National minorities, nationalizing states, and external national homelands
in the New Europe
Rogers Brubaker
Twice in this century, Central and Eastern Europe have undergone a massive and concentrated re
configuration of political space along national lines. In the first phase of this re configuration (which
actually began in the nineteenth century), the crumbling of the great "traditional" multinational land
empires ­ the prolonged decay of the Ottoman Empire and the sudden collapse, in the First World War,
of the Habsburg and Romanov empires ­ left in its wake a broad north-south belt of new states in East
Central Europe, stretching from the Baltic littoral to the Balkan peninsula. In the second phase, the
disintegration of the Soviet Union, Yugoslavia, and Czechoslovakia and the emergence of some twenty
new states in their stead have resulted in the nationalization of political space on a much vaster scale,
extending from Central and Eastern Europe eastward across the entire breadth of Eurasia. Like the
nationalizing settlement that followed the First World War, the most recent reconfiguration of political
space along ostensibly national lines has conspicuously failed to "solve the region s long-refractory
national question. Yet while nationalist tensions have not been resolved, they have been restructured. This
chapter addresses this new phase and form of the national question, focusing on the triadic nexus linking
national minorities, nationalizing states, and external national "homelands," and illustrating its dynamically
interactive quality with a discussion of the breakup of Yugoslavia.
A triadic configuration
The triadic relational nexus has been engendered, or given new urgency, by the new (or at least newly
salient) mismatch between cultural and political boundaries. The massive nationalization of political space
in the region has left tens of millions of people outside "their own" national territory at the same time that
it has subjected the "national" quality of persons and territories to heightened scrutiny. Foremost among
these nationally "mismatched" persons, as suggested in Chapter 2, are the 25 million ethnic Russians,
abruptly transformed from state-bearing nationality in a vast and powerful state into vulnerably situated
minorities of uncertain identity and loyalty in weak and struggling successor states. But many other groups
have a similar, structurally ambivalent member-ship status, belonging by residence and (in most cases) by
formal citizenship to one state and by putative ethnonational affinity to another. These include ­ to name
only a few of the more important ­ some 3 million ethnic Hungarians in Romania, Slovakia, Serbia, and
Ukraine, whose relations with Hungary, limited during the communist era, have multiplied and intensified
in recent years; the 2 million Albanians in Serbia, Montenegro, and Macedonia, whose ties to neighboring
Albania have been renewed and strengthened; the nearly 2 million Serbs living (before the war) in Croatia
and Bosnia-Hercegovina, who, as Yugoslavia began to disintegrate, looked to Serbia as their external
national homeland; the nearly one million Turks in Bulgaria; the Armenians in Azerbaijan, especially in
Nagorno-Karabakh; the Uzbeks in Tajikistan and the Tajiks in Uzbekistan; and the Poles in Lithuania and
other Soviet successor states.1 All of these groups, as well as numerous smaller groups (or potential groups,
since in many cases their "groupness" is more a political project than a social fact),2 must contend not
only with political and economic reconfiguration and dislocation but also with two mutually antagonistic
1
I provide numbers in the text only as a very rough indicator of the orders of magnitude involved. The figures are
invariably contested; indeed disputes concerning the size of putative nations and national minorities have long been central to
nationalist politics. It is an illusion ­ and one that contributes to the prevalence of ethnic nationalism ­ to think that one could
somehow arrive at objectively "correct" figures. Nationality is not a fixed, given, indelible, objectively ascertainable property;
and even subjective, self-identified nationality is variable across time and context of elicitation, and therefore not measurable
as if it were an enduring fact that needed only to be registered.
2
In sheer numbers, Ukrainians ­ that is, those who identified their nationality as such in the 1989 Soviet census ­ in
Soviet successor states other than Ukraine are, at more than 6 million, more numerous than any of the other minority groups
except Russians. But the "groupness" suggested by this distinct statistical existence is, from a sociological point of view,
largely illusory. Both in the Russian Federation, where over 4 million self-identified Ukrainians lived in 1989, and in other
successor states, Ukrainians have tended to assimilate linguistically to, and intermarry with, Russians. Although some
political entrepreneurs have tried to mobilize Ukrainians as a national minority distinct from Russians, this "group-making"
project is unlikely to succeed.
nationalisms ­ the "nationalizing" nationalisms of the states in which they live, and the "homeland"
nationalisms of the states to which they belong, or can be construed as belonging, by ethno-cultural
affinity though not (ordinarily) by legal citizenship. All are, therefore, inscribed in the triadic nexus linking
the minority communities themselves, the states in which they live, and their external national
"homelands."
That relationship is not everywhere and always conflictual. In the case of the residual, though still large,
German minority in Eastern Europe and the former Soviet Union, 3 for example, the triangular
relationship has a unique and largely nonconflictual configuration. This contrasts starkly with the interwar
period. Then, too, there was a dynamic triangular interplay between the huge German national minority,
the newly nationalizing states of East Central Europe in which they lived, above all Poland and
Czechoslovakia, and Germany as their external national "homeland." That relation was deeply conflictual,
even in the Weimar period, and it became fateful after the Nazi seizure of power. Today, by contrast,
guaranteed immigration and citizenship rights in a prosperous and stable external "homeland," together
with the new freedom of exit from the countries of Eastern Europe and the former Soviet Union, act as a
powerful solvent, and magnet, on German minority communities, causing their steady depletion through
heavy emigration. Within a generation, these rights extended to ethnic co-nationals by the Federal
Republic of Germany are likely to lead to a final dissolution of the centuries-old German presence in
Eastern Europe.
In cases where the triangular relationship is more deeply conflictual, however, the new Europe, like
interwar Europe, confronts a potentially explosive ­ and in some cases actually explosive ­ dynamic
interplay between a set of new or newly reconfigured nationalizing states, ethnically heterogeneous yet
conceived as nation-states, whose dominant elites promote (to varying degrees) the language, culture,
demographic position, economic flourishing, or political hegemony of the nominally state-bearing nation;
the substantial, self-conscious, and (to varying degrees) organized and politically alienated national minorities
in those states, whose leaders demand cultural or territorial autonomy and resist actual or perceived
policies or processes of assimilation or discrimination; and the external national "homelands" of the minorities,
whose elites (again to varying degrees) closely monitor the situation of their co-ethnics in the new states,
vigorously protest alleged violations of their rights, and assert the right, even the obligation, to defend
their interests.
