This article was downloaded by: [University of Haifa Library]

On: 29 September 2013, At: 14:39

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Southeast European and Black Sea

Studies
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/fbss20

Political cleavages in Serbia: changes

and continuities in structuring
left–right orientations
a
Willy Jou
a
Department of Political Science, University of California, Irvine,
California, USA
Published online: 04 Aug 2010.

To cite this article: Willy Jou (2010) Political cleavages in Serbia: changes and continuities in
structuring left–right orientations, Southeast European and Black Sea Studies, 10:2, 187-206, DOI:
10.1080/14683857.2010.486947

To link to this article: http://dx.doi.org/10.1080/14683857.2010.486947

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or
arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions
 Southeast European and Black Sea Studies
Vol. 10, No. 2, June 2010, 187–206

Political cleavages in Serbia: changes and continuities in structuring

left–right orientations
Willy Jou*

Department of Political Science, University of California, Irvine, California, USA

(Received 30 April 2009; final version received 10 November 2009)
Taylor and Francis
FBSS_A_486947.sgm
Downloaded by [University of Haifa Library] at 14:39 29 September 2013

Southeast
10.1080/14683857.2010.486947
1468-3857
Original
Taylor
202010
10
jouw@uci.edu
WillyJou
000002010
&Article
Francis
European
(print)/1743-9639
and Black(online)
Sea Studies

This study examines the structuring of political attitudes in Serbia through the
framework of the left–right schema. Using data from public opinion surveys taken
during the 1990s and early 2000s, the impact of socio-demographic variables,
economic interests, democratic support, social values, and nationalistic sentiments
on respondents’ left–right orientations is analysed, as well as the relationship
between left–right placements and support for individual political parties. Results
show that in addition to age and religiosity, evaluation of communist rule and
satisfaction with the state of democratic development form the main axis of
contestation, while there is little evidence for economic, authoritarian–libertarian,
or nationalist cleavages.
Keywords: political cleavage; left–right schema; regime divide; Serbia

Introduction
Publics everywhere face the need to decide which party to vote for and which issue
preference to hold when making political choices. This is challenging particularly in
new democracies, where most parties’ lack of organization and programmatic cohe-
sion on the supply side is often matched by voter confusion over a wide array of
largely unfamiliar choices on the demand side. One device that can serve functions of
both orientation for individual voters and communication for the party system is the
left–right schema (Fuchs and Klingemann 1989; Inglehart and Klingemann 1976).
Although left–right semantics originated in Western Europe, these concepts have been
utilized in studying new democracies in recent years (e.g. McAllister and White 2007;
Shin and Jhee 2005). This encapsulation of political cleavages is especially pertinent
for the period immediately following democratization, as the contexts of transition
from authoritarianism and subsequent reforms provide an additional dimension of
contestation (Klíma 1998; Moreno 1999; Myagkov and Ordeshook 1998).
Scholars have examined political cleavages in post-communist countries both
theoretically (Evans and Whitefield 1993, 2000; Kitschelt 1992) and empirically
(Evans and Whitefield 1998; Kitschelt 1995; Kitschelt et al. 1999; Markowski 1997;
Tóka 1998). However, most empirically based comparative research employ data
from a single point in time and tend to focus on the Visegrad countries. These studies
elucidate what factors underlie cleavage dimensions in post-communist settings, and
how these cleavages relate to partisanship and vote choice. At the same time, both

*Email: jouw@uci.edu

ISSN 1468-3857 print/ISSN 1743-9639 online

© 2010 Taylor & Francis
DOI: 10.1080/14683857.2010.486947
http://www.informaworld.com
 188 W. Jou

voters’ experiences of political and economic changes and parties’ strategies and ideo-
logical appeals may undergo considerable shifts under volatile conditions that
followed the end of communist rule (Hanley 2004; Whitefield and Evans 1999), thus
calling for cross-temporal studies. The present study seeks to supplement works on
post-communist cleavages by tracing changes and continuities in the anchoring of left
and right identification through an examination of public opinion in the 1990s and
early 2000s in Serbia, a country that has received relatively scant attention in the
comparative literature (for a recent exception, see Todosijević 2008), with an empha-
c[acue]t

sis on analysing trends before and after the regime change in 2000.
In exploring whether and how political orientations are structured in Serbia, weak
party identification and group-based interest articulation common to most post-
communist countries are accompanied by the impact of state disintegration and war
Downloaded by [University of Haifa Library] at 14:39 29 September 2013

on the mobilization of political cleavages. Questions of nationhood became entangled

with, and likely impeded, transition toward democracy to a greater degree and for a
longer period than in most of its neighbours, a condition engendering the possibility
that social structural factors and issue preferences would align differently in the space
of political competition compared with other ex-communist countries in Central and
Southeast Europe. Similar to other democratizing transitional societies, expectations
of weak linkages between ideology and party choice (Mair 1997) are compounded by
low institutionalization at both the regime and party system levels (Fink-Hafner 2008;
Zakošek 2008) and a propensity toward personalization of politics (Orlović 2008).
ocasnr[] c[acue]t

These conditions may hinder the development of structured cleavages, as encapsu-

lated in overarching left–right orientations that provide stable heuristics to voters’
party and policy preference. Such a scenario in turn carries implications not only for
Serbia but also for other new democracies facing simultaneous challenges of regime
change and state building.
This study is organized as follows: After reviewing the literature on left–right
identification in general and its application in post-communist settings in particular,
an account of political developments in Serbia in the 1990s with respect to salient
political conflicts is given. Section on data and hypotheses for the study is followed
by results and analysis of empirical testing on factors that shape left–right identifica-
tion as well as extremist orientations. Next, attention turns to the relationship between
left–right placements and vote choice. The final section summarizes the findings and
discusses the possible implications of cleavage patterns for democratic consolidation.

The left–right schema and the structuring of cleavages in post-communist

contexts
The seminal work by Downs (1957) introduces a spatial representation of politics,
placing parties and voters along a left–right spectrum and hypothesizing that voters
make rational choices by selecting parties closest to their own positions. While polit-
ical conflict does not necessarily revolve around only a single set of issues, Inglehart
and Klingemann argue that ‘the left-right dimension is more often than not used by
political elites and mass communication to label the most important issues of a given
era’ (1976, 244). These ideological labels constitute meaningful heuristics for both
voters seeking information shortcuts and parties framing their appeals to the public
(Conover and Feldman 1984; Sniderman, Brody, and Tetlock 1991).
The terms left and right have long been popularly associated with the role of
government in the economy (Downs 1957; Knutsen 1995; Lipset 1960), but they are
 Southeast European and Black Sea Studies 189

not solely defined by these issues. Rather, Inglehart refers to the left–right schema as
a ‘super issue’ incorporating ‘whatever major conflicts are present in the political
system’ (1990, 273), and Huber and Inglehart (1995, 90) characterize it as an ‘amor-
phous vessel’ carrying meanings that vary with political and economic conditions in
different societies. This flexibility and absorptive capacity permit the application of
the schema in different political contexts. While it is hardly surprising that party iden-
tification exerts a considerable impact on voters’ self-placement on the ideological
spectrum (Inglehart and Klingemann 1976), one should distinguish between the two
concepts. Fuchs and Klingemann (1989) state that voters’ understanding of the left–
right schema is increasingly shaped by value-related rather than partisan orientations,
and Knutsen (1997) corroborate this by demonstrating a sizeable value- (instead of
party-) driven component in left–right identification by voters in more advanced soci-
Downloaded by [University of Haifa Library] at 14:39 29 September 2013

eties, as well as countries with a greater number of parties.

