. 5
( 15)


ferring political power and control over the institutions of governance would
foster “the emergence of politically moderate parties and credible leaders”
and a general “political maturity” conducive to an acceptable ¬nal status
But the process of developing the Constitutional Framework for Provi-
sional Self-Government in Kosovo (the second half of the blueprint) illus-
trates the dif¬culties involved. From the start, Albanians fought to create a
constitutional framework “as close to an independent state as possible.”64
Serbs, fearing the ¬nal status implications, largely boycotted the two-month
consultation process.65 Even the name of the document was contentious.
Kosovar Albanians wanted to call it a constitution; the SRSG and other inter-
nationals saw that as implying a judgment on Kosovo™s ¬nal status. “Con-
stitutional Framework” was the compromise.66 More importantly, “none of
the local participants agreed to the text as ¬nally adopted “ a ˜compromise™
that had to be forced on them by Haekkerup [then the SRSG].”67
The Constitutional Framework provided for the “gradual transfer of
responsibilities to Provisional Institutions of Self-Government” in a broad
range of internal areas, from economic policy to good governance and
human rights. The hope has been that the transfer would “enhance demo-
cratic governance and respect for the rule of law.” But the Framework is
an UNMIK regulation, and ultimate authority continues to reside with the

back to some parts of Kosovo.” International Crisis Group, Return to Uncertainty: Kosovo™s
Internally Displaced and the Return Process, December 13, 2002, at 3. Conversely, Serbs
are reluctant to return to what may be a future independent state in which they will be a
minority. Id.
62 United States Institute of Peace, Kosovo Decision Time: How and When?, February 2003, at
4, available at http://www.usip.org/pubs/specialreports/sr100.pdf (last accessed January 27,
63 Chesterman, supra note 31, at 6.
64 International Crisis Group, Kosovo: Landmark Election, November 21, 2001, at 3.
65 Id.
66 Id.
67 Simon Chesterman, Kosovo in Limbo: State Building and “Substantial Autonomy,” August
2001, at 6“7, available at

SRSG, who retains a “virtually unlimited prerogative to override” the new
On the critical issue of majority“minority relations, the Constitutional
Framework is more minority rights oriented than consociationalist. The new
governing institutions include a president (seen by Albanians as a future
head of state) and an assembly. The assembly is structured to provide for
overrepresentation of minorities, particularly Serbs, but stops well short of
affording them veto power. The assembly has 120 seats, 100 to be chosen by
proportional representation; 10 are set aside for Serbs and another 10 for
other minorities. Thus, if Serbs participated fully, they might command over
20 seats.69 In theory, Serbs might exercise signi¬cant in¬‚uence given the split
between the two major Albanian parties, each of which might wish to secure
Serb votes for its programs. In practice, however, the Albanian parties tend to
agree on most issues and are unlikely in the foreseeable future to wish to con-
fer much in¬‚uence on Serbs by courting their votes.70 Other features of the
Assembly attempt to require interethnic compromise. In a vaguely consocia-
tional structure, the seven-member presidency of the assembly requires inclu-
sion of Serb and minority representatives. Similarly, at least one Serb and one
other minority must be included in ministerial positions in the executive.
The Constitutional Framework also contains a catalog of minority rights,
designed “to preserve, protect and express their ethnic, cultural, religious,
and linguistic identities.” These rights include language, education, cultural,
and employment rights, as well as obligations to preserve religious sites and
guaranteed access to public media. The Constitutional Framework also pro-
vides that the “the SRSG will retain the authority to intervene as necessary
in the exercise of self-government for the purpose of protecting the rights of
Communities and their members.”
Unfortunately, these provisions for minorities do not go far enough to
render Serbs secure in an independent Kosovo. At the same time, they do go
far enough to reinforce the salience of ethnicity in Kosovo politics, which
“continues to be fought strictly along ethnic lines.”71 This is the underlying
problem of trying to resolve intercommunal con¬‚ict through legal protec-
tion of minority rights. It is a viable approach in societies that already have
a tradition of democratic accommodation and respect for human rights. But
in deeply divided societies that lack such traditions, minority rights simulta-
neously offer too little and too much. For most Serbs, the paper protections
afforded by the Constitutional Framework offer little or no real protection
against Albanian animosity. Serb politicians warn that independence will
provoke a mass exodus. The end result may be the ethnically homogenous
68 Kosovo: Landmark Election, supra note 64, at 4.
69 Id. at 8.
70 Id. at 12.
71 Chesterman, supra note 31, at 7.

territory that Albanian separatists have long sought but that would run
counter to western insistence on creating a multiethnic Kosovo. At the same
time, so long as power and resources are divided along ethnic lines, politics
will continue to turn on ethnicity.
Multiethnic harmony, the goal underlying the minority rights instruments
of the 1990s and ostensibly the objective of western involvement in the for-
mer Yugoslavia, cannot ¬‚ourish so long as ethnic identity crowds out other
interests (such as class, ideology, and profession) and precludes the forma-
tion of interethnic coalitions. Put another way, dividing power among ethnic
communities represents an acceptance of the logic of nationalism, applied
at the (nominally) substate level. This is the paradox at the heart of western
efforts to achieve viable political settlements to ethnic con¬‚icts in Bosnia,
Kosovo, and other deeply divided societies. The price of such settlements is
often the perpetuation of the differences that necessitated the settlements in
the ¬rst place. In Kosovo, the emphasis on ethnicity in politics is less than in
Bosnia, because of the imbalance in ethnic group numbers. The blueprints
are therefore different, but the underlying obstacle in each case is the same,
getting hostile national communities to live together on equitable terms.
Despite continuing interethnic antagonism, the status issue is clearly com-
ing to a head. In November 2005, the UN Secretary-General appointed for-
mer Finnish President Martti Ahtisaari as his special envoy for ¬nal status
talks. The two most prominent options remain autonomy and independence.
But Ahtisaari and others have called for further progress on building demo-
cratic institutions and protecting minority rights as a precondition. Thus, at
the time of writing this chapter, the blueprint for Kosovo remains incomplete.


Afghanistan and Iraq do not ¬t neatly into the blueprint categories described
above. Intervention was not driven by large-scale internal con¬‚ict over
resources or the identity of the state but rather by the perceived security
interests of the United States and its coalition partners. Nonetheless, condi-
tions associated with the identity- and resource-driven con¬‚ict models exist
in both Afghanistan and Iraq and raise many of the same issues seen in other
post-con¬‚ict environments. In Afghanistan, the power base of most political
elites largely tracks ethnic and tribal af¬liations, rendering aspects of the
identity-con¬‚ict models relevant. Nonetheless, ethnicity does not dominate
political behavior in the way it does in places such as Bosnia and Kosovo.
Thus, the blueprint for Afghanistan emphasizes majoritarian elections and
institution-building. By contrast, in Iraq, Kurdish desires for independence,
and increasing tensions between Shiites and Sunnis, have brought to the
fore many of the issues associated with identity-based con¬‚icts, producing

a complicated constitutional politics that renders straight majoritarian
arrangements untenable.

A. Afghanistan: Center and Periphery
On October 7, 2001, the United States responded to the September 11 ter-
rorist attacks on the World Trade Center and the Pentagon by launching
Operation Enduring Freedom to eliminate al-Qaeda forces from their bases
in Afghanistan and to drive their Taliban hosts from power. Working closely
with Northern Alliance forces, the United States and its allies quickly took
control of most of the country. After twenty-¬ve years of nearly unremitting
warfare, state institutions were in shambles, the country™s infrastructure was
decaying or destroyed, millions of Afghans lived abroad as refugees, and
the country was riven by ethnic, tribal, ideological, and religious ¬ssures.
The interveners, especially the United States, feared a deep and protracted
involvement in Afghanistan and consciously rejected the transitional admin-
istration model employed in Kosovo and East Timor. Where in Kosovo the
UN and other international actors played a dominant role in rebuilding the
state, taking the lead in pursuing security, reconstruction, and democratiza-
tion, the United States insisted in Afghanistan on a “light footprint.” The
goal of this approach was “local ownership of the revival process as much
as possible. . . .”72
The U.S. motives for the light-footprint approach were clear. The United
States did not want to tie down large numbers of troops attempting to secure
and rebuild the country. Instead, the United States wanted to help establish a
new Afghan government, one sympathetic to U.S. interests, and to place on it
primary responsibility for post-con¬‚ict reconstruction. Other international
actors, including the United Nations, did not want to undertake large-scale
nation-building activities without full U.S. involvement and support. In par-
ticular, they did not want to attempt to restore security outside Kabul without
U.S. military backing.
Instead, the goal, as articulated at a November 2001 meeting of the “Six
plus Two” group (the six states bordering Afghanistan, the United States,
and the Russian Federation), was to create “a broad-based, freely chosen
Afghan government,” through which international assistance could be chan-
neled.73 Under UN auspices, and with active encouragement from the United
States and other coalition members, various Afghan factions met in Bonn in
late November to hammer out an agreement on transitional arrangements.
The participating parties included Afghan military commanders, ethnic
72 Robert McMahon, UN: ˜Light Footprint™ in Afghanistan Could Hold Lessons for
Iraq, Radio Free Europe/Radio Liberty, April 7, 2003, available at http://www.rferl.org/
73 See UNAMA/OCPI, The United Nations in Afghanistan: 11 September 2001“11 September
2002, August 31, 2002, at 2.