Since the last term is vulnerable to misunderstanding, I will characterize it a bit more fully. By
"homeland" I do not mean the actual homeland of the minority, in the sense that they or their ancestors
once lived there. That is not necessarily the case. Nor need the minority even think of the external state,
or the territory of that state, as its homeland. External national homelands are constructed through
political action, not given by the facts of ethnic demography. A state becomes an external national
"homeland" for "its" ethnic diaspora when political or cultural elites define ethnonational kin in other
states as members of one and the same nation, claim that they "belong," in some sense, to the state, and
assert that their condition must be monitored and their interests protected and promoted by the state; and
when the state actually does take action in the name of monitoring, promoting, or protecting the interests
of its ethnonational kin abroad. Homeland politics takes a variety of forms, ranging from immigration and
citizenship privileges for "returning" members of the ethnic diaspora, through various attempts to
influence other states' policies towards its co-ethnics, to irredentist claims on the territory of other states.
National minorities, nationalizing states, and external national homelands are bound together in a single,
interdependent relational nexus. Projects of nationalization or national integration in the new nation-states,
for example, "exist" and exercise their effects not in isolation but in a relational field that includes both
the national minority and its external national homeland. In this relational field, minority and homeland
elites continuously monitor the new nation-state and are especially sensitive to any signs of projects of
"nationalization" or "national integration." When they perceive such signs, they seek to build up and
sustain a perception of the state as an oppressively or unjustly nationalizing state. And they might act on
this perception. The minority might mobilize against the perceived projects of nationalization and might
seek autonomy or even threaten secession. The homeland, claiming the right to monitor and protect the
3
It is difficult to give an even approximate estimate of the size of the German minority. The immigration and
citizenship rights extended by Germany to Germans in Eastern Europe and the former Soviet Union have created
a strong incentive for people with any familial connection to Germany or Germanness to identify themselves as
German. In this way the large-scale migration of East European and ex-Soviet Germans to Germany, by
inducing ethnonational reidentification as German, may for a time increase rather than decrease the size of the
German minority in the region.
interests of its ethnic co-nationals abroad, might provide material or moral support for these initiatives
and might lodge protests with the nationalizing state or with international organizations against the
perceived projects of nationalization. This protest activity will react back on the nationalizing state,
although it will not necessarily dissuade it from its nationalizing projects, and it might even lead to their
intensification. The minority might be accused of disloyalty, the homeland of illegitimate interference in
the internal affairs of the nationalizing state.
The dynamic interdependence linking national minorities, nationalizing states, and external national
homelands in a single web of relations calls for analysis in its own right. This requires a reorientation in
the study of nationalism. While the burgeoning corpus of work on nationalism includes large, if dated,
literatures on national integration in new states and on national minorities, as well as a smaller, more
recent literature on state intervention on behalf of co-ethnics in other states, these have remained isolated
from one another. I know of no studies that develop an explicit analytical or theoretical account of the
relational nexus linking national minorities, nationalizing states, and external national homelands.4 Some
studies of particular nationalist situations ­ especially historical ones ­ are sensitive to interactive dynamics
of this sort,5 but none, to my knowledge, has worked out an explicit `model or provided a sustained
analytical discussion of the relational field and its interactive dynamics. To begin to do so is the task of this
chapter.
National minorities, nationalizing states, and external national homelands as arenas of
struggle
To invoke, as I have done, a relationship between three terms might suggest that the terms themselves
are fixed and given. They are, however, not fixed entities but variably configured and continuously
contested political fields. Thinking of what we summarily call national minorities, nationalizing states, and
external national homelands as political fields is a useful way of making explicit the fact that these are
dynamic and relational concepts and should not be reified or treated in a substantialist fashion.
National minority
A national minority is not simply a "group" that is given by the facts of ethnic demography. It is a
dynamic political stance, or, more precisely, a family of related yet mutually competing stances, not a static
ethno-demographic condition. Three elements are characteristic of this political stance, or family of
stances: (1) the public claim to membership of an ethnocultural nation different from the numerically or
politically dominant ethnocultural nation; 6 (2) the demand for state recognition of this distinct
4
Myron Weiner, "The Macedonian Syndrome," World Politics 23, no. 1 (1970), comes closest to developing such an
account, outlining a "syndrome" of predictably covarying characteristics typically found when an irredentist state confronts
an anti-irredentist neighboring state in connection with a border-straddling ethnic group. The account offered here differs
from Weiner's chiefly in three respects. First, Weiner is concerned only with irredentist claims and disputed borders, while I
am concerned with the broader field of homeland politics, in which irredentism is a limiting case. Second, Weiner is
concerned with all border-straddling ethnic groups associated with border disputes, while I consider only national minorities
whose co-ethnics are numerically or politically dominant in another state that can, for this reason, be construed as their
"external national homeland." Third, while Weiner specifies a "syndrome" of covarying characteristics, I emphasize the
contingency and variability of the relations between national minorities, nationalizing states, and external national homelands
­ contingency and variability that follow from treating each of these three "elements" as fields of struggle among competing
positions or stances, and from seeing the relations between these three fields as closely intertwined with relations internal to
the fields. More recently, an emergent literature on diasporas in international politics has begun to explore the triadic relation
between diasporas, host states, and home states, but it focuses on migrant diasporas rather than consolidated national
minorities, settled, in considerable part, in compact areas directly adjoining their respective national homelands. See Gabriel
Sheffer, "A New Field of Study: Modern Diasporas in International Politics," in Sheffer, ed., Modern Diasporas in
International Politics (London and Sydney: Croom Helm, 1986).
5
For sophisticated studies of the national question in interwar Europe, alert to this dynamic, see Joseph Rothschild, East
Central Europe Between the Two World Wars (Seattle: University of Washington Press, 1974); Ronald M. Smelser, The
Sudeten Problem, 1933-1938: Volkstumspolitik and the Formulation of Nazi Foreign Policy (Folkestone, UK: Dawson, 1975);
Rudolf Jaworski, Vorposten oder Minderheit? Der sudetendeutsche Volkstumskampf in den Beziehungen zwischen der
Weimarer Republik und der CSR (Stuttgart: Deutsche Verlags-Anstalt, 1977); C. A. Macartney, National States and National
Minorities (London: Oxford University Press; 1934); and C. A. Macartney, Hungary and Her Successors: The Treaty of
Trianon and Its Consequences, 1919-1937 (London: Oxford University Press, 1937).
