Deciding to Keep Peace: An Analysis
of International Inﬂuences on the
Establishment of Third-Party Peacekeeping
Missions
MARK J. MULLENBACH
University of Central Arkansas
Since the end of World War II, and particularly since the end of the Cold
War, there has been an expansion in the number of third-party peacekeeping
missions established throughout the world. Most of the expansion
in peacekeeping missions in the past decade or so has occurred in
states experiencing intrastate or civil conﬂicts. The questions addressed in
this study are under what conditions do third-party actors either decide to
establish or decide not to establish peacekeeping missions in intrastate
disputes, and specifically, what effect do international-level factors have on
the likelihood that third-party peacekeeping personnel will be deployed
in an intrastate dispute? The previous literature on third-party peacekeeping
and interventions is used to derive a set of theoretical arguments
and hypotheses regarding the establishment of peacekeeping missions by
third-party actors (the United Nations, regional organizations, and ad hoc
groups of states) during the post-World War II period. Specifically, I
argue that several factors originating at the level of the international system
inﬂuence the occurrence of third-party peacekeeping missions. The
results of statistical analyses of the hypotheses largely support the notion
that a set of international-level factors significantly inﬂuences the decisions
of third-party actors to establish or not establish third-party peacekeeping
missions, that international-level factors are more important than
state-level factors, that these factors often have different effects on the
likelihood of different types of third-party peacekeeping.
Under what conditions does a third-party actor either decide to establish or decide
not to establish a peacekeeping mission in an intrastate dispute?1
What effect do
Author’s note: An earlier version of this article was presented at the Annual Meeting of the Peace Science Society
(International) in Ann Arbor, Michigan, November 14–16, 2003. I would like to thank Bill Dixon, Gary Goertz, Jody
McMullen, anonymous reviewers, and ISQ editors for their helpful comments and suggestions. The data used in
this article can be found at the following location: http://www.isanet.org/data_archive.html.
1
Peacekeeping is deﬁned in this study as military and/or civilian personnel deployed by one or more third-party
states, frequently but not necessarily under the auspices of a global or regional organization, into a conﬂict or postconﬂict
situation for the purpose of preventing the resumption of military hostilities between two parties and/or for
the purpose of creating an environment conducive for negotiations between two parties (for examples of similar
definitions of peacekeeping, see Rikhye et al., 1974:10–1; Rikhye, 1984:1–2; James, 1990:1–10; Diehl, 1993:4–14;
Durch, 1993:3–4; Goulding, 1993:452–55; Jett, 1999:13–9). The definition of peacekeeping provided above includes
the following broad functions: maintaining law and order, monitoring or verifying ceaseﬁre or disengagement
agreements, supervising the disarmament or demobilization of combatants, and protecting humanitarian
assistance. This definition of peacekeeping does not include ‘‘peace enforcement’’ efforts (e.g. US-led military forces
deployed in Korea in the 1950s or the US-led military forces deployed in the Persian Gulf region in 1990–1991) or
r 2005 International Studies Association.
Published by Blackwell Publishing, 350 Main Street, Malden, MA 02148, USA, and 9600 Garsington Road, Oxford OX4 2DQ, UK.
International Studies Quarterly (2005) 49, 529–555
international-level factors have on the likelihood that third-party peacekeeping
personnel will be deployed in an intrastate dispute?2
These questions are important
for contemporary scholars and practitioners of international relations for
two reasons. First, the conditions under which third-party actors are motivated
to establish (or the conditions under which third-party actors are constrained
from establishing) peacekeeping missions have not been adequately examined even
though there has been a significant increase in the number of third-party peacekeeping
missions initiated throughout the world since the end of World War II
and particularly since the end of the Cold War (Thakur and Schnabel, 2001:3;
Mason, 2003:19–20).3
Not only has the number of peacekeeping missions increased
in the past several decades, but the complexity of these missions has also increased
(Goulding, 1993:456–60; Mason, 2003:29; Fortna, 2004:269). In addition, most
of the expansion in peacekeeping missions in the past decade or so has taken place
in states experiencing intrastate disputes or civil conﬂicts (James, 1995:242;
Oudraat, 1996:490; O’Connor, 2001:57; Mason, 2003:30). An examination of
the frequency of third-party peacekeeping during the 60 years since the end
of World War II suggests that there has been a significant amount of variation in the
establishment of peacekeeping missions during this period.4
However, despite
the variation in the deployment of third-party peacekeeping personnel during the
past 60 years, scholars have paid relatively little attention to theoretical explanations
for the establishment of these missions in civil conﬂicts. A broader understanding
of why third-party peacekeeping as a means of managing conﬂicts is used
relatively more frequently in some circumstances might ultimately provide scholars
with some additional insight into the duration of civil conﬂicts and the resolution of
intrastate disputes.
Second, many scholars have analyzed the effectiveness of third-party peacekeeping
missions, including the factors inﬂuencing the success or failure of peacekeeping
missions (e.g., Mackinlay, 1989; Diehl, 1993, 1994; Durch, 1993; James, 1995;
Shaw, 1995; Jett, 1999; Fortna, 2004). However, the implications of some of these
studies of peacekeeping effectiveness may be limited without a better understanding
of the factors that inﬂuence the formation of peacekeeping missions in the ﬁrst
place, particularly those studies that focus solely on international conﬂicts during
which peacekeeping missions have actually been established.5
In fact, the success or
failure of a third-party peacekeeping mission may at least partly be inﬂuenced by
the initial decision to establish or not establish the mission. For example, the United
Nations (UN) chose to deploy peacekeeping personnel in Cambodia, El Salvador,
Mozambique, and Namibia in the early 1990s, but it chose not to send peacekeeping
personnel to several other states also experiencing civil conﬂicts (e.g., Bangladesh,
Colombia, Congo-Brazzaville, Ethiopia, India, Laos, Lebanon, Philippines,
South Africa, Sudan, Suriname, and Uganda). Some have argued that the UN
established a number of peacekeeping missions in the early 1990s in states experiencing
ideology-based civil conﬂicts because of the end of the Cold War between
post-conﬂict ‘‘peacebuilding’’ efforts (e.g., election monitoring, human rights monitoring, civilian police assistance/
monitoring, refugee repatriation, humanitarian assistance, temporary administration, de-mining, etc.).
2
International-level factors refer to inﬂuences that originate from the interstate or global level of analysis, such
as a military alliance between the government of the target state and a major power.
3
Virginia Page Fortna (2003:97) argued that the ‘‘most important innovation in conﬂict management in the last
50 years is the practice of peacekeeping: the concept of sending personnel from the international community to help
keep the peace in the aftermath of war.’’
4
For example, Dennis Jett (1999) describes the variation in the establishment of UN peacekeeping missions by
dividing the post-World War II era into seven periods: (1) the Nascent Period, 1946–1956; (2) the Assertive Period,
1956–1967; (3) the Dormant Period, 1967–1973; (4) the Resurgent Period, 1973–1978; (5) the Maintenance Period,
1978–1985; (6) the Expansion Period, 1986–1993; and (7) the Contraction Period, 1994–present.
5
Fortna (2004:269) suggested that the ‘‘vast literature on peacekeeping compares cases and missions, but
generally examines only cases in which the international community intervenes, not cases in which belligerents are
left to their own devices.’’
Deciding to Keep Peace530
the Soviet Union and the United States.6
While UN peacekeeping missions in
Cambodia, El Salvador, Mozambique, and Namibia were largely considered successful,
it is possible that UN peacekeeping missions would not have been as successful
in one or more of the other states experiencing civil conﬂicts based largely
on ethnicity, race, or religion.
On the other hand, it is also possible that third-party actors are arguably motivated
to establish peacekeeping missions in intrastate disputes involving significant
humanitarian problems. If so, this might account for some of the
ineffectiveness of third-party peacekeeping missions in intrastate disputes since
conﬂicts involving large numbers of fatalities and displaced persons might be
among the most difﬁcult conﬂicts to manage and resolve. In either case, it is reasonable
to conclude that understanding the factors that motivate third-party actors
to establish peacekeeping missions might provide some insight into the effectiveness
or ineffectiveness of peacekeeping missions that are actually established.
Most scholarly studies of peacekeeping have focused on the UN primarily because
the organization has been in the forefront in deploying ‘‘peacekeeping’’ personnel
during the past 60 years.7
In fact, a total of 59 peacekeeping (or ‘‘peace
observation’’) missions have been established by the UN in interstate and intrastate
disputes since its founding in 1945, including 46 missions since the beginning of
1988.8
However, these studies completely ignore at least two other types of thirdparty
peacekeeping missions: peacekeeping missions by regional inter-governmental
organizations (IGOs) and peacekeeping missions by states or ad hoc groups of
states. Several peacekeeping missions have been formed by regional IGOs throughout
the world since 1945. These organizations vary from multi-regional organizations
such as the North Atlantic Treaty Organization (NATO) and the
Commonwealth of Nations (CON) to sub-regional organizations such as the Economic
Community of West African States (ECOWAS) and the Commonwealth of
Independent States (CIS). This category also includes organizations with members
located primarily in one particular region of the world, such as the Organization of
American States (OAS) in the Western Hemisphere and the Organization of African
Unity (OAU) in Africa.9
For example, the OAS deployed some 14,000 peacekeeping
personnel in the Dominican Republic beginning in May 1965.10
In addition,
several peacekeeping missions have been established by states or ad hoc groups of
states outside of the ofﬁcial auspices of the UN or a regional IGO (although many of
these missions were authorized by the UN or a regional IGO). This category includes
peacekeeping missions consisting of personnel from one state (unilateral
peacekeeping) and peacekeeping missions consisting of personnel from two or
more states (multilateral peacekeeping). For example, France, Italy, and the U.S.
formed a multinational peacekeeping missionFwhich was known as the Multinational
Force (MNF I)Fin Lebanon in July 1982.
Overall, the study of peacekeeping is limited by the relative overemphasis on
evaluations or assessments of particular UN peacekeeping missions and by the
relative lack of emphasis on empirical analysis of the establishment of third-party
peacekeeping missions in general. While previous studies of UN peacekeeping
6
For example, Fortna (1993:356) noted that ‘‘Namibia was very much tied up in the Cold War’’ and that until
the late 1980s ‘‘each superpower was more concerned with minimizing its rival’s inﬂuence than with promoting a
peaceful solution to the conﬂict.’’
7
The UN Charter does not mention the concept of ‘‘peacekeeping,’’ and the UN did not establish its ﬁrst ofﬁcial
‘‘peacekeeping mission’’ until 1956 when the UN General Assembly established the UN Emergency Force (UNEF) to
supervise the withdrawal of foreign troops from the Suez Canal area. However, the UN had established ‘‘peace
observation’’ missions in Greece, Palestine, Dutch East Indies, and Kashmir beginning in the mid-1940s (Rikhye,
1983:5–6).
8
See UN homepage: http://www.un.org/Depts/dpko/dpko/home.shtml.
9
The Organization of African Unity (OAU) was renamed the African Union (AU) in 2002.
10
Schoonmaker (1990:114–15).
MARK J. MULLENBACH 531
have contributed much descriptive information about such missions, they have
contributed little theoretical understanding of the conditions under which thirdparty
actors are motivated to establish peacekeeping missions. Therefore, the
primary focus of this study is to empirically analyze the conditions under which
third-party peacekeeping missionsFincluding missions established by the UN, regional
IGOs, and statesFare more or less likely to be established in states experiencing
civil conﬂicts. The emphasis on civil conﬂicts, as opposed to interstate
conﬂicts, stems from the fact that most third-party peacekeeping personnel since
the end of World War IIFand almost all peacekeeping personnel since the end of
the Cold WarFhave been deployed to prevent military hostilities between groups
within states (James, 1994:30). In addition, it is possible that the factors explaining
the establishment of third-party peacekeeping missions in intrastate disputes differ
from the factors explaining the establishment of third-party peacekeeping missions
in interstate disputes.
Following a brief review of the prior literature on third-party peacekeeping and
intervention, several testable hypotheses are derived from international-level explanations
of peacekeeping, controlling for the potential effects of a number of
state-level inﬂuences.11
The hypotheses are statistically tested against the historical
record of all intrastate conﬂicts and third-party peacekeeping missions between
1945 and 2002. The results of these empirical analyses will be assessed in terms of
their implications for subsequent, and more narrowly focused, studies of the establishment
and effectiveness of third-party peacekeeping missions in intrastate
disputes. Since there have been few prior studies of the establishment of third-party
peacekeeping, this study is designed to be broad in its scope in order to provide a
theoretical and empirical foundation for subsequent analyses of peacekeeping.