group leaders, expatriate Afghans, and representatives of Afghanistan™s
exiled king.74 On December 6, under considerable pressure from the United
States and other interested states, the negotiating factions adopted the Bonn
Agreement on Provisional Arrangements in Afghanistan Pending The Re-
establishment of Permanent Government Institutions.75 The Security Coun-
cil endorsed the agreement the following day.
The Bonn Agreement offered a simple blueprint for Afghanistan™s future.
The agreement provided for an immediate transfer of power to an Interim
Authority, chosen by participants in the Bonn negotiations. The Interim
Authority was charged with convening an emergency loya jirga within six
months, which in turn was tasked with creating a broad-based Transitional
Authority “to lead Afghanistan until such time as a fully representative gov-
ernment can be elected through free and fair elections to be held no later than
two years from the date of the convening of the Emergency Loya Jirga.”76
The agreement also provided for the convening of a constitutional loya jirga
within eighteen months of the establishment of the Transitional Authority,
to adopt a new constitution for Afghanistan. The agreement recognized the
need for “broad representation in these interim arrangements of all segments
of the Afghan population” and noted that “these interim arrangements are
intended as a ¬rst step toward the establishment of a broad-based, gender-
sensitive, multi-ethnic and fully representative government. . . .”77 Annex I
requested the Security Council to deploy a UN-mandated force to “assist in
the maintenance of security for Kabul and its surrounding areas” and “as
appropriate” to expand “to other urban centres and areas.”78 As far as it
went, the Afghanistan blueprint was reasonably sound. It offered a chance
for Afghans, backed by international support, to establish a democratic gov-
ernment in Kabul and to begin the reconstruction process working from the
center out.
At the outset, the political process launched with the Bonn agreement was
not representative of Afghan society as a whole. The Northern Alliance dom-
inated the talks leading to adoption of the Bonn Agreement and took control
of the ministry of defense and other key security services at the conclusion
of those talks.79 The Emergency Loya Jirga, most of whose 1600 members

74 Human Rights Watch, Afghanistan™s Bonn Agreement One Year Later: A Catalog of Missed
Opportunities, December 5, 2002, available at http://www.hrw.org/backgrounder/asia/
afghanistan/bonn1yr-bck.htm accessed through http://www.hrw.org/press/2002/12/afghan-
1205.htm (last accessed January 28, 2006).
75 Available at http://www.unama-afg.org/docs/ nonUN%20Docs/ Internation-Conferences&
Forums/Bonn-Talks/bonn.htm (last accessed January 28, 2006).
76 Id., para. 4. A loya jirga, or national assembly, has been used at various times in modern
Afghan history to resolve political crises.
77 Id., Preamble.
78 Id., Annex I.
79 McCool, supra note 6, at 8.

were elected under sometimes questionable conditions, constituted a more
broadly representative body, but it “was a rowdy, disorganized, politically
combative affair operating generally without regard to the rules of proce-
dure that had been developed.”80 The Constitutional Loya Jirga was itself
reasonably representative of Afghanistan™s fractured society and managed to
reach consensus on a new constitution, but it had little time to review and
reach agreement on an extremely complex set of issues and opened the door
to perhaps excessive executive branch input. Still, the results of the process
can be deemed “within tolerance levels” for a country emerging from total
Most of the steps outlined in the blueprint have now been completed.
On December 15, 2001, the ¬rst troops of the UN-mandated International
Security Assistance Force (ISAF) arrived in Kabul.82 A week later, the Interim
Administration, headed by Hamid Karzai, took of¬ce. In late January 2002,
prospective donors meeting in Tokyo pledged $5 billion in reconstruction
assistance over a six-year period.83 For the next few months, the Special Inde-
pendent Commission for the Convening of an Emergency Loya Jirga, with
help from the United Nations Assistance Mission in Afghanistan (UNAMA),
engaged in a broad-based consultative process leading to the election of can-
didates for the Loya Jirga, which was held in June in Kabul. The Loya Jirga
elected Karzai as president of the new Transitional Administration. At the
same time, work proceeded on a draft constitution, which was adopted in
December 2003, with elections for a permanent government set for June
2004. To date, the political reform timetable set out in the Bonn Agreement
has largely held, although elections were conducted in October rather than
June 2004.
Moreover, Afghanistan has not relapsed into large-scale civil war; more
than two million refugees have returned; efforts to rebuild schools, clinics,
and markets are underway; international aid has helped ameliorate a dire
humanitarian situation; and the Karzai government has grown increasingly
coherent.84 Further, the currency is reasonably stable, the Afghan National
Army is improving, and approximately 60,000 former combatants have
demobilized.85 But this apparent march toward a revived political life in
Afghanistan masks deep structural problems. Afghanistan is still a country

80 Id., at 10.
81 Id., at 5.
82 UNAMA/OCPI, supra note 73, at 3. In August 2003, NATO took over ISAF™s role.
83 Id.
84 Ray Salvatore Jennings, United States Institute of Peace, The Road Ahead: Lessons in Nation
Building from Japan, Germany, and Afghanistan for Postwar Iraq, April 2003, at 20, avail-
able at http://ics.leeds.ac.uk/papers/pmt/exhibits/2165/pwks49.pdf (last accessed January
28, 2006).
85 International Crisis Group, Rebuilding the Afghan State: The European Union™s Role,
November 30, 2005, at 4.

with minimal state institutions, an average life expectancy of forty-¬ve, a
population that is 70 percent illiterate, little domestic revenue or economic
growth outside the drug economy, and a continuing insurgency.86
By all accounts, the core problem in Afghanistan is insecurity. ISAF, with
some 4500 troops (later expanded to 9000), has been of only modest help.
By way of comparison, “in Kosovo, Bosnia, and East Timor the interna-
tional peacekeeping force amounted to one peacekeeper per seventy or so
people,” whereas “in Afghanistan that ratio was one peacekeeper per ¬ve
thousand people.”87 Worse, U.S. and international timidity largely con¬ned
ISAF to Kabul, at least in the early post-con¬‚ict period. To promote stability
in rural areas, the United States and other countries dispatched civil“military
provincial reconstruction teams (PRTs) to various cities around the country.
Although their impact has been positive in some cases, they have not as yet
provided the Karzai government with a substantial lever by which to rein in
and marginalize warlord activity, and their ultimate success will be measured
by the extent to which they can hand their jobs off to provincial and district
The efforts of the United States and its coalition partners to help the
Afghan government extend its authority throughout the country have been
complicated by the United States™s continued pursuit of Taliban and al-Qaeda
elements operating in the hinterlands. Although keeping the extremists at bay
has given some provinces and districts more breathing space than they would
otherwise have, it is also true that the United States has enlisted the support
of some of the same warlords who defy central authority and intimidate the
local populace. The United States and the Karzai government are attempting
to marginalize these warlords gradually, and in a few cases (such as Herat
province) they have succeeded. But armed illegal groups still operate with
impunity in some areas.
The continued power of local warlords renders uncertain the long-term
prospects for the Afghan transition. The central government™s writ is still
weak or directly challenged in many rural areas. Insecurity and lack of fund-
ing continue to undermine reconstruction efforts.88 “And as infrastructure
decays from lack of maintenance, so the status of local Afghan authori-
ties who have unilaterally identi¬ed the community™s needs is enhanced,”
even though the same individuals are “often part of the architecture of inse-
curity and oppression.”89 Farmers unable to obtain access to credit have

86 Id.
87 Sam Zia-Zari¬, Human Rights Watch Report, Losing the Peace in Afghanistan, at 6“7
(2004), available at http://hrw.org/wr2k4/download/5.pdf (last accessed January 28, 2006).
88 “[I]n Kosovo the international community spent twenty-¬ve times more money, on a per
capita basis, than it has pledged in Afghanistan. . . . [W]hile Iraq received U.S. $26 billion in
reconstruction aid in 2003, Afghanistan received less than $1 billion.” Id., at 2.
89 Jennings, supra note 84, at 24.

increasingly turned back to poppy cultivation, and the heroin trade “has
blossomed again in Afghanistan, generating billions of dollars for forces
outside the control of any legitimate authority,” including local warlords
“who use this money to increase their military capability and gain indepen-
dence from the central government and any international troops working
with them.”90 In this environment, many worry that the government and
constitution created through the Bonn process will rule on paper only.
The new constitution, and the process that created it, are not without their
strengths. The Constitutional Review Commission, a diverse body of experts,
community leaders, religious leaders, and others, held nationwide consulta-
tions, soliciting the views of some 150,000 Afghans at over 500 public meet-
ings. The draft was the subject of vigorous debate in the Constitutional Loya
Jirga among a diverse group representing a broad spectrum of Afghan society.
On its face, the constitution establishes a modern, democratic presidential
system with a bicameral legislature. Although the constitution declares Islam
to be the state religion, it also promises to respect religious freedom, free
speech, and other internationally protected human rights. The constitution
eschews the consociationalist approach to ethnic divisions but recognizes
the equality of all ethnic groups and tribes and promises “balanced devel-
opment in all areas of the country.”91 Women and men are to be treated
Underneath the surface, however, problems are apparent in both constitu-
tional process and substance. Although the Constitutional Review Commis-
sion was mandated to consult widely with the public, and a serious public
education and consultation process was conducted, the time allotted for con-
sultations was insuf¬cient in a country with little infrastructure and many
isolated communities. Moreover, the draft constitution was prepared in rel-
ative secrecy and released late, and the members of the Constitutional Loya
Jirga had only three weeks to consider it. Perhaps more importantly, Pres-
ident Karzai, a Pashtun, had a strong hand in determining the member-
ship of the drafting committee, and the draft not surprisingly concentrated
power in the president™s hands.93 Karzai and his allies (with support from the
United States) argued that a strong presidency was essential to restore order
in a country riven by factionalism. This approach favored Pashtuns as the
largest ethnic group and revived fears among regional leaders and smaller

