Goldsmith-Limits of Idealism

8/13/2019 Goldsmith-Limits of Idealism

1/18

The Limits of IdealismAuthor(s): Jack Goldsmith and Stephen D. KrasnerSource: Daedalus, Vol. 132, No. 1, On International Justice (Winter, 2003), pp. 47-63Published by: The MIT Press on behalf of American Academy of Arts & SciencesStable URL: http://www.jstor.org/stable/20027822

Accessed: 16/07/2010 15:19

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at

http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless

you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you

may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at

http://www.jstor.org/action/showPublisher?publisherCode=mitpress.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed

page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

The MIT PressandAmerican Academy of Arts & Sciencesare collaborating with JSTOR to digitize, preserve

and extend access toDaedalus.

http://www.jstor.org
http://www.jstor.org/stable/20027822?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/action/showPublisher?publisherCode=mitpresshttp://www.jstor.org/action/showPublisher?publisherCode=mitpresshttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/20027822?origin=JSTOR-pdf


2/18

JackGoldsmith & Stephen D. Krasner

The limits of idealism

In 1939 E. H. Carr published what wasto become amodern classic on international relations, The Twenty Years Crisis,1919 -1939. Carr has usually been seen asa defender of realism and a debunker ofidealism, but his thinking was muchmore subtle. He believed that power and

Jack Goldsmith isa professor of law at theUniversity of Chicago. He haswritten widely inpublic international law and U.S. foreign relations law.This essaywas completed before he took a leave ofabsence to serve as special counsel to the General

Counsel of theU.S. Department ofDefense.StephenD. Krasner isGraham H. Stuart Professor of International Relations at Stanford University.A Fellow of theAmerican Academy since1991, he has written widely ina variety of areas,addressing issuesof market failure and distributional conflict in international political economy,and thehistorical practices of sovereignty especiallywith regard todomestic autonomy and nonintervention.His books include Defending the

National Interest (1978), Structural Conflict:The ThirdWorld Against Global Liberalism(1985), and Sovereignty: Organized Hypocrisyf>(1999)'

The views expressed in this article are those of theauthors and do not reflect the official policy or position of theDepartment of Defense or theU.S.government.

interest - the bread and butter of realism - were the primary determinants ofstate behavior. But he also believed that

peoples and their nations were motivated by normative values and aspirations,not merely by a desire tomarshal powerand defend material interests. Carr concluded that Utopia and reality are thusthe two facets of political science. Soundpolitical thought and sound political lifewill be found only where both have theirplace.For Carr the problem of the interwaryears was not international idealism itself, but rather international idealismrun amuck. At the core of the international idealism he criticized was the assumption that right-minded humanbeings could agree on abstract normative principles to guide national behavior, and that these principles, once understood and embodied in internationallaw, would influence nations to act withgreater justice. By his account, international idealism discounted other factors,including the distribution of power andeconomic and political interests.Carr famously argued that such idealism was self-defeating. Some nations,such as Germany, failed to comply withthe principles of reason embodied by theLeague of Nations and similar institutions, and appealed instead to compet

D dalus Winter 2003 47


3/18

JackGoldsmith &Stephen D.Krasneroninternational

justice

ing principles of law and morality to justify their self-interested and rapaciousacts. Other nations, such as Britain andFrance, relied too heavily on the paperguarantees of international law, and noton a

clear-eyed analysisof power and interest (both their own and Germany's),to secure international harmony. Carr attributed the growing international crisisin 1939 (his book was sent to the printerin July of that year) to the idealistic international institutions that were sup

posed tomake a second world war impossible.The kind of idealism that Carr understood to be so damaging to internationalpeace and stability in the interwar yearsis again informing many aspects of international politics. Three developments inparticular - the rise of universal jurisdiction, the creation of a new InternationalCriminal Court, and recurring demandsfor humanitarian intervention - reflect arenewed commitment to internationalidealism. Supporters of these institutionsand policies tend to believe that justice isbest served when it is isolated from politics and power. Only by insulating international institutions and practice fromthe bargaining and compromise thatcharacterize political decision-making,and from the domestic political pressureto which politicians must always bealert, can justice be fully realized. Onthis view, institutions and principles thatminimize the influence of power betterachieve justice than those inwhich power plays an important role ;and decisionsmade by unaccountable actors, especially judges, are more likely to be just thandecisions made by political leaders responsible to their electorates.We believe the new international idealism suffers from four fundamentalflaws:

First, it assumes the Utopian premisethat a global consensus can be reached,

not just on normative principles, butalso on when and how they should beapplied.Second, itminimizes considerationsof power, and assumes that norms ofright behavior can substitute for national capabilities and material interests.

Third, it neglects political prudence :itoffers a deontological rather than aconsequentialist ethics.Fourth, it consistently slights the valueof democratic accountability.Our claim is not that idealism in inter

national politics is irrelevant or inherently harmful. With Carr, we believethat normative ideals can provide a hopefor progress, an emotional appeal, and aground for international action. But wealso agree with Carr that ideals can bepursued effectively only if decisionmakers are alert to the distribution of power,national interests, and the consequencesof their policies. The lesson Carr teachesis that when idealism is not tempered byattention to these factors, the best canbecome the enemy of the good, and aspiration the enemy of progress.

