BOOK REVIEW 
Law, War, and Four Modes of Conflict 
THE INTERNATIONALISTS: HOW A RADICAL PLAN TO OUTLAW WAR 
REMADE THE WORLD. Oona A. Hathaway and Scott J. Shapiro. New 
York, NY: Simon & Schuster, 2018. 
ISBN13: 978-1-50110987-4.  Pp. 640.  $20.00 
Reviewed by Gregory Brazeal* 
ABSTRACT 
The Internationalists argues that the outlawing of war as an 
instrument of state policy in the Kellogg-Briand Pact of 1928 “was 
among the most transformative events in human history, one that has, 
ultimately, made our world far more peaceful.” The authors’ evidence 
for this provocative thesis is ultimately unpersuasive. First, changes 
that they attribute to the Pact began before its creation, as illustrated, 
for example, by the ban on debt collection wars in the second 
convention of the Hague Conventions of 1907. Second, their own 
historical data on wars of conquest suggests that the year 1928 was not 
a turning point between eras, even when viewed in light of the post-
World War II reversals of recent conquests. Rather, the evidence is 
more consistent with the Allies generally reversing the conquests of the 
Axis powers, and preserving their own conquests, regardless of 
whether any conquest took place before or after 1928. Third, their 
interpretation of legal history does not establish that the Kellogg-
Briand Pact played a causally significant role in the creation of the 
postwar legal order. 
* Forrester Fellow, Tulane Law School; Harvard Law School, JD, magna cum laude,
2010; Captain, United States Army Reserve JAG Corps. Thank you to Christopher Szabla 
and David Kennedy for their responses to an earlier draft. The views here are not intended 
to reflect those of the United States Army. 
[531] 
532 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
But The Internationalists offers much more than its central thesis, 
including a fruitful exploration of the idea of law and war as alternative 
forms of conflict resolution. In particular, the authors present a theory 
of “outcasting” as a mechanism for enforcing law even where there is 
no single, centralized government. This review generalizes from the 
authors’ argument to propose a four-mode model of conflict and 
conflict resolution. Against the widespread assumption of a binary 
opposition between anarchy and the state, the four-mode model draws 
attention to two legal-institutional options that lie between pure 
anarchy and the centralized authority of an idealized state.  
Where the parties to a conflict share a law, but not a judge or 
executive, the characteristic method of enforcing the law will be tit-for-
tat retaliation against violations. Armed conflict in this setting will tend 
to resemble a feud. Where the parties to a conflict share a law and a 
judge, but not an executive, community responsibility for enforcing the 
law, such as through outcasting, becomes more feasible. This review 
offers brief historical illustrations and considerations of the likely 
shapes of conflict in each of the four modes. 
INTRODUCTION 
There are basically two forms of conflict resolution: 
administered rules and fighting. Law and war. 
—Paul Bohannan1 
W here should we begin in thinking about the relationship between law and war? 
On the one hand, there is a long tradition, stretching from 
Thucydides’ Melian Dialogue through Hobbes and Clausewitz to 
Kennan and the realist school of international relations theory, that 
views law as essentially irrelevant to war making.2 In this view, states 
1 Paul Bohannan, Introduction, in LAW AND WARFARE xi, xiii (Paul Bohannan ed., 
1967). 
2 See THUCYDIDES, HISTORY OF THE PELOPONNESIAN WAR 402 (Rex Warner trans., 
1972) (presenting argument by the more powerful Athenians to the less powerful Melians 
that “the standard of justice depends on the equality of power to compel and that in fact the 
strong do what they have the power to do and the weak accept what they have to accept”); 
THOMAS HOBBES, LEVIATHAN 86–100 (Richard Tuck ed., 1996) (presenting natural law as 
ineffectual in the anarchic state of nature, leaving only “Warre of every man against every 
man”); CARL VON CLAUSEWITZ, ON WAR 75 (Michael Howard & Peter Paret eds., 1989) 
(1832) (“Attached to force are certain self-imposed, imperceptible limitations hardly worth 
mentioning, known as international law and custom, but they scarcely weaken it.”); GEORGE 
F. KENNAN, AMERICAN DIPLOMACY 95 (expanded ed. 1984) (criticizing “the legalistic-
2019] Law, War, and Four Modes of Conflict 533 
pursue their self-interest in matters of security, and law can do little to 
constrain them. The view of war as fundamentally lawless is also 
echoed in the contemporary pacifist sense that war is “a senseless 
collective slaughter, a descent into irrational barbarism, a horror.”3 
On the other hand, many scholars have found, upon closer 
examination, that wars are often very rule-bound, at least in the sense 
of being highly culturally regulated on one or both sides. Michael 
Howard argues that “Rousseau was surely right when he stated, in 
contradiction to Hobbes, that war without social organization is 
inconceivable.”4 John Keegan frames his A History of Warfare as an 
argument against the Clausewitzian notion of war as a product of pure 
rational calculation and instead, emphasizes the many ways in which 
war is shaped by culture, especially the culture of warriors.5 By 
emphasizing the apparently irrational, passionate, sometimes ritualized 
aspects of warfare, Keegan shows “how much more persistent culture 
is than political decision as a military determinant”6 and thus, how war 
is often constrained—and aggravated—by cultural norms. More 
recently, the legal scholar David Kennedy has tried to draw attention to 
the pervasiveness of law in contemporary war.7 Against his earlier 
sense, as a conscientious objector during the Vietnam era, that the 
military was “all that international law was not—violence and 
moralistic approach to international problems,” for “the belief that it should be possible to 
suppress the chaotic and dangerous aspirations of governments in the international field by 
the acceptance of some system of legal rules and restraints”); John J. Mearsheimer, The 
False Promise of International Institutions, 19 INT’L SECURITY 5, 13 (1994) (“For realists, 
the causes of war and peace are mainly a function of the balance of power, and institutions 
largely mirror the distribution of power in the system. . . . [I]nstitutions are merely an 
intervening variable in the process.”). Sometimes the Latin phrase “inter arma silent leges” 
(the law is silent in times of war) is used to gesture toward the perceived irrelevance of law 
to war. See, e.g., GEOFFREY BEST, WAR & LAW SINCE 1945 3 (1994). 
3 JAMES Q. WHITMAN, THE VERDICT OF BATTLE 1 (2012). 
4 Michael Howard, Temperamenta Belli: Can War Be Controlled?, in RESTRAINTS ON 
WAR 1, 1 (Michael Howard ed., 1979); accord BEST, supra note 2, at 4. Howard also notes 
that despite Clausewitz’s well-known dismissal of international law in ON WAR, 
Clausewitz’s even more well-known assertion of the continuity between war and politics 
implies a recognition that war is not “in its essence ‘uncontrollable.’” Howard, supra note 
2, at 1; see also von Clausewitz, supra note 2. Howard elsewhere observes that Clausewitz 
seems to recognize that “the conduct of war was subject to considerably greater and more 
perceptible limitations in his own time than it had been in the days of, say, Genghis Khan.” 
Michael Howard, Constraints on Warfare, in THE LAWS OF WAR 1, 2 (Michael Howard ed., 
1994). 
5 JOHN KEEGAN, A HISTORY OF WARFARE 3, 5–12, 24 (1993). 
6 Id. at 39. 
7 See generally DAVID KENNEDY, OF WAR AND LAW (2006). 
534 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
aggression in contradiction to our reason and restraint,”8 he describes 
the many ways in which “[l]aw has crept into the war machine. The 
battlespace is as legally regulated as the rest of modern life.”9 A number 
of scholars have also argued that international humanitarian law has 
played a significant legitimizing role in modern war—helping to enable 
it as much as to constrain it.10 
Whether or not the initiation of war can be significantly constrained 
by the international law of jus ad bellum, or the conduct of war by the 
international law of jus in bello, there is a separate sense in which we 
might think about law and war being intimately related. This is the 
sense suggested by the anthropologist Paul Bohannan in the epigraph 
above. War and law can be seen as substitutes, as alternative responses 
to the problem of human conflict. The legal scholar James Q. Whitman 
has argued that the “probably. . . universal human institution” of the 
“contained pitched battle”11 can be understood as a “conflict resolution 
mechanism,”12 and one that “was arguably a more valuable means of 
civilizing war than anything we can offer in our modern law of war.”13 
Oona Hathaway and Scott Shapiro’s recently published book, The 
Internationalists, offers contributions to both threads of thinking about 
the relationship of law to war.14 At the center of their book is the self-
consciously provocative argument that the Kellogg-Briand Pact of 
1928, which purported to outlaw war as an instrument of state policy, 
and which has been almost universally maligned since its apparent 
failure in World War II, was in fact “among the most transformative 
events in human history, one that has, ultimately, made our world far 
more peaceful.”15 Part I of this review will argue that Hathaway and 
Shapiro, or “H&S” as I will call them, ultimately fail to make a 
8 David Kennedy, War and International Law: Distinguishing Military and 
Humanitarian Professions, in THE LAW OF WAR IN THE 21ST CENTURY: WEAPONRY AND 
THE USE OF FORCE 1, 1–2 (Anthony M. Helm ed. 2006). 
9 David Kennedy, Modern War and Modern Law, 16 MINN. J. INT’L L. 471, 472 (2007). 
10 See, e.g., MARTTI KOSKENNIEMI, FROM APOLOGY TO UTOPIA (2006). 
11 WHITMAN, supra note 3, at 6. 
12 Id. at 3. 
13 Id. at 5. 
14 OONA A. HATHAWAY & SCOTT J. SHAPIRO, THE INTERNATIONALISTS 9–11 (2017). 
15 Id. at xiii. The authors refer to the Pact as “the Peace Pact” rather than as “the Kellogg-
Briand Pact,” which is its customary name in the United States. See id. at ix, xi–xii. There 
may be more than cosmopolitanism at work in the choice. One of the book’s arguments is 
that United States Secretary of State Frank Kellogg, who received the 1929 Nobel Peace 
Prize for his work on the Pact, unfairly deprived Salmon Levinson of credit for the Pact. See 
id. at 129–30.  
2019] Law, War, and Four Modes of Conflict 535 
convincing case for the efficacy of the Pact. First, they downplay the 
fact that a significant part of the transformation they describe took place 
before the signing of the Pact in 1928.16 Second, their interpretation of 
historical data unpersuasively minimizes the significance of territorial 
boundary changes that conflict with their thesis.17 Third, their 
interpretation of legal history does not establish that the Kellogg-
Briand Pact played a causally significant role in the creation of the 
postwar legal order, which might be another way of defending the 
significance of the Pact.18 
But the book contains much more than its central thesis on the world-
historical significance of the Kellogg-Briand Pact. As H&S note, the 
“book is, at its core, a work of intellectual history,”19 and this history 
begins with the figure of the early modern jurist Hugo Grotius. H&S 
present Grotius, like Bohannan and Whitman above, as perceiving war 
and law as substitutes: “War is a substitute for courts, Grotius argued, 
because courts are the original substitutes for war.”20 By embedding 
an intellectual history of the use of war and law as substitute modes of 
dispute resolution within a broader history of the attempt to regulate 
war through law, H&S create a productive conceptual tension. They 
simultaneously distinguish and blend the enduring human institutions 
of law and war. Their book could be said to present war as an 
alternative to law that is itself constrained by law.  
This productive tension will be the starting point for the 
investigation of war and law in Part II of this review. Drawing in 
particular on H&S’s discussion of “outcasting” as a mechanism for 
enforcing international law, and their structural analogy between the 
legal institutions of the post-World War II world and of medieval 
Iceland, Part II proposes a novel way of thinking about the general 
relation between war and the law. By disaggregating “the law” as an 
institution into three primary parts—the substance of the law, a judge 
to apply the law, and an executive to enforce the law—it becomes 
16 See infra Part II.A. 
17 See infra Part II.B. 
18 See infra Part II.C. 
19 HATHAWAY & SHAPIRO, supra note 14, at xx. The source of many of the problems 
discussed in Part II may be, at root, that H&S have attempted to graft a causal narrative 
involving several centuries of global history onto materials that are, as they acknowledge 
here, for the most part concerned with the thoughts and writings of a relatively small number 
of practitioners, advocates, and scholars of international law. H&S admirably attempt to 
make up for this methodological conundrum through the statistical analysis of historical 
data, as discussed in Part I.  
20 Id. at 11; cf. id. at 106 (“[w]ar as a legal institution”). 
536 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
possible to distinguish four legal-institutional settings in which a 
conflict might unfold. First, the parties might share neither a law, a 
judge, nor an executive. Second, the parties might share a law, but not 
a judge or an executive. Third, the parties might share a law and a 
judge, but not an executive. Fourth, the parties might share a law, a 
judge, and an executive. Each of these four settings, Part II argues, will 
tend to involve characteristic mechanisms of enforcement and patterns 
of violence. Contrary to the traditional assumption in international 
relations scholarship of a rigid binary distinction between anarchy and 
the state,21 the “four modes of conflict” model proposed here suggests 
that between anarchy and the state there are two fairly discrete and 
stable conditions of legal order. 
The four-mode model of conflict only receives a relatively brief 
sketch in Part II. To the extent that the model is a generalization of 
insights from The Internationalists, it suggests some of the richness of 
the legal and historical materials contained in the book, despite what 
this review argues is the unpersuasiveness of the book’s central claim. 
No standard-length review could summarize all of H&S’s excursions, 
which, in addition to subjects addressed below, include: insightful and 
often entertaining treatments of the transplantation of Grotian 
international law into Japan;22 the legal procedures at the Nuremberg 
tribunals;23 the conflict between Carl Schmitt and Hans Kelsen;24 and 
the rejection of international law in the thought of the Islamic 
fundamentalist Sayyid Qutb.25 
A final aspect of the book that this review will not explicitly address 
is its ideological orientation. The Internationalists is a work of 
liberalism, an expression of Wilsonian optimism in the potentially 
benevolent power of the law.26 Its heroes are Western lawyers who 
believe in the abolition of war through legal rules, and its villains are 
opponents of liberal tolerance, especially on the Right, such as Schmitt 
and Qutb. The book presents the United States as generally a force for 
21 For a critique of the distinction in international relations, see Helen Milner, The 
Assumption of Anarchy in International Relations Theory: A Critique, 17 REV. INT’L STUD. 
67 (1991). 
22 See HATHAWAY & SHAPIRO, supra note 14, at 131–57. 
23 See id. at 276–97. 
24 See id. at 215–43. 
25 See id. at 396–414. 
26 See, e.g., id. at 422 (arguing that “[l]aw creates real power,” that “the New World 
Order is better than the Old,” and that “[d]espite the many challenges, there is reason for 
optimism”). 
2019] Law, War, and Four Modes of Conflict 537 
good in the world, especially through its creation and defense of 
multilateral and international legal institutions.27 It presents 
international economic markets as a tool for encouraging peace and 
punishing aggression.28 By contrast, it labels some (unspecified, but 
presumably illiberal) contemporary states as “evil.”29 The book’s 
Conclusion is a rhetorically heightened work of policy advocacy that 
appears intended to persuade Western elites, and especially lawyers 
and lawmakers in the United States, to stay the course and continue to 
believe that international law is worth improving and defending.30 
Much of this will appear bafflingly naive, or even offensive, to 
critics of liberalism on the Left and the Right. They may be astonished 
to read claims like the following, about the postwar order: “Compulsion 
by war was over. The era of global cooperation had begun.”31 A reader 
approaching the book from the critical, post-Marxist Left may wish to 
object that H&S pay inadequate attention to what the Left might 
perceive as liberalism’s hypocrisies and crimes against subordinated 
groups, and the global South in general, from the racist origins of 
international law as a professional movement in the later nineteenth 
century; through the CIA’s Cold War adventures in Iran, Guatemala, 
Cuba, and elsewhere; through the human and environmental 
devastation caused by coercive Washington Consensus policies in the 
1990s; to the dismissal of international law concerns in the 1998 
Kosovo and 2003 Iraq invasions, and currently in Syria—not to 
mention the United States’ vacillating hostility to the International 
Criminal Court or its failure to punish many of those responsible for 
the use of torture during the Bush administration.32 When H&S present 
27 See, e.g., id. at 418–19 (arguing that the continued success of the New World Order, 
including the decline of interstate war and the near disappearance of conquest “depends on 
the willingness of the United States to continue to play a central role in maintaining the legal 
order”). 
