclusive, goals. To make matters more complicated, as the chaos in Syria clearly demonstrates, benefactors and proxies never really know—or at least sometimes misinterpret—the other’s interests and intentions.16 To the extent that divergence is not clear, both parties risk making bad judgments and commitments that could prolong a conflict and undermine the just cause for which it is fought. The remedy is to either align interests, so that the realization of a proxy’s interest should realize the benefactor’s as well or lead to the end of the proxy relationship. This means that once the proxy has achieved its objective, the relationship—at least as it relates to the conflict in question—should end.

Underestimating costs and risks of violence. As mentioned earlier, proxy relationships allow both benefactors and proxies to mitigate their costs while still, at least in theory, addressing their security needs.17 With this reduction in cost, as the discussion on moral hazards suggested, there often comes an increased readiness on both sides of the relationship to accept risk. For benefactors, the proxy’s direct involvement allows them to address more distant security threats, especially when the urgency for the benefactor to engage directly is lagging.18 The best way to mitigate the effects of this moral hazard is for both sides to consider not just the total costs of the conflict, but also the cost of the conflict if they had to wage it themselves.

This point does not entail that they must be able to do so, but it does entail what both the benefactor and proxy should consider if the consequences of a partner abandoning them in the future are worth risking their situation in the present. Even then, it could still be permissible to proceed, but that decision rests on how trustworthy they perceive the other to be. Ethical decision-making does not have to be risk free; however, there are at least some standards of prudence that officials charged with making such decisions should observe.

Escalation. Related to the concern of underestimating the actual costs of a conflict is the potential for escalation the proxy relationship entails. This concern is exacerbated when both sides directly involved in a conflict have benefactors of their own. In fact, one study concluded, “in the 114 civil wars between 1946 and 2002 where at least 900 people were killed, no rebel group was transferred major conventional weapons without the government also receiving arms from another source.”19 This suggests that escalation in proxy conflicts is more often the rule rather than the exception.

The fact that a proxy conflict escalates does not necessarily undermine the justice of one’s proxy relationship. Given a just cause, the fault of the escalation arguably lies with the side that responded. As Pattison notes, “those facing violations of their basic human rights still retain their right of self-defense, irrespective of how others will react to the exercise of that right.”20 The remedy here is to ensure escalation dominance prior to any intervention. For example, it is likely any lethal support for the Ukraine Government would result in escalation on the part of the Russians. If the United States is not prepared to match that escalation, then it should not start it in the first place.21

Diffusion problem. Related to escalation is the diffusion problem. The concern here is that the capabilities a benefactor may provide a proxy may not stay with the proxy. This diffusion can thus create new conditions for instability. In fact, the diffusion of weapons in a post-conflict environment has demonstrably led to “higher homicide rates, more violent crime, and further conflict.”22

Probably the best example of the hazards associated with this concern is the spread of Stinger anti-aircraft missiles following the Soviet withdrawal in Afghanistan in 1989. After the conflict ended, these missiles were found as far afield as Bosnia, Iran, Kashmir, Tunisia, and the Palestinian territories. To control this diffusion, the U.S. Government initiated a $65 million buy-back program; however, only a small fraction has been recovered and a reported 300 to 600 missiles remain unaccounted.23

There is, of course, no one way to manage this particular moral hazard. However, it is typically better to be proactive than reactive. This point suggests that benefactors should pay attention in advance to how they are going to control the distribution of assistance, especially lethal assistance, as well as prevent its diffusion when the conflict is resolved.

Dirty hands. The indirect nature of the proxy relationship means not only that benefactors can get someone to do their dirty work, it also means they can get someone to do their dirty work without getting their hands dirty. Unfortunately, the bar that international law sets regarding state responsibility for a proxy’s actions is not only high, it is limited in its application as it applies only to direct guidance or instruction by the benefactor and only to those proxies who are nonstate actors.24 As a result, it does not account for all of our moral intuitions on this point. As Michael Walzer notes, there are some things we want to judge wrong even if sometimes we judge them as also necessary.25 Reconciling these competing intuitions requires some kind of accountability if one is to maintain the moral legitimacy of the proxy relationship.

Ideally, one would withhold support from a proxy who committed war crimes or other immoral acts. However, sometimes, as in the case of Iraq’s struggle against the Islamic State, withholding support can lead to greater injustice. For instance, if the Iraqi Security Forces fail, the Islamic State will continue to hold territory, enslave minority populations, and conduct terrorist attacks worldwide. The fact such conditions hold, however, does not entitle a benefactor to turn a blind eye.

Walzer, in his well-known article on the subject, argues that sometimes we can judge an act as both immoral and necessary.26 In such instances, we want the politician to get his hands dirty on our behalf as, in Walzer’s words, “that’s what we pay him for.” The way out of the apparent dilemma is to hold politicians accountable for whatever crime he or she commits. Using Walzer’s example, we want the politician to order a terrorist’s torture to prevent an imminent attack, especially if it means saving large numbers of innocent lives. What we do not want is to condone torture. To keep both intuitions intact, politicians should be held accountable and charged with the crime.27 Those judging the case are free to mitigate any punishment; however, the potential for such accountability is the sacrifice the politician makes.

