Colecţia ACADEMICA 44

Seria

Coordonator serie: Advisory Board:

© 2008 Institutul European, Iaşi www. euroinst.ro INSTITUTUL EUROPEAN Iaşi, str. Lascăr Catargi nr. 43, 700107, C.P. 161 [email protected]

Descriere CIP a Bibliotecii Naţionale a României

Reproducerea (parţială sau totală) a prezentei cărţi, fără acordul Editurii, constituie infracţiune şi se pedepseşte în conformitate cu Legea nr. 8/1996. Printed in ROMANIA

BOGDAN VOICU and MĂLINA VOICU (Editors)

The Values of Romanians: 1993-2006 A sociological perspective

INSTITUTUL EUROPEAN 2008

Table of contents

Bogdan Voicu, Mălina Voicu Introduction. Romania and the comparative study of values / 11 Defining values / 11 Uses of the study of values / 13 Contemporary approaches to values and culture / 18 The structure of the current volume / 22 References / 26 Claudiu D. Tufiş Comrades or citizens? Support for democracy and market economy / 31 The simultaneity of transition to democracy and market economy / 31 A model for the analysis of support for democracy and market economy / 34 Political culture / 34 Support for the political-economic system / 35 Diffuse support for democracy / 37 Predictors of diffuse support for democracy / 38 Diffuse support for market economy / 42 Predictors of diffuse support for market economy / 43 Results / 44 Conclusions / 55 References / 57

5

Mircea Comşa Ideological self-placement: identification, sophistication, bases / 63 The theoretic relevance of the left and right concepts / 63 How the left and right concepts are used / 67 What can be understood by left and right? / 74 Left or right? / 77 The relevance of self-placement on the left-right axis / 80 Social bases of self-placement 85 Partisan bases of self-placement 88 Attitudinal-axiological bases of self-placement 92 Ideological self-positioning: between party loyalty and values / 97 Conclusions / 104 Annexes / 107 Reference list / 111 Databases used / 114 Claudiu D. Tufiş Institutional trust – victim of the postcommunist transition / 115 Trust – theoretical aspects / 117 Trust and risk / 117 Trust and social capital / 119 Institutional trust and interpersonal trust / 120 Trust and uncertainty / 120 Trust – analysis model / 123 Dependent variables / 123 Socio-economic characteristics / 124 Values / 124 Social position / 125 Psychological factors / 126 Evaluations / 126 Treatment of missing data / 127 Results / 128 Conclusions / 139 References / 142

6 corectura 2 28 iulie 2008

Mălina Voicu Religiosity and religious revival during the transition period in Romania / 144 Determinants of religious diversity in Eastern Europe / 145 The strategy for analysis and the employed indicators / 150 Religiosity and religious revival in post-communist Romania / 154 Religious revival: economic decrease or religious diversity? / 160 Conclusions / 165 Annex / 167 References / 168 Raluca Popescu Family values in Romania and in Europe / 170 The family between change and stability / 171 Changes at a demographic level / 171 Changes in family lifestyles / 172 The importance of the family / 173 The place the family occupies in the individual’s life/ 173 The importance of marriage: is marriage an outdated institution? / 177 The importance of marriage: tolerance for different aspects regarding marriage and sexuality / 178 Alternative lifestyles: single mothers / 182 Alternative lifestyles: consensual couples / 183 Roles and statuses in the family / 185 Satisfaction with the family life / 187 European patterns of family value orientations / 188 Conclusions / 191 References / 192 Paula A. Tufiş Social status and child-rearing values / 193 The role of child-rearing values in social reproduction / 197 Socio-demographic characteristics and parental values / 198 7

Mechanisms mediating the relationship between social status and parental values / 200 Data and methodology / 205 Model and sample / 205 The measurement of parental values / 207 Determinants of parental values / 209 Results / 211 Conclusions / 220 Appendix / 223 The comparability of results between urban and rural areas / 223 Additional results / 224 References / 228 Horaţiu Rusu Identity and axiological profile: value identifications for Romanian young people / 232 Theories on collective identities or “How can we establish that a certain bird is a duck?” / 233 The axiological dimension of identity / 237 Indicators and methodology / 239 What kind of value identifications can we find from young people? / 241 Young people’s values identification: which way? / 245 Conclusions/ 250 Annex / 252 Part A – The used indexes in cluster and ANOVA analysis and in their way of construction / 252 Part B – The items in the Schwartz scale / 256 References / 257 Bogdan Voicu Between tradition and postmodernity? orientations in Romania: 1993-2005 / 261

a dynamic of value

On modernization and post-modernization / 263 Attitudes towards environment protection / 268 Between tolerance and normativism / 269 Work ethos 272 8 corectura 2 28 iulie 2008

Religiosity / 274 Traditional order or autonomy? / 275 Democracy of authoritarianism? / 279 Family, marriage and gender relations / 280 Post-materialism – at the end of the tunnel? / 283 Romania 2005: (still) at the beginning of modernity / 284 A few technical considerations / 284 A map of Europe / 284 Romania 2005: differences between groups / 292 Conclusion. The fifth transition: from communism postmodernity? / 297 Annex / 301 References / 305

to

Appendix: List of Romanian political parties & map of the regions / 309 Romanian political parties / 309 The Map of Romania and the location of the main regions referred in the text / 310

9

Introduction. Romania and the comparative study of values1 BOGDAN VOICU, MĂLINA VOICU

This volume discuses the situation of the Romanian society after more than fifteen years of major social changes. We are interested in social values, as determinants and products of the economic and social organization and also constitutive elements of the social development mix. We place our discussion in a comparative perspective, often making comparisons to other European countries. Also, we are interested in the dynamics and the way values have changed in this entire post communist period. Before starting to discuss about Romania and Romanians’ values, there are some issues that need clarification: what values are, what they are good for, what is the importance of studying these values, who and how studies values in the contemporary sociology. These are the topics that we address in the current introduction. The order in which we have listed them gives the order of the following sections. In the end of this introductory chapter, we discuss the structure of the volume and the data we use.

Defining values Talking about values was and will always be a very difficult task because of the different connotations that are given to the term in various scientific domains, connotations further propagated into common language. The most frequent meaning is the one coming from the humanities. In esthetics, philosophy and literature, the term tends to have a normative meaning. Values become criteria that people and collectivities use in order to make the

1

We thank Dumitru Sandu for reading very attentively an earlier version of this chapter and making very useful comments. 11

The Values of Romanians: 1993-2006. A sociological perspective

distinction between good and evil, between beautiful and ugly or between desirable and undesirable (Ester and others, 1994). After the first half of the 20th century, a still vivid debate in the social sciences reflected the tendency of western countries to renounce normativism in favor of accepting diversity. Without acquiring the consensus in what concerns the way they manifest themselves, in our days, sociology, psychology, anthropology place values at the level of the individual, but still as a main factor in the nature of our society. They are not those clear guidelines that establish an absolute good or an absolute beauty anymore. For psychologists, values represent anchors that allow individuals to become oriented in the surrounding world, acting as decoding systems of the bonuses of the potential directions of action, benefits deriving from everyone’s scale of preference, from personal aspirations or needs2. In turn, sociology defines values as inner to the individual, but with an important social determinant, as expressed and at the same time induced by norms, habits or ideologies3. In the field of sociology as in the one of psychology too, values act like an engine in everyday choices, helping in tiding up the individual’s priorities and personal life. At individual level, they are consistently structured in value systems, (Rockeach, 1973), values from a specific domain (for instance, religion, family, social relations, work etc.) being determined and also determining the values of all the other domains. Kluckhohn (1951:411) talks not only about values, but also about values orientations, more generally speaking, acting as organized and generalized conceptions, influencing behavior, regarding nature, the human place and role, the relations with other people, what is desirable or undesirable, no matter if these last ones are related to environment or to inter-human relations. Values cannot be directly observed. No one can precisely describe which the values of somebody else are. They behave as latent realities, inner to individuals, but with an important social determinant. Human collectivities develop common sets of values that insure social cohesion and the possibility of living together in the same environment, but also allow value orientations specific of some subgroups and even of individuals who nevertheless adhere to common values. The inclusion of values in values systems and the pronounced social determinacy lead to their stability. Values do not change from day to day; they need a long time to restructure, every change practically affecting all the other spheres of one’s life. This leads to the fact that, in studying value dynamics, long time intervals are needed (5-10 years) so that the possible fluctuations become visible.

2 3

See van Deth and Scarbrough (1994:22). Deth and Scarbrough (1994:22), Parsons (1964), Voicu and Voicu (2002) and so on.

12 corectura 2 28 iulie 2008

Introduction. Romania and the comparative study of values

Uses of the study of values During the 20th century, several paradigms dominated the discourse about change and development4. The oldest one asserts the determinant role of the economy and of the production relations for all the other components of the social system. Progress is mainly dictated by the changes in the economic system, by the political organization, relations and social values with dynamics dictated by the dynamics of economy. Max Weber’s works would emphasize another factor, namely culture, with the modern rationality as the spark of progress. Finally, the paradigm of factor complementariness is the one that becomes salient in contemporary literature. Built on older or newer papers, from Parsons to Inglehart and Giddens, such an approach discusses the complementarily of the dynamics of economy, culture and political organization, human capital and so on in generating development. The current volume deals mainly with culture. Like values, the term is poly-semantic, being used mainly with three meanings that overlap5 to a large extent. Human sciences had the chronological anteriority. Philology and philosophy defined through culture only the noblest ideas and manifestations of human beings, mostly the artistic ones, which would lead to the highest sublimation of life and beauty. Herder would start a true revolution in knowledge, noticing that anyone, any group has a culture. In fact, it was a type of denormativization. Beauty was not absolute beauty anymore, being relativized in a way that was meant to rule the intellectual debate of the 20th century, becoming dominant at its end. In this context, culture was redefined as a way of living and thinking. Contemporary sociology uses culture with this meaning, mainly discussing rules and institutions as ways in which values and values orientations manifest6, as reflection of the patterns of thinking specific to each community. Anthropology and history add some perceptible and visible material elements, products of a non-material culture, understood in the two denotations already mentioned. These elements of culture are the rituals and the objects produced by human communities or societies. Material culture plays the role of symbols in which ideas, values, rules organizing the non-material culture, mirror. Values and also culture, in their sociological meaning, are constitutive elements of the social development mix. Their study facilitates the knowledge of the evolution of society, of its dynamics, of the potential current needs that the human collectivities might manifest. Some examples allow us to better describe this contribution.

4

See Voicu, 2001. See Berger, 1995. 6 Hofstede (2001), Schwartz (2003, 1994a), Inglehart (1997) and others explicitly define values as the central element of culture. 5

13

The Values of Romanians: 1993-2006. A sociological perspective Figure 1. The Cultural Map of the World, according to Inglehart typology

Data source: EVS/WVS 1999-2001. The positioning of every country on the two axes must be interpreted in a relative way (for example: Sweden is more modern than the USA, but not necessarily completely modernized). The two displayed dimensions are the full factor scores described by Inglehart (1997), not their reduced version used by Inglehart-Welzel (2005).

We will start with the cultural map of the world (Inglehart, Welzel, 2005). The literature discussing about social change has been dominated for many decades by the analysis of two major processes: modernization and postmodernization. Modernization can be schematically reduced to the preference for rationalization and planning to the detriment of tradition as a generation element of the majority of everyday decisions. Likewise, the reductionist perspective of cultural postmodernization can be represented by a striking orientation for self expression, for the satisfaction of superior needs. The type of analysis proposed by Inglehart (1993, 1997) presumes taking into account the two dimensions built as aggregates7 of the values in the following domains: work, family, relations between sexes, social relations, politics, environment protection, religion and so on. Based on data provided by the 7

Factor scores.

14 corectura 2 28 iulie 2008

Introduction. Romania and the comparative study of values

European Values Survey and World Values Survey (see the following section), such an approach allows structuring a cultural map of the world (figure 1). This facilitates the comparative analysis of the contemporary human societies. For example, Romania’s position is somewhere at the periphery of the excommunist countries group, being more traditional and less oriented towards superior needs. Besides, the differences between the East and the West and between the North and the South can be easily recognized on the map. The American exceptionalism is also visible: The US, where modernization developed differently from Europe in what concerns, for example, secularization (Lipset, 1996; Baker, 2005; Inglehart and Welzel, 2005), display a much more traditional society than Western European countries and even some ex-communist countries. Except for this outlier, the cultural map of the world roughly overlaps the map of development, no matter if the development is measured through the human development index or through its components, such as GDP/capita, life expectancy, or level of education. A second example is connected with the evolution of the ideas of the political parties’ supporters. For a long time, the left-right distinction has been a common place in social sciences and also in common language. Placement on this continuum defines, in a reductionist way, the parties’ and electors’ opinion on most political problems. Figure 2 considers the sympathizers of the main Romanian political parties and shows the dynamics of their self-positioning on the left-right scale between 1996 and 20038. Several major tendencies can be noticed: at the beginning, on average, the groups of voters of all the major parties seem to go toward the center. In the second half of the 90s, the crossparty differentiations become more and more visible. The voters of PSD (former FDSN, PDSR and so on) represent the group that heads towards the left and strengthen its position in this area. At the opposite, PNL voters gravitate towards the right, occupying alone this area, after PNŢ-CD stops attracting supporters. PRM voters oscillate around the middle value on the scale (5.5), reflecting the concern of the sympathized party for combining nationalist, xenophobic discourse, specific to the extreme right with the communist sort of demagogy, specific to the extreme left.

8

For details related to the main Romanian parties, please check the appendix at the end of the current volume. This book also includes an excellent chapter written by Mircea Comşa which focuses on the self-positioning on the left-right scale. In this introduction we do not go into so much detail like the mentioned analysis, but we offer an example about the way values structure the domain of political choices. 15

The Values of Romanians: 1993-2006. A sociological perspective Figure 2. The evolution of self identification with the political left or right of the main Romanian political parties’ electors between 1993 and 2006

PRM/ PUNR FDSN

July 1993 1

2

3

4

5

Nov. 1997 PDSR 1

2

3

4

1

2

3

4

1

FSN

5

UDMR

2

3

4

2

3

4

May 2007 1

2

3

LEFT

4

5

8

9

10

8

9

10

8

9

10

7

8

9

10

7

8

9

10

PNL

PRM/ PUNR

7

PNŢ-CD

PNL

UDMR 6

PD

UDMR

PRM PNG

PRM

10

7

PNŢ-CD

6

5

PSD

9

PNL

Oct. 2006 1

8

PD

5

PSD

7

UDMR

6

PRM

PSD

PNŢ-CD

PD

UDMR PDSR

Nov. 2005

6

PRM/ PUNR 5

Nov. 1999

PNL

6

PNG

6

PNL

PD

UDMR PD

7

PNL

RIGHT

Sources: EVS&WVS Romania, the waves in July, 1993, November, 1999, November, 2005, respectively the survey Values ’97 realized by the University of Bucharest (coordinator Dumitru Sandu) in November, 1997 and the November 2006 wave of the Public Opinion Barometer of the Foundation for an Open Society (BOP-FSD). The scores represented graphically are the averages for every party’s electors, of the answers given to the question: Related to politics, people talk about “left” and “right”. Generally speaking, where would you place yourself on the below scale? 1 – “left”… 10 – “right”. The electors are the ones who claimed that they would vote with the respective party in case of organizing elections right in the following weekend. Among PNL electors, we counted all the liberal parties’ electors (PNL–AT, PL, PL’93, NPL, PAC, PNL–CD, PNL–C, UFD and so on depending on the reference year). For 1993, 1997 and 1999, PRM and PUNR electors were considered part of the same group.

16 corectura 2 28 iulie 2008

Introduction. Romania and the comparative study of values

The three groups represent the constants of these dynamics. At the opposite, UDMR and PD voters are the ones that change the way they define themselves depending on the context. UDMR electors head towards the left or right in their self identifications, anticipating the changes of the alliances made by the governing party that holds the office all the way since 1996. The cooperation with PNL and PNT-CD is marked by positioning rather to the right. However, this modifies during the second part of the 90s. Electors head towards the left, their position in 1999 announcing the government alliance’s potential after 2000 (PSD-UDMR). The presence in the government after 2004, together with the PNL and PD also marks a new movement to the right. Figure 3. The dynamics of values orientations towards the maximal or minimal state for the supporters of several Romanian political parties between 1993 and 2006. should take more The 10 state responsibility to ensure that everyone is provided for 9

8

7

6

PRM

PSD

PSD PRM UDMR

5

PD UDMR

4

PD

PNL

PNL

3

1993

1997

1999

2005

2006

2

Individuals should take more responsibility for providing for 1 themselves

Sources: EVS&WVS Romania, the waves in July, 1993, November, 1999, November, 2005, respectively the survey Values ’97 realized by the University of Bucharest (coordinator Dumitru Sandu) in November, 1997 and the November 2006 wave of the Public Opinion Barometer of the Foundation for an Open Society (BOP-FSD). The scores represented graphically are averages, every party’s electors, of the answers given to the question: Which of the two remarks written on the vertical axis is more representative for the one who answers? The answers can be granted a number from 1 to 10. The electors are the ones who claimed that they would vote with the respective party in case of organizing elections right in the following weekend. Among PNL electors, we took into consideration all the liberal parties’ electors (PNL – AT, PL, PL’93, NPL, PAC, PNL – CD, PNL – C, UFD and so on, depending on the reference year). For 1993, 1997 and 1999, PRM and PUNR electors were considered part of the same group. 17

The Values of Romanians: 1993-2006. A sociological perspective

In the case of the PD, the party’s identity oscillations are accompanied by similar oscillations in the way electors define themselves. Initially placed rather to the left side of the political spectrum, they advance gradually to the right, with a recoil lately manifested and probably given by the separation attempt from the PNL and also by the tendency of the sympathizers’ rising number and – implicitly – of their diversification. Finally, but not accidentally, PNG voters have an average of self identification with the left or the right that does not significantly differ from the one of the PRM electors. PSD electors assume the extreme left position, with a more accentuated involvement of the state in the social life than in the case of other parties. At the counter pole, there is the PNL, whose electors assume rather a liberal position, allowing the individual to be more responsible for his/her own wellbeing. In the early ‘90s, there was a general tendency of decreasing the orientation towards the maximal state. An increasing number of people claimed that individuals should be more responsible for their own welfare, while the state should reduce the scope of its intervention. However, such a position was threatened at its turn in the following years, and the change modified again its direction by inverting it. From this point of view Romania could be characterized as being in the process of modernization9. On the other hand, orientation changes in case of the voters of the parties are also accompanied by changes of the socio-demographic structure of these electorates. This is the case of the PD and PRM, parties for which the value differences are the result of the actual change in the voters’ body.

Contemporary approaches to values and culture Currently, the comparative analysis of the social values in different societies is highly influenced by a few major streams. In the following, we do not intend to be exhaustive, but to present the basic principles underlying the analysis of some of the groups active in the study of values, in a world in which this kind of enterprise suffered a real boom10. The most important stream is the one clustered around the European Values Survey and World Values Survey (EVS/WVS). The EVS program was initiated in the late 70s with the main intention of investigating the religious 9

Traditionally, the state has a reduced role in providing welfare, because people, families and communities take care of it. Modern state, in its European version, tended to become “a wellbeing state”, with slight variations resulting from its liberal, sociodemocratical, conservatory or Mediteranean character, where the state assumed a very important role in social services promotion and redistribution. 10 For a relevant discussion about the increasing number of value studies and on the increasing impact of studying values in various fields, see Ester, Braun, Mohler, (2006). 18 corectura 2 28 iulie 2008

Introduction. Romania and the comparative study of values

values in Euro-Atlantic area. In this process, the opinions of some representative samples for the countries in Western Europe, the USA and Canada11 were investigated through the first wave of the survey (1981). Other waves of the values surveys were realized in 1990-1993, 1995-1997, 19992002, 2005-2007, including more and more societies. Notably for this book, Romanian national representative samples were investigated in each wave starting with the 1990-1993 one. Rather personal than methodological motivations led to the settlement of two different teams, one of EVS and the other of WVS. The EVS rule consists of repeating the surveys once every 9 years and keeping part of the questionnaire unchanged for allowing the longitudinal analysis. The WVS differs from the EVS through the fact that it proceeds data collecting every 5 years. The main part of the questionnaire is common and the waves 1990-1993, respectively 1999-2002 are almost identical. Until now, the 1999-2002 wave is the most comprehensive, covering 81 societies and including samples which stand 95 per cent of the Earth’s population. The value surveys, as the EVS and WVS are known, use the same type of starting assumptions, originating in modernization and postmodernization theories. Individual level variations in the spheres of family, religion, work, politics, social relations, environment protection, equality, redistribution and so on are usually interpreted through variations resulting from the modernization process, or late modernization process, anticipating postmodernism. The empirical defining of the two value axes usually follows the analysis proposed by Inglehart (1990, 1997). Inglehart (1991, 1990, 1997) builds his conceptual explanation starting with two main hypotheses: the scarcity hypothesis argues that individuals’ priorities are determined by the socio-economical environment in which they develop and that individual scales of preferences tend to give priority to the things that are rare. These ones can satisfy the basic needs (in poor societies) or superior needs (in communities defined by abundance). The connection to Maslow (1954) and his hierarchical pyramid of needs is obvious. The second hypothesis, the socialization one, claims that individuals’ values and the way they organize their preferences depend on the socio-economical conditions in their primary socialization stage. The individuals who experienced an abundance of goods in their pre-grown up stage, will tend to express rather superior needs, their values orienting them towards satisfying such needs. Hence, there is a tendency to express superior, postmaterialistic needs when economic and scientific progress occur and makes basic needs meet. The paired relation is valid, too: modern and postmodern values adoption emancipates individuals, eliminating some of the constraints that institutionally or morally 11

For detailed description of EVS/WVS, see Halman (2001), Inglehart and others (2004), Voicu and Voicu (2002), Arts and others (2003) and so on. 19

The Values of Romanians: 1993-2006. A sociological perspective

lock liberty. Modernity involves the secularization of authority, while postmodernization leads to authority emancipation (Inglehart, Welzel, 2005: 2526). Creativity, innovation capacity, inventiveness are released and this leads to technical progress and increases of productivity. Inglehart’s empirical analyses of social values bases on factor analyses using the EVS/WVS data sets. The cultural space is empirically described as having two dimensions. The first axis opposes traditionalism to a secularrational way of relating to the world. The second one, opposes survival values, the ones of the orientation towards satisfying basic needs and insuring order, and self-expressing, self achievement values, orientations to superior needs. Hagenaars and his colleagues from the group located at Tilburg University propose a similar approach (Hagenaars et al., 2003), but they label the two resulting dimensions as normative-religious, and respectively, as autonomy social-liberalism. These two factors are analogous to the Inglehart’s ones. EVS and WVS continue to be the most famous and cited international comparative research projects. An internet search using the key words “European Values Survey” provides 14.900 more links. Searching for “World Values Survey” adds 126.000 other links12. At the same time, other comparative research projects hit lower scores: ”European Social Survey” – 86.300; International Social Survey Program – 26.800. The second most important group in the study of values is the one developed around the works of the social psychologist Shalom Schwarz. Conceptually, he talks about values and culture, looking for those values universally acknowledged. Following Rokeach (1973), his initial works (Schwartz, 1994) discuss 10 basic values, structured on two axes: openness to change versus conservatorism and egocentrism (hedonism, strength, self achievement) versus universalism, benevolence, (transcendental) interest towards the others13. Later on, Schwartz (2004) re-groups again the 10 basic values in seven important dimensions defined as higher-order or cultural-level values: egalitarianism (equality and social justice), intellectual autonomy (curiosity and broadmindedness), affective autonomy (hedonism, search of quality of life), mastery (controlling the social and natural environment; involves ambition and daring), hierarchy preference (and authority), embeddedness (strict social order, obedience and respect for traditions), harmony (world peace, communion with nature and so on). Revising these seven dimensions, the polarities induced by modernization and postmodernization processes can be observed. In fact, Schwartz identifies three polar axes: autonomy versus society dependency, egalitarianism versus preference for hierarchy and domination versus harmony.

12 13

The internet search was realized on the 12th of January, 2006, using Google. Self-enhancement versus self-transcending.