6
This suggests why it is difficult to assert a status as national minority in states such as the United States that do not have
clear dominant ethnocultural nations. If the nation that legitimates the state as a whole is not clearly an ethnocultural nation
ethnocultural nationality; and (3) the assertion, on the basis of this ethnocultural nationality, of certain
collective cultural or political rights.
Nationality-based assertions of collective cultural or political rights, although similar in form, vary
widely in their specific content. They range, for example, from modest demands for administration or
education in the minority language to maxima list claims for far-reaching territorial and political autonomy
verging on full independence. Other aspects of the stance of national minorities are also highly variable.
While some favor full cooperative participation in the institutions of the host state, including participation
in coalition governments, others may favor a separatist, noncooperative stance. And while some may shun
overtures to external parties, believing it important to demonstrate their loyalty to the state in which they
live and hold citizenship, others may actively seek patronage or protection from abroad ­ whether from a
state dominated by their ethnic kin or from other states or international organizations. This variation in
specific claims to collective rights, and in overall "stance," occurs not only between but within national
minorities. The full range of stances just sketched could be found, for example, among the Sudeten
Germans of interwar Czechoslovakia.7 This variation in stances within a single national minority, this
spectrum of related yet distinct and even mutually antagonistic stances adopted by different segments of
"the same" ethnonational group, suggests the analytical usefulness of the notion of field. Using this notion,
developed and employed by Pierre Bourdieu in an impressive variety of studies,8 we can think of a
national minority not as a fixed entity or a unitary group but rather in terms of the field of differentiated and
competitive positions or stances adopted by different organizations, parties, movements, or individual political
entrepreneurs, each seeking to "represent" the minority to its own putative members, to the host state, or
to the outside world, each seeking to monopolize the legitimate representation of the group.9
Competition in the representation of the group may occur not only among those making different
claims for the group qua national minority, but also between those making such claims and those rejecting
the designation "national minority" and the family of claims associated with it. This is no mere academic
possibility. Think, for example, of "Russians in Ukraine" (and bracket for the moment the difficulties
inherent in the very expression "Russians in Ukraine" ­ the fact that this expression, with its clean syntax,
designates something that does not in fact exist, namely a definite, clearly bounded group of Russians in
Ukraine).10 There are different ways of conceiving what it means to be a Russian in Ukraine, only some of
which are consistent with conceiving Russians in Ukraine as a national minority. Thus Russians in Ukraine
can be understood as persons of Russian ethnic origin, most of whom speak Russian as their native
language, who nonetheless belong to the Ukrainian nation, understood as a political, territorial, or civic
nation, as the nation of and for all its citizens, regardless of language and ethnicity, not as the nation of
and for ethnic Ukrainians. Were this the prevailing self-understanding of Russians in Ukraine, there would
be no Russian "national minority." There would be persons of Russian ethnic origin and persons speaking
Russian as a native language, but they would not claim to be members of the Russian nation or
but a political nation open, in principle, to all, then the background condition against which the claim of national minority
status makes sense is missing. Collective self-representation as a national minority presupposes a certain type of collective
representation of the majority.
7
On the 1920s struggle among Sudeten Germans between "activists," favoring cooperation with Czechoslovak parties and
participation in coalition governments, and "negativists," rejecting these forms of cooperation, see Jaworski, Vorposten oder
Minderheit?; and Johann Wolfgang Brügel, Tschechen und Deutsche 1918 ­ 1938 (Munich: Nymphenburger, 1967). On
competition among Sudeten Germans after the Nazi seizure of power in Germany, see Ronald M. Smelser, The Sudeten
Problem.
8
See Pierre Bourdieu and Loc Wacquant, An Invitation to Reflexive Sociology (Chicago: University of Chicago Press,
1992), pp. 94ff., and the references cited there.
9
Although Bourdieu has not written on national minorities as such, his essay on regionalism as well as a more general
article on group-making contain suggestive formulations about the importance of representational struggles in the effort to
make and remake groups. See Pierre Bourdieu, "L `identité et la représentation: éléments pour une réflexion critique sur
I'idée de région," Actes de la recherche en sciences sociales 35 (1980), and "Social Space and the Genesis of Groups,"
Theory and Society 14 (1985).
10
During the 1989 census, some 11.4 million residents of Ukraine identified their "nationality" (natsional'nost) as
Russian. A larger number ­ nearly 17 million ­ identified their native language as Russian. See Gosudarstvennyi komitet po
statistike, Natsional'nyi sostav naseleniia SSSR (Moscow: Finansy i statistika, 1991), p. 78. There are no fixed identities here,
but rather a fluid field of competing identities and identifications. One should be skeptical of the illusion of bounded
groupness created by the census, with its exhaustive and mutually exclusive categories. For an argument suggesting that
divisions of language are more significant in post-independence Ukraine than divisions of ethnic nationality, see Dominique
Arel, "Language and Group Boundaries in the Two Ukraines," paper presented at a conference on "National Minorities,
Nationalizing States, and External National Homelands in the New Europe," Bellagio Study and Conference Center, Italy,
August 1994.
nationality.11 There is, of course, no chance of this view monopolizing the field of competing identities.
Indeed, it may recently have been losing ground, as support for independent Ukrainian statehood among
eastern Ukrainian Russians has waned with the rapid deterioration of the Ukrainian economy. But it does
belong to the field of competing stances.
Where does this leave us? If we rethink the concept of national minority along the lines sketched here,
the apparent clarity and simplicity of the concept dissolve. National minorities are not the internally
unified, externally sharply bounded groups that our ordinary language suggests. I will continue to speak of
"national minorities" for convenience, but it should be understood that this is a loose and imperfect
designation for a field of competing stances, and that the "stakes" of the competition concern not only
what stance to adopt as a national minority but whether the "group" (or potential group) in question should
understand and represent itself as a national minority.
Nationalizing state
A similar set of points can be made about the concept of "nationalizing state." I choose this term rather
than "nation-state" to emphasize that I am talking about a dynamic political stance ­ or family of related
yet competing stances ­ rather than a static condition.12 Characteristic of this stance, or set of stances, is
the tendency to see the state as an "unrealized" nation-state, as a state destined to be a nation-state, the
state of and for a particular nation, but not yet in fact a nation-state (at least not to a sufficient degree);
and the concomitant disposition to remedy this perceived defect, to make the state what it is properly and
legitimately destined to be, by promoting the language, culture, demographic position, economic
flourishing, or political hegemony of the nominally state-bearing nation.