It is a separate question whether such accountability is realistic. Certainly mature democracies can hold their leaders accountable to some degree. While not exactly related to dirty hands, a number of U.S. officials were convicted of various crimes related to the Contra scandal.28

Moreover, it is worth pointing out that few wars of any kind are pure or clean. Members of our own forces have, from time to time, committed war crimes. This point again does not entail permission to ignore violations. However, if violations are insufficient to warrant abandoning an otherwise just cause, then they are likely insufficient to warrant withholding support. Again, this point does not suggest turning a blind eye. In such cases, the right thing to do is not to abandon the cause, but rather to do one’s best to hold violators responsible. This point only holds as long as there is a reasonably good accountability scheme in place for those who do commit war crimes such that the amount of injustice caused by fighting is less than the amount without fighting. Ensuring such a system is in place, especially with proxies whose culture and capabilities may make accountability difficult, likely means benefactors will have to bear some of the moral burden of preventing the commitment of such crimes and, failing that, holding perpetrators accountable.

Therefore, in recognizing that perfect compliance with jus in bello norms is an elusive and unlikely goal, this analysis would permit a benefactor-proxy relationship when the following conditions hold:

  • Proxy leadership intends to abide jus in bello norms and hold violators accountable;
  • A greater injustice will arise if the proxy fails than the injustice represented by the jus in bello violations of the proxy; and,
  • The benefactor is willing to take extra measures to mediate the likelihood of any jus in bello violations.


All these considerations combined suggest that the following norms should guide proxy relationships:29

  • Benefactors bear the greater moral burden to ensure conformity to jus ad bellum and jus in bello norms. Because of this burden, benefactors should make good faith efforts not just to seek non-violent solutions first, but also to ensure the option for such solutions is always open.
  • Where the proxy has resorted to war and has a just cause, benefactor intervention is permissible to the extent it is prudent.
  • Where the proxy’s just resort to war depends on benefactor intervention, the benefactor’s cause must also be, in some sense, just as well as necessary. While it may not be a direct response to aggression against itself, it must serve some overriding good; such as preserving a just international order or deterring future aggression.
  • In cases where the proxy’s cause is unjust, benefactors may intervene only to avoid some gross violation of human rights, humanitarian disaster, or set conditions for a rapid and just resolution to the conflict.
  • Benefactor’s causes, interests, and intentions should align in a way that attains the proxy’s objective and either achieves the benefactor’s goals or at least ends the need for the proxy relationship. When the proxy wins, the proxy relationship should end and transform into something that contributes to order rather than set conditions for future chaos.
  • Benefactors must articulate a reasonable connection between the assistance they provide and the political objectives they and any putative proxy would achieve.
  • Benefactors should enable proxies to fight justly and take measures to minimize costs to all affected parties.
  • Where abandoning a proxy who fights unjustly creates more harm than good, benefactors should take extra measures to hold violators accountable and ensure jus in bello norms are upheld.
  • Benefactors should account for all costs to those affected by the war. Because such cost projections are unreliable, both parties must prepare in advance to favorably end any escalation and limit any diffusion of military assistance to the hands of other bad actors.
  • Benefactors and proxies may keep their relationship secret, but must subject it to reasonable oversight.


Even with these principles, standards, and rules of thumb in mind, the cases discussed here clearly demonstrate how morally perilous proxy relationships are. At the time of this writing, conflicts in Syria and Ukraine are frozen, yet escalating. Saudi operations in Yemen continue to implicate the United States and undermine its moral credibility, despite legitimate efforts to manage this particular moral hazard. Meanwhile, the complicated proxy relationships Iraqi Shia militias have with not just their own government, but also with the Iranians, will continue to impede state consolidation as well profoundly affect the implementation of U.S. security cooperation and assistance programs after the Islamic State’s defeat. The point here, however, is not to avoid such relationships, but rather establish, promulgate, and follow norms that treat partners with respect, align interests with the greater good, and reduce the suffering wars inevitably bring.


     1. John J. Mearsheimer and Stephen M. Walt, “The Case for Offshore Balancing: A Superior U.S. Grand Strategy,” Foreign Affairs, July/August 2016, p. 73. The authors observe that an April 2016 Pew poll found that 57 percent of Americans believed that the United States should “deal with its own problems and let others deal with theirs the best they can.” The authors further attribute this finding to frustration over failures in U.S. military and political interventions. See also Seyom Brown, “Purposes and pitfalls of war by proxy: A systemic analysis,” Small Wars & Insurgencies, Vol. 27, Iss. 2, April 2016, p. 2. Brown describes the implications of a polyarchic world order and proxy relationships.

2. Hal Brands, The Limits of Offshore Balancing, Carlisle, PA: Strategic Studies Institute, U.S. Army War College, 2015, p. 1. In the remainder of this work, Brands argues against over-relying on offshore balancing.