20 corectura 2 28 iulie 2008

Introduction. Romania and the comparative study of values

The empirical work on various convenience samples, including mainly students and teachers, from different cultures and nations, allowed Schartz and his followers to validate the proposed model as universal. The lack of empirical validation on larger samples made Schwartz theory difficult to test and develop. This was also complicated by the oscillations in defining the dimensions of the cultural space. Recently, the 2002 wave of the ESS and the 2005-2007 wave of the WVS included a set of items permitting social values measuring and testing Schwartz’s theory on larger samples (Ramos, 2006; Petterson, 2006). These ones, like the presence of the scale in a survey realized on a representative sample in Spain (2004), permitted some analyses14 which discuss the possible convergence or complementariness between the two dimensions 15 proposed by Inglehart (1990, 1997) and the two (three) dimensions proposed by Schwartz. Unfortunately, there was very little communication between the two groups of cross-cultural study of values. (Vinken and others, 2004). Inglehart’s analyses are realized especially at the level of the entire world samples of different WVS waves. With few exceptions, his theory was not subject to empirical testing in the case of specific individual societies. In exchange, Spini (2003) realizes a confirmatory analysis on data coming from Schwartz’s samples of students and teaching staff and he verifies the structural consistency of the explanatory model for almost each of the considered societies. Another influential group of theories is the one initiated by Hofstede (1980, 2001). Hofstede’s theory about values is derived rather empirically and is developed through the systematical study of IBM employees in numerous societies16. For Hofstede, culture is universally defined through five value orientations. The first one is the power distance, the inequality defined bottomup, referring to the individuals’ in an inferior position acceptance of inequality as a matter of course. The second dimension is given by individualism or autonomy, as opposed to collectivism or integration in extended, usually closed, social networks. Masculinity refers to the principle in accordance to which men are more likely than women to be in decision-making positions. Uncertainty avoidance, the forth value dimension resides in the orientation towards usual things and the avoidance of what is new. Finally, long-term time orientation is defined through thrift, planning, perseverance, as opposed to the prevalence of respect for tradition, social duties accomplishment, “protecting the social image”. Hofstede’s studies, realized mainly inside a multi-national corporation, are important particularly because they show that, in globalizing environment conditions, the values of the employees depend on the culture of the collectivity 14

Petterson, 2004, 2006. Traditional – secular and survival values – self expression values. 16 Hofstede (2001) relates about 70 investigated societies. 15

21

The Values of Romanians: 1993-2006. A sociological perspective

where they live. For Hofstede, this dependency is not a local one, but it represents an evidence for the influence of the “national” culture over the value orientations of the individuals. All these above-described approaches try to reduce the complex space of values, trying to theoretically and/or empirically build a reduced number of dimensions, more easily to analyze and interpret. The difficulties that such attempt encounters are, undoubtedly, remarkable, especially in what concerns values measurement, whose latent character is one of the few points of methodological consensus (Jagodinski, 2004). Apart from “dimensionalists”, there are some other groups approaching the problem of values cross-culturally. Such a group is the one which studies “cultural traumas” as sources of social change, marking some communities’ cultural identity for long-term periods, if not even permanently (Smelser & Alexander, eds., 1999; Sztompka, 2000; Alexander et al., 2004). Another group is the one built around the discussion about diversity of modernity, different from a culture to another (see Sachsenmaier, Riedel, Eisenstadt, eds., 2002). These streams focus for the moment on theoretical construction of their explanations, generally avoiding to empirically validate their assumptions at the individual level. In social sciences literature written in Romania, all these streams from the study of social values have had rather a minor impact so far. Early ‘90s did not bring almost any global systematic approach of cultural modernity or values dynamics. During the last years, a few writings on this topic have been published, some of them (Roth, 2002; Chiribucă, 2004; Vlăsceanu, 2001) taking over and presenting modernization and postmodernization theories, others (Sandu, 1996; Voicu, 1999, 2001, 2005; Vlăsceanu, 2007) drawing out personal syntheses and more elaborate explanatory frameworks. They had already been anticipated by punctual empirical analyses, promoted mainly by Dumitru Sandu (1999, 2003) and also by the team working on the study of values in Romania (see the appendix). It is interesting to remark that all these analyses use, as a theoretical basis, only the modernization and postmodernization theories17.

The structure of the current volume The present volume does not intend to build a theory of values or to develop one of the existing theories. What we intend to do is to use the knowledge we already have through the different currents that mark the contemporary study of values and to describe a part of contemporary Romania, 17

The only Romanian article that we know, presenting some other types of dimensional approaches belongs to us (Frunză, Voicu, 1996). In the respective paper, we mention Schwartz’s and Triandis’ approaches rather vaguely. Also, Voicu (2005) uses the arguments in the theory of cultural trauma, as they appear at Sztompka. 22 corectura 2 28 iulie 2008

Introduction. Romania and the comparative study of values

the way it is today and the way it developed through the communist period. We are interested in changes at the level of social values in the spheres of family, voting behavior, religion or attitude towards democracy and so on. Each of the materials included in the volume is conceived as an analysis focusing on the case of Romania. However, the general principle that guides our approach is to refer, whenever it is necessary, a larger context, namely the European framework. We do not discuss the impact of values in all its possible facets, in all imaginable domains. It would be practically impossible to achieve such a goal. What we do is to characterize the dynamics of the social values in Romania between 1993 and 2006 in few domains connected to our research interests. A first grouping includes three articles dedicated to value orientations towards the element of social organization of the public space. Claudiu Tufiş opens the volume with a study dedicated to the orientations towards democracy. Mircea Comşa discusses in detail an element of the political behavior in a study that provides – for the first time in Romania – an empirical and conceptual material dedicated to electors’ self positioning on the left-right scale. Then, Claudiu Tufiş devotes a chapter to confidence in institutions. A second set of articles begins with Mălina Voicu’s analysis over religiosity in Romania, seen from a comparative perspective. This marks the passage to value orientations which manifest particularly in sphere of private relations. Raluca Popescu discusses the social values inside the family, also referring briefly to gender relations problems. Paula Tufiş brings to foreground parents’ value orientations in what concerns children’s socialization, another domain practically unexplored up to the moment in Romanian literature. At the end of the volume, Horaţiu Rusu tries to describe the way in which our days teen-agers differ from teen-ager generations in 1999 and in the early ’90s. A more global perspective over values evolution closes the volume. Bogdan Voicu comes up with an aggregate analysis over Romania’s position and dynamics, from a cultural modernity point of view, in comparison to the rest of the European countries. In our studies we use several data sets, paying the most attention to the EVS/WVS Romanian waves. The first data are from 1993, because, in the 1990 wave of the EVS/WVS, Romania collected data two years later than the rest of the Europeans. The survey was conducted by the the Research Institute for the Quality of Life, under Ioan Mărginean’s coordination, while Marian Preda assured the coordination of the network of field operators. In the 1995-1997 wave of the WVS, because of some communication deficiencies, Romania is practically absent. A mixed team from the University in Bucharest and the Research Institute for Quality of Life under Dumitru Sandu’s18 coordination received a CNCSIS (National University Research 18

Mălina Voicu was also part of the team. She was the coordinator of the data collecting process and she also participated in designing the questionnaire together with 23

The Values of Romanians: 1993-2006. A sociological perspective

Council) grant, collecting data in November 1997 according to a very different questionnaire from the WVS questionnaire; as a consequence, the data base has never been integrated with the others. In 1998, the Independent Center for Social Studies, coordinated by Pavel Câmpeanu, also collected some data according to the WVS questionnaire. However, since the quality of the respective data is very difficult to control (Sandu, 2003:42), in the current volume we preferred not to use that wave of the WVS. In November 1999, a team from the Research Institute for the Quality of Life, coordinated by Cătălin Zamfir, Mălina Voicu and Lucian Pop19 led the data collecting, and the subsequent analysis of the Romanian version of the 1999-2002 wave of the EVS. A failed attempt of put together a volume in which all the results would be presented, eventually materialized through a series of articles that appeared in the 2001 and 2002 issues of the Romanian Sociology journal. The Romanian team for the study of values, developed mainly inside the Research Institute for Quality of Life, acquires consistency and a sort of stability during 1999-2005. The group’s website20 certifies the members’ activity. Since 2005, Mălina Voicu has been a member of the Executive Board of the EVS. In 2005, data for WVS 2005-2007 are collected, with Bogdan Voicu and Mălina Voicu as Romania’s principal investigators. The Research Institute for the Quality of Life (RIQL) assured an institutional support, but the data collecting process was realized with the institutional, logistic and financial aid coming of the Open Society Foundation (OSF) through the program provided by the Public Opinion Barometer (BOP), coordinated by Ovidiu Voicu and assisted by Mihaela Ştefănescu. The Romanian team, including the authors of the current volume, would discuss and finalize the questionnaire variant applied together with BOP members and OSF representatives21. Thus, it resulted in the inclusion of the Romanian variant of the WVS 2005-2007 in BOP-FSD wave from November, 2005, in the first data collecting of this type realized without external financing. The OSF contribution was determinant from a financial point of view, but particularly due to the members of the board of the BOP-FSD and of the coordinator of the program, Mr. Ovidiu Voicu, all of them being active in the critical analysis and the polishing of the Romanian version of the

Dumitru Sandu, Lucian Pop and Mihai Surdu. Among the authors of the current volume, Paula Tufiş and Bogdan Voicu collaborated during the different stages of the data collecting process. 19 Bogdan Voicu, Paula Tufiş, Claudiu Tufiş, Cosmina Rughiniş, Ruxandra Noica, Elena Gheorghiu and Monica Şerban were also part of the team. 20 http://www.iccv.ro/romana/teme/EVS/evs.htm. 21 Dumitru Sandu, Gabriel Bădescu, Manuela Stănculescu, Cosima Rughiniş, Ovidiu Voicu, Mihaela Ştefănescu. 24 corectura 2 28 iulie 2008

Introduction. Romania and the comparative study of values

WVS 2005 questionnaire. The Gallup Organization Romania was the data collecting agency for the respective 2005 wave. A new wave of European Values Survey is scheduled to take place in 2008. The team, hosted by RIQL and led by Mălina Voicu, includes all the authors of this volume, to which adds Marian Vasile. Data collection is scheduled for April-May 2008, and the results will start to be presented to the public in the autumn of 2008. The series of the data sets resulting from the value surveys allows the analysis of the dynamics of the Romanian values. It also provides an empirical support for the comparison to the rest of the European countries22. This is the main data source for the papers collected in this book. A few other sources complete the empirical material of the current volume: the complete series of the Public Opinion Barometer of the Foundation for an Open Society (BOP-OSF), data bases of the Research Institute for the Quality of Life, etc. The present volume does more than just exploiting these numerous empirical resources. At the beginning, each one of the chapters tries to make an incursion in the scientific information already existing in that field of expertise. The main concepts and explanatory theories structuring the perspective that we adopted are presented. At least from two points of view – ideological self positioning and values of children’s socializing – the theoretical information touched upon in the volume is new in the Romanian literature. In their turn, the rest of the chapters offer interesting syntheses too for the analyzed domains. The conceptual framework that we provided serves to the elaboration of some hypotheses that are tested empirically through various statistic procedures. In the entire volume, we tried to avoid technical language and to offer intuitive explanations to the readers who are not familiar with the different statistic procedures that we used and, also, to maintain the text at a reasonable standard. Most of the commentaries referring to analysis procedure that we used are usually included in footnotes and they appear properly in the body text only in the short sections, where the discussion is rather methodological. Even on these occasions, technical explanations are rather succinct. Still, the authors are very open to offer details and any comment is welcome. There are several colleagues that supported us in our current project. To a large extent, the volume is due to the contribution of all those who helped at data collecting in the previous waves of values surveys, either the ones we already mentioned – coordinators, members of the research team, or field 22

Waves 1990-1993 and 1999-2001 are more useful in this respect because they cover the majority of the European Countries. The wave in 2005-2007 had not finished the data collecting process when we were writing this article and that is why data were available only for Romania, Italy, Poland, France, Slovenia, Finland, Sweden, Great Britain, Germany, Holland, Andorra and Russia. 25

The Values of Romanians: 1993-2006. A sociological perspective

operators. We thank them all. Special regards are directed to the people involved in the program for the Public Opinion Barometer of the Foundation for an Open Society, the one that contributed to the financing of the WVS 2005 data collecting. The Research Institute for the Quality of Life was the proper environment for developing a group to study social values and the help coming from its management board (Cătălin Zamfir and Ioan Mărginean) was very important from this point of view. Finally, two CNCSIS grants (AT-102/2006 and ID-56/2007), coordinated by Bogdan Voicu and including all the authors of the current volume as team members, supported writing, translating and publishing the volume. This English edition of the book reproduces the Romanian version (published in 2007). The initial English translation was done by the authors for chapters 1, 3, 4 and 7. Irina Nicula initially translated the rest of the book, with the authors substantially reviewing the translation. Dean Hufstetler further reviewed the entire book, contributed to this version with the final language proofing, and with several valuable suggetsions. As compared to the Romanian edition, we have added an appendix including a map of Romania, with its main regions, as well as a description of the Romanian political parties. Both the regions and the political parties are often referred to in the book, and the short presentation from the Appendix may be useful for the non-Romanian readers.

References Alexander, Jeffrey C., Ron Eyerman, Bernhard Giesen, Neil J. Smelser, Piotr Sztompka. 2004. Cultural Trauma and Collective Identity, Berkeley, Los Angeles, London: University of California Press. Arts, Wil, Jacques Hagenaars, Loek Halman, editors. 2003. The Cultural Diversity of European Unity. Findings, Explanations and Reflections from the European Values Study, Leiden: Brill. Baker, Wayne E.. 2005. America's crisis of values: reality and perception. Princeton, N.J., Princeton University Press. Berger, Bennet M. 1995. An Essay on Culture: Symbolic Structure and Social Structure, Berkeley: University of California Press Chiribucă, Dan. 2004. Tranziţia postcomunistă şi reconstrucţia modernităţii în România, Dacia, Eikon, Cluj-Napoca. Deth, Jan van, Elinor Scarbrough. 1994. The Concept of Value în Jan van Deth, Elinor Scarbrough, eds., The Impact of Values (Beliefs in Government, vol. 4), Oxford University Press. Ester, Peter, Loek Halman, Ruud de Moor. 1994. The Individualizing Society. Value Change in Europe and North America, Tilburg University Press.

26 corectura 2 28 iulie 2008

Introduction. Romania and the comparative study of values Ester, Peter, Michael Braun,, Peter Mohler, eds. 2006. Globalization, Value Change, and Generations. A Cross-National and Intergenerational Perspective, Leiden, Boston: Brill. Frunză, Mălina, Bogdan Voicu. 1997. Statul şi ţăranul român. Un studiu de caz, Revista de Cercetări Sociale, no. 3/1997: 118-130. Hagenaars, Jacques, Loek Halman, Guy Moors. 2003. Exploring Europe’s basic values map, in Arts and others, eds., p. 23-66. Halman, Loek. 2001. The European Values Study: A Third Wave. Source book of the 1999/2000 European Values Study Surveys, EVS, WORC, Tilburg University, 2001. Hofstede, Geert. 1980. Culture's Consequences, International Differences in WorkRelated Values, Beverly Hills: Sage. Hofstede, Geert. 2001. Culture's Consequences, Comparing Values, Behaviors, Institutions, and Organizations Across Nations, Thousand Oaks: Sage Publications. Inglehart, Ronald, 1990. Culture Shift in Advanced Industrial Societies, Princeton University Press. Inglehart, Ronald, 1997. Modernization and Post-Modernization. Cultural, Economic and Political Change in 43 Societies, Princeton University Press. Inglehart, Ronald, Christian Welzel. 2005. Cultural Change and Democracy: The Human Development Sequence, New York and Cambridge: Cambridge University Press. Inglehart, Ronald, Miguel Basañez, Jaime Diez-Medrano, Loek Halman, Ruud Luijkx, eds. 2004. Human Beliefs and Values, a cross - cultural sourcebook based on the 1999 - 2002 values surveys, Siglo XXI Editores. Inglehart, Ronald. 1971. The Silent Revolution in Europe: Intergenerational Change in Post-Industrial Societies, American Political Science Review, 65 (4), p. 991– 1017. Jagodzinski, Wolfgang. 2004. Methodological problems of value research, în Vinken ş.a., eds.: p. 97-121. Kluckhohn, Clyde. 1951. Values and Value Orientations in the Theory of Action, în Parsons, Talcott, Edward Shils (coord.) – Toward a General theory of Action, Harper & Row Publishers, New York, 1951. Lipset, Seymour Martin. 1996. American exceptionalism: a double-edged sword. New York, W.W. Norton. Parsons, Talcott. 1964. Evolutionary Universals in Society, American Sociological Review, Vol. 29, No. 3 (Jun., 1964), pp. 339-357. Petterson, Thorleif. 2004. A third way in values research? A comparative analysis of the approaches of Ronald Inglehart and Shalom Schwartz, World Values Survey Conference Budapest, Hungary, September 2nd-5th, 2004 Petterson, Thorleif. 2006. Basic Value Orientations towards the MDGs and Global Development, paper presented at World Values Conference Society, Politics and Values: 1981-2006, Istanbul, 3 -4 November, 2006 Ramos, Alice. 2006. Social values dynamics and socio-economic development, Portuguese Journal of Social Science, Volume 5, Number 1: 35-64. Rokeach, Milton. 1973. The Nature of Human Values, New York: The Free Press, London: Collier Macmillan Publishers. Roth, Andrei. 2002. Modernitate şi modernizare socială, Polirom, Iaşi. 27

The Values of Romanians: 1993-2006. A sociological perspective Sandu, Dumitru. 1996. Sociologia tranziţiei. Valori şi tipuri sociale în România, Bucharest, Staff Publishing House. Sandu, Dumitru. 1999. Spaţiul social al tranziţiei, Polirom Publishing House, Iaşi. Sandu, Dumitru. 2003. Sociabilitatea în spaţiul dezvoltării. Încredere, toleranţă şi reţele sociale, Polirom Publishing House, Iaşi. Schwartz, Shalom. 1994. Are there Universal Aspects în the Content and Structure of Values?, Journal of Social Issues, 50, p. 19-45. Schwartz, Shalom. 2004. Mapping and Interpreting Cultural Differences around the World, in Vinken and others eds., 2004, p. 43-73. Smelser, Neil J., Jeffrey C. Alexander, eds. 1999. Diversity and Its Dscontents: Cultural Conflict and Common Ground in Contemporary America, Princeton: Princeton University Press. Spini, Dario. 2003. Measurement equivalence of value types across twenty-one countries. Journal of Cross-Cultural Psychology, 33, 3-23. Sztompka, Piotr, 2000. Cultural Trauma. The Other Face of Social Change, European Journal of Social Theory, 3(4): 449-466. Vinken, Henk, Soeters, Joseph, Peter Ester, eds. 2004. Comparing Cultures: Dimensions of Culture in a Comparative Perspective, Leiden: Brill. Vinken, Henk, Soeters, Joseph, Peter Ester, eds. 2004. Cultures and dimensions. Classic perspecives and new opportunities in ‘dimensionalist’ cross-cultural studies, in Vinken and others., eds., 2004, p. 3-27. Vlăsceanu, Lazăr, 2001. Politică şi dezvoltare. România încotro?, Trei Publishing House, Bucharest. Vlăsceanu, Lazăr. 2007. Sociologie şi modernitate. Tranziţii spre modernitatea reflexivă, Polirom Publishing House, Iaşi. Voicu, Bogdan, 1999. Modernitatea între tradiţie şi postmodernism, Revista de Cercetări Sociale, no. 3-4/1999, 36-59. Voicu, Bogdan, 2001. România pseudo-modernă, Sociologie Românească, 1-4/2001: 36-59. Voicu, Bogdan. 2001. Dezvoltare socială, în Luana-Miruna Pop, coord., Dicţionar de politici sociale, Bucharest, Expert Publishing House, p. 258-275. Voicu, Bogdan. 2005. Penuria Pseudo-Modernă a Postcomunismului Românesc. Volumul I. Schimbarea socială şi acţiunile indivizilor, Iaşi: Expert Projects. Voicu, Mălina, Bogdan Voicu. 1999. Programe sociale ale partidelor politice româneşti, în Cătălin Zamfir (coordonator), Politici sociale în România: 1990-1998, Expert Publishing House, Bucharest, 1999, p. 583-691. Voicu, Mălina, Bogdan Voicu. 2002. Studiul valorilor europene: un proiect de cercetare internaţională, Calitatea Vieţii no. 1-4/2002.

Appendix. A selection of products of the group of studying the social values in Romania Mălina Voicu (1999) - „Între cratiţă şi parlament – legitimitatea politicilor de suport pentru femei în România”, in Revista de Cercetări Sociale, no. 3-4/1999 Bogdan Voicu (1999) - „Modernitatea între tradiţie şi postmodernism”, in Revista de Cercetări Sociale, no. 3-4/1999 28 corectura 2 28 iulie 2008

Introduction. Romania and the comparative study of values Mălina Voicu (2000) - "Atitudinea fata de impozite in Romania", in Sociologie Romaneasca, new series, no. 3-4/2000 Bogdan Voicu (2001) - „România pseudo-modernă”, în Sociologie Românească,no. 14/2001, p. 36-59. Mălina Voicu (2001) - „Modernitate religioasă în societatea românească”, in Sociologie Românească,no. 1-4/2001, p. 70-96 Mălina Voicu, cu Bogdan Voicu. 2002. Gender values dynamics. Towards a common European pattern?, Romanian Journal of Sociology, XIII (2002), 1-2, p. 42-63. Mălina Voicu, Bogdan Voicu – „Studiul valorilor europene: un proiect de cercetare internaţională”, Calitatea Vieţii, no. 1-4/2002. Paula Tufiş (2002) - Structură socială şi etnicitate, Sociologie Românească, 2001, 1-4, p. 97-123. Cosima Rughiniş (2003) - Valori europene în relaţiile intime. Studiu comparativ, Sociologie Românească, 2002, 1-2, 38-75. Mălina Voicu, Bogdan Voicu (2003) – „Studiul valorilor europene: un proiect de cercetare internaţională”, Calitatea Vieţii no. 1-4/2002 Mălina Voicu (2003) – „Egalitate, inegalitate şi roluri tradiţionale. O analiză comparativă a valorilor implicate în legitimarea politicilor de suport pentru femei în ţările europene”, Calitatea Vieţii no. 1-4/2002. Mălina Voicu, Bogdan Voicu (2003) - „Volunteering in România: a rara avis” , in Paul Dekker şi Loek Halman (editori) – „The Values of Volunteering. CrossCultural Perspectives”, Kluwer Publishers, 2003, 143-160. Bogdan Voicu , Mălina Voicu (2003) – “Volunteering in CEE: one of the missing links?”, presented at the round-table Globalization, Integration, and Social Development in Central and Eastern Europe, University “Lucian Blaga” of Sibiu, 6-8 September 2003 Gabriel Bădescu (2003) - Încredere şi democraţie în ţările în tranziţie, Sociologie Românească, 2003, I, 1-2, 109-128 Mălina Voicu, Bogdan Voicu. 2004. Promoting volunteering in Eastern Europe, European Conference and Exchange Forum about Volunteering, 7-9 November, Maastricht. Mălina Voicu. 2004. Women Work and Family Life: Value Patterns and Policy Making, în Will Arts şi Loek Halman (editori) – European Values at the turn of the Millenium, Brill Book, Leiden Mălina Voicu. 2004. How secular is Romania?, presented at the Annual Conference of the International Study of Religion in Eastern and Central Europe Association ( ISORECEA), organized by the Catholic University Pázmány Péter , Budapest, December, 2004 Bogdan Voicu. 2005. Social capital: bonding or bridging Europe?, in Horaţiu Rusu, Bogdan Voicu eds. 2005. EU Integration Process from EAST to EAST: Civil Society and Ethnic Minorities in a Changing World, Sibiu: Psihomedia: in print. Mălina Voicu. 2005. Religiousness and gender across Europe. in Horaţiu Rusu, Bogdan Voicu eds. EU Integration Process from EAST to EAST: Civil Society and Ethnic Minorities in a Changing World, Sibiu: Psihomedia: in print. Bogdan Voicu, Tanya Basina. 2005. Social capital and civic participation in Ukraine and Romania, presentation at EU Integration Process from EAST to EAST: 29

The Values of Romanians: 1993-2006. A sociological perspective Civil Society and Ethnic Minorities in a Changing World. Roundtable for young social scientist, Lejpaia, 10-13 June, 2005. Bogdan Voicu. 2005a. Penuria Pseudo-Modernă a Postcomunismului Românesc. Volumul I. Schimbarea socială şi acţiunile indivizilor, Iaşi: Expert Projects. Bogdan Voicu. 2005b. Penuria Pseudo-Modernă a Postcomunismului Românesc. Volumul II. Resursele, Iaşi: Expert Projects.

30 corectura 2 28 iulie 2008

Comrades or citizens? Support for democracy and market economy CLAUDIU D. TUFIŞ

The fall of communism represented a defining moment for the Romanian society. Although repeating this statement ad nauseam has transformed it into a veritable cliché, this does not make it any less true. The 1989 anti-communist revolutions represent “a major cultural and civilizational break […] the slow emergence of the new postcommunist culture and civilization” (Sztompka, 1996: 120). The year 1989 marked, thus, the beginning of a complex transition, of a radical social change. If the transitions to democracy in the Latin American countries, those that indicated the start of the third democratization wave (Huntington, 1993), required only political reforms, in the case of the postcommunist countries transition required not only political but also economic and social reforms. All three types of reforms are necessary, because democracy has survived only in countries with a market economy and it has never survived in countries with other types of economic systems (Dahl, 1998: 166). Vitányi showed that if the required political changes can be implemented in only a few years, the economic changes are much more complicated, while the cultural ones require even more time (Vitányi, 1999: 187-188).

