Correspondence: The Responsibility to Protect

Correspondence Gareth Evans and

Ramesh Thakur
Humanitarian Intervention and the Robert A. Pape
Responsibility to Protect

To the Editors (Gareth Evans and Ramesh Thakur write):

As cochair (Evans) and member (Thakur) of the International Commission on Interven-

tion and State Sovereignty (ICISS), and principal authors of its 2001 report The Responsi-
bility to Protect (R2P),1 we read Robert Pape’s article with great interest—but also with
growing surprise and ultimately considerable disappointment.2 Intervention can be stud-
ied as an analytical concept or as a political project, and Pape’s article clearly falls into the
latter category. His purpose is to advance his so-called pragmatic standard of humanitar-
ian intervention against the standard of the genocide convention (which, in his view, sets
the bar much too high) and R2P (which he thinks is loose and permissive, setting the bar
much too low). For an article proposing to advance humanitarian intervention as a politi-
cal project, however, it is remarkably disconnected from political reality.3
Pape completely overlooks the emergence of R2P over the last decade as the norma-
tive instrument of choice for converting shocked international conscience about mass
atrocity crimes into decisive collective action. His forty-page article devotes just two
pages to R2P, focusing entirely on its original artiulation in the ICISS report and totally
ignoring its subsequent intellectual and political evolution. “[S]ome policy advocates
and scholars,” he states, “have argued for the adoption of the ‘responsibility to protect’
standard” (pp. 50–51). Pardon? R2P was actually adopted by the United Nations
General Assembly, sitting as the 2005 World Summit, the largest gathering of the
world’s heads of state and government ever convened,4 and subsequently in multiple

Gareth Evans is Chancellor and Honorary Professorial Fellow at the Australian National University and
the author of The Responsibility to Protect: Ending Mass Atrocity Crimes Once and for All
(Brookings Institution Press, 2008). Ramesh Thakur is Professor at the Crawford School of Public Policy,
Director of the Centre for Nuclear Non-Proliferation and Disarmament at the Australian National Univer-
sity, and the author of The Responsibility to Protect: Norms, Laws, and the Use of Force in Interna-
tional Politics (Routledge, 2011) and People vs. the State: Reºections on UN Authority, US Power,
and the Responsibility to Protect (United Nations University Press, 2011).

Robert A. Pape is Professor of Political Science and Director of the Chicago Project on Security and Terror-
ism at the University of Chicago.

1. The International Commission on Intervention and State Sovereignty (ICISS), The Responsibility
to Protect (Ottawa: International Development Research Centre for ICISS, 2001).
2. Robert A. Pape, “When Duty Calls: A Pragmatic Standard of Humanitarian Intervention,” Inter-
national Security, Vol. 37, No. 1 (Summer 2012), pp. 41–80. Further references appear parentheti-
cally in the text.
3. We say this with Gareth Evans having spent twenty-one years in active national politics, eight
as Australia’s foreign minister and another nine leading a major international conºict-resolution
nongovernmental organization, the International Crisis Group, and with Ramesh Thakur having
spent nine years in a senior position in the United Nations system.
4. UN General Assembly, sixtieth session, World Summit Outcome, 2005, resolution adopted by the
General Assembly, October 24, 2005, pars. 138–139.

International Security, Vol. 37, No. 4 (Spring 2013), pp. 199–214

© 2013 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.

199
 International Security 37:4 200

resolutions of the Security Council.5 Despite this completely authoritative statement of

the principle (or “standard,” in Pape’s preferred terminology), he concludes, astonish-
ingly, that “the international community is unlikely to embrace the R2P movement”
(p. 52). The 2005 World Summit Outcome Document and its subsequent translation into
shared understandings in intergovernmental circles have simply been airbrushed from
history in Pape’s account.
We do not pretend that there is now anything close to unanimous consensus in the
international community as to how R2P should be applied in every case where mass
atrocity crimes are threatened or occurring, especially at what might be called the
“sharp end” of the R2P response spectrum, where a situation is, prima facie, so grave as
to compel consideration of not just lesser measures (e.g., diplomatic persuasion and
pressure, targeted sanctions, or the threat of International Criminal Court prosecution),
but the extreme option of coercive external military force. There was such consensus
when the Security Council, speciªcally invoking R2P, authorized military action in
Libya in March 2011,6 but it fell apart later in the year as the “BRICS” countries charged
that the NATO-led forces had exceeded their civilian protection mandate.7 This in turn
has contributed signiªcantly to the paralysis of the Security Council in the face of
the even more grievous situation that subsequently unfolded in Syria. It will clearly
take time for trust to be restored between the major players, although—as will be
explained—we are optimistic that it can be.
In what follows, we ªrst outline the evolution of R2P since 2001, totally neglected in
Pape’s analysis. Second, we spell out ªve objections to the analysis he does offer—that
it resurrects unacceptably divisive “humanitarian intervention” discourse; opens the
door to unilateral interventions; ignores prevention and rebuilding responsibilities;
wholly overstates the permissive scope of R2P; and exaggerates the obligations it cre-
ates. Finally, we discuss where R2P stands in the wake of Libya and Syria.

the evolution of r2p since 2001

In current international policy discourse on the question of mass atrocity crimes, it is
the multidimensional and nuanced concept of R2P—not the older one-dimensional mil-
itary concept of humanitarian intervention—that dominates real-world debate. There
have been a number of crucial way stations in the evolution of the concept from its
original formulation by our ICISS commission—all ignored by Pape—starting with the