90 Zia-Zari¬, supra note 87, at 9.
91 Afghan Constitution, Ch. 1, Art. 6, available at http://www.unama-afg.org/docs/Docs.htm
(last accessed January 28, 2006).
92 Chapter 2, Article 1 of the constitution provides: “Any kind of discrimination and distinction
between citizens of Afghanistan shall be forbidden. The citizens of Afghanistan, man and
woman, have equal rights and duties before the law.”
93 See International Crisis Group, Afghanistan: The Constitutional Loya Jirga, December 12,
2003, at 1.

ethnic groups of renewed Pashtun domination. Ultimately, a compromise
was reached, giving the legislature the power to approve key presidential
appointments but otherwise retaining a powerful executive.
The draft constitution favored Pashtuns in other respects as well. Pashto
and Dari, the languages of the largest ethnic groups, were listed as the two
of¬cial languages, and apart from general language about proportional rep-
resentation by region, the draft contained few provisions designed to ensure
that ethnic group interests were accommodated. Indeed, the constitution as
adopted forbids the formation of political parties on ethnic, linguistic, or
regional bases.94 Up to a point, this approach makes sense. Overt organiza-
tion or distribution of political power on ethnic lines can aggravate rather
than ameliorate ethnic divisions. At the same time, integrative provisions
could have been included to ensure that minority group interests are taken
into account in governance. The Constitutional Loya Jirga almost collapsed
over the national language issue, until a last-minute compromise yielded a
provision specifying that other minority group languages could be designated
as national languages in regions in which the speakers of that language were
in a majority. The exclusion of ethnicity as a basis for political organization
appears to re¬‚ect the political self-interest of the Karzai government but will
likely pose long-term problems in a country fragmented in part on ethnic
A related problem is the relation between the center and the provinces
in Afghanistan. Many regional and ethnic group leaders favored a fed-
eral structure or at least substantial devolution of power. The constitution
instead focuses on the central government and its powers, in signi¬cant
part because the Karzai government was reluctant “to formalise a situa-
tion in which regional administrations . . . retain signi¬cant independence.”96
Although this approach has the advantage of lessening the prospects of
regional and ethnic division, it also heightens the prospect of a too-powerful
and perhaps ultimately authoritarian presidency.
The role of religion in governance proved another contentious issue during
the Constitutional Loya Jirga. The inability to resolve differences between
religious hardliners and reformers produced constitutional provisions that
appear at odds with each other. The constitution mandates respect for inter-
national human rights, including religious freedom and equality for women.
But it also precludes political parties with a program that “contravene[s] the
Holy religion of Islam”97 and provides that “no law shall contravene the
tenets and provisions of the holy religion of Islam.”98 The latter provision

94 Afghan Constitution, Ch. 2, Art. 14.
95 See Afghanistan: The Constitutional Loya Jirga, supra note 93, at 5.
96 Id.
97 Afghan Constitution, Ch. 2, Art. 14.
98 Id., Ch. 1, Art. 3.

in particular may open the door for Afghanistan™s conservative judiciary to
invalidate laws incompatible with a strict interpretation of Shari™a. In an
apparently deliberate ambiguity, the constitution does not specify the out-
come of a con¬‚ict between Islamic law and human rights.
If Afghanistan™s past constitutional history is a guide, these possible ¬‚aws
may pale in signi¬cance in the face of a larger problem. Afghanistan™s numer-
ous previous constitutions (seven since 1923) either concentrated near total
power in the head of state or “were products of governments that never exer-
cised signi¬cant control of the Afghan hinterland, and that between 1977
and 1992 were bogged down in ¬erce and unsuccessful counter-insurgency
efforts.”99 The September 2005 parliamentary elections sent mixed signals
concerning Afghanistan™s future. The elections went relatively smoothly, and
although voter turnout was lower than in the 2004 presidential elections,
represented the culmination of the Bonn transitional process. But weak polit-
ical parties combined with ethnic, sectarian, and other divides to produce a
National Assembly with little coherence, no clear mandate, and little capac-
ity to balance executive power.100 Unless the Afghan transition can achieve
greater momentum, the country may ¬nd itself with a paper democracy in
which central government institutions are weak, fractious, and unable to
exert much positive in¬‚uence over large sections of the country.
The present level of international commitment seems inadequate to chal-
lenge fundamentally Afghanistan™s centrifugal forces. Although the govern-
ment has registered some successes, Afghanistan™s warlords continue to be a
major problem in some areas; the Taliban remains a threat; reconstruction
efforts are moving at a glacial pace; the efforts to build a national army and
police will take years to have a lasting impact; and meanwhile, the opium
trade, with its corrupting in¬‚uence, is booming. The blueprint adopted at
Bonn, though reasonably sound, requires much stronger international sup-
port than has yet been provided. The Bush Administration and other interna-
tional actors routinely proclaim that failure is not an option in Afghanistan.
But absent the political will to provide greater security and reconstruction
assistance over the long haul, success is by no means assured.

B. Iraq: Braking Constitutionalism
Iraq™s history, geography, size, political makeup, and religious and ethnic
divisions render it an enormously dif¬cult challenge for a democratic rule
of law makeover. Indeed, Iraq “has all the characteristics that have impeded
99 International Crisis Group, Afghanistan™s Flawed Constitutional Process, June 12, 2003,
at 2.
100 For a detailed look at the divisions in National Assembly following the voting, see
Andrew Wilder, A House Divided? Analyzing the 2005 Afghan Elections, available at

democratic transitions elsewhere: a large, impoverished population deeply
divided along ethnic and religious lines; no previous experience with democ-
racy; and a track record of maintaining stability only under the grip of
a strongly autocratic government.”101 Unlike Afghanistan, however, Iraq
is at least a modern state, with a “population [that] is largely educated,
sophisticated, and urban,” a previously functioning government and bureau-
cracy, and “little history of inter-ethnic or communal violence.”102 Moreover,
although Iraq is currently burdened with a large foreign debt and even larger
reparations obligations stemming from Iraq™s 1990 invasion of Kuwait, Iraq
also possesses the world™s second-largest oil reserves and, despite a crum-
bling infrastructure, at least the future possibility of economic strength. But
even with these advantages, Iraq poses a more dif¬cult challenge than any
of the cases previously discussed.
The second Gulf War ended just weeks after it began on March 19, 2003.
The political and security vacuum created by the rapid collapse of Iraq™s
government and military left the United States and the United Kingdom, as
occupying powers, with enormous and interrelated challenges. They had to
provide security, restore basic services, organize a transitional administra-
tion, and devise a strategy for the ultimate transfer of power back to Iraqis.
Moreover, they had to do so in a deeply divided country, with competing
groups espousing incompatible visions of the country™s future. Minority Arab
Sunnis, approximately 20 percent of the population but long dominant polit-
ically and economically, feared the loss of power and privilege democratic
rule might bring. Iraqi Shiites, perhaps 55 percent of the population, wel-
comed the opportunity to assume a role in governance commensurate with
their numbers but were divided themselves on what a future Iraq should
look like and how closely it should be aligned with a theocratic Iran. Eth-
nic Kurds, perhaps another 20 percent of the population, were prepared to
remain part of Iraq only if they could count on autonomy at least as com-
plete as currently exercised in northern Iraq. Other fault lines, within and
across each group (including tribal, family, and economic ties), and the var-
ied pattern of geographic concentration and intermingling of these groups,
rendered the task of forging a cohesive, democratic whole mind-numbingly
Yet before the war, the United States had bold plans for the democratic
transformation and economic reconstruction of Iraq. In the weeks leading
up to the war, “[t]he plans presented . . . exuded con¬dence that the United
States had the capacity not only to replace Iraqi President Saddam Hussein™s
101 Marina Ottaway, From Victory to Success: One Country, Two Plans, 135 Foreign Policy
55 (2003).
102 Center for Strategic and International Studies, A Wiser Peace: An Action Strategy for a
Post-Con¬‚ict Iraq, at 10 (January 2003).