Such a stance may be an avowed and expressly articulated "position." But it need not be avowed or
articulated for it to be "real" in the sense that matters for this chapter, namely exercising a real effect on the
minority and "homeland" political fields. This may be the case if policies, practices, symbols, events,
officials, organizations, even "the state" as a whole are perceived as nationalizing by representatives of the
national minority or external national "homeland," even if this characterization is repudiated by persons
claiming to speak for the state. To ask whether such policies, practices, and so on are "really" nationalizing
makes little sense. For present purposes, a nationalizing state (or nationalizing practice, policy, or event) is
not one whose representatives, authors, or agents understand and articulate it as such, but rather one that
is perceived as such in the field of the national minority or the external national homeland.
This raises a further complication. What does it mean for a state to be perceived as nationalizing in the
political field of the national minority or that of the external national homeland? It is not sufficient for
anyone who acts in those fields to perceive and characterize the state as nationalizing. The perception has
to be "validated" or socially "sustained." The perception and characterization of the host state and its
practices and policies are themselves crucial objects of struggle within the political fields of the national
minority and the external national homeland.
A national minority ­ to return for a moment to this concept ­ is a field of struggle in a double sense. It
is (as we saw earlier) a struggle to impose and sustain a certain kind of stance vis-d-vis the state; but at the
same time it is a struggle to impose and sustain a certain vision of the host state, namely as a nationalizing
or nationally oppressive state. The two struggles are inseparable: one can impose and sustain a stance as a
mobilized national minority, with its demands for recognition and for rights, only by imposing and
sustaining a vision of the host state as a nationalizing or nationally oppressive state. To the extent that this
vision of the host state cannot be sustained, the rationale for mobilizing as a national minority will be
undermined.
I do not want to give the impression that all that matters are the external perceptions of a host state's
policies and practices as nationalizing. Such external perceptions ­ and the political stance they help justify
11
See Roman Szporluk, "Reflections on Ukraine after 1994: The Dilemmas of Nationhood," The Harriman Review 7, nos.
7-9 (1994).
12
A nationalizing state is precisely not a nation-state in the widely used sense of an ethnoculturally homogeneous state,
the very large majority of whose citizens belong to the same ethnocultural nation. Quite the contrary. Although it does not
presuppose ethnocultural heterogeneity (for nationalizing projects can be, and have been, advanced even in ethnoculturally
homogeneous settings), nationalizing states are ordinarily ethnoculturally heterogeneous. A further reason for preferring the
term "nationalizing state" to "nation-state" is that the latter implies an achieved or completed condition, while the former
usefully implies that this completed condition has not been achieved. A nationalizing state is one conceived by its elites as a
specifically unfinished state (cf. the German conception, current in the Bismarckian period, of the unvoilendet or
"incomplete" nation-state). I discuss the concept of nationalizing state in more sustained fashion in Chapter 4, using interwar
Poland as an example.
and sustain ­ are indeed more important than the self-understanding of participants in the political field of
the nationalizing state, but they are not independent of the political idioms used by participants in that
field. When nationalization is an explicit project rather than merely a perceived practice, when host state
policies and practices are expressly avowed and articulated as nationalizing, the perception of the state as a
nationalizing state will be much more likely to prevail in the external fields ­ among the national minority
or in the external national homeland.
Nor is it unusual for participants in the host state to articulate projects of nationalization, to conceive
and justify policies and practices in a nationalizing idiom. Such an idiom is not only eminently respectable
but virtually obligatory in some contexts. This is often the case in new states, especially those that, for
historical and institutional as well as ethnodemographic reasons, are closely identified with one particular
ethnocultural nation.13 This is the case in almost all Soviet and Yugoslav successor states, thanks to the
legacy of Soviet and Yugoslav nationality policy, which (as I argued in Chapter 2 with respect to the Soviet
case) fixed and crystallized ethnocultural nations and endowed them with "their own" territorial "polities,"
that is, with polities (or pseudopolities) that were deliberately constructed as belonging to particular
ethnocultural nations.
Whether we are talking about perceived nationalizing stances or openly avowed nationalizing projects,
there is a great deal of variation among such stances and projects, not only between states, but within a
given state. The notion of field can be useful here too. It brings into analytical focus the wide range of
nationalizing stances within a single state, the spectrum of related yet distinct and even mutually
antagonistic stances adopted by differently positioned figures in and around the complex inter- and intraorganizational
network that we call, for convenience, "the state."
We can think of a nationalizing state not in terms of a fixed policy orientation or a univocal set of
policies or practices but rather in terms of a dynamically changing field of differentiated and competitive
positions or stances adopted by different organizations, parties, movements, or individual figures within
and around the state, competing to inflect state policy in a particular direction, and seeking, in various and
often mutually antagonistic ways, to make the state a "real" nation-state, the state of and for a particular
nation. 14
13
In the twentieth century, new states have been created in three great bursts ­ after World War I, when the territories of
the great European and Eurasian multinational empires were divided and reconfigured; during mid-century decolonization,
when new states were carved out from most of the overseas territories of the Western European colonial empires; and in the
post-Cold War present, when, in a continuation of the process of national reconfiguration of political space begun in the
nineteenth and early twentieth centuries, incipient nation-states have been formed from the territories of the multinational
Soviet and Yugoslav states and binational Czechoslovakia. All of these states have been conceived by their dominant
political and cultural elites as nation-states and, in a very broad sense, as nationalizing states. But there is an important
difference between the new states that succeeded to multinational territorial states at the beginning and end of the twentieth
century on the one hand, and most of those that were carved out of overseas empires ­ especially in sub-Saharan Africa ­ on
the other. Almost every one of the former was conceived and justified, in the nationalist movements preceding their
independent statehood as well as after statehood was achieved, as the state of and for a particular ethnonational group, which,
though in no case coincident with the entire state population, in almost all cases constituted the majority, and usually the
substantial (though seldom the overwhelming) majority of the state population. Why this was the case would require a
lengthy historical excursus; that it was and is the case is clear. By contrast, most states carved out of overseas colonial
empires were not conceived in the same way ­ before or after independence ­ as the states of and for particular ethnonational
groups. Of course, in practice, some states ­ or portions of the state apparatus, such as the army ­ did come to "belong" to
particular ethnonational groups (not always the same groups that had been favored by colonial administrators). See for
example Donald Horowitz, Ethnic Groups in Conflict (Berkeley: University of California Press, 1985). But given the general
discrepancy in scale between colonial units and ethnic groups, the rhetoric of anticolonial nationalism ­ the claims to
nationhood made during anticolonial struggles ­ was framed in a territorial (and expressly supra-ethnic) rather than an
ethnonational idiom. And leaders of newly independent states also framed their nationalizing projects in territorial and civic
rather than ethnonational terms, hoping to build up a "modern" territorial national identity. See Anthony Smith, State and
Nation in the Third World: The Western State and African Nationalism (New York: St. Martin's Press, 1983). In fact, of
course, politicized ethnicity has flourished at least as much in postcolonial sub-Saharan Africa as elsewhere. But in large part
because of the discrepancy in scale between political and ethnocultural units, state-backed nationalizing projects could not be
as easily linked to one particular ethnonational group as was the case in the new states formed from the continental
multinational empires. The point of this digressive footnote (a point I return to in Chapter 4) is to emphasize that
nationalizing idioms ­ more precisely, idioms of ethnic or ethnocultural nationalization ­ were widely employed in the new
states of interwar Europe, and they are widely employed in the new states of post-Cold War Europe.