The simultaneity of transition to democracy and market economy This dilemma of simultaneity is characteristic to postcommunist transitions. With the exception of Germany and Japan, where democracy and market economy have been implemented by an external actor after the Second World War (Crawford, 1995: 3), democracy and market economy have never 31

The Values of Romanians: 1993-2006. A sociological perspective

been introduced at the same time in any other country (Schopflin 1994; Hall 1995; Offe 1997; Pickel and Wiesenthal 1997). Given the record of previous attempts to simultaneously switch to democracy and market economy, scholars have been rather skeptical about the success of the postcommunist transitions: “many scholars have identified the economic decline that accompanies economic restructuring as the essential dilemma of the dual transition, arguing that if the well-being of the majority of a population is substantially harmed by reforms, popular support for democracy will erode” (Kullberg and Zimmerman, 1999: 326). Similar arguments could also be found in Przeworski (1991), Diamond (1992), Haggard and Kaufman (1995), Nelson (1995), Gati (1996), or Mărginean (1999). These warnings were extremely important, if we take into account the fact that, according to the literature, “if democracy and capitalism are to take root in the former communist states, it is necessary not only to create the institutions and processes intrinsic to those systems, but also to foster popular attitudes that are accepting and supporting of them” (Mason and Kluegel, 2000: 11). The citizens’ support is not only necessary for the good functioning of the new systems; it is necessary for the existence of the democratic systems (Easton 1965; Miller 1974; Norris 1999). This is, in fact, one of the key ideas in the literature: democratic consolidation is not possible in a society that does not accept democracy’s ideals. The stability of a democracy depends on the degree of consistency between the basic principles of the political system and the political values of the citizens (Almond and Verba 1963; Dahl 1989; di Palma 1990; Diamond 1993; Sørensen 1993; White, Gill and Slider 1993; Hahn 1995). This argument is not without its critics. Barry (1970: 48-52) argued that a country can become a stable democracy even if it lacks, initially, a sufficiently large number of people with democratic values, because democracies educate their citizens with such values. Przeworski (1991) completely eliminates citizens from the democratization process, arguing that all democracy needs in order to survive is a self-enforcing equilibrium (although many times this equilibrium is not a real democracy). How people react when faced with major social changes is perhaps one of the most important elements during the transition: “the people are the ultimate movers of reforms” (Sztompka, 1996a: 127) and, at the same time, they are “the main obstacle to reform” (Przeworski, 1993: 185). In addition to this attitudinal congruence, the behavior of the citizens also has an important role in the process of democratic consolidation (Bunce, 2003: 170). Democracy requires the active involvement of its citizens (Dalton and Wattenberg 2000; Norris 2002). Democracy needs democratic citizens. During the last days of 1989, twenty-two million comrades witnessed the fall of the communist system under which they lived for more than forty years. Fifteen years later, in 2004, twentytwo million citizens voted for their fifth time, choosing their third 32 corectura 2 28 iulie 2008

Comrades or citizens?Support for democracy and market economy

postcommunist president. It seems we got used to playing the democratic game. But are we playing it because we like it or because we are forced to play it? Did we manage to transform from communist comrades to democratic citizens? To answer this question one needs to understand what a democratic citizen is. What I understand by this is a person that is attached to the idea of a democratic political system and who considers that, despite all its problems, democracy is a governing system that is better than all others that have been tried (the Churchill hypothesis). This is, however, a minimal definition, to which other attributes can be added (trust, participation, tolerance) to define different types of democratic citizens. In the Romanian case there seems to be an additional dimension that is relevant for this definition: attitudes toward planned economy and market economy. The importance of this dimension can be explained by the influence of two factors. First, at the public level, there is a clear correspondence between communism and planned economy, and between democracy and market economy. Second, external pressures have defined democracy and market economy as the two sides of the same coin, suggesting that a country cannot have a democracy without a market economy. International financial institutions (especially the International Monetary Fund) have always conditioned their assistance on the implementation of specific economic and financial reforms. The majority of these reforms originated from what has come to be known as “the Washington consensus” (for more details see Williamson 1990, 2000). Under these conditions, the simultaneous implementation of both a new political system and a new economic system requires analyzing people’s attitudes toward both systems (Centeno 1994; Simai 1999a). Rose presents a similar argument, considering that since it is difficult finding a democracy without a market economy one cannot talk about a consolidated democracy unless the public accepts the values of both systems (Rose 1992). Previous studies show that democracy and market economy are not always simultaneously accepted, because economic and political reforms are interpreted differently by the public (see, for instance, McIntosh et al, 1994: 507). This is even more important, given that it is generally accepted in the literature that economic reforms add a series of constraints to the democratization process. Przeworski (1991) offers a detailed analysis of the negative effects of the transition to market economy on democratization1.

1

See, also, Diamond (1992), Haggard and Kaufman (1995), Nelson (1995), or Gati (1996) for similar discussions. 33

The Values of Romanians: 1993-2006. A sociological perspective

A model for the analysis of support for democracy and market economy Following the discussion presented above, I analyze in this chapter the attitudes of the Romanians toward democracy and market economy. The main question I address is: to what extent do Romanians accept, more than fifteen years after the fall of communism, the basic principles of democracy and market economy? Although often we do not hesitate to blame the politicians for all the problems we faced throughout the transition (and we are usually right when making these accusations), we rarely stop to think that maybe part of the blame belongs to us. Are we, Romanians, truly democratic citizens, are we tolerant, are we willing to behave democratically, or, rephrasing, do we have a democratic political culture?

Political culture Talking about political culture change from a culturalist perspective is somewhat problematic, this approach being often criticized for its difficulties in explaining change. Eckstein (1988) acknowledged that the postulates of culturalism (oriented action, orientational variability, cultural socialization, and cumulative socialization) do indeed lead to an expectation of continuity but, at the same time, he argued that this is not an unsolvable problem and presented a general culturalist theory of change. His theory deals with both gradual (normal, every-day) and drastic (social discontinuity) changes. Eckstein suggests that, in the case of drastic changes, “cognitions that make experience intelligible and normative dispositions (affect, evaluative schemes) must be learned again, and learned cumulatively […] changes in political cultures that occur in response to social discontinuity should initially exhibit considerable formlessness” (Eckstein, 1988: 796)2. Starting from Merton’s (1957: 141-157) types of individual adaptation – ritualism, innovation, retreatism, rebellion, and conformism – Eckstein presents the following as a set of strategies available for dealing with drastic changes: ritual conformity (characterized by an individual’s rejection of the cultural goals and acceptance of the norms), self-serving conformity (the new goals are accepted but the norms are not), retreatism (both goals and norms are rejected), and intransigent resistance to authority (the goals and norms are not only rejected, they are replaced). These strategies lead, eventually, to new cultural models and themes. Swidler proposed a similar theory, built around the concept of strategy of action, defined as patterns of organizing action, allowing the achievement of different life goals. In a departure from the traditional view of culture, Swidler 2

Similar arguments can be found in Lipset (1960) and Huntington (1968).

34 corectura 2 28 iulie 2008

Comrades or citizens?Support for democracy and market economy

argued that strategies of action are more stable than goals, the result of this being that during unsettled lives people might prefer goals for which they already have a strategy of action (Swidler, 1986: 277). This view may be helpful in explaining, for instance, why some groups accept the ideals of a market economy, while others reject them or accept them only after significant delays (needed for updating their strategies of action). These two theories suggest that changes in political culture are possible and they complement each other to indicate a likely scenario for this change. The fall of the communist regime led to new goals (democracy and market economy) and new functioning rules for the society, and these need to be accepted by the citizens. The adaptation from the old goals and norms to the new ones is a long-time process, characterized by competition among different ideologies. During this process, people can use any of the strategies discussed by Eckstein. Given the duality of the postcommunist transition, it is possible that different combinations of strategies will be used, depending on the relative importance of different goals and norms for the citizens. It should be noted that some combinations have a higher probability of being used than others. For instance, conformism with respect to both democracy and market economy is likely to be a very common combination. Partial adaptation, especially through conformism with respect to democracy and ritualism with respect to market economy can also be a common combination, characteristic to those that are afraid of the negative effects of the transition to market economy. A third frequent combination is defined by ritualism with respect to both democracy and market economy, characterizing those who are not convinced by the goals of the new ideology but that are following its norms. Over time it is expected that only some of the combinations will function and that those that fail will be abandoned. It is also expected that more and more people will adopt the ideal combination for the success of the transition: conformism with respect to both democracy and market economy.

Support for the political-economic system Using WVS/EVS 2005 data, I analyze in this chapter Romanians’ attitudes toward democracy and market economy, trying to discover if the transition to conformism discussed above can be observed in Romania. Attitudes toward democracy and market economy can be interpreted, using Easton’s (1965, 1975) theory, as a form of diffuse support for the principles that create the new political and economic system in Romania3. These attitudes represent, in fact, a form of unconditional support: diffuse support is defined as

3

For a detailed discussion of the political regime as an object of support, see Easton (1965: 190-211). 35

The Values of Romanians: 1993-2006. A sociological perspective

“evaluations of what an object is or represents […] not of what it does” (Easton, 1975: 444) or as “a reservoir of favorable attitudes” (Easton, 1965: 273). Easton’s model includes three objects of support (political community, political regime, and authorities) and two types of support (diffuse and specific). In contrast to specific support, the diffuse type is more durable, it is generated through both socialization and direct experience, and it is expressed through the belief in the legitimacy of political objects4. At the intersection of the two dimensions there are different types of support, that become more and more important for the stability of the political system, from specific to diffuse support and from support for authorities to support for the political community. A country’s chances of having a stable democracy are significantly influenced by the degree of diffuse support for democratic values, goals, and norms among its population (Dahl, 1996: 3). Similarly, Merkel (1996) considers attitudinal consolidation (defined as specific and diffuse support) as an important component for democratic consolidation, next to structural and representative consolidation. Morlino and Montero (1995) modified this model and showed that diffuse support is mainly determined by the perceived efficacy of democratic practices, considering that in the context of countries in transition the lack of other alternatives can increase diffuse support. Przeworski extended this idea: “what matters for the stability of any regime is not the legitimacy of this particular system of domination, but the presence or absence of preferable alternatives” (Przeworski, 1986: 51). This type of support, determined by the lack of alternatives, is defined by Valenzuela (1992) as “inverse legitimation”. It should be noted, however, that this type of legitimation is more fragile than the one determined by the belief in the regime’s qualities. The application of this concept in the Romanian case suggests that democracy benefited right from the start of the transition from a plus of legitimation, generated by the consensus over the type of political system that had to be implemented. Although communism or nationalism could be interpreted as possible alternatives, democracy, as an ideology, became the preferred solution relatively easy. Discussing the chances of communism surviving in the former communist countries, Dalton was asking: “After all, what does a communist say after attending communism’s funeral?” (Dalton, 1994: 481). In the case of market economy, the public debates regarding Romania’s future economic system have continued for a long time, generating a certain uncertainty among the population and blocking the appearance of an initial reservoir of attitudes favorable to market economy. Countless opinions have been formulated during these debates, covering a vast ideological space, from 4

Easton (1965: 267-340) and Easton (1975: 444-453) offer a detailed discussion of diffuse support.

36 corectura 2 28 iulie 2008

Comrades or citizens?Support for democracy and market economy

those considering that the state economy could survive the fall of communism to those that were arguing for gradual economic reforms and to those proposing a “shock” reform as the only solution to Romania’s economic problems.

Diffuse support for democracy Linde and Ekman (2002) and Canache, Mondak, and Seligson (2001) discuss in detail different ways to measure support for democracy. I use in this chapter two variables that refer to support for democracy as a political system: how good is it for Romania having a democratic political system and how important is it for the respondent living in a democratically-governed country. These two variables are significantly correlated (r = 0.307) and can be grouped in an additive index of diffuse support for democracy that I use as dependent variable in the models presented in this chapter. These variables are also used in Maravall (1997), Plasser, Ulram and Waldrauch (1998), Fuchs, Guidorossi and Svensson (1999), Klingeman (1999), and Hofferbert and Klingeman (2001). Given that attitudes toward democracy can be influenced by social pressure, in addition to the dependent variable discussed above I also present results for three additional dimensions that can clarify to what degree the support indicated by the dependent variable is real or only declarative. The first dimension measures attitudes toward alternatives to democracy: an authoritarian regime, a military regime, or a technocratic regime. The correlations among these three variables vary between 0.200 and 0.289, and a factor analysis solution groups them into a separate factor from the variables included in the dependent variable. These variables are used in a similar manner in Hahn (1991), Gibson, Duch and Tedin (1992), McIntosh et al (1994), Gibson (1996b), Miller, Hesli and Reisinger (1997), Miller, White and Heywood (1998), Plasser, Ulram and Waldrauch (1998) and Voicu (2005). The second dimension represents tolerance toward population subgroups that differ from the majority group: people belonging to other races, of other ethnicities, with other religions, with different sexual preferences, gypsies, immigrants, cohabitating couples, and people suffering from AIDS. The analysis of this dimension is required by the fact that tolerance is identified as one of the defining characteristics of a democratic political culture. The third dimension includes characteristics that Romanians consider to be important for democracy: governments should tax the rich and help the poor, religious leaders should interpret the laws, free elections, having unemployment benefits, the army should rule when the government is incompetent, civil rights should protect people’s freedom, prosper economy, harsh sentences for criminals, people should be able to change laws through referenda, and women should have the same rights as men.

37

The Values of Romanians: 1993-2006. A sociological perspective

Predictors of diffuse support for democracy Although a large proportion of the Romanian population is characterized by high levels of diffuse support for democracy, there are also people that are indifferent, or even opposed, to democracy. What are the factors that affect diffuse support for democracy? During periods of significant social changes, people characterized by modernism have a more positive attitude toward the new regime than people characterized by traditionalism5. The interpretation of society’s modern groups as the most pro-democratic groups is consistent with Swidler’s argument that strategies of action are more important than goals, leading to the possibility that people will prefer those goals they can achieve using the strategies of action they have. These strategies of action depend on characteristics that allow an easier adaptation to change (age, education, residential area). People belonging to this group are also more interested and more active in political life. Thus, a first group of variables that can affect support for democracy is composed by respondents’ socio-economic characteristics: gender, age, education, income, ethnicity, religion, and the size of the locality of residence. Gender is used as a control variable because, although women are usually more traditional than men, I do not believe there are significant differences between the two groups. Previous studies showed either that men support democracy more than women (Finifter and Mickiewicz 1992; Gibson, Duch and Tedin 1992; McIntosh et al 1994; or Miller, Hesli and Reisinger 1994) or that there are no significant differences (Dalton 1994; Gibson 1996b; or Mishler and Rose 2002). The only significant differences have been found in studies that used data from the beginning of the transition in Russia. The variable is coded 0 for females and 1 for males. Age can be interpreted as an indicator for different socialization experiences. A person reaching adulthood during Romania’s period of industrialization and economic development (between the 1950s and the 1970s) will have different attitudes toward democracy and communism compared to a person who reached adulthood during the last years of the communist regime. Age can also have another type of influence: young people have more resources to successfully adapt to the new regime, while older people have only limited resources. Even more, attitudinal change becomes more and more difficult as a person gets older. I expect, then, age to have a weak negative effect on diffuse support for democracy. Gibson (1996), Bahry (1997), and Jacobs, Müller and Pickel (2003) have also found significant negative effects in similar studies. Education is probably the variable with the strongest effect in this group. It is agreed in the literature that people with a high level of education

5

Bogdan Voicu writes about traditionalism, modernism, and postmodernism in the Romanian society in his chapter in this volume.

38 corectura 2 28 iulie 2008

Comrades or citizens?Support for democracy and market economy

have smaller chances of being negatively affected by the transition, are more modern, and have more resources to adapt to the new system, all these suggesting that the level of support for democracy increases with education. Hahn (1991), Finifter and Mickiewicz (1992), Gibson, Duch, and Tedin (1992), and Bahry (1997) showed that education has significant positive effects on support for democracy. The education variable has five categories: four grades or less, five – eight grades, nine – eleven grades, high school or post-high school, and university or post-university. The next variables in the group are used as control variables. Income is measured as deciles of income per capita. Ethnicity is coded 1 for Romanians and 0 for other ethnicity. Religion is coded 1 for Christian-Orthodox and 0 for other religions. The last variable in the group is the size of the locality of residence, measured through six categories: village, commune center, small city, medium city, large city, and very large city. This variable also distinguishes between rural and urban areas. Rural areas are characterized by traditionalism, low education, high poverty, and an aged population, all factors that create an environment that is hostile to change and that suggest that people living in rural areas support democracy less. The second group of variables I use in the model measures respondents’ activities in the political life: interest in politics, NGO membership, and protest activities. All variables in this group should have significant positive effects on diffuse support for democracy because they indicate to what degree respondents use standard methods of expressing their opinions in a democratic regime. Interest in politics can be interpreted as “an indicator of citizens’ cognitive involvement in the political process” (Plasser, Ulram, and Waldrauch, 1998: 130). It also seems to be related to attitude formation and to political participation (van Deth 1990; Niedermayer 1990; Maravall 1997). From a theoretical perspective, “it is a central tenet of classical democratic theory that, if popular sovereignty is to have meaning, citizens should be informed about the issues confronting society and should care about their resolution. […] some minimal levels of interest, knowledge and participation would appear to be essential” (Hahn, 1991: 415). This view is consistent with Almond and Verba’s (1963) understanding of the different types of political culture. I measure interest in politics using an additive score composed of two variables: interest in politics and the importance of politics in the respondent’s life. The correlation between the two components is 0.526. The variable ranges between 0 and 6. The two variables included in this index are standard measures that have also been used in previous studies (see Gabriel and van Deth 1995; Plasser, Ulram and Waldrauch 1998; Mishler and Rose 2001; or Martin 2003). In developed democracies citizens have at their disposal a series of points of access to the political system, through which they can express their wishes or their discontent: voting, party membership, NGO membership, lobby activities, protest activities, etc. Although most of these points of access also 39

The Values of Romanians: 1993-2006. A sociological perspective

exist in Romania, they do not function with the same efficiency. As I have shown in the chapter on trust in this volume, political parties are the institutions Romanians trust least. Even more, most of the parties are highly centralized, most of the decisions being made by the party leadership without taking into account the opinions of the regular party members. Romania also does not have a lobbying law, paving thus the way toward semi-legal forms of lobbying. The election turnout, which decreased significantly between 1990 and 2004, is, probably, the most widespread political activity among the general population, but the fact that a person votes every four years does not say very much about its political activism. Membership in NGOs is one of the best indicators of participation to the social life. Given that in Romania membership in NGOs is lower than in developed democracies, it can be said that the Romanian civil society is underdeveloped (Carothers 1996; Verdery 1996; Rupnik 1997; Tismăneanu 1998). Even more, it can also be characterized as being dominated by elites. It should be noted that several important political figures have entered the political arena coming from the civil society. Some of the most significant examples include Emil Constantinescu (founding member of the Civic Alliance who became president), Victor Ciorbea (former trade union leader who became prime minister), Miron Mitrea (former trade union leader who became minister), and Monica Macovei (former president of the Romanian Helsinki Committee who became minister). Regular citizens seem to have more difficulties in organizing into groups meant to protect and promote their own interests. Howard (2002), analyzing civil society in former East Germany and Russia, suggests that postcommunist citizens refuse to become active members of civil society because of their experiences during the communist regime, when voluntary participation in different organizations was mandatory. Based on the previous discussion, I consider NGO membership as a strong indicator of a person’s activism. The variable that I use is coded 0 for people that do not belong to any voluntary organization and 1 for people that are members (active or inactive) in any of the following organizations: religious or church organizations, sport or recreation, education, art or music, trade unions, political parties, environmental organizations, professional associations, charitable organizations, consumer organizations, or other. The third variable in this group is represented by protest activities, coded 0 for people that never took part in a protest action, and 1 for a person that participated in at least one of the following types of protest: signing a petition, boycotting, or participating in demonstrations. The fact that a person has been active in protest activities shows that the person decided to express its dissatisfaction using one of the few means at his/her disposal. I consider such a person to be much more attached to democracy, compared to an individual that is just as dissatisfied, but considers the expression of its lack of satisfaction to be a useless act. If the first person considers they have a chance to solve their 40 corectura 2 28 iulie 2008

Comrades or citizens?Support for democracy and market economy

problems by their actions, the second one has a more cynical view on democracy. The next set includes three variables that indicate respondents’ general value orientations: preference for left or right6, preference for liberty or equality, and position on the materialism – postmaterialism dimension. The variable of self-identification with left or right is used as a control variable. Theoretically, there are no arguments to sustain that supporters of the left are more pro-democratic than supporters of the right or the opposite. On the other hand, it is possible that both extreme left and extreme right supporters are less attached to democracy, but both extremes are rarely visible in the Romanian society. Preference for liberty or equality is coded 0 for those that prefer equality and 1 for those that consider liberty as being more important. This variable indicates different value profiles and should have a significant effect on diffuse support for democracy, with those preferring liberty being more oriented toward democracy. A respondent’s postmaterialism level should also influence support for democracy. The materialism – postmaterialism index is composed of 12 items, using the standard method, and it ranges between 0 (the respondent did not choose any of the postmaterialism values) and 6 (the respondent selected all six postmaterialist values). For more details on this index, see Inglehart (1971, 1977, 1981, 1990) and Inglehart and Abramson (1999). Although people characterized by postmaterialism may be dissatisfied with the functioning of democracy and may want some aspects of the political system to change, the goal of their demands is improving democracy, postmaterialists being more attached to democratic values than materialists. Returning to Easton’s model, it should be noted that diffuse support for democracy is the most stable form of support. Yet, Easton shows that diffuse support can be influenced by specific support (which is determined by evaluations of the performance of the political system). Starting from this idea, the fourth group of variables in the analysis model includes a series of variables testing the significance of these relationships. The first variable in this group is an index of trust in the state’s institutions7. This index measures specific (or generalized, according to some scholars) support for the structure of the political system and should have a significant positive effect on diffuse support for democracy. The second variable in this group is satisfaction with the way democracy functions in Romania. This variable, which measures specific support for the actions of the political system and is determined by how people evaluate the functioning of democracy, should have a significant positive effect. 6

See Mircea Comşa’s chapter in this volume for the analysis of the left-right dimension in the Romanian society. 7 For details on this variable see my chapter on trust in this volume. 41

The Values of Romanians: 1993-2006. A sociological perspective

The third variable is an index of satisfaction with the cabinet’s activity in different areas: standard of living, public order, jobs, agriculture, privatization, health, education, housing, industry, and environmental protection. The more satisfied with the cabinet’s actions a person is the more reasons for this person to have favorable attitudes toward democracy. Although Easton refers in his model only to the political system, in the Romanian case, as I argued above, the political system cannot be separated from the economic system. It is thus possible for diffuse support for democracy to be influenced by both diffuse and specific support for market economy. I discuss these variables in the following section, devoted to support for market economy. The last variable I use in the model starts from Anderson and Tverdova’s (2001) observation that support for democracy is significantly influenced by how the respondent voted in the last elections: those who voted for the governing party (or parties) support democracy more than those who voted for the parties that lost the elections. Although the WVS questionnaire does not have an item referring to the vote during the last elections, respondents are asked who they would vote for if elections were held next Sunday. I recoded this variable so that 1 corresponds to those who would vote for one of the governing parties and 0 corresponds to those who would vote for an opposition party.

Diffuse support for market economy The transition to market economy represents the second dimension of the postcommunist transition in Romania. In measuring diffuse support for market economy I take into account four variables that show the respondents’ acceptance of basic principles of the capitalist economic system: competition, private property, income differentiation, and responsibility for personal welfare. I recoded the variables so that the minimum value (1) on each scale represents acceptance of a value opposed to market economy (competition is a bad thing, state property should extend, income differences should be reduced, and the state should be responsible for individual welfare), and the maximum value (10) indicates acceptance of values characteristic to market economy (competition is good, private property should extend, income differences should increase, and individuals should be responsible for their own welfare). These variables have also been used in previous studies: Finifter and Mickiewicz (1992), Duch (1993), McIntosh et al (1994), Miller, Hesli and Reisinger (1994), Gibson (1996b), Firebaugh and Sandu (1998), and Hofferbert and Klingemann (1999). The more a person accepts market economy values, the more attached to the ideals of this economic system the person is. Factor analysis shows that the four variables are grouped not in one, but in two factors (the correlation between the two factors is 0.173). The first factor (41.27% of the variation) groups attitudes toward income differentiation and responsibility for personal welfare, suggesting it captures an orientation towards 42 corectura 2 28 iulie 2008

Comrades or citizens?Support for democracy and market economy

a social-democratic model of market economy or towards some form of protection against the inequalities produced by a pure market economy. The second factor (28.37% of the variation) groups attitudes towards competition and private property, being, thus, a factor of orientation towards the classical (liberal) model of market economy. Starting from the solution offered by the factor analysis, I use the two indices (built as additive scores) as dependent variables in the models of diffuse support for market economy. I will use the same set of independent variables for both dependent variables, letting the results show the differences between the two components.

Predictors of diffuse support for market economy Just like in the previous model, the first group of variables in this model is composed of respondents’ socio-economic characteristics. Gender is a control variable, although it is possible that women are less attached to the ideals of market economy if we consider that gender inequality on the job market increased during the transition. Such a negative effect was found in McIntosh et al (1994), Miller, Hesli, and Reisinger (1994), and Gibson (1996b). Age is one of the variables that create a significant distinction between the winners and the losers of the transition. The changes in economic structure required adaptation through re-specialization, something that gets more and more difficult as a person gets older. Financially, retired persons lost more during the transition compared to employed people. Even more, although official statistics do not indicate unemployment as an important problem for older people, this happened only because the unemployment rate for this age group has been artificially reduced by decreasing the minimum retiring age. Thus, older persons moved from the active to the inactive group, evading unemployment. All these suggest that losses during the transition are associated with age. McIntosh et al (1994), Gibson (1996b) and Firebaugh and Sandu (1998) report age as having a negative effect on support for market economy. Previous studies have usually shown a positive relationship between education and attitudes toward market economy8. This relationship can be explained by the fact that education offers people the necessary tools to successfully adapt to changes in the economic structure. I expect education to have a significant positive effect on diffuse support for market economy. Income will also have a positive effect: to fully enjoy the advantages of a market economy a certain level of income is necessary. Those who do not reach this level have all the motives to be unhappy about market economy. The size of the locality of residence should have a significant positive effect on support for market economy. The bigger the locality, the more opportunities of finding a good job. Even more, most of the foreign investments 8

See, for instance, Duch (1993), McIntosh et al (1994), Firebaugh and Sandu (1998), or Gibson (1996: 968). 43

The Values of Romanians: 1993-2006. A sociological perspective

were directed toward large and very large cities, leading to their rapid economic development, while smaller cities and the rural areas had to survive only with the economy they inherited from the communist regime. Respondents’ values are represented by the second set of variables, which includes preferences for left or right and preferences for liberty or equality. Those that feel closer to the left should have more support for the social-democratic model of market economy, while followers of the right should be more attracted to the liberal model of market economy. With respect to preferences for liberty or equality, I expect this variable to have an effect only on the social-democratic model (with those preferring equality supporting more this model), but not on the liberal model of market economy. Evaluations of the current personal, local, or national economic situation and optimism or pessimism about the future personal economic situation can have a significant effect on diffuse support for market economy. The more satisfied a person is with the current situation and the more optimistic about his/her future situation, the more attached to the values of market economy this person should be. The variables included in this group correspond to the theories developed in the literature to explain voting decisions: the pocketbook theory (prospective and retrospective) and the sociotropic theory (see Fiorina 1981, Kiewiet 1983, Lewis-Beck 1988, and MacKuen, Erikson and Stimson 1992). The last group of variables used in this mode attempts to capture the effects of other types of support on diffuse support for market economy. This group includes: trust in the state’s institutions (as an indicator of specific support for the structure of the political system), satisfaction with the functioning of market economy in Romania (as indicator of specific support for market economy), and satisfaction with the cabinet’s actions in different areas. The last two variables should have the strongest effects on diffuse support for market economy. Starting with the next section I present the results of statistical analyses for three models: diffuse support for democracy, diffuse support for market economy (the social-democratic type) and diffuse support for market economy (the liberal model). In addition to these models I also present some results related to tolerance for minority groups and to the way Romanians define democracy.