3. Oona A. Hathaway, Emily Chertoff, Lara Domínguez, Zachary Manfredi, and Peter Tzeng, “Ensuring Responsibility: Common Article 1 and State Responsibility for Non-State Actors,” Texas Law Review, Vol. 95, No. 3, February 2017. According to Hathaway et al., the legal standard, however, requires that the benefactor have “effective control” over a proxy’s operations, which entails that the benefactor has to not just intend but also direct any harm the a proxy might commit for the benefactor to be held responsible. For example, in the case of U.S. support to the Contras, the International Court of Justice found that while the United States had “financed, organized, trained, supplied, equipped, and armed” the Contras, it was not ultimately responsible for Contra violations of international humanitarian law, which included killing of unarmed civilians.

4. Andrew Mumford, Proxy Warfare, Cambridge, UK: Polity Press, 2013, p. 11.

5. Ibid., pp. 21-22.

6. Martin Cook, The Moral Warrior, Albany, NY: State University of New York Press, 2004, p. 28. Cook also includes “a better state of peace” as an additional criterion for just war. I have chosen not to include it here since I do not think it will shed additional light on proxy relationships.

7. Barack Obama, National Security Strategy, Washington, DC: The White House, February 2015, p. 2, available from, accessed November 28, 2017.

8. James Pattison, “The Ethics of Arming Rebels,” Ethics & International Affairs, Vol. 29, Iss. 4, Winter 2015, p. 456. Pattison is specifically addressing the arming of nonstate actors engaged in rebellion. However, many of the lessons he derives from his analysis would apply to proxy relationships in general.

9. Thomas Hurka, “Proportionality in the Morality of War,” Philosophy & Public Affairs, Vol. 33, No. 1, Winter 2005, p. 41.

10. Brown, pp. 250-251. See also, Ross Cheit, proj. sup., Sara Chimene-Weiss, Sol Eppel, Jeremy Feigenbaum, Seth Motel, and Ingrid Pangandoyon, researchers, Ingrid Pangandoyon, web des., “Understanding the Iran-Contra Affairs: The Legal Aftermath,” n.d., available from, accessed February 7, 2017.

11. Cook, p. 29.

12. Hooman Majd, “Think Again: Iran’s Green Movement,” Foreign Policy, January 6, 2010, available from, accessed January 23, 2017.

13. Kenneth J. Arrow, “Uncertainty and the Welfare Economics of Medical Care,” The American Economic Review, Vol. 53, Iss. 5, December 1963, pp. 941, 961; see also Matthew McCaffrey, “Moral Hazard: Kenneth Arrow vs. Frank Knight and the Austrians,” Mises Wire, March 14, 2017, available from

14. Arrow, p. 962.

15. Mumford, p. 40.

16. Andreas Krieg, “Externalizing the burden of war: the Obama Doctrine and US foreign policy in the Middle East,” International Affairs, Vol. 92, Iss. 1, January 2016, pp. 109-110.

17. Mumford, p. 100.

18. Krieg, p. 102.

19. Pattison, pp. 458-459.

20. Ibid., p. 459.

21. Barbara Starr and Laura Koran, “U.S. shifts stance on military aid to Ukraine,” CNN, February 2, 2015, available from, accessed February 1, 2017; Peter Foster, “US to increase military aid to Ukraine including Humvees and drones,” The Telegraph, March 11, 2015.

22. Pattison, p. 460.

23. Mumford, p. 109.

24. Hathaway, et al., p. 9.

25. Michael Walzer, “Political Action: The Problem of Dirty Hands,” in Marshall Cohen, Thomas Nagel, and Thomas Scanlon, eds., War and Moral Responsibility, Princeton, NJ: Princeton University Press, 1974, p. 63.

26. Ibid., pp. 63-64.

27. Ibid., p. 81.

28. For a list of convictions, see Cheit et al.

29. This list is not intended to be exhaustive.


The views expressed in this Strategic Insights piece are those of the author and do not necessarily reflect the official policy or position of the Department of the Army, the Department of Defense, or the U.S. Government. This article is cleared for public release; distribution is unlimited.


Organizations interested in reprinting this or other SSI and USAWC Press articles should contact the Editor for Production via email at All organizations granted this right must include the following statement: “Reprinted with permission of the Strategic Studies Institute and U.S. Army War College Press, U.S. Army War College.”


All Strategic Studies Institute (SSI) and U.S. Army War College (USAWC) Press publications may be downloaded free of charge from the SSI website. Hard copies of certain reports may also be obtained free of charge while supplies last by placing an order on the SSI website. Check the website for availability. SSI publications may be quoted or reprinted in part or in full with permission and appropriate credit given to the U.S. Army Strategic Studies Institute and U.S. Army War College Press, U.S. Army War College, Carlisle, PA. Contact SSI by visiting our website at the following address:

“Strategic Insights: Proxy War Norms” by C. Anthony Pfaff (