Introduction:

Thinking About Mediation

Jacob Bercovitch
A PERSPECTIVEON MEDIATION
The ending of the Cold War and the fundamental changes that have taken
place in international relations over the past few years may have altered
the character or occurrence of conflict.They have not removed the causes
of conflict, however, nor affected its intensity or the need to deal with it
effectively.The new international system is as conflict-prone, many would
argue even more so, as any previous system. The world today is literally
covered with ethnic, religious, territorial, and nationalist conflicts that are
as serious, costly, and intense as any in the past. And somehow they need
to be managed or resolved.
Although the Cold War ended, neither history, as we were led to be-
lieve, nor conflict ended in 1990.The global changes of that year spawned
a myriad of new problems, few of which require more urgent attention
than the need to maintain peace and security within and between states.
Regardless of whatever changes were brought about and how the distri-
bution of power took shape, we quickly realized that conflicts were going
to remain with us whether the international system was unipolar, bipolar,
or multipolar. The expectation that conflicts would diminish or even dis-
appear proved to be as erroneous as it was ephemeral. As long as con-
flicts are with us-and it is truly impossible to conceive of a system that
will be conflict-free-the possibility of serious damage exists, as does the
need to prevent or resolve such conflicts.
How, then, can we manage or resolve our conflicts? Throughout his-
tory, individuals, groups, communities, and more recently states, have
searched for methods of dealing with conflict in more constructive and
peaceful ways than the seemingly inevitable resort to hurling stones at
each other. Some of these ways have been fairly ingenious (such as the
whistling or singing competitions that we see in some communities) and
others have been fairly obvious (such as talking to each other). Lying
I
somewhere mid-point on a spectrum ranging from violence to the use of
ritual as a way of managing conflict is the practice of mediati0n.Thi.s book
is devoted to an analysis of this important form of conflict management.
It is hoped that the concepts and case studies presented here will enable
us to formulate better theories of mediation and will contribute to a more
effective response to conflicts in international relations.
It is worth noting at the outset that as a response to conflict there is
little that is novel about mediation. It has been used everywhere and has
a rich and varied history. References to it abound in the Bible, Homer's
Iliad, and ~o~hoc l e f ' Ajax. It is also found in arenas as diverse as neigh-
borhood and personal conflicts, group conflicts, and organizational, envi-
ronmental, and policy conflicts. In the present international system, where
the sophistication and destructive capability of weapons make the violent
pursuit of conflict both costly and irrational and where there is no adher-
ence to a generally accepted set of rules or a central authority with the
power to regulate international behavior, mediation can be seen as an
ideal way of dealing with differences and settling conflicts between an-
tagonistic and fiercely independent states.This is one of the main reasons
why studies of mediation have proliferated in the last two decades or so.
All these studies, despite differences in focus, orientation, and methodol-
ogy, purport to describe how mediation can prevent the spread of a con-
flict and contribute to its resolution.
While there may be no definitive answers, scholars from disciplines
as diverse as anthropology, psychology, political science, sociology, law,
and communications have all attempted to further our understanding of
how mediation works and under what conditions is it effective.Their col-
lective efforts have often been bedeviled by the immensity of the scope of
mediation, the secrecy that normally surrounds it, and the difficulty of
studying mediation, and especially international mediation, in its natural
setting. The upshot of all this is minimal consensus on a definition of me-
diation or on the best method of conducting research on mediation, and
little relevant information on its frequency and success in international
relations. Scholars of mediation, whether engaged in case studies of an
anecdotal nature or large-scale comparative studies, appear to inhabit their
own little islands.To ensure that they are not out of touch with each other
and that their own little islands actually converge to form a body of knowl-
edge, students of the field, whichever research strategy they adopt, have
to define mediation and indicate its unique features, find connections be-
tween structures and mediation behavior, and learn to compare and evalu-
ate the contribution of mediation in a large number of cases. All these
avenues of research are represented in this book.
Whether we pursue research that uses quantitative methods to study
many instances of mediation and achieve a high degree of generalizability,
laboratory experiments or computer simulations to achieve a high degree
of control in the precision and measurement of variables, single cases of
mediation to achieve a high degree of descriptive accuracy, or theoretical
modeling to increase universal understanding, is a less important issue
than the necessity of making our concepts and research strategies explicit
and of communicating across disciplines, theories, and methods. Neither
the relevance nor the findings of any research on mediation can be re-
jected a priori. Multiple approaches-both theoretical and methodologi-
cal-are required to achieve a convergence of findings that can improve
both our understanding and the practice and performance of mediation.
We can make this kind of progress by adopting, as we do here, a research
perspective that permits us to evaluate how mediation affects, and in turn
is affected by, the context, the participants, the strategies, and the nature
of a conflict.
Our approach to mediation recognizes that it is practiced in numer-
ous arenas; that it is, in many ways, a continuation of the parties' own
conflict-management efforts; and that it involves the noncoercive inter-
-
vention of a third party (who may be an individual, an ad hoc group, an
organization, or a state) who seeks to influence or resolve a particular
conflict.This is its primary objective, which mediators fulfill through reli-
ance on persuasion, appeals to logic, the use of information, and the ap-
plication of social-influence strategies.The mediators' objective of chang-
ing, reducing, or resolving a conflict legitimates their intervention. The
material, political, or other resources mediators invest in the process pro-
vide the rationale for their own motives and interests.The intertwining of
the parties' interests, the mediators' interests, and the overall interest of
changing the course or outcome of a conflict (without which interest me-
diation would be neither invited nor accepted) is one of the unique fea-
tures of mediation.
The perspective suggested here recognizes that mediation involves
(1) a relationship between two protagonists and a mediator, (2) behavior
of some sort, within a context, and (3) the outcomes consequent to that
behavior.The outcomes may have been caused by direct mediator behav-
ior, or they may have been facilitated indirectly by mediators removing
barriers to their occurrence, permitting them to occur, or not preventing
them from occurring.This may appear to be a fairly broad conception of
the possible impact of mediation, but it is one with decided advantages. It
allows us to embrace a spectrum of outcomes (and not just the binary
success versus failure), it permits us to analyze the relationship between
different kinds of mediation behavior and outcomes, it reveals the com-
plexity of the causal link between a mediator, the parties in conflict, and
an emerging outcome, and it acts as a guide in evaluating mediation out-
comes from the perspective of the parties, the mediators, or the observer
or scholar.
4 INTRODUCTION
INTRODUCTION
The inclusion of a mediator, in any arena, turns a dyadic conflict into
a triadic relationship. This transformation may be wholly planned or to-
tally accidental. At its heart, however, are the mediator's attempts to af-
fect the behavior, perception, and choice of outcome of the adversaries.
A mediator enters a conflict and becomes part of a voluntary and unique
conflict system. That system comprises two parties in conflict, each with
its own concerns and interests, a set of issues that may or may not be clear
to all, divergent attitudes, and behavior that is at best incompatible and at
worst downright hostile. No mediator enters such a system for altruistic
reasons only. Whether individuals, organizations, or states, mediators en-
ter a conflict system to do something about it, passively or assertively, and
to promote or protect any interests they may have. In this sense, a media-
tor is not unlike another party in the conflict-management process whose
behavior and performance-what it wants to do, chooses to do, or is per-
mitted to do-are as conditioned by the context and circumstances as the
behavior of the adversaries themselves. To understand mediation and its
impact, we have to know something about the context, the issues, and the
parties involved.
The peculiarities and distinctive features of mediation relate to the
fact that it is not a quasi-legal process, nor is it extraneous to the parties'
own conflict-management efforts. To be successful, mediation must be
congruent with, and complementary to, a given conflict in its context, and
to treat mediation as either a legal process or a disinterested input is to
miss important features that explain the relationship between a mediator
and the parties. This volume makes that relationship very explicit.