14
This competitive field also includes stances that reject principles and programs of nationalization, for example, in favor
of some form of cultural pluralism. But for the historical and institutional reasons alluded to above, successor state elites are
strongly disposed to adopt nationalizing stances of one kind or another.
An example of competition among nationalizing stances might help make this a bit less abstract.
Consider the question of language. Elites in all Soviet successor states believe it necessary and desirable to
promote the language of the nominally state-bearing nation. This is a nationalizing stance that all share.
Yet there have been vigorous struggles, in all successor states, about how this should be done. Should
knowledge of the national language be required for citizenship or for certain types of employment? If so,
what level of knowledge? How should the legacy of linguistic Russification be combated, and knowledge
of the national language promoted, when a substantial fraction of the majority nationality does not speak
the national language (as is the case, for example, in Ukraine and Kazakhstan)? In what circumstances
should the use of other languages be permitted, or required, in public life, in the school system, or in the
associational sphere of civil society? What mix of incentives and authoritative measures should be
employed to promote the national language?15
External national homeland
Since the analytical points to be made are similar to those made about national minorities and
nationalizing states, the concept of external national homeland can be treated more briefly. It, too, denotes
a dynamic political stance ­ or, again, a family of related yet competing stances ­ not a static condition,
not a distinct "thing." Common to "homeland" stances are the axiom of shared nationhood across the
boundaries of state and citizenship and the idea that this shared nationhood makes the state responsible,
in some sense, not only for its own citizens but also for ethnic co-nationals who live in other states and
possess other citizenships. These shared assumptions and orientations define a "generic" homeland stance.
But there is great variation among particular homeland stances, great variation in understandings of just
what the asserted responsibility for ethnic co-nationals entails: Should ethnic co-nationals living as
minorities in other states be given moral support, or also material support? What sorts of ties and relations
with the homeland or mother country should be fostered? What sort of immigration and citizenship
privileges, if any, should co-ethnics abroad be offered? What sort of stance should they be encouraged to
take, vis--vis the states in which they live? And what sort of stance should the homeland adopt toward
those states? How forcefully should it press its concerns about their policies toward minorities? What
weight should those concerns be given in shaping the homeland state's overall relations toward the states
in which its co-ethnics live? And how forcefully should it press its concerns in the various international
forums that monitor and set standards for policies toward minorities? These are all contested questions in
homeland states.
The various homeland stances compete not only with one another but with stances that reject the basic
premise of homeland politics, or at least set sharp limits on the permissible forms of homeland politics.
According to these anti-homeland stances, which are more consonant with classical understandings of
interstate relations and international law, a state may, indeed must, protect its own citizens even when they
live in other states. But it cannot legitimately claim to protect its ethnic co-nationals who live in another state
and hold the citizenship of that state. The field of struggle to inflect state policy is therefore constituted by
struggles over whether and how a state should be a homeland for its ethnic co-nationals in other states.
The triadic nexus: a relation between relational fields
As I have argued, national minority, nationalizing state, and external national homeland should each be
conceived not as a given, analytically irreducible entity but rather as a field of differentiated and competing
positions, as an arena of struggle among competing stances. The triadic relation between these three
"elements" is, therefore, a relation between relational fields; and relations between the three fields are closely
intertwined with relations internal to, and constitutive of, the fields. The approach to the national question
adopted here is consistently and radically relational.
A central aspect of the triangular relational nexus is reciprocal inter-field monitoring: actors in each
field closely and continuously monitor relations and actions in each of the other two fields. This process
of continuous reciprocal monitoring should not be conceived of in passive terms, as a registering or
transcription of goings-on in other fields. Rather, the monitoring involves selective attention,
interpretation, and representation. Often, the interpretation of other fields is contested; it becomes the
object of representational struggles among actors in a given field. Such struggles among competing
representations of an external field may be closely linked to struggles among competing stances within the
given field. Thus, the struggle to mobilize a national minority may be linked to a struggle to represent the
host state as a nationalizing or nationally oppressive state. Conversely, proponents of nationalization may
15
On the politics of language in Soviet successor states, see the special issue of Nationalities Papers 23, no. 3 (1995), on
"Implementing Language Laws: Perestroika and its Legacy in Five Republics."
seek to represent the national minority as actually or potentially disloyal, or the homeland as actually or
potentially irredentist. The breakup of Yugoslavia illustrates both linkages: efforts to mobilize the Serb
minority in Croatia depended on efforts to represent Croatia as a dangerously nationalizing state, while
nationalizing elites in Croatia sought to represent the mobilizing Serb minority as disloyal and Serbia as an
irredentist homeland.
Perceptions and representations of an external field may be linked with stances within a field in two
ways. On the one hand, the stances within a field may be prior and governing. In a strong sense, this
occurs when a stance to which one is already committed "requires" a certain representation of the external
field, and therefore generates efforts to impose or sustain it through deliberately selective interpretation or
outright misrepresentation and distortion of developments in that external field. In a weaker but still
significant (and very widespread) sense, it occurs when a particular stance to which one is already at least
provisionally committed disposes one, in entirely "sincere" and noncynical fashion, through well-known
mechanisms of selective (mis-)perception and (mis-)representation, to accept a particular representation of
an external field, a representation congruent with one's own (already provisionally adopted) stance or
position.