Third-Party Peacekeeping in Intrastate Disputes
Although there have been numerous case studies of specific instances of third-party
peacekeeping throughout the world,12
very few statistical analyses have been carried
out on the conditions under which third-party actors are more or less likely to
decide to establish peacekeeping missions in intrastate disputes. Most of these case
studies have largely consisted of descriptions of the events that led to the formation
of one or more peacekeeping missions in one or more states, as well as assessments
of the effectiveness of peacekeeping missions in achieving their particular mandates.
For example, William Durch (1993) edited a book examining in detail some
20 UN peacekeeping missions between 1945 and 1992, including the origin
and an assessment of each of the missions. While there was little analysis of the
decisions by the UN to establish the peacekeeping missions, Durch did suggest that
peacekeeping missions ‘‘require complimentary political support from the Great
Powers and the local parties’’ and that if ‘‘either is missing or deﬁcient, an operation
may never get underway or may fail to achieve its potential once deployed’’
(1993:22–3).
Since third-party peacekeeping tended to be one of the most expensiveFand
one of the most controversialFof the various tools of conﬂict management
throughout the post-World War II period, it is understandable that most previous
studies of peacekeeping focused on the success or failure of such missions. Ramesh
Thakur (1984), for example, studied the involvement of Canada, India, and Poland
in a peacekeeping missionFthe International Commission for Supervision and
Control (ICSC)Fin Vietnam between 1954 and 1973. He argued that ‘‘peace-
11
State-level factors refer to characteristics of the target state (i.e., states experiencing the intrastate conﬂict) or
characteristics of the dispute, such as the level of human suffering resulting from a civil conﬂict.
12
For example, see Curtis (1964), Farris (1994), Adeleke (1995), Alden (1995), Olonisakin (1996), Schmidl
(1999), Sesay (1991), Thakur (1984), and Thakur (1994).
Deciding to Keep Peace532
keeping is successful where it is limited to narrow, precisely deﬁned tasks of overseeing
a military disengagement upon the cessation of hostilities, but fails when
extended to embrace political tasks of conﬂict resolution, and is not viable against
the self-deﬁned vital interests of a superpower’’ (1984:2). Examining six cases of
third-party peacekeeping in his study of the conditions that contribute to the effectiveness
of peacekeeping operations, Paul Diehl (1988:502–03) concluded that
one of the main obstacles to successful peacekeeping has been the ‘‘opposition of
third-party states and subnational groups’’ to the peacekeeping mission. He also
concluded that the ability of peacekeepers to adequately separate the combatants
and the perceived neutrality of the peacekeepers both inﬂuence the effectiveness of
third-party peacekeeping missions (1988:503). Similarly, Frank Gregory (1984)
examined several cases of ‘‘multinational forces’’ during the post-World War II
period, including the UN peacekeeping mission in the Congo in the 1960s and the
OAU peacekeeping mission in Chad in the early 1980s. Gregory concluded that the
success or failure of peacekeeping operations is largely dependent upon the ‘‘impartiality’’
of the peacekeepers (1984:36). These and other studies generally supported
conventional wisdom that third-party peacekeeping could only be successful
if the peacekeeping personnel were perceived by the disputants to be neutral, the
mandate of the peacekeeping mission was limited in its scope, and all relevant
parties in the target state were supportive of (or at least not opposed to) the mission.
Some scholars have sought to contribute to the development of a theory of thirdparty
peacekeeping through the categorization of peacekeeping missions. For example,
Alan James (1990:14–5) suggested ﬁve different categories of peacekeeping
missions based on the ‘‘political circumstances’’ that provide opportunities for such
missions, ranging from ‘‘backyard problems’’ peacekeeping to ‘‘dangerous crossroads.’’
The former category refers to peacekeeping missions that take place within
the sphere of inﬂuence of a major power, such as the Dominican Republic in the
Caribbean region. The latter category refers to missions that take place in situations
where the risk of military hostilities breaking out between two states is relatively
high, such as the demilitarized zone (DMZ) between North Korea and South Korea.
Examining some 57 opportunities for third-party peacekeeping within the ﬁve
categories, James concluded that ‘‘peacekeeping is an activity which can be utilized
in all types of international conﬂict’’ and that ‘‘the kind of international care and
attention which peacekeepers offer can be in demand for long periods irrespective
of the sort of issue to which they are applying their skills’’ (1990:363). Meanwhile,
Diehl, Druckman, and Wall (1998:39–40) suggested twelve categories of peacekeeping
operations on the basis of particular functions, including traditional peacekeeping,
observation, collective enforcement, election supervision, humanitarian
assistance, state/nation building, paciﬁcation, preventive deployment, arms control
veriﬁcation, protective services, intervention in support of democracy, and sanctions
enforcement. While these and other efforts have provided several useful insights
into the phenomenon of peacekeeping, they have arguably extended the
definition of peacekeeping well beyond what is commonly regarded as peacekeeping
in practice.
Although there have been limitations in the prior literature on third-party
peacekeeping, some scholars have recently examined one or more dimensions of
third-party peacekeeping in a more systematic manner. For example, Michael
Gilligan and Stephen John Stedman (2003) statistically analyzed the effects of a set
of independent variables on the duration of time from the beginning of a civil war
to the establishment of a UN peacekeeping mission during the post-Cold War
period. The authors found that the UN was more likely to deploy peacekeeping
personnel in states where there was a high level of fatalities, and that the UN was
less likely to deploy peacekeeping personnel in states with large government military
forces (2003:44–8). These results suggest that both power-related motivations
and humanitarian-related motivations may inﬂuence decisions regarding UN
MARK J. MULLENBACH 533
peacekeeping. On the other hand, Gilligan and Stedman found no evidence that
the UN was more or less likely to establish a peacekeeping mission in secessionist
conﬂicts, when the disputants have signed a peace treaty, in target states with
democratic governments, or in former colonies of permanent members of the UN
Security Council (2003:49–50). The authors concluded that ‘‘considerations of
power are at least as important as considerations of sovereignty in constraining the
UN’s universalism’’ (2003:53).
More recently, Virginia Page Fortna (2004) statistically analyzed the effects of a
set of factors on the deployment of UN and non-UN peacekeeping missions following
civil wars between 1947 and 1999, as part of a broader study of the effectiveness
of peacekeeping missions in ‘‘keeping peace’’ following civil wars. She
found that UN peacekeeping, particularly ‘‘consent-based peacekeeping,’’ was less
likely in cases in which the civil war ended with a victory by the government or
rebels (as well as in cases in which the civil war ended with a formal peace treaty)
and in cases in which the state experiencing the civil war had a large army
(2004:280). Fortna concluded that ‘‘the answer to the question where do peacekeepers
get sent is quite complicated,’’ and added that it ‘‘depends on whether we
are talking about UN peacekeeping or missions by other actors. . .’’ (2004:281). For
the most part, these and other recent empirical analyses of third-party peacekeeping
have focused on the inﬂuence of particular characteristics or attributes of the
target state or the civil conﬂict on the establishment of peacekeeping missions.
Third-Party Interventions in Intrastate Disputes
Notwithstanding the limitations in the literature on third-party peacekeeping, there
has been a considerable amount of empirical research conducted on third-party
interventions in intrastate disputes, including military, economic, and diplomatic
interventions. Since peacekeeping is one particular type of third-party military
intervention, we should be able to gain some theoretical insights into the formation
of peacekeeping missions from these studies.13
Unlike the literature on third-party
peacekeeping, studies of third-party intervention have tended to focus on explanations
from both the international system level and the state level. The previous
literature on third-party interventions can be divided between analyses of state
interventions and analyses of UN interventions.
State Interventions
According to Hans Morgenthau, intervention ‘‘is as ancient and well-established an
instrument of foreign policy as are diplomatic pressure, negotiations and war’’ and
‘‘some states have found it advantageous to intervene in the affairs of other states
on behalf of their own interests and against the latter’s will’’ (1967:425). Morgenthau
argued that major powers, specifically the Soviet Union and US, intervened
in other states during the Cold War to support former colonies that were
lacking in political, military, and economic viability; to support the government or
opposition groups during revolutions and civil conﬂicts; and to support or oppose
governments or opposition groups on the basis of ideology (1967:426–28). Since
the 1960s, Morgenthau’s traditional realist perspective of third-party intervention
has, to a considerable extent, inﬂuenced the literature on state interventions in
other states.14
For example, Frederic Pearson (1974:453) examined foreign military
interventions by states between 1948 and 1967 and found that ‘‘major power’’
military interventions, as well as ‘‘middle or small power’’ military interventions,
13
The scholarly literature on third-party interventions is rather extensive, so I limit my discussion to those
studies that specifically address the causes or motivations of third-party interventions in intrastate disputes.
14
For examples, see Tillema and van Wingen (1982), Feste (1992), Pearson et al. (1994) and Tillema (1994).
Deciding to Keep Peace534
were primarily inﬂuenced by concerns for regional power balances or ideological
interests.
More recently, particularly since the end of the Cold War, scholars have increasingly
examined both strategic or power motivations and non-strategic or affective
motivations of state interventions (Heraclides, 1990; Cooper and Berdal, 1993;
Carment, James, and Rowlands, 1997; Regan, 2000).15
For example, Daniel Byman,
Chalk, Hoffman, Rosenau, and Brannan (2001:23–33) examined the motivations
of state interventions in support of insurgent movements within other states
and found that states tend to support insurgencies in rival neighboring states as a
means of increasing their inﬂuence within a region and weakening the inﬂuence of
their rivals. In addition to ethnic and ideological motivations of state support for
insurgents in other states, Byman et al. (2001:36) also found that ‘‘religion can be a
powerful motivation for states to support insurgencies.’’ Similarly, Jonathan Fox
(2001:526) found that ‘‘religious conﬂicts are associated with more intervention and
the interveners in religious conﬂicts tend to have religious afﬁnities with the minority
on whose behalf they intervene.’’
Some scholars have also found evidence to support the argument that states
sometimes intervene for humanitarian reasons. For example, Martha Finnemore
(2004:102) argued that humanitarian intervention by states during the post-Cold
War period ‘‘cannot be understood apart from the changing normative context in
which it occurs.’’ Specifically, she suggested that the extent of ‘‘humanitarian
norms’’ existing at the level of the international system inﬂuences the extent of
humanitarian military intervention by states (2004:102–03). In his study of state
interventions in civil conﬂicts in the post-World War II period, Patrick Regan
(2000:61) also provided some empirical evidence that unilateral interventions by
states were more likely in civil conﬂicts when there were concerns about an impending
humanitarian crisis.
UN Interventions
Some three decades ago, Oscar Schachter (1974:415–24) suggested that decisions
by the UN to become involved (or not become involved) in internal conﬂicts were
directly inﬂuenced by one or more of the following factors: (1) the interests and
positions of the great powers; (2) the ‘‘territorial integrity’’ of a member-state of the
UN; (3) the involvement of regional organizations; (4) the consent of the government
of the target state; (5) legal restrictions on the UN in regard to ‘‘domestic
matters’’; and (6) attitudes with respect to revolutionary movements and human
rights. According to Linda Miller (1967:19), the ‘‘conditions under which the UN
expresses its concerns in cases of domestic strife and the means by which it responds
to these disorders are determined by the interplay of national and international
interests and inﬂuences.’’ Specifically, Miller concluded that the ‘‘status of
these conﬂicts in world politics’’ inﬂuences the extent to which the UN intervenes in
such internal conﬂicts (1967:35). Similarly, John Ruggie (1974:495–96) argued that
characteristics of the international political environment largely inﬂuenced the
likelihood of a UN intervention in an international dispute. According to Ruggie,
during the Cold War period, the UN was less likely to intervene in an international
dispute involving a ‘‘cold war issue’’ (i.e., an issue stemming from the rivalry between
the U.S. and Soviet Union), less likely to intervene when the dispute involved
the national interests of a permanent member of the UN Security Council, and
more likely to intervene when the dispute involved two ‘‘non-aligned’’ states
(1974:498–503).
15
Mitchell (1970:184–85) suggested that ‘‘affective linkages’’ between internal social groups and external parties
involved ‘‘congruent values, attitudes, ideologies, and self-images’’ such as religion and ethnicity.
MARK J. MULLENBACH 535
More recently, Chantal de Jonge Oudraat (1996:518–19) also argued that the
likelihood of UN intervention in an internal conﬂict, particularly during the postCold
War period, is higher when one or more of the permanent members of the
UN Security Council have national interests in the conﬂict and when one or more
of the permanent members of the UN Security Council perceive the conﬂict to be a
threat to international peace and security. Oudraat concluded that whether or not
the UN Security Council takes action regarding an international conﬂict depends
on one or more permanent members taking the ‘‘diplomatic lead’’ and assuming
the ‘‘costs of diplomatic, economic, or military action’’ (1996:519–20). Overall, these
and other scholars have concluded that the likelihood of the UN intervening in an
international dispute depends largely on the national interests and involvement of
major powers in the international system. However, some scholars have also concluded
that there are circumstances in which the interests of the international
community as a wholeFas well as international norms and lawFare also potentially
important inﬂuences on the likelihood of UN intervention.