5. UN Security Council Resolution, No. 1674 “[r]eafªrms the provisions of paragraphs 138 and 139
of the 2005 World Summit Outcome Document regarding the responsibility to protect populations
from genocide, war crimes, ethnic cleansing and crimes against humanity” (emphasis in original).
UN Security Council Resolution, No. 1674, April 28, 2006. See also UN Security Council Resolu-
tion, No. 1970, February 26, 2011; UN Security Council Resolution, No. 1973, March 17, 2011, on
Libya; UN Security Council Resolution, No. 1975, March 30, 2011, on Côte d’Ivoire; UN Security
Council Resolution, No. 2014, October 21, 2011, on Yemen; UN Security Council Resolution,
No. 1996, July 8, 2011, on South Sudan; and UN Security Council, statement by the president of the
Security Council, September 22, 2011.
6. UN Security Council Resolution, No. 1973.
7. The BRICS countries are Brazil, Russia, India, China, and South Africa—all of which were
members of the Security Council during 2011. On the BRICS as a grouping in world affairs, see
Andrew F. Cooper and Ramesh Thakur, “The BRICS in the New Global Economic Geography,” in
Thomas G. Weiss and Rorden Wilkinson, eds., International Organizations and Global Governance
(London: Routledge, forthcoming).
 Correspondence: The Responsibility to Protect 201

important reports of the UN Secretary-General’s High-Level Panel (disclosure: Evans

was a member) and Secretary-General Koª Annan himself articulating more precisely
the obligations involved prior to the 2005 summit.8 Then, importantly, the “World Sum-
mit Outcome Document” itself narrowed the focus from the ICISS benchmark (rightly
criticized by Pape as too broad) of “population suffering serious harm,” so that only
“four crimes” could trigger R2P—“genocide, war crimes, ethnic cleansing, and crimes
against humanity.”
Further evolution occurred with the secretary-general’s report to the General Assem-
bly in 2009, which helpfully characterized R2P responsibilities in terms of “three pil-
lars”: ªrst, the responsibility of each sovereign state itself to protect its own populations
from the atrocity crimes in question; second, the responsibility of other states to assist it
in doing so; and third, the responsibility of the wider international community to re-
spond in a “timely and decisive” fashion and by all appropriate means (not excluding
coercive military action), in accordance with the UN Charter, if this becomes necessary
because the state in question is “manifestly failing” to protect its people. It has become
obvious in successive annual General Assembly debates—in 2009, 2010, 2011, and 2012
(even with the latter two occurring in the context of signiªcant dissatisfaction with the
way R2P had been applied in the later stages of the Libya operation)—that this frame of
reference is now overwhelmingly accepted.9 As Secretary-General Ban Ki-moon put it
in September 2011, “[O]ur debates are now about how, not whether, to implement the
Responsibility to Protect. No government questions the principle.”10
The rapid acceptance of R2P in international political settings has been accompanied
by an exuberant intellectual debate—again almost wholly ignored by Pape—about its
scope, limits, and mode of implementation, led not only by those associated with ICISS
(in addition to us, most notably Thomas Weiss, past president of the International
Studies Association), but by serious scholars in serious journals and by public intellec-
tuals in the opinion pages of print and online commentary. Edward Luck, previously at
Columbia University, for ªve years the special adviser to Secretary-General Ban on R2P,
and now back in academe as dean of the Kroc School at the University of California,
San Diego, has been particularly inºuential.11 There is a book series on R2P under the
Routledge imprint and a journal devoted exclusively to R2P, the Global Responsibility to

8. High-Level Panel on Threats, Challenges, and Change, A More Secure World: Our Shared Respon-
sibility (New York: United Nations, 2004), http://www.un.org/secureworld/; and UN General As-
sembly, ªfty-ninth session, In Larger Freedom: Towards Development, Security and Human Rights
for All, report of Secretary-General Koª A. Annan, March 21, 2005, http://www.un.org/
largerfreedom/.
9. See UN General Assembly, sixty-third session, Implementing the Responsibility to Protect, report of
the secretary-general, January 12, 2009; UN General Assembly, sixty-fourth session, Early Warning,
Assessment, and the Responsibility to Protect, report of the secretary-general, July 14, 2010; UN
General Assembly, sixty-ªfth session, The Role of Regional and Subregional Arrangements in Imple-
menting the Responsibility to Protect, report of the secretary-general, June 28, 2011; and UN General
Assembly, sixty-sixth session, Responsibility to Protect: Timely and Decisive Response, report of the
secretary-general, July 25, 2012. For descriptions and assessments of the annual General Assembly
debates around these reports, see http://www.globalr2p.org.
10. UN Secretary-General Ban Ki-moon, address to ministerial roundtable “Responsibility to Pro-
tect: Responding to the Imminent Threats of Mass Atrocities,” New York City, New York, Septem-
ber 23, 2011, http://www.un.org/News/Press/docs/2011/sgsm13838.doc.htm.
11. See, for example, Edward C. Luck, “The Responsibility to Protect: The First Decade,” Global Re-
sponsibility to Protect, Vol. 3, No. 4 (2011), pp. 387–399.
 International Security 37:4 202

Protect (GR2P). Many Ph.D. candidates around the world, in law as well as in political
science and international relations, are writing their dissertations on aspects of R2P—as
we well know because we are constantly being asked for advice and interviews. Addi-
tionally, a number of new and inºuential civil society organizations—for example, the
Global Centre for R2P, the International Coalition for the Responsibility to Protect, and
the Asia-Paciªc Centre for R2P—have contributed vigorously to the evolution of the
R2P norm. The explanation for the wide and continuing interest in R2P might well be a
broadly shared sentiment that it is one of the most signiªcant and consequential ad-
vances in the normative architecture of world order: indeed the British historian Sir
Martin Gilbert has described it as “the most signiªcant adjustment to national sover-
eignty in 360 years.”12