On the other hand, perceptions and representations of developments in an external field may
strengthen or undermine existing stances or evoke or provoke new ones. In this case, instead of already
committed stances governing perceptions and representations of the external field, commitments to
stances emerge interactively, in response to perceived and represented developments in the external field.
Thus, stances may shape (and distort) perceptions and representations of an external field, or they may
take shape in response to perceptions and representations of developments in that external field. The two
processes, although analytically distinct, are often intertwined in practice. The Yugoslav case, for example,
abundantly illustrates how strong initial nationalist dispositions or stances among some Serbs and Croats
shaped and distorted perceptions and representations of the other, through both sincere selective
perception and cynical mis-representation; it also illustrates how others, initially indifferent to nationalism,
came to adopt nationalist stances in reaction to perceptions and representations of seemingly threatening
developments in other fields. This dual linkage exemplifies three general features of the relational nexus
with which we are concerned: (1) the close interdependence of relations within and between fields; (2) the
responsive and interactive character of the triadic relational interplay between the fields; and (3) the mediated
character of this responsive interplay, the fact that responsive, interactive stance-taking is mediated by
representations of stances in an external field, representations that may be shaped by stances already
provisionally held.
The triadic nexus and the breakup of Yugoslavia
Having sketched the triadic relational nexus in abstract terms, I would like to conclude with a more
concrete illustrative discussion. Volumes have been written about the collapse of Yugoslavia, and many
more are sure to follow. My aim here is not to provide even a summary account of the collapse, but rather
to highlight its crucially triadic form and to indicate ­ if only programmatically ­ how the relational
approach out-lined above might illuminate its bloody dynamics. I limit my attention here to the first phase
of the breakup, involving Croatian and Slovenian moves toward independence and culminating in the war
in Croatia; I do not discuss the war in Bosnia.
The first phase of the Yugoslav collapse was presented in the American press as a dyadic struggle. On
one side stood Serbia, determined to reassert centralized control (and therefore Serb hegemony) over
Yugoslavia as a whole, or, failing that, to carve out a "greater Serbia" from the ruins of the state. On the
other side stood Slovenia and Croatia, seeking autonomy and ultimately independence in the face of the
Serbian push for hegemony.16 Yet while the Slovenian issue was indeed dyadic, the Croatian conflict was,
from the beginning, fundamentally triadic, involving a tension-fraught dynamic interplay between an
incipient national minority (Serbs in Croatia), an incipient nationalizing state (Croatia), and an incipient
external national homeland (Serbia). Seeing the core dynamic in this way is not simply a matter of
"adding" the Croatian Serbs to the equation. Rather, it directs our attention to differing underlying
processes. The dyadic view of the Serb-Croat conflict construes it as involving a push for Serb hegemony,
a responsive Croatian secessionist movement, and a subsequent war of aggression against independent
Croatia. The triadic view, by contrast, focuses on the complex interplay of three overlapping and mutually
intensifying processes: the nationalization of the Croatian incipient state (both before and after
16
For a sophisticated statement of this View, see Branka Magaš, The Destruction of Yugoslavia (London: Verso, 1993).
independence was formally declared); the increasing disaffection, and nationalist mobilization, of Serbs in
the ethnic borderlands of Croatia; and the development of a radical and belligerent "homeland" stance in
the incipient Serbian state, leading eventually to the intervention of the increasingly Serb-dominated
Yugoslav army in Croatia on the side of plans to salvage a "Greater Serbia" from the rubble of the
federation.
The dyadic view rightly sees the Croatian drive for autonomy and independence as responding, in
significant part, to Serbian nationalist assertiveness. Milosevic's use of nationalist rhetoric to usurp
leadership of the Serbian Communist Party in September 1987 and to mobilize mass support thereafter especially
his emphasis on Serb victimization in overwhelmingly Albanian Kosovo and on the need to
reassert Serbian control over it by curtailing its constitutionally guaranteed autonomy ­ represented a
fundamental and destabilizing challenge to the precarious national equilibrium constructed by Tito. The
key to that equilibrium lay in the institutional restraints on the power of Serbia, preventing the Serbs from
reacquiring the political dominance they had exercised, to disastrous effect, in the interwar Yugoslav state.
The Serbian push to reassert control of Kosovo (and of the likewise formally autonomous Serbian
province of Vojvodina) directly challenged those constraints and the fragile equilibrium built on them.
While the resurgent Croatian nationalism of the late 1980s certainly had deep historical roots, and in many
respects could be seen as reenacting (though going beyond) the Croatian nationalist movement of 1967-71,
it was in crucial part a response to this destabilizing Serbian bid for hegemony within Yugoslavia.17
While the dyadic view illuminates the causes and antecedents of the Croat drive for autonomy and
independence, it obscures the nature and consequences of that drive. Construing it as a secessionist
movement, the dyadic view obscures the extent to which it was also, and inseparably, a nationalizing
movement ­ a movement to assert Croat "ownership" and control over the territory and institutions of
Croatia, to make Croatia the state of and for the Croatian nation.
This was evident in the campaign rhetoric with which Franjo Tudjman, with strong financial backing
from nationalist Croat émigrés, swept to victory in the spring 1990 elections, especially in his stress on the
deep cultural differences between Serbs and Croats and the need to replace Serbs, heretofore
overrepresented in key cultural, economic, and administrative positions in the republic, with Croats. It was
evident in the iconography of the new regime, notably in the ubiquitous display of the red-and-white
checkered armorial shield that had been an emblem of the medieval Croatian state but also of the
murderous wartime Ustasha state (which the new leadership failed categorically and publicly to denounce).
It was evident in the official, and ludicrous, "Croatization" of language. It was evident in the rhetoric of
the new Croatian constitution, which claimed "full state sovereignty" as the "historical right of the
Croatian nation" and symbolically demoted Serbs from their previous status as co-"owners" of the
Republic. And it was evident, perhaps most significantly, in a substantial purge, concentrated in the state
administration but extending beyond it as well, in which many Serbs lost their jobs.18
The significance of these and similar events, discourses, and practices lay not in themselves but in the
representations and reactions they evoked among Croatian Serbs ­ especially village and small-town Serbs
of the Krajina region ­ and in Serbia. The dynamic of nationalization, though partial and incipient, was
real ­ and troubling ­ enough. But through varying mixes of selective appropriation, exaggeration,
distortion, and outright fabrication, Serb nationalist politicians in Croatia and in Serbia proper represented
these nationalizing moves in a sinister light as heralding the establishment of an ultranationalist regime
that threatened the liberties, livelihoods, and ­ if Croatia were to opt for full independence ­ even the lives
of Croatian Serbs.