Theoretical Arguments
Although a good deal of the emphasis in the previous literature, particularly the
literature on third-party peacekeeping, has been on particular characteristics of the
target state and conﬂict, I have chosen to emphasize factors inﬂuencing the likelihood
of third-party peacekeeping missions at the level of the international sys-
tem.16
As one of several conﬂict management techniques available to third-party
actors, peacekeeping is perhaps the one technique that is most inﬂuenced by pressures
and constraints originating from the international system. In fact, third-party
decisions to deploy peacekeeping personnel in civil conﬂict situations typically involve
the governments of two or more states in the international system, global, or
regional organizations, large numbers of military and civilian personnel from two
or more states, and large amounts of ﬁnancial resources. Of course, these decisions
often involve a number of domestic considerations as well, such as national interests
and domestic political conditions. According to Alan James (1990:12), the thirdparty
actors involved in the establishment of a peacekeeping mission ‘‘will have
taken their decisions about the suggested operation in the light of its impact on
their positions and policies, at both the domestic and international levels, and on
balance have decided on a positive response.’’
Assuming that decisions by third-party actors to establish peacekeeping missions
are partially, if not largely, inﬂuenced by international politics, I have identiﬁed
seven international-level explanations of peacekeeping in intrastate disputes. First,
third-party peacekeeping can be explained by the presence of military alliances
between major global or regional powers and states experiencing civil conﬂicts. The
military alliance hypothesis suggests that a third-party peacekeeping mission is less
likely to be established in a target state if the government of the target state has a
formal military alliance with the government of a major global or regional power
(Ruggie, 1974:503; Schachter, 1974:416).17
This hypothesis is based on the assumption
that the deployment of third-party peacekeeping personnel in a target
state allied with a major power would hinder the primary goal of suppressing
16
Although it is tempting to categorize the explanations of third-party peacekeeping according to the two major
theoretical approaches in the ﬁeld of international relations (i.e. realist and liberal), there are several good reasons
not to do so. Referring to explanations of the establishment of UN peacekeeping, Michael Gilligan and Stephen
John Stedman (2003:41) suggested that ‘‘from an empirical standpoint distinguishing between the idealist and
realist explanations will not be as easy as it ﬁrst appears,’’ particularly because in some instances the same prediction
of the occurrence of UN peacekeeping might be made by both realism and liberalism. There is also the problem that
each of these two broad theoretical approaches includes specific variants (e.g. neoclassical realism and structural
realism) that might make contradictory predictions about third-party peacekeeping in particular circumstances.
17
In this study, a ‘‘target state’’ refers to a state that is experiencing, or has recently experienced, a civil conﬂict.
Deciding to Keep Peace536
opposition to the government of the allied state. It is also assumed that a major
power would oppose the involvement of third-party peacekeeping personnel in
what would be considered the major power’s sphere of inﬂuence. In either case, a
major power would be expected to use its political inﬂuence to prevent the deployment
of third-party peacekeeping personnel on the territory of a military ally.
Second, third-party peacekeeping can be explained by the power status of the
target state in the international or regional system. The major power status hypothesis
suggests that a third-party peacekeeping mission is less likely to be established in a
target state if the target state is a major global or regional power (Fortna, 2003:102).
This hypothesis is based on the assumption that the government of a major power
has sufﬁcient military capabilities to suppress or defeat an internal opposition
group. As a result of the power capabilities that a major power government possesses
relative to an internal opposition group, such a government would have little
incentive to permit the formation of a third-party peacekeeping mission on its
territory. This hypothesis also assumes that a major power has sufﬁcient political
inﬂuence to prevent a third-party actor, particularly the UN or a regional organization,
from establishing a peacekeeping mission on its territory since international
organizations are largely reﬂections of the distribution of power in the international
or regional system.
Third-party peacekeeping can also be explained by previous military involvement
of a major power in a target state. Specifically, the military intervention hypothesis
suggests that a third-party peacekeeping mission is less likely to be established in a
target state if a major global or regional power has previously militarily intervened
during the conﬂict in support of, or in opposition to, one of the parties to the
dispute (Forsythe, 1972:1076). This hypothesis is based on the assumption that
once a major power has militarily intervened in support of or against the government
of the target state, it is likely to oppose the subsequent formation of a peacekeeping
mission in the target state by any other third-party actor. The primary goal
of the major power in these situations would be to assist one of the parties to the
dispute in militarily defeating the other party, and a third-party peacekeeping
mission would arguably hinder the achievement of that goal. As with the previous
argument, it is assumed that a major power has the ability to prevent the formation
of a third-party peacekeeping mission when it is not in the national interest of the
major power.
Similarly, third-party peacekeeping can be explained by previous intermediary
involvement of a major power in a target state. The intermediary intervention hypothesis
suggests that a third-party peacekeeping mission is more likely to be established in a
target state if a major global or regional power has previously intervened as an
intermediary (e.g., mediated negotiations between the parties) during the conﬂict.
This hypothesis is based on the assumption that it is occasionally in the national
interest of a major power to facilitate the management or resolution of a conﬂict in
a target state, particularly when the major power neither has a military alliance with
the government of the state nor has militarily intervened for or against one of the
parties to the dispute. In some situations, the primary goal of a major global or
regional power may be to prevent a civil conﬂict in a neighboring state, or any state
for that matter, from undermining the security and stability of the international
system or a regional subsystem. This hypothesis is also based on the assumption
that if a major power has intervened as an intermediary in a civil conﬂict, the major
power is likely to favor (or at least not oppose) the deployment of third-party
peacekeeping personnel in the target state.
A ﬁfth explanation of third-party peacekeeping emphasizes the involvement of
international institutions, such as the UN and regional IGOs, in states experiencing
civil conﬂicts. The institutional involvement hypothesis suggests that a third-party
peacekeeping mission is more likely to be established in a target state if the UN or a
regional IGO has previously been involved as a ‘‘conﬂict manager’’ in the target
MARK J. MULLENBACH 537
state.18
This hypothesis is based on the assumption that once the international
communityFthrough international institutions such as the UN or a regional
IGOFhas invested resources in order to manage or resolve a conﬂict within a
target state, it will rationally seek to protect that investment through continued
involvement following a cessation of military hostilities. From this perspective, one
of the most effective ways to prevent the resumption of military hostilities between
disputants is to deploy third-party peacekeeping personnel in the target state.
Sixth, third-party peacekeeping can be explained by the existence of a nonintervention
norm in the international system.19
Specifically, the non-intervention
norm hypothesis suggests that a third-party peacekeeping mission is less likely to be
established in a target state when there is a prevailing norm of non-intervention in
the international system. This hypothesis is based on the assumption that state and
non-state actors in the international system are constrained, as a result of a norm of
non-intervention, to intervene in the domestic affairs of sovereign states (Ball,
1961; Scott, 1968; Little, 1975; Slater and Nardin, 1986; Krasner, 1995). According
to Vincent (1974:14), the norm of non-intervention ‘‘identiﬁes the right of states to
sovereignty as a standard in international society and makes explicit the respect
required for it in abstention from intervention.’’ In fact, the existence of a nonintervention
norm is reﬂected in Article 2.7 of Chapter I of the UN Charter ratiﬁed
on October 24, 1945, which states that ‘‘nothing contained in the present Charter
shall authorize the UN to intervene in matters which are essentially within the
domestic jurisdiction of any state’’ (Goodrich, Hambro, and Patricia, 1969:60).20
The charters of several regional IGOs, including the OAS and OAU, also contain
references to a norm of non-intervention.21
Notwithstanding the prevalence of a norm of non-intervention during the postWorld
War II period, several scholars have argued in recent years that the end of
the Cold War marked the emergence of a conﬂicting international normFa norm
of humanitarian interventionFwhich permits interventions in the domestic affairs
of states in order to protect human rights and provide humanitarian assistance to
those adversely affected by civil conﬂict (Blechman, 1995; Weiss and Chopra, 1995;
Finnemore, 1998; Clarke, 2001; Kardas, 2001).22
Indeed, Karns and Mingst
(2001:219) argued that ‘‘the reality . . . is that the UN is now operating under a
de facto norm of intervention in civil conﬂicts, using limited force to stop anarchy
and humanitarian disaster.’’ Likewise, Finnemore (1998:198) suggested that
since the end of the Cold War, states ‘‘consider intervening in small remote
states for largely humanitarian reasons because the organization and normative
structures of international politics have created the means and justiﬁcation to do
so.’’ The norm of non-intervention that was prevalent during the Cold War period
has arguably been replaced by the norm of humanitarian intervention during the
18
See Schachter (1974:424–45) for a discussion of 10 different ‘‘ways and means’’ for the UN to play a role in
internal conﬂicts, including good ofﬁces, conciliation, fact-ﬁnding, economic assistance, and sanctions.
19
For a thorough review of the study of the inﬂuence of ‘‘international norms’’ on the behaviors of states in the
international system, see Raymond (1997).
20
The UN General Assembly also adopted the ‘‘Declaration on the Inadmissibility of Intervention in the Domestic
Affairs of States and the Protection of their Independence and Sovereignty’’ on December 21, 1965 (Onuf,
1971:218).
21
For example, Article 19 of Chapter IV of the OAS Charter states: ‘‘No state or group of states has the right to
intervene, directly or indirectly, for any reason whatever, in the internal or external affairs of any state’’ (Ball
1961:23). In addition, Article III(2) of the OAU Charter requires member states to declare their adherence to the
principle of ‘‘non-interference in the internal affairs of states’’ (Skjelsbaek, 1986:150).
22
Finnemore (2004:113) argued that the ‘‘humanitarian intervention norms that have evolved . . . appear to allow
intervention in cases of humanitarian disaster and abuse, but with at least two caveats. First, they are permissive
norms only. They do not require intervention, as the cases of Burundi, Sudan, and other states make clear. Second,
they place strict requirements on the ways in which intervention, if employed, must be carried out: Humanitarian
intervention must be multilateral if states are to accept it as legitimately and genuinely humanitarian.’’
Deciding to Keep Peace538
post-Cold War period as a result of the changes that occurred in the international
system following the collapse of the Soviet Union and the end of the superpower
rivalry. As a result, it is assumed that the deployment of third-party peacekeeping
personnel was generally less constrained during the post-Cold War period compared
with the Cold War period.
Finally, third-party peacekeeping can be explained by the existence of a territorial
integrity norm in the international system. The territorial integrity norm hypothesis
suggests that a third-party peacekeeping mission is less likely to be
established in a target state if there is a threat to the territorial integrity of the state,
specifically if there is a movement by one particular region to secede from the target
state. Referring to the norm of territorial integrity, Alexis Heraclides (1990:342)
argued that third-party actors ‘‘refrain from supporting secessionist movements
because of existing normative regime against involvement, particularly involvement
with groups that threaten the territorial integrity of states.’’ The norm of
territorial integrity arguably extends to the establishment of peacekeeping missions
by third-party actors in civil conﬂicts involving secessionist movements. For example,
David Forsythe (1972:1077) indicated that, in the case of the Biafran secessionist
conﬂict in Nigeria in 1967–1970, ‘‘calls for a UN peacekeeping force . . . were
rejected by Lagos since a ceaseﬁre policed by a UN force would tend to consolidate
Biafra’s separate position besides drawing international attention to its claim of
independence.’’
The territorial integrity norm hypothesis is based on the assumption that states in
the international system are committed to, and in fact have a shared interest in,
preserving the territorial integrity of all sovereign states in the international system.
For example, the UN Charter explicitly prohibits members from violating the territorial
integrity of any state in the international system.23
This hypothesis also
assumes that the formation of a third-party peacekeeping mission in a state with a
secessionist movement would enhance the prospects of the secessionists achieving
their goals. Therefore, the common interest in preserving the territorial integrity of
all states in the international system is an important constraint on third-party decisions
to establish peacekeeping missions in these situations.
Research Design
In order to test the preceding hypotheses regarding the establishment of thirdparty
peacekeeping missions, some 213 civil conﬂicts occurring between January 1,
1945 and December 31, 2002 were identiﬁed.24
Unlike many previous studies of
‘‘civil wars’’ involving 1,000 or more fatalities, the cases of civil conﬂict in this study
correspond to periods of military hostilities between government security personnel
and members of one or more armed opposition groups within a state lasting 10
23
Chapter I, Article 2.4 of the UN Charter states, ‘‘All members shall refrain in their international relations from
the threat or use of force against the territorial integrity or political independence of any state, or in any other
manner inconsistent with the Purposes of the UN.’’