lUniversal jurisdiction is the power of adomestic court to try foreign citizens,including government officials, for certain egregious international crimes committed anywhere in the world. This authority

ispremised

on the idea that human rights violations are an affront to allhumanity and thus may be punishedanywhere, regardless of the defendants'nationality or the place of the crime.Universal jurisdiction aims to strengtheninternational human rights law by marshaling politically independent domesticcourts to enforce that law. The classic

modern example is the Pinochet case, in

48 D dalus Winter 2003


4/18

which Spain attempted to extradite Pinochet from England (where he was undergoing back surgery) to stand trial inSpain for torture and related international crimes he allegedly committed inChile. (The extradition request originally charged Pinochet with crimes againstSpaniards as well, but these charges weredeemed inadmissible, thus making thecase one of 'pure' universal jurisdiction.)The House of Lords ruled that international law required England to extraditePinochet to Spain for these crimes, butthe government of Great Britain eventually sent Pinochet back to Chile afterdetermining that he was unfit to standtrial.1The Princeton Principles of UniversalJurisdiction, a document drafted by leading scholars and jurists from around the

world,2 are a comprehensive statementof the nature and scope of universal jurisdiction. The Principles extend universal jurisdiction to piracy, slavery, warcrimes, crimes against peace, crimesagainst humanity, genocide, and torture.They specify

that nationaljudicial

organs may rely on universal jurisdictioneven if their national legislation does notspecifically provide for it. They strip alldefendants - including sitting heads ofstate - of any official immunities. Andthey maintain that amnesties in particular are generally inconsistent with theobligation of states to provide accountability for serious crimes under interna

tional law. In short, the Princeton Prin- The limitsciples aim to replace impunity with ac- ?fldeallsmcountability by extending universal jurisdiction as broadly as possible.The Princeton Principles reflect conventional wisdom among idealists aboutthe shape and direction that international law should take. The Principles willlikely influence future universal jurisdiction prosecutions, because nationalcourts interpreting international law givespecial deference to the views of scholarsand jurists. In our view, however, thePrinceton Principles are an unfortunatedevelopment that exemplifies the newidealism's failure to take seriously thecontested nature of international norms,the importance of prudence, and thepossibility of abuse exacerbated by theabsence of democratic accountability.International criminal law is extraordinarily vague. Virtually everyone agreesthat genocide and torture and crimesagainst humanity are internationalcrimes. But when we attend to the details of what acts constitute these crimes,and of when these crimes can properlybe tried by courts, there ismuch disputeand little definitive guidance. Considerthree of many examples :

Among the most clearly defined of international crimes is torture, which theTorture Convention defines to includeany act inflicted by a public official bywhich severe pain or suffering, whether physical or mental, is intentionallyinflicted on a person to obtain infor

mation, punish, or intimidate.3 Amnesty International claims that theUnited States violates this principlewhen its police use stun guns, peppersprays, and restraint chairs, and when


i Regina v. Bow Street Magistrate, Ex Parte Pinochet, 2WLR 827 (HL) (1999).2 The drafting committee was comprised ofseven jurists from American universities, andthe meeting at which the Principles wereadopted was attended by scholars and juristsfrom Canada, Ghana, the United Kingdom,China, and Turkey as well as the United States,and included former presidents of the International Court of Justice, the American Bar Association, and Tokyo University.

3 Convention Against Torture and OtherCruel, Inhuman or Degrading Treatment orPunishment, art. 1.


5/18


justice

its prison officials use solitary confinement and related maximum securitydetention techniques.4 The UnitedStates disagrees ; it believes these practices are legitimate and do not constitute torture within the meaning of theTorture Convention. There is no definitive source or judicial decision thatcan resolve this disagreement. Underuniversal jurisdiction, any nationalcourt could try these U.S. officials if it,like Amnesty International and manyother human rights groups, viewedthese police practices as torture.

A crucial issue in any universal jurisdiction prosecution iswhether the defendant has an official immunity fromprosecution under international law.The existence and scope of these immunities as they apply to universal jurisdiction prosecutions are contestedand unsettled. The House of Lords interpreted international law to lift Pinochet's immunity as a former head ofstate. More recently, the InternationalCourt of Justice (icj) interpreted international law to hold that the Congoleseforeign minister was immune from auniversal jurisdiction prosecution inBelgium for alleged war crimes andcrimes against humanity he committed in his country.5 The ICJ decisiontechnically has no precedential effectbeyond the case it decided. So thescope of official immunity from a universal jurisdiction prosecution remainsan open question. Under universal jurisdiction, each national court gets todetermine the proper scope for itself.

When the United States and its NATOallies bombed Yugoslavia in 1999, theyviolated the UN Charter's prohibitionon the use of force against sovereignnations in the absence of SecurityCouncil authorization. Under thePrinceton Principles, NATO officials

might be subject to universal jurisdiction prosecutions for crimes againstpeace. But they might not; many international lawyers believe there is adeveloping customary exception to theUN Charter for certain humanitarianinterventions. In addition, AmnestyInternational and an independentgroup of law professors have concluded that NATO countries committedserious violations of the laws of war

when they purposefully destroyed civilian targets (such as a television station and electricity grids) and whenthey killed civilians by droppingbombs from no lower than fifteenthousand feet.6 The prosecutor at theInternational Criminal Tribunal in The

Hague investigated these allegationsand concluded, after much internalwrangling, that they did not warrantprosecution.7 Under a regime of universal jurisdiction, a court in any nation of the world could prosecuteNATO leaders and military membersand decide whether such actions constitute acceptable humanitarian intervention or criminal acts.Because the content of international

human rights law is so contested, courtsexercising universal jurisdiction in goodfaith are likely to interpret and enforcethis law inways that affected groups willview as unconvincing, self-serving, and


4 Amnesty International, Rights for All (NewYork :Amnesty International, 1999), chaps. 3-5.

5 Democratic Republic of Congo v. Belgium, judgment of 14 February 2002, available at .

6 Amnesty Report at


6/18

discriminatory. A universal jurisdictionprosecution can do more than provokeresentment among the affected groups ;it can also provoke domestic unrest orinternational conflict. Until recently,Belgium was considering universal jurisdiction charges against both Ariel Sharon and Yassar Arafat for human rightsviolations each allegedly committed inthe Middle East. (Such a prosecution remains a possibility.8) A decision by aBelgian court that Sharon or Arafat, orboth, are war criminals will not likelydampen discord in the Middle East. It ismuch more likely tomake matters worseby legitimizing views of extremists onboth sides.