28 See, e.g., id. at 420 (“Free trade not only channels productive activity away from war 
making, it provides a legal tool for disciplining states who violate the rules.”); id. (stating 
that adherence to the WTO’s free trade rules “creates jobs and raises wages overall”).  
29 Id. at 369. 
30 See, e.g., id. at 423 (“Those who shape the law are the hydraulic engineers of the 
political world. . . . Each of us, even those outside the halls of government, has the capacity 
to make a difference.”). 
31 Id. at 305. 
32 H&S mention some of these issues in passing. See, e.g., id. at 329–30 (noting the 1953 
coup in Iran and the 2003 invasion of Iraq as exceptions to the general picture of state 
security from foreign interference increasing in the postwar era); see also id. at 381 
(acknowledging that outcasting “often favors larger, stronger states over smaller, weaker 
ones”). Even if post-Marxist historical perspectives are set aside, H&S’s account of liberal 
538 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
the post-World War II decline of interstate wars and especially 
territorial conquest in a positive light, only partially dimmed by the 
related rise of intrastate conflict, a reader on the Left will probably 
object that this largely Whiggish history leaves out the voices of the 
victims of past and continuing Western economic and military 
oppression and exploitation. 
H&S ultimately offer little engagement with the critical perspective 
of the Left, and I will not attempt to construct a dialogue here. If 
anything, The Internationalists is more productively read as a liberal 
response to the critique of liberal international institutions from the 
Right, such as by realist theorists of international relations.33 Central to 
H&S’s argument is the claim that law has mattered to the decline of 
interstate war, and especially wars of conquest, as described in the next 
section. 
I 
THE INTERNATIONAL LAW OF CAUSATION 
One way of approaching The Internationalists’ treatment of the 
Kellogg-Briand Pact is to view it as an answer to the question: Why 
have wars of conquest declined in frequency and size since the middle 
of the twentieth century? 
As H&S note, scholars from a variety of disciplines have proposed 
explanations for the post-World War II decline of interstate war,34 a 
phenomenon that is sometimes described, in a phrase from the historian 
John Gaddis, as “the Long Peace.”35 The cognitive psychologist Steven 
international law and American power is remarkably less critical than what one finds in, for 
example, MARK MAZOWER, GOVERNING THE WORLD 65–93 (2012) (describing the racist 
underpinnings of the rise of international law in the later nineteenth century); EREZ 
MANELA, THE WILSONIAN MOMENT 25 (2009) (describing Wilson’s racist vision of 
delayed self-determination for non-European peoples). See generally TIM WEINER, LEGACY 
OF ASHES: THE HISTORY OF THE CIA (2007) (recounting various, often violent episodes of 
CIA interference in other states’ sovereignty); and JOSEPH STIGLITZ, GLOBALIZATION AND 
ITS DISCONTENTS (2002) (drawing attention to the failures of the Washington Consensus). 
33 For one suggestion of the underlying genealogical and conceptual affinities between 
international relations realism and the Right, see William E. Scheuerman, Carl Schmitt and 
Hans Morgenthau: Realism and Beyond, in REALISM RECONSIDERED: THE LEGACY OF 
HANS J. MORGENTHAU IN INTERNATIONAL RELATIONS 62 (Michael C. Williams ed., 2007). 
34 HATHAWAY & SHAPIRO, supra note 14, at 334 (noting that political scientists have 
emphasized the importance of “the invention of nuclear weapons, the spread of democracy, 
[or] the rise of global trade” as explanations). 
35 STEVEN PINKER, THE BETTER ANGELS OF OUR NATURE 190 (2011) (citing John 
Gaddis, The Long Peace: Elements of Stability in the Postwar International System, 10 
INT’L SECURITY 99 (1986)). H&S ally themselves with the “declinist” hypothesis regarding 
2019] Law, War, and Four Modes of Conflict 539 
Pinker summarizes much of the research in his interdisciplinary, and 
highly contested, 2011 work The Better Angels of Our Nature.36 Pinker 
notes that attempts to understand the unexpectedly enduring aspects of 
postwar peace began to appear as early as the 1980s.37 The main 
explanations, none of them mutually exclusive, have been: the role of 
nuclear weapons in deterring war,38 the rise of democracy and the 
relative peace between democratic states,39 the rise of liberalism and 
the pacific effects of international commerce,40 and the pacifying 
growth of intergovernmental organizations.41 
The extraordinary novelty of The Internationalists’ contribution to 
the literature of the Long Peace lies in its argument that international 
law in general, and the Kellogg-Briand Pact in particular, played a 
central causal role.42 No prominent scholarly work before The 
Internationalists has so unapologetically adopted the opposite stance 
from the realist school of international relations theory, asserting not 
only that international law generally played some causally significant 
war and violence, favorably citing Nils Petter Gleditsch’s claim that “[d]espite the various 
critiques, there is wide agreement on the decline of war and other forms of violence. . . . 
However, the reasons for the decline are less clear.” Nils Petter Gleditsch, The Decline of 
War, 15 INT’L STUD. REV. 396 (2013), cited in HATHAWAY & SHAPIRO, supra note 14, at 
533 n.37. 
36 See PINKER, supra note 35, at 189–294. 
37 Id. at 253–55 (citing works and observations by Werner Levi, John Gaddis, Kalevi 
Holsti, Evan Luard, John Mueller, Francis Fukuyama, James Lee Ray, Carl Kaysen, Robert 
Jervis, John Keegan, Michael Howard, and Azar Gat). 
38 Id. at 268–78. 
39 Id. at 278–84. Related to the thesis of a democratic peace is the idea that American 
hegemony within the liberal international order ensured peace—first within the United 
States’ sphere of influence during the Cold War, then in the world as a whole after the Cold 
War. Cf. G. John Ikenberry, The End of Liberal International Order?, 94 INT’L AFF. 7, 8 
(2018). 
40 See PINKER, supra note 35, at 284–88. 
41 See Id. at 288–94. To these, Pinker adds his own explanation, which might be 
described as “constructivist” if it appeared in a work of international relations scholarship: 
it is grounded in the power of culture to reshape interests and norms. He argues that the 
Long Peace is  
a result of one of those psychological retunings that take place now and again over 
the course of history and cause violence to decline. In this case it is a change within 
the mainstream of the developed world (and increasingly, the rest of the world) in 
the shared cognitive categorization of war. 
Id. at 251. Pinker suggests that this cultural shift, which has been especially pronounced in 
Europe, may underlie many of the other variables such as democracy and trade that have 
been shown to correlate with the decline of interstate war. See id. at 290–91. 
42 The authors themselves note the novelty of their central claim. See HATHAWAY & 
SHAPIRO, supra note 14, at xii–xiii. 
540 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
role in shaping states’ decisions regarding the initiation of armed 
conflict during the twentieth century, but that the Kellogg-Briand Pact 
specifically played one of the causally central roles. One of the authors 
of The Internationalists, in an earlier review, referred to a particularly 
skeptical work of international law scholarship as a “minimal 
rationalist theory.”43 In contrast, The Internationalists might be 
described as a “maximal legalist theory.” It places law at the causal 
center of international affairs.44 
To be sure, H&S acknowledge that the adoption of the Kellogg-
Briand Pact was not sufficient by itself to cause the decline of wars of 
conquest.45 But they go on to state that the Pact was “a necessary 
start.”46 Although “the Pact was not the only factor responsible for the 
transformation” of the international order, its “outlawry of war was a 
crucial . . . trigger. It sparked a series of events that would lead to the 
construction of a new global order.”47 If the Kellogg-Briand Pact (or, 
presumably, another legal instrument like it) had not come into 
existence, H&S seem to argue, the decline of wars of conquest would 
not—perhaps could not—have taken place.  
In fact, the details of H&S’s argument suggest that they do not view 
the Pact as merely a necessary condition for the decline of wars of 
conquest. They do not view it as one condition among many, perhaps 
a necessary hurdle to be overcome but not the most significant 
transformation in the broader history. Rather, their narrative places law 
in general, and the Pact in particular, at or near the causal center of the 
history of armed conflict in the postwar era. As already noted, they 
present the Pact as “among the most transformative events of human 
43 Oona A. Hathaway & Ariel N. Lavinbuk, Rationalism and Revisionism in 
International Law, 119 HARV. L. REV. 1404, 1407 (2006) (reviewing JACK L. GOLDSMITH 
& ERIC A. POSNER, THE LIMITS OF INTERNATIONAL LAW (2005)). 
44 The Internationalists could also be seen as “legalist” in the sense of adhering to the 
traditional ideology of the legal profession, with its liberal aspiration toward the resolution 
of conflicts through ostensibly neutral, apolitical rule following. See JUDITH SHKLAR, 
LEGALISM 18 (1986); accord MARTHA FINNEMORE, THE PURPOSE OF INTERVENTION: 
CHANGING BELIEFS ABOUT THE USE OF FORCE 38–46 (2003). 
45 “The Pact did not bring about the end of conquest and interstate war on its own; no 
treaty, no law could have.” HATHAWAY & SHAPIRO, supra note 14, at 335. 
46 Id. (emphasis added). Elsewhere, H&S again acknowledge that the Pact “was not 
sufficient by itself” to bring about the decline of wars of conquest, but rather required Allied 
victory in World War II, the postwar reversal of conquests since 1928, and the development 
of the U.N. Charter. See id. at 331. On another occasion, H&S also make the more modest 
claim that the interwar changes in international law, and in some state practices, “resulted 
at least in part from the Pact.” Id. at 330. 
47 Id. at xviii. 
2019] Law, War, and Four Modes of Conflict 541 
history.”48 It “marked the beginning of the end” of “war between 
states,” and “the replacement of one international order with another.”49 
The “cascade of events” made possible by the Pact include the 
reshaping of the world map in favor of smaller, weaker states; the 
human rights revolution; “the use of economic sanctions as a tool of 
law enforcement”; and the “explosion in the number of international 
organizations.”50 It would not be an exaggeration to say that the 
historical narrative presented by The Internationalists is one in which 
international law changes, and in its wake international history, 
including the history of war and peace, unfolds. The Kellogg-Briand 
Pact, far from being a largely futile or perhaps perverse gesture of 
interwar utopian liberalism, becomes an indirectly but profoundly 
efficacious turning point in world history: a necessary and central, 
though obviously not sufficient, condition for the existence of the world 
we live in today.51 
The overarching outline of H&S’s history is easily summarized. It is 
the story of a transformation from what they call the “Old World 
Order” to the “New World Order,” with the Kellogg-Briand Pact acting 
as the hinge between the two. The Old World Order arose in early 
modern Europe, and its rules are reflected in the writings of Hugo 
Grotius and his successors. It was “the legal regime European states 
adopted in the seventeenth century and spent the next three centuries 
48 Id. at xiii. 
49 Id. at xiii; see also id. at 334 (“The outlawry of war not only led to the end of conquest. 
It precipitated the end of international war itself.”); id. at 418 (noting that the Pact has 
“clearly” worked, in the sense that “interstate war has declined precipitously, and conquests 
have almost completely disappeared.”); id. at xiv (“The Pact was aimed at ending war 
between states and, in that, it proved remarkably successful.”). 
50 Id. at xv. H&S even present World War II in language that comes close to suggesting 
it was fought over differences regarding the Kellogg-Briand Pact. The war, they write, 
“would become a contest between two competing visions of the world: between one that 
saw the Pact as a piece of paper and one that saw it as a new legal reality.” Id. at 182; see 
also id. at 188, 213; id. at 193 (“The war was no longer about who would control what 
territory. It was about what rules would govern when the war was over.”); id. at 332 (“[T]he 
winners had just fought a war in which the rallying cry was the rejection of territorial 
aggrandizement by force. That rallying cry was rooted in the Pact.”). But see id. at 250 
(noting that “the Allies were far more concerned with victory than with its legal 
consequences.”). 
51 Another example of the explicitly causal nature of H&S’s claims regarding the 
significance of international law arrives in their claim that “[f]rom our bird’s-eye view, it is 
possible to see what observers on the ground too often miss: that what was once 
frighteningly common is now thankfully infrequent”—that is, wars of conquest—“because 
what was once seen as the embodiment of international law is now understood as its 
repudiation.” Id. at 329. 
542 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
imposing on the rest of the globe.”52 One of the central rules of the Old 
World Order was “the right of conquest”: “Any state that claimed it had 
been wronged by another state, and whose demands for reparations 
were ignored, could retaliate with force and capture territory as 
compensation.”53 In the absence of a shared legal authority to resolve 
disputes through litigation, states resorted to war, and the rules of the 
Old World Order effectively declared that “Might was Right,” meaning 
that “success creates legal rights in war.”54 When a state succeeded in 
conquering the territory of another state, “[i]t owned all public property 
and possessed the legal authority to rule over its subjects.”55 H&S show 
how Grotius, after initially attempting to condition the legal rights of 
the conquering state to some extent on the justice of the state’s cause,56 
eventually came to accept in his most influential work, The Law of War 
and Peace, that under the law of nations, the victorious state’s cause 
must be treated as if it had been just.57 
H&S emphasize that the rules of the Old World Order were 
“binding,” “understood to be obligatory,” and that “sovereigns largely 
obeyed them.”58 The first part of The Internationalists (“Old World 
Order”) features a number of historical anecdotes illustrating political 
figures, and not only theoreticians such as Grotius, appealing to and 
apparently complying with the rules of the Old World Order.59 The 
book helpfully draws attention to the once-routine state practice of 
issuing a war manifesto containing the assertion of a right that the war 
is intended to uphold.60 H&S show that as late as the 1840s, United 
States President James Polk was careful to justify his war against 
52 Id. at xv. 
53 Id. 
54 Id. at xv, 23–24. H&S note that “Grotius was not the first to propose the Might is 
Right Principle.” See also id. at 25 (earlier articulations of the principle in the works of 
Raphael Fulgosius and Balthazar Ayala). 
55 Id. at xv. 
56 See id. at 11, 23–24. 
57 Id. at 23–24, 53–55 (“Winning in war meant winning in law”). 
58 Id. at xv. 
59 It should be noted that H&S also gesture occasionally toward alternate explanations 
for state behavior that do not give a central role to international law. See id. at 171 (“Wright 
would later speculate that the attraction of the new policy stemmed less from an 
interpretation of international law and more from practical imperatives.”); id. at 177 
(“Favoring one side in the fight, [Woodring] warned, would lead to war, regardless of what 
Roosevelt (or international legal scholars) said.”). 
60 See id. at 31–55; id. at 46 (“The function of manifestos was to explain the legal basis 
of the war being waged.”). H&S have also created an online compendium of war manifestos, 
available at documents.law.yale.edu/manifestos. 
2019] Law, War, and Four Modes of Conflict 543 
Mexico in part as a means of collecting unpaid debts—a perfectly 
legitimate basis for a war of conquest under the rules of the Old World 
Order.61 They also provide examples of diplomats such as Thomas 
Jefferson, when he was secretary of state during the Washington 
administration, insisting on the duty of neutral states to maintain strict 
impartiality between warring parties.62 As H&S emphasize, the notion 
that neutrality entailed strict impartiality, and thus prohibited the use of 
economic aid or sanctions, was another important aspect of the Grotian 
law of nations for most of the period of the Old World Order.63 
A skeptic of the efficacy of international law might question some 
of H&S’s phrasing concerning the legal rules of war and peace laid out 
in treatises by writers such as Grotius and Vattel. H&S state that “each 
stage” of war in the Old World Order was “carefully regulated.”64 But 
is it a meaningful “regulation” of the decision to go to war to require 
that a certain kind of justification be formally articulated, even if the 
justification can be utterly pretextual? Such “regulation” does not in 
fact prevent a state from initiating war whenever it chooses to do so, 
for whatever actual reason. The obligatory formalities at the start of a 
war under the Old World Order might just as well be categorized as jus 
in bello regulations, and the “Might is Right” principle as a regulation 
of jus post bellum procedures placing no constraint on the decision to 
go to war, leaving the category of jus ad bellum effectively empty.65  
In other words, to the extent that states were free under the Old 
World Order to initiate war against any state at any time and on any 
actual basis, so long as they carried out the proper formalities, it is 
difficult to see how the substance of Old World Order legal doctrine 
“regulated” the decision to go to war in any meaningful sense. When 
H&S state that under the Old World Order, “[w]ars could not be entered 
for any reason whatsoever” but “had to be made in the name of 
justice”66 the second claim may be true, but the first appears to be 
61 See id. at 33–35. But see infra Part II.A (prohibition of debt collection wars in 1907 
Hague Conventions). 