24
The complete list of intrastate conﬂicts, including the beginning and ending dates of each of the conﬂicts, is
located at: http://faculty.uca.edu/$markm/tpi_intrastateconﬂicts.htm. More than 97% of the 213 cases (207 out of
213) used in this study are included in one or more of the following conﬂict datasets: Ayres, R. William, ‘‘Violent
Intrastate Nationalist Conﬂicts (VINC), 1945–1996,’’ University of Indianapolis (http://facstaff.uindy.edu/$bayres/
vinc.htm); Doyle, Michael W. and Nicholas Sambanis. (2000) International Peacebuilding: A Theoretical and
Quantitative Analysis. American Political Science Review 94 (December): 779–801; Fearon, James D. and David D.
Laitin. (2003) Ethnicity, Insurgency, and Civil War. American Political Science Review 97 (February): 75–90; Gleditsch,
Nils Peter, et al. (2002) Armed Conﬂict 1946–2001: A New Dataset, Journal of Peace Research 39 (5): 615–637;
Marshall, Monty G. 2003. ‘‘Major Episodes of Political Violence, 1946–2002,’’ Center for Systemic Peace (http://
members.aol.com/CSPmgm/warlist.htm); Regan, Patrick M. 1996. ‘‘Conditions of Successful Third-Party Intervention
in Intrastate Conﬂicts.’’ Journal of Conﬂict Resolution 40: 336–359; and Sarkees, Meredith Reid. 2000. ‘‘The
Correlates of War Data on War: An Update to 1997,’’ Conﬂict Management and Peace Science 18:123–144 (COW IntraState
War Data, 1816–1997, Version 3.0, http://cow2.la.psu.edu/).
MARK J. MULLENBACH 539
or more days, regardless of the number of fatalities.25
The threshold for including a
civil conﬂict in this study is intentionally low in order to account for the wide range
of scenarios, including both short periods of military hostilities and long periods of
military hostilities, in which a third-party actor might decide to establish or not
establish a peacekeeping mission. Cases of civil conﬂict, including the post-conﬂict
periods that follow the end of military hostilities, are appropriate units of analysis
for this study since they represent distinct opportunities for third-party peacekeeping.
In other words, third-party actors generally do not have the opportunity to
deploy peacekeeping personnel in a target state unless the parties to an intrastate
dispute have engaged in military hostilities.26
Of the 213 civil conﬂicts that were identiﬁed for this study, 111 conﬂicts (52%)
ended during the Cold War period, and 102 conﬂicts (48%) endedFor were still
ongoingFduring the post-Cold War period.27
The conﬂicts took place in each of
the ﬁve regions of the world, including Asia/PacificF69 (32%); Europe/Former
Soviet UnionF25 (12%); Middle East/North Africa/Persian GulfF38 (18%); SubSaharan
AfricaF59 (28%); and Western HemisphereF22 (10%). Figure 1 shows
that the number of ongoing civil conﬂicts varied considerably during the period
from 1945 to 2002. During the period from 1945 to 1960, the number of ongoing
conﬂicts was relatively stable in the range of 3–15 per year. The number of ongoing
conﬂicts increased sharply during the period from 1961 to 1973, and then stabilized
in the range of 25–30 per year through 1978. During the period from 1979 to
1991, there was another sharp increase in the number of ongoing conﬂicts, peaking
at 46 in 1991. Since 1992, there has been a decrease in the number of ongoing
0
5
10
15
20
25
30
35
40
45
50
1947 1950 1953 1956 1959 1962 1965 1968 1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001
# of Conflicts
FIG. 1. Number of Ongoing Intrastate Conﬂicts
25
The average duration of the 213 intrastate conﬂicts used in this study is approximately 74 months. The
duration of the shortest conﬂict was 11 days, and the duration of the longest conﬂict was 568 months. Eight out of
the 213 conﬂicts lasted less than 1 month, and 67 of the 213 conﬂicts lasted less than 12 months.
26
The only exceptions would be cases of ‘‘preventive deployment’’ of peacekeeping personnel; however, there
have been very few such cases in the post-World War II period.
27
The cases that ‘‘ended during the Cold War period’’ correspond to intrastate conﬂicts that ended on or before
December 31, 1989.
Deciding to Keep Peace540
conﬂicts, but the number of conﬂicts remains at a level higher than much of the
pre-1979 period.
Each of the 213 civil conﬂicts was examined for cases of third-party peacekeeping
missions. One or more third-party peacekeeping missions were observed during
or following a total of 76 (35.7%) of the 213 civil conﬂicts, including 28 (25.2%) of
the 111 conﬂicts that ended during the Cold War period and 48 (47.1%) of the
102 conﬂicts that occurred during the post-Cold War period. A total of 111
different third-party peacekeeping missions were established during or following
the conﬂicts, including 43 UN peacekeeping missions, 35 regional IGO peacekeeping
missions, and 33 state peacekeeping missions.28
Figure 2 shows the
number of third-party peacekeeping missions initiated between 1945 and 2002.
This graph indicates that the number of peacekeeping missions established by
third-party actors during 5-year periods ranged between two and seven from 1945
to 1989, but the number of peacekeeping missions established increased sharply
after the end of the Cold War. The number of peacekeeping missions established
during both 5-year periods in the 1990s ranged between 30 and 35. However, there
were only 12 peacekeeping missions established during the 3-year period between
2000 and 2002.
Overall, there are at least two interesting observations regarding the information
shown in Figure 2. First, the UN established no peacekeeping missions in intrastate
disputes between 1970 and 1984, after establishing eight missions during the earlier
period from 1945 to 1969. Despite the lack of UN missions (or perhaps as a
result of the lack of UN missions), regional IGOs and ad hoc groups of states
established eleven peacekeeping missions during the period from 1970 to 1984.
Second, the significant increase in peacekeeping missions during the ﬁrst decade of
the post-Cold War period was the result of the establishment of increasing numbers
of peacekeeping missions by all three types of third-party actors. Although the UN
established 29 peacekeeping missions in intrastate disputes during the 1990s, re-
0
5
10
15
20
25
30
35
1945-
1949
1950-
1954
1955-
1959
1960-
1964
1965-
1969
1970-
1974
1975-
1979
1980-
1984
1985-
1989
1990-
1994
1995-
1999
2000-
2002
# State Missions
# Regional IGO Missions
# UN Missions
FIG. 2. Third-Party Peacekeeping Missions in Intrastate Conﬂicts
28
The complete list of third-party peacekeeping missions in intrastate conﬂicts, including the dates of each of the
peacekeeping missions, is located at: http://faculty.uca.edu/$markm/tpi_intrastateconﬂicts.htm.
MARK J. MULLENBACH 541
gional IGOs and ad hoc groups of states established a combined 34 peacekeeping
missions in intrastate disputes during the same period.
Dependent Variables
In order to analyze the formation of peacekeeping missions by the three different
types of third-party actors, four different models are estimated in this study: (a) a
model explaining the establishment of any type of third-party peacekeeping mission;
(b) a model explaining the establishment of UN peacekeeping missions; (c) a
model explaining the establishment of regional IGO peacekeeping missions; and
(d) a model explaining the establishment of state (or ad hoc groups of states)
peacekeeping missions. For each of these models, the dependent variable is coded
‘‘1’’ when a peacekeeping mission was established during or within 12 months of
the end of a civil conﬂict and is coded ‘‘0’’ otherwise.
In addition, four modelsFusing the same four variations of the dependent variable
discussed aboveFare estimated with alternative indicators of ﬁve of the seven
independent variables. Specifically, four of the independent variables are divided into
a ‘‘global power’’ indicator and a ‘‘regional power’’ indicator. For example, the military
alliance variable is divided into an indicator that accounts for a military alliance
between the government of the target state and a major global power and an indicator
that accounts for a military alliance between the government of the target
state and a major regional power. In addition, the institutional involvement variable
is divided into an indicator for the ‘‘UN’’ and an indicator for ‘‘regional IGOs.’’
Independent Variables
Military AllianceFcoded ‘‘1’’ if the government of the target state has a military alliance
with a major global power29
or major regional power,30
and coded ‘‘0’’ otherwise.
The source of the data on military alliances is the Correlates of War (COW)
Project, Formal Interstate Alliances Data Set (version 3.03).31
Only ‘‘defense pacts’’
(Type I category in the COW Project’s alliances data) are coded as ‘‘military alliances’’
in this study.
Major Power StatusFcoded ‘‘1’’ if the target state is a major global or regional
power, and coded ‘‘0’’ otherwise (see footnotes 29 and 30).
Military InterventionFcoded ‘‘1’’ if a major global or regional power intervened
militarily in support of, or in opposition to, one of the parties during a civil conﬂict,
29
With one exception, the "major global powers" refer to the ﬁve permanent members of the UN Security
Council during the period from 1945 to 2002: United States, Soviet Union/Russian Federation, Republic of China/
People’s Republic of China, France, and United Kingdom. The People’s Republic of China (PRC) (or ‘‘mainland
China’’) is coded as a "major global power" during the period from 1950 to 2002, even though the Republic of
China/Taiwan held the permanent seat on the UN Security Council during part of that period.
30
The ‘‘major regional powers’’ refer to the four to seven most militarily powerful states in each of the ﬁve
regions of the world in terms of level of military expenditures and number of military personnel, including external
states with substantial military forces based or deployed in each of the regions. The following countries are considered
major "regional powers" during part or all of the period from 1945 to 2002: (1) Asia/Pacific RegionFAustralia
(1958–2002); China (1945–2002), France (1945–1954), India (1947–2002), Japan (1952–2002), South Korea
(1958–2002), Russia/Soviet Union (1945–2002), United Kingdom (1945–1958), United States (1945–2002); (2)
Europe/Russia/Former Soviet Union RegionFFrance (1945–2002), Italy (1945–2002), Russia/Soviet Union (1945–2002),
United Kingdom (1945–2002), United States (1945–2002), West Germany/Germany (1949–2002); (3) Middle East/
North Africa/Persian Gulf RegionFEgypt (1945–2002), France (1945–1962), Iran (1945–2002), Israel (1962–2002),
Saudi Arabia (1945–2002), Turkey (1945–2002), United Kingdom (1945–1970), United States (1970–2002); (4) SubSaharan
Africa RegionFEthiopia (1945–2002), France (1945–2002), Nigeria (1960–2002), Portugal (1945–1975),
South Africa (1945–2002), United Kingdom (1945–1980); and (5) Western Hemisphere RegionFArgentina (1945–
2002), Brazil (1945–2002), Canada (1945–2002), Mexico (1945–2002), United States (1945–2002) (Sources: National
Material Capabilities Data, Version 2.1, Correlates of War (COW) Project, University of Michigan; World Military
Expenditures and Arms Transfers, U.S. Arms Control and Disarmament Agency/U.S. Department of State).
31
See Gibler, Douglas M. and Meredith Sarkees (2004) Measuring Alliance: the Correlates of War Formal
Interestate Alliance Data Set, 1816–2000. Journal of Peace Research 41: 211–222.
Deciding to Keep Peace542
and coded ‘‘0’’ otherwise. In this study, the notion of ‘‘military intervention’’
includes any type of military involvement, including military assistance (e.g., military
airlift, weapons, and supplies), military advisory, and use of military force. The source
of data on military interventions in intrastate disputes is the Third-Party Interventions
in Intrastate Disputes Project, Military Interventions Data Set (version 1.0).32
Intermediary InterventionFcoded ‘‘1’’ if a major global power or major regional
power intervened as an intermediary during a civil conﬂict, and coded ‘‘0’’
otherwise. The notion of ‘‘intermediary intervention’’ includes any neutral or nonpartisan
effortFincluding fact-ﬁnding, good ofﬁces, mediation, and conciliationFfor
the purpose of assisting the parties in peacefully managing or resolving
the conﬂict. The source of data on intermediary interventions in intrastate disputes
is the Third-Party Interventions in Intrastate Disputes Project, Intermediary Interventions
Data Set (version 1.0).
Institutional InvolvementFcoded ‘‘1’’ if the UN or a regional IGO previously intervened
to manage or resolve a civil conﬂict, and coded ‘‘0’’ otherwise. A wide
variety of conﬂict management activities by the UN or a regional IGO in a target state
constitutes ‘‘institutional involvement,’’ including fact-ﬁnding, good ofﬁces, mediation,
conciliation, economic sanctions, military sanctions, humanitarian assistance,
repatriation assistance, human rights monitoring, and election monitoring. The
source of data on UN and regional IGO interventions in intrastate disputes is the
Third-Party Interventions in Intrastate Disputes Project, Intermediary Interventions
Data Set (version 1.0) and Participatory Interventions Data Set (version 1.0).