regime but to alter the character of the state and the very social fabric of
Iraq.”103 The U.S. forces and contractors would “vet the Iraqi civil service,”
“create and train a Baath-free military and police force,” rapidly restore basic
services, and rehabilitate Iraq™s physical infrastructure.104 The blueprint for
political reconstruction was even more ambitious. The United States would
appoint a transitional U.S.-led administration (¬rst military, later civilian)
to run the country, and it would quickly transform Iraq “from a central-
ized, hierarchical country into a model of participatory democracy.”105
A new constitution would be drafted, one in which the national govern-
ment would “be limited to essential national functions, such as defense and
security, monetary and ¬scal matters, justice, foreign affairs, and strategic
interests such as oil and gas.”106 The rest would be left to local govern-
ment, which would be “required to operate in an open, transparent, and
accountable manner.”107 The transfer of sovereignty to an elected govern-
ment would occur only after a constitution had been drafted and rati¬ed by
Unfortunately, this plan for Iraq rested on a set of assumptions having no
grounding in local conditions. The United States thought it could remove
Saddam Hussein and the most senior Iraqi of¬cials, but that most of the
government bureaucracy would remain in place and continue to operate.109
The United States also expected that organized resistance would end quickly.
In fact, the Iraqi government evaporated almost overnight. Occupying forces
encountered sustained opposition, and efforts to restore basic services and
provide minimal security ¬‚oundered. As a result, the United States soon
began talking about “a short, light-handed occupation and the swift transfer
of power to an Iraqi interim authority.”110
Unlike many of the other cases considered, the con¬‚ict in Iraq did not
end in a peace agreement or a compact among warring parties, leading natu-
rally to the formation of an interim government. Nor was there a victorious
internal faction prepared to assume power. The United States might have
attempted to pull together a broad cross-section of Iraqi political leaders
to form an interim government, along the lines of the All Liberia Confer-
ence that created Liberia™s 1990 Interim Government of National Unity. But
years of repression by the Baath party left little in the way of credible politi-
cal leaders or civil society organizations to participate in such a conference.
103 Ottaway, supra note 101, at 56.
104 Id.
105 Id.
106 Id. (quoting a statement from USAID).
107 Id. (quoting a statement from USAID).
108 The term sovereignty is employed here in a political rather than a legal sense.
109 Celeste Ward, USIP Report, The Coalition Provisional Authority™s Experience with Gover-
nance in Iraq Lessons Identi¬ed, May 2005, at 2.
110 Ottaway, supra note 101, at 56.

As a coalition of Iraqi opposition ¬gures accurately predicted in a report
prepared as part of a State Department planning project in November 2002:

As a result [of the oppressive rule of Saddam Hussein], there are no recognized
domestic political institutions, groups or individuals that can step forward, invoke
national legitimacy and assume power. . . . Many groups and individuals will even-
tually emerge and compete for power, but this will only happen gradually, as the
environment becomes safe for public participation.111

The United States recognized that “premature elections risked empower-
ing forces seen as most hostile to it “ the Islamists and remnants of the old
regime.”112 Appointing an interim government dominated by Iraqi exiles,
as some in the Pentagon wished to do, would have been equally problem-
atic, because the exiles lacked any genuine internal constituency.113 Thus, the
United States understandably wanted to allow time for moderate political
forces to emerge before surrendering signi¬cant power to Iraqis. Accord-
ingly, the United States decided to continue to exercise plenary executive
and legislative powers, notwithstanding its rhetoric about a “light-handed
occupation,” until such time as a national constitution could be drafted and
a new government elected, a process expected to last at least two years.
With this in mind, and after a brief experiment with military administration,
Washington established the Coalition Provisional Authority (CPA) with Paul
Bremer as the lead civilian administrator and de facto proconsul.
The CPA™s work was hampered from the start by inadequate security,
communications and logistical problems, poor planning, lack of knowledge
of Iraq and Iraqis, insuf¬cient advance preparation, chaotic local condi-
tions, and a substantial legitimacy de¬cit. Ill-conceived decrees (especially the
abrupt dismantling of the Iraqi army and clumsy efforts at de-Baathi¬cation
of the civil service) poorly communicated to the people of Iraq made mat-
ters worse. As local expectations of rapid progress went unmet, the United
States came under growing pressure, from Iraqis, the UN, and the inter-
national community, to increase Iraqi responsibility for governance and to
establish a legitimate interim Iraqi authority.
In July 2003, the CPA responded by creating the twenty-¬ve-member Iraqi
Interim Governing Council (IGC). In choosing council members, the CPA
was driven by con¬‚icting imperatives. On the one hand, the United States
111 Democratic Principles Working Group, Final Report on the Transition to Democracy in
Iraq, November 2002, at 16, quoted in Thomas R. Pickering, James R. Schlesinger, & Eric
P. Schwartz, Report of an Independent Task Force Sponsored by the Council on Foreign
Relations, Iraq: The Day After, April 16, 2003, at 28.
112 International Crisis Group, Governing Iraq, August 25, 2003, at 10, available at
http://www.crisisweb.org//library/documents/report archive/A401098 25082003.pdf. Early
elections in post-intervention Bosnia, which entrenched intransigent nationalists, offered a
clear lesson in the dangers of early elections.
113 Id.

wanted to “develop an interim authority that would have legitimacy in Iraq
and abroad, appease the population and de¬‚ect criticism of the occupation
forces.”114 On the other hand, the United States did not want to empower
groups overtly hostile to U.S. political goals in Iraq.115 In an effort to strike an
appropriate balance, the CPA “consulted broadly not only with Iraqi politi-
cal and social forces, but also with the UK and the once-suspect UN.”116 At
the same time, “both the CPA and major Iraqi political groups” were given
a tacit veto over the Governing Council™s ¬nal membership. The result was
“a broadly diverse body, including various strands of Iraqi society “ Islamist
and secular, modern and traditional, old notable families and tribes.”117 But
it was also an unelected body that consisted principally of “political lead-
ers with weak popular followings, very little in common between them, no
bureaucratic apparatus and a clumsy nine-person rotating presidency.”118
Worse, in an understandable and in some respects laudable effort to re¬‚ect
Iraq™s diverse population, it was a body whose members were chosen in sig-
ni¬cant part on the basis of ethnic and religious af¬liations, leading observers
to warn that “[e]thnic and religious con¬‚ict, for the most part absent from
Iraq™s modern history, is likely to be exacerbated as its people increasingly
organize along these divisive lines.”119 Not surprisingly, most Iraqis per-
ceived the Governing Council as a tool of the United States, and insofar as
Bremer retained ¬nal executive and legislative authority, they were not far
But as attacks on coalition forces mounted, some Bush Administration
of¬cials pushed for a quick transfer of power “to lessen the U.S. footprint
and diminish popular resistance to the occupation.”120 They were joined
by disgruntled Iraqis of “all shades,” and by “governments that opposed
the U.S. war in Iraq such as France and Germany. . . .”121 In October 2003,
the Security Council, in Resolution 1511, urged the CPA “to return govern-
ing responsibilities and authorities to the people of Iraq as soon as prac-
ticable” and “invited” the Governing Council to produce a timetable and
program for drafting a constitution and holding elections under it “no later
than 15 December 2003.” Yielding to the pressure, the CPA agreed with the
Governing Council to a phased transition involving the adoption of interim
governing principles, the selection of a constitutional convention, and even-
tual election of a government to assume power under a new constitution.

114 Id., at ii.
115 For an insightful discussion of U.S. legitimacy problems in Iraq, caused in part by U.S. efforts
to maintain control of the political process there, see Larry Diamond, What Went Wrong
in Iraq?, 83 Foreign Affairs 34 (2004).
116 Governing Iraq, supra note 112, at 12.
117 Id.
118 Id., at ii.
119 Id.
120 International Crisis Group, Iraq™s Constitutional Challenge, November 13, 2003, at 7.
121 Id., at 6.

But the CPA™s plans, however well intentioned, did not dovetail with
the aspirations of much of Iraq™s population. A late June 2003 edict issued
by Grand Ayatollah Ali Sistani, the most respected Iraqi Shiite cleric, urged
Iraqis to press the coalition for early general elections. Direct elections would
favor the majority Shiites and give them a decisive in¬‚uence on the drafting
of a constitution and the future governance of Iraq. Sistani™s stance helped
force the CPA to abandon its original strategy of holding elections after a
constitution was put in place. Instead, the Iraqi Governing Council, with
the CPA™s acquiescence, invited Lakhdar Brahimi, the Special Advisor to the
UN Secretary-General, to consult with Iraqi leaders on the feasibility of con-
ducting early elections and on possible candidates for senior positions in a
new Iraqi interim government. On March 8, 2004, after extensive negoti-
ations, the Iraqi Governing Council adopted a Transitional Administrative
Law (TAL) designed to set out the basics of a future constitution and lay
the groundwork for the election of a new government by the end of January
2005. The new government, selected on the basis of Brahimi™s recommen-
dations, negotiations among the competing parties, and vigorous (some say
heavy-handed) U.S. input, assumed power on June 28, 2005.
The TAL, and the organization of the interim government, represented
compromises among the competing interests of Iraq™s principal ethnic and
sectarian groups. The central fault line involved the nature of federalism in
Iraq and the extent of autonomy to be given to Iraqi Kurds. In theory, vir-
tually all parties in Iraq rejected the idea that the state should be structured
along ethnic or sectarian lines. They were aware of the dif¬culties that have
frustrated similar efforts in places such as Lebanon, Cyprus, and Bosnia.
In practice, however, Iraqi Kurds insisted that only near total autonomy
could protect their interests. Although they were willing “to cede matters of
foreign, monetary and national defense policy to the Iraqi national govern-
ment,” they were adamant about retaining most of their current autonomy
and governing structure.122 In response, the TAL insisted that government
in Iraq shall be “republican, federal, democratic, and pluralistic” and that
federalism “shall be based upon geographic and historical realities and the
separation of powers, and not upon origin, race, ethnicity, nationality, or
confession.” At the same time, the TAL authorized any three governorates
in Iraq to amalgamate as a political unit and to block adoption of a perma-
nent constitution.
The TAL was problematic in various respects. In particular, it empha-
sized Iraq™s ethnic and sectarian divisions. But it nonetheless offered a set of
principles that might be adapted to a new constitution, which in turn might
serve as a vehicle for resisting the centrifugal forces that threatened to tear
Iraq apart. Many hoped that the process of drafting a constitution would