62 HATHAWAY & SHAPIRO, supra note 14, at 85. 
63 See, e.g., id. at 90–91. 
64 Id. at 35. 
65 The Japanese scholars who responded to the importation of the ideas of Grotius seem 
to have perceived the unusual nature of an international “law” in which the actions of the 
stronger party were by definition “lawful.” Some objected to the use of the Japanese term 
“hō” (law) for the word “law” in translations of “international law”: “Understandably, they 
saw Western ‘international law’ as unpredictable and inconsistent, reflecting the value of 
‘might,’ not ‘law.’” Id. at 145. 
66 Id. at 44. 
544 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
nearly the opposite of the practical reality that The Internationalists 
itself describes. Wars could, in fact, be entered for any reason 
whatsoever, provided that they were accompanied by a formal 
declaration facially asserting a just cause. The Old World Order’s 
treatment of the decision to go to war might better be seen as the 
absence of law than as a legal rule. 
The second part of the book, “Transformation,” presents an 
engaging account of how several largely uncelebrated figures helped to 
transform the international law of jus ad bellum after the Grotian order 
failed to prevent the catastrophe of World War I.67 These figures are 
the “Internationalists” of the book’s title: Salmon O. Levinson, who 
formed a social movement in support of outlawing war and shepherded 
the creation of the Kellogg-Briand Pact; James T. Shotwell, who 
ghostwrote the Pact and then incorporated it into the first draft of the 
United Nations Charter; Sumner Welles, who also played a central role 
in the creation of the United Nations; and Hersch Lauterpacht, a legal 
scholar who helped to theorize and extend the changes brought about 
by the Pact.68 
As H&S present the story, the core of Grotius’s rules, including the 
right of conquest, the “Might is Right” principle, and neutrality as strict 
impartiality, remained intact until the signing of the Kellogg-Briand 
Pact in 1928. The signing of the Pact triggered a series of legal 
transformations that ultimately culminated in the creation of the “New 
World Order,” the subject of the third part of the book. H&S argue that 
the outlawry of war in the Pact created practical problems in other areas 
of international law that led some scholars and diplomats in the 
interwar years to argue for broader changes. If aggressive war was no 
longer lawful, then how was the law to be enforced? How was a state 
that initiated an aggressive war—such as Japan in its 1931 invasion of 
Manchuria—to be punished? Under the Old World Order, economic 
sanctions against the aggressive state would constitute a violation of 
neutrality. The Stimson Doctrine, later echoed by the League of 
Nations, departed from the Grotian law of neutrality by declaring that 
the United States would not recognize an aggressive state’s legal right 
67 “The ‘Great War’ was a true Grotian war—launched to right wrongs both real and 
imagined.” Id. at 104. 
68 See id. at xxi, 106–15, 125–30 (Levinson), 115–21, 194–98 (Shotwell), 183–88, 197–
98 (Welles), 238–39, 246–52 (Lauterpacht). Henry Stimson also plays a pivotal role. See id. 
at 163–68, 177–79; id. at 169 (“It would be hard to overemphasize the importance of the 
League’s acceptance of the Stimson Doctrine”). H&S are apparently the first scholars to 
identify Shotwell as the author of the first draft of the U.N. Charter. See id. at 196. 
2019] Law, War, and Four Modes of Conflict 545 
to conquered territory. As H&S put it, “The Stimson Doctrine was the 
first step in dismantling the Old World Order and constructing a new 
system of law.”69 
By the end of H&S’s story of transformation, the New World Order 
in which we currently live had come into being. Under this New World 
Order, the “rules governing conquest, criminal liability, gunboat 
diplomacy, and neutrality” are “precisely the opposite” of what they 
were under the Old World Order:70  
[A]ggressive wars are illegal. And because aggressive wars are
illegal, states no longer have the right to conquer other states; waging
an aggressive war is a grave crime; gunboat diplomacy is no longer
legitimate; and economic sanctions are not only legal, but the
standard way in which international law is enforced.71
The Internationalists makes a persuasive case that since the end of 
World War II, we have lived in a radically different world of 
international law from the one described by Grotius and his successors, 
and that during the same period, the nature of interstate (and intrastate) 
conflict has changed dramatically, including through the sharp decline 
of wars of conquest. But was the Kellogg-Briand Pact in fact a causally 
significant factor in the establishment of the post-1945 world order?  
As noted in the Introduction, there are at least three reasons to doubt 
H&S’s assertions regarding the historical significance of the Pact. The 
first concerns transformations before the signing of the Pact in 1928. 
The second concerns H&S’s statistical evidence for the efficacy of the 
Pact. The third concerns their evidence from legal history. 
A. Transformations Before the Pact
The Internationalists suggests that the Kellogg-Briand Pact was 
responsible for a normative movement away from the view that the use 
of military force is an appropriate way of resolving interstate disputes. 
As an illustration of this change, H&S focus on the use of military force 
to collect debts. They include a lengthy historical vignette on the 
Mexican-American War. “President Polk,” H&S write, “celebrated the 
conquest of Mexican territory and justified it by claiming that the U.S. 
Army was collecting unpaid debts. After 1928, however, such wars 
were no longer considered just.”72 Similarly, they write that “[a]fter 
69 Id. at 169, 163–70. 
70 Id. at xvii. 
71 Id. 
72 Id. at 333. 
546 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
1928,” war was no longer seen as an “appropriate legal means of 
resolving disputes.”73 “Today, war is regarded as a departure from 
civilized politics. But . . . [b]efore 1928, every state accepted the 
opposite position.”74 
But revulsion toward the use of war to collect debts, or more 
generally toward the view of war as civilized politics, did not begin in 
1928. The Kellogg-Briand Pact was not even the first international 
treaty attempting to ban the use of force to collect debts. As H&S 
quietly mention in a single footnote, the second convention of the 
Hague Conventions of 1907 had already prohibited debt collection 
wars.75 The convention begins:  
The Contracting Powers agree not to have recourse to armed force 
for the recovery of contract debts claimed from the Government of 
one country by the Government of another country as being due to 
its nationals. 
This undertaking is, however, not applicable when the debtor 
State refuses or neglects to reply to an offer of arbitration, or, after 
accepting the offer, prevents any compromise from being agreed on, 
or, after the arbitration, fails to submit to the award.76 
As the political scientist Martha Finnemore makes clear in Chapter 
Two of her 2003 book The Purpose of Intervention, 1907 was the 
pivotal year in the outlawry of debt collection wars, not 1928.77 After 
1907, “intervention behavior among states changed. European states 
ceased interventions to collect contract debts from foreign 
governments.”78 In fact, the United States already had a policy from at 
least 1885 against the use of military force to collect debts.79 Neither 
the United States’ policy nor the willingness of the state parties to the 
1907 Hague Convention to agree to the outlawing of debt collection 
wars is consistent with H&S’s presentation of 1928 as the year that 
states stopped viewing debt collection wars as “just,”80 or war in 
general as an “appropriate legal means of resolving disputes.”81 
73 Id. at 335. 
74 Id. at xiv. 
75 See id. at 445 n.27; see also Convention Respecting the Limitation of the Employment 
of Force for the Recovery of Contract Debts, Oct. 18, 1907, 36 Stat. 2241, 1 Bevans 607 
[hereinafter Second Hague Convention of 1907]. 
76 Second Hague Convention of 1907, art. I. 
77 See FINNEMORE, supra note 44, at 24–51.  
78 Id. at 46. 
79 See id. at 46 n.60. 
80 HATHAWAY & SHAPIRO, supra note 14, at 333. 
81 Id. at 335. 
2019] Law, War, and Four Modes of Conflict 547 
Moreover, less than a decade after 1907, the Covenant of the League 
of Nations attempted to establish an even broader prohibition on the 
use of force to resolve interstate disputes, although one with a larger 
loophole. As H&S observe, the League Covenant created a 
“compulsory system of dispute resolution.”82 If either side was 
dissatisfied with the judgment, however, “it could resort to war 
provided it waited three months before doing so.”83 Thus, the League 
Covenant created no new categorical ban on the use of military force 
as an instrument of policy. But it expressed in its preamble a normative 
commitment reflecting the changes that were already well underway 
before 1928: the parties’ “acceptance of obligations not to resort to 
war.”84 Whatever the lack of practical efficacy of the League Covenant, 
it is surely evidence that before 1928, there was already a growing 
sense that war represented “a departure from civilized politics.”85 
Even if one accepted that the Kellogg-Briand Pact (“the Pact”) 
represented the beginning of a legal movement away from the Grotian 
acceptance of aggressive war, it would be hard to deny that in cultural 
terms the Pact was more the result of a transformation than the cause 
of one. H&S suggest that “[l]egal revolutions do not end with the 
passing of a law. They begin with them.”86 It might be more accurate 
to say that legal revolutions usually do not even begin with the passing 
of a law. Especially in the case of the outlawry of war, surely no legal 
revolution would have been possible without the gradual cultural shift 
away from the traditional valorization of war in Western society, 
especially among the elites most closely involved in shaping the law. 
It might also be argued that legal transformations cannot be understood 
in the absence of attention to changes in the distribution of economic 
and political power.87 
82 Id. at 105.  
83 Id.; see Covenant of the League of Nations, art. 12, Dec. 1924. 
84 Covenant of the League of Nations, preamble, Dec. 1924. 
85 HATHAWAY & SHAPIRO, supra note 14, at xiv. 
86 Id. at 331. 
87 On the relative decline of martial culture and the relative rise of abhorrence toward 
war throughout the West since the later eighteenth century, and especially since World War 
I, see, for example, IMMANUEL KANT, Toward Perpetual Peace, in PRACTICAL 
PHILOSOPHY 311 (Mary J. Gregor ed., 1996) (1795); BENJAMIN CONSTANT, The Spirit of 
Conquest and Usurpation and Their Relation to European Civilization, in POLITICAL 
WRITINGS 44 (Biancamaria Fontana ed., 1988) (1815); JAMES J. SHEEHAN, THE 
MONOPOLY OF VIOLENCE: WHY EUROPEANS HATE GOING TO WAR (2007); MARTIN VAN 
CREVELD, THE CULTURE OF WAR 249–332 (2008); PINKER, supra note 35, at 23–24, 242–
43, 263–67; MAZOWER, supra note 32, at 1–38, 66 (noting, among other things, 
international peace conferences in 1849 and 1851). The historiography of transformations 
548 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
To suggest that the norm against aggressive war in general, and 
conquest in particular, likely developed before the Pact and helped 
make it possible to a greater extent than the Pact led to the creation of 
the norm, is not to say that international law plays no role in shaping 
and solidifying norms.88 It is not implausible that the Pact may have 
focused interwar international attention on the growing taboo against 
wars of conquest. But did its signing help to solidify the norm more 
than its failure contributed to the undermining of the norm? Did the 
norm persist into the postwar era more in spite of the Pact than because 
of it? Approaching these questions would probably require fine-grained 
historical attention to how the Pact and the norm against conquest was 
invoked, or not invoked, not only by legal scholars but by the media, 
the public, and government officials. This is not something that The 
Internationalists attempts to do.89 
If H&S’s argument simply used the Kellogg-Briand Pact as a kind 
of synecdoche for the broader cultural transformation of which it was 
a part, their claims could be more easily defended. But, as presented 
throughout the remainder of this Part, H&S’s claims seem to be less 
concerned with symbolic associations and more with literal historical 
causality. 
B. Statistical-Historical Evidence for the Pact’s Efficacy
H&S suggest that a simple, counting-based statistical analysis of 
historical evidence supports their claim that the Pact was a pivotal event 
in the decline of wars of conquest. Drawing on the Correlates of War 
in cultural attitudes toward war remains highly disputed and unsettled, as is the relation of 
those transformations to more specific and concrete military, legal, political, and economic 
developments. See, e.g., LAUREN BENTON & LISA FORD, RAGE FOR ORDER: THE BRITISH 
EMPIRE AND THE ORIGINS OF INTERNATIONAL LAW, 1800–1850 (2016) (arguing that 
scholars have concentrated too narrowly on the role of intellectual history in the 
development of international law). 
88 See, e.g., FINNEMORE, supra note 44, at 72 (noting that “international law, 
international regimes, and the mandates of formal international organizations” 
“[i]ncreasingly since the nineteenth century” have “channel[ed] . . . normative 
coevolution”). 
89 The historian Gary Bass has noted that while H&S “do a great job” of describing 
“norm entrepreneurs” like Lauterpacht, the “next step” in evaluating “the observable 
implications” of their argument would be to study “how the norm cascades outward after 
1928 from their norm entrepreneurs into a wider world,” for example by investigating 
government archives for “the hidden debates and deliberations which result in policy.” Gary 
Bass, The Internationalists Mini-Forum: Why Has War Declined?, JUST SECURITY (Nov. 
15, 2017), https://www.justsecurity.org/47093/internationalists-mini-forum-idea-
international-law-decline-war/. 
2019] Law, War, and Four Modes of Conflict 549 
project, H&S and “eighteen brilliant Yale Law students” built a dataset 
of 254 instances of territorial change involving militarized conflict 
between 1816 and 2014.90 Based on this dataset, H&S conclude that 
“[c]onquest, once common, has nearly disappeared,” and “the switch 
point” was 1928, the year of the Pact.91 After the Pact, H&S write, 
“[m]ight still produced military victories. But it could no longer 
provide lasting legal victories.”92 They readily acknowledge that the 
average number of conquests per year between 1929 and 1948 did not 
change significantly from the average number between 1816 and 
1928.93 This would of course seem to suggest, at least at first glance, 
that the Pact had no effect on the frequency of wars of conquest, as its 
dismissive critics have always maintained, and that any later change in 
the frequency of conquest resulted from events at the end of World War 
II. H&S’s statistical argument for the causal efficacy of the Pact rests
on the fact that “in 1931, states began to refuse to recognize conquests,”
and “[e]ven more remarkabl[y], with the exception of Taiwan, all the
unrecognized transfers of territory between 1928 and 1949 were later
reversed.”94 In other words, “while territory continued to be seized after
the Peace Pact went into effect, the Pact meant that transfers of control
over territory did not, except in rare cases, translate into legal rights
over that territory.”95 In sum, “few conquests have stuck since the Pact
went into effect.”96
90 HATHAWAY & SHAPIRO, supra note 14, at 312–13. The dataset is available at 
www.theinternationalistsbook.com. It excludes certain categories of militarized territorial 
change, such as “a reversal of an earlier unrecognized seizure of the same territory.” Id. at 
313. Thus, China’s 1945 seizure of Manchuria from Japan is not one of the 254 instances of
possible conquest in the dataset, while Japan’s seizure of Manchuria in 1931 is. See id. Also,
H&S self-consciously exclude “nonconquests” where a powerful state uses force to cause
or prevent a revolution in a weaker state, but does not then annex the weaker state’s territory,
such as the CIA’s coup in Iran in 1953, or the USSR’s armed intervention in Hungary in
1956 or Czechoslovakia in 1968. See id. at 329–30.
91 Id. at 313. 
92 Id. at 316. 
93 See id. at 313–14. In fact, the likelihood in any given year that the average state would 
be the victim of conquest went slightly up during the 1929–1948 period as opposed to the 
1816–1928 period, from a 1.33 percent chance to a 1.8 percent chance. See id. For the use 
of 1948 as the year of postwar settlement, see HATHAWAY & SHAPIRO, supra note 14, at 
320 (noting that postwar territorial transfers did not conclude for legal purposes until 1948, 
the deadline for the completion of various treaty obligations under the Paris Peace Treaties 
of 1947). 
94 Id. at 318. 
95 Id. at 319. 
96 Id. at 330. 
550 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
But does H&S’s data in fact support the conclusion that the Pact was 
a causally significant historical turning point in the decline of 
conquest? We might compare the alternative hypothesis that the Pact 
made little difference to state behavior at any point, and that the 
phenomena captured in H&S’s statistics would have taken place 
whether or not the Pact was ever signed. To begin with, the decline in 
conquests after 1948 can be explained without reference to the Pact, as 
the many proposed explanations of “the Long Peace” have attempted 
to do, for example by pointing to some combination of nuclear 
weapons, the spread of democracy, and changing cultural values 
regarding colonialism and the use of military force.97 Even if we seek 
a purely legalistic explanation, it would seem to make more sense to 
present the post-1948 decline of conquest as a result of the U.N. Charter 
rather than of the Kellogg-Briand Pact.98  
The crucial period for evaluating the causal efficacy of the Pact is 
thus the period between 1928 and 1948. Were the conquests between 
1928 and 1948 later reversed as a result of the Pact—that is, because of 
their illegality under the Pact—or for some other reason? An alternate 
explanation might be that the conquests between 1928 and 1948 that 
were later reversed were reversed because they were carried out by 
Axis powers that were defeated in World War II, and the victors in that 
war did not want their defeated enemies to continue to control recently 
conquered territories. What evidence might be relevant to evaluating 
the relative plausibility of this theory in comparison to H&S’s theory 
that the reversals took place as a result of the conquests’ illegality under 
the Pact? 