The distinction between left–right orientation and party identification is especially
pertinent to post-communist new democracies, where the short history of electoral
competition, more volatile party systems, and a top-down model of party organization
all lead one to expect considerably lower levels of party identification compared with
established democracies (cf. Kaase and Klingemann 1994). Evans and Whitefield
(1993, 528–31) argue that the pursuit of egalitarian economic policies and suppression
of organized civil society groups hamper the development of political interests and
allegiances, thereby providing no social or institutional basis for left–right orienta-
tions. Also, Tóka (1998, 598–600) finds a decrease in structural voting (i.e. based on
class and religious affiliation), and that even structural voters are volatile in their
choices due to lack of group identity. One thus expects left–right orientations in new
democracies to be less firmly anchored in socio-structural factors.
Party systems emerging in the wake of the industrial revolution continued to reflect
the owner versus worker cleavage through subsequent decades (Lipset and Rokkan
1967), yet, one may question whether cleavages rooted in historical legacies of
Western Europe could be transplanted to post-communist countries that have followed
vastly different political trajectories, including sudden rather than gradual democrati-
zation and mass enfranchisement (Geddes 1995). Insofar as the end of communist rule
generated opportunities for private economic activities, in most cases accompanied by
moves toward privatization, one may hypothesize contention along lines similar to
Western Europe. While communist policies inhibiting classes and group interest repre-
sentation (Evans and Whitefield 1993) are likely to hamper the formation of left–right
cleavages based mainly on economic interests, some scholars find evidence on the
significance of both economic attitudes (Bielasiak 1997; Dalton 2006; Evans and
Whitefield 1998; Klíma 1998) and the increasing role of class (Szelényi, Fodor, and
Hanley 1997) in defining left and right in post-communist countries.
The dramatic (and in the case of most former Yugoslav republics, traumatic) expe-
rience of regime change is likely to constitute a major cleavage. Moreno (1999)
hypothesizes that the conflict between authoritarian and democratic forms of govern-
ment (hereafter the regime divide) often overshadows economic questions in defining
left and right in new democracies, because the most contentious issues focus on redis-
tribution of power rather than wealth. Unlike established democracies, publics in new
democracies can evaluate their current regime in comparison with the alternatives
based on personal experience (McDonough, Barnes, and Pina 1986; Rose, Mishler,
and Haerpfer 1998). Dissatisfaction with the present system may easily lead to nostal-
gia for communist rule (Ekman and Linde 2005). Therefore, the regime divide may
 190 W. Jou

not only structure left–right orientations in the immediate aftermath of transition

(Evans and Whitefield 1998; Kitschelt, Dimitrov, and Kanev 1995), but persist as a
major cleavage in countries where democratization remains tentative or reversible
(e.g. Haughton and Rybář 2004).
coarn[]

Social values constitute another important basis for structuring left–right orienta-
tion. Particularly, in countries where regime type has ceased to be in dispute, political
elites may find it easier to mobilize traditional cleavages based on religion, ethnicity,
or attitudes toward foreigners, rather than yet undifferentiated class delineations. In
Hungary and Poland, for example, left and right are primarily determined by culture
and religion (as well as opinion on former communist rule) rather than class or
economic interests (Evans and Whitefield 1995; McManus-Czubińska et al. 2003;
n[aceu]t

Mészáros, Solymosi, and Speiser 2007; Szczerbiak 2008). Forecasting cleavages in

Downloaded by [University of Haifa Library] at 14:39 29 September 2013

post-communist settings, Kitschelt (1992) links authoritarian orientations with anti-

market views, and libertarianism with pro-market and cosmopolitan attitudes. Losers
in the process of economic transition would gravitate toward the former position and
winners toward the latter. In short, social values are hypothesized to be systematically
linked with economic interests.
For citizens in new democracies who perceive communism as an externally
imposed regime, nationalist sentiments may overlap with anti-communism on the
same cleavage dimension. However, nationalism is more difficult to incorporate into
the ideological schema where communist rule enjoyed greater legitimacy due either to
indigenous roots (the former Yugoslavia being a prominent example) or material
benefits brought to the public. Ethnicity presents an additional complicating factor,
because regime change in ethnically heterogeneous countries almost inevitably brings
issues of accommodation (or exclusion) of national minorities to the fore. Whitefield
(2002) argues that post-communist cleavages largely resemble those found in Western
Europe except on the ethnic dimension. Stefanovic (2008) specifically points out that
the far right in Eastern Europe differs from its western counterpart by focusing on
internal minorities as well as territorial disputes. In addition, several studies report
notable variations in regime support by different ethnic groups (Dowley and Silver
2002; Lühiste 2008). While national/ethnic identity is not intrinsically linked with left
or right (Inglehart and Klingemann 1976), the extent to which this dimension is
absorbable into ideological orientations in countries where it comprises a major cleav-
age warrants investigation.
What accounts for these different cleavage structures in post-communist societ-
ies? Rivera (1996) singles out legacies of communism as the key determinant of post-
transition patterns of party competition. Kitschelt (1995) also offers a historical
explanation, distinguishing among three types of communist rule – bureaucratic-
authoritarian, national consensus, and patrimonial – and predicting cleavages based
on economic interests, social values (e.g. religion, morality), and democratization or
issues of law and order, respectively. Evans and Whitefield (1993) propose three
decisive influences on post-communist cleavages: market experience, ethnic homoge-
neity, and security of statehood. For example, in established states that are ethnically
homogenous and enjoy some degree of market success, economic cleavages similar
to those observed in Western Europe are likely to emerge, whereas political conflict
may revolve around issues of nationalism and ethnicity in newly established states
containing heterogeneous populations. Finally, institutional rules can indirectly affect
cleavages by influencing political choices available to publics. For instance, presiden-
tial systems may be more vulnerable to clientelism, and proportional electoral laws
 Southeast European and Black Sea Studies 191

allow small parties with narrow appeals to gain legislative seats whereas plurality
rules encourage overlapping cleavages represented by larger parties.

Political conflicts in Serbia

Political developments in Serbia during the 1990s can be characterized as a process of
transition ‘not into a fledgling democracy, but rather into a one-party authoritarian and
populist dictatorship’ (Cohen 2001, 118). The ruling Serbian Socialist Party (Socialist
Party of Serbia, SPS) succeeded the League of Communists of Serbia by inheriting not
only its properties and organizational networks (Sell 2002, 125), but also its anti-market
ideology. In addition to controlling the police and media, two interrelated factors
allowed SPS to remain dominant until 2000: wars in Croatia, Bosnia-Herzegovina, and
Downloaded by [University of Haifa Library] at 14:39 29 September 2013

later Kosovo led to an ‘ethnification of the political sphere’ (Fink-Hafner 2008, 174;
see also Offe 1996),1 and the opposition was often fragmented and unable to offer a
credible alternative program in the face of a government-led nationalist consensus
(Miller 1997). The wars allowed the government of Slobodan Milošević to not only ocsarn[] c[acue]t

politicize ethnic sentiments by depicting successive opponents, such as Croats and

Kosovar Albanians, as oppressors in the past while casting Serbs as victims (Ramet
2007, 47), but also to ‘categorically disqualify political opponents as treasonous, unpa-
triotic, and fomenting division when unity is needed’ (Gordy 1999, 24; see also Sell
2002, chapter 7).2
Nationalist mobilization severely limited the opposition’s room for maneuver.
Under circumstances prevailing throughout most of the 1990s, adopting a stance
diametrically opposite the regime’s nationalist line was not a politically viable option.
Instead of presenting policy objectives distinct from those of the government, opposi-
tion parties could offer little more than different means of pursuing the same nation-
building goals (Slavujevic 1998, 108).3 Just as importantly, antagonism among oppo-
sition parties matched, if not exceeded, that between them and the government (Goati
1998, 28). For example, programmatic heterogeneity among its component parties lay
behind the opposition Democratic Movement of Serbia (DEPOS) coalition’s failure to
elaborate its platforms and led to its decision not to contest the 1992 election, and the
break-up of the Zajedno (‘Together’) coalition was attributed to ‘long-standing
jealousies and political differences’ between leading opposition figures, Vuk
Drašković and Zoran Din d ić (Cohen 2001, 214). Meanwhile, the opponents of the
sc[aorn] c[acue]t D
[orkts] odr[kst] c[acue]t

regime who advocated liberal democratic, non-nationalist programs found themselves

marginalized, or even ostracized as traitors to the nation.
Furthermore, drastic shifts in various opposition parties’ positions, such as that of
Drašković’s Serbian Renewal Movement (SPO) from nationalism to an anti-war
sc[aorn] c[acue]t

stance, and Din d ić’s Democratic Party (DS) from economic liberalism to a hard-line
D
[orkts] odr[kst] c[acue]t

nationalist view, led to confusion and loss of credibility among voters opposed to the
regime (Gordy 1999, 56). Whatever political merits such a tactic might have, this
could not have facilitated party competition over issue-based cleavages. In short, even
as the public gradually became disillusioned with the SPS, prospects of regime change
remained implausible in the absence of a unified alternative government, as many
opposition politicians were as discredited as the government due to their ‘constant
squabbling and highly visible failure to maintain a politically united organizational
structure’ (Cohen 2001, 220).
As alluded to above, a major characteristic of the Serbian party system lies in the
prominent role of party leaders. As both a cause and a consequence of weak party
 192 W. Jou

institutionalization, leaders retained tight control over their parties to such an extent
that in the mind of the public ‘parties are basically the leaders’ (Orlović 2008, 209). c[acue]t