Mediation begins with the interaction of two conflict parties within a
unique context. These parties' reaction to their conflict and to the act of
mediation is the result of their particular experience, society, culture, and
structure. These features in turn affect how a mediator intervenes, what
strategies and outcomes are pursued, and mediation's impact on the par-
ties and the outcome. Studies on mediation should be organized and me-
diation itself should be analyzed in terms of a broad framework that places
mediation at the very heart of a conflict. This sort of framework encour-
ages us to explore how different contexts impinge on mediation, what
conditions are most conducive to success, what behaviors mediators use
in different contexts, and how mediators relate to different parties. Only
when such analysis has taken place can an evaluation of the impact of
mediation-always a tentative and perilous matter-be undertaken.
It is important to bear in mind that neither mediator roles nor media-
tion performance can be stipulated in advance. Generic principles pro-
moting better outcomes should be viewed with caution, as mediation is a
dynamic and flexible process and adaptability is its prized attribute and
its key to success. This adaptability is a source of strength for the practi-
tioner but a source of bewilderment for the scholar. Our ultimate aim
here is to understand this complex, responsive, and heterogenous process
and to provide a solid basis for assessing its performance. Only in this way
can we move the study of mediation from infancy to maturity.
THEMATICORGANIZATION
m e chapters in this book proceed developmentally from theoretical con-
siderations through an examination of the relationship between a media-
tor and the parties, an exploration of the mediation context, an illustra-
tion of the possible range of application in international relations, and
finally to the presentation of how novel approaches can expand the role
and relevance of mediation.
In their chapter, Bercovitch and Houston show just how relevant and
widespread mediation has become in international relations.The authors
review the literature on mediation, establish the background for assess-
ing its occurrence, and examine the conditions-both structural and be-
havioral-that determine its performance and effectiveness in interna-
tional relations.They use a theoretical framework and a large number of
mediation cases in the post-1945 era to show how aspects of a context
such as who the parties are, who the mediator is, and what the conflict is
all about have an impact on both the nature and outcome of mediation.
Mediation is found to be a reciprocal process and a contingent form of
political influence. The best guides to its effectiveness are the issues in-
volved, the intensity of the conflict, the parties' previous relations, and
the mediation strategy.
The second part of the book examines a question in the mediator-
parties relationship that has often been asked in analyzing prerequisites
for successful mediation: that of bias and impartiality. Is it necessary, pos-
sible, or even desirable for mediators to be totally unbiased and impar-
tial? The relationship between a mediator and the parties in conflict is
voluntary. That means that a mediator has to be acceptable to both par-
ties. Mediators need not be perceived as impartial or unbiased (is there
such an entity in international relations?) to be acceptable; they do need
to be seen as having access to resources and the ability to get the conflict
parties out of a no-win situation. It is resources and the ability to effect a
change, not the appearance of impartiality, that are the sine qua non of
effective mediation.
Carnevale and Arad examine the notion and consequences of media-
tor impartiality. In a paper that utilizes laboratory research experiments,
they demonstrate that effective mediation may well be undertaken by
biased or partial mediators, and that decisions concerning the acceptabil-
ity of mediators do not simply reflect the mediators' alignment. Numer-
ous actors, partial and impartial, may serve as mediators in international
\
conflicts. A perception of "fair decisions rairly arrived at" appears to be
more important in the approval of a mediator than the attachment to a
dated notion of bias and impartiality.
A further exploration of the mediator-parties relationship and the
notions of bias and impartiality, this time in actual conflict situations, is
undertaken by the authors of the next two chapters. Wehr and Lederach
analyze the Esquipulas peace process in Central America in 1987 and
find that in a number of cases a mediator from within the conflict envi-
ronment, an "insider and partial," is more likely to succeed than an "im-
partial outsider." This is certainly the case when a mediator is connected
to and known by both conflict parties,shares their culture, and is expected
to live with the consequences of mediation. Such mediators inspire trust,
an effective resource in mediation. The Contadora experience in Central
America suggests that there may be other international conflicts where
mediators need be neither neutral nor even external to the conflict. In
many contexts the events, persons, and attitudes may legitimate a media-
tor in the form of a sympathetic and respectful insider rather than a "neu-
tral" outsider.
Of all the actors in international relations, none could have a stron-
ger claim to impartiality and neutrality than the United Nations. As a
supranational organization, the UN was set up to transcend state inter-
ests and embody global concerns, and in the new political climate it finds
itself with greater decisional latitude and involvement in more and more
conflicts (such as in Sudan, Somalia, Bosnia, Cambodia, Zaire, and oth-
ers). Skjelsbak and Fermann examine the factors that account for the
involvement of the United Nations in so many international conflicts.They
find that the personal skills and moral status (derived from the Charter)
of the Secretary-General and the degree of cooperation between the per-
manent members of the Security Council are the most important assets in
the scope and effectiveness of UN mediation. It would seem that impar-
tiality does not define mediation; nor is it bound up with its successful
performance by individuals, states, or even international organizations.
The importance of the context in international mediation is treated in
Part 2 of the book. Many mediation attempts fail because the parties in
conflict make different assumptions about the process and have different
expectations regarding its outcome. When the parties have different as-
sumptions about conflict and different ideals, goals, and values about social
reality, mediation and conflict management in general are unlikely to be
effective. An exchange of meanings, a shared common ground, and an aware-
ness of cultural differences are prerequisites to successful mediation or con-
flict management. The chapters by Cohen and by Mandell discuss the rela-
tionship between culture and mediation in theory and in practice.
Cohen's chapter raises questions about the need for mediators to be
sensitive to the context and the value differences within a c0nflict.A new
INTRODUCTION /
i
form of mediation, cultural mediation, may be called for to help negotia-
tors remove their cultural blinders and perceive reality in essentially the
same fashion.The value of mediation that ignores cultural forces or con-
signs them to the periphery is very limited indeed.
Mandell focuses his attention on Kissinger's mediation efforts between
Israel and Syria after the 1973 Yom Kippur War. His analysis shows that
Kissinger's task-oriented "toing and froing" could not have produced a
durable settlement. As a mediator Kissinger was tempted by Western ne-
gotiation procedures and by the prospect of playing a central role, but he
completely ignored basic cultural disharmony between the parties and
failed to get them to agree on common norms. Any settlement emerging
from such mediation could only have been temporary and short-term. A
successful mediation is predicated on the awareness of cultural differences
and the creation of shared norms.
The new international environment is characterized by much conflict
and many structural problems. In the absence of an explicit framework
defining international response to these problems, mediation has emerged
as a consistent and effective instrument. It is no longer narrowly focused
on conflicts between states, but has now become relevant in internal, eth-
nic, and global environmental conflicts. The principles and practices of
mediation have become central to the international community's response
to conflict, and Part 3 presents a series of case studies that demonstrate
the range and contribution of mediation in international relations.
Rupesinghe's chapter examines problems of mediation in a protracted
conflict within a state. In such conflicts the human costs, often borne by
the civilian population, are high and the chances of settlement pretty low.
To achieve a successful outcome, mediation must proceed on a number of
tracks. An official, formal track can deal with high politics and achieve
agreement on principles, but only a series of citizen-based, informal tracks,
using a number of channels of mediation, can create the conditions for a
sustainable reconciliation. Mediation in internal conflicts requires a change
of attitudes and behavior of the elites as well as the peoples that can only
be accomplished by multiple mediation avenues. Their absence contrib-
utes to the risk of wider escalation.
Of the conflicts that have erupted since 1989, none has posed a more
serious challenge to the international community than the turmoil in the
former Yugoslavia. A variety of responses and policies have been devel-
oped to deal with the conflict. Individuals, former statesmen, single coun-
tries, groups of countries, and international organizations have all at-
tempted some form of mediation-regrettably without much success.