First, we might be interested to see whether any post-1928 conquests 
by the Allied victors of World War II, as opposed to the defeated Axis 
97 See supra text accompanying n. 35–42. 
98 When Pinker, for example, addresses international law as a factor in the Long Peace, 
he attributes to the U.N. Charter what H&S attribute to the Pact:  
One paradoxical contributor to the Long Peace was the freezing of national borders. 
The United Nations initiated a norm that existing states and their borders were 
sacrosanct. By demonizing any attempt to change them by force as “aggression,” the 
new understanding took territorial expansion off the table as a legitimate move in the 
game of international relations. 
PINKER, supra note 35, at 258. As suggested above, the norm of respecting state borders had 
been growing long before either the U.N. Charter or the Pact. But Pinker’s association of 
the United Nations with the norm reflects the basic truth that the U.N. Charter was sufficient 
by itself to establish the illegality of aggressive cross-border invasions. For reasons 
discussed at infra Part II.C., there is no basis for concluding that the Pact was a necessary 
or causally important precursor to the Charter. 
2019] Law, War, and Four Modes of Conflict 551 
powers, were reversed. If the Pact was truly the reason for the reversals 
of conquests between 1928 and 1948, rather than the reversals resulting 
from victors imposing their preferences on defeated states as they 
would have done regardless of the Pact, then we might expect to see 
Allied conquests, to the extent that there were Allied conquests after 
1928, among the reversals.  
This proposition is difficult to test because, for the most part, it was 
the Axis powers, not the Allies, who carried out conquests between 
1928 and the end of the war. But there are several exceptions. H&S 
acknowledge that China’s claim to Taiwan in 1945 was not recognized 
as legitimate but remained in place after the war—at least until 1949, 
when “the fleeing nationalist army declared Taiwan an independent 
state.”99 In addition, as H&S note, “[i]n June 1940, the Molotov-
Ribbentrop Pact between Nazi Germany and the Soviet Union awarded 
the Soviet Union control of Estonia, Lithuania, and Latvia.”100 
Although “many states, including the United States, refused to 
recognize the transfer,”101 the transfer was not reversed as part of the 
territorial settlements in the post-World War II treaties, and the Baltic 
states remained under Soviet control until 1991.102 Finally, it is 
noteworthy that France and Great Britain, although not the United 
States, recognized their World War I ally Italy’s conquest of Ethiopia 
in 1935.103 The seizure was only reversed after the conclusion of the 
war in which Italy had joined their Axis enemies.104 In all of these 
cases, non-Axis conquests of territory that were illegal under the 
Kellogg-Briand Pact were not reversed after the war, even as the Axis 
conquests were reversed. In fact, H&S provide no examples of illegal 
post-1928 non-Axis conquests being reversed after the war. 
Second, we would be interested to see whether the Allies allowed 
themselves to take new territories at the conclusion of the war, in 
violation of the Pact. Not every Allied state might want new territory. 
But to the extent that the Pact was causally significant, it would have 
presumably made postwar conquests less likely, and any states that did 
carry out or permit postwar conquests would have been more likely to 
grapple with their illegality under the Pact. In the event, France seized 
99 HATHAWAY & SHAPIRO, supra note 14, at 316. 
100 Id. at 318. 
101 Id. at 318–19. 
102 See id. at 319. 
103 Id.  
104 Id.  
552 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
a portion of Italy, including the towns of Tenda and Birga, as part of 
the 1947 Paris Treaty of Peace—a “minor adjustment of the border” in 
H&S’s words.105 Greece, Yugoslavia, and Albania also took parts of 
Italy.106 Most importantly, the Soviet Union obtained “significant 
territory” from a number of states including Germany, Poland, Japan, 
Romania, and Finland.107  
H&S emphasize that these postwar seizures “did not begin to 
approach those of previous wars.”108 But the relative lack of postwar 
territorial transfers to the Allies may not be evidence of the efficacy of 
the Pact. It could be attributed to the changing norms, especially among 
relatively liberal states such as the United States and the United 
Kingdom, that led those states to sign the Pact in the first place,109 rather 
than to a sense of legal obligation to the Pact. More significant is the 
fact that H&S provide no evidence of postwar officials struggling with 
the illegality under the Pact of the Allied seizures of Italian territory, or 
of the Soviet Union’s multiple conquests. This is hardly surprising. 
After the war, the Allies were developing a new international legal 
order to replace the failed order of the League of Nations and the 
Kellogg-Briand Pact. It would be unusual if concerns about strict 
adherence to the failed, soon-to-be-replaced interwar order had played 
a significant role in their postwar decision-making.110 
More generally, there is a notable silence in The Internationalists 
concerning the role of the Kellogg-Briand Pact in the decision-making 
processes of the postwar officials who were responsible for establishing 
the territorial lines that H&S treat as evidence of the efficacy of the 
Pact.111 This silence is especially noteworthy considering the extensive 
historical sleuthing that seems to have gone into the book, as reflected, 
for example, in its discovery of Shotwell’s role in the drafting of the 
105 Id. at 321–22. 
106 See id. at 321. 
107 See id. at 322. 
108 Id. at 321. 
109 See supra Part II.A. 
110 It is true that the Kellogg-Briand Pact is still in effect. See Jack Goldsmith & Daryl 
Levinson, Law for States: International Law, Constitutional Law, Public Law, 122 HARV. 
L. REV. 1791, 103 (2009). But its prohibition on the use of war to resolve interstate disputes
plays no operative role in the contemporary international humanitarian law of jus ad bellum.
See, e.g., THE HANDBOOK OF INTERNATIONAL HUMANITARIAN LAW xxviii, 1, 759–70 (2d
ed. 2008) (noting that “[t]he modern ius ad bellum is of relatively recent origin and is based
upon Article 2 (4) and Chap. VII of the UN Charter,” and including no reference to the
Kellogg-Briand Pact in its Table of International Instruments or Index).
111 See Bass, supra note 89. 
2019] Law, War, and Four Modes of Conflict 553 
U.N. Charter, and as confirmed by the heroic labors described in the 
book’s acknowledgments.112 If there were evidence of the postwar 
planners and negotiators viewing the Pact, or even the year 1928, as a 
crucial dividing line between those conquests that were legal and 
should legally be upheld and those that were illegal and should legally 
be reversed, one suspects that H&S’s small army of researchers would 
have uncovered it, and H&S would have included it in the book. The 
fact that no such evidence appears strengthens the sense that any 
difference in the postwar treatment of conquests before and after the 
Pact may simply be a coincidence related to the timing of the Axis as 
opposed to the Allied powers’ conquests, rather than a sign of the Pact 
making a difference.113 
Third, we might also be interested to see how territories conquered 
by the Axis and Allied powers before 1928 were treated after World 
War II. If the Axis conquests were reversed, even though they were 
legal under the pre-Pact international law of the Old World Order, 
while the Allied conquests were not, this would be another piece of 
evidence weighing in favor of the conclusion that the postwar reversals 
of post-1928 conquests had little to do with the 1928 Pact, and were 
instead the result of victorious states acting on their interests as they 
112 See HATHAWAY & SHAPIRO, supra note 14, at 425–30. 
113 Indeed, it would be difficult to reconcile H&S’s suggestion of a deep Allied 
commitment to the underlying principles of the Pact with, for example, Churchill’s proposal 
to Stalin in 1944 that various Eastern European countries be divided by arbitrary percentages 
into British and Soviet spheres of influence. See Geoffrey Roberts, Stalin’s Wartime Vision 
of the Peace, 1939–1945, in STALIN AND EUROPE: IMITATION AND DOMINATION, 1928–
1953 250 (Timothy Snyder & Ray Brandon eds., 2014).  
It is noteworthy that when H&S intimate that World War II was fought not only against 
the Axis powers but against the Old World Order, their rhetorical evidence comes not from 
the Allies invoking the Pact, but rather by way of a quote from the Atlantic Charter. See 
HATHAWAY & SHAPIRO, supra note 14, at 213. As H&S acknowledge elsewhere, the 
Atlantic Charter makes no reference to the Pact. See id. at 330–31; see also id. at 191 
(summarizing drafting of Charter). Nor does the Pact state that the Allies intend to restore 
the territorial lines existing in 1928. See Atlantic Charter, U.S.-U.K., Aug. 14, 1941, 55 Stat. 
1603. If the Pact, or the year 1928, weighed heavily in the minds of wartime officials, one 
would have expected them to refer to it, at least privately. 
Similarly, despite H&S’s suggestion that the Pact’s causal significance to the New World 
Order is evident from “what those involved said,” the evidence that the officials responsible 
for creating the postwar order, including the postwar territorial settlements, acted as they 
did as a conscious result of the existence of the Pact is very thin. Id. at 330. Indeed, H&S 
themselves seem to recognize that the Pact was held in wide disrepute by the end of the war. 
See, e.g., HATHAWAY & SHAPIRO, supra note 14, at 188 (“Writing at the tail end of the 
Second World War, Welles would look back and—like many who didn’t yet recognize its 
revolutionary implications—blame the Pact for giving ‘the delusion of a great body of the 
American people that the mere formulation of a wish is equivalent to positive action.’”).  
554 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
would have regardless of the Pact. Testing this proposition is made 
difficult by the fact that the Axis powers had far fewer pre-1928 
colonial conquests than the Allies: “The Axis powers had largely 
missed out on the colonial land grab.”114 But not entirely. H&S note 
that as part of the postwar settlement, “Germany, Italy, and Japan did 
lose some territory they had held before 1928”115—including a portion 
of eastern Germany that the Allies returned to Poland, and the reversal 
of the Italian conquests of Libya and Somaliland and the Japanese 
conquest of Korea116—while the Allied victors, not surprisingly, did 
not give up any of their pre-1928 conquests.117 
H&S acknowledge the evidence in the preceding paragraphs. But 
they minimize its significance.118 They give the impression that the 
failure to reverse post-1928 Allied (but not Axis) conquests after World 
War II, the new Allied (but not Axis) territorial gains after the war, and 
the reversal of pre-1928 Axis (but not Allied) conquests after the war, 
are exceptions to the rule that the Pact was generally efficacious in the 
long run—examples of the unavoidable difference between the clear 
logic of legal ideals and the crooked timber of interstate politics.119 As 
they note regarding the efficacy of international law in general, the fact 
that a law is sometimes violated does not mean that the law is not 
effective, much less that it is not a law.120  
But it seems more accurate to view these cases not as exceptions to 
an otherwise well-established general rule, but as some of the only 
available evidence for evaluating H&S’s thesis that the Kellogg-Briand 
Pact was highly effective against the contrary thesis that it made little 
difference. H&S raise the question: “Why did most of the borders after 
the Second World War snap back to the lines that had existed when the 
Peace Pact was signed?”121 But what H&S attribute to the causal force 
of the Pact could be just as easily, and more consistently, explained as 
114 HATHAWAY & SHAPIRO, supra note 14, at 192. 
115 Id. at 321. 
116 See id. at 321–22 (also noting that Japan “withdrew its claims to a number of islands 
in the South China Sea”). 
117 It is irrelevant to the current argument that various Allied states began relinquishing 
pre-1928 colonies of their own accord soon after the war. See id. at 322. 
118 See, e.g., id. at 321 (noting that the amounts of pre-1928 conquest territory taken 
away from Axis powers after the war “did not begin to approach those of previous wars”). 
119 See, e.g., id. at 322 (“These concessions to Stalin were seen by the other Allied 
powers as regrettable deviations from accepted law, not precedents to be followed in the 
future.”). 
120 Id. at 418. 
121 Id. at 332. 
2019] Law, War, and Four Modes of Conflict 555 
the result of the fact that nearly all territorial conquests between 1928 
and the postwar settlement were by the Axis powers. Where the 
conquests were not Axis-led, they did not snap back. The lines also did 
not snap back to 1928 in several cases where the Axis powers had 
conquered territory before 1928.  
Did the Pact play a significant causal role in the decline of wars of 
conquest, as H&S suggest, or was it causally insignificant, as 
generations of historians and political scientists have assumed? In the 
few instances where H&S’s historical evidence of conquest allows us 
to test their hypothesis against the contrary hypothesis that states acted 
as they would have done regardless of the Pact, the evidence seems to 
support the latter. 
C. Legal-Historical Evidence for the Pact’s Efficacy
In addition to their cultural and statistical arguments for the Kellogg-
Briand Pact’s significance, H&S argue that the Pact was indirectly 
responsible for the transformation of international law that resulted in 
the “New World Order” of the postwar era: 
By prohibiting states from using war to resolve disputes, it began a 
cascade of events that . . . reshaped the world map, catalyzed the 
human rights revolution, enabled the use of economic sanctions as a 
tool of law enforcement, and ignited the explosion in the number of 
international organizations that regulate so many aspects of our daily 
lives.122 
The trouble with this causal argument is that it is not clear that the 
Pact was a necessary step at all. If anything, it appears to have been a 
false start. For the same reasons that we do not celebrate the Articles of 
Confederation as a necessary first step toward the creation of the United 
States Constitution, it would seem odd to celebrate the Pact as a 
necessary first step toward the creation of the U.N. Charter and the 
other legal institutions that actually constituted the postwar global legal 
order. The Pact was neither sufficient nor necessary to these later 
developments. It was, if anything, a mistake that the later legal order 
attempted to correct.  
The causal issue can be clarified with a counterfactual thought 
experiment.123 Imagine a world in which Levinson failed to persuade 
Kellogg to push for the Pact, the Pact was never signed, and war was 
122 Id. at xv. 
123 For the relevance of counterfactual thinking to historical explanation, see generally 
Cass R. Sunstein, Historical Explanations Always Involve Counterfactual History, 10 J. 
PHILO. HIST. 433 (2016). 
556 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
not outlawed in 1928. Assuming World War II would have proceeded 
largely as it did regardless of the existence of the Pact—an assumption 
that H&S provide no reason to question—is there any reason to believe 
that the U.N. Charter and the other postwar legal institutions would 
have been materially different? It is true that Shotwell’s first draft of 
the U.N. Charter incorporated the language of the Pact.124 But the rest 
of the State Department team that produced the final draft of the 
Charter rejected the inclusion of this language,125 and also rejected the 
proposal to impose automatic membership in the new organization on 
all states that were parties to the Pact.126 Was the Pact, and the contested 
legal developments that followed it in the interwar years, necessary as 
an inspiration for the underlying ideas in the Charter? 
It might be plausible to argue that the Pact played a significant causal 
role in creating the postwar legal order if it were assumed that the 
drafters of the U.N. Charter would not, or could not, have arrived at the 
rules of jus ad bellum contained in the Charter in the absence of the 
Pact. But there is no reason to believe the Pact, and the interwar legal 
developments that H&S persuasively argue it triggered, were necessary 
conditions to the final shape of the U.N. Charter, or even that they made 
its final shape significantly more likely. The underlying goal of the 
Charter—using an international legal agreement to promote peace and 
security among states—long predated the Pact and did not require the 
Pact as inspiration.127 Based on this goal, it was practically inevitable 
that the Charter would contain some system for the legal regulation of 
war, as the League of Nations had before it. Indeed, against H&S’s 
claim that the United Nations “would be a new and improved Pact,” it 
seems far more appropriate to think of the United Nations as a new and 
improved League—a new international organization to replace the 
failed old one, not merely a new peace treaty.128 If the Pact had never 
existed, it seems more likely than not that the U.N. Charter would have 
been no different. Even the failure of the Pact was arguably an 
124 See HATHAWAY & SHAPIRO, supra note 14, at 195. 
125 See id. at 194–97. 
126 See id. at 197. In other words, H&S present the presence of the Pact’s language in 
Shotwell’s first draft of what would become the U.N. Charter as further evidence of the 
historical significance of the Pact. But the evidence could also be understood as support for 
the opposite conclusion: that the Pact had failed to make a difference, which was why 
Shotwell was attempting to repackage its language in a more effective agreement. 