Results In analyzing the data for this chapter I used the Norm software (Schafer 2000). Thus, for those respondents who did not answer all questions I estimated their answers based on the principles described in the chapter on trust included in this volume. 44 corectura 2 28 iulie 2008

Comrades or citizens?Support for democracy and market economy

Figure 1 presents the evolution of attitudes toward democracy and its alternatives between 1999 and 2005 (the variables were not included in the 1993 and 1997 surveys). The data show that diffuse support for democracy has significantly increased during this period. Almost 95% of the Romanians consider (in 2005) that having a democratic political system is good for Romania, compared to only 75% in 1999. The second indicator of diffuse support is not presented in this figure, because it is available only for 2005. When asked to rate from 1 to 10 the importance of living in a country that is governed democratically, 95% of the respondents give a passing grade (at least 5) and almost half give the highest grade (10). The mean of this variable is 8.5. These two variables show that, 16 years after the fall of the communist regime, Romanians are extremely attached to the idea of living in a democracy (see, also, Mărginean, Precupeţu, and Precupeţu, 2001: 23). Figure 1. Diffuse support for democracy (% of population with positive attitudes towards democracy) 100 Democracy (% pro) Military regime (% con)

75

Technocratic regime (% con) Authoritarian regime (% con)

50

25

0 1999

2005

Data source: WVS 1999 and WVS 2005.

There can be, however, significant differences between what people are saying and what they are really thinking or doing. Is it true that almost all of the Romanians are strongly democratic beings? Figure 1 also presents three variables measuring attitudes toward three alternatives to a democratic regime: a military regime, a technocratic regime (in which experts, and not the 45

The Values of Romanians: 1993-2006. A sociological perspective

government, take the important decisions for the country), and an authoritarian regime (in which there is a strong leader who does not bother with elections and the parliament). These variables are often used in the literature as indicators of (anti-) democratic attitudes. More than three quarters of the Romanians consider that a military regime would not be good for Romania, a result that is consistent with the previous variables. If we look at attitudes toward a technocratic or an authoritarian regime, however, we obtain a completely different image, in which the real support for democracy is lower than the declarative support (see Voicu 2005): only a quarter of the respondents believe a technocratic regime would be bad for the country, and only 22% have a negative opinion about an authoritarian regime that does not have a parliament and elections. How can these differences be interpreted? A possible explanation could be that when respondents are asked about their attitudes toward democracy they automatically offer the answer they consider to be correct (or expected from them) as result of social pressure. The differences could also be explained by the fact that the last two variables do not measure only democratic attitudes; they capture other attitudes as well. In the case of the variable on technocratic regime, a positive answer should be interpreted as indicating a favorable attitude toward decisions made by experts rather than a negative attitude toward decisions made by one of the key institutions of a democratic regime. Similarly, a positive answer on the variable referring to an authoritarian regime should not be interpreted as an antidemocratic answer, but rather as an indicator of a poor opinion about the quality of the Romanian parliament and elections. These two variables do not show acceptance of non-democratic alternatives. They actually show preference for an efficient governing act, with decisions that are made quickly, by people that know what they are doing. They also indicate a negative vote given to Romanian politicians, suggesting that Romanians see the MPs as wasting time on useless debates rather than adopting necessary legislation and do not consider the cabinet as being composed of experts. If one remembers cases such as the fight to replace Adrian Năstase from the leadership of the House, then it is easy to understand why Romanians prefer a strong leader that does not waste time with the parliament. It should also be noted that, for three of the four variables presented here, the percentage of people with democratic attitudes has significantly increased between 1999 and 2005. Tolerance is one of the most important attitudes for a democratic regime. In the 2005 survey, respondents received a list of ten groups of people and were asked to indicate those they would not like having as neighbors. I have eliminated from this analysis two groups, because in their case intolerance can be justified and does not represent an anti-democratic behavior: people that are dependent on drugs (76% of Romanians would not like having them as

46 corectura 2 28 iulie 2008

Comrades or citizens?Support for democracy and market economy

neighbors) and alcoholics (more than 68% of the Romanians would not like such neighbors). The rest of the results are presented in Figure 2. Less than 20% of the respondents have an intolerant attitude with respect to five of the eight groups included in analysis: cohabitating couples, immigrants, and people of other race, ethnicity, or religion. These results are consistent with those presented above and create the image of a relatively tolerant population. Figure 2. Intolerance to minority groups (% intolerants in population) 75

50

25

0 Homosexuals

Gypsies

AIDS infected

Other race

Immigrants Cohabitating couples

Other religion

Other ethnicity

Data source: WVS 2005.

Unfortunately, Romanians do not manage to show the same level of tolerance with respect to the other three groups: around 40% would not like AIDS infected people as neighbors, percentage that increases to 50% in the case of gypsies and to 60% in the case of homosexuals. If in the case of AIDS infected people intolerance can be explained by the lack of education regarding how this disease can be contacted, the reluctance to accept gypsies or homosexuals as neighbors is a clear indicator of an anti-democratic attitude. Intolerance toward gypsies is a classical example of racial discrimination, and it can be explained by the almost daily promotion of negative stereotypes through mass media. Intolerance toward homosexuals can be largely explained by the fact that Romanians are, generally, a very religious people that consider such a sexual preference at odds with their faith. Regardless of its causes, however, intolerance is characteristic of a large proportion of the population.

47

The Values of Romanians: 1993-2006. A sociological perspective

Additional information on Romanians’ democratic attitudes is presented in Figure 3. For each of the ten characteristics, respondents have been asked to rate the importance for democracy on a scale from 1 to 10. Only four of these characteristics (free elections, equal rights for men and women, protection of individual rights, and the possibility of changing laws through referenda) belong to the classical definition of democracy and in all cases respondents have identified them as such (three of these variables have a mean score higher than 9.25 and the fourth has a mean score of 8.4). The results, however, suggest that Romanians define democracy in a much wider sense than those studying democracy, including in this definition aspects related to economy and maintaining order. Figure 3. Characteristics essential for democracy (average score on a ten-point scale)

Data source: WVS 2005. The variables are measure on a scale from 1 (not essential for democracy) to 10 (essential for democracy).

Surprisingly, Romanians consider having a prosperous economy as the most important characteristic of a democracy (average score 9.46). The existence of unemployment benefits is also seen as a very important characteristic of a democracy (mean score 8.84). The third characteristic that belongs to the economic domain (taxing the rich and helping the poor) has a lower average score (6.79), being considered of only average importance for democracy. Although these results show that democracy and market economy need to be analyzed together (as I suggested above), they also demonstrate a certain level of confusion among the public with respect to what democracy is. This conclusion is strengthened by the large average score (9.32) obtained by another characteristic that, normally, is not relevant at all for a democratic system: harsh sentences for criminals. Just like in the 48 corectura 2 28 iulie 2008

Comrades or citizens?Support for democracy and market economy

case of support for an authoritarian or a technocratic regime, these results suggest that people think of some of these characteristics as areas where the transition failed to meet their expectations, rather than thinking of them as characteristics that are essential to democracy. The results presented so far show that Romanians declare a very high level of support for democracy that increased over time. But this is happening in a context in which a large proportion of the population defines democracy in a peculiar way and does not hesitate to exhibit an intolerant (anti-democratic) behavior towards some minority groups. Table 1. Diffuse support for democracy Model 1

Model 2

Model 3

Model 4

Model 5

Model 6

Intercept

4.760 ***

4.647 ***

4.494 ***

4.040 *** 3.699 *** 3.719 ***

Male

0.154 **

0.087

0.096 +

0.095 +

0.066

0.067

Age

0.010

0.016

0.018

0.020

0.026

0.027

Education

0.061 +

0.024

0.015

0.022

-0.005

-0.005

Deciles income per capita

0.063 ***

0.058 ***

0.054 ***

0.043 **

Ethnicity: Romanian

-0.018

Religion: Orthodox

0.042

Urban

0.093

-0.032

0.028 *

0.027 *

-0.051

-0.080

-0.076

-0.083

0.045 +

0.041

0.027

0.026

0.024

0.112 +

0.101

0.152 *

0.123 *

0.120 +

Interest in politics

0.047 *

0.045 *

0.018

0.009

0.010

NGO member

0.240 ***

0.233 ***

0.215 **

0.173 *

0.175 *

Protest

0.304 ***

Ideological self-placement to right

0.293 *** -0.004

0.323 *** 0.296 *** 0.293 *** -0.022

-0.027

-0.031

Prefers liberty to equality

0.205 **

0.135 *

0.077

0.078

Postmaterialism

0.070 +

0.063 +

0.050

0.049

Trust in state’s institutions

0.023 **

0.024 **

0.024 **

Satisfaction with functioning of democracy

0.142 *** 0.136 *** 0.135 ***

Satisfaction with cabinet’s actions

-0.010

-0.014 *

-0.015 *

Support for market economy (liberal)

0.045 *** 0.045 ***

Support for market economy (social-democrat)

0.023 *** 0.023 ***

Satisfaction with market economy

-0.024

-0.025

Vote for governing party 2

Adjusted R 2

R change

Notes:

0.074 5.0%

7.8%

8.9%

14.2%

17.5%

17.5%

5.0%

2.9%

1.1%

5.3%

3.2%

0.1%

The coefficients in the table are unstandardized regression coefficients. Significance levels: *** p < 0.001 ** p < 0.010 * p < 0.050 + p < 0.100. 49

The Values of Romanians: 1993-2006. A sociological perspective

Table 1 presents the results of regression analyses explaining diffuse support for democracy9. The final model (model 6) explains 17.5% of the variation in diffuse support for democracy. Five of the six groups of independent variables used in the model contribute to explaining some of the variation in the dependent variable, the most important being the socioeconomic characteristics and other types of support for components of the political system. Among the groups of socio-economic variables, income is the only one with a significant effect: the higher a person’s income, the higher its level of diffuse support for democracy. The total effect of this variable, however, is relatively small. Residence in urban areas also has a marginally significant positive effect. The coefficients for this group of variables lead to a very important conclusion: diffuse support for democracy does not depend on respondents’ socio-economic characteristics. Men and women, young and old people, people with low or high levels of education, all of them support democracy to the same extent. If there were any differences in support that could be explained by these variables, they have disappeared during the transition. The second set of variables in the model measures the effect of being involved in the political life on support for democracy. Two of these variables have significant positive effects: both membership in NGOs and previous participation in different forms of protest increase diffuse support for democracy. These results confirm the theories arguing that democracies need active citizens to develop. Given that in Romania both types of behavior are relatively rare, it is possible that support for democracy will increase to the extent that people will become more involved in the political life. Most probably this increase will be determined by increased participation to voluntary activities (the Romanian transition has been characterized by low levels of protest). Although two of the variables measuring respondents’ values have significant effects when they are introduced in the model (model 3), they lose their significance in the final model, which includes all independent variables. Preferences for left or right, preferences for liberty or equality, and the level of 9

I estimated the models in SPSS 14.0.2. using the OLS (Ordinary Least Squares) regression. The starting model includes as independent variables only the socioeconomic characteristics of the respondents. For each group of variables I introduced in analysis I estimated a new model. The coefficients in the table are unstandardized regression coefficients (b). The last row in the table (R2 change) shows the increase in the explanatory power of the current model compared to the previous model. The coefficients on this row are a measure of the importance of different groups of variables in explaining diffuse support for democracy. For instance, in Table 1, Model 2 differs from Model 1 by the political activism variables. By adding these variables, compared to Model 1, the explanatory power of Model 2 increased with 2.9%, from 5.0% to 7.8%. 50 corectura 2 28 iulie 2008

Comrades or citizens?Support for democracy and market economy

postmaterialism do not distinguish among people with different levels of support for democracy. Diffuse support for democracy seems to be influenced by other forms of support for the political system. Both trust in the state’s institutions (as indicator of specific support for the structure of the political system) and satisfaction with the way democracy works (as indicator of specific support for the actions of the political system) significantly increase the level of diffuse support for democracy. These results confirm Easton’s hypothesis that diffuse support can be affected by specific support. The third variable in this group (satisfaction with the cabinet’s actions in different areas) has a significant but negative effect which is difficult to interpret because, theoretically, this coefficient should have been positive. This negative coefficient could be partially explained if one interprets satisfaction with the cabinet’s actions as an indicator of respondents’ critical attitudes. If this is true, then those who are critical with respect to the actions of the cabinet should have higher levels of diffuse support for democracy, an interpretation that corresponds to our image of democratic citizens as well-informed, critical people that play the role of guardians of democracy. It should be noted, however, that post hoc explanations are rarely convincing. The coefficients of the variables in the next group, diffuse and specific support for market economy, show that people who support market economy (either the liberal or the social-democratic type) are also more likely to support democracy. This finding suggests that Romanians understand that democracy cannot exist without market economy. It is interesting to note that satisfaction with the functioning of the market economy does not have a significant effect on support for democracy, indicating that Romanians are willing to support democracy even if they may be dissatisfied with the current situation of the economy. The last variable in the model, voting intention for a governing party, does not have a significant effect, suggesting that people do not condition their support for democracy on who is governing the country. The results in Table 1 show that diffuse support for democracy is not affected by respondents’ socio-economic characteristics (with the exception of income), confirming the results presented before, which showed democracy as being accepted by most of the population. Democracy benefits from citizens’ involvement in the political life, although their numbers are not very large. Finally, diffuse support for democracy is significantly influenced by diffuse support for market economy and by specific support for the structure and actions of the political system. Moving on to diffuse support for market economy, Figure 4 presents the evolution of attitudes toward four essential values of market economy from 1993 to 2005. The attitudes toward competition and private property represent diffuse support for the liberal model of market economy, while attitudes toward income differences and responsibility for individual welfare represent diffuse 51

The Values of Romanians: 1993-2006. A sociological perspective

support for the social-democratic model of market economy (one in which the state intervenes to eliminate the inequalities generated by the market economy). Figure 4. Diffuse support for market economy 100

75

50 Individual welfare Income differences

25

Competition Private property

0 1993

1997

1999

2005

Data source: WVS 1993, WVS 1997, WVS 1999, and WVS 2005. The data represent percentages of population with positive attitudes toward the four components of market economy.

Diffuse support for the liberal model of market economy presents minor fluctuations, but it remains at approximately the same level throughout the whole period of time: around 90% of the Romanians have a positive attitude towards competition and about 63% have a positive attitude towards private property. Support for the social-democratic model of market economy, however, increases significantly between 1993 and 2005. If in 1993 only 40% of the population considered that the state should be responsible for individual welfare and that income differences should be reduced, by 2005 almost two thirds of the population supports these values. It should be noted that the social-democratic model, which asks the state to intervene on the market to protect its citizens, is ideologically opposed to the liberal model, which considers that the state should not intervene (or it should do so only to a small extent) on the market. It is interesting then to find out that even if support for the social-democratic model increases, this is not followed by a decrease of support for the liberal model, suggesting that Romanians support market economy and, at the same time, require state intervention to help those who suffered during the transition. 52 corectura 2 28 iulie 2008

Comrades or citizens?Support for democracy and market economy

The results of the regression analyses presented in Table 2 confirm this interpretation. In reading the results in this table one should take into account the coding of the two dependent variables. The variable for the liberal model is coded so that a high score indicates acceptance of the liberal values and a low score indicates their rejection. In the case of the variable for the social-democratic model, a high score shows acceptance of state intervention on the market, which is opposed to the principles of the liberal model. As a result of this coding scheme, the coefficients of the independent variables will have opposed signs for the two dependent variables. The final model explains 14% of the variation in diffuse support for the liberal model of market economy and 17% in the case of the socialdemocratic model (reminder: I use the liberal and social-democratic terms referring to ideologies and not to the political parties with the same names). In contrast to support for democracy, in these models the respondents’ socio-economic characteristics explain the largest part of the variation in the dependent variables. This suggests the existence of significant differences in support for market economy among different population subgroups; whereas in the case of support for democracy only minor differences have been found (income was the only variable with a significant effect in that model). Taking into account the effects of the independent variables included in the model, it can be seen that support for the two dependent variables has different determinants. Men show more support for the liberal model than women, but there are no significant differences between the two groups with respect to support for the social-democratic model. The age coefficients are not significant, showing that age does not generate differences of support for market economy. In the case of the social-democratic model of market economy, age has significant coefficients in the first two models, but this effect disappears when the evaluations of the personal and national economic situation are included in analysis. Education has a strong significant effect in both models. The more educated a person is, the more that person will support the liberal model of market economy and oppose the social-democratic model. The liberal model of market economy has more support from people with high income (significant positive coefficient) but in the final model for the social-democratic model income has only a marginally significant effect. The results for education and income confirm the theoretical expectations: those who had the least to lose during the transition (people with high education and high income) have all the reasons to support the liberal model of market economy and, at the same time, do not feel the need for the state to intervene on the market. The last variable in this group, residence in urban / rural, shows that urbanites have more support for the liberal model of market economy, but do not differ significantly from those living in rural areas with respect to support for the social-democratic model. The two variables indicating the value profile of the respondents have coefficients that are consistent with theoretical expectations. With respect to the left-right dimension, supporters of the right favor the liberal model, while 53

The Values of Romanians: 1993-2006. A sociological perspective

supporters of the left seem to be more attached to the ideals of the socialdemocratic model. At the same time, those who consider liberty more important than equality have higher levels of support for the liberal model, while those who value equality more have higher levels of support for the social-democratic model (moreover, the effect of this variable on support for the social-democratic model is twice as large than in the case of the liberal model). Table 2. Diffuse support for market economy The liberal model

The social-democratic model

Model 1 Model 2 Model 3 Model 4

Model 1

Model 2

Model 3

Model 4

Intercept

8.697*** 7.756*** 7.408*** 7.363*** 13.117*** 14.732*** 18.494

18.230

Male

0.642**

-0.399+

-0.403+

0.073

0.081

Age

-0.072

0.641** -0.055

0.638** -0.041

0.626** -0.064

-0.390+

-0.387+

0.262*** 0.234**

Education

0.471*** 0.445*** 0.435*** 0.460***

-0.605*** -0.565*** -0.517*** -0.520***

Deciles income per capita

0.300*** 0.285*** 0.274*** 0.256***

-0.276*** -0.252*** -0.146** -0.113*

Ethnicity: Romanian

-0.019

-0.017

-0.005

-0.016

-0.136

-0.141

-0.224

-0.193

Religion: Orthodox

0.033

0.025

0.026

0.001

-0.123

-0.110

-0.093

-0.074

Urban

0.535*

0.474*

0.501*

0.562*

-0.090

0.006

-0.155

-0.166

Ideological self-placement to right

0.111*

0.108*

0.083+

Prefers liberty to equality

0.839*** 0.838*** 0.769**

-0.185** -0.141*

-0.121*

-1.303*** -1.261*** -1.191***

Current personal economic situation

0.016

-0.062

-0.792*** -0.693***

Future personal economic situation

0.145

0.032

-0.611*** -0.517***

Current national economic situation

-0.050

-0.184

0.100

0.225

Trust in state’s institutions

0.035

Satisfaction with cabinet’s actions

0.088***

-0.083**

Satisfaction with market economy

0.073

-0.418**

2

Adjusted R 2

R change

Notes:

0.041

11.3%

12.6%

12.6%

13.8%

9.7%

12.4%

15.9%

16.5%

11.3%

1.3%

0.0%

1.2%

9.7%

2.6%

3.5%

1.2%

The coefficients in the table are unstandardized regression coefficients. Significance levels: *** p < 0.001 ** p < 0.010 * p < 0.050 + p < 0.100.

The next group of variables in the model makes a clear distinction between the supporters of the two models of market economy I analyze. The evaluations of the personal or national economic situation do not have any effect on the level of support for the liberal model of market economy. The support for the “pure” model of market economy is not influenced by how respondents evaluate the state of the economy. In the case of support for the social-democratic model of market economy, however, the evaluations of the 54 corectura 2 28 iulie 2008

Comrades or citizens?Support for democracy and market economy

personal economic situation play an important role: people who consider they are worse off now than a year ago show higher levels of support for state intervention on the market, and this effect is amplified if the same person expects its economic situation to get worse during the next year. Evaluations of the national economic situation do not have a significant effect on support for the social-democratic model. These results confirm both variants of the pocketbook theory (retrospective and prospective) but do not offer any evidence for the sociotropic theory. The last group of variables in the model measures the effect of other types of support on support for market economy. Satisfaction with the cabinet’s actions increases support for the liberal model of market economy and decreases support for the social-democratic model. The latter is also influenced by satisfaction with the functioning of market economy in Romania: the more satisfied a person is with how market economy is working, the more it rejects the social-democratic model and the opposite. Given that some of the main reasons for being dissatisfied with the transition to market economy include the increasing inequalities and the decreasing standard of living, one can understand why these factors generate favorable attitudes toward state intervention on the market. Overall, the results presented in Table 2 suggest that the differences in support for the two models of market economy are determined by the respondents’ experiences during the transition. The more a person lost during the transition, the more reasons it has to support the social-democratic model of market economy, in which the state should be responsible for individual welfare and should reduce income differences. If the economic situation in Romania will improve (and this scenario seems less a fantasy in 2007 than it seemed only years ago) and if this improvement will not be reflected only in macroeconomic indicators but also in everyday life, then it is likely that people will not feel such an acute need for the state’s corrective intervention on the market and will support the liberal model more and more.

Conclusions What are the answers to the questions stated at the beginning of this chapter? Are Romanians truly democratic citizens or are they only declaring themselves as such? As I have shown, most of the Romanians support the idea of a democratic political system in Romania, and the percentage of supporters has significantly increased between 1999 and 2005 (see Figure 1). Although many Romanians still support non-democratic alternatives, I believe this can be explained by the fact that these indicators capture not only support for democracy but also dissatisfaction with some problems that were common during the transition. It should be noted, however, that these problems are not 55

The Values of Romanians: 1993-2006. A sociological perspective

problems of democracy; they are rather the result of having an insufficiently developed political class. Much more disturbing is the tendency of more than half of the Romanians to have intolerant attitudes toward some minority groups in the Romanian society: gypsies and homosexuals. This result suggests that the transition from communist comrades to democratic citizens has not finished yet, and it is possible that this transition will end only when the generations that reached adulthood under the communist regime will be completely replaced by younger generations. On the other hand, even the most developed democracies have a small percentage of the population that is characterized by discrimination against minority groups. Of course, this does not make them morally acceptable. A society that considers itself to be democratic has a duty to fight for the elimination of these types of attitudes but, at the same time, has to acknowledge that complete elimination is impossible: there will always be people that are ignorant enough not to be able to abstain from discriminating on the basis of race or sexual preferences. Another interesting result is shown by Figure 3: Romanians define democracy in a wider sense than the definitions used in the literature. In addition to the standard attributes of a democracy, which are correctly identified, Romanians also include in the definition characteristics related to market economy (prosper economy and social protection) and a characteristic related to maintaining order (harsh sentences for criminals). It is very likely that the definition given to democracy is influenced by the serious problems the Romanian society faced during the transition: economic decline and increased criminality. Based on these results it can be argued that Romanians have already travelled a long way towards becoming democratic citizens, but they still have some way to go. I also believe, although I did not analyze relevant data in this chapter and I draw this conclusion based on personal observations, that, in 2005, Romania’s citizens were more democratic than its institutions. Fortunately, the last years have shown a tendency toward increasing the transparency and the responsibility of the state’s institutions, leading me to believe that, in time, these institutions will become as democratic as the citizens they serve. The results presented here suggest that although democracy and market economy are two distinct systems, the need for their simultaneous implementation has generated at the population level a complex of attitudes that links the two systems to one another: a person that supports democracy has a higher probability of supporting market economy as well. The regression analyses presented in this chapter lead to a series of interesting conclusions. First, it is evident that market economy is a much more controversial issue than democracy: if support for democracy is influenced only by the respondent’s income, support for market economy divides the population 56 corectura 2 28 iulie 2008

Comrades or citizens?Support for democracy and market economy

in multiple subgroups based not only on income, but also on gender, education, and residential area. It seems that market economy still has some way to go in order to gain the same level of support that democracy has. Moreover, there is not yet a clear consensus with respect to the type of market economy preferred by Romanians. Some support the liberal model, while others (and the results suggest they are primarily those who had more to suffer during the transition) prefer the social-democratic model, according to which the state has to intervene in order to correct the disequilibria generated by the market economy. It is possible that support for this model represents in fact a type of delay in adapting from the planned economy to the market economy, but the results presented in Figure 4 show that more and more people support it, suggesting that the explanation resides in the economic problems of the transition and not in the inadequate adaptation to the new economic system. According to these results, the new political and economic system seems to enjoy diffuse support coming from a large proportion of the population. If one also takes into account the fact that during the last several years Romania has become a country that is governed more and more democratically, and that the state is increasingly withdrawing from the economy, paving the way for a “real” market economy, it can be argued that the story of the Romanian transition, although marked at times by dramatic moments, is approaching a happy end.

References Almond, Gabriel and Sidney Verba. 1963. The Civic Culture: Political Attitudes and Democracy in Five Countries. Princeton: Princeton University Press. Anderson, Christopher and Yuliya Tverdova. 2001. „Winners, Losers, and Attitudes about Government in Contemporary Democracies.” International Political Science Review. 22 (4): 321-338. Bahry, Donna, Boaz, Cynthia and Stacy Gordon. 1997. „Tolerance, Transition, and Support for Civil Liberties in Russia.” Comparative Political Studies. 30 (4): 484-510. Barry, Brian. 1970. Economists, Sociologists, and Democracy. London: CollierMacmillan. Bunce, Valerie. 2003. „Rethinking Recent Democratization: Lessons from the Postcommunist Experience.” World Politics. 55 (2):167-192. Canache, Damarys, Mondak, Jeffrey and Mitchell Seligson. 2001. „Meaning and Measurement in Cross-National Research on Satisfaction with Democracy.” Public Opinion Quarterly. 65 (4): 506-528. Carothers, Thomas. 1996. Assessing Democracy Assistance: The Case of Romania. Washington, DC: Carnegie Endowment for International Peace. Centeno, Miguel Antonio. 1994. „Between Rocky Democracies and Hard Markets: Dilemmas of the Double Transition.” Annual Review of Sociology. 20: 125-147.