Webb, Koutrakou, and Walters examine the European Community's ability
to mediate a conflict in its very midst. They find that most mediation at-
tempts have been ill-fated, whether because they were dictated by do-
mestic political considerations or because they proceeded along one di-
ci INTRODUCTION INTRODUCTION 9
\
mension only.To be effective in complex internal conflicts, mediation must
proceed along several avenues, develop clear objectives, and use all the
tools available to it. Complex conflicts call for complex mediations, and
any other response will simply fail.
Threats to international peace and security can now come from a va-
riety of sources, among the most important of which are environmental.
Issues such as pollution, soil erosion, desertification, and deforestation
generate conflicts as intense as any political issue. Environmental con-
flicts also usually transcend national jurisdiction and involve multiple is-
sues and parties, technical uncertainty, and considerable disagreement on
their implications. Shmueli and Vranesky examine the structure of me-
diation that is most pertinent to environmental conflicts and highlight the
roles mediators enact in such conflicts. Successful mediation in environ-
mental conflicts requires a search for multilateral consensus and the man-
agement of an enormous complexity of information and organization.
Given such difficulties, environmental mediation is best embedded in an
international structure such as the Organization of Economic Coopera-
tion and Development (OECD) or the United Nations Conference on
the Environment and Development (UNCED).
The final chapters of this book seek to expand our conception and
the theoretical range of mediation. These chapters challenge some of our
basic assumptions about conflict, resources, goals, and mediators' identi-
ties and interests. They represent something of a departure from tradi-
tional conceptions of mediation and a significant step in the evolution of
a wider scope for mediation roles.
Kriesberg argues for a conception of mediation that is context-de-
rived. The structure of a conflict, its context, the presence or absence of
other actors, political pressures rising from within, and the overall rela-
tions between the parties all have a considerable impact on who can me-
diate a conflict and how that mediator can behave. The chapter suggests
that mediation services may be provided by different actors, both formal
governments and informal individuals (or quasi-mediators), each imple-
menting a unique role to "unblock contextual problems" at different phases
of the conflict. Moving from a conflict to a settlement requires a series of
steps and transitions; quasi-mediators may have a crucial role in the ini-
tial phases, and having mapped the road, may allow formal mediators to
take the parties through the negotiation and implementation phases.
Keashly and Fisher expand our conception further by introducing the
consultation, or problem-solving, model and arguing for the matching of
mediation models to conflict situations.The consultation approach is based
on generic principles of conflict; it emphasizes an analysis of conflict, in-
terests, and needs, focuses on perceptions and cognitions, requires honest
communication, and is facilitated by a skilled third party or mediator. At
the heart of this chapter is the contingency approach, which suggests that
effective mediators who can properly read the conflict cues begin with
one mediation strategy, usually consultation, and when they achieve the
results anticipated move on to a more formal strategy. Ideal mediation
should be matched to the sequential stages of escalation until the parties
themselves are in a position to coordinate their de-escalation.
The chapters in this volume make it evident that resolving conflict
through mediation requires the conscious efforts of human beings work-
ing in a very fraught and complex arena. How well their efforts are under-
stood by people in the mediating profession and scholars determines how
successful mediation is going to be. Mediation does have its own forms,
logic, rules, and mechanisms. Understanding its many facets requires an
understanding of the parties' positions on different issues, their overall
relationship, their resources, and their motivation to settle their conflict.
Mediation cannot be viewed simply as an addendum to negotiation, or as
a set of techniques imported by someone completely removed from the
dispute to correct the perceptual errors of those in conflict. It is an inte-
gral part of the process of negotiation and conflict management in which
each actor, the mediator included, interacts with the others, exerts an in-
fluence, and seeks to promote a specific outcome. This is both the objec-
tive and rationale of international mediation.
To understand how mediation works means understanding the con-
text of a conflict as well as the background, interests, needs, and resources
of the participants.The nature of the influence process and the modalities
of influence used by a mediator are dependent on these factors.The form
the mediation takes, the role of a mediator, even the criteria used to evalu-
ate mediation outcomes are all functions of the broader context of a con-
flict situation. This is why we often describe mediation as a contingent or
reciprocal aspect of conflict management. One misses a lot if one ignores
the reciprocal interaction between the mediators and the mediated.
Mediators in international relations are agents of influence, promot-
ers of cooperation, and catalysts for change.They achieve their objectives
through the use of information, communication, facilitation, interpreta-
tion, and the waving of carrots and sticks. Mediation should not be con-
fused with altruism; mediators are usually cognizant of their own inter-
ests and they have motives, consciously expressed or not, that they wish
to see promoted or protected. This interplay of actors, motives and inter-
ests, resources, and context gives mediation its unique features.
International mediation is an effective form of conflict management.
The chapters that follow try to create a shared arena of discourse on me-
diation and convey the richness and complexity of the process by illus-
trating how, and under what conditions, mediation works. It is hoped that
this book can also help us to develop clear and practical political policies
for conflict management, and to define the mediation instruments most
effective in achieving these objectives.
The Study of International
Mediation: Theoretical Issues
and Empirical Evidence
Jacob Bercovitch and Allison Houston
C~f l i c t t s one of the most pervasive-and inevitable-ceatures of all so-
cial systems, however simple or complex they may be and irrespective of
their location in time and space. This is true of personal, group, and orga-
nizational, as well as international systems. Wherever it occurs, conflict is
significant, newsworthy, and challenging. It can lead to mutual satisfac-
tion and growth or it may produce acrimony, hostility, and violence. Our
interest in the study of conflict is undoubtedly related to our desire to
manage A -- it in a - way that maximizes its potential benefits and minimizes its-
destructive consequences.
There are a number of ways of dealing with or managing conflict.
These may range from avoidance and withdrawal, through bilateral ne-
gotiation, to various forms of third-party intervention. Third-party inter- -"
vention in conflict, particularly of the nonbinding, noncoercive kind, is in
many ways as old as conflict itself. It has played an important role in in-
dustrial and preindustrial societies. Its popularity as a way of dealing with
conflict grows each year, as does its applicability to different reaIms. Un-
resolved problems and conflicts create the conditions for third-party in-
tervention of one form or another.
Yet notwithstanding its popularity, longevity, ubiquity, and importance,
we know far less about %his form of conflict management than we imag-
ine. Systematic analyses, let alone empirical studies, of third-party inter-
vention in general and mediation in particular have been very rare. The
phenomenon has for too long remained little studied and poorly under-
stood. The purpose of this chapter is to redress this balance somewhat
and present an overview of the behavioral literature on mediation. We
intend to do so by organizing our material around the question that we
believe is central to an understanding of this problem: namely, what fac-
" '
tors and conditions determine the success or failure of mediation? Al-
a -
~ n c ~ I U U Y ut ll~ltliNAllONAL MEDIATION
i
though the focus of this chapter is the international environment, many
of the findings presented below can contribute to enhancing the effec-
tiveness of third-party intervention and mediation in other contexts.
THE NATUREOF INTERNATIONALMEDIATION
The practice of settling disputes through intermediaries has had a rich
history in all cultures, both Western and non-Western (Gulliver, 1979).
Although there are considerable differences in the way mediators from
different cultures deal with a conflict, all the approaches have value in
terms of managing or settling disputes. In the international arena, with its
perennial challenges of escalating conflicts, shrinking resources, and ris-
ing ethnic demands, and with the absence of generally accepted "rules of
the game," the potential application of mediation is trulv limitless.