127 See, e.g., KANT, supra note 87, at 311. 
128 On the United Nations as a replacement for the League of Nations, see generally 
MAZOWER, supra note 32, at 191–213 (“The League Is Dead, Long Live the United 
Nations”). 
2019] Law, War, and Four Modes of Conflict 557 
unnecessary lesson, because the failure of the League would have been 
sufficient to suggest the potential value of a prohibition on the 
unauthorized use of force (article 2(4)) backed by the threat of 
sanctions (article 41) and ultimately the threat of force (article 42).129  
To maintain that the Pact was a necessary or even a causally 
significant step (that is, a step making a later event more likely, even if 
the step was not necessary to the occurrence of the later event) toward 
arriving at the postwar international order, it might be argued that the 
practical logic of the U.N. Charter only appears as an obvious option 
to us because we live in the world created by the Pact, a world in which 
the assumptions of the Old World Order, such as the right of conquest 
and the unfeasibility of sanctions, have been swept away. But such an 
argument seems to rest on implausible assumptions regarding the 
practical difficulty of imagining alternatives to the Grotian rules. One 
need only look at the debates that H&S describe leading up to the Pact 
in order to appreciate that it was entirely possible to think through 
U.N.-like practical alternatives to the Old World Order without needing
the experience of the Pact and its consequences in the interwar years.
Indeed, while Levinson may have held back from supporting sanctions
against aggressive states in part based on the Grotian assumption that
such sanctions would constitute acts of war,130 Shotwell experienced
no such qualms and advocated for the Permanent Court of International
Justice to “hear disputes over aggression” and issue “financial
sanctions.”131 In a counterfactual history with no Pact, we can imagine
Shotwell bringing these ideas with him to the State Department
subcommittee on international organizations, from which they could
easily have ended up shaping the U.N. Charter, just as the article 2(4)
outlawing of aggressive war could easily have been drafted without the
Pact having ever existed. Imagining practical alternatives to the legal
arrangements of the Old World Order hardly requires a Copernican
revolution.
As to the distinct question of whether the United Nations and the 
other international legal institutions of the postwar world have been 
causally significant in shaping state and non-state behavior—a position 
that might be defended even if the efficacy of the Pact were rejected—
H&S adopt a characteristically maximalist-legalist position. They 
attribute several features of the postwar Long Peace to the powerful 
129 See U.N. Charter arts. 2(4), 41, 42. 
130 See HATHAWAY & SHAPIRO, supra note 14, at 114. 
131 Id. at 117–18. 
558 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
influence of international law. Why have nuclear weapons never been 
used as threats to achieve territorial aggrandizement? “The rule against 
conquest prohibits it.”132 Against, or at least beside, the view that the 
spread of international markets played a central role in causing the 
Long Peace,133 H&S suggest “the rise of global free trade was at least 
as much a consequence as a cause of the outlawry of war.”134 (One 
might wonder how to explain the rise of global free trade during the 
“first globalization” before World War I, when war had not yet been 
legally outlawed.135) Why, beginning in the 1940s, did states stop 
growing as they generally had since the early modern period, and 
instead start to split apart into smaller units? “The answer, again, starts 
with the outlawry of war.”136 Why is there still so much conflict, despite 
132 Id. at 332. 
133 See, e.g., PINKER, supra note 35, at 284–88. H&S deserve credit for drawing attention 
to a point that is often neglected in the liberal case that trade can have a pacifying effect. 
Against the Leninist idea that capitalism leads naturally to colonialist exploitation and war 
as states pursue greater profits than their domestic markets can provide, liberalism argues 
that trade between states can increase the costs of interstate war, to the extent that war 
disrupts the ordinary functioning of business; and that trade can decrease the benefits of 
such wars, in the sense that trade allows states to gain access to the resources and markets 
of other states without having to resort to force. Compare VLADIMIR LENIN, IMPERIALISM, 
THE HIGHEST STAGE OF CAPITALISM 91 (Resistance Books 1999) (1917) (presenting 
imperialism “as the development and direct continuation of the fundamental characteristics 
of capitalism in general”); WENDY BROWN, UNDOING THE DEMOS 47, 50–51, 142–46 
(2015) (noting that “critical intellectuals” in the late 1970s and early 1980s saw 
“neoliberalism” as “something that resecured the South as a source of cheap resources, 
labor, and production in the aftermath of colonialism”); see, e.g., SHEEHAN, supra note 87, 
at xvii (presenting the liberal argument that post-1945 industrialization in Europe “expanded 
the connections among peoples and nations, weaving a web of interdependent relationships 
that required and sustained peaceful exchange.”). H&S add to the standard liberal argument 
by emphasizing that interstate trade can also contribute to peace by laying the groundwork 
for outcasting, “a legal tool for disciplining states who violate the rules.” HATHAWAY & 
SHAPIRO, supra note 14, at 420. Once a state is dependent on access to international markets, 
other states can discipline it for violations of unrelated international legal rules by depriving 
it of access to those markets. 
134 HATHAWAY & SHAPIRO, supra note 14, at 333. But see HATHAWAY & SHAPIRO, 
supra note 14, at 344 (acknowledging that technological change and institutions of private 
international law likely played a role in the rise of global free trade as well). 
135 Although the 6.2% average annual growth in global trade between 1950 and 2007 is 
higher than the 3.8% growth rate between 1850 and 1913, the earlier period represented a 
significant departure from prior history. See WORLD TRADE ORGANIZATION, WORLD 
TRADE REPORT 2008: TRADE IN A GLOBALIZING WORLD 15 (2008). H&S concede that the 
transformation of international law in the wake of the Pact “was not the only reason global 
free trade boomed” between 1945 and the present, but offer no explanation for how it was 
possible for free trade to boom in the “Old World Order” prior to World War I. See 
HATHAWAY & SHAPIRO, supra note 14, at 344. 
136 Id. at 338. 
2019] Law, War, and Four Modes of Conflict 559 
the outlawing of war? “The answer is that these conflicts are not 
prohibited by the Pact.”137 In each of these cases, alternative 
explanations could be offered that do not depend on the efficacy of 
international law or the significance of the Pact. For example, a liberal 
international relations theory explanation for the United States having 
not used the threat of nuclear annihilation to conquer nonnuclear states 
may be that doing so would conflict with the basic normative 
preferences of American voters and policymakers. An interest-based 
explanation might involve the high reputational costs of such a gesture, 
or the likelihood that it would encourage other states to enter alliances 
in opposition to United States power.  
Neither of these plausible, and in fact compatible, explanations 
necessarily depend on the fact that the New World Order prohibits 
conquering states from obtaining the legal right to the resources of 
conquered territories. H&S emphasize this fact as a deeply significant, 
often overlooked source of changed incentives since World War II that 
has powerfully altered state behavior, along with the threat of 
outcasting.138 A skeptic might question whether H&S exaggerate the 
importance of legal title. Even without legal title to conquered territory, 
a conquering state still possesses the conquered resources and can 
profit from them through internal use or trade on the black market. 
II 
WAR AND LAW AS MODES OF CONFLICT RESOLUTION 
Even if The Internationalists is unpersuasive regarding the 
significance of the Kellogg-Briand Pact, it remains a rich resource for 
thinking about international law and the possible shapes of the 
relationship between law and war. As already noted, H&S echo their 
colleague James Whitman by emphasizing how European states of an 
earlier era perceived and used war as “a legitimate means of righting 
wrongs.”139 Against the idea of war as an eruption of lawless rage, H&S 
draw attention to an earlier tradition, stretching back to Cicero, passing 
137 Id. at 352. It would be tempting, but unfair, to say: “What a shame that no one thought 
to ban unjust intrastate conflict in the Pact as well!” Even H&S do not suggest that 
international law is capable of achieving every goal, although they believe it is capable of 
many things. 
138 See, e.g., id. at 315–16; see also Oona Hathaway & Scott J. Shapiro, What Realists 
Don’t Understand About Law, FOREIGN POL’Y, Oct. 9, 2017 (arguing that “the interests of 
powerful states changed around 1928 . . . because the law had changed”). 
139 HATHAWAY & SHAPIRO, supra note 14, at xv. 
560 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
through Aquinas, and reflected in Grotius and his successors, that 
viewed “war as a substitute for courts.”140  
H&S emphasize that after the transformation from the Old World 
Order to the New World Order, states are no longer allowed to use war 
to settle disputes. But in the absence of a centralized international court 
with mandatory jurisdiction and an effective means of enforcing its 
judgments, how are interstate disputes meant to be resolved? 
The Internationalists focuses on the practice of outcasting as one 
way in which states can enforce norms in the absence of a centralized 
global authority.141 H&S introduce the idea through a description of the 
medieval Icelandic assemblies known as “Things.”142 The most 
important Thing “met each spring . . . to hear lawsuits and resolve 
administrative issues.”143 Eventually, a national Thing began to meet, 
presided over by a Lawspeaker whose “main task was to recite the laws 
from memory.”144 Because Iceland had no public prosecutors, and in 
fact “no executive branch at all,” “[v]ictims seeking justice had to 
prosecute the accused in a Thing. If the victim was successful, the 
Thing would declare the defendant guilty and sentence him to one of 
several penalties,” the most severe of which was “full outlawry,” in 
which the outlaw could be “killed with impunity.”145 “Indeed, the 
prosecutor of the case was often . . . obliged to carry out the punishment 
himself, assuming the outlaw didn’t flee first.”146 Furthermore, 
“[a]nyone who helped an outlaw could be punished, even outlawed.”147 
140 Id. at 11 (“War is a substitute for courts, Grotius argued, because courts are the 
original substitutes for war”). 
141 As H&S note, outcasting has been “largely ignored by scholars.” Id. at 375. H&S 
introduced the concept of outcasting in an earlier work: Oona A. Hathaway & Scott J. 
Shapiro, Outcasting: Enforcement in Domestic and International Law, 121 YALE L.J. 252 
(2011). The general conceptual relation of H&S’s typology of outcasting in the earlier article 
to my typology of modes of conflict here might be summarized by noting that all of their 
examples of outcasting either take place in Type 2 or Type 3 institutional settings, with the 
division between Type 2 and 3 mirrored by their distinction between adjudicated and 
nonadjudicated outcasting. See generally id. 
142 See HATHAWAY & SHAPIRO, supra note 14, at 373–75. In light of the discussion of 
feuds later in this review, infra text accompanying n. 170–72, it might be worth noting, as 
Hathaway and Shapiro recognize, that feuding existed in medieval Iceland alongside the 
Thing-system. See Hathaway & Shapiro, supra note 141, at 289–90.  
143 HATHAWAY & SHAPIRO, supra note 14, at 373. 
144 Id. at 373. 
145 Id. at 374. 
146 Id. 
147 Id. 
2019] Law, War, and Four Modes of Conflict 561 
H&S propose that medieval Icelandic outlawry be seen as an 
example of the general phenomenon of outcasting, which “occurs when 
a group denies those who break its rules the benefits available to the 
rest of the group.”148 Significantly, outcasting does not require a 
centralized institution of law enforcement. H&S suggest that 
international law today, lacking any “world police” or “global courts 
with compulsory jurisdiction,” is often enforced through outcasting.149 
For example, under the rules of the World Trade Organization, “[i]f a 
state breaks the rules, another state can file a complaint and prosecute 
its case before a tribunal. If this tribunal rules in its favor, . . . the WTO 
authorizes the state that filed the complaint to break the rules in 
return. . . . The WTO is like a global Thing.”150 
H&S can be taken as suggesting that outcasting is a mechanism of 
law enforcement that is particularly suited to situations in which the 
parties to a conflict share a law and a judge but lack a shared executive. 
This way of presenting their claim raises the possibility of a more 
general model. Is there, for example, a mechanism of law enforcement 
that is particularly suited to disputes in which the parties share a law, 
but not a judge or executive? In fact, it is possible to distinguish four 
distinct modes of conflict, each with its own characteristic mechanisms 
of conflict resolution. In practice, of course, actual historical conflicts 
will rarely fit perfectly within a single idealized mode, and there will 
be a great deal of variation within any mode along a number of 
variables. But we can start by specifying the four modes in abstract 
form. 
In what follows, my primary example of conflict will be war 
between modern states, but I will gesture toward the wider applicability 
of the modes of conflict. Looking beyond even H&S’s analogy to the 
premodern Icelandic state, I include passing references to 
anthropological studies of conflicts between non-state communities. 
These very brief illustrations are, of course, meant as nothing more than 
speculative provocations toward further thought. My use of historical 
and anthropological materials is in the spirit of a very preliminary and 
tentative gesture toward the kind of abstract structuralist models that 
appear in the writings of a historical sociologist like Charles Tilly.151 
148 Id. at 375. 
149 Id. 
150 Id. at 379. 
151 See, e.g., CHARLES TILLY, THE POLITICS OF COLLECTIVE VIOLENCE 15, Fig. 1.1 
(2003) (presenting a transhistorical two-axis typology of interpersonal violence). 
562 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
I will assume that to recognize a shared executive enforcer of the 
law ordinarily implies recognizing a shared judge to determine how the 
law is to be enforced, and that to recognize a shared judge ordinarily 
implies recognizing a shared law. Thus, I ignore the possibility of a 
conflict in which the parties perceive themselves as sharing, for 
example, a judge, but do not perceive themselves as sharing the law 
that the judge is supposed to apply.152 Also, to be clear, by treating the 
law, the judge, and the executive as disaggregable, I am not suggesting 
that they must be separate entities. It is of course possible for one entity 
to create, apply, and enforce the law, as with an idealized Leviathan. 
Finally, I alternate freely below between discussion of “laws” and 
“norms.” The extent to which a norm applicable to a conflict is law-
like—regardless of how the family resemblance concept of “law” is 
conceived—will certainly have significant and likely predictable 
effects on the shape of the conflict and how it might be resolved.153 But 
that is not my focus here. 
A. Type 1
In this mode of conflict, the parties share neither a law applicable to 
the conflict, nor a judge authorized to apply the law, nor an executive 
authorized to enforce the law. In other words, the conflict takes place 
in a state of pure anarchy between the parties, at least regarding the 
subject matter of the conflict.  
152 It might be objected that there is a common scenario in which two parties recognize 
themselves as sharing a judge with legal authority to resolve a dispute, but do not recognize 
themselves as sharing a law capable of resolving the dispute. In this scenario, the parties 
appear before a tribunal that they both believe has jurisdiction over the dispute, but they 
disagree entirely about which law governs the dispute. For my purposes, however, the 
agreement on jurisdictional law is a sufficient basis for saying that the parties share a law 
capable of resolving the dispute. They agree as a matter of law that they are bound by the 
judge’s decision, even if they disagree about the substance of the law to be applied. It is of 
course possible to devise counterexamples in which we might be tempted to say that the 
parties to a dispute shared a judge but not a law—for example, if the parties disagreed about 
which substantive law applied, and each recognized the judge’s jurisdiction only on 
condition that the judge applied that party’s preferred substantive law. But in general, it will 
be possible to recharacterize these scenarios either as situations in which the parties share a 
law in some sense, despite appearances to the contrary, or as situations in which the parties 
do not, in fact, recognize a shared judge, despite each side’s claim to do so. 
153 See, e.g., Kenneth W. Abbott, Robert O. Keohane, Andrew Moravcsik, Anne-Marie 
Slaughter & Duncan Snidal, The Concept of Legalization, 54 INT’L ORG. 401 (2000) 
(defining legalization in terms of three variables and considering the effects of an institution 
being located at different points along the three variables). Like H&S’s concept of 
outcasting, the authors note that their “conception of legalization . . . [moves] away from a 
narrow view of law as requiring enforcement by a coercive sovereign.” Id. at 403. 
2019] Law, War, and Four Modes of Conflict 563 
The primary alternative to force, whether we think of it as 
negotiation, diplomacy, persuasion, or reason, will tend to be effective 
as an alternative only if the parties share the kinds of norms that have 
been postulated as absent here. Negotiation usually depends, implicitly 
or explicitly, on an appeal to some shared standard. Thus, if the conflict 
is not resolved by one side giving in, and the parties are capable of 
using force, there will be a tendency for the conflict to escalate to the 
use of force. 