122 Edward Wong, Governing Council Parties Are Said to Back Broad Autonomy for Kurds,
The New York Times, January 10, 2004, at A6.

help heal some of Iraq™s divisions and pave the way for the emergence of
a national consensus on the future of the state. Unfortunately, the process
operated instead to exacerbate Iraq™s divisions,123 and the adoption of the
constitution, whatever its merits viewed in the abstract, may yet come to be
seen as a turning point on the road to dissolution and possibly civil war.
The problems with Iraq™s constitutional process began when Sunni Arabs
largely boycotted elections to the Transitional National Assembly (TNA) in
January 2005. Although ¬fteen Sunni Arabs were added to the ¬fty-¬ve exist-
ing members of the Constitutional Committee in July, they had little in¬‚u-
ence over the course of the subsequent negotiations and were often wholly
excluded from key discussions between Kurds and Shiites. The Sunni negotia-
tors charged that their Kurdish and Shiite counterparts refused to accommo-
date core Sunni interests on key issues, including federalism, Iraq™s national
identity, and de-Baathi¬cation; Shiite and Kurdish negotiators viewed the
Sunnis as obstructionist. When the Committee failed to reach consensus by
August 15, the TAL deadline for submitting a draft to the TNA, the drafters
secured a one-week extension and then an additional three-day extension.
When that proved insuf¬cient, negotiations continued, without any clear
legal basis.124 The delays stemmed from belated efforts to devise last-minute
compromises that would attract Sunni support, but when that failed, the
draft was approved by the TNA over the strong objections of the Sunni
negotiators. The TNA could have taken an additional six months under the
TAL, but the United States and some interim government of¬cials insisted
on adherence to the original timetable.
As the October 15 referendum drew near, and concerns over Sunni oppo-
sition grew, the TNA approved modest amendments to the constitutional
draft that would, among other things, permit changes to the constitution to
be adopted by the parliament even after the referendum. But the changes
were too little, too late for most Sunnis, who voted against adoption of the
constitution in large numbers. For Sunnis, the constitution represented a
bargain between Shiites and Kurds; the marginalization of Sunnis during the
drafting process con¬rmed their fears that even the most basic Sunni inter-
ests would not be protected in the new order. As a result, the constitution-
formation process, far from healing rifts among Iraqis, exacerbated ethnic
and sectarian divisions.
The process problems were matched by problems in constitutional sub-
stance. Although Kurds have long been persecuted in Iraq, and Shiites denied
political power proportionate to their numbers, ethnic and sectarian divides

123 See, e.g., Jonathan Morrow, Iraq™s Constitutional Process II: An Opportunity Lost, Novem-
ber 2005, available at http://www.usip.org/pubs/specialreports/sr155.pdf (last accessed Jan-
uary 28, 2006).
124 International Crisis Group, Unmaking Iraq: A Constitutional Process Gone Awry, Septem-
ber 26, 2005, at 2.

have not in the past dominated Iraqi politics to the extent that would
render a Bosnian-style solution inevitable. Nonetheless, the constitution,
though formally eschewing ethnic federalism, permits both Kurdish and
Shiite governorates to amalgamate into autonomous regions, with control
over most aspects of daily life within them. Sunnis feared that allowing the
three Kurdish provinces to form a region in the oil-rich north and the nine
or so Shiite provinces to form a region in the oil-rich south would pave the
way for the partition of Iraq, with Sunnis left in charge of a resource-poor
center. Even if Iraq does not break apart, Sunnis fear that ambiguous
language relating to the disposition of Iraq™s oil resources might be inter-
preted by a Shiite-dominated government in ways that would leave Sunnis
with little. Sunnis also objected to provisions on national identity and de-
Baathi¬cation, among others, fearing they were further steps to marginalize
the Sunnis.
In both process and substance, Iraq™s new constitution re¬‚ects a “growing
ethno-sectarianism in which Iraqis identify strictly with their own preferred,
self-de¬ned community and interpret events exclusively through an ethno-
sectarian lens.”125 This trend is extraordinarily dangerous. Geography in
Iraq does not lend itself to ethnic federalism as easily as might be supposed.
Although northern Iraq is primarily Kurdish, whereas central Iraq is pre-
dominantly Sunni and southern Iraq largely Shiite, the reality is that all
three groups live interspersed to a signi¬cant degree throughout Iraq. Views
concerning the borders of these regions are heavily colored by economic and
strategic interests, especially the location of contested oil ¬elds. As a result,
any effort to divide Iraq along ethnic lines is likely to deteriorate quickly.
Discussions between Kurds, Sunnis, and Shiites on the possibility of a
coalition government in early 2006 hold out at least the possibility of com-
promise and cooperation among Iraq™s contending communities. But in the
absence of any fundamental consensus on the future of the state or the legit-
imacy of its new constitution, it remains an open question whether Iraq will
slide into full-scale civil war.


What should be evident from this chapter™s review of post-con¬‚ict blueprints
is that no general template exists and that every approach carries substantial
risks. At the same time, the nature and extent of the risks vary with the
type of con¬‚ict and the extent to which interveners are prepared to insist
on a workable blueprint. Identity-based con¬‚icts present greater political
engineering challenges than resource-driven con¬‚icts, but blueprints for both
125 Id., at 11.

will fail if interveners place higher priority on satisfaction of short-term
con¬‚ict termination interests over long-term political settlement interests.
r The process of designing a blueprint may be as important as the substance
of the blueprint. A poorly managed process will exacerbate rather than
mitigate the internal divisions that produce con¬‚ict in the ¬rst place. The
problem is particularly acute when it comes to designing post-con¬‚ict
r Post-con¬‚ict blueprints are not the product of a single, coherent design.
Instead, they emerge through a process of bargaining among actors with
widely varying goals, sharply differing degrees of commitment, and lim-
ited information. The substantive viability of a blueprint will therefore
vary depending on the dif¬culty of the case, the degree to which inter-
veners are prepared to insist on a workable design, and the willingness
of the principal local actors to cooperate. Too often, the temptation is to
pursue a quick end to the ¬ghting through unstable accommodations of
powerful local actors.
r When a state is divided against itself, and its population lacks a shared
commitment to a single national identity, blueprint options are limited
and unattractive. Unless one party to the con¬‚ict wins outright (as in
Rwanda), or can be separated from other parties through autonomy (as in
Kosovo) or independence (as in East Timor), interveners must ¬nd ways
to promote power-sharing among the contending groups. This presents
an almost irresolvable dilemma: consociationalist blueprint provisions
suf¬cient to insulate minorities from the threat of majoritarian power
perpetuate the differences that precipitated con¬‚ict in the ¬rst place and
render government at the center cumbersome or unworkable. But more
limited minority rights measures more conducive to effective governance
are likely to be rejected as insuf¬cient by minority groups whose coop-
eration is essential to the construction of a viable, rule of law oriented
r Blueprints for resource-driven con¬‚icts and other instances of state failure
that do not center on state identity are easier to create and implement.
But the temptation in such cases is to rush to an early electoral exit,
before combatants are disarmed and viable state institutions and norma-
tive commitments to democracy and the rule of law are in place. Elections
by themselves are not a blueprint for success.
The better that interveners understand the risks of different blueprint
options, the easier it will be to avoid the pitfalls associated with each, and
the easier it may be for interveners to resist seemingly attractive short-term
options with disastrous long-term consequences.
Post-con¬‚ict blueprints provide the foundation, for better or worse, for
subsequent efforts to build and strengthen the rule of law. Without such a

foundation, interveners and local reformers will lack a framework for tran-
sitioning to effective governance, for building justice systems, for providing
security for the population, and for working to strengthen and reinforce
cultural commitments to the rule of law. Subsequent chapters examine in
detail a number of speci¬c aspects and challenges of blueprint implementa-
tion. The next chapter, in particular, will examine the enormous challenge
of establishing security as the essential precondition “ the sine qua non “ for
building the rule of law after con¬‚ict.