Even a cursory survey of recent international disputes reveals the
extent and heterogeneity of mediation. In the last decade or so, we have
seen the involvement of such parties as the United Nations (in the Viet-
nam-Kampuchea dispute, the Iraq-Kuwait dispute, and the Yugoslav dis-
pute), the pope (in the Beagle Channel dispute), the organization of Af-
rican Unity (in the Tanzania-Uganda dispute), the Organization of
American States (in the Nicaragua dispute), the Arab League and Alge-
ria (in the Iraq-Kuwait dispute), the United States (in numerous efforts
in the Middle East). Less formal mediation efforts (by the Quakers, for
example, or by prominent politicians such as President Carter, Lord Owen,
or Prince Sihanouk) occur on a daily basis.
A? a form of international conflict management, mediation is likely
--
to occur when {llf a conflict has gone on for some time, (2jjthe efforts of
the individuals or actors involved have reached an impasse, (?)neith;ri
actor is prepared to countenance further costs or escalation of the disr
pute, andQ-bot h parties welcome some form of mediation and are ready
to engage in direct or indirect dialogue (Bercovitch. 1984).
, - Whatever its specific characteristics, mediation mus; in essence be
seen as an extension of the negotiation process whereby an acczpfable
third party intervenes to change the course or outcome of a particular
conflict. B e third party, with no authoritative decisionmaking power, is
there to assist the disputants in their search for a m u t u a l l ~ a ~ c ~ e
agreement. As a form of conflict management, mediation is distinguish-
able from the more binding forms of third-party intervention, such as ar-
bitration and adjudication, in that it is initiated upon request and it leaves
the ultimate decisionmaking power with the disputants (Folberg and Tay-
lor, 1984; Moore, 1986).
- -
In an arena such as the international one, where a largeand highly
diverse number of actors coexist, where each guards its sovereignty and
THE STUDY OF INTERNATIONALMEDIATION
13
i_ndep_endence zealously, and where each views the resort to violence as a
viable option, mediation with its ad hoc basis, voluntary nature, and non-
binding character offers a relevant and useful response to the problems
posed by ethnic, regional, and global conflicts.
The scope .---- of mediation activities in the international arena is truly
immen"Se.This is reflected in the abundance of definitions offered by vari-
&US students of the discipline. Doob, one of the "founding fathers" of the
field, uses the term in a very broad sense indeed. He defines it as "the
efforts of one or more persons to affect one or more other persons when
. . . the former, the latter or both perceive a problem requiring a resolu-
tion" (1993:l). Mediation thus purports to offer a solution to any prob-
lem the disputants perceive as such. Other definitions of mediation stress
its objectives. Mitchell defines it as "any intermediary activity . . . under-
taken by a third party with the primary intention of achieving some com-
-
promise settlement of issues at stake between the parties, or at least end- .
ing disruptive conflict behavior" (1981:287). Some definitions highlight
specific attributes in mediation or its dynamic phase structure. Folberg
and Taylor see mediation as "the process by which the participants, to-
gether with the assistance of a neutral person or persons, systematically
isolate disputed issues in order to develop options, consider alternatives,
and reach a consensual settlement that will accommodate their needs"
(1984:7).
The reality of international mediation is in some ways closest to Doob's
definition, but it is also more restricted. Mediators may intervene to pro-
tect the parties or to promote their own-interests. They may or may not
possess the required attributes of neutrality and visibility. They may be
external to the dispute or they may come from within the dispute envi-
ronment.Their mediation may be passive or active, and it may purport to
change some aspects of behavior or to affect the perceptions of the par-
ties. Mediators may act as individuals or as members of a larger ad hoc'
group. Taking all these features into account, we see international media-
tion as a reactive process of conflict management whereby parties seek :
the assistance of, or accept an offer of help from, an individual, group, or
,
organization to change their behavior, settle their conflict, or resolve their /'
problem without resorting to physical force or invoking the authority 6f
.the law (see Bercovitch et al., 1991:8; but compare Dryzek and Hunter,
1987; Wall, 1981).
This may be a broad definition, but the terminology here is quite sig-
nificant. Such a definition permits us to study cases in different contexts,
to employ different methods, and to make comparisons between cases.
Above all, we feel it forces us to recognize that mediation is a dynamic
and complex social process comprising parties in dispute, a social envi-
, : -
ronment or a context, a particular dispute or problem, and a mediating
agent. Our basic contention here is that the success or failure of any me-
I 6 THE STUDY OF INTERNATIONAL MEDIATION
THE STUDY OF INTERNATIONAL MEDIATION I /
to evaluate the impact of different kinds of mediation and assess the rela-
tionship between dispute characteristics and mediation (Bercovitch, 1989),
or evaluate the most successful mediation strategies (Bercovitch and Wells,
1993).
Mediation must not be analyzed or understood in terms of a simple
cause and effect model in which a particular strategy invariably produces
a desired outcome. Mediation in general, and international mediation in
particular, is not merely an exogenous input that can be applied uniformly
and indiscriminately to all disputes. Nor can it be presented only as a set
of rules, the rigid pursuance of which can affect or influence the dispu-
tants, their behavior, or their perceptions. The relationship between me-
diators and the disputing parties is reciprocal.Those involved in a dispute
wish to influence the mediator and the mediator certainly hopes to influ-
ence the parties. The contingency approach enables us to focus on this
reciprocal relationship through either a detailed study of a single media-
tion or the utilization of a large number of mediation cases.
We will pursue the latter course in the hope that our efforts consti-
tute a valuable step on the road to theory building.
METHODOLOGY
To start with, we should obtain information on the incidence of interna-
tional disputes and international mediation. A great many quantitative
studies describe the occurrence and analyze important patterns of inter-
national disputes. These are well summarized by Maoz (1982), Cioffi-
Revilla (1990), and Bremer (1993). None of these studies, however, ad-
dresses itself specifically to the question of how disputes are managed or
terminated. It is with this aspect that we are particularly concerned.
Our empirical examination was conducted through the utilization of
an original data set of international disputes from 1945 to 1990, which
was developed by us as part of a larger research project on the correlates
of international mediation. For operational purposes, we define an inter-
national dispute, as Singer and Small do (1982), as an organized and con-
tinuous militarized conflict involving at least one state and resulting in at
least 100 fatalities (Bercovitch et al., 1991). We identified 241 disputes
that meet our criteria. The geographical distribution of these disputes is
shown in Figure 1.2.
Compiling a list of all mediation events in international disputes is an
ambitious task. Routine informal, institutional mediations are carried out
behind closed doors between various internatiqnal actors on a daily basis.
Our task here is to examine the "nonroutine" mediation attempts that
were mentionedin public sources.:Of the 241 international disputes iden-
tified, 137 were actually mediated. Some disputes were mediated once,
Figure 1.2 Geographic Region of Disputes
30 o
others experienced multiple mediations. A thorough search of the (Lon-
don) Times, New York Times, and Keesing's Archives revealed 593 sepa-
rate mediation cases. That is, 593 reports of formal or institutionalized,
nonviolent, and nonjudicial interventions of an outsider or third party of
some sort occurred in these disputes. Figure 1.3 shows the relationship
between the number of disputes and the rate of mediation at successive
historical periods.
If we examine the prevalence of mediation vis-a-vis other conflict-
management activities, we get a fairly clear picture, in Figure 1.4, of just
how widespread the resort to mediation as a form of international con-
flict management really is.
Overcoming the problem of data availability brings us to the second
major issue: namely, exploring outcomes and relating them to a whole
host of contextual factors. Devising an index of successful mediation out-
comes is a complicated matter. There is little or no agreement on which
episode constitutes the outcome or how to identify a terminal point in
dealing with a dynamic and ever-changing process. Furthermore, media-
tion outcomes may be perceived and in turn defined very differently in-
deed by an observer, by the parties involved, by the international com-
Figure 1.3 Disputes and Mediation Cases by System Period
300
m
U1
Q 200
B
s
(r
6
S
Er loo
t
0
1945-1955 19561965 1966-1975 1976-1985 1986-1990+
System Period
Disputes initiated
Mediation cases
Conflicts in progress
Figure 1.4 Conflict Management Type
munity, or by the mediator himselflherself. Outcomes may be considered
successful at one point only to be deemed unsuccessful a few years later.