If a conflict in pure anarchy results in the use of force, the force will 
tend, at least in the long term, not to be constrained by laws or norms 
of good conduct. The latter laws or norms, such as jus in bello rules in 
warfare, impose a burden by preventing a party to a conflict from taking 
steps that the party might want to take. Even if a party has its own self-
restraining norms, there will often be practical and psychological 
pressures to set aside the norms in a conflict where the other side is 
perceived as not sharing them. The norms will often be seen as 
interfering with the pursuit of victory in the conflict, because the norms 
will limit the range of permissible action. In warfare, for example, 
complying with norms may be perceived as putting one’s own forces 
at greater risk of violence, and even as helping the enemy. It will often 
be easier to rationalize an exception to the norms than to take the steps 
necessary to enforce them. Why should an enemy that refuses to play 
by the rules receive the benefit of the rules? 
Historical examples that approximate the purely anarchic mode of 
conflict are not difficult to find. In hunter-gatherer societies, tribes 
often invaded neighboring territory in raids. Unlike the ritualized, 
highly symbolic, low-fatality battles sometimes associated with 
societies without states,154 these raids appear to have been rationally 
calculated to achieve a maximum amount of killing and destruction 
with minimum risk. The military historian Azar Gat summarizes the 
apparently universal pattern of “primitive warfare”: “face-to-face 
confrontations were usually mostly demonstrative and low in 
casualties, but a great deal of killing was done by surprise, mostly 
during unilateral actions.”155 Further, “the most lethal and common 
form of warfare was the raid.”156 The most killings took place in large-
scale raids in which “the camp of the attacked party could be 
154 See KEEGAN, supra note 5, at 29 (describing ritualized Zulu battle); PINKER, supra 
note 35, at 43.  
155 AZAR GAT, WAR IN HUMAN CIVILIZATION 116, 118 (2006). 
156 Id. at 117. 
564 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
surrounded, and its unprepared, often sleeping, dwellers massacred 
indiscriminately (except for women who could be abducted).”157 In 
fact, the practice of quasi-genocidal raiding may predate the existence 
of anatomically modern human beings. It has been observed among 
chimpanzees, some of our species’ genetically closest living 
relatives.158 
It is also not difficult to find examples of unconstrained, potentially 
genocidal warfare between human groups in more recent history. In 
medieval Europe, writings on war explicitly distinguished between 
rule-bound bellum hostile, “the norm within Western Christendom,” 
and a different mode of warfare for use “against outsiders, infidels, or 
barbarians.”159 This latter kind of warfare was “a brutal legacy from the 
Romans which they termed bellum romanum, or guerre mortelle, a 
conflict in which no holds were barred and all those designated as 
enemy, whether bearing arms or not, could be indiscriminately 
slaughtered.”160 
Especially clear modern European examples of such conflicts can be 
found in the early modern wars of the English against the Irish,161 many 
of the wars of colonial Americans against the indigenous peoples of 
America,162 many of the wars of European colonialists against non-
white, non-Christian indigenous peoples in general,163 and the Nazi war 
against the Soviet Union.164 
157 Id. 
158 See PINKER, supra note 35, at 37–38; GAT, supra note 155, at 8–9. Not all raiding, it 
should be emphasized, was large scale. See GAT, supra note 155, at 117. Raids were also 
conducted in the course of feuds, as discussed below. See id. at 116–17; see also infra text 
accompanying n. 170–72 (feuding as Type 2 conflict).  
159 Michael Howard, Constraints on Warfare, in THE LAWS OF WAR: CONSTRAINTS ON 
WARFARE IN THE WESTERN WORLD 1, 3 (Michael Howard, George J. Andreopoulos & 
Mark R. Shulman eds., 1994). 
160 Id. 
161 See Harold E. Selesky, Colonial America, in THE LAWS OF WAR: CONSTRAINTS ON 
WARFARE IN THE WESTERN WORLD 59, 59–61 (Michael Howard, George J. Andreopoulos 
& Mark R. Shulman eds., 1994); Geoffrey Parker, Early Modern Europe, in THE LAWS OF 
WAR: CONSTRAINTS ON WARFARE IN THE WESTERN WORLD 40, 56 (Michael Howard, 
George J. Andreopoulos & Mark R. Shulman eds., 1994). 
162 See Selesky, supra note 161, at 62–73. 
163 See Parker, supra note 161, at 56–57. 
164 See, e.g., id.; TIMOTHY SNYDER, BLOODLANDS: EUROPE BETWEEN HITLER AND 
STALIN 166, 175 (2010). By contrast, the Nazi and Soviet campaigns of extermination 
against “kulaks,” Jews, and other groups living on territory under their control during the 
1930s and 1940s fit the model of a Type 4 conflict, as described below, better than they fit 
the Type 1 model. Generally, these campaigns were state atrocities against individuals who 
recognized themselves to be under the state’s authority rather than acts of warfare against a 
2019] Law, War, and Four Modes of Conflict 565 
B. Type 2
In this mode of conflict, the parties recognize themselves as 
governed by a shared law that at least one side believes is applicable to 
the conflict, but they do not fall under the compulsory jurisdiction of a 
shared judge capable of settling how (or if) the law in fact applies, nor 
do they perceive themselves as falling under a shared executive with 
authority to enforce the law. A shared law could be as simple as a 
bilateral agreement, or as wide-ranging as a shared religious tradition 
with binding moral codes.  
In the binary terms of realist international relations theory, parties 
that share a law but lack a shared judge or enforcer of the law remain 
in a state of anarchy.165 But this binary model obscures the potential 
difference that a shared law, even standing alone, can make to the shape 
of a conflict. A shared law can serve as a focus of coordination or 
cooperation, allowing the parties to resolve an actual or potential 
conflict through negotiated agreement.166 The shared law can also be 
enforced through self-help processes. When one side perceives the 
other side as violating the shared law, the former may punish the 
violation through reprisals and other tit-for-tat mechanisms.167 
Of course, in the absence of a shared judge, there will always be a 
risk that the parties may disagree over whether a violation of the shared 
law took place, and this disagreement may lead to conflict. As Hobbes 
recognized more than four centuries ago, in a state of anarchy “when 
there is a controversy . . . the parties must of their own accord, set up 
for right Reason, the Reason of some Arbitrator, or Judge, to whose 
sentence they will both stand, or their controversie must either come to 
blowes, or be undecided.”168 But if the rules are clear enough,169 history 
group engaged in armed conflict against the state based on a rejection of the state’s authority. 
When viewed in legal-institutional context, the campaigns resemble a genocidal policing 
more than war. 
165 See, e.g., KENNETH N. WALTZ, THEORY OF INTERNATIONAL POLITICS 79–106 
(1979) (defining the relation between states as an anarchic “state of war”). For an analysis 
of the binary opposition between anarchy and the state in Waltz, see Milner, supra note 21. 
166 See GOLDSMITH & POSNER, supra note 43, at 12; see generally ROBERT O. 
KEOHANE, AFTER HEGEMONY: COOPERATION AND DISCORD IN THE WORLD POLITICAL 
ECONOMY 31–46 (1984). 
167 See HATHAWAY & SHAPIRO, supra note 14, at 77–80 (describing the role of reprisals 
in the early modern enforcement of the laws of war). 
168 HOBBES, supra note 2, at 32–33; see also id. at 90, 96–101. 
169 Indeed, one of the advantages of creating the legal institution of a shared judge is that 
the existence of the judge facilitates the use of more complex, flexible, or finely targeted 
rules. 
566 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
shows that some degree of law-based coordination is possible even in 
the absence of a shared judge or enforcer.  
One of the classic examples of the Type 2 mode of conflict is the 
feud, an ancient and apparently universal pattern of violence.170 In the 
feud, the parties to the conflict recognize themselves as having different 
identities in some narrow, strong sense—for example, as belonging to 
different clans—while at the same time recognizing themselves as 
having a shared identity in a broader, weaker sense—for example, as 
belonging to the same religion, or descending from the same 
mythological ancestors.171 The latter sense of community will often 
supply norms that regulate the violence of the feud. The norms can both 
provide a basis for violence by establishing wrongs that can or must be 
avenged and constrain that violence by placing limits on acceptable 
conduct. Feuds do not typically devolve into no-holds-barred 
exterminatory warfare. At the same time, the judgeless structure of the 
feud may allow it to continue indefinitely without resolution. The 
conflict may only be resolved when the parties voluntarily turn to a 
neutral arbiter in the community, or the community places pressure on 
the parties to end the conflict.172 
Another ancient and apparently universal form of conflict made 
possible by the existence of shared norms is the ritualized, pitched 
170 One definition of feuding proposes five essential elements: 
(1) kinship groups are involved, (2) homicides take place, (3) the killings occur as
revenge for injustice (the terms duty, honor, righteous, and legitimate appear in
discussions of the motivation for the homicides), (4) three or more alternating killings
or acts of violence occur, and (5) the acts of violence and killing occur within a
political entity, such as a tribe, nation, or country.
Keith F. Otterbein, Feuding, in ENCYCLOPEDIA OF CRIME AND PUNISHMENT (David 
Levinson ed., 2002). See also JARED DIAMOND, GUNS, GERMS, AND STEEL 286 (1999) 
(“Each murder in band and tribal societies usually leads to an attempted revenge killing, 
starting one more unending cycle of murder and countermurder”); Edward L. Queen, 
Customs of Wrongs, in ENCYCLOPEDIA OF LAW & SOCIETY: AMERICAN AND GLOBAL 
PERSPECTIVES 1597 (David S. Clark ed., 2007) (noting “the blood feud has a near 
universality about it”). 
171 See Otterbein, supra note 170; accord Margaret Hasluck, The Albanian Blood Feud, 
in LAW AND WARFARE, supra note 1, at 381 (noting that the largely self-governing Albanian 
communities during World War I “consisted in the narrower sense of the family, and in the 
wider sense of the tribe”). 
172 This is not to say that the broader community surrounding a feud is necessarily a 
force for peace. The surrounding factions might want the parties to engage in feuding for 
their own reasons, even when the parties themselves would prefer not to do so. Cf. Oren 
Falk, Bystanders and Hearsayers First: Reassessing the Role of the Audience in Dueling, in 
A GREAT EFFUSION OF BLOOD?: INTERPRETING MEDIEVAL VIOLENCE 98, 108 (Mark D. 
Meyerson, Daniel Thiery & Oren Falk eds., 2004). 
2019] Law, War, and Four Modes of Conflict 567 
battle.173 Gat offers the following sketch of the battle among hunter-
gatherers: 
[T]he place and time . . . were normally agreed upon in advance. . . .
[T]he combatants hardly ever closed in on each other. The two
opposing dispersed lines stood at a spear-throwing distance, about 50
feet, hurling spears at one another while dodging the enemy’s spears.
In some cases, such battles were intended in advance to put an end to
a conflict and were thus truly ‘ceremonial’, with the spear throwing
restrained and mixed with ceremonial dances. Once blood was spilt,
or even before, the grievances were seen as settled, and the battle was
terminated.174
In order for such a ritualized form of battle to be carried out, the two 
sides must of course have at least a generally shared understanding of 
what the rules are. 
Contrary to the contemporary image of the battle as a locus of 
unrestrained brutality, Whitman emphasizes how pitched battles have 
often served to contain violence: 
[A] pitched battle . . . is a contained and economical way of resolving
a dispute between two warring groups or countries. . . . [T]he result
of fighting a pitched battle is to limit violence in the community at
large: if a conflict can be decided by a day of concentrated killing on
the battlefield, then violence can be prevented from spilling over to
the rest of society.175
In order for a battle to serve its role as a kind of “trial” or proceeding 
for settling a conflict, the parties must have a shared understanding 
beforehand of how the battle will serve that role. They must share 
certain norms regarding how the battle will proceed, how its outcome 
will be understood, and what will follow it. 
Just as The Internationalists notes structural similarities between the 
medieval Icelandic Thing and institutions of the New World Order such 
as the WTO, so an analogy might be drawn between the feud and 
European military conflicts from the rise of the modern state onward, 
at least until the establishment of the Concert of Europe in 1815.176 As 
173 See WHITMAN, supra note 3, at 6 (“In the societies of the past, the contained pitched 
battle was probably a universal human institution.”). 
174 GAT, supra note 155, at 117. As Gat also notes, these ritualized battles “could 
escalate into real battles, in the heat of conflict, by accident, or by treachery,” and sometimes 
“true battles were intended from the start.” Id.  
175 WHITMAN, supra note 3, at 3. 
176 For a comparison of the Concert of Europe to other international institutional orders, 
see MAZOWER, supra note 32, at 3–12; FINNEMORE, supra note 44, at 97–98. In the 4-type 
scheme of conflict I am sketching here, the Concert might be seen as lying ambiguously 
between a Type 2 and Type 3 institution. It possessed a shared law in the form of various 
568 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
in a feud, the incessant warring within Europe took place between 
entities that, in a strong but narrow sense, viewed themselves as having 
distinct and conflicting identities, but in a broader and looser sense 
viewed themselves as belonging together in Western Christendom. The 
Christian just war tradition, and the Grotian laws of war that eventually 
arose out of it,177 provided a set of shared norms that shaped and 
constrained conflict. The result was the pattern of endless but 
increasingly limited war that reached its purest expression in what 
military historians sometimes present as the eighteenth-century 
European “golden age of war . . . the last age of civilized battle warfare 
before the disasters of modern war took hold.”178 These wars were 
treaties and implicit agreements existing against the background of the laws of nations. 
Through the discussions of the major powers at its center, the Concert also offered a kind of 
shared judge. In the words of the Austrian official Friedrich von Gentz, the major powers of 
the Concert provided a “high tribunal . . . whose members guarantee to themselves and to 
all parties concerned, the peaceful enjoyment of their respective rights.” MAZOWER, supra 
note 32, at 4–5. Mazower suggests that the Concert system began to fall apart after the 
Crimean War, and then especially after the Franco-Prussian War of 1870–71. See id. at 11. 
By 1913, Europe could be seen as having reverted to a Type 2 system of shared law without 
any shared judge to adjudicate disputes, despite the rise of the international law movement 
and its dreams of peace through arbitration. See id. at 44 (noting the later nineteenth-century 
shift from Concert to bilateral negotiation in commercial matters); id. at 66 (describing the 
arbitration movement). 
177 See HATHAWAY & SHAPIRO, supra note 14, at 10; Howard, supra note 159, at 2. 
178 WHITMAN, supra note 3, at 12; see also MAZOWER, supra note 32, at 4 (describing 
modern European war before the Concert of Europe as “endless alliances of states and 
princes, constantly shifting in order to preserve or disrupt the prevailing balance of power”). 
But see BEST, supra note 2, at 21 (offering a reminder that eighteenth-century European 
warfare was hardly humane). 
Another, more recent example of feud-like warfare can be seen in the ongoing civil war 
in Afghanistan. As the journalist Dexter Filkins writes, the people of Afghanistan in 2001 
had been fighting for so long, twenty-three years then, that by the time the Americans 
arrived the Afghans had developed an elaborate set of rules designed to spare as many 
fighters as they could. . . . Men fought, men switched sides, men lined up and fought 
again. War in Afghanistan often seemed like a game of pickup basketball, a contest 
among friends. . . . Battles were often decided . . . not by actual fighting, but by 
flipping gangs of soldiers. . . . The fighting began when the bargaining stopped, and 
the bargaining went right up until the end. . . . Even the fighting, when there was 
fighting, had a desultory feel. 
DEXTER FILKINS, THE FOREVER WAR 50–52 (2008); cf. KARL MARLANTES, WHAT IT IS 
LIKE TO GO TO WAR 228–30 (2011) (describing how German and British tank commanders 
developed war constraining rules, including a daily ceasefire at 5:00 p.m., during World 
War II in North Africa). Once again, we see how war can serve as an alternative to rule-
based negotiation and can at the same time be constrained by its own set of rules. But the 
development of such rules may be more difficult where the opponent is perceived as 
belonging to a radically different group adhering to radically different moral or legal norms. 