57

The Values of Romanians: 1993-2006. A sociological perspective Crawford, Beverly. 1995. „Post-Communist Political Economy: A Framework for the Analysis of Reform.” In Markets, States, and Democracy: The Political Economy of Post-Communist Transformation, edited by Beverly Crawford. Boulder: Westview Press. Dahl, Robert. 1989. Democracy and Its Critics. New Haven: Yale University Press. Dahl, Robert. 1996. The Future of Democratic Theory. Estudio/Working Paper 90. Madrid: Juan March Institute of Study and Research. Dahl, Robert. 1998. On Democracy. New Haven: Yale University Press. Dalton, Russell. 1994. „Communists and Democrats: Attitudes towards Democracy in the Two Germanys.” British Journal of Political Science. 24 (4): 469-493. Dalton, Russell and Martin Wattenberg. 2000. Parties without Partisans: Political Change in Advanced Industrial Democracies. Oxford: Oxford University Press. Diamond, Larry. 1992. „Economic Development and Democracy Reconsidered.” American Behavioral Scientist. 35 (4): 450-499. Diamond, Larry (ed.). 1993. Political Culture and Democracy in Developing Countries. Boulder: L. Rienner Publishers. di Palma, Giuseppe. 1990. To Craft Democracies: An Essay on Democratic Transition. Berkeley: University of California Press. Duch, Raymond. 1993. „Tolerating Economic Reform: Popular Support for Transition to a Free Market in the Republics of the Former Soviet Union.” American Political Science Review. 87 (3): 590-608. Easton, David. 1965. A Systems Analysis of Political Life. New York: Wiley. Easton, David. 1975. „A Re-Assessment of the Concept of Political Support.” British Journal of Political Science. 5 (4): 435-457. Eckstein, Harry. 1988. „A Culturalist Theory of Political Change.” American Political Science Review. 82 (3): 789-804. Finifter, Ada and Ellen Mickiewicz. 1992. „Redefining the Political System of the USSR: Mass Support for Political Change.” American Political Science Review. 86 (4): 857-874. Fiorina, Morris. 1981. Retrospective Voting in American National Elections. New Haven: Yale University Press. Firebaugh, Glenn and Dumitru Sandu. 1998. „Who Supports Marketization and Democratization in Post-Communist Romania?” Sociological Forum. 13 (3): 521-541. Fuchs, Dieter, Guidorossi, Giovanna and Palle Svensson. 1999. „Support for the Democratic System”. In Citizens and the State, edited by Hans-Dieter Klingeman and Dieter Fuchs. Oxford: Oxford University Press. Gabriel, Oscar and Jan van Deth. 1995. „Political Interest.” In The Impact of Values, edited by Jan van Deth and Elinor Scarbrough. Oxford: Oxford University Press. Gati, Charles. 1996. „If Not Democracy, What? Leaders, Laggards, and Losers in the Postcommunist World.” In Postcommunism: Four Perspectives, edited by Michael Mandelbaum. New York: Council on Foreign Relations. Gibson, James. 1996. „Political and Economic Markets: Changes in the Connections between Attitudes toward Political Democracy and a Market Economy within the Mass Culture of Russia and Ukraine”. The Journal of Politics. 58 (4): 954-984. 58 corectura 2 28 iulie 2008

Comrades or citizens?Support for democracy and market economy Gibson, James, Duch, Raymond and Kent Tedin. 1992. „Democratic Values and the Transformation of the Soviet Union.” Journal of Politics. 54 (2): 329-371. Haggard, Stephan and Robert Kaufman. 1995. The Political Economy of Democratic Transitions. Princeton: Princeton University Press. Hahn, Jeffrey. 1991. „Continuity and Change in Russian Political Culture.” British Journal of Political Science. 21 (4): 393-421. Hahn, Jeffrey. 1995. „Changes in Contemporary Russian Political Culture.” In Political Culture and Civil Society in Russia and the New States of Eurasia, edited by Vladimir Tismăneanu. Armonk, NY: M. E. Sharpe. Hall, John. 1995. „After the Vacuum: Post-Communism in the Light of Tocqueville.” In Markets, States, and Democracy: The Political Economy of Post-Communist Transformation, edited by Beverly Crawford. Boulder: Westview Press. Hofferbert, Richard and Hans-Dieter Klingemann. 2001. „Democracy and Its Discontents in Post-Wall Germany”. International Political Science Review. 22 (4): 363-378. Howard, Marc Morjé. 2003. The Weakness of Civil Society in Post-Communist Europe. Cambridge: Cambridge University Press. Huntington, Samuel. 1968. Political Order in Changing Societies. Cambridge: Harvard University Press. Huntington, Samuel. 1993. The Third Wave: Democratization in the Late Twentieth Century. Norman, OK: University of Oklahoma Press. Inglehart, Ronald. 1971. „The Silent Revolution in Europe: Intergenerational Change in Post-Industrial Societies.” American Political Science Review. 65 (4): 991-1017. Inglehart, Ronald. 1977. The Silent Revolution: Changing Values and Political Styles among Western Publics. Princeton: Princeton University Press. Inglehart, Ronald. 1981. „Post-Materialism in an Environment of Insecurity.” American Political Science Review. 75 (4): 880-900. Inglehart, Ronald. 1990. Culture Shift in Advanced Industrial Society. Princeton: Princeton University Press. Inglehart, Ronald and Paul Abramson. 1999. „Measuring Postmaterialism.” American Political Science Review. 93 (3): 665-677. Jacobs, Jörg, Müller, Olaf and Gert Pickel. 2003. „Persistence of the Democracies in Central and Eastern Europe: Consolidation, Stability, and People’s Power of Resisting.” In Political Culture in Post-Communist Europe: Attitudes in New Democracies, edited by Detlef Pollack et al. Aldershot: Ashgate. Kiewiet, Roderick. 1983. Macroeconomics & Micropolitics: The Electoral Effects of Economic Issues. Chicago: University of Chicago Press. Klingeman, Hans-Dieter. 1999. „Mapping Political Support in the 1990s: A Global Analysis”. In Critical Citizens: Global Support for Democratic Government, edited by Pippa Norris. Oxford: Oxford University Press. Kullberg, Judith and William Zimmerman. 1999. „Liberal Elites, Socialist Masses, and Problems of Russian Democracy.” World Politics. 51 (3): 323-358. Lewis-Beck, Michael. 1988. Economics and Elections: The Major Western Democracies. Ann Arbor: University of Michigan Press. Lipset, Seymour Martin. 1960. Political Man: The Social Bases of Politics. Garden City: Doubleday.

59

The Values of Romanians: 1993-2006. A sociological perspective MacKuen, Michael, Erikson, Robert and James Stimson. 1992. „Peasants or Bankers? The American Electorate and the U.S. Economy.” American Political Science Review. 86 (3): 597-611. Marraval, José María. 1997. Regimes, Politics and Markets: Democratization and Economic Change in Southern and Eastern Europe. Oxford: Oxford University Press. Martin, Irene. 2003. „Interest in Politics and the Political Culture Approach: The Case of the New Democracies of Southern and Eastern Europe.” In Political Culture in Post-Communist Europe: Attitudes in New Democracies, edited by Detlef Pollack et al. Aldershot: Ashgate. Mason, David and James Kluegel. 2000. „Introduction: Public Opinion and Political Change in the Postcommunist States.” In Marketing Democracy: Changing Opinion about Inequality and Politics in East Central Europe, edited by David Mason and James Kluegel. Lanham: Rowman & Littlefield Publishers. Mărginean, Ioan. 1999. „Suportul social pentru democraţie.” Sociologie Românească. 1 (2): 3-18. Mărginean, Ioan, Precupeţu, Iuliana and Marius Precupeţu. 2001. „România în cadrul celui de-al treilea val al democratizării.” Sociologie Românească. 3 (1-4): 20-34. McIntosh, Mary et al. 1994. „Publics Meet Market Democracy in Central and East Europe, 1991-1993.” Slavic Review. 53 (2): 483-512. Merkel, Wolfgang. 1996. Institutions and Democratic Consolidation in East Central Europe. Madrid: Juan March Institute of Study and Research. Merton, Robert. 1957. Social Theory and Social Structure. Glencoe: Free Press. Miller, Arthur. 1974. „Political Issues and Trust in Government: 1964-1970.” The American Political Science Review. 68 (3): 951-972. Miller, Arthur, Hesli, Vickie and William Reisinger. 1994. „Reassessing Mass Support for Political and Economic Change in the Former USSR.” American Political Science Review. 88 (2): 399-411. Miller, Arthur, Hesli, Vickie and William Reisinger. 1997. „Conceptions of Democracy among Mass and Elites in Post-Soviet Societies.” British Journal of Political Science. 27 (2): 157-190. Miller, William, White, Stephen and Paul Heywood. 1998. Values and Political Change in Postcommunist Europe. London: Macmillan Press Ltd. Mishler, William and Richard Rose. 2001. „Political Support for Incomplete Democracies: Realist vs. Idealist Theories and Measures.” International Political Science Review. 22 (4): 303-320. Mishler, William and Richard Rose. 2002. „Learning and Re-Learning Regime Support: The Dynamics of Post-Communist Regimes.” European Journal of Political Research 41 (1): 5-36. Morlino, Leonardo and Jose Montero. 1995. „Legitimacy and Democracy in Southern Europe.” In The Politics of Democratic Consolidation: Southern Europe in Comparative Perspective, edited by Richard Gunther, Nikiforos Diamandouros, and Hans-Jürgen Puhle. Baltimore: Johns Hopkins University Press. Nelson, Joan. 1995. „Linkages between Politics and Economics.” In Economic Reform and Democracy, edited by Larry Diamond and Mark Plattner. Baltimore: Johns Hopkins University Press. 60 corectura 2 28 iulie 2008

Comrades or citizens?Support for democracy and market economy Niedermayer, Oscar. 1990. The European Citizens’ Interest in Politics and Their Attitudes and Behavior Concerning the EC and European Integration. Reports of the Zentrum für Europäische Umfrageanalysen und Studien. 90 – 6. Mannheim: Zentrum für Europäische Umfrageanalysen und Studien. Norris, Pippa. 1999. Critical Citizens: Global Support for Democratic Government. Oxford: Oxford University Press. Norris, Pippa. 2002. Democratic Phoenix: Reinventing Political Activism. Cambridge: Cambridge University Press. Offe, Claus. 1997. Varieties of Transition: The East European and East German Experience. Cambridge: MIT Press. Pickel, Andreas and Helmut Wiesenthal. 1997. The Grand Experiment: Debating Shock Therapy, Transition Theory, and the East German Experience. Boulder: Westview Press. Plasser, Fritz, Ulram, Peter and Harald Waldrauch. 1998. Democratic Consolidation in East Central Europe. New York: St. Martin’s Press. Przeworski, Adam. 1986. „Some Problems in the Study of the Transitions to Democracy.” In Transitions from Authoritarian Rule: Comparative Perspectives, edited by Guillermo O’Donnell, Philippe Schmitter and Laurence Whitehead. Baltimore: Johns Hopkins University Press. Przeworski, Adam. 1991. Democracy and the Market: Political and Economic Reforms in Eastern Europe and Latin America. Cambridge: Cambridge University Press. Przeworski, Adam. 1993. „Economic Reforms, Public Opinion, and Political Institutions: Poland in the Eastern European Perspective.” In Economic Reforms in New Democracies: A Social Democratic Approach, edited by Luis Carlos Bresser Pereira, Jose Maria Maravall and Adam Przeworski. Cambridge: Cambridge University Press. Rose, Richard. 1992. „Escaping from Absolute Dissatisfaction: A Trial-and-Error Model of Change in Eastern Europe.” Journal of Theoretical Politics. 4 (4): 371-393. Rupnik, Jacques. 1997. „The Politics of Transition.” In Democracy in Romania: Assessment Mission Report. Stockholm: International IDEA. Schafer, Joe. 2000. NORM: Multiple Imputation of Incomplete Multivariate Data under a Normal Model, Version 2.03. Software available online at http://www.stat.psu.edu/~jls/misoftwa.html. Schopflin, George. 1994. „Postcommunism: The Problems of Democratic Construction.” Daedalus. 123 (3): 127-141. Simai, Mihály. 1999. „Issues and Experiences in the Practice of Democratization: Models and Paradigms.” In The Democratic Process and the Market: Challenges of the Transition, edited by Mihály Simai. New York: United Nations University Press. Sørensen, Georg. 1993. Democracy and Democratization: Processes and Prospects in a Changing World. Boulder: Westview Press. Swidler, Ann. 1986. „Culture in Action: Symbols and Strategies.” American Sociological Review. 51 (2): 273-286. Sztompka, Piotr. 1996. „Looking Back: The Year 1989 as a Cultural and Civilizational Break.” Communist and Post-Communist Studies. 29 (2): 115-129. 61

The Values of Romanians: 1993-2006. A sociological perspective Tismăneanu, Vladimir. 1998. Fantasies of Salvation: Democracy, Nationalism, and Myth in Post-Communist Europe. Princeton: Princeton University Press. Valenzuela, Samuel. 1992. „Democratic Consolidation in Post-Transitional Settings: Notion, Process, and Facilitating Conditions.” In Issues in Democratic Consolidation: The New South American Democracies in Comparative Perspective, edited by Scott Mainwaring, Guillermo O’Donnell and Samuel Valenzuela. Notre Dame: University of Notre Dame Press. van Deth, Jan. 1990. „Interest in Politics.” In Continuities in Political Action. A Longitudinal Study of Political Orientations in Three Western Democracies, edited by Kent Jennings and Jan van Deth. Berlin: de Gruyter. Verdery, Katherine. 1996. What Was Socialism, And What Comes Next? Princeton: Princeton University Press. Vitányi, Iván. 1999. „Issues and Experiences in the Practice of Democratization: Models and Paradigms.” In The Democratic Process and the Market: Challenges of the Transition, edited by Mihály Simai. New York: United Nations University Press. Voicu, Mălina. 2005. „Economic Efficiency or Ideology? Social Support for Democracy in Central and Eastern Europe.” In European Integration from East to East: Civil Society and Ethnic Minorities in a Changing World, edited by Horaţiu Rusu and Bogdan Voicu. Sibiu: Psihomedia. White, Stephen, Gill, Graeme and Darrell Slider. 1993. The Politics of Transition: Shaping a Post-Soviet Future. Cambridge: Cambridge University Press. Williamson, John. 1990. „What Washington Means by Policy Reform.” In Latin American Adjustment: How Much Has Happened? Edited by John Williamson. Washington, DC: Institute for International Economics. Williamson, John. 2000. „What Should the World Bank Think about the Washington Consensus?” The World Bank Research Observer. 15 (2): 251-264.

62 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases MIRCEA COMŞA

Fairly often we here media talk about left and right wing politics: “that party is moderate left; that candidate is right wing; that policy is left wing, and so on”. Gradually, in Romania too, the actors’ positioning on the left-right axis has become somewhat common practice, at least in the media or in the politician’s speeches1. But what do these labels mean for the common people? Do these concepts have a (well defined) content, a shared meaning? And, if so, how close are the laymen’s views and those of the specialists? What is the percentage of those that self-position on the left-right dimension and, related to them, what is the percentage of left-oriented, right-oriented or centre persons? Has the Romanians’ self-placement on the left-right axis changed over time? In comparison with the citizens of other countries (either former communist countries or from the EU), where are Romanians placed on the left-right axis? Is there any connection between self-placement and the individuals’ sociodemographic characteristics? What about a connection with values or preferences for the parties? These are the main questions that the current text tries to answer to. In the beginning we will analyze, at a theoretical level, some relevant aspects of self-placement on the left-right dimension.

The theoretic relevance of the left and right concepts In every society there are a few social cleavages (Lipset and Rokkan, 1967), some of them being predominant in each important historical period. Each of these cleavages sets its mark upon society’s systems, and, by extension, upon the political system shaping it at least by setting the relevant dimensions 1

This statement represents more of an intuition, an approximation, rather than an observation of some content analysis of the political programs or media speeches. 63

The Values of Romanians: 1993-2006. A sociological perspective

of the political conflicts. Some examples of possible dimensions are: left-right (Evans et al., 1996), materialist-postmaterialist (Inglehart and Abramson, 1994; Inglehart, 1990, 1997), libertarian-authoritarian (Kitschelt, 1992), liberty vs. equality (Noelle-Neumann, 1998) or “the new social movements” that mainly target aspects concerning European integration (pro-integration vs. antiintegration, positive integration vs. negative integration) (Hix, 1999; Scharf, 1996; Gabel, 1998), peace and ecology. In some authors’ opinion (Sani and Sartori, 1983), it is possible that, over time, most of these dimensions be incorporated in the left-right dimension, and, according to some analysis, this process is already happening (for example Noelle-Neumann, 1998). As we suggested earlier, the concepts of ideological cleavage and of ideological dimension refer to different things. And so, the cleavage is an opposition of values or interests in a society (for example, between the working class and the capitalists, between the State and the Church, between rural and urban), as opposed to the ideological dimension, that is a heuristic instrument used by politicians, citizens and analysts to organize and shape the political controversies and the behavior of political actors. The fact that most citizens are poorly informed in politics is the main assumption of most analyses. In this context, unfavorable to the correct functioning of the democratic system, ideology, or, to be exact, ideological labels are targeted onto two main goals, and these are: it facilitates political communication (a more brief and efficient way of political communication), and, closely connected, it helps individuals at making political decisions, at forming political opinions in insufficiently informed conditions. Therefore, the left and right ideological labels are tools that people use to orient in a complex political context. They work as a mechanism that helps one understand what happens in the political scene by reducing its complexity (Fuchs and Klingemann, 1990: 205; Knutsen, 1998b: 393). Even more so, the left-right dimension is a useful tool for the experts and politicians as well, by being a useful way in which they analyze and understand events (Gabel and Huber, 2000). But what happens to the political left or right? If in the specialized literature there is a high consensus regarding the fact that the left-right dimension is important for understanding the political ideology and attitudes (it is often used as an independent variable to explain partisanship, the position adopted in a controversy, voting behavior, etc), but regarding the contents of this dimension, a low consensus may be observed (Potter, 2001). With this aspect in mind, two main theoretical directions can be identified: one that gives this dimension a meaning (any meaning), and one according to which there is a variable content of this dimension (depending on the population and timeframe). It has been often thought that the left and right terms are connected to social and political specific conflicts (the class conflict), or to types of approaches to solving problems (state intervention vs. free market), or to certain 64 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases

moral values (equality vs. freedom). Surely, in certain times and contexts, the self-placement process on the left-right axis was closely connected to these conflicts, approaches and values (van der Eijk et al., 2006). Consequently, according to the first perspective, left and right politics can refer to ideal types of socialism and capitalism (Kitschelt, 1992: 12), to the distribution of material and economical goods (Inglehart, 1990, 1994), or to the opposition between equality and freedom (Noelle-Neumann, 1998: 329). Many occidental empirical studies make the distinction between two meanings associated with the left-right axis: one tied to economy and the other to the socio-political domain. From the economical perspective, the left refers to the equal distribution of resources and state intervention, and the right refers to the free market and to justified inequality. From the political perspective, to the left, the focus is on rights and social freedoms, and to the right, on order, discipline, tradition (Evans and Whitefield, 1998). Other analysis (Knutsen, 1995a) lead approximately to the same conclusions: the left is associated with traditional, secular values, to sociodemocracy and post-materialistic directions, whereas the right is associated to religious values, economic liberalism and post-materialistic directions. On the other hand, the meaning of the left and right terms is the collective result of the way in which the various social actors interact politically (what matters is their verbal and symbolic interactions) (van der Eijk et al., 2006: 181). Consequently we can not refer to a constant definition of these terms, as they are being continuously redefined by the social actors through interacting, as a consequence of new problems, conflicts and strategies. Briefly, the meaning of the left and right terms is politically built through the sustained process of cooperation and political conflict (van der Eijk et al., 2006). Consequently, according to the second perspective, the left and right concepts are devoid of all content, and given meaning by each particular individual (Knutsen, 1998b: 393). The political left and right are viewed as some kind of “empty containers ready to be filled with political content, which happens at different times and among different political and social groups” (Tarchi, 1995: 187, apud Knutsen, 1998a). Of course, the two perspectives are not self-exclusive, and it’s possible that a part of the meanings associated with the left and right concepts be relatively widespread (in both space and time), whereas the other part be contextually definable (in accordance to space, time and social class). Some of the analyses show that, even if common elements do exist, the meaning of the left and right political labels vary on more than one dimension: between countries, over time and in certain social classes. Thus, the meanings vary in accordance to age and political group (Inglehart, 1984; Fuchs and Klingemann, 1989); to the elderly the left is associated with the attitudes and the controversies of socioeconomic nature (social program support, class interests, the influence of syndicates), and the right with limiting the intervention of the government, support for the middle class and the influence of the business 65

The Values of Romanians: 1993-2006. A sociological perspective

sector. For the youth, the protection of the environment, social inequality and lifestyles are connected together with socio-economic interests, the left being associated with the opposition of nuclear energy, sex-equality support, prodisarming and the approval of social programs (Dalton, 2002). The content of the left and right concepts varies not only by country (as opposed to Europe, in the US the left is tied to liberalism and the right to conservatory politics), but even in the same country, by time period (especially for countries on the path of becoming democratic). Some analyses uphold this kind of variation. In Russia, the meanings of these labels were inverted during the 90s (with the passing from the communist regime to capitalism and an open market economy); the left came to be associated with free market, democracy and liberalism, and the right to the meaning of the communist and socialist regime (Evans and Whitefield, 1998). Thus, the associated meanings are flexible in times of political change, and fixed in times of stability. A similar conclusion obtained through a different kind of analysis (the analysis of party documents and the investigations of political elites) shows that the placement of parties and politics on this dimension at different moments in time or in different countries is subject to change (Gabel and Huber, 2000). In stable democratic systems, once set, the meanings associated with the left-right dimension become relatively persistent, as they are being sustained by multiple mechanisms, both institutional and informal (parties, syndicates, the media, the family unit, the local networks, etc) being propagated through these mediums as well (Evans and Whitefield, 1998). No matter the chosen perspective, if they are to be used as a means of obtaining knowledge, the ideological labels have to meet certain criteria. Often, the criteria mentioned by some authors are different in number and specificity. And so, Zechmeister (2006) proposes two generally formulated conditions: “first, they must reference relevant political divisions and actors and, second, they must be understood and used in consistent manners” while Todosijevic (2004) identifies five conditions that he formulates empirically2. The most often used arguments for supporting the relevance of self-placement in the left-right 2

These are the following: (c1) the elite uses the left-right dimension in a coherent manner (in agreement with the parties positioning on the axis; with a correlation between the preferred positions and politics); (c2) the public uses the left-right dimension in a coherent manner as well (the voters are to perceive the position of their party on the axis; a correspondence is to be between the perceived positions of the parties on the left-right axis, and their real positions); (c3) the public and the elites coherently use the dimension in the same way; (c4) the relevance of the dimension for voting (that is for a relation to exist between the preference in politics and voting; the public is to prefer the parties situated as close as possible to their own position on the axis); (c5) to be a congruence between the voters preferences in politics and those of the elites of the voted parties (the electors are to vote with the parties that promote the politics they prefer themselves).

66 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases

dimension are the usage of this direction by the population (the percentage of individuals that self-position) and the positioning of parties on the axis (the percentage of individuals that correctly position themselves on this axis) and by how “correct” one should understand “according to the expert opinion or to the party program”. Another possible criterion is the one of coherence between selfplacement and the voted party. The coherence position is valid when the voter prefers one of the parties situated on the axis closest to his own position (if another choice is made, a situation of incoherence appears) (Boy et al., 1997). There are two ways to find out the degree in which Romania’s population correctly uses the concepts of left and right, based on the available data (BOP, EVS & WVS). The first method is the identification of the meanings associated with these ideological labels through an open question that, even though limited3, can offer a direct representation of the aforementioned concepts. Another possibility is the analysis of the relationship between selfplacement and certain variables with which, according to the theory and to empirical analysis performed in other European countries, it should be associated, in a particular frame. The possible variation categories are those socio-demographical (gender, age, education, income, social status, job, living conditions), different attitudinal and axiological scales (economic liberalism, attitude towards democracy, communism, authoritarianism, nationalism, etc), the political actors (the party or leader vote, the party loyalty, the faith in the party and the leaders, etc). In a more general way, our approach follows the search for an answer to the question: “In what degree can we speak of stability and consistency in the belief and political values system in Romania?” To answer this question, we will analyze one by one, aspects of the left-right dimension usage, the meanings of the left and right terms, self-positioning on the left-right axis, and its social, axiological and partisan bases.

How the left and right concepts are used In Romania, the left-right opposition was, at least until 2000, scarcely used, as the communism-anticommunism, reformatory-anti-reformatory axis were preferred (Sandu, 1996, 1997; Bădescu and Sum, 2005). This was to be expected, given the particular context in which the Romanian parties appeared and developed (the slow breach from communism and the centralized

3

The percentage of those who answer the open questions is generally lower in comparison to the same questions, but with different possible answers (Comşa, 2003). What is more, in the case of certain population categories, the meanings may exist in a form that is difficult to put into words, and even at a level of low consciousness. 67

The Values of Romanians: 1993-2006. A sociological perspective

economy), as well as the early overall stage of democracy.4 Consequently, it is to be expected that the usage of the left-right dimension in describing and analyzing the Romanian political space be reduced, both with the political actors and with the population. Indeed, the notoriety of the left and right political concepts is relatively low with the Romanian adult population, according to BOP-FSD data (November 2003), where only half of the respondents said they had heard of these concepts (Figure 1). Figure 1. In politics, left and right are commonly used terms. Have you heard about these things?