Non-Intervention NormFcoded ‘‘1’’ if a civil conﬂict ended during the Cold War
period (on or prior to December 31, 1989), and coded ‘‘0’’ otherwise.33
Territorial Integrity NormFcoded ‘‘1’’ if a civil conﬂict involves a secessionist/separatist
movement, and coded ‘‘0’’ otherwise. A secessionist movement refers to an
effort by one or more groups within a state to separate themselves, as well as part of
the territory of that state, from the control of the government of the state. The goal
of a secessionist group is to establish a new state or attach to another state, usually
for ethnic, religious, or economic reasons (Spencer, 1998: 2–3). Of the 213 civil
conﬂicts used in this study, 69 (32.4%) involve secessionist movements. The source
of data on intrastate disputes is Third-Party Interventions in Intrastate Disputes
Project, Intrastate Disputes Data Set (version 1.0).
Control Variables
Four sets of variables are included in each of the models to control for various
attributes of the target state or the civil conﬂict that may inﬂuence a third-party
actor to decide to deploy peacekeeping personnel in a target state.
First, the manner in which a civil conﬂict ends might affect the likelihood of the
establishment of a third-party peacekeeping mission. Specifically, I control for the
possibility that the likelihood of a peacekeeping mission varies depending on
whether or not the parties formally agree to a cessation of military hostilities. The
ceaseﬁre agreement variable is coded ‘‘1’’ when there is a formal ceaseﬁre agreement
and coded ‘‘0’’ otherwise. Along the same lines, I control for the possibility that the
likelihood of a peacekeeping mission varies depending on whether or not one of
32
The data sets of the Third-Party Interventions in Intrastate Disputes Project are located at http://facul-
ty.uca.edu/$markm/tpi_homepage.htm.
33
In addition to intrastate conﬂicts that began during the post-Cold War period or after December 31, 1989
(and therefore, occurred entirely during the post-Cold War period), conﬂicts that began during the Cold War
period but ended during the post-Cold War period are coded ‘‘0’’. The later conﬂicts are cases that would have been
inﬂuenced by both the non-intervention norm and the humanitarian intervention norm. This coding scheme would
potentially pose a problem if there were cases in which a third-party peacekeeping mission was established during
the Cold War period in an intrastate conﬂict that began during the Cold War but ended during the post-Cold War
period. However, there are no such cases in the data set.
MARK J. MULLENBACH 543
the parties to a dispute is militarily defeated by the other party (Fortna, 2004:278).
The military defeat variable is coded ‘‘1’’ when one of the parties to the dispute was
militarily defeated at the end of the conﬂict and coded ‘‘0’’ otherwise.
Second, the duration, consequence, and intensity of a civil conﬂict might inﬂuence
the likelihood of the formation of a third-party peacekeeping mission. Specifically,
I control for the possibility that the likelihood of a peacekeeping mission is
different depending on whether or not the conﬂict was relatively short. For example,
if a third-party actor wanted to minimize the risk of peacekeeping failure, it
might not want to establish peacekeeping missions during or following relatively
lengthy conﬂicts. The conﬂict duration variable is coded ‘‘1’’ when a civil conﬂict
lasted less than 12 months and coded ‘‘0’’ otherwise. Also, there is the possibility
that the extent of human suffering caused by a civil conﬂict affects the likelihood of
a peacekeeping mission (Regan, 2000:50–1; Gilligan and Stedman, 2003:44; Mason,
2003:32). The human suffering control variable is coded ‘‘1’’ if there was an
average of 500 or more annual fatalities and 50,000 or more total displaced persons
resulting from the conﬂict; it is coded ‘‘0’’ otherwise.34
In addition, there is the
possibility that the intensity of a civil conﬂictFin terms of the number of combatant
and non-combatant fatalities suffered per monthFinﬂuences the likelihood of a
peacekeeping mission (Regan, 2000:49–50). Highly intense (or extremely violent)
conﬂicts might encourage third-party actors to establish peacekeeping missions in
order to end the extreme violence or might discourage third-party actors from
establishing peacekeeping missions as a result of the extreme violence. The conﬂict
intensity control variable is coded ‘‘1’’ when there was an average of 1,000 or more
monthly fatalities and coded ‘‘0’’ otherwise.35
Third, the type of intrastate dispute might inﬂuence the deployment of thirdparty
peacekeeping personnel. Specifically, there is the possibility that the presence
or absence of one or more dispute characteristics such as religion, ethnicity, and
ideology inﬂuences the likelihood of peacekeeping. The religion, ethnicity, and ideology
control variables are coded ‘‘1’’ if these characteristics are present in an intrastate
dispute and coded ‘‘0’’ otherwise.
Finally, the region of the world in which an intrastate dispute is occurring might
affect the likelihood of the establishment of a third-party peacekeeping mission
(Gilligan and Stedman, 2003:43). Four dummy variables are included in the models
for the following regions: Asia/Pacific; Europe/Former Soviet Union; Middle East/North
Africa/Persian Gulf; and Sub-Saharan Africa (the Western Hemisphere region is the
residual or omitted category).
Analysis
As a ﬁrst step in the analysis of the inﬂuence of international-level factors on the
establishment of third-party peacekeeping missions in civil conﬂicts, the percentages
corresponding to the number of conﬂicts during which third-party peacekeeping
missions were established given the occurrence of each of the
international-level factors are examined. Table 1 provides these percentages for
each of the different types of international inﬂuences. For example, the ﬁrst entry
in column 1 indicates that third-party peacekeeping personnel were deployed in
about 28% of the civil conﬂicts (23 out of 81) in which the government of the target
state had a military alliance with a major global or regional power. Furthermore, the
second entry in column 1 indicates that third-party peacekeeping personnel were
deployed in about 25% of the civil conﬂicts (14 out of 57) in which the government
34
Two alternative measurements of this control variable are also tested (100 annual fatalities/25,000 displaced
persons and 1,000 annual fatalities/75,000 displaced persons).
35
Two alternative measurements of this control variable are also tested (average of 500 or more monthly
fatalities and average of 1,500 or more monthly fatalities).
Deciding to Keep Peace544
of the target state had a military alliance with a major global power, and the third
entry in column 1 indicates that third-party peacekeeping personnel were deployed
in about 29% of the civil conﬂicts (23 out of 80) in which the government of the
target state had a military alliance with a major regional power. Although somewhat
lower, these percentages are not significantly different from the overall rate at
which one or more third-party peacekeeping missions were established in the total
number of civil conﬂicts (76 out of 213, or approximately 36%). On the other hand,
the percentages corresponding to the number of cases in which the target state was
a major power or major regional power (approximately 12%) are significantly lower
than the overall rate of about 36%. In addition, the percentages corresponding to
the number of cases in which a major power intervened as an intermediary (68%)
and in which the UN or a regional IGO was previously involved (approximately
58%) are significantly higher than the overall rate of 36%.
Although the information obtained from the ﬁgures in Table 1 is somewhat
useful, a significant deﬁciency regarding the interpretation of the percentages is
that there is no means of assessing the separate effects of each of the international
inﬂuences on the formation of third-party peacekeeping missions. Because there is
a significant amount of overlap among the various international inﬂuences, it is
nearly impossible to assess the independent effect that any one particular international-level
factor had on the establishment of third-party peacekeeping missions.
In order to properly analyze the separate effects of each of the independent variables
controlling for other international-level and state-level factors, statistical
techniques are used. Specifically, the effects of the international inﬂuences on the
likelihood of third-party peacekeeping missions are estimated using logistic
TABLE 1. International Inﬂuences and the Establishment of Third-Party Peacekeeping Missions
International Inﬂuence
(1) (2) (3) (4) (5)
Third-Party
Peacekeeping
United Nations
Peacekeeping
Regional IGO
Peacekeeping
State
Peacekeeping Totals
Military alliance 28.4% (23) 9.9% (8) 12.3% (10) 9.9% (8) 38.0% (81)
Global power 24.6% (14) 8.8% (5) 10.5% (6) 8.8% (5) 26.8% (57)
Regional power 28.8% (23) 10.0% (8) 12.5% (10) 10.0% (8) 37.6% (80)
Major Power Status 11.6% (5) 2.3% (1) 2.3% (1) 6.9% (3) 20.2% (43)
Global power 22.2% (4) 0.0% (0) 5.6% (1) 16.7% (3) 8.4% (18)
Regional power 11.6% (5) 2.3% (1) 2.3% (1) 6.9% (3) 20.2% (43)
Military intervention 36.3% (29) 16.3% (13) 12.5% (10) 12.5% (10) 37.6% (80)
Global power 35.3% (24) 17.6% (12) 13.2% (9) 10.3% (7) 31.9% (68)
Regional power 37.8% (28) 17.6% (13) 12.2% (9) 13.5% (10) 34.7% (74)
Intermediary intervention 68.0% (34) 32.0% (16) 20.0% (10) 32.0% (16) 23.5% (50)
Global power 68.6% (24) 34.3% (12) 20.0% (7) 28.6% (10) 16.4% (35)
Regional power 68.2% (30) 29.5% (13) 22.7% (10) 34.1% (15) 20.7% (44)
Institutional involvement 57.7% (64) 30.6% (34) 21.6% (24) 19.8% (22) 52.1% (111)
United Nations 66.2% (51) 41.6% (32) 22.1% (17) 20.8% (16) 36.2% (77)
Regional IGOs 59.8% (52) 27.6% (24) 27.6% (24) 19.5% (17) 40.8% (87)
Non-intervention norm 25.2% (28) 11.7% (13) 9.9% (11) 8.1% (9) 52.1% (111)
Territorial integrity norm 31.9% (22) 11.6% (8) 8.7% (6) 15.9% (11) 32.4% (69)
Totals 35.7% (76) 16.9% (36) 13.1% (28) 14.1% (30) (213)
Note: Entries in columns 1 through 4 represent the percentages of intrastate conﬂicts during which third-party
peacekeeping missions were established given each type of international inﬂuence. The number of intrastate conﬂicts
during which third-party peacekeeping missions were established in each circumstance is in parentheses.
Entries in column 5 indicate the percentage of intrastate conﬂicts during which each international inﬂuence was
observed. The sum of these percentages does not equal 100 because the international inﬂuences are not mutually
exclusive. The total number of cases (out of the total of 213 intrastate conﬂicts) during which each international
inﬂuence was observed is in parentheses.
IGO, inter-governmental organizations.
MARK J. MULLENBACH 545
regression analysis since each of the variations of the dependent variable is measured
dichotomously.
The logistic regression results are presented in Tables 2 and 3. Models 1 through
4 (Table 2) correspond to the four variations of the dependent variable using the
basic set of independent variables, while Models 5 through 8 (Table 3) correspond
to the four variations of the dependent variable using the alternative set of independent
variables. Each of the eight models provides at least some evidence in
support of one or more of the seven hypotheses. For example, the results in Model
1 provide support for three of the seven hypotheses. As predicted, third-party
peacekeeping missions are significantly less likely when the target state is a major
power. In addition, third-party peacekeeping missions are significantly more likely
when a major power has previously intervened as an intermediary during a civil
conﬂict and when the UN or a regional IGO has previously been involved during a
civil conﬂict. The coefﬁcients for military alliance and non-intervention norm are in
the hypothesized directions (negative), but they are not statistically significant. Finally,
the coefﬁcients for military intervention and territorial integrity norm are
neither in the hypothesized direction nor statistically significant in Model 1.
Similarly, the results in Model 5Fusing the alternative set of independent variables,
but the same dependent variable as in Model 1Falso provide support for
three of the seven hypotheses. As predicted, third-party peacekeeping missions are
TABLE 2. Logit Analyses of the Establishment of Third-Party Peacekeeping Missions in Civil Conﬂicts
Variable
Model 1
(Third-Party)
Model 2
(United Nations)
Model 3
(Regional IGO)
Model 4
(State)
Military alliance À .200 (.603) À 1.875nn
(1.137) .920 (.736) .658 (.558)
Major power status À 1.345nn
(.780) À 2.195nn
(1.080) À 1.775nn
(.977) À .615 (.596)
Military intervention .643 (.633) À .372 (.688) .185 (.632) .201 (.581)
Intermediary
intervention
1.761nnn
(.554) .724 (.590) .065 (.518) 1.226nnn
(.506)
Institutional
involvement
2.396nnn
(.547) 3.369nnn
(.983) .938 (.759) 1.082nn
(.636)
Non-intervention
norm
À .327 (.675) 1.563n
(.990) .143 (.664) À 1.305nn
(.585)
Territorial integrity
norm
.291 (.781) .150 (.928) À .419 (.939) À .462 (.617)
Ceaseﬁre agreement 2.210nnn
(.497) 2.192n
(1.224) 1.462n
(.838) 1.404n
(.745)
Military defeat À .432 (.748) À 1.735 (1.433) À .752 (.860) .288 (1.043)
Human suffering .841 (.530) 1.488nn
(.689) .965n
(.535) À .400 (.456)
Conﬂict intensity À .325 (.597) .654 (.496) À .283 (.549) .238 (.502)
Conﬂict duration 1.510nnn
(.585) .308 (.620) 2.047nnn
(.542) À .262 (.547)
Religion À .063 (.691) .278 (1.208) 1.277n
(.741) À .182 (.785)
Ethnicity À 1.102n
(.580) .856 (1.231) À 1.825n
(.936) À .330 (.801)
Ideology À .473 (.665) .950 (1.043) À .371 (.824) À .595 (.831)
Asia/Pacific À .221 (1.004) À 3.488nn
(1.664) a 2.594nnn
(.878)
Europe 1.095 (1.143) À 2.662 (1.709) 1.420 (.865) 1.919n
(1.067)
Middle East À 1.110 (.920) À 2.726n
(1.635) .297 (.847) a
Sub-Saharan Africa À .184 (1.107) À 2.368 (1.521) 2.230nnn
(.867) .939 (.900)
Constant À 3.670nnn
(1.317) À 5.771nn
(2.564) À 5.218nnn
(1.544) À 4.290nnn
(1.083)
N 213 213 213 213
Log likelihood À 71.513 À 50.613 À 55.897 À 65.455
Wald w2
83.5 66.16 65.16 42.82
Pseudo-R2
0.4847 0.4770 0.3256 0.2440
Note: Cell entries are logit coefﬁents. Robust standard errors, which are adjusted for clustering on the intrastate
dispute, are in parentheses. Significance levels: n
po.10; nn
po.05; nnn
po.01; one-tailed tests.