ªve problems with pape’s analysis

The consequence of Pape failing to take into account any of R2P’s political and intellec-
tual evolution since the ICISS report, especially since 2005, is that he takes the debate
straight back to the deeply divisive, problematical, costly (in blood and treasure), and
utterly ineffectual pre-2001 status quo ante. In this respect, we have ªve speciªc objec-
tions to his analysis.
First, Pape resurrects the language and discourse of “humanitarian intervention.”
ICISS was successful in repositioning the international consensus because we made the
core, sustaining idea not the “right to intervene” but the “responsibility to protect.” We
quickly discovered the visceral hostility across the developing world to any so-called
right of intervention, for any purpose, rooted in these countries’ experience of Western
missions civilisatrice in the era of colonialism. The developing countries remain deeply
suspicious of the self-serving hidden agenda of geopolitical and commercial interests
behind such claims. To dismiss their claims is to deny their history and disrespect their
collective memory. Unlike humanitarian intervention, R2P puts the needs and interests
of the victims of atrocities ahead of those of the intervening powers. It is victim and
people centered, whereas “humanitarian intervention” privileges the perspectives,
preferences, and priorities of the intervening states. Unlike humanitarian intervention,
which is only about military coercion, R2P embraces a whole spectrum of preventive
and reactive responses, with coercive military action reserved only for the most ex-
treme and exceptional cases.
Second, not only does Pape take his readers back to the rightly rejected and dis-
carded world of humanitarian intervention, but he would also take us back to the
unsustainable world of unilateral interventions. The task for ICISS was to address a
critical protection gap between complicity, paralysis, and illegality. If atrocities are be-
ing perpetrated and members of the international community have the capacity and
opportunity to stop them but choose to look the other way, they are part complicit even
though not the prime perpetrator in the atrocity crimes. To insist on absolute state sov-
ereignty and nonintervention in domestic affairs under the UN Charter regime, how-
ever, is to accept a paralysis of international action and to give tyrants the license to kill.
Yet, to undertake unilateral intervention—that is, one not authorized by the United
Nations—is to violate the existing body of international law that restricts the use of inter-

12. Martin Gilbert, “The Terrible 20th Century,” Globe and Mail (Toronto), January 31, 2007.
 Correspondence: The Responsibility to Protect 203

national force to defense against armed attack or when authorized by the United
Nations. R2P successfully ªnessed this protection gap.
Moreover, Pape ignores the reality that acting unilaterally adds to the transaction
costs of the enterprise, and that the exercise of national power is made more efªcient
and more effective when it is grounded in international legitimacy, with all the encour-
agement of reciprocally accommodating behavior this tends to promote. The principled
underpinnings for adopting an R2P rather than a humanitarian intervention perspec-
tive are reinforced by the reality of the gradual but steady shift of power and inºuence
from the West to the rest. The folly of disregarding the global South’s sensitivities and
preferences has only grown in the decade since R2P was ªrst articulated.
Of course, no major power will commit itself in advance either never to use force un-
less UN-authorized or always to use force when the UN Security Council so decides.
Nevertheless, it is very much in the U.S. interest, especially as its relative power and
inºuence begin to wane, to bind the rising powers to global norms and international
law on their international behavior. The United States cannot fashion a world in which
all others have to obey international law and norms, but Washington can opt out of any
of them, whenever and for however long it chooses.
Third, going backward on humanitarian intervention means a reluctance to embrace
the responsibilities to prevent and rebuild, which are core to R2P but not normally part
of humanitarian intervention discourse, as Pape’s contribution makes clear. If interven-
tions are embedded conceptually in the rights and privileges of the intervening actors,
then of course the fewer the constraints and obligations on them, the better. In that case,
however, they can hardly be called “humanitarian.” Conversely, if interventions are
genuinely motivated by humanitarian concerns as the primary goal (accepting, as Pape
does, that the real world is often characterized by mixed-motive situations), then their
implementation implies solidarity across borders. Such solidarity, however, cannot be-
gin and end with military intervention. It must also ªnd expression at the precrisis
point and be continued after the immediate crisis is over.
Pape, in fact, effectively concedes the existence of a responsibility to rebuild after a
military intervention for humanitarian purposes when he identiªes as one of the three
requirements of his pragmatic standard for such interventions “a workable strategy for
creating lasting local security.” We agree with his rationale: “so that saving lives in the
short term does not lead to open-ended chaos in which many more are killed in
the long term” (p. 43). As well as the conceptual incoherence of arguing otherwise, this
is practical common sense. Of course efforts should be made to help to build or rebuild
institutions and conditions that will prevent a relapse into the kind of murderous situa-
tion that required outside intervention in the ªrst place.
Fourth, Pape would take the normative architecture back to the pre-R2P status quo
on a false premise. The charge against R2P—that it is too permissive and would em-
broil the United States and the West in interventions without end all over the world—
is wrong in theory and demonstrably false in practice. R2P, as endorsed by world lead-
ers at the UN in 2005, would restrict military interventions to protect at-risk popula-
tions only in the context of the speciªed “four crimes” of genocide, war crimes, ethnic
cleansing, and crimes against humanity, and only when such interventions are autho-
rized under Chapters 7 or 8 of the UN Charter. By ignoring the 2005 outcome document
and the subsequent reports from the secretary-general and their reception in the
General Assembly, Pape is able to paint a false picture. Our 2001 report was an advo-
 International Security 37:4 204

cacy document, not the ªnal word, and it succeeded admirably in its objective of pro-
ducing an authoritative political response. That response, and the only authoritative
document for evaluating and judging R2P as the intervention standard, is the 2005 out-
come document. Ironically, the dominant scholarly criticism of the 2005 iteration was
that it was “R2P lite” and had set the bar too high.13
To be sure, many calls have been made for R2P to be invoked militarily in various sit-
uations. Even wrong-headed calls have had the unintended beneªt of clarifying both
the limits and the permissive circumstances of R2P, as argued by Cristina Badescu and
Thomas Weiss.14 This is analogous to imitation being the sincerest form of ºattery
(Myanmar), and perhaps also hypocrisy being the tribute that vice pays to virtue
(Russia in South Ossetia). To date, however, the reality is that Libya in 2011, and the
less-noticed Côte d’Ivoire resolution at the same time,15 are the sole examples of coer-
cive action being authorized by the United Nations under the rubric of R2P. Pape’s cat-
alogue in table 1 (p. 76), which tests the three alternative standards against actual crises,
is irrelevant for R2P. His last column—mislabeled, as is all too unhappily common,
“‘Right’ to Protect”—implies that R2P would require coercive military intervention in
every one of the cases listed, but that is to totally misunderstand the limits of the doc-
trine as it has always been formulated.
Pape claims incorrectly that “R2P sets the bar for intervention so low that virtu-
ally every instance of anarchy and tyranny—or indeed, every potential instance—
represents an opportunity for the international community to violate the sovereignty of
states” (p. 43). As noted, R2P as adopted by world leaders in 2005 restricts intervention
to the four speciªed crimes of genocide, war crimes, ethnic cleansing, and crimes
against humanity. In addition, the phrase “every potential instance” is also badly mis-
leading. These crimes must be either occurring or expected to occur. If a genocide is
about to break out—as indeed was reasonably anticipated in Rwanda in 1994—it does
not make sense to wait until it does break out before acting, especially if more lives can
be saved by acting preemptively. This is different, however, from acting on the suspi-
cion that it might possibly break out at some undeªned point in the future.
On the question of criteria for the use of force, we do not take issue with the central
theme of Pape’s article that, when it comes to military intervention for human protec-
tion purposes, the bar needs to be set high (though not so impossibly high as it tends to
be under the Genocide Convention). Our concern, rather, is to emphasize that all of the
necessary intellectual and policy apparatus to do just that is already there with R2P, and
that it is neither necessary nor helpful to reinvent the wheel for this purpose, particu-
larly when any embrace of “humanitarian intervention” language in any form, even
with the proposed new “pragmatic” modiªer, is bound to make international consen-
sus impossible.
The speciªc prudential criteria for the use of force that we would endorse are the ªve