NA 6% Yes 49%

No 45% Source: BOP-FSD, October 2003; NA means “no answering”.

Even though the notoriety of the labels is approximately 50%, the degree of those who self-position on the left-right axis is somewhat larger, varying between 50%-60%5. Thus, in 2006, 61% of the population had selfpositioned (51% in 2005), and 36% answered that they didn’t know where they

4

„ In young democracies, where the components of democratic politics are relatively new and often in flux, the meanings of ideological labels are likely to be less developed and evolving.” (Zechmeister, 2006). 5 The fact that the percentage of those who self-position is larger in comparison to those who have heard of these labels is not surprising and is found in other democracies as well (Lambert et al., 1986). The difference can be explained through the tendency of the individuals to respond even when they do not have an answer (the first question of notoriety acts as a filter) or through the memory anchor (the explicit connection of the left and right concepts with the position of some parties or with their own individual position). 68 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases

had placed themselves on the axis, and 3% didn’t answer at all (BOP-FSD, October 2006)6. Who are those that do self-position? Are they any different than those that don’t, and if yes, how? According to other empirical and theoretical analyses we expect that the two population groups would be very different, at least in what concerns the gender, age, education level, social class, living conditions, media exposure and interest towards politics. More to the point it is to be expected that this percentage is higher with men, and that it would rise along with all of the aforementioned factors. The analyzed data (Fig 2) confirm these differences7. Thus, men self-position in a percentage of 60% and women in a percentage of 44%; people having only middle school self-position in a percentage of 34% whereas those with higher education 68%; the people from the lower classes are at 36%, those from the middle classes are at 67%; the people not interested in politics are at 27%, whereas those with a high interest in politics are at 80%; those who use one information source at most are at 23% and those who use more than 5 are at 78%. The data show associations with the age (the elderly use self-positioning less) as well as with the type of locality (the percentage is higher in the urban areas), but these are mainly the result of some differences associated with urban – rural divide. Following from this difference in the structure of the population (and because of the different media access), the percentage of people who self-position is different in these two living environments. Similarly, the elderly are less educated and less interested and informed (they have lower access because of the rural living environment) and consequently they use the discussed dimension in a smaller extent.

6

In conformity to some analysis (Krosnick and Berent, 1990), the usage of the ordinal scale with possible answers labeled only in the extremes increases the percentage of the non-answers. Consequently, it is possible that part of those who do not self-position will have difficulties in this respect. Data shows that, even when clearly defined labels are offered (right, center-right, center, center-left, left), 43% of the subjects are not able to identify with one of the five offered ideological options, considering that the ideology is irrelevant (Mungiu-Pippidi, 2002) (the percentage of those who’s self-position grows by a small amount). On the other hand, when the ideological labels take the form of the main ideological orientations (social-democratic, liberalism, socialism, etc.), the percentage of those who are able to declare their attachments is larger. Thus, when looking at the closed questions (that offer possible answers), the percentage of those who feel they are near an ideological orientation is somewhere between 64-72% (MMT research, 2000). 7 All associations are statistically significant (p = 0.000). The intensity of associations (given the fact that the variables are nominal, the volume of the population constant and the number of degrees of freedom is relatively close, I used the contingent coefficient) is ordered by interest (0.37), information (0.29), education(0.26), subjective class (0.20), locality type (0.19), gender (0.15) and age (0.14). 69

The Values of Romanians: 1993-2006. A sociological perspective Figure 2. The variation in the percentage of individuals who self-position, depending on certain characteristics (%) 51

total

44

female

57

18-34 years

34

apprentices / unfinished college college / unfinished faculty

36

upper-middle class

52 76 80

high interest in politics

68

middle class

64 27

medium interest in politics

61

superior studies low class

46

little interest in politics

56

labor class / down-middle class

47

small urban

no interest in politics

42

55+ years

rural - center

big urban

57

35-54 years

at most comprehensive school

41

rural - peripferal

60

male

information sources 0-1 information sources 2-3

53

information sources 4-5

57 67

information sources 6-7

23 45 62 72

Source: Analysis based on data EVS&WVS 2005

The existence of the differences discussed earlier is not in itself a proof of a causal relation between the analyzed characteristics and the usage of ideological self-positioning. To achieve this, a multivariate dependent data analysis is necessary, as well as the ordering (before and after) of variables and relations in a coherent logical model, both theoretical and statistical. Such a model can present itself in the form of Figure 38.

8

The SEM model (structural equation modeling) built with AMOS 7. The values presented in the model are standard regression coefficients. In case of each dependent variable, the overall explained variance (R2) is marked in the upper right, in relation to the input field. All the variables have been measured at an ordinal level (4-5 categories), with the exception of the “the usage of the left-right axis” and gender variables, who both had 2 categories. The reference categories for the gender variable was “female” and for the dependent variable “usage”. 70 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases Figure 3. A model explaining the usage of the left-right dimension age -.30

-.05

.05

e4 .09

.14 .15

.30 e1

.06

education .46

-.10

.15 .34

.35

-.22

.06

.33

e6

-.07

gender -.09

political interest

.23 .29

-.06

L-R usage .07

subjective class e3 .14 .12

.29 .35

.36 income

.42

e2

.26

.08

information e5

.06 .08 locality type

GFI = 0,996; AGFI = 0,981; delta2 IFI = 0,994; CFI = 0,993; RMSEA = 0,035; pclose = 0,937

Source: Analyses based on EVS&WVS 2005 data. The values associated with the arrows represent the standardized coefficients; the upper right values, related to each dependent variable, show the total explained variance.

The model we constructed explains 23% of the variance of the dependent variable (the usage of the left-right scale), with the predictability principle represented by the interest for politics (0.34). Consequently, the capacity to use the left-right scale is dependent mainly on the level of interest for politics. Significant effects also appear in case of the education, location, sex, subjective class and information variables.9 Thus, middle class individuals from large cities, with a higher level of education and information have a larger capacity for ideological self-placement. Age and income both determine the capacity to self-placement, but to a smaller extent. If the effect of the political interest is nearly entirely direct, gender, education and type of locality influence more indirectly the usage of the left-right scale (approximately two thirds from 9

The hypothesis of the dependence of ideological self-placement on the level of education and the degree of political information (Inglehart and Klingemann, 1976) is confirmed by the data in Romania’s case too. 71

The Values of Romanians: 1993-2006. A sociological perspective

the total effect is indirect) (Table 1). The largest part of indirect influences of gender and education take place by means of interest towards politics (women and individuals with less education are less interested in politics, and therefore, a lower percentage of them are capable of self-positioning on the left-right axis). In the case of gender, a significant part of the influence is transmitted through variable information (men use more sources of information, and therefore a higher percentage of them manage to self-position). Similar conclusions appear in the case of other analysis referring to Romania as well (Comşa, 2006: 146-7). In these analysis too, differences in gender are maintained even when a series of socio-demographic variables are controlled. Different interest towards politics and the ability to use political concepts are the result of differences in gender socialization (on the other hand, the differences tend to decrease from a generation to the next one). Table 1. A model explaining the usage of the left-right dimension (the standardized effects significant for p=0.000) Effect type

Locality type

Sex (female)

Age

Educatio n

Individual income

Subjectiv e class

Interest in politics

Information

Direct Indire ct Total

0.06

-0.07

-0.05

0.06

0.00

0.07

0.34

0.12

0.11

-0.10

-0.04

0.15

0.04

0.04

0.01

0.00

0.17

-0.17

-0.09

0.21

0.04

0.11

0.35

0.12

Source: Analysis based on EVS&WVS 2005 data.

But does the percentage of individuals who self-position vary over time? Data show that we can talk rather of stability than about some dynamics of this percentage.10 Thus, the percentage of individuals who self-position is relatively stable in time after 1990, with a few exceptions (1993 and 1998) (Fig 4).11 It is very probable that the greatest part of the variation is caused by errors in the samples, by the somewhat different methodology in collecting data and by differences in style in data collection of the institutes that lead the public surveys.12

10

The same stability in self-positioning (although at higher rates) can be seen in the case of countries in Western Europe between 1973 and 1990 (Knutsen, 1998a). 11 In this research (MMT), following a similar methodology, the percentage of individuals who chose to self-position has been somewhat close to the same value (66% in 1999, and 57% in 2001). 12 These kinds of differences between institutions, given the conditions of similar methodologies and similar questions, have been observed in other analysis as well (Voicu, 1999; Sandu, 2004, in another internal analysis over BOP-FSD data). 72 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases Figure 4. The percentage of individuals that position on the left-right axis (1993-2006)

82

76 63

1993

1994

1995

1996

66

63

1997

55

1998

1999

61 51

2000

2001

2002

2003

2004

2005

2006

Source: Analysis based on BOP-FSD 2006 data; EVS&WVS 1993, 1998, 1999 and 2005; ICCV 1997; CSES 1996 and 2004. The figures represent the percentages of the total sample group.

Related to other countries, Romania places at the end of the hierarchy based on the percentage of individuals that self-position on the left-right axis (CSES, EVS&WVS).13 In comparison with other ex-communist countries (presently placed at similar levels with the capitalist countries), Romania is placed at the end (only the countries of the former USSR have the same low values) (Figure 5). What is interesting is the fact that, at the beginning of democracy, from all the ex-communist countries, the Czech Republic and Romania had some of the highest percentages of individuals that were capable of self-positioning on the left-right axis (EVS/WVS 1990-1993). If the Czech Republic holds this position in 2005, too, Romania is placed at the end of the hierarchy. If the 1993 Romanian data have been well collected (it is fairly probable that individuals with higher studies are overrepresented), by using the same methodology (sampling and interrogation), the tendency (in opposition to other countries) is difficult to explain. A possible explanation could be the fact that during the 1990-1992 periods, discussions took place with regard to different society models that Romania should follow (like Austria, Sweden, and Korea).

13

In Canada from the 80s, almost the same value could be observed, regarding both self-positioning and the usage of this dimension in the description of the political parties (Lambert et al., 1986). 73

The Values of Romanians: 1993-2006. A sociological perspective

87

92

95

Albania

Armenia

84

Croatia

Poland

80

Bosnia & Herzegovina

79

Georgia

78

Slovakia

77

Hungary

72

77

Macedonia

Estonia

69

76

Czech Republic

68

Serbia

67

Slovenia

67

Montenegro

Bulgaria

67

Latvia

67

Moldavia

67

Azerbaijan

64

Ukraine

Belarus

63

Russian Federation

55

62

Lithuania

52

Romania

Figure 5. The percentage of individuals capable of self-positioning on the left-right axis in the former communist countries

Data Source: EVS/WVS 1999-2000, and WVS 1995-1997 for Armenia, Azerbaijan, Georgia.

The clearer the public policies presented by the parties are, the highest the degree in which the individuals’ left-right axis orientation’s are structured (according to the social factors, controversies and policies). It is difficult to establish what the causal direction is: is this taking place because the parties’ policies are more clearly defined or because individuals have a more structured conception of the left and right concepts and therefore are able to better differentiate the parties in this dimension? It seems that the leading part is now the way in which parties present their position regarding certain policies (how structured these positions are) (Freire, 2006). It is highly probable that the low percentage of individuals that are capable of self-positioning on the left-right axis is determined by the fact that the opinion leaders (political individuals, journalists and political analysts) have used these terms too little or not at all in the post-December public space, preferring positional concepts more familiar to the public when explaining certain attitudes, events or public policies. This way of positioning includes the communist–ex-communist and reforming–antireforming dimensions, and more recently (starting mainly with the general elections of 2000), the corruption – anti-corruption dimension as well.

What can be understood by left and right? The fact that half of Romanians have heard of the concepts of political left and right or the fact that 60% of them self-position on the left-right axis does not mean that all of them know what these labels refer to, not even in a broad sense. Taking into account the 49% of Romanians who declared they have heard of the terms of political left and right, only three out of five are able to associate the labels with some content. Related to the entire population, this 74 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases

means 29%. 14 Consequently the percentage of individuals who cannot truly operate with this dimension is 71% (they either did not hear of the terms or, if they did, they are not able to provide even a minimal definition).15 What are the meanings the population associates with these terms? The semantic contents attributed by the population to the political left and right include aspects that generally go together from a theoretical-logical perspective, with a single significant exception (the undiscriminating superposition of the left-right axis over the power-opposition axis), and a few other small ones (left means wealthy, but also poor, which can be both true and false, in accordance with the meanings for poor and wealthy, as well as the living level of the individual; right means liberalism, but also nationalism, although this association is sometimes found in reality and theory, too). Figure 6. What do people understand by political left and right? other

left

7

1

conservatives

1

conservatives

poverty

1

corruption

conflicts

1

something negative

3

for rich persons

3

PNTCD

3

corruption

2

equality

2

PRM / Vadim

2

extremism / radicalism

3

democracy welfare / something positive social protection

Christians

4

2

4

democracy

9

extremism / nationalism / PRM

9

governing party

9

5 capitalism / market economy

11

liberalism

11

12

governing party

14

political opposition

14

political opposition welfare / something positive communism / socialism

right

7

other

13 15

31

Source: BOP-FSD, October 2003; % of those who provided a definition (29%)

14

The number is close enough to the one obtained in another research project (Mungiu-Pippidi, 2002). Reported to the entire adult population, in 1999, the percentage of individuals to be considered ideologically competent (the ones that have made correct associations with regard to the labels of political left and right) is estimated to be 25% at the most. 15 We do not believe that this situation is specific to Romania; it is very probable that this is the case in other countries too (Bulgaria, the countries of the former USSR); the values are relatively the same with the ones registered, under similar conditions, in Canada, during the 80s (40% of the electorate was able to provide a definition; Lambert et al., 1986). 75

The Values of Romanians: 1993-2006. A sociological perspective

As expected, the definitions for the concepts of political left and right are, for the most part if not entirely, antonymic, as belonging to opposing political currents. Thus, for the sample group, left means „communism / socialism” (31%) and right means „liberalism” (11%).16 As it can be seen, the ideological content is predominant, more in the case of the left concept. This happened because, in the case of the left, the form of political organization and that of economic organization are the same (sometimes not only at the level of the common people), whereas in case of the right concepts, the two are different, but connected aspects, with the same degree of importance. Of course, at the level of the popular definitions, there are some aspects that can be found in the vicinity of one of the two poles, as well of some common conceptual aspects. Between the contents that only relatively appear at one of the two poles, in case of the left concepts, there is an association with the social protection (12%) and equality (2%); in case of the right concept an association with capitalism and market economy (11%). With regard to the common notes, both the left and the right concepts are identified (sometimes in different degrees) with the opposition (14-13%), with the governing party / power (14-9%), with welfare (5-15%), democracy (4-9%), extremism (5-9%) and corruption (2-2%). If, in case of some attributes like welfare, extremism, corruption and democracy, the association is (or could be) correct and correspondent to reality, what is surprising is the high percentage (28% for the left and 22% for the right) of individuals that associate the two poles with the power and opposition (in almost the same degree). On the other hand, in relation to the moment of the data collecting process, almost a half of those that identify the concepts of power-opposition are right (indeed, PSD, the governing party in 2003, is a left oriented party and a large part of the opposition is right oriented). The meanings provided by the respondents for the left and right labels can be analyzed from another perspective as well. According to a somewhat spread typology, ideological labels can have three significant cognitive components: symbolic, of public policy and of valence (Zechmeister, 2006).17 The aspects connected to these components are: the symbolic component, which refers to a political group and even to an important actor (especially in the countries in which the political space is highly personalized); the aspect of public policy signals the preference for a certain opinion, related to a public policy; then, the valence signals the importance of a goal (the reduction of 16

In the 80s, the Canadians associated in a large extent the left with socialism and the right to conservatism and free initiative (Lambert et al., 1986). 17 To this, an affective component can be added. This refers to the attachment towards a certain label, the ideological label in this case (this can be used in a non-ideological manner, whereas individuals can react only to hearing the name of the label) (Jacoby, 2001). 76 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases

poverty, the eradication of corruption, social justice) and is characteristic to the new democratic systems, mainly during the elections (the actors communicate mostly the targeted goals and far less the means by which these goals will be attained; therefore, the final goals are underlined and not the political and public means through which they are achieved). By applying this typology to the left-right dimension, as Romanians understand it, some interesting conclusions emerge. The symbolic component is dominant (66% for the left and 50% for the right), followed by the valence component (16% and 23%) and only after the component of public politics (12% and 11%; the preference for social protection and market economy). In conclusion, for Romanians, the left-right axis is connected mainly to the political actors and political goals, and only secondly to the means by which these goals are achieved; the conclusion isn’t a surprise, given the low level of political information in the population and the way the political stage is, especially in what concerns electoral campaigns in Romania.

Left or right? Today, at most 6 out of 10 Romanians are capable of self-positioning on the left-right axis. Although it is expected that they understand different things to a certain extent, with regard to these labels (see previous data), an important part of the population that self-position, generally has a correct definition of the concepts; this definition is close to what the political science understands them to be. Consequently it is relatively fair to group positioning and to identify them with the proper labels. To simplify the analysis we have chosen to group the ten possible positions in groups of two, resulting in five categories of respondents. Related to the total number of individuals who selfposition, these elector categories have the following percentages: 9% for the left, 16% for the centre-left, 41% for the centre, 20% for centre-right and 15% for the right (BOP–FSD 2006). Related to this population, these percentages are much lower (due to the high percentage of those who do not self-position): 5% for the left, 10% for the centre-left, 25% for the centre, 12% for the centre right and for the right 9%. The meanings the population attribute to the left and right probably differ over time, but we believe that the differences are more or less significant.18 Therefore it is correct to compare the evolution of the percentages for the different orientations over time (Figure 7). 18

For Romania, the empirical data necessary to uphold this affirmation are lacking. The ones we did have access to (BOP-FSD and the research of Metro Media Transylvania) cover a small time period, (2000-2003), insufficient to sustain or infirm this affirmation. But with regard to this small timeframe, we observed no differences in the meanings attributed to the terms. 77

The Values of Romanians: 1993-2006. A sociological perspective Figure 7. The percentages of the ideological categories determined by the selfplacement on the left-right axis left

centre-left

2006

9

2005

9

10

1999

9

10

centre

16

1997

5

10

1993

4

15

centre-right

41 43

right

20

15

21

49 55 53

17 19

13 20 23

10 6

Source: Analyses based on BOP-FSD 2006; EVS&WVS 1993, 1997, 1999 and 2005 (the percentages were calculated related to the ones who expressed an option).

The data we presented show that, independent of the reference point, almost half of the population self-positioned in the centre, almost a third to the right and a fifth to the left (the percentages are calculated related to all the persons who self-positioned – most frequently, 60%; in order to see the percentages in the total number of the population, these values should be divided by two). Even if the changes that happened gradually are not significant, some tendencies can be noticed. With a high rate of probability19, we can say that, between 1996 and 2006, in Romania, we attend a process of ideological polarization. The existence of such a tendency is supported by noticing the fact that, if the percentage of the ones who self-position varies around 55 % of the population, the relative percent of the population who declare as “centre” lowers in time. It should also be added that the tendency that was noticed can be the result of two different processes (the result of both processes or the result of one of them). On one hand, it is possible that the meanings of the concepts of left and right become clear over time and, in this situation, choosing the centre variants signify the lack of a position in 1993 in comparison with 2006 (the hypothesis of ideological clarification). On the other hand, it is possible that the individuals, who, in 1993, were choosing deliberately the centre, have gradually modified their preferences and, now, a 19

The results of the regression equations defined by the values of every ideological orientation (dependent variable) and time (independent variable) support these tendencies. With the exception of the categories centre-left and centre-right, the other categories show a significant evolution from a statistical point of view in the mentioned direction (F test shows that they are significant and R2adj takes values between 75% and 85%). 78 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases

part of them orientate towards the right (the hypothesis of ideological polarization). In order to determine which one of these two hypotheses corresponds to a larger extent to the real situation, more analyses are needed. The answer to this question doesn’t represent an objective of this chapter. On the other hand, the process of ideological polarization is relatively slow and the percentage of the centre electorate is not much lower at the end of the interval in comparison with its beginning (41% in comparison with 53%). The tendency of polarization observed in Romania can also be noticed in other former communist countries20, too, but not in Western Europe21 as well. On this background, the percentages of left and right populations grow, but in a different way. Therefore, if the percentage of the “left-orientated” population seems to lower until 1997 and then to rise slightly, (the differences are not statistically significant), the percentage of the “right-orientated” population rises relatively constantly in time, but not a lot (in fact, what rises is the percentage of the “right-orientated” population and not the centre-right one). Figure 8. Self-placement on the left-right axis in the former communist countries (average on a 1 to 10 scale) left 5,5

5,5

5,5

Montenegro

Slovenia

Bosnia & Herzegovina

Hungary

Slovakia

Macedonia

Albania

Croatia

Poland

Armenia

Lithuania

Azerbaijan

Serbia

Ukraine

Moldavia

Belarus

Bulgaria

5,8

5,8

5,9

5,9

6,0

Georgia

5,4

Czech Republic

5,3

Estonia

5,3

Latvia

5,3

Romania

5,1

5,2

5,8

5,1

5,7

5,0

5,1

5,6

4,9

5,4

4,7

Russian Federation

right

Source: Analyses on the basis of the data provided by EVS/WVS 1995-1997 and 19992000 (values related to the ones who self-positioned).

20

In Russia, in conditions of the high stability of the self-placements on the axis (as percentages), a slight tendency of polarization is observed (as a result of the percentage decrease of the ones who cannot use these concepts) (Evans and Whitefield, 1998). 21 For 1973-1990 timeframe, in Western Europe, on the contrary, a rising tendency of the ones who self-position in the centre of the scale is observed. This tendency is explained through socio-demographic characteristics and political involvement. The rising of self-positioning in the centre is higher in the case of the persons with less education and in women’s case (but not among the persons who are the most politically uninvolved) (Knutsen, 1998b). 79

The Values of Romanians: 1993-2006. A sociological perspective

As we saw before, the meanings of the terms left and right show variations from one country to another. As a consequence, the comparisons among countries (especially among countries which, generally speaking, are very different from a cultural point of view and, particularly, they are different also in terms of political culture) are just partially right. Still, we present informatively a few figures and conclusions resulting from the analysis of some data provided by some international comparative researches (EVS & WVS 1995-1997 and 19992000). Although we would expect Romania’s place rather among the leftorientated countries (where most of the former communist countries place), the data show a different situation. Therefore, from a total of 78 countries (both EVS&WVS waves), Romania placed 55th with an average of the registered values of 5.8 (on a scale from 1 to 10, where 1 represents the left and 10, the right), that means that it is among the “centre-right” countries (equal to Finland, Iceland, New Zealand, Malta, Lithuania, Turkey and Bulgaria). Even if we take into account only the communist countries, Romania places rather to the right, together with Bulgaria, Lithonia, Estonia and the Czech Republic (Figure 8). We should also take into account the fact that in the years following exactly after the fall of the communist system, Romania’s position was almost the same. Among the countries that participated in the 1989-1993 wave of the EVS/WVS, only Romania, the Czech Republic and Poland had electorates with an average of self-position above 5.5, which is rather an orientation of the right. In Romania’s case, it is very likely that the average of a rather right selfplacement be, at least partially, the combined result of the low percentage of the ones who self-position and of the differences in self-placement depending on education (more educated persons are able to self-position in a larger extent and they self-position, on average, to the right). Besides this, in Romania, right-wing self-defining seems to be desirable (Chiribucă, 1996). Consequently, the data show a population more oriented to the right than in reality.

The relevance of self-placement on the left-right axis Does the fact that the voters prefer self-placement in the centre, left or right have any relevance? Or, to express it differently, is there any connection between self-placement on the left-right axis and the respondents’ sociodemographic characteristics? And is there any connection between selfplacement and the casting of a vote for a party or between self-placement and the attitudinal-axiological orientations? In the following, we will investigate the relation between self-placement and these aspects, using the data provided by BOP-FSD and EVS&WVS research. For the beginning, it is necessary to present a few theoretical aspects and empirical results related to our topic. In the specialized literature, there is an agreement regarding the distinction between the three components (or fundaments) of self-placement on the left-right axis: the social component, the value component and the partisan 80 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases

component (Inglehart and Klingemann, 1976). The social component refers to the localization of individuals in the social structure and to the correspondent social identities; the value component (the controversy and problems issue) refers to the citizens’ attitudes towards the current conflicts (socioeconomic, religious, “the new policy”) in the Western society; the partisan component refers to the identification with a specific party. Related to the three components, Inglehart’s and Klingemann’s reference study (1976: 264-9) showed that the partisan dimension is dominant22. Most of the analyses realized by them were bivariate. As a consequence of the results obtained, the social dimension (“social bases”) of self-placement was ignored in the previous studies and most of the analyses was limited only to the partisan and valuebased component; similar conclusions were reached when the analyses were renewed. As a result of the analysis of the data provided by the inquiry realized in five countries for a period of approximately 20 years (van der Eijk et al., 2006: 172) the conclusion is that the association between the social structure (this time, measured by the following indicators: church attendance and being a trade union member) and self-placement is little (around 0.10). At the same time, the association between the vote for a party and self-placement is consistently bigger than the association between self-placement and social position (0, 30) (van der Eijk et al., 2006). Of course, these are informational values at the level of many elections and for many countries. Meanwhile, the association with social bases seems to remain constant and the one with the partisanship seems to lower. Generally, the differences among countries are relatively big. As it often happens in social sciences, the initial analyses started from a simple idea, namely one of the differences occurring in the operationalization of the concepts.23 Therefore, Freire (2006) shows in a longitudinal comparative analysis (countries in EU, 1976-2002) that the social dimension is still important. According to him, the previous analyses reached the same conclusions due to some problems of specifying the models, more exactly due to the way in which the social dimension was measured and operationalized. In the same author’s opinion, the social component of self-placement is not composed of the same dimension (the structural one), but of three dimensions: structural, organizational and subjective. A synthetic presentation of the 22

According to the Inglehart and Klingemann analysis (1976), in the Europeans’ case, the feeling of belonging to the left or to the right reflects more toward the affiliation to a party and less concerning the preference for a certain policy or certain values. 23 An illustrating example is the following: in the framework of an analysis related to the function of associative membership in shaping democratic civic and political attitudes or, more generally speaking, of the social capital, a modification of the operationalization of the associative membership (taking into account the former membership) triggers different conclusions (associative membership has an effect over democratic attitudes), concordant with the theoretical expectations (Hooghe, 2003). 81

The Values of Romanians: 1993-2006. A sociological perspective

operational definition realized by Freire can be represented as it follows (the indicators used for measuring the social component and self-placement are positioned at the crossroads of the rows and columns): Table 2. The operational model of the social component of self-placement Dimension

Class

Religious

Structural

Occupational status

How often one goes to church

Organizational

Member of trade unions

Member of organizations / religious associations

Subjective

Confidence in trade unions Confidence in big companies

Confidence in Church

Source: Cf. Freire, 2006.