Mediation outcomes may also be defined as successful or unsuccessful
depending on the extent to which they meet certain normative criteria
(for example, has mediation produced greater fairness, efficiency, legiti-
Each of these perspectives has its limitations. Here we have decided
modify Haas's success index (1986) and focus on the behavioral conse-
on rather than on such factors as efficiency, legitimacy,
or short- or long-term success. We define a mediation as suc-
ade a considerable and positive difference to the
conflict and the subsequent interaction between the
is deemed to be partially successful when it has initi-
negotiations and a dialogue between the parties. It is defined as be-
en it has achieved a cease-fire or a break in hos-
this with a second outcome category, failure, which
hen mediation has had no discernible or reported
te or the parties' behavior (Bercovitch et al., 1991).
iation outcomes can be seen in Figure 1.5 (we
excluded from our analysis "no mediation" and "mediation offered
t is inappropriate to regard these as either failed or
ccessful attempts).
It might be tempting to assume that mediation can produce a success-
1 outcome in most types of international disputes. Yet as shown in Fig-
re 1.5 Mediation Outcomes
Full Settlement 7.8%
ure 1.5, an analysis of the data indicates !t this is not the case. Fully 55
percent (325 cases) of mediation attempts were unsuccessful, compared
with only 45 percent (267 cases) of successful cases (we collapse all three
gradations of successful outcomes into one measure). The relatively low
number of successful mediation outcomes may be explained in terms of
our very strict behavioral definition of success in international disputes.
Others (for example, Frei, 1976) consider mediation successful when it is
offered only.
The basic question we wish to address concerns the kinds of media-
tion cases that are successful. Which contextual attributes or factors ex-
plain successful mediation? Can such attributes or factors be used as "in-
formation inputs" by decisionmakers in determining whether or not to
initiate mediation? We would like to suggest that these questions and oth-
ers are best answered within the conceptual framework presented above
in Figure 1.1.
CONTEXTVARIABLES
Characteristics of the Parties
Parties'political context. A traditional hypothesis in the study of inter-
national relations suggests that democratic states are less likely to initiate
conflict than nondemocratic states. Mack and Snyder (1957) conclude that
the greater ability of democratic states to channel and accommodate in-
ternal discontent makes them less likely to exhibit external aggression.
Maoz and Abdolali (1989) found little empirical support for this hypoth-
esis. What they did determine was that democratic states are unlikely to
find themselves in a dispute with one another. What we want to know is
just how responsive democratic states, and indeed other states, are to in-
ternational mediation when they are in conflict. Does the type of political
system affect the chances of successful conflict management?
The political context can be divided into five regime types: monar-
chies, one-party states, military regimes, multiparty democratic states, and
others. Although we find that democratic states account for 30 percent of
all those states involved in mediation attempts, they usually resort to
mediation only when their adversaries are nondemocratic states. Only 6.7
percent of all mediation attempts involved disputes where both parties
were democratic states. Overall, we find that the kind of political system
does not greatly affect the chances of mediation success.
However, if we distinguish between symmetric dyads (disputes where
both parties share the same political system) and asymmetric dyads (where
the disputants are from different political systems) we find that there is a
53 percent probability of successful mediation in symmetric disputes com-
I
pgred to only 41 percent in asymmetric disputes. Mediation is easier when
the parties share a political system or have a basic adherence to the same
set of cultural norms and values. Shared norms and sociopolitical similar-
ity minimize misperception and facilitate a successful conclusion to a con-
flict.
We ---C-cP*5).-., can also - examine how such factors as internal composition, cul-
t upl and ethnic differences, and degree of homogeneity affect the suc-
diss or failure of mediation. Conflict management by third parties can
occur only between adversaries that have well-defined and legitimate iden-
tities. A mediator's job is hardly likely to prove easier if the incumbent
government of one of the adversaries is experiencing an insurgency, re-
bellion, or any other serious internal threat. Mediation has a better chance
of success when each disputant is accorded legitimacy. Disunity or lack of
cohesion within a state makes it difficult for both the adversaries as well
a5'a niediator to engage in any meaningful form of conflict management.
successive failures of mediation attempts in Lebanon, Cyprus, and
the former Yugoslavia illustrate this point only too well.
The relationship between internal unity and successful conflict man-
agement has been alluded to by many (for example, Burton, 1968;
Modelski, 1964). Raymond and Kegley state that "the greater the cultural
differences between disputants, the less likelihood of successful media-
tion" (1985: 38). Kressel and Pruitt (1989) support this argument by sug-
gesting that internal discord within a state has a negative impact on its
interactions with other states.The greater the fractionation within a state,
the greater the cha . This may seem intuitively
pTusible. When we between homogeneity, eth-
nic composition, and mediation outcomes we find that mediations where
one or both of the parties are culturally fragmented have a 54.4 percent
chance of success. Where one or both of the parties have no significant
cultural minority, the chances of successful mediation are as high as 64.4 a
percent. Where one or both parties have a significant cultural minority,
the chances of mediation success are only 38.4 percent.
s'power. Another contextual factor relates to the relative power
status of the parties in conflict. Is there a relationship between the power
and capabilities of states, or their discrepancies, and mediation effective-
ness? Ott (1972) and Young (1967) suggest that the smaller the power
differences between the adversaries, the greater the effectiveness of in-
ternational mediation. ~ o ~ i c a l l ~ this seems quite obvious. In cases of clear'?
power disparity, the stronger adversary may not be prepared to make any
concessions or compromises that are essential to mediation success. Yet
others, such as Deutsch, argue that conflicts in which there is "a mutual
recognition of differential power and legitimacy" (1973:46) will be more
22 THE STUDYOF INTERNATIONALMEDIATION
easily resolved. The presence of a fairly unambiguous advantage by one
of the parties may well create a clearer incentive toward a settlement.
The idea that mediation is most effective in disputes involving adver-
saries with equal power receives strong empirical support from a reinter-
pretation of Butterworth's data (1976). In a study examining power re-
sources and the impact of international mediation, a clear pattern emerged
showing high mediation impact (that is, abatement or settlement of a dis-
pute) when power capabilities were evenly matched and low impact or no
- - - - - - - - - - - --- ---
impactwhen -- power disparity - was high (Bercovitch, 1985). In this chapter,
---___
we wish to replicate this examination with a much larger data set. The
parties' relative power was measured using the Cox-Jacobson scale, in-
corporating measures of states' gross national product (GNP), military
spending, GNP per capita, territorial size, and population (Cox and Jacob-
son, 1973). We find that where the power disparity between parties is small
(0-4 on our scale), that is, where there is little difference in the power
resources possessed by each party, the chances of successful mediation
are 51.4 percent, compared to only 33.3 percent where power disparity is
great (11+ on our scale).
Previous relations between the parties. An international dispute is not
an isolated event; it has a past and presumably some sort of a future. It
may occur between parties who have had a history of friendship or one of
enmity. Past events and interactions cannot be discounted. Indeed, the
previous relationship between the parties is cited by Deutsch (1973) as
one of the main variables affecting the course and outcome of a conflict.
To assess the impact of the dynamics of a relationship, we divided the
parties' previous relationships into five distinct categories: friendly, an-
tagonistic (unfriendly but without previous conflict), conflictual (previ-
ous low-level conflict), one previous dispute between the parties, and more
than one previous dispute. We found that where the parties' previous re-
---
lationship was friendly the probability of successful mediation was 80 per-
cent. Where the partiesliad had one or more than one previous dispute
(such as in the Korean, Arab-Israeli, and Pakistan-Indian conflicts) the
probability of success was only 40 percent and 46 percent respectively.