13. See, for example, Alex J. Bellamy, Responsibility to Protect (Cambridge: Polity, 2009), p. 67; and
Thomas G. Weiss, Humanitarian Intervention: Ideas in Action, 2d ed. (Cambridge: Polity, 2011),
p. 127.
14. Cristina Badescu and Thomas G. Weiss, “Misrepresenting R2P and Advancing Norms: An Al-
ternative Spiral?” International Studies Perspectives, Vol. 11, No. 4 (November 2010), pp. 354–374.
15. UN Security Council Resolution, No. 1970, No. 1973, and No. 1975.
 Correspondence: The Responsibility to Protect 205

that have emerged from the recommendations of the ICISS, the High-Level Panel, and
Secretary-General Annan.16 First, seriousness of risk: Is the threatened harm of such a
kind and scale as to justify prima facie the use of force? Second, primary purpose of the
proposed military action: Is it to halt or avert the threat in question, whatever other sec-
ondary motives might be in play for different states? Third, last resort: Has every non-
military option been fully explored and the judgment reasonably made that nothing
less than military force could halt or avert the harm in question? Fourth, proportional-
ity: Are the scale, duration, and intensity of the proposed military action the minimum
necessary to meet the threat? Fifth, and usually the toughest legitimacy test, balance of
consequences: Will those at risk ultimately be better or worse off, and the scale of suf-
fering greater or less?
Without entering into a detailed discussion of Pape’s own criteria, on many smaller
points of which we would take issue had we the space to do so,17 it is evident that at
least in some major respects his concerns mirror ours. His requirement of an “ongoing
campaign of mass homicide” (p. 43) is one way of articulating the need for the serious-
ness of the risk to be very great, and “a viable plan for intervention” (p. 43) is a not very
different way of saying that an intervention must do more good than harm. We think it
was a mistake not to include a reference to criteria of legitimacy for the use of force in
the 2005 resolution; their adoption by the General Assembly, or even just as informal
guidelines by the Security Council, remains unªnished international business.
Fifth, we can only regard as an egregious straw man, built for the sole purpose of
knocking down, Pape’s assertion that R2P “would effectively obligate” states “to com-
mit vast resources to provide for the welfare of foreigners even if this came at the ex-
pense of obligations to their own citizens” (p. 52). To our knowledge, no advocate,
supporter, or sympathizer of R2P—and, before this, no critic of R2P—had made this
claim. It may be worth adding in this respect that R2P argues for a political responsibil-
ity on the part of the international community to help populations at risk of atrocity
and creates no new legal obligation whatsoever.

r2p after libya and syria

Where does R2P stand now, after the controversies surrounding Libya and Syria? Ap-
plying R2P principles, we agree with Pape (in his case applying his own pragmatic hu-
manitarian intervention standard) that Libya in March 2011 was a textbook case for
coercive military intervention, and that tens of thousands of lives, in Benghazi and else-
where, were almost certainly saved by it. In a speech on January 18, 2012, to a confer-
ence to honor ICISS on the tenth anniversary of the R2P report, Secretary-General Ban
noted that, historically, the international community’s “chief failing” has not been too
much intervention, but rather “the reluctance to act in the face of serious threats”—the

16. See ICISS, The Responsibility to Protect, pp. 32–37, 74–75; High-Level Panel on Threats, Chal-
lenges, and Change, A More Secure World, p. 67; and UN General Assembly, In Larger Freedom,
p. 43. These recommendations differ a little in their language and presentation, but their core con-
cepts are the same.
17. For example, his “separable target population” and “lucrative targets for over-the-horizon
power” conditions for “low-cost intervention” (pp. 58–59) seem at odds with the realities on the
ground in Rwanda in 1994, when Tutsi and Hutu were completely intermixed throughout the
country.
 International Security 37:4 206

same argument as Pape’s regarding the genocide standard of intervention. In Ban’s

view, Libya in 2011 “demonstrated that human protection is a deªning purpose of the
United Nations.”18 Had the international community acted as quickly and robustly as it
did in the 1990s, the 8,000 men and boys murdered outside Srbrenica and most of the
800,000 men, women, and children hacked to death in Rwanda would be alive today.
We are not so sure, however, that the NATO-led operation in Libya remained a text-
book R2P case for its duration.19 If the objective genuinely was, and remained through-
out, “the protection of civilians and civilian populated areas” and not regime change as
such, why—at least after the initial defense of Benghazi—were cease-ªre offers that
may have been serious rejected outright without exploration? Why were ºeeing per-
sonnel posing no immediate risk to civilians, and locations of no obvious military sig-
niªcance, targeted? Why did the interveners break their own arms embargo in
supplying the rebels? The Western powers had answers to most if not all of these
questions—for example, in the argument that protecting civilians in areas such as
Tripoli that long remained under Muammar Gaddaª’s control could only in practice be
accomplished by overthrowing the regime. They did not satisfy the BRICS countries—
among others on and outside the Security Council. Moreover, the R2P consensus
underpinning Resolution 1973 fell apart over the course of 2011, damaged by gaps in
expectation, communication, and accountability between those who mandated the op-
eration and those who executed it.
An important result of these gaps was a split in the international response to the
worsening crisis in Syria. Draft Security Council resolutions introduced by Arab and
Western countries have been vetoed by China and Russia, still smarting from what they
perceived to be overreach in the implementation of the Libyan mandate. These two per-
manent members remained deªantly opposed to any resolution, even involving com-
pletely nonmilitary forms of pressure, arguing—with more political than intellectual
force—that this could set in train a sequence of events leading to a Resolution 1973–
type authorization for external military intervention in Syria.20
Although the backlash against the Libyan intervention decision is unquestionably
a setback for R2P, it does not sound its death-knell. As is clear from the UN General
Assembly debates already mentioned, there remains overwhelming support for the gen-