Proponents of universal jurisdictionclaim that these leaders should be heldaccountable for their internationalcrimes, no matter what the consequences. This argument presupposes aconsensus on the nature of the international crimes we have just questioned.The argument also overlooks the possibility that a universal jurisdiction prosecution may cause more harm than theoriginal crime it purports to address.Universal jurisdiction courts and prosecutors possess neither the competencenor the incentive to fully consider theseharms. They are doubly unaccountablein the sense that they are relatively unaccountable to their own government (tothe extent that they are politically independent), and they are completely unaccountable to the citizens of the nationwhose fate they are ruling upon. Itdoesn't matter that they act with benevolent intent. What matters is that theymay do something that harms people to

whom they have no real connection and The limitswhose interests they are poorly posi- ?*l ea lsmtioned to assess. Because relevant constituencies cannot hold courts exercisinguniversal jurisdiction accountable forthe negative consequences of their rulings, the courts themselves will invariably be less disciplined and prudent thanwould otherwise be the case.

The inability of universal jurisdictioncourts to consider the consequences oftheir actions in affected countries is aparticular threat to amnesties, reconciliations, truth commissions, and similarprograms that can successfully facilitatetransitional justice. Modern international idealists tend to see these programs asa rejection of accountability. In fact,such programs often contain elements ofindividual accountability. More importantly, these programs are best viewed as

prudential arrangements that sacrificesome benefits - such as punishment ofthe guilty and restoration of the respectand integrity of victims - for the sake ofother values, including the minimizationof human suffering, closure, a stablepeace, and the like. In recent years, amnesties have been an important component in several peaceful settlements ofbloody civil conflicts, including ones inChile, Haiti, Sierra Leone, and South Africa.

As Michael Scharf correctly notes, arejection of amnesty and an insistenceon criminal prosecutions can prolong... conflict, resulting inmore deaths, destruction, and human suffering. 9 Consider the Truth and Reconciliation process in South Africa. Under the Princeton Principles, this process would notpreclude a universal jurisdiction prosecution, in a court outside South Africa,


8 The original prosecution against Sharon wasthrown out on the grounds that universal jurisdiction criminal prosecutions in abstentia wereprohibited under Belgian law. The Belgium Parliament is currently considering amending thatlaw to permit such prosecutions.

9 Michael Scharf, The Amnesty Exception tothe Jurisdiction of the International CriminalCourt, Cornell International Law Journal 32 (3)(1999): 507


7/18

Jack of Apartheid-era governmental officials.i 1 1 l '-fitC??^tmir\ This insistence on individual accountabtepnen D.Krasner bility at any cost could have terrible efon 1 fects on the still-fragile South Africaninternational .,. . A , . . , ,justice reconciliation. And itmight have precluded the reconciliation altogether (orat least made it even more rocky) haduniversal jurisdiction been widely practiced in the 1990s. In this way, universal

jurisdiction can make political solutionsto already difficult transitions to peaceand democracy even more difficult.The inability of universal jurisdictionprosecutors to weigh judiciously theconsequences of their actions distinguishes them from purely domesticprosecutors, and attests to the importance of democratic accountability inthe enforcement of criminal law. In a do

mestic prosecution, at least in the UnitedStates, the prosecutor is accountable tothe community inwhich she serves inthe sense that she is either elected (as in

many states) or (as in the federal system) appointed and subject to removalby elected officials. As a result, in deciding whether and how to prosecute acrime, a domestic prosecutor will oftentake into account the consequences ofthe prosecution for community health,safety, and morale.10 Inmany instancesthe adverse community consequences ofholding an individual accountable for apast crime can lead prosecutors to forgoprosecution, or to strike a plea deal favorable to the accused. (And of coursepolitical accountability also dampens thelikelihood that this discretionary processwill be abused.) Because universal jurisdiction prosecutions take place outsideaffected communities, universal jurisdiction courts and prosecutors lack theincentive, or the institutional capacity,to consider such tradeoffs.

The discussion thus far has proceededon the optimistic assumption that nations will apply universal jurisdictionprinciples in good faith. But there is noreason to believe this will be true. It isnot only the House of Lords and theBelgian courts that can prosecute underuniversal jurisdiction. Corrupt courtsthat lack political independence can aswell. And many nations will have incentives to engage in politically motivateduniversal jurisdiction prosecutions.The Princeton Principles rely on legalnorms to preclude such prosecutions.They insist that a state shall exerciseuniversal jurisdiction in good faith, andadd that a state and its judicial organsshall observe international due processnorms, including... the independenceand impartiality of the judiciary. Thereliance on legal norms in this context is

wholly unconvincing. The Principles failto consider why a nation with bad-faithmotives to prosecute a universal jurisdiction crime would care about such dueprocess principles - principles that, inany event, are manipulable in opportunistic ways.lo date, the costs of universal jurisdiction have not been obvious - at least inthe United States and Europe - becausemost universal jurisdiction prosecutionshave been brought by Atlantic alliancenations against offenders inweak countries. But there is no reason to think thispattern will continue. The rate of universal jurisdiction prosecutions has increased in recent years. And, as their potential and scope become clear, as human rights groups continue to pressurenations to bring such prosecutions, andas weaker countries realize that universal jurisdiction can be a tool for creatingpolitical mischief on the internationalstage, especially against more powerfulcountries, such prosecutions will in


?o See generally Darryl K. Brown, Third-PartyInterests in Criminal Law, Texas Law Review 80(6) (2002): 1383.


8/18

crease. Enthusiasm for universal jurisdiction might dampen in light of theicj's recent ruling on immunity for the

Congolese foreign minister. If not, weexpect that the many adverse consequences of universal jurisdiction we havediscussed will become more apparent.