The cynical and opportunistic manipulation involved in the more extreme of these representations and
misrepresentations, irresponsibly evoking the specter of the Ustasha regime to discredit every
manifestation of Croatian nationalism, is often stressed. But the emphasis on elite manipulation cannot
explain why representations of a prospectively independent Croatia as a dangerously nationalizing state
17
For a lucid and sustained analysis of the resurgent Serbian nationalism of the 1980s, see Veljko Vujacic, "Communism
and Nationalism in Russia and Serbia," Ph.D. dissertation, University of California at Berkeley, 1995.
18
See Misha Glenny, The Fall of Yugoslavia: The Third Balkan War (London: Penguin, 1992), esp. pp. 12-13, 77, 81-2;
Leonard J. Cohen, Broken Bonds: The Disintegration of Yugoslavia (Boulder, Colo.: Westview Press, 1993), pp. 96-8, 208;
Bette Denich, "Dismembering Yugoslavia: Nationalist Ideologies and the Symbolic Revival of Genocide;," American
Ethnologist 21, no. 2 (1994), 377-81; Robert Hayden, "Constitutional Nationalism in the Formerly Yugoslav Republics,"
Slavic Review 51, no (1992); Eugene A. Hammel, "The Yugoslav Labyrinth," in Eugene Hammel, Irwin Wall, and Benjamin
Ward, Crisis in the Balkans (Berkeley: Institute of International Studies, University of California, 1993), pp. 16-17; and
Bogdan Denitch, Ethnic Nationalism: The Tragic Death of Yugoslavia (Minneapolis: University of Minnesota Press, 1994), p.
45.
were sufficiently resonant, and sufficiently plausible, among certain segments of the Krajina Serb
population, to inspire genuine fear and induce militant mobilization, and eventually armed rebellion,
against the Croatian regime.19
While the dyadic view treats Croatian and particularly Krajina Serbs as passive dupes, vehicles, or
objects of manipulative designs originating in Serbia, the triadic view sees them as active participants in
the intensifying conflict and as political subjects in their own right, construing (and misconstruing) the
dangers of the present in the light of the atrocities of the past. The complex process through which
representations of Croatia as a dangerously nationalizing, even protofascist, state emerged, took root, and
became hegemonic among Serbs in certain parts of Croatia's ethnic borderlands cannot be reduced to a
story of outside manipulation. Efforts by nationalist radicals in Serbia to mobilize grievances and fears
among Croatian Serbs were indeed an important part of the process. But the bulk of the work of
mobilizing grievances and fears was undertaken locally by Croatian Serbs. And the grievances and fears
were there to be mobilized. Although representations of wartime atrocities ­ often greatly exaggerated were
indeed widely propagated from Belgrade, memories of and stories about the murderous wartime
Independent State of Croatia, and especially about the gruesome fate of many Croatian `and Bosnian
Serbs (Bosnia having been incorporated into the wartime Croatian state), were not imports. They were
locally rooted, sustained within family and village circles, and transmitted to the postwar generations,
especially in the ethnically mixed and partly Serb-majority borderland regions where (outside of Bosnia)
most atrocities against Serb civilians had occurred, and where (again excluding Bosnia) the main Partisan
as well as the few Chetnik strongholds in Croatia had been located. It was among village and small-town
Serbs in just these regions ­ and not, for example, among the cosmopolitan Serbs of Zagreb ­ that
encounters with the incipient Croatian nationalizing state, interpreted through the prism of revived
representations of wartime trauma, generated intransigent opposition to Croatian independence.20
These mutually alienating encounters between the nationalizing and increasingly independent Croatian
state and the fearful and increasingly radicalized Serb borderland minority thus had their own destabilizing
logic; they were not orchestrated from Belgrade. But Serbian "homeland politics" was crucial to the
overall relational nexus. Homeland stances ­ involving identification with, assertions of responsibility for,
and demands to support or even "redeem" and incorporate ethnic Serbs out-side Serbian state territory have
a long tradition in Serbian politics. 21 The relation between the expansionist "small Serbia"
(established as an independent kingdom, though still under nominal Ottoman suzerainty, in 1829 and
recognized as fully independent in 1878) and the large Serb communities in the Ottoman and AustroHungarian
empires was a burning issue in the decades before World War I, and one that touched off the
war when a Bosnian Serb nationalist revolutionary assassinated Archduke Francis Ferdinand, heir to the
Habsburg throne, in Sarajevo. With the formation of a Serb-dominated South Slav state after the war,
incorporating the great majority of former Habsburg and Ottoman Serbs, the problematic of homeland
politics receded. Nor did it reemerge openly after World War II in Tito's reconstructed (and more
nationally equilibrated) Yugoslavia. Just as Russians viewed the Soviet Union as a whole (and not just the
Russian republic) as "their" state, so Serbs viewed the Yugoslav state as a whole (and not just Serbia) as
their own, regarding internal boundaries as insignificant or "merely administrative." Yet homeland politics
revived in Serbia, and emerged in Russia, when the "nationalization" of constituent units of Yugoslavia
and the Soviet Union eroded Serbs' and Russians' sense of being "at home" throughout the state.
The revival of Serbian homeland politics ­ of politicized concern with Serbs outside Serbia ­ centered
initially on Kosovo. Although it was formally part of the Serbian Republic, its constitutional promotion to
near-republic status in 1974, together with its gradual but thoroughgoing "Albanianization" (through
differential fertility, Serb out-migration, and preferential treatment in cultural and administrative positions),
were perceived by Serb intellectuals as a "quiet secession" that had, in practice, stripped Serbia of its
historic heartland.22 The dwindling Serb community in Kosovo was represented as a physically and
psychologically harassed national minority, forced increasingly to emigrate, subject to "genocide," in the
19
Glenny, The Fall of Yugoslavia, p. 11; and Denich, "Dismembering Yugoslavia," 381.
20
Denich, "Dismembering Yugoslavia"; Glenny, The Fall of Yugoslavia; and Denitch, Ethnic Nationalism, p. 33.
21
Ivo Lederer, "Nationalism and the Yugoslavs," in Peter F. Sugar and Ivo John Lederer, eds., Nationalism in Eastern
Europe (Seattle: University of Washington Press, 1994 [1969]).