18. Ban Ki-moon, “Address to Stanley Foundation Conference on the Responsibility to Pro-
tect,” New York City, New York, January 18, 2012, http://www.un.org/apps/news/infocus/
sgspeeches/statments_full.asp?statID⫽1433.
19. See Gareth Evans, “Civilian Protection and the New Geopolitics of Intervention,” SIPRI Year-
book 2012 (Oxford: Oxford University Press for SIPRI, 2012), pp. 25–29; and Ramesh Thakur,
“Libya and the Responsibility to Protect,” Situation Report (Pretoria: Institute for Security Studies,
March 2, 2012).
20. See China’s ambassador to the United Kingdom Liu Xiaoming, “China Believes Syria Needs a
Peaceful Solution,” Guardian (London), February 10, 2012; Vladimir Isachenkov, “Moscow Support
for Assad Well-Calculated,” San Francisco Chronicle, February 5, 2012; Walter Russell Mead, “Rus-
sia’s Syrian Bet Explained,” American Interest, blog, February 5, 2012, http://blogs.the-american-
interest.com/wrm/2012/02/05/russias-syrian-bet-explained/; Dmitri Trenin, “Russia’s Line in
the Sand on Syria,” Foreign Affairs, February 5, 2012, http://carnegieendowment.org/2012/02/05/
russia-s-line-in-sand-on-syria/9g77; Yezid Sayigh, “China’s Position on Syria” (Washington, D.C.:
Carnegie Endowment for International Peace, February 8, 2012), http://www.carnegieendowment
.org/2012/02/08/china-s-position-on-syria/9iyy; and Fyodor Lukyanov, “Why Does Russia’s
Stance on Syria Bafºe Everyone?” Russia in Global Affairs, February 23, 2012, http://eng
.globalaffairs.ru/redcol/Why-Does-Russias-Stance-on-Syria-Bafºe-Everyone-15478.
 Correspondence: The Responsibility to Protect 207

eral principles of R2P, including all three of its pillars. There will always be disagreement,
sometimes intense, about what precise responses are appropriate to particular situations,
and those disagreements will almost invariably become more acute as the debate moves
to the sharp end of the response spectrum and the stakes increase. Only very rarely will
all of the stars align in favor of UN-supported military intervention, as in Libya.
The most encouraging feature of the present international political debate post-Libya
is the attention being devoted to a proposal made by Brazil, one of the ªercest global
South critics of the course of the Libyan intervention, aimed at ªnding a new basis for
consensus among Security Council members in responding to the most extreme mass
atrocity situations: what it calls “responsibility while protecting” or “RWP.”21 Designed
to supplement rather than supplant R2P, its two key elements are ªrst, for the Security
Council to embrace, formally or informally, an agreed set of criteria or guidelines (along
the lines of those noted above, including “last resort,” “proportionality,” and “balance
of consequences”) to help it reach consensus in any debate before an R2P military inter-
vention is authorized; and second, for the Council to accept some form of monitoring or
review mechanism to ensure that it has a reasonable chance to maintain that consensus
throughout the duration of an implementation operation.
In the end, decisions are made based on the particular exigencies of the day. Any de-
cision to intervene militarily will be contingent, made on a case-by-case basis rather
than following a prescriptive formula, although we strongly believe that the prudential
criteria for the use of force supported by most R2P advocates would genuinely assist
that process. Conversely, any authorization can be misused, with the likely result of
fostering perceptions that it is a convenient tool in the service of foreign imperialism.
In this sense, Pape is falling into the error of blaming the normative tool for the ills of
those using the tools to pursue their own agendas. That can happen to R2P. And it can
happen to any substitute.

conclusion
Both we and Pape want to achieve and maintain a genuine shared normative under-
standing of underlying principles, to maximize the prospects of support for coercive
military intervention when it is warranted and resistance to it when it is not, and above
all to save the maximum possible number of innocent lives at risk from mass atrocity
crimes. Where we disagree is in our ªrm conviction that, in the actual world of
policymakers and those who hope to inºuence them, the responsibility to protect is
now, and will remain, the only credible frame of reference. The proper course is not to
ignore, abandon, misrepresent, or circumvent the new R2P norm, but to consolidate
and strengthen it.
—Gareth Evans
Canberra, Australia
—Ramesh Thakur
Canberra, Australia

21. UN General Assembly, sixty-sixth session, “Responsibility while Protecting: Elements for the
Development and Promotion of a Concept,” November 11, 2011.
 International Security 37:4 208

Robert A. Pape Replies:

In my article “When Duty Calls,” I advance a new, pragmatic standard for humanitar-
ian intervention that speciªes when the United States and other members of the inter-
national community should intervene militarily to stop a government from harming its
own citizens.1 The pragmatic standard has three requirements for the use of force:
(1) an ongoing campaign of mass homicide sponsored by the government; (2) a viable
plan for international intervention with reasonable estimates of low casualties for the
intervening forces; and (3) a workable strategy for creating lasting local security for
the threatened population. In addition, I explain why adopting the pragmatic standard
would save more lives than the two prevailing alternatives: the so-called genocide stan-
dard (i.e., the moral imperative to intervene to prevent genocide), which sets the bar for
intervention too high, and the “responsibility to protect” (R2P), which sets the bar so
low that virtually every instance of anarchy or tyranny would create unbounded obli-
gations beyond the capacity of states to fulªll.
Gareth Evans and Ramesh Thakur, both long-standing proponents of the responsibil-
ity to protect standard, argue that my article pays insufªcient attention to R2P, ignores
the evolution of R2P after its founding document in 2001, and fails to appreciate that
R2P is now the “normative instrument of choice” guiding “collective action” by the in-
ternational community to stop governments from harming their citizens. They see no
need for a new standard because, in their words, there is already “rapid acceptance of
R2P in international political settings.”
Evans, Thakur, and I agree that the international community has a broad responsibil-
ity to protect innocents threatened by their governments. We disagree, however, that
such a broadly framed responsibility entails an obligation or a duty to intervene mili-
tarily. In a world where military intervention is increasingly considered a means to stop
ongoing campaigns of mass homicide, a narrow focus on when and how it should be
used is not only appropriate, but necessary. Further, the absence of discussion in my ar-
ticle of the “evolution” of R2P beyond the 2001 founding document was appropriate—
there are more statements and documents by R2P proponents (which I did examine),
but these do not change or clarify R2P’s standards for the use of force. Finally, Evans
and Thakur exaggerate the extent to which the international community has embraced
R2P, particularly with regard to armed intervention in humanitarian crises. Accord-
ingly, my response clariªes the objective for humanitarian intervention, the practical
implications of R2P and the status of R2P in the international community, and the spe-
ciªc objections raised by Evans and Thakur.