2In July of 2002, international idealistsrealized a long-held dream :the creationof an International Criminal Court (ice)

with jurisdiction over genocide, crimesagainst humanity, war crimes, and, potentially, the crime of aggression.11

In some respects, the ICC is an improvement over a regime of universaljurisdiction by national courts. The ICCis a centralized institution. Its treaty defines the international crimes within itsjurisdiction. It also rejects universal jurisdiction, requiring instead a nexus tothe territory or persons of a treaty signatory.And yet the ICC has most of the othercharacteristics - and flaws - of universaljurisdiction. Its norms are still much tooopen-ended and contested to permit aconsensus on proscribed behavior; itsuppresses considerations of power; itlacks democratic accountability; and itcannot reliably balance legal benefitsagainst possible political costs.The ICC defines the crimes within itsjurisdiction. But these definitions rely agreat deal on contested international lawnorms, and they leave the ICC great interpretive flexibility. For example,crimes against humanity include imprisonment or other severe deprivationof physical liberty in violation of fundamental rules of international law. Unfortunately, international law provides

little concrete guidance about what The limitsthese fundamental rules require. After ?fldeahsmlisting other examples of crimes against

humanity, the ICC treaty describes as afinal one other inhumane acts of a similar character intentionally causing greatsuffering, or serious injury to body or to

physical or mental health. Such a criminal prohibition would almost certainlybe void for vagueness under U.S. law.To take another example, the ICC includes dozens of prohibitions under theheading of war crimes, including

willfully causing great suffering, or serious injury to body or health of civilians, and destroying or seizing the ene

my's property unless... imperatively demanded by the necessities of war. Thescope of these prohibitions is obviouslyuncertain, but it is easy to imagine thembeing applied to NATO actions in Kosovoand U.S. actions inAfghanistan. The ICCtreaty is chock-full of many similarlyvague and indeterminate criminal prohibitions.

One reason these vague norms are particularly troublesome is that the ICCprosecutor and court are unaccountableto any democratic institution or electedofficial. The ICC prosecutor is, to be sure,elected by a secret ballot by amajority ofthe signatory nations, each of which getsa single vote. But such an electoral system is problematic because, among other things, the vast majority of ICC ratifiers are weak nations that are never seriously involved in international police actions and thus have no incentive to consider the costs of zealous prosecutions.12Even more importantly, the prosecutorcan initiate investigations and prosecutions on his own, or at the suggestion ofthe UN or any signatory nation - all


il The ice's charter is available at ;allsubsequent quotes come from this document.

12 As of November 15, 2002, ICC ratifiers were :Andorra, Antigua and Barbuda, Argentina, Australia, Austria, Belgium, Belize, Benin, Bolivia,

Bosnia and Herzigovina, Botswana, Brazil, Bui


9/18


justice

without review, or the threat of review,by political actors. His prosecutions aresubject to legal review by the trial andappellate courts of the ICC, but thesecourts are similarly unaccountable toany democratic institution.This lack of accountability means thatthe ICC presents many of the dangers ofuniversal jurisdiction. Its structure is re

markably similar to the much-malignedU.S. Independent Counsel statute. Byguaranteeing independence at the priceof political control, it invites questionable and even politically motivated prosecutions. Legal restrictions and definitional limitations are not likely to provide real checks on the ice's behavior,for the ICC itself is the ultimate interpreter of these norms. Experiences withthe more accountable international tribunals in The Hague and Rwanda haveshown that international courts will notbe bound by the letter of their governingrules when justice as they conceive it requires otherwise. ICC jurisdiction canonly be expected to expand.In addition, the ICC, like a universal jurisdiction court, lacks the institutionalcapacity to identify and balance properlythe consequences of a prosecution onpotentially affected groups. The ICC

treaty insists that the most seriouscrimes of concern to the internationalcommunity as awhole must not go unpunished and their effective prosecutionmust be ensured. Here again we seemodern international idealism's com

mitment to individual accountability atthe expense of national amnesties andother forms of political reconciliation.The ICC theoretically permits the prosecutor to decline to investigate whenthere are substantial reasons to believethat an investigation would not serve theinterests of justice. But the final callrests with the prosecutor, who there isno reason to think has the perspective,information, or incentives tomake thisdecision wisely. (When Richard Goldstone, the Yugoslav Tribunal's first prosecutor, was asked if he worr[ied] aboutthe consequences to the Bosnian peaceprocess of indicting Radovan Karadzicand Ratko Mladic, he responded thatthe indictment was really done as, ifyou like, as an academic exercise_Because our duty was clear. 13)It is true that the ICC treaty requiresthe court to dismiss a case if it is alreadyunder investigation in national court,unless the State is unwilling or unableto genuinely carry out the investigationor prosecution. But the ICC has the finalword on what counts as a genuine investigation based on its perception ofwhether the domestic proceedings areinconsistent with an intent to bring theperson concerned to justice, a provisionthat opens the possibility of double jeopardy if the prosecutor decides that anational conviction or investigation istoo lenient and therefore not genuine. Itis natural to expect the ICC to interpretits charter inways that support its jurisdiction.


garia, Cambodia, Canada, Central African Republic, Columbia, Costa Rica, Croatia, Cyprus,Democratic Republic of Congo, Denmark, Djibouti, Dominica, Ecuador, Estonia, Fiji, Finland, France, Gabon, Gambia, Germany, Gha

na, Greece, Honduras, Hungary, Iceland, Ireland, Italy, Jordan, Latvia, Lesotho, Liechtenstein, Luxembourg, Macedonia (FYR), Mali,

Marshall Islands, Malawi, Mauritius, Mongolia,Namibia, Nauru, New Zealand, Niger, Nigeria,Norway, Panama, Paraguay, Peru, Poland, Por

tugal, Romania, Samoa, San Marino, Senegal,Sierra Leone, Slovakia, Slovenia, South Africa,Spain, Sweden, Switzerland, Tajikistan, the

Netherlands, Timor-Leste, Trinidad and Tobago, Uganda, United Kingdom, United Republicof Tanzania, Uruguay, Venezuela, and Yugoslavia.