Clearly the historical context of a dispute exerts a strong influence on the
manner of its management and likely outcome.
The Nature of the Dispute
-. " - -- - --
There is general agreement in the literature that "the success or failure of
mediation is largely determined by the nature of the dispute" (Ott, 1972:
597). Naturally the choices of conflict management modes and the chances
of successful mediation are affected by the importance each adversary
attaches to the issues in the dispute. When vital interests are affected (for
-- - -
example, issues of sovereignty or territorial integrity), intermediaries will
be quite unlikely to have any impact. Can we, howevego beyond this
&her obvious point and identify dispute aspects more specifically and
assess their effects on international mediation? In this section we try to
do that.
ration and timing of intervention. To begin with, the duration of a
nd the timing of initiating mediation may to a large extent deter-
e likelihood of its success. To be effective, mediation must take
a propitious moment. ow ever, little agreement occurs on how
recognize when a conflict is "ripe" for mediation. Edmead (1971) claims
at mediation is more likely to succeed if it is attempted at an early stage,
ertainly well before the adversaries cross a threshold of violence and
begin to inflict heavy losses on each other. Others, such as Northedge and
Donelan (1971), Ott (1972), and Pruitt (1981), sugge&&t_mediation is
ore effective when a dispute has gone through a few phases and must
r@inly not be initiated before each side has shown a willingness to mod-
rate its intransigence and revise its expectations.
An analysis of the mediation data set and direct interviews with ex-
perienced intermediaries both lend support to Northedge and Donelan's
proposition (Bercovitch, 1984). It seems that mediation is slightly more
effective when it follows, rather than precedes, some "test of strength"
ween the disputants. At some stage in a dispute, here at around seven
hirty-six months (by which stage the adversaries presumably have ex-
usted other modes), mediation efforts show the greatest degree of suc-
ess (50.2 percent), compared to a 47.4 percent chance of success for ear-
r interventions. Attempts made at over thirty-six months are considerably
s successful in mediation (38.6 percent).
lities and intensity at time of intervention. Closely related t o the tim-
g of mediation is dispute intensity. The costs incurred by t he parties
from continuing a conflict may at some point become so overwhelming
G t e t h e r losses are deemed intolerable and unacceptable. Her e again
he literature on mediation offers two contradictory points of view. Jack-
son (1952) and Young (1967,1968) suggest that the greater the intensity
f a dispute, the higher the likelihood that mediation will be accepted and
e successful (as a way of cutting losses, if nothing else). An opposing
view contends that the greater the intensity and the higher the losses, the
more polarized the parties' positions will become and the more deter-
mined each party will be to reject any mediation effort and attempt to
"win" at all costs (Brockner, 1982; Burton, 1969; Modelski, 1964).
To test these hypotheses, we evaluated the intensity of a n interna-
tional dispute by the number of fatalities and related this to mediation
outcomes. We found a clear and significant relationship between low fa-
24 THE STUDY OF INTERNATIONALMEDIATION
I
talities and successful mediation. Just over 39 percent of mediation at-
tempts had any degree of success in disputes with more than 10,000 fatali-
ties, compared to 64 percent of successful outcomes in low-fatality (100-
500 fatalities) disputes. Although we examined fatalities in isolation from
other variables, they are clearly an important factor in determining the
success or failure of mediation.
Issues. The literature on mediation abounds with ideas linking media-
tion effectiveness to the nature of the issues in dispute. Ott sees the "ab-
sence of vital national security interests, particularly questions of territo-
rial control" (1972:616) as a necessary precondition for successful
mediation. Randle contends that "should a dispute affect vital security
interests of the parties, no amount of mediation by a third party is likely
to prevent the outbreak of hostilities" (1973:49).And Lall, himself a prac-
titioner as well as a student of international mediation, contends that "it
is one of the principles of international negotiation that when territory is
at stake, the party in possession tends to resist third party involvement"
(1966:lOO). What they all seem to say is that the adversaries' perception
of the issues is a key factor in determining whether or not to accept a
mediation initiative and in influencing whether it will have much success.
Issues in conflict are the underlying causes of a dispute.They may not
always be clear, but they refer to what the dispute is all about. Often more
than one issue may be involved, and the parties themselves may not agree
on what constitutes a disputed issue or on its relative importance. Here
we suggest five categories to describe and reflect the tangible and intan-
gible types of issues that may characterize international disputes: sover-
eignty, ideology, security, independence, and all other issues (including
ethnicity).
Sovereignty disputes refer here to disputes where the adversaries have
incompatible claims to a specific piece of territory (for example, Argen-
tina-England dispute over the Falklands/Malvinas, Iraq-Kuwait Gulf War).
Ideology disputes are defined as disputes based on strong disagreements
over the nature of a political system, basic values, or beliefs (for example,
Iran-Iraq, Korean, U.S.-Panama conflicts). Security disputes are disputes
over frontiers, borders, and territories (for example, 1967 Arab-Israeli war,
India-Bangladesh dispute, U.S.-Libya dispute). Independence disputes are
fought by countries seeking to liberate themselves from another state and
to determine their own national selfhood (for example, Mozambique-
Portugal, Lithuania-USSR, Bouganville-PNG disputes).
Looking at the data set, we see that sovereignty (36.3 percent) and
security (23.6 percent) are the most prevalent issues in contention in in-
ternational disputes. A more interesting feature is that ideology disputes
(50.4 percent chance of success) are more amenable to mediation than
security disputes (where only 40.7 percent of mediation attempts had some
success) and sovereignty disputes (44.7 percent success). Where disputes
were over issues of resources and ethnicity, the chances of successful me-
diation were even higher (70 percent and 66.7 percent respectively).The
nature of the issue can, and does, affect mediation outcomes.
There can be no doubt that issues in dispute, and how they are per-
ceived, make a difference to the probability of achieving a successful or
unsuccessful outcome.This argument is implicit in much of the writing on
tion. Here we provide some preliminary evidence to suggest the
of issues that are best handled by international mediation. Other
ies (for example, Kochan and Jick, 1978; Kressel, 1972) offer differ-
ature of t he Mediator
e identity and characteristics of a mediator have been cited by some
rett, Drieghe, and Shapiro, 1986; Carnevale, 1986;Young, 1968) as pre-
dictors of success. Others, such as Harbottle (1979), Kockan and Jick
(1978), and Ott (1972), do not view the mediator as a critical determi-
t,relegating himiher to a secondary position.Thus it seems possible to
ue, on one hand, that the personal characteristics of mediators as ma-
agents differentiate effective from ineffective mediation, or on the
er hand that personal traits are largely irrelevant. What really matters
nature of the dispute. Either way, it would be useful to investigate
lationship between mediators' characteristics and the effectiveness
diation is a voluntary process. '$is means that mediators cannot
"ess they are perceived as reasonable, acceptable, knowledge-
able to secure the trust and cooperation of the disputants. Jack-
self an experienced international mediator, makes this quite clear:
t would be difficult, if not impossible, for a single mediator, who was
trusted by one of the parties, to carry out any useful function" (1952:
ediation also depends not only on the mediator's knowl-
d conflict management but also on hisiher prestige and
iFjr"iiiginality of ideas, access to resources, and ability to act
un%t"iu'sively. h_rl_ _ _ In a theoretical discussion, Wehr (1979) lists the attributes
required for successful mediation as including knowledge about conflict
ituations, an ability to understand the positions of the antagonists, active ,
listening, a sense of timing, communication skills, procedural skills (for
example, chairing meetings), and crisis management.
The list of desired personal attributes for a successful international
mediator is very long indeed. Among the attributes that experienced in-
ternational mediators cite as particularly important are intelligence,
Gu I I IL dl uu r vr Il u 1 cl <l V/ - \ I I VI VHL MtUIHI I WN
personal qualities are associated wi t h success in other areas of human
endeavor; they are, of course, no less important in international media-
tion.Trust, credibility, and a high degr ee of personal skill and competence
in the mediator are also necessary preconditions for effective mediation
(Karim and Pegnetter, 1983; Landsberger, 1960).