Especially, when faced with external threats such as the NATO bombing campaign,
‘the differences between the national, the regime, and the leader became blurred’
(Cohen 2001, 285; italics added). Personalization concealed, and probably abetted, the
fact that party programs on topics most salient to the public were either vague or simi-
lar to one another. The highly personal nature of politics continued even after the
regime change in 2000, as the struggle between the president, Vojislav Koštunica, and sc[aorn]

the prime minister, Din d ić, intensified (Pribicevic 2004) following the overthrow of
D
[orkts] odr[kst] c[acue]t

Milošević. One can expect leader-centered competition to dilute both the ideological
sc[aorn] c[acue]t

identification and the heuristic value of the left–right schema for party and issue
preference.
Downloaded by [University of Haifa Library] at 14:39 29 September 2013

Another possible complicating factor for left–right identification in Serbia was

the tacit ‘red-brown’ coalition between the nominally socialist SPS and the Serbian
Radical Party (SRS). Zakošek (2008) describes this alliance of ideological extremes
sc[aorn]

dominating the party system and marginalizing centrist opposition. The paradox of
cooperation between parties professing diametrically opposing ideologies can be
explained by SRS campaigns appealing to patriotism and support for the creation of
Greater Serbia – precisely the themes through which SPS mobilized voters throughout
the 1990s. Evidence that ‘the nominally left-wing SPS and the right-wing SRS …
were closely associated in the minds of their supporters’ (Gordy 1999, 46), and that ‘a
large segment of either party found it easy to transfer its support to the other’ (Cohen
2001, 167) is provided by survey results from the mid-1990s showing that while SPS
and SRS voters differed in terms of nostalgia for communism (much higher among
SPS adherents), they shared similar attitudes on nationalism, xenophobia (high), liber-
alism, and post-materialism (low) (Pantic 1998). After the fall of Milošević, SRS has sc[aorn] c[acue]t

become ‘a party that combines elements of both left and right extremism’ (Slavujevic
2003, cited in Konitzer 2008, 739).
What cleavages prevailed in Serbia during the first post-communist decade, and
what changes can one expect after the regime change in 2000? Orlović mentions that c[acue]t

‘during the 1990s political competition was reduced to the conflict between national-
ists and westerners’ (2008, 207), yet the latter camp was hardly present on the elec-
toral stage. In all elections throughout this decade, Goati observed that ‘issues of
national and state identity were at he center of political controversies, while the issues
of living standards were relegated to the background’ (2000, 68), leaving little room
for the development of a socio-economic cleavage. In contrast, whereas privatization
was blocked under the Milošević regime due at least in part to ideological reasons
sc[aorn] c[acue]t

(Goati 2000, 71), governments in the 2000s have not favoured the maintenance of a
large state sector (Hudson 2003, 147). It is possible that as economic reform proceeds,
parties could mobilize support by focusing on disparities between winners and losers
of this process.
Nikolic-Ristanovic (1998, 251–2) characterized Serbian society as divided
between supporters and opponents of the government, but cautioning against equating
the latter with opposition party backers. In other words, this represented a social
divide that was not entirely (at least not successfully) mobilized into a political cleav-
age. Pantic (1998, 81) observes a cleavage between traditional and modern values,
likely linked to different socio-structural constituencies, prompting Gordy’s observa-
tion of ‘the deep social division between older and younger people; between urban and
rural people; and between the small group of highly educated people and the much
 Southeast European and Black Sea Studies 193

larger group of less educated people’ (1999, 51–2). Todosijević identifies two factors,
c[acue]t

labeled alienation-egalitarianism and pro-communist nationalism, with the latter

encompassing ‘a mixture of leftist and rightist anti-liberal views’ (2008, 888–90). In
addition, he illustrates that party system fragmentation notwithstanding, competition
was essentially uni-dimensional and bipolar.

Data and hypotheses

Data for this study are taken primarily from two surveys, Yugoslav Public Opinion
1996 (ZA 2911) and the fourth wave of the World Values Survey conducted in Serbia
in November 2001. Despite differences in question wording and response scales, these
surveys contain the most complete range of questions for testing.4 They also permit
Downloaded by [University of Haifa Library] at 14:39 29 September 2013

comparison of changes and continuities in the contents underlying left–right orienta-

tions before and after the 2000 regime change. Two additional surveys from 1990 and
2002 (ZA 2903, ZA 3893), while lacking many questions probing general political
attitudes, include a few items that allow analysis of longer-term evolution of the
understanding of left and right. Only respondents in Serbia are included in the analysis
below.5
Five clusters of independent variables are tested for their effect on respondents’
self-placements on the left–right spectrum, which is standardized around a seven-
point scale (1 = left, 7 = right). Socio-demographic variables include age, gender,
education, size of community, and religiosity. Dummy variables are created for
respondents who belong to the occupational categories of worker (skilled and
unskilled manual labour) and owner (employers in private enterprises). One expects
that, similar to other post-communist societies, younger cohorts with a shorter period
of socialization under communism would lean less toward the left than their elders.
Goati (1998, 29) notes a division between ‘two Serbias’, a rural/suburban portion with
strong socialist support versus an urban segment backing the opposition. Similarly,
Gordy points to the concentration of SPS voters among older voters and those residing
in less developed parts of the country, as well as ‘a linear decrease in support for SPS
… with educational attainment’ (1999, 52–4). However, in view of the aforemen-
tioned lack of programmatic distinction between government and opposition, these
socio-structural differences may not necessarily be reflected in a relationship between
community size or educational attainment and left–right placement. Significance of
either occupation dummy variable would indicate a class cleavage, though this is not
expected if issues of nationalism and ethnicity rather than economic distribution domi-
nated political discourse.
Other indicators of a cleavage based on economic interests include attitudes
toward privatization and confidence in trade unions and private enterprises. Even
when implementing privatization on a limited scale, the government ‘remained
committed to some elements of a socialist economy’ (Hudson 2003, 140), so this
represented a salient issue in the debate over economic reform. Also, if transition
‘winners are generally oriented toward liberal democratic parties … [and] losers are
more oriented to social national parties’ (Orlović 2008, 220), and assuming that
c[acue]t

winners and losers are at least in part defined by the impact of transition on living stan-
dards, then respondents’ level of satisfaction with personal financial conditions may
be significantly related to their left–right positioning. However, this would only be
true if egocentric economic evaluations and liberal democratic orientations coincide
on the same cleavage.
 194 W. Jou

In categorizing stages of state and societal pluralization in post-communist coun-

tries, Simon describes a scenario in which:

[T]he political force originating in the former socialist state will remain strong for a long
time; it will be able to hold its position even after elections and either the democratiza-
tion process of state institutions will be protracted or will stop. Such a position means
that even after elections, the power and influence of the former communist and authori-
tarian forces will be strong. Democratization can be protracted and the state will preserve
many authoritarian features. (1997, 363)

This account accurately portrays the state of incomplete transition in Serbia during
the 1990s, when many opposition parties denied legitimacy not only to the incumbent
authorities but also the constitutional order itself (Goati 2000, 72). At the same time,
Downloaded by [University of Haifa Library] at 14:39 29 September 2013

and unlike its Yugoslav predecessor, one can consider the Milošević government a
sc[aorn] c[acue]t