22
Dennison Rusinow, "Nationalities Policy and the `National Question,'" in Pedro Ramet, ed., Yugoslavia in the 1980s
(Boulder, Colo.: Westview Press, 1985), pp. 146-7. On the background to Serb concern about Kosovo, see Vujacic,
"Communism and Nationalism," pp. 204-30.
scandalously hyperbolic language of the first major statement of the Serb nationalist revival.23 Having
again been "lost," Kosovo was in need of redemption, of reincorporation into a restored, strengthened,
unitary Serbia ­ a program taken up, with great mobilizational success, by Milosevic.
As Slovenia, Croatia, and later Bosnia-Hercegovina moved toward independence, Serbian homeland
politics ­ as articulated by Milosevic, by his even more radically nationalist opponents, and by the statecontrolled
broadcast media ­ was extended to, and came to focus increasingly on, Serb minorities in
Croatia and Bosnia. Through the prevalence in the media and public discourse of what one anthropologist
has called "narratives of victimization and of threat, linking the present with the Past and projecting onto
the future,"24 the plight of Kosovo Serbs was represented in generalized terms as a threat to Serbs in
minority Positions everywhere. After the election of Tudjman, this threat was seen as particularly acute in
Croatia, which was increasingly represented as a protofascist successor to the wartime Ustasha state.
Croatian claims to self-determination and sovereign statehood were met with counterclaims that Serbs,
too, had the right to self-determination, the right to a state of their own ­ if not Yugoslavia, then an
enlarged Serbia. The secession of Croatia, Milosevic bluntly warned throughout 1990 and the first half of
1991, would require the redrawing of its boundaries. Croatia's borderland Serbs were encouraged to take a
stand of intransigent opposition to the new Croatian regime and to its bid for independence, and, as the
crisis intensified, were provided with arms and logistical support.
The increasingly ominous tenor of Serbian homeland politics was doubly destabilizing, provoking both
the Croatian government and Croatian Serbs to adopt more intransigent stances. Just as the reassertion of
central Serbian control over Kosovo, by upsetting the precarious national equilibrium in Yugoslavia,
helped spark Croatian secessionism, so Serbian claims to speak for Croatian Serbs, by challenging Croatian
sovereignty and reinforcing representations and fears of aggressive Serb hegemony, helped push the
Croatian government toward a more uncompromising stance ­ toward the pursuit of full independence
(rather than a restructured federal or confederal arrangement) and toward the more vigorous assertion of
its authority in the rebellious borderlands (which occasioned armed clashes that led to the intervention of
the army, initially as a peacekeeping force, but increasingly as an ally of local Serb forces). At the same
time, the pan-Serb rhetoric, anti-Croat propaganda, and talk of border revisions emanating from Belgrade,
together with the more uncompromising Croatian government stance, pushed Croatia's borderland Serbs
toward greater intransigence ­ toward such steps as the formation of a "Serbian National Council" (July
1990), the holding of a referendum on autonomy for Croatian Serbs despite its prohibition by Croatian
authorities (August 1990), the establishment of the "Serbian Autonomous Region of Krajina" (December
1990), and the proclamation of that region's "separation" from Croatia (February 1991).
It is not possible here to discuss in detail the interactive dynamic that led to the outbreak of a war
pitting the heavily Serbianized "Yugoslav People's Army" and various Croatian Serb militias against the
over-matched Croatian army, resulting in the occupation for several years of :nearly a third of Croatian
territory (including parts in which Serbs had been only a small minority) and sealing the final dissolution
of Yugoslavia before spreading, with still more devastating consequences, to Bosnia-Hercegovina. I have
had to limit my discussion to a general sketch of the interplay between the incipient Serb national minority
in Croatia, the incipient Croatian nationalizing state, and the incipient Serbian homeland, locked in an
intensifying spiral of mistrust, mis-representation, and mutual fear. I have had to ignore not only the
detailed interactive sequence of that interplay, but also the struggles among competing stances internal to
the minority, nationalizing state, and homeland. Enough has perhaps been said, however, to suggest the
potential fruitfulness of a relational, dynamic, interactive approach to nationalist conflict.
Conclusion
The fears and fault lines, the resentments and aspirations, the myths and memories that defined the
national question in Yugoslavia have long been well known. But they did not foreordain the bloody
breakup of the state. That was a contingent outcome of the interplay of mutually suspicious, mutually
monitoring, mutually misrepresenting political elites in the incipient Croatian nationalizing state, the
incipient Serb national minority in that state, and the incipient Serbian "homeland" state.
23
This was the "Memorandum" of the Serbian Academy of Arts and Sciences, prepared in 1986. 1\ French version has
been published in Mirko Grmek et al., eds., Le nettoyage ethnique: documents historiques sur une idéologie serbe (Paris:
Fayard, 1993). On the memorandum, see Vujacic, Communism and Nationalism, pp. 257-67.
24
Bette Denich, "Unmaking Multi-Ethnicity in Yugoslavia: Metamorphosis Observed," Anthropology of East Europe
Review II, nos. 1-2 (1993), 51.
The relational and interactive perspective outlined in this essay, and illustrated with respect to the
breakup of Yugoslavia, makes it possible to give due weight to both structure and contingency in the
analysis of the national question in Eastern Europe and post-Soviet Eurasia. The relational field in which
the national question arises is a highly structured one. In the post-Soviet case, for example, it was
predictable ­ for the historical and institutional reasons analyzed in Chapter 2 as well as for conjunctural
reasons linked to economic and political crisis ­ that nationalizing stances of some kind would prevail
among successor state elites; that successor state Russians would tend to represent themselves as a
national minority; and that Russian Federation elites would engage in "homeland" politics, asserting
Russia's right, and obligation, to protect the interests of diaspora Russians. In the Yugoslav case, again for
historical and institutional as well as conjunctural reasons, the emergence of nationalizing, minority, and
homeland stances was similarly predictable. But what could not be predicted in these or other cases ­ and
what cannot be retrospectively explained as structurally determined ­ was just what kind of nationalizing
stance, what kind of minority self-understanding, what kind of homeland politics would prevail in the
struggles among competing stances within these three relational fields, and just how the interplay between
the three fields would develop. Here, social science and history must acknowledge, and theorize, the
crucial causal significance of the contingency inherent in social and political action, without neglecting the
powerful structuration of the relational fields in which action and struggle occur.
In: Brubaker, Rogers. Nationalism Reframed. Cambridge: Cambridge University Press, 1996. 55-76