Filkins notes that in contrast to the Afghans’ almost ritualized Type 2 armed conflicts, “[t]he 
2019] Law, War, and Four Modes of Conflict 569 
relatively rule-bound not only in their focus on battles as a potential 
mechanism of decision,179 but also through the gradual development of 
jus in bello regulations concerning, for example, the use of poison, 
treacherous killing, and the treatment of prisoners.180 
Yet the states lacked any shared judicial or executive authority for 
interpreting and enforcing these rules. As a result, H&S note, many of 
“the familiar laws of war arose from [a] tit-for-tat process.”181 States 
entered agreements, and then enforced the agreements through 
reprisals against violators. For example, the Dutch and the Spanish 
entered a treaty in 1599, in the midst of the Eighty Years War, to 
ransom soldiers of different ranks at specified prices rather than 
hanging them.182 “This cartel was reissued several times during the war 
and soon thereafter was copied by other European nations. . . . These 
agreements were enforced through reprisals: Breach of the cartel by 
one side led to breach by the other.”183 States used “self-help” to 
enforce agreed-upon laws in the absence of a shared judge or executive. 
More recently, the conflict between the United States and the Soviet 
Union in the Cold War could be seen as an instance of a Type 2 conflict. 
It is true that the conflict nominally took place within the jurisdiction 
of the U.N. Security Council, which has the authority to play the role 
of a kind of shared international “judge” under the U.N. Charter. But 
because both of the main parties in the Cold War conflict were 
permanent, veto-holding members on the Security Council, they were 
able to ensure that it never rendered a judgment against either of 
them.184 In the absence of an effective judicial entity with compulsory 
jurisdiction over them, the United States and the Soviet Union 
nevertheless entered a variety of formal and informal agreements, 
above all the tacitly developed agreement not to invade each other’s 
one group of people who really took fighting seriously were the foreigners—that is, the 
Americans and Al-Qaeda. They came to kill.” FILKINS, supra, at 53.  
179 See WHITMAN, supra note 3, at 20 (arguing that the law of war of pitched battles 
from the sixteenth through the early nineteenth century involved both sides agreeing “to be 
bound by the result . . . of a day of combat”). 
180 See HATHAWAY & SHAPIRO, supra note 14, at 72, 77–80. 
181 Id. at 78. 
182 Id.  
183 Id.  
184 In the early years of the Cold War, the Soviet Union was especially dependent on its 
veto, because the United States “could count on a majority.” DONALD KAGAN, ON THE 
ORIGINS OF WAR 441 (1995). 
570 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
core spheres of influence.185 The latter norm developed through a tit-
for-tat process during the first decades of the Cold War, as one would 
expect in a Type 2 conflict.186 
C. Type 3
In this mode of conflict, the parties agree that they share a law 
applicable to their dispute, and both recognize some shared judge with 
the authority to decide how the law applies,187 but they do not recognize 
any shared executive entity with the authority to enforce the law, 
including the judge’s decision in their case. Each party can attempt to 
enforce the law through self-help measures such as retaliation and 
reprisal, as in a Type 2 conflict, and the shared law itself may authorize 
such measures. But the existence of a shared judge increases the 
feasibility of an alternative, potentially more effective, efficient, and 
even just mode of enforcement: community punishment of 
violations.188 As H&S suggest, depriving violators of the law of the 
benefits of the community—“outcasting”—can be a particularly 
effective example of community sanction.189 
185 Arms treaties provide another example. See, e.g., WALTER ISAACSON & EVAN 
THOMAS, THE WISE MEN 630–33 (1986) (describing Averell Harriman’s negotiation of a 
nuclear test ban treaty with the Soviet Union). The Soviets tried to minimize the importance 
of a sticking point in the negotiations by noting that “[t]he Leninist view of treaties . . . 
always permitted them to be abrogated if they threatened the self-interest of the Soviet 
state.” Id. at 632. At first glance, this position might seem to call into question the purpose 
of entering any treaty. Yet from another perspective, the “Leninist view” might be said to 
reflect the reality of any agreement in a Type 2 legal-institutional setting. Each party is free 
to interpret its agreements however it wishes, or to abandon them, provided it is willing to 
accept the consequences from other parties. 
186 Many of the familiar episodes of the first years of the Cold War, such as the 
announcement of the Truman Doctrine, the American provision of military aid to Greece 
and Turkey, the Soviet rejection of the Marshall Plan, the American refusal to intervene in 
Czechoslovakia in 1948, the Berlin blockade and airlift, and the establishment of NATO 
and the Warsaw Pact, can be understood as steps in the formation of the shared United 
States-Soviet norm against certain interventions in each other’s spheres of influence. See 
KAGAN, supra note 184, at 437–47. Once the norm was in place, the United States and the 
USSR enforced it through actual or threatened tit-for-tat retaliation against violations, as 
illustrated most dangerously during the Cuban Missile Crisis. See id. at 447–548. In the 
Helsinki Accords of 1975, the norm of nonintervention became legally formalized. See 
TONY JUDT, POSTWAR 501 (2006).  
187 As the example of the medieval Icelandic Thing illustrates, the shared judge does not 
need to be an individual person. It can be a group, such as the lay judges we know as juries. 
188 Cf. Hathaway & Shapiro, supra note 141, at 324 (noting that “outcasting by third 
parties almost always requires adjudication”). 
189 See generally id. at 141. I note that sanctions by a community are analytically distinct 
from sanctions of a community. On the latter, see Daryl J. Levinson, Collective Sanctions, 
56 STAN. L. REV. 345, 348–49 (2003) (presenting the collective sanctioning of groups as a 
2019] Law, War, and Four Modes of Conflict 571 
Why does the existence of a shared judge open the door to new 
modes of law enforcement? Without a judge (whose authority is 
recognized by all sides in the conflict) any attempt by those who are 
not parties to the conflict to determine how the shared law applies, and 
then to carry out a remedy or punishment as appropriate, will run a very 
high risk of simply expanding the conflict rather than resolving it. A 
party to the conflict that perceives itself as disadvantaged by the non-
party’s decision or subsequent actions will likely view the non-party 
not as neutrally interpreting and enforcing the law, but as joining the 
conflict on the side of the disadvantaged party’s opponents.190 
Especially if the intervening non-party takes punitive action, the 
punished party may respond by retaliating. Rather than ending, the 
conflict may grow in size. In addition, a shared judge’s decision can 
help to coordinate community action by serving as a trigger for it, or 
even by explicitly ordering it. In the case of outcasting, for example, 
the judge’s decision imposes a duty on all members of the community 
to exclude the law violator. 
There are a variety of reasons why it might be preferable to enforce 
the law through a shared judge backed by community sanctions rather 
than through self-help measures. Above all, disputes will be less likely 
to fall into potentially endless, feud-like patterns. When two parties 
disagree about the application of the law, and there is no judge, any 
action taken by one party may be taken as a violation of the law by the 
opposing party, prompting retaliation, and the disagreement may never 
be resolved. Once a shared judge has made a decision, however, the 
costs of rejecting the decision will generally be greater. 
potentially useful, and in fact used, regulatory strategy in which the responsibility for 
deterrence is delegated to those who are well positioned to monitor and control potential 
wrongdoers). While sanctions by a group are difficult without something like a shared judge 
to coordinate group behavior, Levinson’s examples show that sanctions of a group by 
another entity do not depend on the sanctioning entity sharing a judge with the sanctioned 
group. A blood feud, for example, is a paradigmatic example both of punishing a group for 
the wrongs of individuals in the group, and, as argued above, of a Type 2 conflict in which 
there exists a shared law but no shared judge. See Levinson, supra note 189, at 352–53 
(describing blood feud as collective punishment); HATHAWAY & SHAPIRO, supra note 14, 
at 74–84 (feud as Type 2 conflict). 
190 The Grotian law of nations rested in part on the premise that “it was uncertain which 
side waged war rightfully, . . . as there was no common judge above the parties by whom 
this could be ascertained in terms of civil law.” HATHAWAY & SHAPIRO, supra note 14, at 
25 (quoting Raphael Fulgosius); see also infra note 210–16 (discussing how H&S’s logic 
connecting principles of the Old World Order only applies in the absence of a judge: “they 
might have presented a central aspect. . . .”). 
572 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
Community sanctions are especially superior to the use of reprisal 
as a form of self-help law enforcement. As H&S discuss,191 reprisal—
the deliberate violation of the law in retaliation for an opponent’s 
violation—generally involves behavior that the community believes 
should be prohibited. In warfare, this may involve inhumane conduct 
such as torture or the execution of prisoners. 
As suggested above, the medieval Icelandic Thing—a historical 
anomaly in the sense of being a state without a centralized executive—
and various international legal institutions of the postwar world are 
examples of Type 3 modes of conflict resolution. H&S’s use of the 
World Trade Organization as the central example of a Thing-like 
postwar institution, however, is problematic. In their discussion of 
outcasting, H&S suggest that the WTO “is like a global Thing.”192 “If 
a state breaks the rules, another state can file a complaint and prosecute 
its case before a tribunal. If this tribunal rules in its favor, . . . the WTO 
authorizes the state that filed the complaint to break the rules in 
return.”193 In other words, “the WTO agreements entitle the victorious 
party to suspend the benefits of membership in the community.”194 
But attention to the legal-institutional structures highlighted by the 
Type 4 model of conflict suggests an imperfection in the analogy 
between the Thing and the WTO, and in the notion of the WTO’s 
enforcement mechanism as an example of outcasting. When a WTO 
tribunal determines that a state has broken the rules, the tribunal does 
not authorize all members of the community to sanction the rule-
breaking state. The rule-breaking state is not outcast from the benefits 
of membership in the WTO community as a whole—even in some 
narrowly targeted way—but only from the benefits of free trade with 
the complaining state or states. The latter are authorized to retaliate 
against the rule-breaking state through the imposition of tariffs. Self-
help measures like retaliation, as we have seen, are the signature mode 
of law enforcement in a Type 2 conflict.195 To be fair, H&S’s 
191 HATHAWAY & SHAPIRO, supra note 14, at 79. 
192 Id. at 379. 
193 Id. 
194 Id. 
195 The fact that the designers of the WTO created a Type 3 institution that relies on self-
help rather than community enforcement also illustrates that the Four-Type Model 
represents a set of claims about institutional potentials and tendencies, not practical 
necessities. A shared judge makes collective enforcement of the law more feasible, but does 
not require it. Similarly, it is possible to imagine a law in a Type 2 setting requiring 
community enforcement of its violation, or a Type 4 institution delegating enforcement of 
the law to the community or to harmed individuals. These arrangements are less likely than 
2019] Law, War, and Four Modes of Conflict 573 
discussion of the WTO in their earlier article on outcasting recognizes 
precisely this distinction between the Thing and the WTO, noting that 
the former relies on third parties for enforcement, while the former 
permits enforcement only by the victimized first party.196 
A number of post-1945 international institutions fit the Thing-like 
Type 3 model better than the WTO. One example is the U.N. Security 
Council.197 Although it was originally envisioned that the Security 
Council would control a military force198 and thus arguably become 
the ones in the chart summarizing the Four-Type Model because, generally speaking, where 
policing is feasible, it is preferred to community responsibility for law enforcement, which 
is in turn preferred to self-help. But variations are possible. See infra text accompanying n. 
167.  
196 See Hathaway & Shapiro, supra note 141, at 327. It might be noted that some of 
H&S’s examples of “outcasting” strain the ordinary usage of the term. For example, they 
present bilateral retaliation, such as through customary countermeasures, as a form of 
“simple outcasting,” a form of law-enforcement that is “permissive, nonadjudicated, in-
kind, and proportional, and only first parties are included.” Id. at 326. But when one party 
retaliates in this way against another party, we would not ordinarily say that the other party 
has thereby been outcast from the community. 
197 An example from before 1945 would be the League of Nations, which had a Type 3 
structure, with the ability to render judgments and order member states to impose sanctions, 
but no armed force of its own to compel action by member states if they failed to act. See 
HATHAWAY & SHAPIRO, supra note 14, at 105. During the later nineteenth century and 
through the 1899 and 1907 Hague Conventions negotiations, there was a movement to 
promote peace through compulsory international arbitration, which would also have fit the 
Type 3 structure. See FINNEMORE, supra note 44, at 34–36; see also MAZOWER, supra note 
32, at 66. By contrast, voluntary arbitration, such as through the Permanent Court of 
Arbitration established by the 1899 Hague Conventions, is probably best seen as a Type 2 
rather than a Type 3 institution. It is functionally similar to an agreement between the parties 
to resolve a conflict through the verdict of battle or by flipping a coin. The parties may 
delegate a one-off decision to a judge, but the conflict does not take place in an institutional 
context in which a shared judge already exists. 
In practice, whether a judge has mandatory jurisdiction over a conflict, such that a party 
does not have the option of refusing to submit the conflict to the judge, may not always be 
clear. One of the earliest scenes of judging in Western literature reflects this ambiguity. The 
Shield of Achilles passage in Homer’s Iliad describes a quarrel “over the blood-price for a 
kinsman just murdered” that is resolved by submitting the dispute to the judgment of the 
city elders. HOMER, THE ILIAD 483–84 (Robert Fagles trans., 1990). But it is unclear 
whether either party could have refused to submit the conflict to the elders. For a 
contemporary, nonfictional scene reflecting the same ambiguity, see ANAND GOPAL, NO 
GOOD MEN AMONG THE LIVING 79 (2014) (describing tribal elders resolving a conflict in 
the mountains of Afghanistan). More generally, Gopal’s account of the changing forms of 
conflict resolution in the Pashtun countryside illustrates the practical logic and historical 
disruptions that could drive a community to submit to shared norms, then increasingly 
assertive shared judges, and ultimately a shared executive—whether a tribal strongman 
backed by a colonial power, or a militant movement of religious judges and students. See 
generally ANAND GOPAL, NO GOOD MEN AMONG THE LIVING 73–82 (2014). 
198 See U.N. Charter art. 43. 
574 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
something like a Type 4 global Leviathan—authorized to create, apply, 
and enforce law at least in some contexts—it has instead relied on U.N. 
member states to enforce its decisions.199 Just as the Thing makes a 
collective decision as to whether someone has violated the shared 
Icelandic law, and then obligates or permits members of the community 
to take various responsive actions,200 so the U.N. Security Council 
makes a collective decision that an entity has violated international law, 
and then obligates or permits member states to take various acts.201 
Other international institutions in the post-1945 world order could also 
be described as creating a shared judge and relying on community 
enforcement of the judge’s decisions, including the European Union.202 
The increasing use of community enforcement of international 
norms after World War II, and especially after the end of the Cold 
War,203 has gone hand-in-hand with the rise of multilateralism. As 
199 See U.N. Charter ch. VII; HATHAWAY & SHAPIRO, supra note 14, at 213, n.62. 
200 See HATHAWAY & SHAPIRO, supra note 14, at 373–74.  
201 The Security Council permits rather than obligates U.N. member states to take 
military action, but other responsive actions such as economic sanctions may be obligatory. 
To take a recent example, on December 22, 2017, the Security Council adopted Resolution 
2397, which declares that “all Member States shall prohibit” the transfer to North Korea of 
all crude oil, unless an exception is made. U.N. Security Council res. 2397 (2017) (emphasis 
added). 
202 E.U. states have generally complied with judgments of the European Court of Justice 
(ECJ). See Laurence R. Helfer & Anne-Marie Slaughter, Toward a Theory of Effective 
Supranational Adjudication, 107 YALE L.J. 273, 276, 290–98 (1997). But the 1992 Treaty 
of Maastricht also granted the ECJ the authority to impose fines on states that refused to 
comply. See Press Release, European Union, Financial Penalties for Member States Who 
Fail To Comply with Judgments of the European Court of Justice: European Commission 
Clarifies Rules (Dec. 14, 2005). If a member state persistently refused to comply with an 
ECJ judgment or pay a resulting fine, the ultimate mechanism for enforcing the law would 
be for the members of the European Council (excluding the accused state) to agree to 
suspend various rights of the violating member state—that is, to outcast the violator. See 
Consolidated Version of the Treaty on European Union art. 7, 2010 O.J. C 83/01; see also 
Maïa De La Baume, Brussels Puts Warsaw on Path to Sanctions over Rule of Law, 
POLITICO.EU, Dec. 21, 2017. Thus, ultimate responsibility for enforcing EU law rests in the 
hands of the community of member states. 
Interestingly, in the early 1950s, the Western European states came close to creating a 
European Defense Community (EDC) with a European army. See JUDT, supra note 186, at 
244–45. If the French National Assembly had not rejected the EDC treaty in 1954 (by the 
relatively close vote of 319 to 264) Western Europe might have become a Type 4 system 
with regard to internal security conflicts. In any case, the hegemonic security guarantee of 
the United States arguably created a similar result. See id. 