‘semi-democracy’ which held elections and permitted both opposition parties and
some independent media (Cohen 2001, 423), so that debate over regime preference
could, and did, take place. Under these circumstances, one expects the regime divide
to form a major cleavage, in terms of attitudes toward the former communist regime
(and, by implication, its successor party that remained in power), satisfaction with the
extant tentative democratic process, and confidence in the current government.
A related, though theoretically distinct, concept from preference or nostalgia for
the former communist regime is that of authoritarian attitudes. Such orientations exist
even in established democracies, often finding expression in far right parties stressing
law and order issues mixed with a xenophobic agenda. Kuzmanović (1995) character- c[acue]t

izes Serbia’s political culture as entailing inclination toward authoritarian values, will-
ingness to support strong leadership, and preference for collectivist nationalism, and
Nikolic-Ristanovic perceives a ‘peasant-based, patriarchal society’ with a ‘collectivist
mentality’ and desire for firm leadership (1998, 250–1) particularly susceptible to
authoritarian appeals. Todosijević corroborates this by concluding that ‘the overarch-
c[acue]t

ing left-authoritarian dimension … is the crucial ideological division’ in Serbia (2008,

894), though in the absence of an opposite right-libertarian pole, one may raise the
question whether this constitutes the cleavage defining left and right or merely a
valence issue. This dimension is captured by variables measuring obedience to author-
ity (or whether children should be inculcated with this value) and the desirability of
strong leaders.
Finally, the government’s cultivation of an ethnically exclusive Serbian identity
(rather than a more inclusive Yugoslav one) led to clear demarcation of previously
blurred boundaries among ethnic groups and sharp increases in ethnic distance
(Golubovic 1995, cited in Stefanovic 2008, 1199). By engaging in historical revision-
ism and fostering a sense of Serbian victimhood, the government sought justification
for targeting other ethnic groups. Paradoxically, the ascendance of nationalism as the
single predominant political issue might prevent it from forming the basis of a cleav-
age, since not only the ruling SPS but also opposition parties of all ideological shades
fought to don the nationalist cloak, leaving no space for of alternative stances (Cohen
2001; Gordy 1999). As nationalism became linked with anti-western sentiments due
to sanctions and NATO bombings, ‘the cleavage of a pro-European orientation (pro-
modernization) vs. traditionalism (anti-modernization)’ emerged (Fink-Hafner 2008,
176). This justifies testing the impact of Serb nationality, views on the Kosovo
dispute, national pride, and confidence in the EU on the structuring of left–right orien-
tations in the following section.
 Southeast European and Black Sea Studies 195

Anchors of left–right orientations

Table 1 shows results of ordinary least square regression with left–right self-
placement as the dependent variable. On a seven-point scale, average left–right posi-
tions move from 3.66 (1990), 3.89 (1996), 4.01 (2001), to 4.12 (2002), with standard
deviations between 1.40 and 1.51. While results from four surveys are shown for the
purpose of cross-temporal comparison, levels of significance for 1990 and 2002
should be interpreted with caution due to the absence of several attitudinal measures
that are found to have substantial impact on respondents’ left–right positioning in the
more complete surveys of 1996 and 2001. For example, while it appears that those
who favour private ownership in 1990 are significantly more likely to lean toward the
right, but that salience of this issue declined in later years, one can offer an alternative
explanation positing a linkage between support for privatization and dissatisfaction
Downloaded by [University of Haifa Library] at 14:39 29 September 2013

with communist rule, so the former may cease to be significant if the latter were
included as an independent variable. The same can be said for the worker dummy vari-
able in 2002.
One can observe three sets of variables that have exerted significant influence on
left–right orientations across time. Similar to many post-communist societies, older
respondents who have spent most of their lives under communism are inclined toward
the left. This distinction held true as early as 1990, when even the youngest cohort had
no experience of non-communist rule. Religion also plays a significant role, either in

Table 1. Impact on left–right orientations.

1990 1996 2001 2002
Age −0.12 ** −0.14 ** −0.10 ** −0.15 **
Gender 0.02 −0.10 * −0.04 −0.04
Education −0.03 −0.10 0.01 0.07
Town size 0.00 0.02 0.04 0.01
Religiosity −0.21 ** −0.04 −0.16 **
Worker (dummy) −0.05 0.03 0.06 0.12 **
Owner (dummy) 0.05 0.03 0.05
Financial satisfaction −0.11 * 0.02
Private ownership −0.10 ** −0.03 −0.07
Confidence in Unions 0.13 * 0.06
Confidence in Companies −0.09 −0.02
Past regime evaluation −0.17 ** −0.31 **
Democratic satisfaction −0.10 −0.13 ** 0.02
Confidence in Government 0.27 ** −0.02 −0.02
Strong leader 0.05 −0.07 0.00
Respect authority 0.02 0.07 0.06
Confidence in Churches −0.08 −0.16 ** −0.03
Serb national 0.01 0.12 ** 0.03
Kosovo 0.04 0.04 0.01
National pride 0.07 * −0.04
EU −0.15 ** 0.05 0.02
Adjusted R2 0.03 0.41 0.25 0.05
Standardized coefficients shown: *p < 0.05; **p < 0.01.
 196 W. Jou

the form of personal religiosity (1996, 2002) or confidence in churches (2001). It

should be noted that, unlike other ex-Yugoslav republics such as Croatia, Macedonia,
and Slovenia, no political party in Serbia during the period examined ran election
campaigns by appealing to religious values.6 Thus one sees evidence of a divide
rooted in social structure anchoring left and right identification without explicit polit-
ical mobilization. In contrast, despite greater support for SPS-led governments in rural
areas through the 1990s, community size did not have an independent effect on left–
right placements.
Evaluation of the political system under communism is statistically and substan-
tially significant in both 1996 and 2001, with respondents rating communist rule nega-
tively aligned on the right. This confirms that the regime divide persisted not only
through the period of halted democratization under Milošević, but also after the 2000
sc[aorn] c[acue]t
Downloaded by [University of Haifa Library] at 14:39 29 September 2013

regime change. Furthermore, while evaluation of communism is related to satisfaction

with the democratic process (more satisfied respondents lean toward the right),7 these
two variables are by no means synonymous, and exert independent influences on left–
right orientations (p-value of the coefficient for democratic satisfaction just slightly
exceeds 0.05 in 1996). Some leading figures within opposition parties during the
1990s began their careers in the communist party, which inhibited them from credibly
repudiating the former regime. On the other hand, being unable to differentiate them-
selves from the ruling SPS on nationalist issues, their opposition to the government
focused on its undemocratic means of exercising power, giving rise to a separate
cleavage centered on criticism of democratic deficit.
Two variables that turn out to be significant for anchoring left–right orientation in
1996 but not longer so in 2001 are notable. Whether respondents expressed confidence
in the Serbian government had a decisive impact in the mid-1990s,8 when polarization
between government and opposition constituted a cleavage independent of evaluation
of communist rule and satisfaction with democratic performance. In other words,
reasons for opposing the regime comprised not only of its communist past and dicta-
torial tactics. As seen in both 2001 and 2002, confidence in government no longer struc-
tured ideological placement after 2000, presumably because the new round of power
struggle primarily involved personalities and parties within the former opposition alli-
ance (Democratic Opposition of Serbia, DOS) rather than between it and the SPS.
The implication of this development for democratic consolidation is ambiguous:
while an end to sharp ideological polarization creates an environment conducive to
compromise, some would argue that the quality of democracy is enhanced when
parties in both government and opposition adopt clear programmatic positions identi-
fiable with either left or right (rather than a vaguely defined center), so that supporters
and opponents of government policies should be distinguishable by left–right orienta-
tion. At the same time, institutional rules aimed at promoting consensus and broad
coalition formation are not uncommon in post-conflict societies, in order to bridge
disagreements over provocative issues such as those that sharply divided the Serbian
public. Since no single party has commanded a parliamentary majority in Serbia since
the late 1990s, coalitions became a regular feature of government, possibly leading to
decreased ideological distance among parties. An encouraging sign for democracy can
be found in the fact that satisfaction with democratic performance no longer formed a
cleavage dividing left and right in 2002, which suggests that the axis of political
contention has moved beyond debate over the desirability of democratic rules.
The other notable difference between 1996 and 2001 is found with respect to
confidence in the EU. Since the Serbian government in the 1990s not only resorted to
 Southeast European and Black Sea Studies 197