IGO, inter-governmental organizations; a, dropped due to lack of variance.
Deciding to Keep Peace546
significantly less likely when the target state is a major regional power. Furthermore,
third-party peacekeeping missions are significantly more likely when a major
regional power has previously intervened as an intermediary during a civil conﬂict
and when the UN or a regional IGO has previously been involved during a civil
conﬂict. The coefﬁcients for military alliance/global power, intermediary intervention/global
power, and non-intervention norm are in the hypothesized directions,
but they are not statistically significant. The coefﬁcients for military alliance/regional
power, military intervention/global power, military intervention/regional
power, and territorial integrity norm are neither in the hypothesized directions nor
statistically significant. A more thorough discussion of the statistical results and
ﬁndings from each of the models is provided in the next section.
Since logistic coefﬁcients are difﬁcult to interpret, predicted probabilities for each
of the independent variables are provided in Tables 4 and 5. Specifically, the tables
TABLE 3. Logit Analyses of the Establishment of Third-Party Peacekeeping Missions in Civil Conﬂicts
Variable
Model 5
(Third-Party )
Model 6
(United Nations)
Model 7
(Regional IGO)
Model 8
(State)
Military alliance
Global power À .464 (1.945) À 5.300nnn
(2.042) 1.939n
(1.308) À 1.833 (2.232)
Regional power .374 (1.871) 3.023n
(1.854) À .776 (.910) 2.329 (2.242)
Major power status
Global power a a a a
Regional Power À 1.532nn
(.768) À 4.725nnn
(1.879) À 2.736nnn
(1.140) À .310 (.653)
Military intervention
Global power .280 (.999) .319 (1.763) 2.281nn
(1.085) À 1.003 (.804)
Regional power .750 (.949) .538 (1.905) À 2.048nn
(1.147) 1.193n
(.903)
Intermediary intervention
Global power .757 (.719) 1.574 (1.267) À .262 (.684) À .447 (.944)
Regional power 1.266nn
(.723) À .998 (1.481) 1.345nn
(.771) 1.327nn
(.737)
Institutional involvement
United Nations 2.033nnn
(.463) 5.703nnn
(1.451) À .853n
(.633) .569 (.650)
Regional IGOs 1.278nnn
(.514) À .933 (.759) 2.500nnn
(.998) .240 (.659)
Non-intervention
norm
À .162 (.669) 2.851nn
(1.244) .747 (.807) À 1.614nn
(.696)
Territorial integrity
norm
.452 (.790) À .260 (.980) À .157 (.871) À .599 (.686)
Ceaseﬁre agreement 2.118nnn
(.527) 2.782nnn
(1.011) 1.772nn
(.786) 1.848nn
(.821)
Military defeat À .518 (.808) À 2.965nn
(1.433) À .851 (.961) .594 (1.070)
Human suffering .635 (.507) .727 (.821) 1.672nnn
(.588) À .216 (.444)
Conﬂict intensity À .397 (.639) .584 (.653) À .179 (.615) .194 (.491)
Conﬂict duration 1.348nn
(.601) .368 (.644) 2.800nnn
(.654) À .296 (.536)
Religion .339 (.681) .766 (1.406) .774 (.848) À .034 (.819)
Ethnicity À .897 (.716) 2.953nn
(1.468) À 2.854nnn
(.904) À .301 (.739)
Ideology .431 (.721) .382 (1.759) À .622 (1.081) À .203 (.777)
Asia/Pacific À .549 (.948) À 6.893nnn
(2.101) a 3.750 (2.361)
Europe .953 (1.059) À 6.570nnn
(2.016) 2.666nn
(1.320) 3.274 (2.436)
Middle East À 2.129 (2.071) À 10.218nnn
(3.077) 2.225 (1.446) a
Sub-Saharan Africa À .348 (1.102) À 4.481nn
(2.052) 2.944nn
(1.199) 2.007 (2.196)
Constant À 3.500nnn
(1.286) À 5.762nn
(2.528) À 7.324nnn
(2.421) À 5.653nn
(2.511)
N 213 213 213 213
Log likelihood À 69.566 À 39.363 À 47.264 À 64.629
Wald w2
95.80 67.09 53.67 39.21
Pseudo-R2
0.4987 0.5932 0.4298 0.2536
Note: Cell entries are logit coefﬁents. Robust standard errors, which are adjusted for clustering on the intrastate
dispute, are in parentheses. Significance levels: n
po.10; nn
po.05; nnn
po.01; one-tailed tests.
IGO, inter-governmental organizations; a, dropped due to lack of variance.
MARK J. MULLENBACH 547
provide estimates of the probabilities of the various types of third-party peacekeeping
missions given the inﬂuence of each of the international-level factors. The
baseline probability represents the probability of a third-party (UN, regional IGO,
or state) peacekeeping mission being formed when all of the international-level
factors (independent variables) and state-level factors (control variables) are set at
‘‘0,’’ except for ceaseﬁre agreement, human suffering, conﬂict duration, ethnicity,
and Sub-Saharan Africa, which are set at ‘‘1.’’ For example, column 1 of Table 4
indicates that the probability of the establishment of a third-party peacekeeping
mission is substantially greater than the baseline probability of 42% when a major
power has intervened as an intermediary in a target state (74%) and when an
international organization has previously been involved as a conﬂict manager in the
target state (86%). In addition, the probability of a third-party peacekeeping mission
is substantially lower than the baseline probability of 42% when the target state
is a major power (21%).
Although hypotheses were not stipulated with respect to the control variables, it
is interesting to note that the only state-level factors that had at least somewhat
consistent effects across the models on the likelihood of third-party peacekeeping
missions were ceaseﬁre agreement and conﬂict duration. The coefﬁcients for
ceaseﬁre agreement were positive and statistically significant in each of the eight
models, while the coefﬁcients for conﬂict duration were positive and statistically
significant in four of the eight models. Not surprisingly, third-party peacekeeping
missions are significantly more likely to be established when the parties to the
conﬂict have formally agreed to a cessation of military hostilities,36
but peacekeeping
missions are also generally more likely to be established when the duration of a
conﬂict is less than 12 months.
TABLE 4. Predicted Probabilities of the Establishment of Third-Party Peacekeeping Missions
Variable
(1) (2) (3) (4)
Third-Party
Peacekeeping
United Nations
Peacekeeping
Regional IGO
Peacekeeping
State
Peacekeeping
Baseline probability .42 (.21) .06 (.07) .43 (.21) .07 (.07)
Military alliance .38 (.22) .02 (.04) .62 (.19) .13 (.13)
Major power status .21 (.20) .02 (.04) .18 (.19) .05 (.06)
Military intervention .55 (.26) .06 (.10) .48 (.26) .10 (.11)
Intermediary intervention .74 (.20) .12 (.13) .44 (.20) .19 (.14)
Institutional involvement .86 (.10) .51 (.15) .64 (.13) .16 (.10)
Non-intervention norm .35 (.18) .18 (.12) .46 (.21) .02 (.03)
Territorial integrity norm .47 (.21) .07 (.08) .35 (.21) .05 (.05)
Note: Cell entries are probabilities of the establishment of a peacekeeping mission when each of the independent
variables is separately set at 1, with all other independent variables set at 0. Standard errors are in parentheses. The
baseline probability is the probability of the establishment of a peacekeeping mission using all 213 intrastate conﬂict
cases when all of the independent variables are set at 0 (the baseline probability assumes that the conﬂict ended with
a ceaseﬁre agreement; the conﬂict involved a high level of human suffering; the conﬂict had an ethnic dimension;
the duration of the conﬂict was less than 12 months; and the conﬂict took place in Sub-Saharan Africa). Probabilities
are estimated using Clarify 2.1 with Stata 8.0.
IGO, inter-governmental organizations.
36
Gilligan and Stedman (2003:50), who found no evidence that the UN was more likely to establish a peacekeeping
mission when the parties to a civil war had negotiated a peace treaty, indicated that conventional wisdom
suggested that ‘‘the UN requires a peace treaty before it will send in a peacekeeping mission.’’ Actually, the conventional
wisdom is that the UN requires a ceaseﬁre agreementFnot a peace treatyFbefore it will send in a
peacekeeping mission. The results of this study strongly suggest that the UN is in fact more likely to establish a
peacekeeping mission when the parties have formally agreed to a cessation of military hostilities.
Deciding to Keep Peace548
Findings
The statistical results presented in Tables 2 through 5 suggest seven preliminary
ﬁndings regarding the inﬂuence of international-level factors on the likelihood of
third-party peacekeeping missions in intrastate disputes. A summary of the ﬁndings
is provided below.
Finding 1: There is some evidence that under certain circumstances, the formation
of a third-party peacekeeping mission is less likely when the government of a
target state has a military alliance with a major power (Models 2 and 6). Specifically,
Model 6 indicates that UN peacekeeping missions are significantly less likely if the
government of a target state has a military alliance with a major global power. This
result is consistent with the military alliance hypothesis. On the other hand, Model 6
also indicates that UN peacekeeping missions are significantly more likely if the
government of a target state has a military alliance with a major regional power. In
addition, Model 7 indicates that regional IGO peacekeeping missions are significantly
more likely if the government of a target state has a military alliance with a
major global power. Each of these results is also substantively significant, as indicated
in Tables 4 and 5. According to column 2 of Table 5, the probability of a UN
peacekeeping mission is significantly lower than the baseline probability of 7% if the
government of a target state has a military alliance with a major global power (1%).
Furthermore, the probability of a UN peacekeeping mission is more than six times
higher than the baseline probability of 7% if the government of a target state has a
military alliance with a major regional power (44%).
Why are UN peacekeeping missions less likely, but regional IGO peacekeeping
missions more likely, when the government of a target state has a military alliance
TABLE 5. Predicted Probabilities of the Establishment of Third-Party Peacekeeping Missions
Variable
(1) (2) (3) (4)
Third-Party
Peacekeeping
United Nations
Peacekeeping
Regional IGO
Peacekeeping
State
Peacekeeping
Baseline probability .37 (.19) .07 (.09) .32 (.22) .10 (.10)
Military alliance
Global power .35 (.31) .01 (.04) .66 (.24) .09 (.18)
Regional power .46 (.31) .44 (.29) .23 (.23) .46 (.32)
Major power status
Global power a a a a
Regional Power .15 (.15) .01 (.05) .07 (.12) .09 (.10)
Military intervention
Global power .43 (.27) .13 (.20) .68 (.28) .05 (.07)
Regional power .53 (.25) .19 (.26) .11 (.15) .27 (.22)
Intermediary intervention
Global power .52 (.24) .21 (.20) .28 (.24) .09 (.12)
Regional power .62 (.23) .07 (.16) .57 (.22) .27 (.20)
Institutional involvement
United Nations .77 (.15) .86 (.13) .19 (.17) .15 (.12)
Regional IGO .63 (.18) .02 (.04) .79 (.13) .11 (.09)
Non-intervention norm .33 (.17) .38 (.19) .44 (.20) .03 (.03)
Territorial integrity norm .46 (.20) .05 (.09) .28 (.20) .05 (.04)
Note: Cell entries are probabilities of the establishment of a peacekeeping mission when each of the independent
variables is separately set at 1, with all other independent variables set at 0. Standard errors are in parentheses. The
baseline probability is the probability of the establishment of a peacekeeping mission using all 213 intrastate conﬂict
cases when all of the independent variables are set at 0 (the baseline probability assumes that the conﬂict ended with
a ceaseﬁre agreement; the conﬂict involved a high level of human suffering; the conﬂict had an ethnic dimension;
the duration of the conﬂict was less than 12 months; and the conﬂict took place in Sub-Saharan Africa). Probabilities
are estimated using Clarify 2.1 with Stata 8.0.