Security as Sine Qua Non

As George Tanham, an international security and counterinsurgency expert,
wrote of Vietnam, “[s]trange as it may seem, the military victory is the easiest
part of the struggle. After this has been attained, the real challenge begins:
the reestablishment of a secure environment opens a new opportunity for
nation building.”1 Tanham™s observation could just as easily be applied to
the cases studied in this book. The terminology has evolved “ the United
States, in particular, now prefers “stability operations” to “nation build-
ing.” But Tanham™s point still holds. Military intervention marks only the
¬rst, and usually the simplest, phase of the much larger and more com-
plex task of restructuring the governing institutions of the affected state and
encouraging the ascendance of actors and social norms capable of making
those institutions successful.
As Tanham suggests, the reestablishment of a secure environment is “the
sine qua non of post-con¬‚ict reconstruction.”2 Absent basic security, efforts
to reform political institutions, adopt new laws, promote national reconcil-
iation, and jump-start economic growth are destined to fail. In most cases,
however, military victory does not, as Tanham seemed to assume, correlate
directly with the establishment of the secure environment that in turn “opens
a new opportunity for nation building.” In fact, most of the cases studied in
this book do not entail a clear military victory. In many cases, interveners
used force selectively against one or more parties to the con¬‚ict, but their
primary goal was to compel a negotiated settlement rather than to achieve an
outright military victory. Even when interveners did seek and secure a clear

1 George K. Tanham, War without Guns: American Civilians in Rural Vietnam (1966),
at 138, quoted in Erwin Schmidl, Police Functions in Peace Operations: An Histori-
cal Overview, in Policing the New World Disorder: Peace Operations and Public
Security (Robert Oakley, Michael Dziedzic, & Eliot Goldberg eds., 1998), at 19, 35.
2 Center for Strategic and International Studies and the Association of the U.S. Army, Play to
Win: Final Report of the bi-partisan Commission on Post-Con¬‚ict Reconstruction, January,


military victory, as in Afghanistan and Iraq, “winning” marked only a shift
from major combat operations against ¬elded forces to counterinsurgency
campaigns and low-intensity con¬‚ict against hit-and-run enemies.
In almost all cases, then, post-intervention security efforts take place in
a demanding and often hostile environment. Internal warring factions typ-
ically retain considerable power and weaponry and pose a danger to the
larger blueprint implementation process even when not engaged in open
con¬‚ict with interveners and their local allies. Moreover, in the immediate
post-intervention period, internal security forces, to the extent they remain
intact, are more likely to be part of the problem than part of the solution.
In most of the cases studied, the pre-intervention security and police forces
of the state operated primarily to impose or support the ruling elite™s hold
on power (or, in the case of Afghanistan, its draconian social vision); far
from following the idealized serve-and-protect mandate familiar to viewers
of U.S. television police dramas, these security forces terrorized regime oppo-
nents, persecuted minorities, and instilled lasting distrust of state authority
in the general population. Such forces can scarcely be relied on to maintain
order in a manner consistent with respect for the rule of law. In a few cases,
as in Somalia, local police may retain considerable popular support. But
even then, the police cannot maintain order in a vacuum; their efforts must
be supported by well-functioning courts and prisons and by a government
minimally capable of providing its population with basic services.
In general, for post-con¬‚ict security efforts to succeed, four conditions
must be met. First, security cannot depend solely or even primarily on coer-
cion. Force or the threat of force is often essential to the maintenance of
order, particularly in the ¬rst days of an intervention. But in the long term,
public order, at least outside of a police state, rests on a societal consensus
about the legitimacy of state institutions and con¬dence in the capacity of
such institutions to deliver basic services. Accordingly, long-term progress
on security depends on and must be matched by progress in political and
economic reconstruction. Even a government that protects its people from
attack cannot function effectively, at least not from a rule of law standpoint, if
the same people are starving, freezing, or dying of easily preventable disease.
To achieve lasting security, interveners and their local allies must therefore
do more than restore order and protect civilians from crime. They must also
provide minimally acceptable levels of basic public goods, including func-
tioning infrastructure (such as power, water, sewage treatment, health care,
telecommunications, and the like) and basic humanitarian assistance (such
as food and medicine) until the local governing institutions and economy
can be restored suf¬ciently to provide those goods on their own. Otherwise,
spoilers will ¬‚ourish, public support for the interveners and their reforms will
dwindle, local elites cooperating with the interveners will be discredited, and
demobilized ¬ghters will likely return to their former livelihood.

Second, security requires from interveners a mix of capabilities signi¬-
cantly different from the war-¬ghting mix required for an initially successful
military intervention. Winning wars and maintaining order are two very dif-
ferent tasks. The mix of forces needed for the latter will vary widely from
one country to another, depending on the status and capabilities of local
security forces and the nature and extent of post-intervention political vio-
lence and criminal activity. In some cases, the appropriate force mix will vary
widely in different locations and at different times within the same country,
as the intensity of opposition to the establishment of a new political order
waxes and wanes. In general, however, the demand for security is particu-
larly acute in the immediate post-intervention period, when local belligerents
retain their combat capabilities and indigenous security institutions are in
disarray; at that point, only the military can provide order. As the situation
stabilizes, different skills are required, and interveners must deploy some mix
of international civilian police and indigenous police.
Third, efforts to promote security must be part of and subordinate to
the larger peace process. When spoilers threaten to derail the peace pro-
cess through violence, interveners cannot remain neutral. Instead, early and
vigorous opposition to spoilers may prove essential to building a lasting
peace. At the same time, decisions on whether and when to combat spoilers
remain political decisions, to be taken by those managing the overall peace
process. Premature attempts to disarm a warring faction, for example, may
undermine or derail diplomatic efforts designed to engage key local actors
in support of an evolving peace process.3 In short, security is part of but not
a substitute for efforts to implement the overall blueprint for post-con¬‚ict
Fourth, interveners must work with local actors from the beginning to
rebuild indigenous security institutions. Even the most determined interven-
ers typically seek to leave as quickly as local conditions permit, and only
the establishment of an effective indigenous security capability offers inter-
veners a legitimate exit option. Moreover, although interveners can impose
order temporarily, interveners lack the knowledge of local laws, customs,
and culture required for effective policing over the long term. Ideally, local
actors will participate fully in all decision-making pertaining to security from
the outset and assume at every stage of the post-con¬‚ict period as much of
the security responsibilities for their state as circumstances permit. Involving
local actors can help interveners adapt security practices to conform to local
norms and simultaneously build capacity for the eventual full transfer of
security responsibilities to local forces. Moreover, fostering local ownership
of security measures will contribute to the social consensus that underpins

3 See Michael Dziedzic & Benjamin Lovelock, An Evolving “Post-Con¬‚ict” Role for the Mil-
itary: Providing a Secure Environment and Supporting the Rule of Law, in Post-Con¬‚ict
Justice (M. Cherif Bassiouni, ed., 2002), at 851, 851.

order in democratic societies. At the same time, great care must be taken
to avoid empowering political elites whose interests are not consistent with
long-term rule of law objectives.
Meeting the challenge of providing security is a complex, costly, and time-
consuming task. In pursuing security, the synergistic approach outlined in
Chapter 3 should be kept in mind. Too often, interveners focus on numeric
outputs “ numbers of police deployed, judges trained, and spoilers killed or
incarcerated “ without articulating adequately the overall strategic objec-
tives sought and ensuring that resources are deployed in a coherent and
mutually reinforcing way. Similarly, the immediate demands of a hostile
post-intervention environment often distract interveners from focusing ade-
quately on how to adapt western models of policing and law enforcement
to local conditions and norms, even though security institutions that lack
popular legitimacy are unlikely to endure. Finally, interveners often fail to
approach security issues systemically, focusing instead on individual com-
ponents. But rapid progress in one area, such as police reform, will soon
be undercut if judges are corrupt, incompetent, or intimidated or if prison
facilities cannot keep pace with arrests. Just as security must move forward
in tandem with political and economic reform to ensure public order, so
too must the individual components of an effective security system move
forward in a mutually supportive fashion.
Notwithstanding the dif¬culty of achieving basic security in a post-con¬‚ict
environment, substantial progress in restoring public order and rebuild-
ing basic rule of law institutions has been achieved in several post-con¬‚ict
societies, including Kosovo, East Timor, Bosnia, and Sierra Leone. Although
the establishment of a secure environment does not guarantee success, the
inability to establish a secure environment does guarantee failure.


The provision of order is the ¬rst task of any government. No government,
least of all one committed to the rule of law, can function effectively if its
people cannot go about their daily life without fear of being shot, tortured,
raped, robbed, or bombed. Unfortunately, post-intervention conditions ren-
der the task of ensuring physical security extremely dif¬cult. The task is
doubly dif¬cult in states such as Iraq and Afghanistan, where armed con¬‚ict
continues even after interveners secure their initial military victory.