inciting prejudice against other ex-Yugoslav ethnic groups but also denouncing the
West for imposing sanctions (and later for NATO bombings), attitudes toward the EU
became a symbolic issue dividing nationalistic traditionalists from cosmopolitan
modernizers. Since Serbia had no prospect of EU membership during this period, one
must assume that confidence in the EU entailed abstract notions of infringement on
sovereignty (for opponents) or democracy and the market economy (for supporters)
rather than concrete, interest-based issues such as agricultural subsidies or environ-
mental regulations.9 After the DOS coalition came to power 2000, pro-European
views became more widespread. While attitude toward the EU continues to be politi-
cally divisive, with SPS and SRS adopting an anti-European stance in contrast to all
other major parties, it ceased to influence left–right orientations.
It is notable that emphasis on respecting authority and preference for strong lead-
Downloaded by [University of Haifa Library] at 14:39 29 September 2013

ers did not play a role in structuring left–right orientations, indicating that ideological
positioning is not influenced by an authoritarian–libertarian axis. This (non-)finding
is not attributable to the overlapping of libertarian values with negative evaluations
of communist rule or dissatisfaction with lack of democratization, since even in the
absence of the latter two variables in 1990, authority and leader variables had no
impact. There is also little evidence pointing to cleavages based on class or economic
interests. Whereas one may expect economic issues to assume greater salience after
2000 as polarization over the Milošević regime subsided, in fact the two variables that
sc[aorn] c[acue]t

attained significance in 1996, satisfaction with personal financial conditions and

confidence in labour unions, no longer influenced left–right placements by 2001.10 It
is open to question whether the worker dummy in 2002 reached significance only in
the absence of other variables; moreover, the fact that workers lean to the right does
not accord with normal concepts of a class cleavage.
The relationship between nationalism and left–right orientation has long raised
theoretical difficulties, since how nationalist sentiments are mobilized often depends
on the identity of the ‘other’, the alleged enemy of the nation. Since the 1990 and 1996
surveys do not contain a question on national pride, it is impossible to gauge the impact
of this variable on left–right placement during the 1990s, though results in 2002 suggest
that it is has ceased to form a major political cleavage. The significance of Serbian
ethnicity in 1996, during a phase when political discourse was mainly conducted
through an ethnic prism, was less surprising than the positive sign of the coefficient,
meaning that ethnic Serbs tended to identify with the right despite the Miloś ević regime
sc[aorn] c[acue]t

that proclaimed itself protector of the Serbian nation being nominally socialist. It is
important to note here that ethnic Serbs comprised approximately four-fifth of respon-
dents in each survey meaning that there is limited variance on this measure.11
The status of Kosovo had been (and continues to be) an issue galvanizing intense
debate in Serbia. Since the status of the autonomous province underwent important
changes during the period covered by the surveys, response categories in each ques-
tionnaire are dissimilar. I have classified respondents’ opinion on the Kosovo issue
into three categories, representing soft-line (coded 1), middle-of-the-road (2), and
hard-line (3) positions in the context of circumstances at each time point. In none of
the three survey years did views on Kosovo correlate significantly with left–right
placements. What Table 1 does not show, however, is a consistently curvilinear rela-
tionship: respondents supporting both soft- and hard-line stances occupy almost the
exact same location on the left–right spectrum, while those backing a middle position
lean more toward the left. This trend was particularly remarkable in 1990 and 1996,
when the difference exceeded 0.5 on a seven-point scale.
 198 W. Jou

Lastly, one observes that variance in left–right placements accounted for by all
independent variables is considerably lower in 2001 than 1996, suggesting that ideo-
logical orientation became less structured after 2000. One possible explanation lies
in the disappearance of a major cleavage, namely the sharp conflict between the
Milošević regime and its opponents throughout the 1990s. Concomitantly, as the
sc[aorn] c[acue]t

series of wars that lacerated much of the former Yugoslavia came to an end with
the forced Serbian withdrawal from Kosovo, the tactic of inciting nationalist senti-
ments and ethnic tensions lost credibility. Whether political conflict would become
more firmly anchored in new issue dimensions as a consequence of post-2000 demo-
cratic opening, or whether questions linked to the past such as attitude toward the
former communist system would persist as a major cleavage, awaits testing of public
opinion data collected at later dates.
Downloaded by [University of Haifa Library] at 14:39 29 September 2013

As described above, one interpretation of Serbian political history in the 1990s

consists not of left versus right, but instead extremes versus the center, with the label
‘extreme’ applied to both socialists on the left and radicals on the right. Did this so-
called ‘red-brown’ alliance only reflect short-term strategic choices of party elites,12
or was it buttressed by shared outlooks among supporters of both parties? A dependent
variable measuring extremism has been created by recoding placements at both ends
of the left–right spectrum as 2, penultimate positions as 1, and all other locations of 0.
Respondents classified in the most extreme category ranged between 9.1% and 12.5%,
while those in the next most extreme positions comprised between 7.4% and 17.9%
of the samples, leaving 7–8 out of 10 respondents in the residual category. Utilizing
the same independent variables as described above, regression results are shown in
Table 2.
A quick glance through the table suggests that most independent variables had no
impact on ideological extremism, and the very low variance accounted for in all
surveys confirms this conclusion. The only factor consistently differentiating extrem-
ists from moderates is the former’s preference for strong leadership. Extremists on
both sides of the ideological spectrum were characterized by a greater number of
common attitudes not during the 1990s, but after the overthrow of the Milošević sc[aorn] c[acue]t

regime in 2000. One may speculate that since the government throughout the 1990s
professed a leftist ideology, its opponents would include both centrist and rightist
forces. After 2000, as the new DOS government sought to steer a center (or center–
right; see below) course, SPS on the left and SRS on the extreme right now found
themselves sharing the role of opposition, thus creating a basis for a pattern of polar-
ized pluralism. However, this does not necessarily imply that extreme rightist and left-
ists hold similar views on most issues, nor does it herald formal cooperation between
parties on the margins of the ideological spectrum.

Left–right orientations and vote choice

Since the casual arrow between left–right placement and party identification can point
in both directions, party support is not included as an independent variable explaining
the structuring of left–right orientations. Nevertheless, it is important to examine the
relationship between party preference and left–right positioning. Table 3 lists correla-
tions between these two factors for respondents voting for each party or coalition from
1996 to 2002.13 The table confirms that despite the predominance of ethnic national-
ism in political discourse during the 1990s, attitudes on this question did not define
left and right, in the absence of an anti-nationalist pole. Instead, one observes a clear
 Southeast European and Black Sea Studies 199

Table 2. Impact on extremist left–right placements.

1990 1996 2001 2002
Age −0.02 −0.07 −0.01 0.03
Gender −0.11 ** −0.01 −0.01 0.01
Education 0.05 −0.12 −0.02 −0.03
Town size −0.02 0.11 −0.03 −0.06
Religious 0.02 −0.01 −0.11 *
Worker (dummy) 0.04 −0.08 −0.02 0.06
Owner (dummy) −0.03 0.01 0.02
Financial satis. −0.06 0.02
Private ownership 0.04 0.01 −0.12 **
Downloaded by [University of Haifa Library] at 14:39 29 September 2013

Conf. Union 0.03 −0.02

Conf. Companies −0.02 −0.01
Post-regime evaluation 0.06 0.09 *
Democratic satisfaction −0.03 0.10 * 0.01
Conf. Government −0.01 0.08 −0.02
Strong leader −0.09 * −0.22 ** −0.09 *
Respect authority 0.05 0.10 −0.08 *
Conf. Churches −0.03 −0.01 0.03
Serb national 0.07 * 0.00 0.01
Kosovo 0.03 0.06 0.12 *
National pride 0.01 −0.07
EU 0.07 0.02 0.01
R2 0.02 0.02 0.03 0.03
Note: Dependent variable: 2 = most extreme positions; 1 = penultimate extreme positions; 0 = other.
Standardized coefficients shown: *p < 0.05; ** p < 0.01.

Table 3. Left–right orientation and party choice.