13 Gary Bass, Stay the Hand of Vengeance(Princeton, N.J. :Princeton University Press,2000), 6-7.


10/18

Xerhaps the most troubling element ofthe ICC is its relationship to the U.N. Security Council. The United States arguedthat the ICC should prosecute only onthe basis of referrals from the SecurityCouncil. The ICC drafters rejected theU.S. proposal on the grounds that itwould inject international power politicsinto the decision whether to prosecute,and would give each of the Big Five powers a veto over any prosecution. Thedrafters viewed power politics, and theopportunistic use of Security Councilvetoes, as an obstacle to individual accountability under international humanrights law.

The ICC in its final form does permitthe Security Council to delay a prosecution for twelve-month renewable terms.But this just means that an ICC case cango forward so long as a single permanentmember vetoes a resolution of delay.And even if the Security Council votes todelay an ICC initiative (as it did when itgranted UN peacekeepers a twelvemonth immunity from prosecution inJuly of 200214), many

commentatorsbelieve the ICC has the power to engagein 'judicial review' of the Security Council and possibly to disregard its decision.There are at least two problems withthis attempt to eliminate power politicsfrom the enforcement of internationalcriminal law and to subvert the recognition of national power incorporated inthe UN Security Council. The first parallels a problem with universal jurisdiction :the ICC could initiate prosecutionsthat aggravate bloody political conflictsand prolong political instability in theaffected regions. Relatedly, the possibilities for compromise that exist in a political environment guided by prudentialcalculation are constricted when political deliberation must compete with anindependent judicial process. Many

believe that the threat of prosecution by The limitsthe international tribunal in The Hague ?*l ea lsmmade it practically impossible for NATOto reach an early deal with Milosevic,thereby lengthening the war and the suffering in the Balkans in the summer of1999. The best strategy for stability oftendepends on context and contingent political factors that are not reducible to arule of law. There is no reason to thinkthat a politically unaccountable prosecutor and court will make such difficult,context-specific calls wisely, even assuming they had the discretion to do so.The second problem results from whatCarr would have described as a chasmbetween theory and practice. Proponents of the ICC believe that itmay, inthe words of Human Rights Watch'sKenneth Roth, save many lives. 15 Thisiswishful thinking. Even if the ICC turnsout not to have the disruptive effects described above, and even if it is somehowable to prosecute low-level human rightsabusers, it is hard to see how the ICC canstop, or even affect, persons responsiblefor

large-scalehuman

rights abuses.The main reason for this conclusion isthat the ICC can only prosecute personsit can get custody over. The Milosovics,Mullah Omars, and Pol Pots of the

world, however, tend to hide behind national borders, where they are hard toreach. Moreover, the most notorious hu

man rights abusers have been motivatedby their own sense of mission and justice. They have seen themselves as saviors, not sinners. They have been determined to cling to power and they believe, as all leaders with amission do,that they can reshape the world in theirown image. If they have not been deterred by the threat of U.S. military intervention, they are unlikely to worrymuch about an ICC that lacks any real


14 Sec. Coun. Res. 1422.15 The Court the US Doesn't Want, The New

York Review of Books, 19 November 1998.


11/18

Jack enforcement mechanism of its own and?fephTnD

& thatmust dePend on itsmembers,Krasner whose decisions are uncertain, to arrestoninternational

justice rriand surrender suspects.lhis brings us to the U.S. refusal to participate in the ICC. There are many reasons for the U.S. stance, most notablythe perception that the United States'sdisproportionate share of internationalpolicing responsibilities exposes it to adisproportionate risk of politically motivated charges being brought before theICC.

Itmay seem odd that an institutionthat will have little effect on rogue human rights abusers could so concern theworld's greatest power. But U.S. troops,unlike rogue government officials, donot hide behind national borders. Hundreds of thousands of them are deployedaround the globe, making them potentially easy to grab and bring to The

Hague. (The United States is trying tocounter this danger by signing bilateralagreements inwhich the signatoriesagree not to surrender nationals of theother to the ICC.)Even if no U.S. defendant is broughtbefore the ice, it can still cause mischieffor the United States by being a publicforum for official criticism and judgment of U.S. military actions. For allthese reasons, the ICCwill more likelyaffect the activities of the generallyhuman-rights-protecting but militarilyactive United States than rogue state actors who hide behind walls of sovereignty (or in ungoverned areas) and care little about world public opinion and international legitimacy.

Despite his opposition to the ICC treaty, President Clinton signed it in 2001,just before he left office, so that theUnited States could participate in ongoing negotiations. InMay of 2002, however, the Bush administration officially

notified the United Nations that theUnited States does not intend to becomea party to the treaty. In August of 2002,President Bush signed the AmericanServicemen's Protection Act (aspa), astatute that enjoyed broad bipartisansupport. ASPA is sometimes called the

Hague Invasion Act because it authorizesthe president to use all necessary meansto release U.S. officials from ICC captivity. It also bars military aid to some nations that support the ICC, and it requires the president to certify that U.S.peacekeepers will be immune from ICCprosecution.U.S. opposition to the ICC is importantbecause U.S. military and financial backing have been crucial to the operation ofad hoc international criminal tribunals.Consider how Milosevic wound up inThe Hague. Itwas not the gravitationalpull of international norms that broughthim there. Rather, the United Stateswielded enormous diplomatic and military power to oust him from office, andthen threatened to withhold some $50

million in aid to the successor regime inYugoslavia until it turned over Milosevicto the Yugoslav tribunal.The Milosevic episode teaches a general lesson. The ICC simply cannot, without U.S. support, fulfill its dream ofprosecuting big-time human rights abusers who hide behind national borders.This iswhy the ICC's alienation of theUnited States may actually hinder ratherthan enhance human rights enforcement. We have already seen this effecton peacekeeping and ad hoc international tribunals. And of course the ICCwill

most likely chill U.S. military action notwhen central U.S. strategic interests areat stake (as in Afghanistan), but rather inhumanitarian situations (like Rwandaand perhaps Kosovo) where the strategicbenefits of military action are low, andthus even a low probability of prosecu



12/18

tion weighs more heavily. In this way,the ICCmay ironically increase ratherthan decrease impunity for humanrights atrocities.The establishment of an ICC that isunacceptable to the world's most powerful nation (and also to other large andpowerful nations, including Russia, China, Indonesia, and India) represents afolly reminiscent of the League of Nations, and portends a similar fate. Theinternational idealists who rejected U.S.demands for Security Council controlover ICC prosecutions aimed to decouplethe enforcement of international criminal law from international politics. Theywanted equal justice under law

- theequal application of international human rights law to weak and powerful nations alike. Both aims are a fantasystrongly reminiscent of the interwar idealism that Carr so effectively and presciently criticized. In demanding a fullloaf of neutral justice rather than a halfloaf of justice that accords with the interests of nations that can enforce it, andin creating an institution that relies onlegal norms wholly removed from considerations of power, international idealists may diminish rather than enhancethe protection of human rights.