More recent analyses (Freire, 2006) show that when the social bases are defined only in accordance with the structural dimension (status and the frequency people go church), the explained variance is really low (most often, below 10%).24 But, when the social bases are defined through the three dimensions, the social component of self-placement is bigger (comparatively with the partisan and axiological components) in 11 out of 24 analyses25. What happens meanwhile with the relation between self-placement and the socio-demographic variables or between the electoral options and the value orientations? The gradual effect of the modernization process is the weakening of the relation between the social structure and the preferences, orientations and political behavior (van der Eijk et al, 2006: 167). Consequently, it is expected that the intensity of the relation between the social structure and the political ground is lower. Empirical analyses realized by different authors show different tendencies from a certain point of view. Therefore, in some analyses, the variance explained by the social structure stays relatively constant (van der Eijk et al, 2006: 172) or lowers in most of the Western European countries (Freire, 2006). Also, it has to be added that, in both analyses, social bases were defined 24

The conclusions are similar with the ones obtained related to previous periods by Inglehart and Klingemann (1976) or van der Eijk (2006). 25 In total, 24 models of multivariate analysis have been built (for each year and country), and the explained variance is larger than 49% for 10 of these models. The identification with a party was replaced in the analysis with the voting option (the first was not part of the questionnaire) and the non-voters have been excluded, which leads to the overestimation of the partisan component. Even in these conditions, in 11 of the cases, the social component was the most important element of self-positioning (in 9 of the cases, the partisan component, and in the other 3 the value component) (Freire, 2006). 82 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases

only through the structural dimension. When the social bases included both organizational and subjective dimensions (Freire, 2006), the variance explained stood relatively constant in time (4 states), lowered (5 states) or rose (2 states), but as an average tendency we may talk about a decrease (erosion of the social determinacy of ideological self-placement). Different conclusions appear in what concerns the effects due to the cohorts: they do not exist (van der Eijk et al., 2006) or, on the contrary, it has been observed a difference between the people born before and after 194526 (Freire, 2006). In what concerns the intensity of the relation between the vote for the parties and self-placement, it has been observed that there has been a gradual decrease at the level of the whole electorate, but this happened on the background of different variations (van der Eijk et al, 2006). Following a simple logic, the perspective presented before can be continued, starting from the same remark: the size of the ideological component is connected to the socio-economic development. According to some analyses, (Dalton, 1988) the size of the ideological component of the left-right dimension is connected to the process of cognitive mobilization. By cognitive mobilization it is understood that there is a process through which the voters come “to possess the political resources and abilities necessary for them to be selfsufficient related to politics” (Dalton, 1988: 18). Briefly, the situation can be described as the following. In typical industrial societies, the voters are characterized by low levels of cognitive mobilization. Middle voters in these societies are little educated and, when he/she encounters the complexity of politics, they base their loyalty mainly on the affiliation to specific social or party organizations (the so-called external mobilization). As a consequence, self-placement on left-right axis is made depending on the identification with a party and the vote is cast for a party in accordance with the affiliation to a specific social group. At the same time with the rise of the educational level (especially of the number of people with superior studies), citizens become more sophisticated from a political point of view and the rise of the access to mass-media makes them even more informed. As a consequence of these transformations, in advanced industrial societies, self-placement and selection of a party are based more on value orientations (this component is more important in these societies) (Knutsen, 1997). A series of analyses (Knutsen, 1997) realized at a bivariate and at a multivariate level at the same time, using data provided by Eurobarometer 16 and EVS II, supported this hypothesis. In this way, the variance explained (as part of a variance analysis model with selfplacement as a dependent variable) by the common component of partisan and axiological dimensions rises between 1980-1990 and, in an interval of only 10 years, it becomes almost as important as the partisan component. At the same 26

In most of the countries, the explanatory power of the social component is bigger in the case of persons born after 1945. 83

The Values of Romanians: 1993-2006. A sociological perspective

time, analyses show a rise of the value-based component (but the rise is little). Generally, within the total of the explained variance, 40-50 % reverts to each of the partisan or combined components, and 10-20 % reverts to the value-based components (Knutsen, 1997). The results of the analyses presented before differ significantly. Mainly, the differences are the result of the different way in which the social component and the type of statistical analysis were operationalized. Therefore, if we include all the three dimensions of the social component, its explanatory power rises significantly and in many cases it surpasses the one of partisanship. At the same time, the differences among the models are connected with the type of statistical analysis (regression, respectively, variance analysis), but especially with the method of specifying the models. The three factors (social, partisan, axiological) can have independent influences over self-placement or we can consider that values and partisanship have a common component (common variance explained). If this common component is attributed to values, on the basis of the assumption that these are previous to partisanship, the variance explained by the value-based component comes to be equal to that of the partisan component (Knutsen, 1997). Consequently, in accordance with the method that was chosen, the empirical results differ and, at a general level, the implications will also differ. If self-placement is less connected with social position and more with the electoral options, this shows not the irrelevance of left and right concepts, but the fact that the political system is, to a certain extent, independent and that political phenomena can be explained through other political phenomena (van der Eijk et al, 2006)27. In this case, self-placement doesn’t represent a simple reflection of the social position or, more generally said, the political phenomena are not a derivation of other phenomena from other spheres (social structure, modernization etc.), but have a high degree of autonomy. The discussion related to the absolute or relative importance of the social basis of ideological selfplacement has an implication at a more general level. The extent to which ideological self-placement represents or does not represent a reflection of the social position is tightly connected with the debate about autonomy, respectively about the systems in a society (or even more than that, the prevalence of the social system, in our case, over the other systems). Theoretical and empirical differences related to the self-placement bases impose the reconstruction of the analyses and interpretations with other data, related to other spaces and political times. This is our main objective in what follows.

27

Even if, at the beginning, the situation was not like this, in the end, either the political conflicts that led to these cleavages were solved, or the people got used to them (van der Eijk et al, 2006). 84 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases

Social bases of self-placement In accordance with the perspective of sociological determination, we expect that we should find differences in what concerns self-placement on the leftright axis, differences (also) depending on the socio-demographic characteristics of our subjects. The data we show (Figure 9) demonstrate what follows: ™ Statistically, self-placement differs significantly among the categories of the most variables, but the differences are not big28; ™ There are no self-placement differences depending on gender; ™ The older persons (more than 55 years old) are, on average, more leftorientated individuals; ™ The more educated persons are, on average, more right-orientated; ™ Persons who subjectively place themselves in a superior class are, on average, more right-orientated; ™ Persons with higher salaries self-position more to the right; ™ Persons living in bigger localities self-position easier to the right (also because of the differences in the population structure); ™ Related to their status, only retired persons tend to self-position to the left; ™ Related to the persons who have a job, a rise was noticed of the right selfplacement together with the rise of the occupational status; ™ Associative affiliation – trade union or professional – is accompanied by self-placement to the right; ™ The persons who have more confidence in trade unions and companies are rather right-oriented; ™ Church attendance is accompanied by the self-placement to the left; ™ There are no differences of self-placement depending on the affiliation to a religious association; ™ Persons with low confidence in the church tend to self-position to the right. Even if there are some self-placement differences among the different categories of population on the left-right axis, very often these differences are little (on a 1-10 scale, their values are below 1). In order to see to what extent socio-demographic characteristics explain self-placement differences, we need to analyze simultaneously the relations among all these variables. For this, we realized a regression model with categorical29 variables, where the dependent variable is the self-placement and the independent variables are all the socio28

Testing the significance of differences was done with chi2 (the association analysis) and also with F (the variance analysis). 29 In SPSS, “optional scaling” from regression; through procedure, the level of variables measuring was defined as nominal. Consequently, we considered that the relations among variables can take any shape, which means that total explained variance is maximal (if we considered the relations are linear, total explained variance would have been significantly smaller). 85

The Values of Romanians: 1993-2006. A sociological perspective

demographic variables we had presented before. The total variance explained by this model is low (6%30), which shows that the connection between the sociodemographic variables and self-placement is weak (the last one depends a little on the first ones). With the exception of the variables gender and trade union member/ professional association member, all the others variables have a statistically significant influence over self-placement on the left-right axis. Among these, the biggest effects occur in the case of the following variables: church attendance, affiliation to religious associations, confidence in church, in trade unions/companies, subjective class and residence. 31 Given the fact that the variance explained by the model is little, the portraits of the left or right oriented people should be looked at skeptically. Given all the other similar conditions, the persons who tend to self-position to the right are the ones less than 55 years old, belonging to the middle class, who seldom go to church. Figure 9. The variation of self-placement on the left-right axis in function of several socio-demographic, attitudinal and behavioral variables (averages on the scale 1=left to 10=right) total

6,0

employed - high

male

6,0

employed - medium

female

6,0

employed - low

18-34 years

6,2

retired person

6,1

housewife, stay-at-home

6,1

student/pupil

6,1

unemployed

6,2

35-54 years 55+ years

5,6

at most comprehensive schools

5,5

apprentices, unfinished college college, unfinished faculty

5,8 6,1 6,5

faculty, M.A., PhD low class

5,6

labor class

5,6

down part of the middle class

0-100 RON / pers.

201-400 RON / pers.

6,0

rural-center

no confidence in trade unions/comp.

5,8 6,1

very high confidence in trade unions/comp.

6,4

going to Church-once a year/not at all

6,3

going to Church-at festivals

5,9

rural-peripheral

6,7 5,9

6,2

5,7

101-200 RON / pers.

400+ RON / pers.

non-member of a trade union/prof. assoc.

high confidence in trade unions/comp.

6,6

5,8 5,6

trade union’s/prof. assoc. member

6,2

upper part of the middle class

6,6 6,2

6,3 5,7 5,9

going to Church-once a month

5,7

going to Church-once a week

5,8

member of a religious organization

6,1

non-member of a religious organization

6,0

low confidence in Church

small urban

6,1

high confidence in Church

big urban

6,1

very high confidence in Church

6,5 5,7 6,0

Source: Analyses based on the data of WVS 2005, Romania.

30

Adjusted R2. The statements are made on the basis of a measure of relative importance (“importance” in SPSS), proposed by Pratt (Meulman and Heiser, 2005). 31

86 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases

Left-oriented persons are mainly old persons, belonging to the middle or working class, who often go to church. In conclusion, even if the connection between self-placement on the left-right axis and socio-demographic characteristics can be found in Romania’s case, too, (the sense of the relations is similar), its total size is low. We bring into attention the fact that our model included more socio-demographic variables than the ones that usually appear in similar analyses (Inglehart and Kligemann, 1976; van der Eijk et al., 2006). At the same time, the model also considered some other dimensions of the social bases, not only the structural one (according to Freire’s analyses, 2006). Despite all these, the explained variance is lower than the one registered in the case of other countries, in the conditions of more simple explanatory models, using a smaller number of predictors32. And this happens when most authors consider that the connection between social bases and ideological self-placement tends to be lower in advanced industrial countries or, to express it differently, selfplacement depends less and less on the social or religious class characteristics. In the case of Romania, it is very unlikely for the explained variance to be the result of such a process and it is possible indeed that the lack of the relation be determined by the low levels of identification and ideological sophistication. A few arguments in this respect were presented before (only half of the Romanians self-position on the left-right axis, respectively only a third correctly understands these labels). The hypothesis of the dilution of self-placement social bases can also be tested in the case of Romania (the evolution of variance explained by the social characteristics). Besides this, given the serious transformations Romania passed through the last 16 years, it is necessary to investigate the evolution of the meanings of the relations; more precisely, which socio-demographic categories were left/right categories at a certain moment of time and which are these categories in the present33. For this we calculated and compared the averages of self-placement in the case of different categories of population for the interval 1993-2005 (EVS&VWS data; Appendixes: Table 12). With the exception of some punctual situations (as a consequence of the relatively different definition given to population categories), the data show a relatively big stability of selfplacement (the differences are not large). Generally speaking, the categories 32

The explained variance is approximately 10% if we take into account only the structural dimension, respectively 10-20% of all three dimensions are taken into account (Freire, 2006). 33 Such an approach may seem strange. But a few analyses (Evans and Whitefield, 1998) show that, in the case of the societies affected by deep changes, a situation also characterizing Romania, the sense of relations can invert. So, in the case of Russia, if, in 1993, the class the most right-oriented was composed of workers, and the most leftoriented one of entrepreneurs, in 1996, the situation is almost the other way round (Evans and Whitefield, 1998). 87

The Values of Romanians: 1993-2006. A sociological perspective

which, at a certain moment of time, tend to be of left/right orientations, keep this position irrespective of the reference year. For example, irrespective of year, the ones who place themselves in the labor class or lower class, selfposition to the right. Similarly, more educated people tend to self-position to the right and the persons who are less educated, to the left. In order to test the hypothesis of the dilution of the social bases of selfplacement, we built four regression models with categorical variables, using the same predictors of socio-demographic nature as much as we could34. The variance explained by these models is almost identical; the variations are very low and don’t follow a specific pattern: 7% (1993), 9% (1998), 8% (1999), and 6% (2005). Consequently, the hypothesis we treat about is not supported by data in the case of Romania, too. The predictors we find in more models and that have a bigger influence in comparison with the others are the socio-economic status (job, subjective class), education, residence, confidence in trade unions and companies, confidence in church and the frequency with which people go to church.

Partisan bases of self-placement Most analyses show that placement on the left-right axis represents one of the main factors determining European citizens’ political elections (Inglehart and Klingemann, 1967; Knutsen, 1997; van der Eijk et al., 2006) in what concerns the vote and other aspects at the same time (government performance, political controversies, leaders, etc.). This doesn’t mean that, irrespective of the country and of the period, there is a strong association between the preference for parties and placement on the left-right axis35. Do we find this relation in Romania’s case, too? Given the previous results and also the low importance attributed to this dimension36 by the population, the answer is most likely “yes”, but we expect the intensity of the association to be significantly lower in comparison with the average association observed in the case of Western European countries. Given the fact that party identification was not measured (generally, it is not measured on the data related to Romania) we used voting intention37 34

In the case of a few models, some predictors lacked (for example, the subjective class was measured just by two sets of data), but this affects the explained total variance very little (due to the strong connection between it and education or salary). 35 For example, the association between the vote and self-placement is very low in Ireland and very high in Denmark (van der Eijk et al., 2006). 36 Explicitly, 73% of the total population considers that the fact that a candidate is left or right doesn’t affect the vote (Mungiu-Pippidi, 2002). 37 This substitution often happens in case of other analyses, too (Knutsen, 1997; Freire, 2006), especially with data referring to European countries, from similar reasons. 88 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases

instead of it. In order to have a clearer image of the relation between the vote and self-placement we calculated the percentages offered to the parties in different types of electorates (Table 3) and also the percentage of the voters of a certain type inside the electorate of a party (Table 4). Also, we reduced the number self-placement categories from 10 to 538. Table 3. Vote for parties depending on self-placement on the left-right axis Party

Left

Centre Left

Centre

Centre Right

Right

Total

D.A. 29 38 33 29 13 23 Alliance PD 8 13 25 19 23 20 PNG 11 3 3 5 4 5 PRM 19 17 11 11 10 12 PSD 44 41 16 10 8 20 PNL 5 2 8 10 16 9 UDMR 0 1 6 6 3 4 Another 0 1 2 2 3 2 one Total 100 100 100 100 100 100 Source: Analyses based on BOP-FSD data in 2006. Way of interpreting: 13 % of the ones who self-positioned to the left vote with A.D.A. The cells on grey background indicate the highest frequencies for each column. Table 4. Self-placement on the left-right axis depending on the vote for parties Center Center Center Right Total Left Right D.A. Alliance 4 12 38 26 19 100 PD 4 10 48 19 20 100 PNG 23 10 29 23 16 100 PRM 15 21 34 17 13 100 PSD 20 32 32 10 7 100 PNL 5 3 38 22 31 100 UDMR 0 4 56 30 11 100 Another one 0 7 43 21 29 100 Total 9 15 39 20 17 100 Source: Analyses based on BOP-FSD data from 2006. Way of interpreting: 48 % of the ones who will vote with PD self-positioned to the centre. The cells on grey background indicate the highest frequencies for each row. Party

38

Left

1-2 = left, 3-4 = centre- left, 5-6 = centre, 7-8 = centre-right, 9-10 = right. 89

The Values of Romanians: 1993-2006. A sociological perspective

The data shown demonstrate that the electors who self-position to the left or in centre-left vote in a bigger extent to PSD, the ones in the centre with DA Alliance and PD (less with PSD), the ones in centre-right with DA Alliance and with PD and the ones who self-position to the right with DA Alliance, PD and PNL (Table 3). If we refer to the electorate of a certain party, the relation is the same clear (Table 4). PD Alliance, PD and PNL are voted by the electorate placed in the centre or to the right (with percentage differences), PSD by the electorate placed in the centre or to the left, UDMR by the electorate in the centre or in centre-right and, finally, PNG and PRM by all the electorates in almost the same extent (even if a little bit more by the centre electorate)39. If we use a synthetic measure of self-placement on the left-right axis (the average of the self-placement of a party’s voters), we can arrange political systems on this axis as it follows: Figure 10. Placement of the parties on the left-right axis on the basis of their voters’ self-placement 7.0 PNL 6.5 UDMR 6.4 D.A. Alliance 6.4 PD

5.9 5.8Total Not decided yet 5.5 PNG 5.4 Do not vote 5.3 PRM 4.5 PSD 1

2

3

left

4

5

6

center

7

8

9

10

right

Source: Analyses based on BOP-FSD data in 2006

Point 5.5 is the centre of the axis and current voters (the ones who indicated a party) place at 5.9, so the average of voters is placed rather to the right of the centre. Only one party clearly placed to the left is PSD, at the centre PRM and PNG and to the right, DA Alliance, PD, UDMR and PNL. The fact is significant that PD is placed much more closely to DA Alliance (the values are almost identical) in comparison with PNL, placed more to the right. The arrangement of the parties on the left-right axis on the basis of the average position of their voters seems to be the same one with the intuitive40 one built on the basis of self-placements made by the parties (through direct or indirect

39

The existence of differences is supported by the high values of standardized residuals. As far as we know, in Romania’s case, there is no analysis of the parties’ programs or of the leaders’ speeches from this perspective (positioning on left-right axis). 40

90 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases

statements, through party documents) or on the basis of the evaluations made by experts. To what extent is the placement of the parties stable over time?41 The stability of the parties positioning depends not only on the population’s perceptions, but also on the changes that take place on the political stage. The parties system in Romania suffered significant modifications (some parties disappeared, coalitions appeared, but they also suffered modifications along the reference period), which makes our analysis even more difficult. If we simplify our analysis a bit and we group the parties according to the affiliation to different coalitions and the “mother party” (the one from which they separated), the image of their positioning on left-right axis becomes much clearer (Appendixes: Table 13). This picture shows a relatively similar positioning of the parties for the analyzed period, which means that the meanings associated with the left and right labels aren’t inverted over time. What happened, during the same period, with the relation between ideological self-placement and the vote for parties? In order to answer this question, I used the same type of statistical analysis (univariate analysis of variance). A measure of the intensity of the relation is given by the explained variance (partial eta2). Given the fact that a big part of the electors do not express a voting option, I calculated this index relating to both all the electors (all the sample) or just to the voters (the ones who express a voting option) (Table 5).42 Generally speaking, almost a sixth part of the self-placement variable is explained by the variance of the vote variable. The percent of the explained variance is significantly bigger than the variance explained by the socio-demographic variables.43 The data we obtained seem to indicate a slightly descendant tendency of the explained variance by vote, even if the distribution oscillates with two decrease points (1997 and 2006). If we group the years, we come to a more simplified variant, which indicates a slight decrease of the variance explained by vote. 44

41

In Russia’s case, an inversion of the sense of the association (Evans and Whitefield, 1998) and implicitly, of positioning the parties on the left-right axis can be noticed. 42 This practice is often met in similar analyses and the explanations are very simple. On one hand, in Europe, its identification in rarely measured, and, on the other hand, the ignorance related to the ones who don’t have any vote option rises artificially the intensity of the relation between the two variables (as it can be seen in the case of this analysis, too). 43 It must not be forgotten the fact that a part of social influence over positioning can be transmitted immediately, through vote. 44 To what extent this tendency is a real one or rather represents the result of the differences in collecting data or in collecting the selection variations is a question to which the answer can be given after testing the relation on other data sets, too. 91

The Values of Romanians: 1993-2006. A sociological perspective Table 5. The intensity of the relation between self-placement and vote (explained variance) 45

Explained variance(%) Voters Electors Voters Electors

1993 18 14 18 14

1997 13 10

1998 20 17 17 13

1999 17 11

2005 18 14

2006 11 7 15 11

average 16 12 16 12

The average of the variance explained by partisanship (the vote for party) at the level of the EU countries is, generally, much higher in EU countries. Consequently, in the 80s, the average was 33% and, in the 90s; the average was 35% (Knutsen, 1997). Among the EU countries that we took into account, only in case of Belgium and Ireland low values of explained variance were registered, relatively close to what we obtained for Romania.

Attitudinal-axiological bases of self-placement Left and right are associated (in the specialized literature, in political actors’ programs, by common persons) with the preference for certain values and social policies. In this way, the left is associated with socialism and the right with capitalism, middle class, private property, liberty, etc. Of course, at the level of each individual these accents can differ, but we also expect that, statistically, these dimensions should partially overlap the left-right dimension.46 In order to test these relations in Romania’s case too, we used a set of items which measure the individuals’ position related to different controversies like the following ones: equality versus liberty, low versus high differences among salaries, state property versus private property, state’s responsibility versus individual’s responsibility, the efficiency of competition versus the inefficiency of competition, success being a matter of luck and relations versus a matter of work, etc. In order to illustrate the relation between these attitudes and self-placement on the left-right axis, we reduced the number of self-placement categories at five and we calculated the average of attitudes within this framework (Figure 11). Also, for making the interpretation easier, we inverted some attitudinal dimensions so that the dimensions indicating a left ideological orientation position to the left of the diagram and the ones that indicate right ideological orientation position to the right of the diagram. As one can notice from the graph, in the case of the most part of attitudinal dimensions, this tendency is present (in the case of the “work versus luck and relations” 45

Eta2parţial obtained through univariate analysis of variance. The data from Germany, but also from other countries show that there is a connection between the respondents’ position on the left-right axis and the preference for liberty versus equality (with the meaning of social justice) (Noelle-Neumann, 1998). 46

92 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases

dimension there cannot be any association noticed), without being very pronounced. Indeed, self-placement to the left can be associated with the preference for values, policies and beliefs like: social equality (either general, or particular, in the case of salaries), the conservation of the function of the state (at a general economic level or through social protection), the limitation of competition, social exploitation as a way of obtaining welfare, the fatalistic view over life. Reciprocally, persons with right affiliations are more oriented to freedom valorization, to higher differences among salaries valorization, to private property, to individual responsibility for one’s own welfare, to the positive role of competition, the rise of the level of richness for everyone and individuals’ self-determinacy. On the basis of the average size of the correlation indexes with a certain orientation (Table 6), we can state that, in the case of Romania’s population, left-right dimension is more connected to two main values: accepted social inequality (low differences versus high differences among salaries, the choice of equality in detriment of freedom) and the relation between the state and individual (the responsibility of welfare comes to the state or to the individual, the preference for private ownership versus state ownership). This conclusion is also sustained by the relative importance of the attitudinal dimensions that shape these values (Annexes: Table 14, Table 15). Has the direction of the association between these attitudes and selfplacement changed meanwhile?47 In order to analyze this relation over time, given the fact that the majority of the variables were measured on an ordinal scale with 10 steps, we used as a measure the correlation coefficient (Table 6). Even if related to a certain attitude, the size of the coefficients varies year by year (and sometimes is statistically insignificant), there is no situation in which the direction of the relation inverts. Consequently, we can state that the relation between self-placement and different attitudes implied by the left or right orientations has not inverted over time. But has the intensity of the relation between attitudes and selfplacement changed over time? Even if there were changes related to the intensity of the correlation between self-placement and a certain attitude or to the size of the average correlation, there is no clearly defined tendency in this respect. For 1993, there is an average correlation fairly higher which decreases during 1997 and 1999 and then remains constant. This observation is truly independent of the way in which we calculate the average correlation (all the items or just the common items).

47

In Russia, the direction of the correlations between the left-right scale and other scales (economic liberalism, political liberalism, nationalism, democracy, etc.) inverted between 1993 and 1996 (Evans and Whitefield, 1998). 93

The Values of Romanians: 1993-2006. A sociological perspective Figure 11. The relation between self-placement on the left-right axis and different value preferences Equality is more important than freedom Salary’s differences should be lower State property should expand and develop The State should assume more responsibility for everyone’s welfare Competition is a bad thing. It highlights the worst part in people. Success is mostly a matter of luck and relations People can become rich only on others’ account Everything in life is determined by destiny.