1996 2001 2002
SPS −0.51 ** −0.41 ** −0.40 **
SRS 0.40 ** 0.02 0.17 **
JUL −0.24 **
DEPOS 0.24 **
DS 0.08
DSS 0.20 **
DOS 0.26 ** 0.13 **
SSJ −0.06
SPO 0.12 **
Note: SPS, Socialist Party of Serbia; SRS, Serbian Radical Party; JUL, Yugoslav Left; DEPOS,
Democratic Movement of Serbia; DS, Democratic Party; DSS, Democratic Party of Serbia; DOS,
Democratic Opposition of Serbia; SSJ, Party of Serbian Unity; SPO, Serbian Renewal Movement.
Pearson correlation coefficients shown: *p < 0.05; **p < 0.01.
 200 W. Jou

government versus opposition cleavage, with SPS and Yugoslav Left (JUL) voters on
the left and most opposition supporters on the right. The DS stands out as the sole
exception, attributable to party’s deliberate efforts to deemphasize ideology and focus
on bread-and-butter issues.14 With reference to Table 1, one may interpret this domi-
nant cleavage as rooted in the evaluation of communist rule (both SPS and JUL are
direct descendants of the former ruling party) and (dis)satisfaction with the extant
state of democracy.
The cleavage pattern pitting SPS against all other parties seems to persist even
after the change of government in 2000. SPS remains firmly anchored on the left,
while the new DOS coalition draws its support from right-leaning voters. Opinion on
the past continues to structure left–right orientations, probably combining evaluation
of both the communist and Milošević regimes. As the main political cleavage polar-
sc[aorn] c[acue]t
Downloaded by [University of Haifa Library] at 14:39 29 September 2013

ized between left-wing adherents of the former authoritarian government and right-
wing backers of the new democratic one, voters of the far right radicals unexpectedly
found themselves close to the center in terms of left–right location. Perhaps SRS
voters were simply unsure of their positions, since standard deviations for their self-
placements are always the highest among all parties, and this difference was largest in
2001. In terms of both vote shares and parliamentary representation, the early 2000s
witnessed the ascendancy of right-wing parties. Yet the mean self-placement of all
voters showed remarkable stability, moving only slightly from 3.87 in 1996 to 4.06
and 4.18 in 2001 and 2002, respectively.15
Table 4 lists correlations between party support and ideological extremism, with
the same three-category coding used in Table 2 as the dependent variable. Since
extremism is frequently linked with political alienation, one expects that supporters of
governing parties or coalitions are least likely to profess such sentiment. This was
indeed the case in 2001 and 2002, when DOS voters displayed more centrist inclina-
tions. In contrast, SPS backers in 1996 were not more moderate than supporters of
then opposition parties. Most observers of Serbian politics would agree that SRS
voters in 1996 were most inclined to place themselves on the extremes of the ideolog-
ical spectrum. Explanation for their turn toward moderation in 2001 lay at least as
much in the shifting anchors of left and right as their own attitudinal volatility and
uncertainty. SPS supporters moved sharply toward the left extreme after their party

Table 4. Extremist left–right placement and party choice.

1996 2001 2002
SPS −0.033 0.202 ** 0.118 **
SRS 0.296 ** 0.033 0.088 *
JUL 0.079
DEPOS −0.044
DS −0.109 *
DSS −0.089
DOS −0.193 ** −0.175 **
SSJ −0.022
SPO 0.046
Note: SPS, Socialist Party of Serbia; SRS, Serbian Radical Party; JUL, Yugoslav Left; DEPOS,
Democratic Movement of Serbia; DS, Democratic Party; DSS, Democratic Party of Serbia; DOS,
Democratic Opposition of Serbia; SSJ, Party of Serbian Unity; SPO, Serbian Renewal Movement.
Pearson correlation coefficients shown: *p < 0.05; **p < 0.01.
 Southeast European and Black Sea Studies 201

lost power in 2000 (from 2.91 in 1996 to 2.27 in 2001). While they were joined by
SRS partisans in 2002, this did not augur a common platform between these two
groups. Notwithstanding the shared nationalistic and Euro-skeptic views, their policy
differences still outnumber similarities.

Discussion and conclusion

The preceding pages have reviewed theories on factors anchoring left–right orienta-
tions in post-communist settings, as well as circumstances unique to Serbia, and tested
them through analysis of public opinion data. While both the unavailability of certain
attitudinal items at various time points and differences in question wording caution
against drawing direct comparisons and definitive conclusions, one can nevertheless
Downloaded by [University of Haifa Library] at 14:39 29 September 2013

discern important patterns regarding changes and continuities in the structuring of

political attitudes. In addition to age and religiosity, the regime divide constituted the
main cleavage in Serbia, as evaluation of communist rule, and to some extent satisfac-
tion with democratic development, proved to be important factors distinguishing and
dividing left and right. An economic or class-based cleavage, if one is present at all,
appeared to play only a minor role, and no authoritarian–libertarian cleavage was
found. Questions on nationalism and ethnicity did not reveal consistent results over
time. Whereas this dimension overshadowed all other issues during most of the 1990s,
it represented a valence rather than positional issue, so did not develop into a political
cleavage. It also bears reiteration that questions of national identity are particularly
difficult to incorporate into the left–right schema.
Cleavage dimensions can also be discerned in part by correlations between vote
choice and left–right placements, confirming that voters are well aware of where
parties stand in relation to the most important political divides. Furthermore, this
offers evidence that even under circumstances that seem unfavourable to program-
matic party competition – disputes over state boundaries and status, mobilization of
ethnic nationalism, incomplete democratization or even reversion to authoritarian rule
– left and right nevertheless provide useful heuristics in guiding vote choice. Such
structuring of preferences corroborates observation by Miller, White, and Heywood
(1998) that attitudes toward parties in post-communist countries are not random but
organized by political values as well as socio-demographic profile and economic
interest. Their finding that attitudes toward parties are influenced by the regime
divide, a government versus opposition divide, and nationalism is particularly appli-
cable in the Serbian context. Yet, if one believes that this axis of competition was a
product of specific circumstances during the 1990s, a decline in its salience – and the
emergence of alternative cleavages, for instance the divergent interests of winners and
losers of economic reforms – can be expected, especially as a new generation of poli-
ticians uninvolved in conflicts of the Milošević era ascends the political stage.
sc[aorn] c[acue]t

One may argue that it is precisely where voters are not offered the same set of choices
on the ballot each time (e.g., shifts in the composition of anti-SPS coalitions, from
DEPOS to Zajedno to DOS) that left–right orientations are most likely to anchor party
and issue preferences, in place of party identifications that have little time to take root
(Pierce 1981). Examples of frequently changing party constellations can be found in
established democracies, but are more often observed in countries where the electoral
market was opened suddenly and unexpectedly. Ideology may thus be particularly
important during the period of transition (e.g. Hanley 2004), which was more protracted
in Serbia than in most other post-communist democracies. The Serbian electorate seems
 202 W. Jou

to resemble its Polish counterpart in demonstrating a relatively stable structure of left–

right orientations despite high inter-party volatility (Szczerbiak 2008).
From a comparative perspective, the most important cleavages in Serbia during the
first decade of post-communism, including not only the regime divide but also the
salience of religiosity, are similar to those found in Croatia. In the latter case, opinions
on abortion and church role in state affairs, as well as level of religiosity, have
consistently influenced left–right placements throughout the 1990s and into the early
2000s. These parallels are hardly surprising in view of the two countries’ common
experiences of war and semi-authoritarian leadership following the break-up of
Yugoslavia. At the same time, Croatian surveys reveal an authoritarian–libertarian
cleavage not found in Serbia, as left and right identifiers are also distinguished by their
views on strongman rule and banning strikes to preserve public order (both favoured
Downloaded by [University of Haifa Library] at 14:39 29 September 2013

by right identifiers). Religiosity is also highly significant in another ex-Yugoslav

republic, Slovenia, though in this case neither preference for strong leader nor satis-
faction with democracy affected left–right orientations at any time during the 1990s.
These comparisons highlight the importance of contrasting contemporary experiences
in structuring axes of political competition, notwithstanding a common historical
legacy of Yugoslav communism.
What implications do different cleavage patterns have for democratization?
Zuckerman (1975) asserts that cross-cutting cleavages are good for democratic devel-
opment, and Lin, Chu, and Hinich (1996) emphasize the positive effect of new cleav-
ages displacing the regime divide. According to McDonough, Barnes, and Pina
(1986), depolarization of democratization itself as a source of political contention
signals consolidation. Regarding the contents of political conflicts, one may argue that
interest-based cleavages (e.g. speed of privatization) allow greater room for negotia-
tion and compromise, in contrast to fixed identity-based cleavages such as religion
and ethnicity. Bochsler’s report that key topics raised in the most recent parliamentary
election involved ‘Serbia’s past, inter-ethnic questions, and Euro-Atlantic integration’
(2008, 161) partly confirms findings in this study, but more importantly, raise ques-
tions on whether party competition over these themes recurs at the expense of issues
closer to the lives of most voters. Also, if a high proportion of extreme left–right posi-
tioning is associated with greater alienation from the regime (Miller and Listhaug
1993), the continued pariah status of SRS, the largest single party in both the 2003 and
2007 elections before an acrimonious split in 2008,16 could only reinforce anti-system
sentiments and cause concern for democratic stability.