A. Causes and Consequences of Government Collapse
In all of the cases considered in this book, with the possible exception of Iraq,
intervention took place in “failed states” or “quasi-states,”4 that is, states
4 For a discussion of “quasi-states,” see Robert H. Jackson, Quasi-States: Sovereignty, Inter-
national Relations and the Third World (1993).

that lacked not only effective governments but also the underlying social and
political cohesion that makes effective governance possible without undue
reliance on coercion. States in the industrialized west have “set the standard
for effective statehood . . . by their demonstrated success in simultaneously
meeting the basic needs of the large majority of their populations, protect-
ing their human rights, and promoting and guaranteeing political participa-
tion.”5 But these are states that “have, by and large, successfully completed
their state-building process, are politically satiated and economically af¬‚u-
ent, and possess unconditional legitimacy in the eyes of the overwhelming
majority of their populations.”6
By contrast, many states in which military intervention typically occurs
have not completed the state-building process; their governments cannot
meet the basic needs of most of their population and do not possess broad
and unquestioned legitimacy. These are states with ineffective governments
that have not consolidated their authority over much of the territory of
their state, have not succeeded in maintaining order within the territory
they do control, and have not managed to use state resources to support
security and policing activities that serve the public interest or “to carry on
routine administration, deepen the state™s penetration of society, and serve
symbolic purposes (taxation).”7 Most importantly, these are states that lack
the core identifying characteristic of an effective state: a government with a
monopoly over coercive power. Many lack a sense of common citizenship
or a political culture that dictates peaceful resolution of disputes.8 In these
states, civil society is weak and disorganized, security forces serve regime
rather than state interests, legal codes are outdated and inadequate, and
courts are corrupt, politicized, and ineffective.9 It is not just that law and
political institutions in these states are ineffective; it is that the faith in law
and political institutions that underpins policing and order in effective states
does not exist.

5 Mohammed Ayoob, State Making, State Breaking, and State Failure, in Turbulent Peace
(Chester Crocker, Fen Hampson, & Pamela Aall, eds., 2001), at 127, 133.
6 Id.
7 Id., at 128. Ayoob points out that European states took centuries to form, and argues
that developing countries have lacked both the time and the free coercive hand necessary to
induce or compel “disparate populations under their nominal rule to accept the legitimacy of
state boundaries and institutions” and to tax and otherwise regulate their lives. Id., at 130.
8 See J. ˜Kayode Fayemi, Governing Insecurity in Post-Con¬‚ict States: the Case of Sierra
Leone and Liberia, in Reform and Reconstruction of the Security Sector (Alan Bryden
& Heinder Hanggi, eds., 2004), at 179, 182“183.
9 See, e.g., James Dobbins et al., Rand, America™s Role in Nation-Building: From
Germany to Iraq, September, 2003, at xxviii, available at http://www.rand.org/pubs/
monograph reports/MR1753/MR1753.pref.pdf (noting that after the “rapid and utter col-
lapse of central state authority” in Somalia, Haiti, Kosovo, and Afghanistan, the “local
police, courts, penal services, and militaries were destroyed, disrupted, disbanded, or dis-
credited and consequently unable to ¬ll the post-con¬‚ict security gap”).

When state institutions are this weak, opportunistic elements in society
are quick to take advantage. Warlords and political entrepreneurs ¬‚ourish
and often ¬nance their private militias through criminal activity, including
traf¬cking in arms and drugs. Simple banditry, “fueled by military deser-
tion, the breakdown of social structures, and demobilization” of government
forces, “is endemic.”10 Ordinary crime also escalates sharply because of a
“combination of economic necessity, social breakdown, the focus of police
forces on political and regime security, and the proliferation of weapons.”11
In some states, a “war economy” takes hold, in which political elites and
con¬‚ict entrepreneurs support continuing violence and political instability
for personal gain.
States that lack the political and social cohesion that stem from a fully
realized state-building process are perpetually vulnerable to government col-
lapse, easily triggered by any shock to the system. Somalia is perhaps the
most obvious example. Somalia never possessed an effective government,
and when the end of the Cold War eliminated U.S. and Soviet incentives
to prop up any particular regime, anarchy quickly followed. But Somalia
was far from unique. Samuel Doe™s government in Liberia disintegrated in
the face of a few hundred ¬ghters; Sierra Leone™s government disappeared
when its military joined forces with the rebels; and Afghanistan™s govern-
ment collapsed in the face of the student-inspired Taliban. Even in states
with apparently effective governments, as in former Yugoslavia and Iraq,
authoritarian rule only masked the weakness of state institutions and an
underlying lack of social cohesion.
The collapse of Somalia™s government in the early 1990s, and the prospect
that much of Somalia™s population might starve in the ensuing con¬‚ict, pre-
cipitated a western military intervention with an unrealistically narrow aim:
to establish a secure environment for the distribution of food. The interven-
ers quickly discovered that security and the provision of food could not be
achieved on a sustainable basis in the absence of viable governing institutions,
and the mission soon morphed into a more ambitious nation-building effort.
But the interveners failed to appreciate the magnitude of the task they faced.
Somalia™s descent into an anarchic maelstrom of warring clans re¬‚ected the
weakness of the state itself, which never commanded a monopoly on coer-
cive power or a coherent social order legitimate in the eyes of the varied
populations inhabiting the state. In this context, super¬cial efforts to impose
order and hold elections held no chance of success, and it was not long
before a chastened United States ¬‚ed Somalia, with the United Nations not
far behind.

10 Id., at 178 (Dobbins et al. make the point in the context of a discussion of Iraq, but the
point holds true in other cases as well.).
11 Id.

What is true of Somalia holds true, albeit to a lesser degree, of the other
cases under consideration. Restoring sustainable order in the aftermath of
intervention in a failed or failing state requires not just the establishment
of the security institutions of an effective state; it requires building the
state itself. The post“World War II reconstruction efforts in Germany and
Japan, the only two unambiguously successful instances of large-scale post-
war reconstruction, had the advantage of taking place in effective states in
which underlying questions about statehood had already long since been
addressed. Although order can be established in contemporary failed or fail-
ing states, it is likely to be transitory unless underlying de¬ciencies in state
structures are also addressed. Thus, in Haiti, for example, the 1994 U.S.-
led restoration of Jean Bertrand Aristide to of¬ce following his ouster by
the military in 1991, and the follow-on UN peacekeeping and peacebuilding
efforts, succeeded only temporarily in restoring order. The new and vetted
police put back on the streets gradually melted away or ceased to matter,
because there was no real change in the underlying political culture and little
in the way of effective or politically legitimate state institutions. As a result,
in February 2004, Aristide was once again forced to ¬‚ee the country by rebel
forces with poor human rights records and in the face of general popular
Attempting to build a viable political and legal order in a state that has
never completed an effective state-building process is a daunting challenge. In
the west, particularly in Europe, the process of building effective states lasted
many years and, in most cases, centuries. Yet, as Charles Call points out,
“in today™s postcon¬‚ict societies, international actors are attempting to take
shortcuts through those historical processes, creating public police forces
and revamped judicial systems in a few months, while external resources are
supplanting internal tax revenues and globalized communications are trans-
mitting new ideas and expectations.”12 It does not follow that interveners in
such states face a hopeless task. Order can be restored. And the restoration
of order can pave the way for a successful political reconstruction effort. But
the process is long, arduous, and expensive.

B. Seeing Security in Context
Because sustainable security requires at least minimally functioning state
institutions, security efforts are meaningful only if undertaken as part of
a larger post-con¬‚ict reconstruction and rule of law project. That larger
post-con¬‚ict project is now often conceived of as involving four distinct but
interdependent and related tasks, commonly referred to as the four “pillars”:

12 Charles T. Call, Introduction: What We Know and Don™t Know about Post-Con¬‚ict Justice
and Security Reform, in Constructing Justice and Security After War (Charles T. Call,
ed., 2006), at 9“10.

security, justice and reconciliation, social and economic well-being, and gov-
ernance and participation. In this framework, security is de¬ned to include
“all aspects of public safety, in particular, creating a safe and secure environ-
ment and developing legitimate and effective security institutions.”13 Secu-
rity includes not just the protection of civilians from violence, but also the
protection of the territorial integrity of the state.14 Although security forms
the foundation on which all other reconstruction tasks must build, efforts
at establishing security in turn depend critically on progress with the other
The justice and reconciliation pillar “addresses the need to deal with past
abuses through formal and informal mechanisms for resolving grievances
arising from con¬‚ict and to create an impartial and accountable legal sys-
tem for the future. . . .”15 Although Chapter 7 of this book examines the
challenges of transitional justice, it is worth noting here that there is an inti-
mate but not necessarily straightforward connection between accountabil-
ity and security. Accountability mechanisms may help marginalize or deter
politicians and other actors opposed to the post-con¬‚ict blueprint, mitigate
pressure for violent redress of past grievances, and foster reconciliation in
some cases, but in other circumstances, the pursuit of accountability may
also provoke violent backlash and jeopardize fragile political bargains and
nascent government institutions.
The government and participation pillar “addresses the need to create
legitimate, effective political and administrative institutions and partici-
patory processes, in particular, establishing a representative constitutional
structure, strengthening public-sector management and administration, and
ensuring the active and open participation of civil society in the formulation
of the country™s government and its policies.”16 Inevitably, military interven-
tion takes place in support of political goals. Often, those goals include sup-
port of a negotiated peace settlement or support for an agreed set of political
processes to form a new government. Although “[n]o peace force can com-
pel reconciliation if the power brokers involved are unalterably opposed,”17
most cases involve relatively ¬‚uid situations in which security and progress
on governance are directly linked. If basic security is not ensured, the par-
ties to the con¬‚ict face an internal security dilemma: any steps they take
toward peace, for example, by disarming, may leave them vulnerable to
their opponents. More generally, absent effective security measures, parties

13 John Hamre & Gordon Sullivan, Toward Post-Con¬‚ict Reconstruction, 25 Wash. Q. 85, 91
(2002), available at http://www.twq.com/02autumn/hamre.pdf.
14 Id.
15 Id.
16 Id., at 92.
17 Robert B. Oakley & Michael J. Dziedzic, Conclusions, in Policing the New World
Disorder, supra note 1, at 509, 535.