3In the last decade alone, many hundredsof thousands of people have died in the

Balkans, central Africa, Afghanistan, Indonesia, Haiti, and elsewhere, some before our eyes on CNN. Humanitarian disasters, of which genocide is the mostappalling, are not pretty things. No reasonable person would argue that theyshould simply be ignored.The question iswhat to do aboutthem. Universal jurisdiction and the ICCare institutions designed to redress and- if deterrence can be made towork - toprevent such gross human rights abuses.

A third practice more directly aimed at The limitsprevention or mitigation is humanitari- ?*l ea lsman intervention.

Technically, humanitarian intervention in the absence of Security Councilauthorization violates the UN Charter.And until recently, many internationalidealists have viewed humanitarian intervention with suspicion on the groundthat nations often use humanitarian intervention as a cover for an unjustifiedinvasion of another country. But today

many international idealists are arguingthat states have a responsibility to act toprevent or rectify humanitarian catastrophes regardless of whether or nottheir material or security interests are atrisk.

Typical of this trend is a report issuedin 2001 by the International Commission on Intervention and State Sovereignty entitled The Responsibility toProtect.16 The Commission was supportedby a secretariat housed in Canada's Department of Foreign Affairs and International Trade and was composed of agroup of international personages cochaired by Gareth Evans, a former foreign minister of Australia, and Mohamed Sahnoun, an Algerian diplomatand special advisor to the UN secretarygeneral. The report argues that each nation has an international responsibilityto avoid or mitigate humanitarian disasters that could result either from conscious policy or from indifference or ineffectiveness in the face of natural calamities. This responsibility rests first

with the domestic government, but ifthat government fails to act then otherstates and international organizationshave a responsibility to protect.


i6 International Commission on Interventionand State Sovereignty, The Responsibility to Protect :Report of the International Commission on Intervention and State Sovereignty (Ottawa : International Development Research Centre, 2001).


13/18

Goldsmith &Stephen D.Krasneroninternational

justice

The Commission members, echoingSecretary-General Kofi Annan's state

ments, aim to undermine the assertion,explicit in the UN Charter, that the principle of sovereignty precludes externalintervention. Their report contends thatsovereignty resides with individuals aswell as states. The major purpose of government is protecting individual rights ;if a government manifestly fails to protect these rights by engaging, for instance, inwidespread killing or ethniccleansing, then others have an obligationto intervene. Sovereignty and the responsibility to protect are mutually constitutive, not contradictory, principles.States that massively fail to protect individuals within their own borders are notproperly exercising their sovereign authority and therefore cannot claim thatexternal intervention is illegitimate.17

No Jo one, regardless of his understanding of international affairs, would arguethat humanitarian concerns should carryno weight in decisions about intervention. The hard issue iswhether nations

have an obligation or responsibility tointervene for humanitarian reasonsalone.

The argument that nations are obligedto intervene ignores, or, at best, minimizes, the fact that electorates in advanced industrialized democracies havebeen reluctant to expend blood and treasure to deal with humanitarian catastrophes that do not affect their material interests. Despite the hundreds of thousands of deaths caused by human rightsabuses during the past decade, despitethe millions of such deaths in the lastcentury, humanitarian intervention has

not generated any wellspring of supportamong domestic publics in the advancedindustrialized democracies that possessthe military muscle tomake a difference.

Germany and Japan have been extremely reluctant to engage in overseasdeployment of their military forces forany purpose, humanitarian or otherwise.No major European state has made a sustained commitment to humanitarian intervention. Indeed, no combination of

European countries has the military capability to conduct a serious military intervention of any kind outside of Europe, and none appears willing to makethe budgetary commitments that wouldmake such interventions possible. European forces do have the ability to participate in peacekeeping operations, buteven here the tolerance for losses can belimited. Belgium, for instance, whichhad several hundred troops deployed inRwanda at the beginning of the 1994 crisis, withdrew them after ten of its soldiers were killed by Hutu militia.The extreme caution with whichAmerican presidents have engaged inhumanitarian interventions suggeststhat they believe that they are walkingon very thin ice when they cannot convincingly tie their activities to materialinterests that the voting public can understand. To be sure, the Clinton ad

ministration undertook humanitarianinterventions in Somalia, Haiti, Bosnia,and Kosovo. But the last two were overtly tied to the viability of NATO andAmerican security, and even here the

United States relied on high-altitude airattacks that minimized the chances forAmerican casualties. In Somalia, Clintonextricated the United States after eighteen soldiers were killed. He did not actin Rwanda where an estimated eighthundred thousand people died - a decision that caused him no discernible political problem.


17 International Commission, The Responsibilityto Protect, 8 ;Fernando R. Tes?n, The Case for

Humanitarian Intervention, Public Law andLegal Theory Working Paper No. 39, FloridaState University College of Law, November2001, p. 2-3.