Another characteristic that has traditionally been cited as being
strongly associated with effective mediation is even-handedness or im-
partiality. Young claims that "a high score in such areas as impartiality
would seem to be at the heart of successful interventions in many situa-
tions" (1967:81). His views are echoed by Jackson (1952), as well as
Northedge and Donelan (1971), who claim that parties will have confi-
dence in a mediator only if helshe is perceived as impartial.
We are, however, doubtful of t he importance of this attribute. The
traditional emphasis on impartiality stems from the failure-to_L-mgnize
mediation as a reciprocal process of social interaction in which the me-
diator is a major participant. It is entirely sensible to see mediation as
"assisted negotiation" (Susskind and Cruickshank, 1987); to regard me-
diation as an exogenous input is bot h erroneous and unrealistic. A media-
tor engages in behavior that is designed to elicit information and exercise
influence. Mediators are accepted by t he adversaries not because of their
impartiality but because of their ability to influence, protect, or extend
the interests of each party in conflict (Faure, 1989; Kressel and Pruitt,
1985; Smith, 1985; Touval, 1985; Touval and Zartman, 1985).
To exercise any degree of influence, mediators need "leverage," or
resources to search for information and move the parties away from rigid
positions. Leverage or resources buttress the mediator's ability to facili-
tate a successful outcome through t he balancing of power discrepancies
and enhancing of cooperative behavior.The mediator's task is also essen-
tially one of reframing and persuasion (Edmead, 1971; Frei, 1976; Gross-
Stein, 1985; Touval and Zartman, 1985). These strategies are most suc-
cessful, as Touval and Zartman (1985) observe, not when a mediator is
unbiased or impartial but when he/she possesses resources that either dis-
putant values. Effective mediation i n international relations is more a
matter of mediators' utilization of resources, leverage, and influence com-
mensurate with their position t o enhance fairness than it is of impartiality
(Brookmire and Sistrunk, 1980).
One of the most effective resources any international mediator can
possess is legitimacy. Leaders of st at es and high-level officials such as for-
eign or prime ministers have legitimacy and can bring it to bear together
with their status and respect (Ott, 1972; Touval and Zartman, 1985). Un-
der the proper auspices and sponsorship of high-ranking mediators, an
environment of credibility, trust, a nd joint interests may be established.
The presence of a powerful and legitimate mediator allows the parties to
I H t SIUVY Wt IN1 t l i NAI I WNHL Mt UI HI I WI U L/
\
back down from fixed positions, make concessions, and "save face" (Pruitt
and Johnson, 1970). Leaders and representatives of large governments
with more resources at their disposal will be more likely to be successful
mediators than other actors.
One way in which we can examine this notion empiri-
11mediators according to rank. Mediators have differ-
ssess different resources, both of which they use in differ-
in different disputes (Bercovitch, 1992). Reflecting the diversity
le mediators in the international environment, we have ranked
iators along a dimension ranging from government leaders and rep-
es of regional and international organizations to private indi-
In relating rank to mediation outcomes, we find that mediators with
ccess rate are leaders and representatives of regional organiza-
ns (62.4 percent and 50 percent respectively). They are followed by
sentatives of small governments (54.8 percent and 56.8
spectively). Interestingly, both leaders and representatives of
s fare rather worse than expected (40 percent and 31.3 percent
ectively), as do representatives of international organizations (only
organizations (such as the Organization of African Unity,
of African States, Contadora group, and the Economic
f West African States) with common ideals, perspectives,
ar to offer the best chances of successful outcomes in
iation. In contrast, international organizations such as
tions have a very poor record in the area of mediation. One
t an organization such as the United Nations usually deals
actable disputes that resist mediation and other forms
nagement, whereas regional organizations deal with less
s. A more compelling argument asserts that regional orga-
s mediate within the same cultural and value system-and
eems, promotes agreement more than any other factor.
relationship with the parties. We can also assess mediators' in-
everage, and legitimacy by examining their previous relation-
the adversaries.The importance adversaries assign to a continu-
tionship with a mediator may well influence their perception of a
and their behavior. Where a mediator is aligned with one of the
rties or shares a common experience or goals with one party and future
ractions are important to both, each disputant may show greater flex-
ty and confidence in the outcome. Mediator alignment, past relation-
lp with the adversaries, and the mediator's own interests affect both
mediator behavior and mediation outcomes.
28 THE STUDYOF INTERNATIONALMEDIATION
1
Frei (1976) found that mediators who shared religious, ideological, or
economic values had a higher chance of success than other mediators. We
examined the consequences of mediator alignment by looking at the po-
litical structure of mediator-parties relationships. Thus mediators may
come from the same political bloc as one or both of the parties, a different
bloc, be unaligned, or have no previous relationship with the parties. We
found that those mediators who came from the same bloc as both adver-
saries had a significantly higher chance of being successful (62.18 per-
cent) than mediators from the same bloc as one of the parties only (31.4
percent) or a mediator from a different bloc (51 percent). Mediators from
the same bloc have a strong self-interest in maintaining peace and stabil-
ity, and a greater chance of doing so.
PROCESS VARIABLES
The Initiation of Mediation
Thus far we have evaluated the impact of different contextual factors on
mediation. Here we consider the mediation process itself-the way it is
initiated and conducted.
A question that invariably confronts all potential mediators and par-
ties to conflict is whether or not they should take the step of initiating
mediation. It is agreed that to be effective mediation must take place un-
der the most propitious conditions, but just whose decision should it be to
intiate mediation?
Since mediation is essentially a voluntary process, it is logical to as-
sume that a conflict will be most constructively and effectively dealt with
through mediation when both the parties are willing to commit them-
selves to the process. Hiltrop (1985) suggests that mediation is most suc-
cessful when both of the adversaries request it. In cases where only one
party is interested in seeking mediation assistance, or interested third par-
ties propose it, the effectiveness of mediation may be reduced consider-
ably.
An analysis of our data set clearly indicates that mediation is most likely
to be successful (62.3 percent) when initiated by both parties to a dispute,
as opposed to only one of the parties requesting mediation (41.3 percent).
Surprisingly, mediation also appears to be highly successful (60 percent)
when regional organizations initiate proceedings, perhaps because of com-
mon interests and relationships shared with the parties. In contrast, media-
tion initiated by a mediator or by international organizations achieved only
41.2 percent and 29.7 percent success rates respectively.
THESTUDY OFINTERNATIONALMEDIATION
29
iation Environment
of the central tasks of mediation is to accentuate cooperation and
encies toward agreement.This is best achieved when the parties' con-
management takes place in a neutral environment, free from the ex-
a1pressures and influences of constituents and media. Such an envi-
ent allows the mediator to have procedural control over the process
es to concentrate on the more substantive issues. In a neutral
t, a mediator is able to create a level playing field by guaran-
-party free and equal access to iriformdion -and resources,
ining the flow of communication between the parties, and, where
essary, balancing power differences between the parties.
Our data confirm the importance of environment in the success of
ediation. Mediation on neutral ground (including mediator territory)
rovides the conditions most conducive to successful mediation, achiev-
g a 49.5 percent and 54.4 percent chance of success respectively. This is
contrast to mediations held on the parties' territories (45 percent) or
ediations that have moved among a number of different sites (36.4 per-
ent). An appropriate site, then, can considerably influence the success or
ating the relationship between what mediators do and the outcomes
ir efforts is, on the whole, based on ex post facto reflections by me-
tors (and they may be quite reluctant either to claim success or to take
ponsibility for failure) or on direct observations of their performance
rely, if ever, available in international relations). Although conceptu-
asuring mediator behavior, roles, and strategies is difficult,
e with Kochan and Jick (1978) and Touval(1982) that it is the
la1 variable affecting mediation outcomes.