clarifying the objective

As I write in my article, proponents of humanitarian intervention have long argued that
the international community has a duty to save lives even when citizens are threatened
by their own governments. Since World War II, the practice of genocide has been the
most well-known standard—deªned by the 1948 Convention as acts “committed with
the intent to destroy, in whole or in part, a national ethnic, racial, or religious group.” In

1. Robert A. Pape, “When Duty Calls: A Pragmatic Standard of Humanitarian Intervention,” Inter-
national Security, Vol. 37, No. 1 (Summer 2012), pp. 41–80.
 Correspondence: The Responsibility to Protect 209

practice, however, the international community has done little to prevent genocide, ef-
fectively standing by in all major instances of state-sponsored efforts to destroy a nation
or an ethnic group from the Holocaust in the 1940s, to Biafra, Bangladesh, Ethiopia, and
Cambodia in the 1960s and 1970s, to Rwanda and the Sudan in recent years.
The main problem has not been a lack of consensus that genocide is a crime. Rather,
it has been the lack of agreement on a standard that would deªne (1) the threshold for
action with sufªcient clarity for a timely response—otherwise, by the time the inter-
national community decides to act, there may be few lives left to save; (2) the cost
threshold that states should accept in responding to the moral crime of mass atrocity—
without consideration of costs, intervention might violate the intervening govern-
ment’s existing responsibilities toward protecting its own citizens, creating a clash of
moral duties; and (3) the long-term obligations that interveners must adopt so that sav-
ing lives in the short term does not lead to ungovernable chaos and a resumption of
mass killing in the future.
Although acknowledging problems with the current genocide standard, R2P focuses
on the wrong objective. Evans and Thakur suggest that broad acceptance by the inter-
national community of the responsibility to protect is itself the key goal. Hence, they
count the number of statements by policymakers that use R2P-related phrases con-
demning mass atrocities and other related crimes as a measure of the power of their
ideas. International consensus on moral rhetoric, however, is the wrong measure of suc-
cess. There already exists a far stronger international consensus against genocide,
which has done little to compel states to act.
What is missing is international consensus on when to intervene militarily to save
lives. This consensus requires agreement on translating normative principles into prac-
tical action in real cases of intervention. My article both develops and clariªes those re-
quirements. It measures success not with regard to the number of times international
policymakers issue statements condemning mass atrocities and related crimes, but with
regard to the number of instances in which the international community has actually
used force to save lives put at risk by a local government—most notably over the past
twenty years, in northern Iraq, Bosnia, Kosovo, and now Libya.

r2p in practice
How much progress has R2P made in deªning the standards necessary to save lives?
Alas, hardly any. My article argues that R2P does not identify the number of atrocities
necessary to justify the use of force, fails to identify the level of costs that interven-
ers should accept or to offer any guidance on how the duty to save others should be
reconciled with prior moral obligations, and would obligate the international commu-
nity to engage in ambitious nation building with fuzzy criteria for starting or ending
these projects.
Evans and Thakur state that this critique is based on R2P’s 2001 founding docu-
ment—which is true—and that the subsequent evolution of R2P solves the problems I
identify—which it does not. To bolster their case, Evans and Thakur list ªve “pruden-
tial criteria for the use of force.” An examination of these criteria, however, shows that
R2P remains hopelessly vague and would effectively result in the same problems
that currently confront the genocide standard. In contrast, the pragmatic standard ad-
dresses all of these criteria in a manner that is signiªcantly more well deªned than the
nebulous genocide and R2P standards for the use of force.
 International Security 37:4 210

Evans and Thakur’s ªrst criterion is “seriousness of risk.” The authors ask, “Is the
threatened harm of such a kind and scale as to justify prima facie the use of force?” But
even if Evans and Thakur mean “the threatened harm of” genocide, war crimes, ethnic
cleansing, or crimes against humanity (see their text and n. 5), the question is still im-
practically opaque. What kind of war crime? What kind of ethnic cleansing? What de-
gree of group killing constitutes genocide? By contrast, the threshold for the pragmatic
standard would normally be crossed when the local government kills 2,000 or more in-
nocents within about a month.
Their second criterion is “primary purpose of the proposed military action: Is it to halt
or avert the threat in question, whatever other secondary motives might be in play for
different states?” Asking this question acknowledges that mixed motives may be in-
volved in intervention decisions, but it fails to establish what would indicate when moral
motives are urgent. The pragmatic standard stipulates that a mass homicide campaign
must have already started and be ongoing against an identiªable pool of victims, thus
stipulating observable conditions when moral motives are likely to be primary.
Third is “last resort: Has every nonmilitary option been fully explored and the judg-
ment reasonably made that nothing less than military force could halt or avert the harm
in question?” This question is as ill-deªned as Evans and Thakur’s ªrst question.
How long does the pre-intervention debate take? At what point have nonmilitary op-
tions been “fully explored?” Leaving the standard for when to intervene up to the polit-
ical judgments of interveners, rather than setting a ªrm standard of harm for when to
intervene, increases the chance that, by the time there is consensus that everything non-
military has been done, it will be too late.
Fourth, “proportionality: Are the scale, duration, and intensity of the proposed mili-
tary action the minimum necessary to meet the threat?” Of course, interveners should
not commit atrocities in the name of stopping them. Evans and Thakur, however, fail
to engage with the moral dilemma that is central in humanitarian intervention—
identifying the level of costs that interveners should accept so that the duty to save
others can be reconciled with prior moral obligations. This is a core purpose of the
pragmatic standard.
Finally, “[b]alance of consequences: Will those at risk ultimately be better or worse
off, and the scale of suffering greater or less?” Again, the question begs a host of imme-
diately apparent practical issues that can and should be tackled by the international
community prior to the onset of humanitarian crises. On what basis will interveners
judge the risk that mass killing will resume in the near or long term? Given R2P’s em-
phasis on nation building, how will it be possible to prevent interventions initially
driven by humanitarian concerns from evolving into greedy adventures to secure
geopolitical and commercial advantages? Such interventions might effectively become
occupations leading to resistance by local populations to interveners now perceived
as transforming the domestic institutions in ways that advance their own national in-
terests. The pragmatic standard addresses the need for lasting security for threatened
populations, but in a manner that emphasizes local self-determination over foreign-
imposed regime change or other projects for reconstructing the national political, social,
economic, and other institutions necessarily in line with the long-term preferences of
the interveners.
Given R2P’s open-ended criteria for the use of force, it is little wonder that it is not
being powerfully embraced by the international community.
 Correspondence: The Responsibility to Protect 211