14/18

The report of the International Commission on Intervention and State Sovereignty recognizes the problematic absence of democratic support for humanitarian intervention. It suggests that

the budgetary cost and risk to personnelinvolved in any military action may in factmake itpolitically imperative for the intervening state to be able to claim somedegree of self-interest in the intervention,however altruistic its primary motivemight actually be. Apart from economicor strategic interests, that self-interestcould, for example, take the understandable form of a concern to avoid refugeeoutflows, or a haven for drug producers orterrorists, developing in one's neighbourhood.18The Commission here acknowledges a

gap between its own prescriptions aboutthe moral obligation to act tomitigatehumanitarian disasters, and the viewsheld by democratic electorates in Europe, Japan, and North America - electorates whose money would be spentand whose sons and daughters could bekilled.

This absence of democratic support isa fundamental problem for those whoinsist that nations should intervene toarrest human suffering in other nations.A basic tenet of the idealistic outlookthat underlies demands for humanitarian intervention is that liberal democracyis the morally preferable form of domestic governance.19 In a democracy, foreign policy must have national supportand be justified in terms acceptable tothe voting public. But this means thatpolitical leaders cannot engage in acts of

altruism abroad much beyond what con- The limitsstituents and/or interest groups will sup- ?*l ea lsnport. This conclusion is fatal to the interventionist project. The most we can expect is that when a nation's strategic interests dovetail with an inclination toward genuine humanitarian intervention, itwill intervene - as the UnitedStates did in Bosnia, Haiti, and Kosovo.

Once again, this means that international justice will depend on the powerand interest of nations, and will often result in uneven patterns of enforcementthat critics deride as hypocritical. Opportunistic interventions are also whatgive rise to the (not unjustified) concernthat many so-called humanitarian interventions are ruses for invasions motivated in large part by strategic ends. A cleareyed analysis of interventions would realize that such mixed-motive cases areprobably the best we can hope for. Thepresence of mixed motives does not detract from the fact that some such interventions might help local populations,as the Kosovo intervention arguably did.

guments for the duty to interveneand prevent human suffering suffer fromanother problem in addition to the democratic deficit: they underplay, even ifthey do not ignore, questions of political

prudence. Political prudence demandsthat foreign policy actions be judged interms of their consequences, not theirintentions.Information affecting the cost of inter

vention, including the state of affairs inthe target country and the price of intervention - inmoney spent, lives lost, andother opportunities forgone - are hard todetermine. Similarly, the consequencesof intervention, including the costs and

A,


i8 International Commission, The Responsibilityto Protect, 36.

19 Cf. Brian Barry, Statism and Nationalism :A Cosmopolitan Critique, in Global Justice, ed.Ian Shapiro and Lea Brilmayer (New York :

New York University Press, 1999) : the domestic aspect of cosmopolitanism is ... liberal de

mocracy.


15/18


justice

likelihood of constructing a social andpolitical order superior to that whichwould exist in the absence of intervention, are hard to know in advance. Thesefactors make it all the more difficult forresponsible democratic leaders to intervene, even if they were willing to ignorethe absence of domestic support.In several articles and awidely praisedbook, Samantha Power has been highlycritical of American policy for failing toprevent or react to the genocidal policyadopted by Hutu extremists in Rwanda.She faults, among others, American Ambassador David Rawson for his failure toanticipate the scale of the killings. Shequotes Rawson as follows: Most of usthought that if a war broke out, itwouldbe quick, that these poor people didn'thave the resources, the means, to fight asophisticated war. I couldn't have knownthat they would do each other inwiththe most economic means. 20

Rawson was, however, far from ignorant about Rwanda. He had grown up inBurundi, the son of an American missionary. He spoke the local language. Hecould not, in Power's words, have beenmore intimate with the region, the culture, or the peril. Yet he totally missedwhat was about to occur. Power arguesthat Rawson and others suffered from

what she calls imaginative weakness.She also claims that US officials who'did not know' or 'did not fully appreciate' usually chose not to. 21 But itwouldbe more straightforward and obvious tosay that policymakers must always makeguesses about alternative states of theworld with limited information andtime - and absent overwhelming information to the contrary, there is no rea

son to reject that state of the world thatismost consistent with the policy options they find most attractive.Ex ante efforts to assess systematicallythe costs and benefits of any intervention are extraordinarily challenging.What is happening on the ground israrely known with certainty. Even afterthe killing has begun, observers mightnot know whether they face a civil waror a systematic effort tomurder members of a particular ethnic group. American leaders thought that bombing Serbiawould provide Milosovic with the coverthat he needed to withdraw from Kosovo ; instead it led him to accelerate efforts at ethnic cleansing. Even ardentsupporters of humanitarian interventionrecognize that there must be some assessment of reasonable cost for the inter veners. But it is usually difficult toknow beforehand what such costs mightbe. How many foreign troops wouldhave been killed if there had been aquick reaction to developments inRwanda? How many NATO soldierswould have been lost if an aggressive,rather than cautious, air and groundcampaign had been conducted againstSerbia? What would the casualties be ifan effective external fighting force weredeployed to the Sudan? If a politicalleader guesses wrong, what would be theimplications for his ability to secure political support from his own electorate?

JLinally, and perhaps most challengingly,

is thequestion

of reconstruction. Juststopping the killing is not enough. If intervention occurs, the InternationalCommission argues, there is an obligation to rebuild. Refugees must be allowed to return; human rights must berespected; judicial systems must be reconstructed; militaries must be demobilized.

6o D dalus Winter 2003

20 Samantha Power, A Problem from Hell (NewYork: Basic Books, 2002).21 Samantha Power, Genocide and America,The New York Review of Books, 14March 2002.