There are many typologies for describing mediation strategy and be-
ovitch, 1984; Kochan and Jick, 1978; Kolb, 1983; Kressel, 1972).
low~Touval and Zartman (1985), who classify mediator be-
- . _----
hav&ralong a continuum ranging from low to high intervention. We iden-
h - three main strategies that encompass the spectrum of-mediator be-
havior (Bercovitch et al., 1991). At the low end of the spectrum are
Communication-facilitation strategies where a mediator takes a fairly pas-
sive role, largely as a channel of communication or go-between for the
parties, and exhibits little control over the process or substance of media- \
tion. In the second set of mediation strategies, procedural strategies, a '
mediator exercises more formal control over situational aspects or the
process of mediation. Here a mediator may determine such factors as the
30 THESTUDYOF INTERNATIONAL b. ATION
mediation environment, the number and type of meetings with the adver-
saries, the agendas covered in those meetings, the control of constituency
influences, and the distribution of information and resources t o the par-
ties. In the most active range of mediator behavior, directive strategies,
the mediator sets out to affect the content and substance as well as the
process of mediation. A mediator may achieve these goals by providing
incentives, offering rewards and punishments, issuing ultimatums, and in-
troducing new proposals.
Which of these strategies is most effective in international disputes?
Kochan and Jick (1978), for instance, found that industrial mediators who
used directive strategies were more successful than mediators who used
communication strategies. Carnevale and Pegnetter (1985), on the other
hand, in their survey of mediators found communication-facilitation strat-
egies to be more effective than other kinds. In an earlier study, Bercovitch
(1986) found that communication strategies were the most commonly used
but directive strategies were the most successful.
Here we examine the relationship between different mediation strat-
egies and mediation outcomes. A clear pattern emerges that shows that
the likelihood of achieving a successful mediation outcome is 52.3 per-
cent when directive strategies are employed and only 32.2 percent when
communication-facilitation strategies are used. This relationship is par-
ticularly strong when disputes are intense (Donohue, 1989; ~ i l t r o ~ , 1989).
Mediators who possess the ability, opportunity, and resources to initiate
and engage in active mediation are more likely to produce a successful
outcome than powerless mediators who put their faith in communication
strategies only. Procedural strategies, while used in only 16 percent of all
mediation attempts, have a 48.4 percent chance of success.
Mediation operates within a system of reciprocal social influence in
which the parties and the mediator seek to influence each other. Media-
tion behavior is not based on a specific predetermined plan of action but
reflects the changing context of a dispute and the interests and needs of
all concerned. A mediator has to be seen as a full, if external, participant
in a conflictual decisionmaking system. As such, it is not at all surprising
that possession of resources and an active strategy provide the basis for
successful mediation.
CONCLUSION
Third-party assistance or mediation is one of the most promising ap-
proaches to constructive conflict management. To understand it better,
we need to study what mediators do, how they do it, and the consequences
of their actions. In this chapter, we have sought to organize what is known
about mediation, extract propositions about how it works, and identify a
number of factors that are hypothesized to affect its outcomes. Although
THE STUDY OF INTERNATIONAL MEDIATION
3 1
we restricted ourselves here to the field of international mediation, our
findings are intended to be applicable to other mediation arenas.
Our understanding of mediation is predicated on the notion that
is related to the overall context in which it occurs. The overall
context affects mediation and is, in turn, affected by it. Hence mediation
can best be seen as an extension of bilateral conflict management, not a
replacement of it. The nature and effectiveness of mediation depend as
much on who the parties are and the character of their dispute and inter-
action, as on who the mediator is and hislher behavior. Mediation is a
contingent form of political influence; its performance and results are
contingent on context. To overlook this point is to mistake the nature of
mediation in international relations.
Here we have gathered information on a large number of mediation
cases and tested various propositions to identify the factors that are closely
correlated with the success or failure of mediation.The results of our analy-
sis offer some exciting insights into the most significant factors and are
summarized in Figure 1.6. It emerges that components of all four of the
clusters of our framework have a direct impact on mediation outcome.
results reveal that all the factors, from the disputants themselves-
they are, their past relationship, their sociopolitical norms and power,
ctual dispute, the issues at stake, its intensity and duration-to the
diator and the way hetshe influences the process of conflict manage-
nt, determine the success of mediation.
gure 1.6 Factors Associated with Successful Mediation
Nature of the
r i
Rank and Identity I
Relationship with
Parties
Nature of the Parties Mediation Process
Political and Cultural Initiator
Environment
Strategies
Previous Relationship
with Parties
Duration and Timing
of Intervention
Fatalities and Intensity
32 THE STUDYOF INTERNATIONAL' '31ATION
The approach and method adopted here are deliberately limited in
their scope. Future research will be directed toward specifying more clearly
how each factor, in itself and combined with others, transforms a conflict
system. We hope that the approach we have taken and the results pre-
sented here take us some way toward viewing mediation in the proper
perspective. By viewing it accurately, studying it in actual situations, and
acknowledging its richness and diversity we can provide the necessary
guidelines for more effective mediation in international relations.
REFERENCES
Bercovitch, J. (1984). Social Conflicts and Third Parties: Strategies of Conflict Reso-
lution. Boulder, Colo.: Westview.
Bercovitch, J. (1985). "International Mediation: Incidence and Outcomes." Un-
published paper. University of Canterbury, Christchurch, N. Z.
Bercovitch, J. (1986). "International Mediation: A Study of Incidence, Strategies
and Conditions of Successful Outcomes." Cooperation and Conflict 21:
155-168.
Bercovitch J. (1989). "International Dispute Mediation." In K. Kressel and D.
Pruitt, eds., Mediation Research. San Francisco: Jossey-Bass, 284-299.
Bercovitch, J. (1992). "The Structure and Diversity of Mediation in International
Relations." In J. Bercovitch and J. Rubin, eds., Mediation in International
Relations: Multiple Approaches to Conflict Management. New York: St.
Martin's.
Bercovitch, J., Agnoson, J. T., and Wille, D. (1991). "Some Contextual Issues and
Empirical Trends in the Study of Successful Mediation in International
Relations." Journal of Peace Research 28: 7-17.
Bercovitch, J., and Houston, A. (1993). "Influence of Mediator Characteristics
and Behavior on the Success of Mediation in International Relations."
International Journal of Conflict Management 4: 297-321.
Bercovitch, J., and Langley, J. (1993). "The Nature of the Dispute and the Effec-
tiveness of International Mediation." Journal of Conflict Resolution 37:
670-691.
Bercovitch, J., and Wells, R. (1993). "Evaluating Mediation Strategies: A Theo-
retical and Empirical Analysis." Peace and Change 18: 3-25.
Bremer, S. N. (1993). "Advancing the Scientific Study of War." International Inter-
actions 19: 1-26.
Brett, J. M., Drieghe, R., and Shapiro, D. L. (1986). "Mediator Style and Media-
tion Effectiveness." Negotiation Journal 2: 277-285.
Brockner, J. (1982). "Factors Affecting Entrapment in Escalating Conflicts." Jour-
nal of Research in Personality 6: 247-266.
Brookmire, D., and Sistrunk, F. (1980). "The Effects of Perceived Ability and
Impartiality of Mediator and Time Pressure on Negotiation." Journal of
Conflict Resolution 24: 311-327.
Burton, J. W. (1968). Systems, States, Diplomacy and Rules. Cambridge: Cambridge
University Press.
Burton, J. W. (1969). Conflict and Communication. London: Macmillan.
Burton, J. W. (1972). "The Resolution of Conflict." International Studies Quar-
terly 16: 5-29.