Evans and Thakur overstate the international community’s commitment to R2P. At

issue is not occasional abstract rhetoric or the number of times that the United Nations
General Assembly has passed resolutions using the words “responsibility to protect.”
What matters is the threshold for military action—whether the international commu-
nity has regularly and consistently endorsed the standards for the use of force embod-
ied in R2P or regularly and consistently engaged in armed intervention to protect lives
on the basis of R2P standards. When viewed in this light, it is clear that, rhetoric not-
withstanding, the international community has not manifestly endorsed R2P’s stan-
dards for the use of force or acted largely in accord with those standards.
The high point of international support for R2P was the 2005 World Summit, a
plenary meeting of the UN General Assembly. Aside from the power to approve the
UN budget, the General Assembly is a deliberative body with no authority to enforce
the resolutions it passes. Assembly resolutions are statements of principle, not treaties,
and are therefore not binding in international law. Indeed, they are often passed by
“consensus” without a formal vote by the member states, making the commitment of
individual states foggy and difªcult to gauge.
The 2005 World Summit resolution cited by Evans and Thakur is a prime example. It
did not create any new obligations for member states, and it refers to the “responsibility
to protect” in only 2 of its 178 paragraphs. At most, these statements have the status of
moral suasion. Some may be persuaded, others not—but it is hard to tell, because the
resolution does not compel members to reveal actual preferences for future policy apart
from public rhetoric motivated by domestic or other political motives. Further, the
inºuence of moral suasion stemming from the 2005 World Summit document’s state-
ment on R2P is particularly weak, given that the document failed to adopt any criteria
for the use of military force—an omission that Evans himself calls “disappointing.”2
Finally, the status of even this weak degree of consensus for R2P in the international
community is questionable. In 2009 the president of the General Assembly issued a
“concept note” and organized a plenary debate underlining opposition to R2P. This led
to a resolution that pointedly failed to endorse concrete plans for implementation of
R2P, disappointing supporters.3 Chief among the concerns was that R2P set such a low
bar for intervention that it could easily be invoked as justiªcation for imperial or other
aggressive military adventures, as in fact happened with supporters of the invasion of
Iraq in 2003 and Russia’s war against Georgia in 2008.4
If rhetorical support for R2P is weak and inconsistent, the connection between the
principles of R2P and the behavior of the international community is virtually nonexis-

2. Gareth Evans, The Responsibility to Protect: Ending Mass Atrocities Once and For All (Washington,
D.C.: Brookings Institution Press, 2008), p. 48.
3. For assessments of the 2009 General Assembly debate, see Global Centre for the Responsibility
to Protect (GCR2P), The 2009 General Assembly Debate: An Assessment (New York: GCR2P, August
2009); International Coalition for the Responsibility to Protect (ICRtoP), Report on the General As-
sembly Plenary Debate on the Responsibility to Protect (New York: ICRtoP, September 15, 2009); and
“Responsibility to Protect: An Idea Whose Time Has Come—and Gone?” Economist, July 23, 2009,
http://www.economist.com/node/14087788.
4. Michael Ignatieff, “Why Are We in Iraq? (And Liberia? And Afghanistan?),” New York Times
Magazine, September 7, 2003; Michael Byers, “High Ground Lost on UN’s Responsibility to Pro-
tect,” Winnipeg Free Press, September 18, 2005; and interview with Minister of Foreign Affairs of
the Russian Federation Sergey Lavrov by the British Broadcasting Corporation, Moscow, August
9, 2008, http://www.un.int/russia/new/MainRoot/docs/warfare/statement090808en.htm.
 International Security 37:4 212

tent. First, intervention in past cases of atrocities considered ripe for intervention ac-
cording to R2P advocates has rarely occurred. When the international community has
intervened, it has done so because the case met the conditions of the new, pragmatic
standard advanced in my article. Alex Bellamy, a staunch advocate of R2P, catalogues
episodes of mass atrocities to clarify “R2P’s preventive agenda,” with a total of twenty-
one qualifying for intervention from 1990 to 2010. Of these, the international commu-
nity intervened in only the ªve cases that would qualify according to the pragmatic
standard, and it failed to act to any meaningful degree to prevent or stop atrocities in
the other sixteen cases.
Second, the connection between R2P and recent cases of mass atrocities in Libya and
Syria is even more problematic. As Evans and Thakur point out, supporters of R2P
were initially eager to claim the international intervention in Libya in March 2011 as an
endorsement of their principles, but they have since distanced themselves from the in-
tervention, even suggesting that the Libya operation is now better thought of as coun-
ter to the principles of R2P. At the same time, Evans and Thakur indicate that mass
atrocities in Syria do qualify for intervention according to R2P principles. In other
words, for proponents of R2P, the responsibility to protect has not only failed to pro-
duce appropriate intervention in both Libya and Syria, but its standards have opened
the door to action that supporters ultimately condemn.