IGO, inter-governmental organizations; a, dropped due to lack of variance.
MARK J. MULLENBACH 549
with a major global power? One possible explanation is that while global powers are
quite reluctant to allow UN peacekeeping in their own spheres of inﬂuence (mainly
due to the fact that no one major global power exercises complete authority over
the UN), they are sometimes willing and able to use particular regional IGOs to
further their own national interests.
Finding 2: There is considerable evidence that the establishment of a third-party
peacekeeping mission is significantly less likely if the target state is a major power, as
predicted by the major power status hypothesis.37
Specifically, there is evidence that
third-party, UN, and regional IGO peacekeeping missions are significantly less
likely when the target state is a major power (the coefﬁcients for the major power
status variables are statistically significant in Models 1, 2, 3, 5, 6, and 7). The
coefﬁcients for major power status variables in Models 4 and 8 (state peacekeeping)
have the predicted sign, but are not statistically significant. Overall, the relationship
between major power status and third-party peacekeeping is also substantively
significant. As shown in column 1 of Table 4, the probability that third-party
peacekeeping personnel will be deployed is significantly lower than the baseline
probability of 42% if the target state is a major power (21%).
Finding 3: There is only minimal evidence that the establishment of a third-party
peacekeeping mission is significantly less likely if a major power has previously intervened
militarily during a civil conﬂict. In fact, the coefﬁcients for the military
intervention variables are in the predicted direction (negative) in only three of the
eight models (Models 2, 7, and 8), and the coefﬁcients are statistically significant in
only one of these models. There is evidence that regional IGO peacekeeping missions
are significantly less likely when a major regional power has previously intervened
militarily in a target state (Model 7), as predicted by the military intervention hypothesis.
However, there is also evidence that regional IGO peacekeeping missions are significantly
more likely when a major global power has previously intervened militarily
in a target state (Model 7), and state peacekeeping missions are significantly more
likely when a major regional power has previously intervened militarily (Model 8).
Once again, these results point to the interesting possibility that the formation of
regional IGO peacekeeping missions is inﬂuenced in different ways by major global
powers and major regional powers. Similar to the results regarding the military
alliance variable (ﬁnding 1), regional IGOs appear to be subject to contradictory
inﬂuences from major global powers and major regional powers intervening militarily
in civil conﬂicts. When major global powers have direct national interests in
civil conﬂicts, they tend to inﬂuence regional IGOs to deploy peacekeeping personnel
in the target states. On the other hand, when major regional powers have
direct national interests in civil conﬂicts, they tend to inﬂuence regional IGOs to
refrain from deploying peacekeeping personnel in the target states. This result is
probably best explained by the possibility that while major global powers are frequently
able to inﬂuence regional IGOs to deploy peacekeeping personnel when it
is in their national interest for such missions to be established, regional IGOs are
frequently unable to do so. On the other hand, major regional powers are often
able to prevent regional IGOs from deploying peacekeeping personnel when it is
not in their national interest for such missions to be established.
Finding 4: There is some evidence that the formation of a third-party peacekeeping
mission is more likely if a major power has previously intervened as an
37
As indicated in column 5 of Table 1, there are a total of 43 cases of targets states that are categorized as ‘‘major
powers.’’ Many of these cases involve colonies controlled by major global powers (e.g., French Indochina–Vietnam).
In each of the 43 cases, the major power was categorized as a ‘‘major regional power.’’ In 18 of those same 43 cases,
the major power was also categorized as a ‘‘major global power’’ (e.g., France was categorized as a major global
power, as well as a major regional power in Southeast Asia, in the 1950s). Since there is no variation among the 18
cases categorized as ‘‘major global powers’’ (i.e., they are all also coded as ‘‘major regional power’’), logit coefﬁcients
cannot be estimated for the ‘‘major global power’’ variables in Models 5 through 8 in Table 3.
Deciding to Keep Peace550
intermediary during a civil conﬂict (Models 1, 4, 5, 7, and 8). Specifically, there is
evidence that the establishment of regional IGO peacekeeping missions and state
peacekeeping missions are significantly more likely if a major regional power has
intervened as an intermediary during a civil conﬂict (Models 7 and 8). These results
are also substantively significant. As shown in column 3 of Table 5, the probability of
a regional IGO peacekeeping mission is significantly higher than the baseline
probability of 32% if a regional power has previously intervened as an intermediary
(57%). Interestingly, the coefﬁcients for the intermediary intervention variable in
Model 6 suggest that UN peacekeeping missions are generally less likely (negative
coefﬁcient) when a regional power has intervened and are generally more likely
(positive coefﬁcient) when a global power has intervened. These results, although
not statistically significant, may reﬂect the conventional wisdom that the UN is
inﬂuenced to a large extent by the national interests of the major global powers with
permanent representation on the UN Security Council (Oudraat, 1996: 518–19).
Overall, these results generally support the intermediary intervention hypothesis. Thirdparty
actors are generally more likely to establish peacekeeping missions in civil
conﬂicts in which one or more major powers have previously signaled their preference
for peaceful management and resolution.
Finding 5: There is considerable evidence that the establishment of a third-party
peacekeeping mission is more likely if the UN or a regional IGO has previously
been involved during a civil conﬂict (Models 1, 2, 4, 5, 6, and 7). As predicted by the
institutional involvement hypothesis, a UN peacekeeping mission is significantly more
likely to be established when one or more agencies of the UN have previously been
involved during a civil conﬂict (Model 6), and a regional IGO peacekeeping mission
is significantly more likely when one or more regional IGOs have previously been
involved during a civil conﬂict (Model 7). Interestingly, the results also suggest that
the UN and regional IGOs may defer to each other in civil conﬂicts when the other
has previously been involved in conﬂict management or resolution efforts. In fact,
UN peacekeeping missions are generally less likely when a regional IGO has been
involved during a civil conﬂict, and regional IGOs are generally less likely when the
UN has been involved during a civil conﬂict. As shown in column 2 of Table 5, the
probability of a UN peacekeeping mission is significantly higher than the baseline
probability of 7% when a UN agency has previously been involved during a conﬂict
(86%). Likewise, the probability of a regional IGO peacekeeping mission is significantly
higher than the baseline probability of 32% when a regional IGO has
previously been involved during a conﬂict (79%), as shown in column 3 of Table 5.
Finding 6: There is some evidence that the establishment of a third-party peacekeeping
mission was less likely in target states during the Cold War when there was
a prevailing norm of non-intervention in the international system (Model 4 and 8).
However, this ﬁnding only applies to peacekeeping missions established by states.
There is no evidence that UN or regional IGO peacekeeping missions were less
likely during the Cold War period. In fact, the results indicate that UN peacekeeping
missions were significantly more likely during the Cold War period when
there was a norm of non-intervention (Models 2 and 6). According to column 2 of
Table 5, the probability of a UN peacekeeping mission is more than ﬁve times the
baseline probability of 7% when there was a prevailing norm of non-intervention
during the Cold War (38%).
Why were state peacekeeping missions significantly less likely to be established
during the Cold War period when there was a norm of non-intervention, while UN
peacekeeping missions were more likely to be established? One possible explanation
is that states were more constrained than the UNFdespite the ‘‘non-intervention’’
provision in the UN CharterFto deploy peacekeeping personnel during
the Cold War period. Indeed, the UN was actually permitted to intervene in the
domestic affairs of other states during the Cold War if there was a perceived ‘‘threat
to international peace and security.’’ A second, and perhaps better, explanation is
MARK J. MULLENBACH 551
that the behaviors of states regarding the practice of peacekeeping have changed
relatively more than the behaviors of IGOs as a result of the end of the Cold War.
The evidence, as shown in Figure 2, indicates that the number of state peacekeeping
missions increased significantly following the end of the Cold War, when the
number of civil conﬂicts also increased significantly. However, the number of state
peacekeeping missions remained at a relatively high level after the late 1990s, even
though the number of UN and regional IGO peacekeeping missions decreased
somewhat. Perhaps, as the norm of non-intervention eroded in favor of a norm of
humanitarian intervention beginning in the 1990s, states have been better able to
adapt to the changing normative environment, whereas the UN and regional IGOs
(retaining the ‘‘non-intervention’’ provisions in their charters) have been less able to
adapt to the changing norms.
Finding 7: There is little or no evidence that the formation of a third-party
peacekeeping mission is less likely if a civil conﬂict involves a secessionist movement.
There is no statistically significant evidence that third-party actors are less
likely to establish peacekeeping missions in conﬂicts involving secessionist movements.
However, the coefﬁcients for the territorial integrity norm variables are in
the predicted direction (negative) in ﬁve out of eight models. Overall, the results
indicate that third-party actors may be reluctant as a result of a norm of territorial
integrity to establish peacekeeping missions during or following secessionist conﬂicts,
but the evidence is not compelling. In fact, these results are essentially consistent
with Gilligan and Stedman’s (2003:49) ﬁnding that there is ‘‘no evidence that
the UN intervenes in secessionist conﬂicts at a different rate than it intervenes in
attempts to take over control of the government.’’ Assuming that these results are
accurate, why are third-party actors not more constrained to intervene as peacekeepers
in secessionist conﬂicts? It is possible, and indeed likely, that a norm of
territorial integrity actually does exist in the international system, but the norm is
applicable to states and IGOs that might be tempted to intervene on the side of the
secessionist group, as opposed to intervening as neutral peacekeepers. This explanation
might account for Heraclides’ (1990:353) ﬁnding that ‘‘the regime against
intervention and, in particular, against intervention with secessionists, applied to
some extent . . . primarily with regard to high-level involvement.’’
Conclusion
This study represents one of the ﬁrst systematic analyses of the conditions under
which third-party actors were more or less likely to decide to initiate peacekeeping
missions in intrastate disputes during the post-World War II period. Overall, the
results from this study suggest that a set of international-level factors inﬂuence
third-party actors to establishFor constrain third-party actors from establishingFpeacekeeping
missions during or following civil conﬂicts. In fact, one of the
best predictors of peacekeeping in the future is the past and current actions and
behaviors of major global and regional powers, as well as the global and regional
IGOs that may be inﬂuenced by these major powers. Except for the positive effect
that ceaseﬁre agreements have on the establishment of third-party peacekeeping
missions in intrastate disputes, the inﬂuence of various state-level factors on the
deployment of peacekeeping personnel (when accounting for international-level
factors) is not particularly strong. This ﬁnding may cause some scholars and practitioners
to reconsider some of the conventional wisdom regarding the likelihood of
peacekeeping missions in civil conﬂicts.
Given the set of preliminary ﬁndings discussed in the previous section, future
research should examine in more detail the various international-level and statelevel
factors that might inﬂuence the establishment of third-party peacekeeping
missions. For example, are there any significant differences with respect to the
effects of these factors in the different regions of the world or between the Cold War
Deciding to Keep Peace552
period and the post-Cold War period? If so, what accounts for these differences?
One of the key ﬁndings of this study was that third-party peacekeeping missions are
more likely to be established when a major power or an IGO has previously intervened
militarily or diplomatically during military hostilities in a target state.
Future research should focus on the factors that inﬂuence states and IGOs to
choose to intervene militarily or diplomatically in some civil conﬂicts and not others,
as well as the factors that motivate third-party actors to choose one method of
intervention over another method of intervention.
Finally, future research should not only examine the relative effectiveness of
peacekeeping missions that the UN, regional IGOs, and states decide to establish in
response to civil conﬂicts, but it should also account for the factors that inﬂuence
the decision to establish peacekeeping missions in the ﬁrst place. In particular,
future research on the effectiveness of peacekeeping should account for both
international-level and state-level factors. As indicated earlier, previous research
on the effectiveness of peacekeeping has focused almost entirely on state-level
factors or attributes of the peacekeeping mission, such as the perceived neutrality
of the peacekeepers, the scope of the peacekeeping mission, and the level of
support of local parties in the target state. The importance of these factors may or
may not hold up when international-level factors are considered. Overall, this study
has shown that at the very least, there are factors that systematically inﬂuence
whether or not a peacekeeping mission will be established in a civil conﬂict, and that
these factors often have different effects on the decisions of the UN, regional IGOs,
and states.
References
ADELEKE, ADEMOLA (1995) The Politics and Diplomacy of Peacekeeping in West Africa: The ECOWAS
Operation in Liberia. The Journal of Modern African Studies 33:569–593.