Notes
1. Goati (2000, 70) notes that this process of ethnification was considerably more intense in
Serbia than in Montenegro.
2. Gordy portrays the decision to go to war as politically motivated: ‘although a large propa-
ganda campaign was under way, war was not engaged until the Serbian regime was seri-
ously threatened by its own internal opposition’ (1999, 59).
3. Opposition parties followed the government line in criticizing international sanctions.
Among leading anti-regime figures, Dind ić ‘believed it necessary to play nationalist card’,
ordk[ts] cea]u[t

while Koš tunica ‘adopted a more explicitly nationalist orientation’ than other opposition
csoa[rn]

leaders (Cohen 2001, 176).

4. While using the third wave of the World Values Survey, also conducted in 1996, would
avoid problems of comparability with the fourth wave, a national survey is chosen because
it contains questions on nationality and Kosovo that permit comparison with surveys from
1990 and 2002.
 Southeast European and Black Sea Studies 203

5. N = 1400 (1990), 1486 (1996), 1200 (2001), 1200 (2002). Sample excludes Kosovo in 2002.
6. A Christian Democratic Party of Serbia (DSS) was founded in 1997, but did not win any
legislative seats.
7. The bivariate correlation is 0.273 in 1996, 0.307 in 2001 (p < 0.001 in both cases).
8. Confidence in the Serbian rather than Yugoslav government is used where both questions
are included in the surveys, because ‘the parliament and the government of the FRY
[Federal Republic of Yugoslavia] still existed only formally, while all power was concen-
trated in the republics’ (Goati 1998, 14), and the Serbian parliament was ‘where most of
the real legislative power lay’ (Birch 2002, 508).
9. Paradoxically, while 71% of respondents in 1996 professed little or no confidence in the
EU, the same survey reveals that 81% were in favour of eventual EU membership.
10. In 1996, the opposition ‘Together’ coalition included the Association of Free and Indepen-
dent Trade Unions of Serbia (Vukomanovic 1998, 41), which may explain the significance
of the confidence in unions variable.
11. 79.2% in 1990; 79.8% in 1996; 83.5% in 2002 (which excluded Kosovo).
Downloaded by [University of Haifa Library] at 14:39 29 September 2013

12. SRS joined other opposition parties in 1993 and became a vehement critic of the SPS
government.
13. The 1996 and 2001 surveys contain question on voting intentions, while the 2002 survey
asks respondents how they voted in the previous (2000) election.
14. While Cohen (2001, 123) places the DS on the center–left of the political spectrum, the
party sought ‘positioning as a center party both on the classic left–center–right spectrum
and a middle position between pro-regime and anti-regime’ as a strategy ‘aimed at attract-
ing dissatisfied but non-radical voters’ (Slavujevic 1998, 103).
15. The electorate expressed a more leftist orientation back in 1990, averaging 3.63 on a seven-
point scale.
16. Long-time SRS leader Vojislav Šeš elj, standing trial at the International Criminal Tribunal
S[coarn] csoa[rn]

for the Former Yugoslavia, disagreed with his successor Tomislav Nikoli ć’s pro-EU
cea]u[t

stance. Expelled from the SRS, Nikolić formed the Serbian Progressive Party (SNS) in
cea]u[t

October 2008, which has adopted a mainstream conservative profile.

References
Bielasiak, J. 1997. Substance and process in the development of party systems in East Central
Europe. Communist and Post-Communist Studies 30: 23–44.
Birch, S. 2002. The 2000 elections in Yugoslavia: The “bulldozer revolution”. Electoral Studies
21: 499–511.
Bochsler, D. 2008. The parliamentary election in Serbia, January 2007. Electoral Studies 27:
160–5.
Cohen, L.J. 2001. Serpent in the Bosom: The rise and fall of Slobodan Milosevic. Boulder,
CO: Westview Press.
Conover, P.J., and S. Feldman. 1984. How people organize the political world: A schematic
model. American Journal of Political Science 28: 95–126.
Dalton, R.J. 2006. Social modernization and the end of ideology debate: Patterns of ideologi-
cal polarization. Japanese Journal of Political Science 7: 1–22.
Dowley, K.M., and B.D. Silver. 2002. Social capital, ethnicity and support for democracy in
the post-communist states. Europe-Asia Studies 54: 505–27.
Downs, A. 1957. An economic theory of democracy. New York: Harper.
Ekman, J., and J. Linde. 2005. Communist nostalgia and the consolidation of democracy in
Central and Eastern Europe. Journal of Communist Studies and Transition Politics 21:
354–74.
Evans, G., and S. Whitefield. 1993. Identifying the bases of party competition in Eastern
Europe. British Journal of Political Science 23: 521–48.
Evans, G., and S. Whitefield. 1995. Social and ideological cleavage formation in post-communist
Hungary. Europe-Asia Studies 47: 1177–204.
Evans, G., and S. Whitefield. 1998. The structuring of political cleavages in post-communist
societies: The case of the Czech Republic and Slovakia. Political Studies 46: 115–39.
Evans, G., and S. Whitefield. 2000. Explaining the formation of electoral cleavages in post-
communist democracies. In Elections in Central and Eastern Europe, ed. H.-D. Klingemann,
E. Mochmann, and K. Newton, 36–67. Berlin: Edition Sigma.
 204 W. Jou

Fink-Hafner, D. 2008. Europeanization and party system mechanics: Comparing Croatia,

Serbia and Montenegro. Journal of Southern Europe and the Balkans 10: 167–81.
Fuchs, D., and H.-D. Klingemann. 1989. The left–right schema. In Continuities in political
action: A longitudinal study of political orientations in three western democracies, ed.
M.K. Jennings, J.W. van Deth, et al., 203–43. Berlin: Walter de Gruyter.
Geddes, B. 1995. A comparative perspective on the leninist legacy in Eastern Europe.
Comparative Political Studies 28: 239–74.
Goati, V. 1998. Introduction: Political development in the Federal Republic of Yugoslavia
(1990–1996). In Elections to the Federal and Republican Parliaments of Yugoslavia
(Serbia and Montenegro) 1990–1996, ed. V. Goati, 12–34. Berlin: Edition Sigma.
Goati, V. 2000. The Federal Republic of Yugoslavia at the historical crossroads. In Transfor-
mation processes in the Yugoslav successor states between marginalization and European
integration, ed. H. Riegler, 65–83. Baden-Baden, Nomos: Verlagegesellschaft.
Gordy, E.D. 1999. The culture of power in Serbia: Nationalism and the destruction of alterna-
Downloaded by [University of Haifa Library] at 14:39 29 September 2013

tives. University Park: Pennsylvania State University Press.