responses to evans and thakur’s ªve objections

Evans and Thakur also make ªve speciªc objections to my article, which are best orga-
nized around two themes. The ªrst three constitute reasons why they believe the term
“responsibility to protect” is better than “humanitarian intervention.” Objections four
and ªve are defenses against my argument that the R2P standard is too permissive.
Evans and Thakur’s ªrst objection is that I use the “language and discourse of ‘hu-
manitarian intervention,’” which is politically less attractive than the “responsibility to
protect,” because the former encourages states to view moral international action as a
pretext for intervention policies based on geopolitical and commercial interests,
whereas the later does not. The term “humanitarian intervention,” however, has long
been used to refer to international action to save lives put at great risk by a local gov-
ernment and remains so to this day. Of course, states may well be suspicious that moral
justiªcations for intervention could serve as cover for selªsh motives, given that na-
tional interests often determine the policies of states. Changing the label is unlikely to
allay those suspicions, however. One of the main criticisms against R2P is that it sets the
threshold for action so low that it can easily justify imperial or other self-interested in-
tervention policies—a criticism strengthened in 2008, when Russia explicitly used the
language of R2P to formally justify its aggression against Georgia. For real progress to-
ward saving lives—and to allay suspicion that any humanitarian intervention is really a
cover for selªsh policies—a better alternative is to specify the requirements for humani-
tarian intervention in depth. This would greatly circumscribe the scope for misappro-
priating the standards for moral action as cover for interest-based policies, which is a
key purpose of my article.
Second, Evans and Thakur claim that “Pape . . . would also take us back to the unsus-
tainable world of unilateral interventions.” This is false. My article contains a section
entitled “An Organized International Coalition” in which I stress the value of multilat-
eral action. I emphasize this point in the conclusion where I state that “an established
 Correspondence: The Responsibility to Protect 213

organization such as the UN Security Council should authorize humanitarian interven-

tions in cooperation with states from the region of the proposed intervention [which]
would lower the risk that the new standard could become a pretext for the selªsh mo-
tives of one or a handful of states” (p. 80).
Third, they claim that focusing on “humanitarian intervention” means focusing in-
ternational solidarity on military intervention and that it implies a reluctance to em-
brace responsibilities to prevent and rebuild. This objection stands in opposition to
Evans and Thakur’s second point, given that both cannot be true. Further, Evans and
Thakur are right that my article focuses on the requirements for armed intervention to
save lives, but they are wrong that this would undermine international efforts to pre-
vent atrocities in the ªrst place or to create the conditions for lasting security for threat-
ened populations. Given the failures of the genocide standard, it is important to
develop new moral alternatives that would actually save lives put at risk by local gov-
ernments. The main centerpiece of any such policy must be the requirements for using
force to safeguard lives at imminent risk, which can then serve as a basis for identifying
the logically consistent and morally coherent policies for prevention and lasting stability.
This is the logic of the pillars of the pragmatic standard for humanitarian intervention.
Indeed, ignoring the need to clarify the standard for saving lives in immediate danger
would likely open the door to many competing interests and dynamics in decisions
about prevention and lasting stability, resulting in endless debate rather than action.
Fourth, Evans and Thakur contend that “Pape[’s] . . . charge against R2P—that it is
too permissive and would embroil the United States and the West in interventions
without end all over the world—is wrong.” Although R2P’s standards for military ac-
tion are ill deªned, there is good reason for thinking that these principles would justify
a vast expansion of international interventions. As mentioned, Alex Bellamy lists
twenty-one cases of mass atrocities and armed conºict from 1990 to 2010. R2P would
obligate intervention in all of them, although precisely how is unclear. Further, this list
focuses on cases where mass atrocities are already occurring, whereas Evans himself
would apply R2P’s principles more broadly to encompass “cases where the risk of mass
atrocity crimes is in the medium or longer term.” His criteria for identifying those cases
are (1) whether the country has a past history of mass atrocities; (2) whether “tensions
of a kind that have given rise to conºict in the past” still persist; (3) the “strength of the
countries’ coping mechanisms when it comes to resolving grievances and tensions”;
(4) the “receptivity of the country or society in question to external inºuence”; and
(5) whether the country has “good” leaders because “bad leaders . . . can make any
problem worse.”5 Given Evans and Thakur’s exceedingly abstract and poorly bounded
principles for identifying cases of future mass atrocities and Bellamy’s inclusion of vir-
tually all instances of armed conºicts around the world, one can reasonably wonder
whether there are any meaningful limits to R2P—and can see why few states would
want to commit themselves to such an open-ended policy.
Finally, Evans and Thakur argue that “Pape’s assertion that R2P ‘would effectively
obligate’ states ‘to commit vast resources to provide for the welfare of foreigners’”
is wrong, and they contend that R2P “creates no new legal obligations whatsoever.”
Evans and Thakur are right that R2P currently creates no new legal obligation for the

5. Evans, The Responsibility to Protect, pp. 73–75.

 International Security 37:4 214

international community, which contradicts their point that the international commu-
nity has embraced their principles. The key word, however, is “currently.” Given the
expansive nature of R2P’s call for armed intervention in virtually any case of anarchy
and tyranny in which there are actual or potential mass atrocities and nonmilitary op-
tions have been exhausted, adoption of R2P by the international community would ob-
ligate states to provide vast military, economic, and other resources. Further, R2P lacks
a serious discussion about the limits of those new obligations, making the doctrine both
politically unworkable and morally questionable. States are unlikely to sign a blank
check with the potential for undermining their existing political and moral obligations.

conclusion
There is a tremendous need for new standards of international action that would effec-
tively translate moral duties into humanitarian realities—doing more to actually save
lives than merely talking about it. To move forward, it is important to have a healthy
debate on competing standards based on their merits. The pragmatic standard for hu-
manitarian intervention advanced in my article may have its limitations. Yet if adopted,
it would help to save more lives than either the existing genocide standard or the alter-
native of the responsibility to protect. It would also provide a sound moral basis for
new international treaties to stop mass homicide campaigns by local governments
against their citizens.
—Robert A. Pape
Chicago, Illinois