ALDEN, CHRIS (1995) The UN and the Resolution of Conﬂict in Mozambique. The Journal of Modern
African Studies 33:103–128.
BALL, M. MARGARET (1961) Non-Intervention v. Human Rights and the Preservation of Democratic
Institutions. International Organization 15:21–37.
BLECHMAN, BARRY M. (1995) The Intervention Dilemma. The Washington Quarterly 18:63–73.
BYMAN, DANIEL L., PETER CHALK, BRUCE HOFFMAN, WILLIAM ROSENAU, AND DAVID BRANNAN (2001)
Trends in Outside Support for Insurgent Movements. Santa Monica, CA: RAND Corporation.
CARMENT, DAVID, PATRICK JAMES, AND DANE ROWLANDS (1997) ‘‘Ethnic Conﬂict and Third-Party Intervention:
Riskiness, Rationality, and Commitment.’’ In Enforcing Cooperation: Risky States and
Intergovernmental Management of Conﬂict, edited by Gerald Schneider and Patricia A. Weitsman,
pp. 104–131. New York: St. Martin’s Press.
CLARKE, JOHN N. (2001) A Pragmatic Approach to Humanitarian Intervention. Journal of Humanitarian
Intervention (September). Available at hhttp://www.jha.ac/articles/a072.pdf i.
COOPER, ROBERT, AND MATS BERDAL (1993) Outside Interventions in Ethnic Conﬂict. Survival 35:
118–142.
CURTIS, GERALD L. (1964) The United Nations Observation Group in Lebanon. International Organization
18:738–765.
DIEHL, PAUL F. (1988) Peacekeeping Operations and the Quest for Peace. Political Science Quarterly
103:485–507.
DIEHL, PAUL F. (1993) Institutional Alternatives to Traditional U.N. Peacekeeping: An Assessment of
Regional and Multinational Options. Armed Forces & Society 19:209–230.
DIEHL, PAUL F. (1994) International Peacekeeping. Baltimore, MD: The Johns Hopkins University Press.
DIEHL, PAUL F., DANIEL DRUCKMAN, AND JAMES WALL (1998) International Peacekeeping and Conﬂict
Resolution: A Taxonomic Analysis with Implications. Journal of Conﬂict Resolution 42:33–55.
DURCH, WILLIAM J., ED. (1993) The Evolution of UN Peacekeeping: Case Studies and Comparative Analyses.
New York: St. Martin’s Press.
FARRIS, KARL (1994) UN Peacekeeping in Cambodia: On Balance, A Success. Parameters 24:38–50.
FESTE, KAREN A. (1992) Expanding the Frontiers: Superpower Intervention during the Cold War. New York:
Praeger Publishers.
MARK J. MULLENBACH 553
FINNEMORE, MARTHA (1998) ‘‘Military Intervention and the Organization of International Politics.’’ In
Collective Conﬂict Management and Changing World Politics, edited by Joseph Lepgold and Thomas
G. Weiss, pp. 181–204. Albany, NY: State University of New York Press.
FINNEMORE, MARTHA (2004) ‘‘Constructing Norms of Humanitarian Intervention.’’ In Essential Readings
in World Politics, 2nd ed., edited by Karen A. Mingst and Jack L. Snyder, pp. 102–118. New
York: W. W. Norton & Company.
FORSYTHE, DAVID P. (1972) Peacekeeping and Domestic Instability. Orbis 15:1064–1084.
FORTNA, VIRGINIA PAGE (1993) ‘‘United Nations Transition Assistance Group.’’ In The Evolution of UN
Peacekeeping: Case Studies and Comparative Analysis, edited by Durch J. William, pp. 353–375. New
York: St. Martin’s Press.
FORTNA, VIRGINIA PAGE (2003) Inside and Out: Peacekeeping and the Duration of Peace after Civil and
Interstate War. International Studies Review 5:97–114.
FORTNA, VIRGINIA PAGE (2004) Does Peacekeeping Keep Peace? International Intervention and the
Duration of Peace after Civil War. International Studies Quarterly 48:269–292.
FOX, JONATHAN (2001) Religious Causes of International Intervention in Ethnic Conﬂicts. International
Politics 38:515–532.
GILLIGAN, MICHAEL, AND STEPHEN JOHN STEDMAN (2003) Where Do the Peacekeepers Go? International
Studies Review 5:37–54.
GOODRICH, LELAND M., EDVARD HAMBRO, AND ANNE PATRICIA SIMONS (1969) Charter of the United
Nations: Commentary and Documents. New York: Columbia University Press.
GOULDING, MARRACK (1993) The Evolution of United Nations Peacekeeping. International Affairs
69:451–464.
GREGORY, FRANK (1984) The Multinational ForceFAid or Obstacle to Conﬂict Resolution. Conﬂict
Studies 170:1–37.
HERACLIDES, ALEXIS (1990) Secessionist Minorities and External Involvement. International Organization
44:341–378.
JAMES, ALAN (1990) Peacekeeping in International Politics. London: MacMillan, Ltd.
JAMES, ALAN (1994) ‘‘Internal Peacekeeping.’’ In Peacekeeping and the Challenge of Civil Conﬂict Resolution
(Proceedings of the Sixth Annual Conﬂict Studies Conference, University of New Brunswick,
September 1992), edited by David A. Charters, pp. 3–24. New Brunswick, NJ: Center for Conﬂict
Studies, University of New Brunswick.
JAMES, ALAN (1995) Peacekeeping in the Post-Cold War Era. International Journal 50:241–265.
JETT, DENNIS C. (1999) Why Peacekeeping Fails. New York: Palgrave Publishers.
KARDAS, SABAN (2001) Humanitarian Intervention: The Evolution of the Idea and Practice. Perceptions:
Journal of International Affairs 6:120–137.
KARNS, MARGARET P., AND KAREN A. MINGST (2001) ‘‘Peacekeeping and the Changing Role of the
United Nations: Four Dilemmas.’’ In United Nations Peacekeeping Operations: Ad Hoc Missions,
Permanent Engagement, edited by Ramesh Thakur and Albrecht Schnabel, pp. 215–237. Tokyo:
United Nations University Press.
KRASNER, STEPHEN D. (1995) ‘‘Sovereignty and Intervention.’’ In Beyond Westphalia? State Sovereignty
and International Intervention, edited by Gene M. Lyons and Michael Mastanduno, pp. 228–249.
Baltimore, MD: The Johns Hopkins University Press.
LITTLE, RICHARD (1975) Intervention: External Involvement in Civil Wars. Totowa, NJ: Rowman and
Littleﬁeld.
MACKINLAY, JOHN (1989) ‘‘The Characteristics of Success and Failure in Modern Peacekeeping Forces.’’
In The Peacekeepers: An Assessment of Peacekeeping Operations at the Arab-Israel Interface, pp. 213–
230. London: Unwin Hyman, Ltd.
MASON, T. DAVID (2003) Globalization, Democratization, and the Prospects for Civil War in the New
Millennium. International Studies Review 5:19–36.
MILLER, LINDA B. (1967) World Order and Local Disorder: The United Nations and Internal Conﬂicts.
Princeton, NJ: Princeton University Press.
MITCHELL, C. R. (1970) Civil Strife and the Involvement of External Parties. International Studies
Quarterly 14:166–194.
MORGENTHAU, HANS J. (1967) To Intervene or Not to Intervene. Foreign Affairs 45:425–436.
O’CONNOR, MICHAEL (2001) ‘‘Policing the Peace.’’ In United Nations Peacekeeping Operations: Ad Hoc
Missions, Permanent Engagement, edited by Ramesh Thakur and Albrecht Schnabel, pp. 57–73.
Tokyo: United Nations University Press.
OLONISAKIN, FUNMI (1996) UN Co-operation with Regional Organizations in Peacekeeping: The Experience
of ECOMOG and UNOMIL in Liberia. International Peacekeeping 3:33–51.
Deciding to Keep Peace554
ONUF, NICHOLAS GREENWOOD (1971) The Principle of Non-Intervention, the United Nations, and the
International System. International Organization 25:209–227.
OUDRAAT, CHANTAL DE JONGE (1996) ‘‘The United Nations and Internal Conﬂicts.’’ In The International
Dimensions of Internal Conﬂicts, edited by Michael E. Brown, pp. 489–535. Cambridge, MA: MIT
Press.
PEARSON, FREDERIC S. (1974) Geographic Proximity and Foreign Military Intervention. Journal of
Conﬂict Resolution 18:432–460.
PEARSON, FREDERIC S., ROBERT A. BAUMANN, AND JEFFREY J. PICKERING (1994) ‘‘Military Intervention
and Realpolitik.’’ In Reconstructing Realpolitik, edited by Frank W. Wayman and Paul F. Diehl, pp.
205–225. Ann Arbor, MI: The University of Michigan Press.
RAYMOND, GREGORY A. (1997) Problems and Prospects in the Study of International Norms. Mershon
International Studies Review 41:205–245.
REGAN, PATRICK M. (2000) Civil Wars and Foreign Powers: Outside Interventions in Intrastate Conﬂict. Ann
Arbor, MI: University of Michigan Press.
RIKHYE, INDAR JIT (1983) ‘‘Peacekeeping and Peacemaking.’’ In Peacekeeping: Appraisals & Proposals,
edited by Henry Wiseman, pp. 5–18. New York: Pergamon Press.
RIKHYE, INDAR JIT (1984) The Theory and Practice of Peacekeeping. New York: St. Martin’s Press.
RIKHYE, INDAR JIT, MICHAEL HARBOTTLE, AND BJORN EGGE (1974) The Thin Blue Line: International
Peacekeeping and Its Future. New Haven, CT: Yale University Press.
RUGGIE, JOHN G. (1974) Contingencies, Constraints, and Collective Security: Perspectives on UN
Involvement in International Disputes. International Organization 28:493–520.
SCHACHTER, OSCAR (1974) ‘‘The United Nations and Internal Conﬂict.’’ In Law and Civil War in the
Modern World, edited by Moore John Norton, pp. 401–445. Baltimore, MD: The Johns Hopkins
University Press.
SCHMIDL, ERWIN A. (1999) The International Operation in Albania, 1913–1914. International Peacekeeping
6:1–10.
SCHOONMAKER, HERBERT G. (1990) Military Crisis Management: U.S. Intervention in the Dominican Republic,
1965. New York: Greenwood Press.
SCOTT, ANDREW M. (1968) Nonintervention and Conditional Intervention. Journal of International
Affairs 22:208–216.
SESAY, AMADU (1991) The Limits of Peace-keeping by a Regional Organization: The OAU Peacekeeping
Force in Chad. Conﬂict Quarterly 11:7–26.
SHAW, CAROLYN M. (1995) Regional Peacekeeping: An Alternative to United Nations Operations? The
Journal of Conﬂict Studies 15:59–81.
SKJELSBAEK, KJELL (1986) Peaceful Settlement of Disputes by the United Nations and other
Intergovernmental Bodies. Cooperation and Conﬂict 11:139–154.
SLATER, JEROME, AND TERRY NARDIN (1986) Non-Intervention and Human Rights. The Journal of Politics
48:86–96.
SPENCER, METTA, ED. (1998) Separatism: Democracy and Disintegration. Lanham, MD: Rowman & Littleﬁeld
Publishers, Inc.
THAKUR, RAMESH (1984) Peacekeeping in Vietnam: Canada, India, Poland, and the International Commission.
Edmonton, Alberta: The University of Alberta Press.
THAKUR, RAMESH (1994) From Peacekeeping to Peace Enforcement: The UN Operation in Somalia.
The Journal of Modern African Studies 32:387–410.
THAKUR, RAMESH, AND ALBRECHT SCHNABEL (2001) ‘‘Cascading Generations of Peacekeeping: Across
the Mogadishu Line to Kosovo and Timor.’’ In United Nations Peacekeeping Operations: Ad Hoc
Missions, Permanent Engagement, edited by Ramesh Thakur and Albrecht Schnabel, pp. 3–25.
Tokyo: United Nations University Press.
TILLEMA, HERBERT K. (1994) Cold War Alliance and Overt Military Intervention, 1945–1991. International
Interactions 20:249–278.
TILLEMA, HERBERT K., AND JOHN R. VAN WINGEN (1982) Law and Power in Military Intervention:
Major States after World War II. International Studies Quarterly 26:220–250.
VINCENT, R. J. (1974) Non-Intervention and International Order. Princeton, NJ: Princeton University
Press.
WEISS, THOMAS G., AND JARAT CHOPRA (1995) ‘‘Sovereignty Under Siege: From Intervention to
Humanitarian Space.’’ In Beyond Westphalia? State Sovereignty and International Intervention, edited
by Gene M. Lyons and Michael Mastanduno, pp. 87–114. Baltimore: The Johns Hopkins
University Press.
MARK J. MULLENBACH 555
556