not committed to the larger peace or reconstruction process can easily sabo-
tage any progress on political and economic reconstruction through violence;
in countries emerging from civil war, spoilers will be free to restart the war if
their demands are not met. Conversely, a lack of progress on achieving politi-
cal institutions broadly acceptable to the most powerful actors in society may
induce those actors to seek power through potentially violent means.
The fourth pillar, social and economic well-being, also correlates closely
with security. Social and economic well-being “addresses fundamental social
and economic needs, in particular, providing emergency relief, restoring
essential services to the population in areas such as health and education,
laying the foundation for a viable economy, and initiating an inclusive and
sustainable development program.”18 At the outset, it entails “protecting
the population from starvation, disease, and the elements”; later, it involves
“long-term social and economic development.”19 Absent progress in this
area, combatants and criminal elements have little reason to forego political
Thus, in addition to deploying military and police forces to maintain
public order, interveners must simultaneously work with local authorities
to deploy the civilian specialists who can restore electricity, rebuild dam-
aged or destroyed basic infrastructure, and jump-start the country™s econ-
omy. In Iraq, General Franks, the commander of coalition ground forces,
and Paul Bremer, head of the Coalition Provisional Authority, argued back
and forth over which took precedence “ security or reconstruction. Bremer
complained that reconstruction could not proceed effectively without better
security; Franks argued that delays in reconstruction undermined efforts to
restore security.20 Both tasks are clearly interrelated and to some extent must
proceed simultaneously.
Overall, though, progress on all four pillars begins with basic security. In
the short term, interveners must control looting, separate antagonists, limit
public violence, and generally maintain a minimally secure environment,
pending the reestablishment of indigenous police and security forces. Unless
and until this happens, everything else gets put on hold. New political parties
cannot take hold, civil society cannot function, and economic activity cannot
¬‚ourish. As Scott Feil notes,
A return to any sense of normalcy depends on the provision of security. Refugees
and internally displaced persons will wait until they feel safe to go home; former
combatants will wait until they feel safe to lay down their arms and reintegrate into
civilian life or a legitimate, restructured military organization; farmers and merchants

18 Hamre & Sullivan, supra note 13, at 91.
19 Id.
20 Michael R. Gordon, The Strategy to Secure Iraq Did Not Foresee a 2nd War, The New York
Times, October 19, 2004, at A1.

will wait until they feel that ¬elds, roads, and markets are safe before engaging in food
production and business activity; and parents will wait until they feel safe to send
their children to school, tend to their families, and seek economic opportunities.21

For the same reasons, an insecure environment “is extremely hostile to long-
term development initiatives.”22 Just as individuals “will think twice about
investing scarce capital into something that is likely to be taken away the next
day by armed bandits,” so too will foreign investors shy away from putting
capital into an economy vulnerable to an imminent renewal of hostilities.23
In a vicious circle, insecurity undermines ordinary political and business
activity in a way that breeds further insecurity.24


Security tasks vary from one post-con¬‚ict environment to another, depending
on the outcome of the con¬‚ict, the status and interests of the belligerents,
the size and geography of the state, the extent to which state institutions
and security forces have disintegrated, local political and economic condi-
tions, the involvement of neighboring states, and the capacity, commitment,
and goals of the interveners. Moreover, security tasks will vary by time and
place. In some states, events may follow a more or less linear trajectory, mov-
ing from combat operations against belligerents to peacekeeping operations
intended to separate warring parties to more traditional police operations
as the post-con¬‚ict situation stabilizes. In other cases, traditional policing in
some areas of the state may coincide with active combat operations in oth-
ers. In many cases, the demand for security may transform from protection
against overt political violence to protection from organized economic crime.
In general, however, security tasks usually entail the following roles: separa-
tion, control, and eventual demobilization of belligerents; protection of the
civilian population; protection of political leaders, especially local partners
in an ongoing peace process; protection of mission participants and the secu-
rity forces themselves; protection of local infrastructure and institutions; and
control of crime and localized violence.25
21 Scott Feil, Building Better Foundations: Security in Post-con¬‚ict Reconstruction, 25 Wash.
Q. 100 (2002).
22 Espen Barthe Eide & Thorstein Bratteland, Norwegian Experiences with U.N. Civilian
Police Operations, in Policing the New World Disorder, supra note 1, at 437, 438.
23 Id., at 438“439.
24 Id., at 439.
25 See, e.g., Feil, supra note 21, at 98“99. Security Council Resolution 1244, which established
an international security force for Kosovo, identi¬ed the relevant security tasks to include:
“Deterring renewed hostilities, maintaining and where necessary enforcing a cease¬re, and
ensuring the withdrawal and preventing the return into Kosovo of Federal and Republic mil-
itary, police and paramilitary forces”; “[d]emilitarizing the Kosovo Liberation Army (KLA)

At the outset of the post-intervention period, the dominant task is usu-
ally to control belligerents and maintain basic order. Rapid progress on this
front is vital, because security efforts at the outset set the tone for the entire
post-con¬‚ict reconstruction effort. Failure to establish a modicum of order
quickly undermines public con¬dence and invites the entrenchment of forces
opposed to the larger peace process or intent on taking personal advantage
of the political and security vacuum created by the collapse of the govern-
ing regime.26 In most cases under study, the initial effort to restore order
has been principally a military task for two reasons. First, the restoration of
order often requires combat capabilities that only the military can provide.
Military forces “are heavily armed, train and deploy as a unit” and “operate
under rules of engagement” suitable to the control of local belligerents.27
Second, the military is already in place when post-con¬‚ict operations begin
and, to the extent specialized military forces not present are needed, they can
be deployed quickly. Recruiting and deploying international civilian police,
and vetting and restoring indigenous military and police forces, takes much
longer, creating a “deployment gap” that usually only the military can ¬ll.28
At the same time, the military is generally not well prepared for the
demands of a major post-con¬‚ict security role. Military forces are trained
and equipped principally to ¬ght and win wars. Maintaining the peace in
the aftermath of war requires a different mix of forces, different doctrine,
different training, and different equipment than ¬ghting the war itself. The
U.S. military in particular has been slow to adjust to the different demands
of the post-con¬‚ict security role, in large part because of the reluctance of
military leaders to let the military get drawn too deeply into nation-building
activities. Although considerable progress has been made in recent years,
much remains to be done, including greater attention to constabulary forces
and improved training for the post-con¬‚ict environment.

and other armed Kosovo-Albanian groups”; “[e]stablishing a secure environment in which
refugees and displaced persons can return home in safety, the international civil presence
can operate, a transitional administration can be established, and humanitarian aid can be
delivered”; “[e]nsuring public safety and order”; “[s]upervising demining”; “[s]upporting,
as appropriate, and coordinating closely with the work of the international civil presence”;
“[c]onducting border monitoring duties as required”; “[e]nsuring the protection and free-
dom of movement of itself, the international civil presence, and other international organi-
zations.” S.C. Res. 1244, para. 9, U.N. Doc. S/RES/1244, June 10, 1999.
26 U.S. Institute of Peace, Establishing the Rule of Law in Iraq, Special Report No. 104, April
2003, at 3 (noting that without immediate progress on security, “international engagement
will be jeopardized by a loss of credibility and an entrenchment of organized crime, extra-
judicial processes, and terrorist activities”) (hereinafter USIP Report).
27 U.S. Army Peacekeeping & Stability Operations Institute, Partnership for Effective Peace
Operations Brie¬ng Note, January 2004 (hereinafter Brie¬ng Note).
28 See, e.g., Alton L. Gwaltney III & Cody M. Weston, Soldiers as Cops, Judges, and Jailers:
Law Enforcement by the U.S. Military in Peace Operations, in Bassiouni, supra note 3, at
863, 875.

A. The Post-Intervention Window of Opportunity
Timing in peace operations is a “crucial determinant of success or failure.”29
When ¬ghting ends, or at least moderates to the point that security becomes
a priority, a critical window of opportunity opens. Interveners have a chance
to demonstrate that a new sheriff is in town and that it is no longer “business
as usual.” Belligerents and other spoilers may be intimidated by the arrival
of professional military forces. Depending on the stage at which interven-
tion takes place, belligerents may be weakened from the ¬ghting, demoral-
ized, and prepared to accept disarmament and demobilization. The general
population, exhausted by years of warfare, crime, and disorder, will often
be prepared and even eager to cooperate with international forces promis-
ing order and a gradual restoration of normalcy. As Robert Oakley and
Michael Dziedzic observe, “[t]his phase of the intervention should not be
squandered, because military presence in signi¬cant numbers and the initial
positive impact on public opinion are of limited duration. The longer an
external military force remains deployed on the ground, the more it is apt to
be perceived as an occupation army.”30 Indeed, the window of opportunity
may last only a few weeks.31
In Iraq, the United States and its coalition partners learned the hard way
the importance of securing order quickly:

In April 2003 U.S. soldiers stood by and watched as looters rampaged through
the streets of Baghdad, and throughout Iraq, demolishing government of¬ces and
destroying valuable records. . . . Looters also damaged hospitals, schools, and basic


. 5
( 15)