Hanley, S. 2004. Blue velvet: The rise and decline of the new Czech right. Journal of Communist
Studies and Transition Politics 20: 28–54.
Haughton, T., and M. Rybář . 2004. All right now? Explaining the successes and failures of the
coar[n]

Slovak Centre-Right. Journal of Communist Studies and Transition Politics 20: 115–32.
Huber, J.D., and R. Inglehart. 1995. Expert interpretations of party space and party locations
in 42 societies. Party Politics 1: 73–111.
Hudson, K. 2003. Breaking the South Slav dream: The rise and fall of Yugoslavia. London:
Pluto Press.
Inglehart, R. 1990. Culture shift in advanced industrial society. Princeton, NJ: Princeton
University Press.
Inglehart, R., and H. Klingemann. 1976. Party identification, ideological preference and the
left-right dimension among western mass publics. In Party identification and beyond, ed. I
Budge, I. Crewe, and D. Farlie, 243–73. London: Wiley.
Kaase, M., and H.-D. Klingemann. 1994. The cumbersome way to partisan orientations in a
“New” democracy: The case of the former GDR. In Elections at home and abroad, ed.
M.K. Jennings and T. Mann, 123–56. Ann Arbor, MI: University of Michigan Press.
Kitschelt, H. 1992. The formation of party systems in East Central Europe. Politics and Society
20: 7–50.
Kitschelt, H. 1995. Formation of party cleavages in post-communist democracies: Theoretical
propositions. Party Politics 1, no. 4: 447–72.
Kitschelt, H., D. Dimitrov, and A. Kanev. 1995. The structuring of the vote in post-communist
party systems: The Bulgarian example. European Journal of Political Research 27: 143–60.
Kitschelt, H., Z. Mansfeldova, R. Markowski, and G. Tóka. 1999. Post-communist party
systems: Competition, representation, and inter-party cooperation. Cambridge: Cambridge
University Press.
Klíma, M. 1998. Consolidation and stabilization of the party system in the Czech Republic.
Political Studies 46: 492–510.
Knutsen, O. 1995. Value orientations, political conflicts and left-right identification: A
comparative study. European Journal of Political Research 28: 63–93.
Knutsen, O. 1997. The partisan and the value-based component of left-right self-placement: A
comparative study. International Political Science Review 18: 191–225.
Konitzer, A. 2008. The Serbian radical party in the 2004 local elections. East European Poli-
tics and Societies 22: 738–56.
Kuzmanović, B. 1995. Authoritarianism. In Society in crisis: Yugoslavia in the early ‘90s, ed.
cea]u[t

M. Lazić, 161–86. Belgrade: Filip Viš njić.

cea]u[t csoa[rn] cea]u[t

Lin, T.-M., Y.-H. Chu, and M.J. Hinich. 1996. Conflict displacement and regime transition in
Taiwan: A spatial analysis. World Politics 48: 453–81.
Lipset, S.M. 1960. Political man. Garden City, NY: Doubleday.
Lipset, S.M., and S. Rokkan. 1967. Cleavage structures, party systems, and voter alignments:
An introduction. In Party systems and voter alignments: Cross-national perspectives, ed.
S.M. Lipset and S. Rokkan, 1–64. New York: Free Press.
Lühiste, K. 2008. Support for strongman rule in ethnically divided societies: Evidence from
Estonia and Latvia. Democratization 15: 297–320.
 Southeast European and Black Sea Studies 205

Mair, P. 1997. Party system change: Approaches and interpretations. Oxford: Oxford University
Press.
Markowski, R. 1997. Political parties and ideological spaces in East Central Europe. Communist
and Post-Communist Studies 30: 221–54.
McAllister, I., and S. White. 2007. Political parties and democratic consolidation in post-
communist societies. Party Politics 13: 197–216.
McDonough, P., S.H. Barnes, and A.L. Pina. 1986. The growth of democratic legitimacy in
Spain. American Political Science Review 80: 735–59.
McManus-Czubi ń ska, C., W.L. Miller, R. Markowski, and J. Wasilewski. 2003. The new
n[aec]ut

Polish “Right”? Journal of Communist Studies and Transition Politics 19: 1–23.
Mészáros, J., N. Solymosi, and F. Speiser. 2007. Spatial distribution of political parties in
Hungary 1990–2006. Political Geography 26: 804–23.
Miller, N.J. 1997. A failed transition: The case of Serbia. In Politics, power, and the struggle
for democracy in South-east Europe, ed. K. Dawisha and B. Parrott, 146–88. Cambridge:
Downloaded by [University of Haifa Library] at 14:39 29 September 2013

Cambridge University Press.

Miller, A.H., and O. Listhaug. 1993. Ideology and political alienation. Scandinavian Political
Studies 16: 167–92.
Miller, W., S. White, and P. Heywood. 1998. Political values underlying partisan cleavages in
former communist countries. Electoral Studies 17: 197–216.
Moreno, A. 1999. Political cleavages: Issues, parties, and the consolidation of democracy.
Boulder, CO: Westview Press.
Myagkov, M., and P. Ordeshook. 1998. The spatial character of Russia’s new democracy.
Public Choice 97: 491–523.
Nikolic-Ristanovic, V. 1998. Serbia between Civil War and Democracy. Peace Review 10:
249–54.
Offe, C. 1996. Varieties of transition: The East European and East German experience.
Cambridge: MIT Press.
Orlović, S. 2008. Parties and the party system of Serbia and European integrations. Journal of
cea]u[t

Southern Europe and the Balkans 10: 205–22.

Pantic, D. 1998. Value orientations of the electorate. In Elections to the Federal and Republican
Parliaments of Yugoslavia (Serbia and Montenegro) 1990–1996, ed. V. Goati, 69–85.
Berlin: Edition Sigma.
Pierce, R. 1981. Left–Right perceptions, partisan preferences, electoral participation, and
partisan choice in France. Political Behavior 3: 117–36.
Pribicevic, O. 2004. Serbia after Milosevic. Southeast European and Black Sea Studies 4: 107–18.
Ramet, S.P. 2007. The denial syndrome and its consequences: Serbian political culture since
2000. Communist and Post-Communist Studies 40: 41–58.
Rivera, S.W. 1996. Historical cleavages or transition mode? Influences on the emerging party
systems in Poland, Hungary and Czechoslovakia. Party Politics 2: 177–208.
Rose, R., W. Mishler, and C. Haerpfer. 1998. Democracy and its alternatives: Understanding
post-communist societies. Cambridge: Polity Press.
Sell, L. 2002. Slobodan Milosevic and the destruction of Yugoslavia. Durham: Duke University
Press.
Shin, D.C., and B.-K. Jhee. 2005. How does democratic regime change affect mass political
ideology? A case study of South Korea in comparative perspective. International Political
Science Review 26: 381–96.
Simon, J. 1997. Electoral systems and democracy in central Europe, 1990–1994. International
Political Science Review 18: 361–79.
Slavujevic, Z.D. 1998. The issues: Dimensions of electoral confrontations. In Elections to the
Federal and Republican Parliaments of Yugoslavia (Serbia and Montenegro) 1990–1996,
ed. V. Goati, 86–108. Berlin: Edition Sigma.
Sniderman, P.M., R.A. Brody, and P.E. Tetlock. 1991. Reasoning and choice: Explorations in
political psychology. Cambridge: Cambridge University Press.
Stefanovic, D. 2008. The path to Weimar Serbia? Explaining the resurgence of the Serbian far
right after the fall of Milosevic. Ethnic and Racial Studies 31: 1195–221.
Szczerbiak, A. 2008. The birth of a bipolar party system or a referendum on a polarizing
government? The October 2007 Polish Parliamentary Election. Journal of Communist
Studies and Transition Politics 24: 415–43.
 206 W. Jou

Szelényi, I., É. Fodor, and E. Hanley. 1997. Left turn in post-communist politics: Bringing
class back in? East European Politics and Societies 11: 190–224.
Todosijević, B. 2008. The structure of political attitudes in Hungary and Serbia. East European
cea]u[t

Politics and Societies 22: 879–900.

Tóka, G. 1998. Party appeals and voter loyalty in new democracies. Political Studies 46:
589–610.
Vukomanovic, D. 1998. A short history of political parties. In Elections to the Federal and
Republican Parliaments of Yugoslavia (Serbia and Montenegro) 1990–1996, ed. V. Goati,
35–51. Berlin: Edition Sigma.
Whitefield, S. 2002. Political cleavages and post-communist politics. Annual Review of Political
Science 5: 181–200.
Whitefield, S., and G. Evans. 1999. Class, markets, and partisanship in post-Soviet Russia:
1993–96. Electoral Studies 18: 155–78.
Zakošek, N. 2008. Democratization, state-building and war: The cases of Serbia and Croatia.
csoa[rn]
Downloaded by [University of Haifa Library] at 14:39 29 September 2013

Democratization 15: 588–610.

Zuckerman, A. 1975. Political cleavage: A conceptual and theoretical analysis. British Jour-
nal of Political Science 5: 231–48.