203 As G. John Ikenberry notes, “It is important to recall that the postwar liberal order 
was originally not a global order. It was built ‘inside’ one half of the bipolar Cold War 
system. It was part of a larger geopolitical project of waging a global Cold War.” Ikenberry, 
supra note 39, at 9. In terms of the Four-Type Model of conflict resolution, it would be at 
least as accurate to say that Type 3 institutions of the postwar order, such as the United 
2019] Law, War, and Four Modes of Conflict 575 
Martha Finnemore notes with regard to humanitarian intervention: “To 
be legitimate in contemporary politics, humanitarian intervention must 
be multilateral.”204 In contrast to the unilateral or strategic multilateral 
interventions of the nineteenth century, most of which “involved 
protection of Christians from the Ottoman Turks,”205 humanitarian 
interventions since 1945 have consistently attempted to present 
themselves as multilateral actions “in defense of ‘generalized 
principles’ of international responsibility.”206 The United States sought 
U.N. authorization and international participation in its humanitarian 
interventions in Somalia and Haiti, for example, even though it did not 
actually need “the involvement of other states for military or strategic 
reasons.”207  
The growing international expectation that all legitimate 
humanitarian interventions must be multilateral fits comfortably with 
the growing use of Type 3 institutions for the enforcement of 
international law. As states have become accustomed to the idea that it 
is sometimes the responsibility of the international community to 
enforce judgments under international law, rather than just the 
responsibility of a harmed party, it may have come to seem natural that 
international norms should be upheld by multilateral state coalitions, 
rather than unilaterally, even when there has been no decision by a 
shared judge. The agreement of multiple states may even function as 
an imperfect proxy for the decision of such a judge. 
D. Type 4
In this mode of conflict, the parties share an applicable law, a judge 
with recognized authority to apply the law to their dispute, and an 
executive with recognized authority to enforce the law. The 
paradigmatic example of a Type 4 institution of conflict resolution 
would of course be the centralized government of a state. Whether the 
parties to a conflict actually engage with the machinery of the state 
Nations, operated against the background of the Type 2 conflict between the United States 
and the Soviet Union, as to say that the latter operated against the background of the former. 
Finnemore phrases the characteristic “modes of diplomacy” of the Cold War era in a 
summary chart using the following terms: “Varied; use of multilateral negotiations becomes 
common but core security concerns handled bilaterally.” FINNEMORE, supra note 44, at 97. 
204 FINNEMORE, supra note 44, at 78. On the significance of multilateralism generally, 
see John Gerard Ruggie, Multilateralism: The Anatomy of an Institution, 46 INT’L ORG. 561 
(1992). 
205 FINNEMORE, supra note 44, at 58. 
206 Id. at 81. 
207 Id. 
576 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
through litigation, or merely negotiate in the shadow of the law, their 
conflict will be shaped by the law of the state. 
Just as a typical method of law enforcement in a Type 2 conflict is 
retaliation, and a typical method in a Type 3 conflict is community 
sanctioning, a typical method in a Type 4 conflict is policing: the 
enforcement of the law by agents of the executive power that both 
parties to the conflict recognize as having authority over them. In an 
institutional setting where the parties to the conflict recognize a shared 
law enforcer, the law enforcer will naturally be a likely mechanism for 
enforcing the law. 
But not all Type 4 institutions are states, and not all conflicts within 
states fit the Type 4 mode. We might consider a religious institution 
with the authority to create law, render judgments, and enforce those 
judgments, such as through exclusion;208 or a family structure in which 
a matriarch or patriarch lays down the rules, determines when one has 
been broken, and may administer a punishment; or a criminal 
organization whose chief plays the role of lawgiver, judge, and 
executioner; or, perhaps, the International Criminal Court, which 
applies an international law that is presumed to be universally shared, 
and has its own security staff and detention center for carrying out 
sentences. Conflicts take place across a variety of institutional contexts 
in which the parties share a law, judge, and enforcer of the law that are 
not the state. 
There are many potential advantages to a Type 4 institutional setting 
over the three types above. Just as centralizing the application of the 
law in a shared judicial authority may both increase the feasibility of 
administering complex laws and facilitate community coordination, so 
centralizing the enforcement of the law in a shared executive authority 
may increase the feasibility of administering complex or otherwise 
difficult remedies or penalties and free up the community from having 
to engage in law enforcement. A range of just resolutions that are 
simply impractical in a Type 3 setting may become feasible in a Type 
4 setting. 
208 Hathaway and Shapiro present the Catholic Church’s use of excommunication to 
enforce canon law as an example of outcasting, emphasizing the ways in which the Church 
externalized the enforcement of canon law to non-Church actors. See Hathaway & Shapiro, 
supra note 141, at 290–99. However, where a religious institution enforces its rules 
internally, through the actions of its own agents, as in the case of “minor” excommunication, 
the conflict between the punisher and the punished might be seen as belonging to a Type 4 
institutional setting. See id. at 296. 
2019] Law, War, and Four Modes of Conflict 577 
Of course, the creation of a centralized law enforcement authority 
can also open the door to tyrannical forms of violence and oppression 
that would not be feasible in a Type 1 through Type 3 setting. Only in 
a Type 4 setting does it become possible for a conflict to develop not 
between two parties under a shared law, but between a party and the 
centralized agent of the law, such as the police. As the history of the 
state from its earliest creation to the present illustrates, a law enforcer 
may be capable of as much violence against those under its authority 
as a military force is capable of violence against a foreign enemy. The 
Holocaust and many other genocides, reigns of terrors, mass 
enslavements, and other atrocities belong to the Type 4 mode of 
conflict between the state and those under its authority. 
Finally, it might be noted that civil wars and related phenomena such 
as insurgencies, counterinsurgencies, and revolutionary terrorism 
present a problem of categorization in the Four-Type Model of conflict. 
One side tends to view the conflict as Type 4, a violation of the 
legitimate laws of the state by those who are subject to the laws, while 
the other (revolutionary) side denies that the state is legitimate and has 
authority over them. What is at issue in the conflict is precisely whether 
the two sides exist under a shared state system of law, judges, and law 
enforcement, or whether they exist under separate state systems that are 
at war.  
It is likely that this legally undecidable structure contributes to the 
exceptional ferocity and lack of constraint in so many civil wars. The 
sides are at least implicitly battling over the right to claim a monopoly 
on force. Because each side views the other’s invocations of law as 
illegitimate and is unwilling to enter agreements under the other’s law, 
it is more difficult to develop agreements to limit the violence, as might 
be possible in a Type 2 or Type 3 setting based on a shared law. The 
fighting may thus devolve into Type 1-style, no-holds-barred warfare, 
even though such warfare is most typical among parties that view each 
other as entirely alien or inhuman. This should not be the case in a 
conflict where the parties, prior to the outbreak of war, lived side by 
side in an ostensibly shared political community. 
*** 
The preceding discussion of the four modes of conflict, and 
accompanying modes of conflict resolution, can be summarized in the 
following chart. In the chart, “S” stands for “recognized by the parties 
to the conflict as shared between them,” while “NS” stands for “not 
recognized as shared.” “LEG,” “JUD,” and “EXEC” stand, 
578 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
respectively, for a law capable of resolving the conflict, a judge with 
the authority to apply the law, and an executive with the authority to 
enforce it.209 
Table 1. Four Modes of Conflict and Conflict Resolution 
Characteristic Characteristic 
mechanisms of forms of 
Type LEG JUD EXEC law enforcement violent conflict 
1 NS NS NS N/A raid,  genocidal war 
self-help by 
2 S NS NS harmed party feud, 
(e.g., reprisal) pitched battle 
community 
3 S S NS responsibility multilateral 
(e.g., outcasting) intervention 
centralized 
4 S S S responsibility state violence 
(e.g., policing) 
While H&S’s discussion of outcasting arguably leads the way to this 
four-mode typology of conflict and conflict resolution, it is unclear 
whether they would agree with the validity or utility of thinking of 
conflict in these terms. In fact, if H&S had agreed that the transition 
from the Grotian legal order to the post-1945 international law of jus 
ad bellum, centered around the United Nations, reflected a shift from a 
Type 2 mode of military conflict without a shared judge toward a Type 
3 mode with a shared judge, they might have presented a central aspect 
of their historical analysis differently.  
209 The tripartite division of legal institutions into law, judge, and executive obviously 
echoes the discussion of separated government powers in MONTESQUIEU, THE SPIRIT OF 
LAWS 154–56 (Cambridge 1989) (1748), although the echo is not intentional or intended to 
signify any conceptual connection between the two very distinct discussions. Montesquieu, 
in any case, did not invent the tripartite division of government powers. See M.J.C. Vile, 
CONSTITUTIONALISM AND THE SEPARATION OF POWERS 24, 31–32, 36 (1998) (noting that 
the distinction between the legislative, judicial, and executive powers of government 
appears first during the English Civil War, and is prefigured as early as the writings of 
Aristotle). 
2019] Law, War, and Four Modes of Conflict 579 
Throughout Part I of The Internationalists, the section of the book 
dealing with the Old World Order, H&S argue that there is a kind of 
practical and legal logic connecting on the one hand, “the privilege to 
use force to enforce their rights,” and, on the other hand, four subsidiary 
legal rules of the Old World Order, including the rights of conquest 
(that is, the right to gain legal rights through conquest, regardless of the 
actual justice of the conquest); the military’s “license to kill” without 
criminal penalty, and regardless of whether the initiation of the war was 
just; the principle that neutrality implies strict impartiality between 
warring parties, regardless of whether the initiator of the war acted 
justly; and the validity of “gunboat diplomacy” (threats of interstate 
force used to coerce agreement from weaker states).210 “Once states 
have a privilege to use force to enforce their rights,” H&S conclude, 
these subsidiary legal rules “inevitably follow[].”211 That is, “[o]nce 
war was legal, . . . there was no alternative to a world where Might 
made Right.”212 
Interpreted charitably, H&S’s argument about the “logical”213 
relationship between the privilege to use force and the subsidiary legal 
rules should be understood as an argument about practical tendencies 
rather than a legal formalist claim about deductive necessities. The 
specific arguments they present suggest that the adoption of contrary 
subsidiary rules would result in such serious practical problems that the 
subsidiary rules would prove unworkable—not that allowing neutral 
states to use economic sanctions, for example, would necessarily be 
logically inconsistent with granting states the privilege to use force.214  
In essence, H&S argue that any legal distinctions that require a 
neutral determination of whether a war was just are impracticable in 
the world of independent European states existing during the Old 
World Order: 
Though he insisted that war could only be waged for justice, Grotius 
also understood that in an international order of sovereign states, 
there are many versions of justice. The stability of legal rights upon 
which global commerce and international cooperation depends 
210 See, e.g., HATHAWAY & SHAPIRO, supra note 14, at 97 (presenting “[t]he logical 
structure” of the Old World Order in chart form). 
211 Id. 
212 Id. 
213 Id. 
214 In fact, these two rules reflect the legal arrangement that Genêt apparently assumed 
to exist when he invited the United States to provide France with what H&S describe as 
“assistance just short of war” in France’s conflict against the British. See id. at 84. 
580 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
would be undermined by the multiplicity of opinions over the justice 
of particular conflicts. The only possible way to allow victims to 
vindicate their rights through war is to allow nonvictims to gain legal 
rights from war, too. When Might is Right, multiple views on justice 
no longer matter, and all those who are engaged in war, victims and 
nonvictims alike, have a license to kill.215 
As in Hobbes’s vision of the anarchic state of nature, so in Grotius’s 
world of sovereign states, each side in a conflict judges its own 
conduct. The practically inevitable result is that their judgments 
conflict even when they believe themselves to be bound by the same 
law—whether it is natural law, divinely revealed law, or the customary 
law of nations. 
Yet once we think in terms of the distinction between a Type 2 and 
Type 3 mode of conflict and conflict resolution, we might question 
whether the four subsidiary rules of the Old World Order in fact follow, 
even as a matter of practical necessity, from the premise of sovereign 
states with a privilege to use force to enforce their rights. A world of 
sovereign states could agree—of their own sovereign free wills—to 
having the justice of their wars decided by an international arbiter, a 
shared judicial entity. The right of conquest, for example, could be 
conditioned on the judicial body’s decision concerning the justice of 
the alleged wrong. If a state claimed that it was initiating a war based 
on a wrong that the judicial body determined did not exist or merit a 
war or did not justify the extent of the state’s conquest, the state could 
be denied legal title to some or all of its conquered territories. In theory, 
the state’s soldiers or officials might even be denied legal immunity for 
their killing. A shared judicial body could also be given the authority 
to determine whether a treaty was coerced through the threat of unjust 
force, thus undermining any practical need to permit gunboat 
diplomacy.216 
215 Id. at 96. 
216 As for “neutrality as impartiality,” it is unclear whether this rule is practically 
necessary even in the absence of a shared judge. Consider how states might have behaved if 
a consensus had developed that, contrary to Grotius, it was permissible under international 
law for a neutral state to provide “assistance short of war” to one side in a conflict—the 
Genêt rule. Sometimes, it is true, the disadvantaged state would have attacked the neutral 
state because of its assistance. This would have been an unjust basis for attack under the 
rules, because the attacked state was neutral. But the justice of an attack makes no legal 
difference under the rules: all interstate attacks, regardless of their justice, are treated alike, 
with the victor receiving legal title to the spoils. If it was not fatal to the Grotian system that 
states sometimes attacked one another without just cause, why would it have been fatal to 
revise the list of just causes so as to exclude “assistance short of war” or economic sanctions? 
In a world where “Might is Right,” the precise location of the legal line between just and 
2019] Law, War, and Four Modes of Conflict 581 
In other words, H&S’s discussion of the practical necessity tying 
together the rules of the Old World Order rests on an unstated premise: 
the states of the Old World Order, although they possess a shared law, 
lack a shared judge. The subsidiary rules of the Old World Order do 
not “inevitably follow[]” simply from the premise of sovereign states 
having “a privilege to use force to enforce their rights.”217 
It might be objected that the absence of a shared judge is implied by 
the premise of sovereignty. A state that has agreed to be legally bound 
by the decisions of an external judge regarding the justice of its wars, 
it might be argued, is no longer sovereign. Yet there is something 
arbitrary in defining sovereignty in this way. Why is a state that 
perceives itself as bound by international law, but only according to its 
own interpretation of that law, sovereign, while a state that exerts its 
sovereign will to agree to abide by the decisions of an outside 
adjudicator, not sovereign? Perhaps this objection will invite some to 
conclude that a truly sovereign state cannot be bound by international 
law at all. But if this is the case, then how can a sovereign state bind 
itself through constitutional law, or public law in general?218 
Indeed, the four-mode typology described above may offer a starting 
point for a general alternative to approaching the identity of the state in 
terms of sovereignty, just as it offers an alternative to approaching 
international relations in terms of a binary distinction between anarchy 
and the state.219 
CONCLUSION 
It is commonplace that the world faces a variety of challenges that 
would benefit from global coordination, from climate change, 
pollution, terrorism, and financial regulation, to the management of 
epidemics and the control of nuclear, biological, and chemical 
weapons. Viewing the menu of institutional arrangements for interstate 
coordination in terms of the Four-Type Model of conflict and conflict 
resolution, if nothing else, offers a reminder that there are choices other 
unjust reasons for interstate attack would seem to be a difference that makes no difference. 
See id. 
217 Id. at 97. 
218 Scholars have noted that constitutional law and international law provide similar 
challenges for certain conceptions of state sovereignty. See Jack Goldsmith & Daryl 
Levinson, Law for States: International Law, Constitutional Law, Public Law, 122 HARV. 
L. REV. 1791 (2009); Christopher A. Whytock, Thinking Beyond the Domestic-International
Divide: Toward a Unified Concept of Public Law, 36 GEO. J. INT’L L. 155 (2004).
219 See Milner, supra note 21. 
582 OREGON REVIEW OF INTERNATIONAL LAW [Vol. 20, 531 
than pure anarchy and the potential tyranny of a Leviathan-like global 
state. The most significant practical implication of the model may be 
one already suggested by the treatment of outcasting in The 
Internationalists: for the enforcement of the law, a state is not 
required.220 
220 This might seem to be an obvious point to some. But discussions of whether 
international law is truly “law” often neglect the point, even when it would be highly 
relevant. See, e.g., BEST, supra note 2, at 5–7 (against the critique that international law is 
not law because it is not enforced, arguing that states nevertheless have reasons to comply 
with it, and do comply with it—but not noting that mechanisms for enforcing international 
law do, in fact, already exist).