16/18

Success requires creating institutionalarrangements towhich all of the relevant local actors will adhere. This is

more easily done in some areas than inothers. The highly developed institutional structures of Europe- the Euro

pean Community, the Organization forSecurity and Cooperation in Europe, andNATO - offer alternatives to conventional sovereignty for the Balkans. These alternatives make it easier tomaintain minority rights and prevent conflict,22 although even here the prospects for longterm success are uncertain. Other neighborhoods, such as ones inAfrica andcentral Asia, are less hospitable. Buildingstable and tolerant societies in these areas is an enormous challenge, and thereis no guarantee of success. Despite aclear security motive for intervention,

widespread international support, andbillions of dollars in assistance, theAmerican-led effort to reconstruct Afghanistan might still fail. It is all themore difficult to sustain such efforts incountries where the direct security interests of powerful and rich states are notengaged.The difficulty of assessing the costsand benefits of intervention and the absence of domestic support for purely humanitarian actions do not rule out suchactivities. But these considerations dosuggest that it iswishful thinking to presume that the responsibility to protectwill become a central norm in statedecision-making. Any decision to engage in humanitarian intervention musttake into account available resources,

domestic support, probabilities of success, the danger of doing more harmthan good, and, most importantly, thematerial interests of the intervener. Thisonce again will lead, at best, to selectivejustice.

In internationalpolitics,

selective justice is the best we can hope for.

4We have offered reasons to be pessimistic about the efficacy of three regimes -universal jurisdiction, the ICC, and (certain conceptions of) humanitarian intervention - that aim to enforce interna

tional human rights norms. Our point isnot to criticize the norms themselves,but to focus attention on pathologiesthat may result from the inadequate institutions inwhich they are embedded.International institutions can damagerather than promote international idealsif they are incompatible with the interests of those states whose support isneeded for their success.

Consider two successful weddings ofideals, interests, and power - the first associated with the beginning of the modern state system, and the second with itspossible transformation. The treaties ofWestphalia (1648) that ended theThirtyYears War are famous for embracing theprinciple that the prince determines thereligion of his territory. But the actualterms of the treaties limited the emperor's right to regulate religious practiceswithin the Holy Roman Empire. Theserestrictions, analogous tomodern human rights, protected some minority religious practices, mandated the sharingof public offices in some cities withmixed populations, and most importantly, altered the domestic institutionalstructure of the Empire by requiring thatreligious questions be decided by amajority of Catholics and Protestants voting separately in the diet and courts of

The limitsof idealism


22 Robert O. Keohane, Political Authority after Intervention: Gradations in Sovereignty,in Jeff Holzgrefe and Robert O. Keohane, eds.,Humanitarian Intervention :Principles, Institutionsand Change (Cambridge :Cambridge University

Press, forthcoming); MichaelIgnatieff, StateFailure and Nation Building, unpublished,Harvard University, Kennedy School of Government, 2002, passim.


17/18


justice

the Empire.23 These protections werelargely efficacious, not because the normof religious toleration motivated leaders(in fact no European leader believed inthis norm), but rather because the Thirty

Years War had shown that efforts to repress religious practices in Germanywere so politically volatile that theycould threaten the very existence of theEmpire.The European Union is another example. The Union has transformed the continent from one riven by war in the firsthalf of the twentieth century to one inwhich war is unthinkable, at least amongmember states. European integrationwas motivated by ideals, by an aspiration among a small number of leaders tobind the states of Europe into a peacefulweb of relations from which they couldnot extricate themselves. An importantelement of this integration was the creation of a human rights regime that fostered democracy and tolerance in the domestic realm. But these ideals could onlybe realized by grounding them in interests, economic and political,

andby

creating institutions that made it possiblefor European leaders to ensure that nonation had an incentive to defect.24

The Peace ofWestphalia and the European Union are institutions that successfully harnessed the power and interest ofnations to enforce moral ideals. Theseinstitutions worked because each nationbenefited from the institution and hadan interest in complying with its terms.

Unfortunately, it is not always, or even

usually, possible to yoke self-interest into such a self-enforcing mechanism topromote moral ideals.When self-enforcement fails, the alternative is a system of selective justice enforced by the powerful,

oneconsequenceof which is effective immunity for the

powerful. What has not proved possiblein international affairs is universal international justice based on legal normsthat operate in the absence of either selfenforcement or hegemonic dominance.This iswhy we believe that the normthat states ought to intervene militarilytomitigate humanitarian catastropheswill not become accepted in practice.Persons motivated to commit the abuseshave nothing to gain from forgoing theabuse out of deference to internationalnorms alone. And the leaders of democratic states - or, perhaps more to thepoint, American presidents - will not beable to secure the domestic political support needed to place lives at risk whentheir states' security interests are not directly at stake.Universal jurisdiction and the ICC, incontrast, can matter, because they establish judicial procedures that rely on theauthority and policing powers of national states for enforcement. The problemhere is not that such institutional arrangements will be ineffectual. As wehave suggested, these institutions can affect the costs of political action, and canhave a special impact on nations like theUnited States that are globally active andcare about

public opinionand interna

tional legitimacy. The problem withthese institutions is that they can domore harm than good.The ICC and universal jurisdiction assume a consensus on human rights ideals and their applicability, and expectthat compliance will follow. But no suchconsensus exists; non-national judicialproceedings will always be open to


23 See Stephen D. Krasner, CompromisingWestphalia, International Security 20 (3) (Win

ter 1995/1996).

24 For a discussion of the importance of interests in the process of European integration see

Andrew Moravcsik, The Choice for Europe :SocialPurpose and State Power from Messina toMaastricht (Ithaca, N.Y. :Cornell University Press,1998).


18/18

charges of bias, an ambiguity that Milosovic has exploited in his trial before theInternational Criminal Tribunal for theFormer Yugoslavia (an institution thatavoids many of the pitfalls of the ICC ).The ICC and universal jurisdiction severthe link between norm enforcement andpolitical accountability. One consequence of this separation is that the institutions are practically, and in somecircumstances legally, discouraged fromengaging in assessments of costs andbenefits that are often so important forthe prevention of human suffering. As aresult, such institutions may worsenrather than alleviate human rights catastrophes.25

The limitsof idealism


25 The authors would like to thank RyanGoodman, Michael Ignatieff, Robert O. Keohane, Tracey Meares, Eric Posner, and Adrian

Vermeule for their comments on earlier draftsof this paper.

Documents

Goldsmith-Limits of Idealism