6.8

7.0

7.4

5.0

5.3

5.7

5.4

6.0

5.1

5.5

5.6

3.2

3.4

3.7

3.5

3.1

3.6

3.4

3.7

3.3

6.2

5.7

6.1

6.3

6,2

5,8

6,4

6,8

centre-left

centre

6.1

5.7

4.1

3.7

4.6

4.5

3.7

3.8

2.7

left

4.6

centre-right

7.2

7,3

Freedom is more important than equality Salary’s differences should be higher Private property should expand and develop Each individual should assume more responsibility for his/her own welfare Competition is good. It helps people to work harder and develop new ideas Long term, work usually makes a better life Welfare can rise for everyone Every person chooses his or her own destiny

right

Source: Analyses based on BOP-FSD data from 2006 * For each polar pair, on the rows, each respondent evaluated its position on a 10 point scale, where 1 means complete accord with left side statement, and 10 complete accord with the right one. For each point on the left-right dimension, the mean evaluation of the respondents from the respective class was computed and graphically represented in the figure. Way of interpreting: The respondents who self-positioned to the left are in a higher degree in accord with the statement „The State should assume more responsibility for everyone’s welfare”, and those who self-positioned to the right are in a higher degree in accord with the statement „Each individual should assume more responsibility for his/her own welfare” (the mean rises from the left to the right).

94 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases Table 6. The correlation between different attitudinal-axiological orientations and ideological self-placement Item

1993

1997

1998

1999

2005

materialism vs. post-materialism

0.23

-

0.10

(0.04)

(0.04)

-

0.10

preference for equality vs. freedom

0.20

0.05

-

-

0.09

0.18

0.13

-

-

-

0.05

-

0.05

-

0.06

-

(0.02)

-

0.04

0.16

0.27

0.19

0.10

0.21

0.21

(0.04)

0.24

(0.05)

(0.05)

0.12

0.13

(0.03)

0.16

0.09

0.11

0.22

0.13

0.07

0.09

(0.00

0.11

0.11

0.08

(0.04)

(0.03)

-

(0.05)

(0.04)

0.03

0.16

0.09

-

(0.00)

0.12

0.10

-

-

-

0.10

0.16

0.13

(0.05)

-

(0.06)

-

-

0.07

-

-

(0.02)

-

-

0.02

-

-

(0.01)

-

-

0.01

-

-

(0.01)

-

-

0.01

0.09

(-0.02)

0.11

-

-

0.08

salary should be equal for everyone economic problems should prevail over the ecologic problems it would be bad if the difference among 0.33 salaries rose state property should prevail over private 0.26 property responsibility for individual welfare should belong to the state vs. to the 0.17 individual competition is a bad thing vs. a good 0.12 thing success in life is a matter of luck and (relations vs. work 0.01) richness can rise only on others’ 0.11 account, cannot rise simultaneously for everyone the way in life is guided: by destiny vs. self-determinacy unemployed persons have the right to 0.11 refuse a job vs. they must accept any job public houses should be controlled by the state vs. have a greater freedom responsible for the pension is the state vs. the individual responsible for the residence is the state vs. the individual major changes should be made with 0.14 attention vs. firm action good ideas are old ideas vs. new ideas 0.12

2006 average

0.08

0.06

-

-

-

0.09

average of all the items

0.16

0.08

0.11

0.06

0.07

0.15

0.10

average of common items

0.22

0.08

0.19

0.08

0.09

0.17

0.14

average of all the items

0.16

0.08

0.11

0.12

average of common items

0.22

0.12

0.13

0.16

Source: Analyses based on BOP-FSD data from 2006; EVS&WVS 1993, 1997, 1998, 1999 and 2005 The sign "-" means that the item wasn’t asked in that year/survey. Parentheses indicate the fact that the relations are not significant (p>0.05). Calculating the averages was done also including some statistically not significant coefficients.

95

The Values of Romanians: 1993-2006. A sociological perspective

Even if in the case of average values of the correlation indexes, we observed rather a slight decrease tendency, it is possible that the variance explained by all attitudes simultaneously show a totally different outcome.48 For this we calculated the variance explained for every year, considering either all the items or the items present in all the research done (Table 7) Table 7. The intensity of the relation between self-positioning and aptitudes (explained variance) Explained variance (%)49 All the items Common items All the items Common items

1993

1997

1998

1999

2005

2006

average

15 13

3 2

11 13

5 4

3 2

9 7

8 7

15 13

6 6

6 5

9 8

Source: Analyses based on BOP–FSD 2006 data; EVS&WVS 1993, 1997, 1998, 1999 and 2005

The multivariate analysis shows the same tendency, which is the total explained variance falling over time. 50 Therefore, if we look just at the common items, the explained variance decreases from 13% in 1993 to 5% in 2005-6. As a conclusion, in relation with the attitudinal-axiological directions, we cannot speak of a clarification in the meaning of left and right concepts, but rather of an opposite process (de-crystallization). 51 The overall explained variance by value orientations related to Romania is placed at approximately half the average of the other UE countries. If in Romania a decrease tendency can be observed in the explained variance of the value orientations, in the other EU countries the situation is reverse. Thus, the

48

In the case of a similar analysis referring to Russia, the relation between left-right scale and other scales (economic liberalism, politic liberalism, nationalism, democracy, etc.) was tested. It was noticed that between 1993 and 1996 there was a rise of the correlation indexes. The same conclusion emerged through the multiple regression analysis (R2 rose) (Evans and Whitefield, 1998), but this concordance is not mandatory. 49 2 R adj was obtained by multiple linear regressions. 50 The data shows a relatively large variation from year to year regarding the total explained variance. It is more likely that the differences are the result of a fluctuation in the sampling, in the construction of the questionnaires (regarding the order and place of the questions, as the actual phrasing of the questions was nearly identical), or in the data gathering process. 51 The decrease of the explained variance over time shows that the left-right dimension has shared meanings, at the level of the population, only in a low degree (either less shared meanings, or the existing ones but to a lower extent). Consequently we cannot speak of a clarification of meaning or definition with regard to this dimension. 96 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases

average of the variance explained by value orientations (religious values,52 leftright materialist values, materialism-postmaterialism) at the level of the EU countries (Knutsen, 1997) was at 19% in 1981 and at 22% in 1990.

Ideological self-positioning: between party loyalty and values The individual analyses presented before have shown that, in the case of Romania, the self-positioning on the left-right axis doesn’t depend in a large degree on other factors. Thus, it depends to a small extent on the social structure (the explained variance is just 6-8%) and in a somewhat larger extent upon the party orientation (12-16%). In comparison with other EU countries, in Romania, the relation between self-positioning and the social structure is almost as strong, but this happens given the conditions in which the model built for Romania includes a much higher number of socio-demographic variables (Freire, 2006). The links between self-positioning and values, respectively between selfpositioning and voting standards are situated at approximately half of the average of the other EU countries. We point out the fact that the obtained values for the explained variance presented before have resulted from individual analyses (the models included one after the other all the socio-demographic variables, as well as voting habits and value orientations). To be presented with a clearer picture we should cross over the bivariate level to the multivariate level. 53 In this new context it is to be expected that the effects of party loyalty and axiological orientations is smaller because of the initial partial overlapping of the influences.54 Moving from the bivariate level to the multivariate level is not simple. The usage of regression analysis, no matter the type, doesn’t take into account the temporal order of the variables, which is an essential condition

52

As opposed to the EU situation, in Romania the religious values are not highly associated (depending on how they are used) with ideological self-positioning; for this reason they are not included in the multivariate analysis. Consequently, in a study (Knutsen, 1997) that took into account EVS data from 13 countries from Western Europe, it has been concluded that religious values (measured through the question: “How important is God in your life?”) correlate with self-positioning to a high degree (0.12-0.30). The correlation of the same variables in the case of Romania is 0.00 (EVS&WVS, 2005). 53 Some of the presented analysis is multivariate as well. What is being attempted here is to estimate the concurrent influence of the variables that are in relation to the party and axiological base of self-positioning; multivariate refers to the fact that we take into account more than one dimension (each with more than one indicator). 54 The value orientations and the voting options are associated, and the univariate analysis of variance shows that there is a significant interaction between them (the percentage of the explained variance is sometimes larger than the interaction between these components, over the value component). 97

The Values of Romanians: 1993-2006. A sociological perspective

for a causal explicative model (Bartle, 1998).55 If, in case of the sociodemographic variables56, it is somewhat clear that they represent the antecedents (preconditions) for the values and political option, the relation between these last two is less clear. Building on the assumption that the self-positioning variable is dependent,57 at least on the theoretical level, the relations between the three variables involved (values, voting options and self-placement, with the last being considered as dependent) may take different forms, two of which are synthesized in Figure12. In the case of model A, the relation between the axiological orientation and the party orientations is ambiguous, with some of the explained variance overlapping (there is a common part). If we introduce an ordering relation between these two variables, using the value orientation anterior to the voting, the relation changes into that of model B. In the case of this model, the common variance is attributed to the axiological orientations. In both models, it is considered that self-placement is a function of two parameters: the value orientation and the voting option. In other words, individuals with certain values, or individuals who prefer a certain party are consequently positioned, without necessarily having a clear picture of what this positioning entails (Inglehart and Klingemann, 1976).

55

If we want to explain self-positioning, we have to go by the condition of variables temporal ordering. If we are only after a prediction, following that condition is not necessary, therefore we may use regression models as well. 56 It may be added that, even among the socio-demographic variables there is a certain temporal order, and therefore their influence is not only direct but indirect as well (for example, the gender variable may influence self-positioning both directly or through the socio-economic status). 57 Theoretically, the relation between the voting option and self-positioning may be of two-ways: (1) “I” prefers the X party, where X is a left party, therefore “I” is positioned to the left, or (2) “I” orientation is for the left, X is a left party, therefore “I” votes with X. The bivariate analysis shows that the influence is indeed bi-directional, but also that the level of influence of voting is larger than the influence of self-positioning (if the dependent variable is vote, then Eta is 0.27, and if the dependent variable is selfpositioning, then Eta is 0.42; EVS&WVS, 2005). 98 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases Figure 12. The alternative causal models of value orientation variables, voting options and ideological self-positioning

A B Experimental classical design Hierarchical decomposition Ambiguous relation between Values before voting values and voting options Values

Values

Left-right selfplacement

Left-right selfplacement Vote

Unique value

Vote

Common component

Unique party

Value component

Party component

Source: After Knutsen, 1997.

But which of the two is of more importance in determining the position on the left-right axis? To answer that question, we built a series of univariate analysis of variance models and multiple linear regression models, and we compared the explained variance, taking each component into account (Table 8: A and B) (we made a decomposition of the total explained variance into explicit variances of each of the three components, two of them unique and one that is common and used for the interaction of the other two).58 To make the longitudinal interpretation of the data easier, and, at the same time, using an attenuation of the variations resulting from the differences in data collection methods, we built a simplified representation of this table as well, separating the data depending on the research year.59

58

The two tables, A and B, are the result of the two different methods used in the explained variance calculation regarding the three components (a partially different method was used in the calculation of the axiological component). 59 Given the relatively high variance of values from year to year (most probably because of the rather small samples on which these analyses have been performed and as a 99

The Values of Romanians: 1993-2006. A sociological perspective Table 8A. The link between voting options values and left-right self-positioning: the variance explained by the unique party component (vote), the unique value component and the compounded component A

Explained variance (%)

The decomposed explained variance (%)

The percentage of the explained variance (%)

Year

values (1)

voting (2)

total (3)

unique to the value (4)

common (5)

unique to the vote (6)

values (7)

common (8)

voting (9)

1993

15

12

20

8

7

5

38

36

26

1997

3

8

10

2

0

8

20

5

75

1998

11

16

22

7

4

11

30

19

51

1999

5

9

12

3

1

7

28

11

61

2005

3

13

15

2

1

12

11

7

82

2006

9

7

12

6

3

4

48

24

28

Average

7

11

15

5

3

8

29

17

54

1993

15

12

20

8

7

5

38

36

26

97-99

6

11

15

4

2

9

27

14

59

05-06

6

10

14

4

2

8

27

15

58

Average

9

11

16

5

4

7

32

23

45

Building this table included the following steps: (a) the explained variances of value is calculated (multiple linear regression, with self-positioning as the dependent variable and the axiological orientations as the independent variables); after that, the variances explained by vote intentions are calculated (using univariate analysis of variance, with self-positioning as the dependent variable and the voting option as the independent variable, were the undecided are included); the explained variances in both voting options and values are simultaneously calculated (using univariate analysis of variance, with self-positioning as the dependent variable and both voting options and axiological orientations as the independent variables); in the table, these variances are shown in the columns 1 to 3; (b) the variance is calculated for the two unique components and for the common one; for the unique component representing the voting option (column 6), a subtraction is made from the total variances of the value component (column 3 – column 1); for the unique axiological component (column 4) another subtraction is made from the total variance of the voting option component (column 3 – column 2); for the common component (column 5), the variance of the unique components (column 4 and column 6) is subtracted from the total variance; (c) the percentages of the explained variances for the three components are calculated (columns 7 through 9) by dividing the variance of each component by the total variance (columns 7 to 9, each divided by column 3). The items used for the measurement of the value component have been different year by year (only 4 of them are common, the ones with the higher degree of explanatory power). For a higher degree of certainty regarding data validity

further result of the small percentage of those who self-position or have a party option) it is more secure to work related to the average values of close years. 100 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases we have redone the analysis using only the common items and the obtained values are nearly identical (the explained variance of values has decreased to a very small extent). Table 8B. The link between voting options values and left-right self-positioning: the variance explained by the unique party component (vote), the unique value component and the compounded component B

2

2

R

Eta

The percentage of the explained variances (%)

partial

Year

total

total

values

interaction

voting

values

interaction

voting

1993 1997 1998 1999 2005 2006

16 10 19 9 15 13 14

21 15 24 14 18 16 18

7 0 8 0 2 8 4

5 5 2 5 3 2 4

9 10 14 8 13 6 10

33 3 32 3 12 48 22

22 32 10 36 16 12 21

45 65 58 61 73 40 57

16 13 14 14

21 18 17 19

7 3 5 5

5 4 2 4

9 11 10 10

33 13 30 25

22 26 14 21

45 61 56 54

Average

93 97-99 05-06 Average

The table presents in a synthetic manner, the results of a group of models used in univariate analysis of variance (SPSS Base 15.0 User’s Guide, 2006; Page et al, 2003; Lawrence et al, 2006). The dependent variable is self-positioning and the independent variables are value orientations (the covariate variable, built through factorial analysis of the common variables of value) and the voting options (the factor variable; included here are the non-voters, the undecided and the non-respondents); the interaction between the two was also included. Starting with the presupposition that the value orientations precede voting options, the used method for the calculation of the sum-ofsquared deviations was hierarchical decomposition (type I in SPSS), with the variable taking part of the model in the following order: values, values * voting, voting. The explained variances appearing in column R2 and Eta2partial differ (the latter is larger than the first) because of the fact that R2 is calculated based on the assumption that the relations are linear. Source: Analyses based on BOP–FSD 2006; EVS&WVS, 1993, 1997, 1998, 1999 and 2005 data.

The data shown here indicates that the self-positioning on the left-right axis is dependent only in a relatively small extent on the value orientations or on voting options (independent or not). Only in two situations (1993 and 1998) the total explained variance was above 20%, with the average on all the other years situated at 15-19% (depending on the measurement methods). From a series of 101

The Values of Romanians: 1993-2006. A sociological perspective

similar models created for other Western European countries (Knutsen, 1997), in the 1990s, only Ireland and Belgium had a total explained variance of similar proportions, with all the other countries rising to as much as 52%, and an average of 40%, double in comparison with Romania. In conclusion, in the case of Romania, the link between voting or values and self-positioning is much weaker. The explained variances of the three components are close enough, but the unique component representing the voting option is clearly the largest. Thus, the average explained variance of the unique component of voting options is 8% (10%), whereas that of the unique value component is at 5% and that of the common component at 3% (4%) (the values differ if the years are compounded). The relative percentage of the explained variance in each of these three components does not differ much in relation to the means of the calculation. Thus, in the case of variant A, the percentages are as follows (between parenthesis the B variant): the unique component of vote option 45% (54%), the unique component of values 32% (25%) and the common component 23% (21%). It can be therefore concluded that, in Romania, self-positioning depends more on the party preference and less on the values orientation. Even more, the influence mediated by the voting procedure of the values over selfpositioning is low too. In comparison with other European countries (Knutsen, 1997), in Romania the unique component of the voting option has a similar percentage (the average over 13 countries, in the 90s is 47%), with the percentage for the unique component of value a little higher (14%) and that for the common component a little lower (39%). As it can be seen in these countries, the influence mediated by the voting process over self-positioning which is comparable with that of voting itself. What follows from these observations is that the causal model that corresponds to Romania is largely similar to a type A model, whereas the B model fits more with Western European countries. Still, taking into account that we consider the values as being anterior to the voting options, the two factors determine in an approximately equal extent the value of self-positioning on the left-right axis; the situations is similar to the averages observed in case of other EU countries (Knutsen, 1997). What is happening to the size of the explained variance over time? In both calculation models, it can be observed that the total explained variance slowly decreases in time, with the fall being concentrated in the 1993-1999 period. Thus, if at the beginning of this period, the total explained variances were at 21% (20%), the average for the 1997 to1999 interval had decreased to 15% (18%), and at the end of the period being as low as 14% (17%). What’s more, this decrease is completely the result of the lowering of the unique and common value variances, given the fact that the variance of the unique component of voting option is constant or slowly growing. The observed tendency in the case of Romania cannot be found in the EU space (where the 102 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases

total explained variance is measured in the 1981-1990 period, with the average over 13 countries at approximately 36%; Knutsen, 1997). Another difference which can be observed is related to the variance of the three components percentages over time. Considering similar analysis (Knutsen, 1997), for the 1981-1990 period, in Western European countries a small growth can be observed in the unique value component and a decrease in the voting option component. What’s more, further observation shows that the percentage of the common component tends to grow when the development level of the country is more evolved (the two aspects have a strong correlation to one another). 60 As a result of these observations, we may expect that, in Romania’s case as well, the percentage of the common component would be relatively smaller at the beginning of the reference period, and also that it would grow over time. The analysis result shows a different tendency instead. The percentage of the common component is high enough at the beginning of the period and, contrary to what was believed, it decreases during the reference period, as opposed to the percentage of the voting option unique component that tends to go slowly up. On the other hand, the aforementioned tendencies take place in the 1993-1999 timeframe. After this period, the level of the explained variance of each of the three components remains approximately unchanged, or it tends to come back to the level it was at in 1993 (in relation to the years average and taking into account both methods of calculus). In conclusion, in comparison with the EU countries, in Romania the situation looks a little different, regarding the following aspects: (1) the total variance explained by voting options and values, related to self-positioning on the left-right axis is much lower in Romania (about half of the EU average); (2) the relative percentages of the three components in relation to the total explained variance differ: the unique components (voting options and value orientations) each has a larger percentage in Romania, whereas the common component has a lower percentage; (3) the absolute size of the unique voting component slowly increases over time, while the size of the common component decreases and (4) the percentage of the unique voting component increases over time, while the size of the common component decreases. What is the significance of this observation? Briefly, it concerns the motion between conflicts based on class and conflicts based on value orientations. If this tendency is present in Western countries (Knutsen, 1995b), in Romania, things seem to be just the opposite, at least in the 1993-1999 timeframe. 61 The cognitive mobility that characterizes advanced industrial

60

They are measured through the PIB indicators in relation to each individual and the percentage of the service sector. 61 They only “seem to be”, given the fact that for the beginning period of democracy in Romania, we have only one set of data available (1993). To be sure that was the state of 103

The Values of Romanians: 1993-2006. A sociological perspective

societies has, over time, determined their citizens to vote more and more by taking into account value orientations and also to self-position themselves on the left-right axis relating to both party preference and value orientations. In the case of Romania, the self-positioning process is carried out mainly related to voting options, without a strong relation between value orientations and voting options. On the other hand, if the common percentage is attributed to values, these tend to have a percentage similar to that of the voting option (directly or not).

Conclusions Romania’s population is characterized by an average level (related to the used scale) of ideological identification.62 Thus, only a little over a half of the sample group manages to self-position on the left-right axis. Related to other cultural environments, the level of identification is very low, and in most of the European countries, the percentage of individuals that are capable of selfpositioning on the left-right axis is significantly larger. Even though only a half of the sample group believe that they know the meaning of the concepts of political right and left, only a third are capable of attributing to these labels one of the meanings that is recognized to be correct. In comparison with other countries, the level of ideological sophistication63 of the Romanians is significantly lower. The level of ideological identification varies to a large extent with the different characteristics of the individual from the sample group. The usage of the left-right axis is larger among men, or among individuals with a higher education, a higher social status, a higher degree of political involvement or information. Although we may see a rise in the level of ideological identification over time, the data appear to be relatively stable. In comparison with other countries (ex-communist or not, with the exception of the countries of the former USSR), Romania is among the ones with the lowest level of ideological identification. The irrelevance of the left and right terms for approximately half of Romania’s population is mainly the result of their absence from the public vocabulary of the opinion leaders (political or nonpolitical) after 1989.

things at that moment, the replication of the analysis would be necessary using another set of data from that year or from a close year. 62 This signifies the usage of the left-right axis in the description of the political space, without necessarily attributing the same meanings to the terms (Fuchs and Klingemann, 1990: 205). 63 This signifies both the usage of the left-right axis as well as the usage of the same meanings for the terms (Fuchs and Klingemann, 1990: 205). 104 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases

The percentage of ideological and pragmatic contents (living levels) differs in the definition of the two poles. 64 Consequently, left means ideology to a larger extent (31%) in comparison with right (22%, if we add the two components: liberalism and market economy). In the case of the left concept, the political dimension dominates, whereas for the right, a differentiation is made between the economical dimension (market economy) and the political one (liberalism). The “final” objectives of the left (social protection) and of the right (generalized welfare) are identified in approximately the same degree (12%, respectively 15%). For Romanian citizens, the left and right concepts mainly target the political actors (the symbolic components), political goals (valence component) and only after these two, the means by which these goals are reached (the public political component). On average, Romanians population more likely places to the right (5.9 on a scale of 1 to 10). In the majority of European societies, the citizens position more to the left. No matter the reference period, about half of the population self-positions in the centre, about a third to the right and one fifth to the left (the percentages are calculated in relation to the complete number of individuals who self-position). In the 1993-2006 timeframe, we see in Romania an ideological polarization process taking place (the percentage centre-orientated decreases as the right percentage increases). The relation between the socio-demographic variable and selfpositioning is weak. Self-positioning on the left-right axis makes a significant statistical distinction between the categories of most socio-demographic variables, but the differences are not too large: there are no self-positioning differences relating to gender; individuals orientated towards the political right are, on average, younger and more educated, with higher incomes, living in larger communities and with a higher social status (the left has the opposite profile). The data shows a large stability in self-positioning related to different socio-demographic categories; generally the categories that are, at some point in time, more right/left orientated, keep their relative positioning over time. Based on the self-positioning of voters, the only party clearly placed to the left is the PSD, with PRM and PNG in the centre and the DA Alliance, PD, UDMR and PNL to the right. The dispersion of the parties on the left-right axis based on the average positions of voters seems to be in conformity with the intuitive placement based on party self-positioning (using direct or indirect affirmations or through party documents) or on expert evaluation. Also, the data show a similar relative positioning of parties in the analyzed period, which means that the meanings associated with the left and right labels have not been reversed over time.

64

The percentages are computed by reporting the total number of individuals that answered with a definition. 105

The Values of Romanians: 1993-2006. A sociological perspective

Self-positioning on the left-right axis is more tightly connected with two main values: the accepted social inequality (small differences vs. large income differences, the option for equality in the detriment of freedom) and the relation between the state and the person (responsibility for the welfare falling on the state vs. on the individual, preference for state vs. private propriety). The relation between self-positioning and the different attitudinal-axiological orientations have kept its meaning over time. On the other hand, related to the attitudinal-axiological orientations, we cannot talk about a clarification of the meanings attributed to the left and right concepts (rather we are witnesses to a reverse process of ambiguity). Self-placement depends only to a small extent on the social structure (the explained variance is only 6-8%) and in a somewhat larger extent on the axiological orientations (8-9%) or on party options (12-16%). In comparison with other EU countries, in Romania, the relation between the social structure and self-positioning is nearly as strong, and the relations between selfpositioning and value orientations, respectively between self-positioning and voting options is situated at approximately half of the average of these other countries. The relation between self-positioning on the left-right axis and value orientations and voting options (independent or not) is relatively small. Only in two situations (1993 and 1998) the total explained variance rises above 20%, with the average over all these years amounting to 15-19%. The total variance is to be distributed into three components, as follows: 8-10% the unique voting component, 5% the unique value component and 3-4% the common component. If we believe that values precede the votes, then the explained variance in voting is nearly equal to the explained variance in values (the effects are summed up both directly and indirectly). In conclusion, in Romania, the ideological self-positioning works as a heuristic tool only for a small percentage of individuals. The main arguments for this are the low level of notoriety and usage of the left and right concepts, as well as the weak connection between self-positioning and the social bases, both party related and axiological.

106 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases

Annexes Table 9. Total standardized effects of the explanatory model of the left-right dimension usage Total effects Education Salary per person Subjective class Interest in politics Information Usage of the dimension

Salary per Subjective person class 0.00 0.00

Interest in politics Information 0.00 0.00

Type of locality 0.42

Sex (feminine) -0.10

Age -0.30

Education 0.00

0.45

-0.05

0.01

0.46

0.00

0.00

0.00

0.00

0.27

-0.05

-0.10

0.46

0.29

0.00

0.00

0.00

0.08

-0.24

0.00

0.18

0.02

0.07

0.00

0.00

0.32

-0.14

-0.12

0.48

0.13

0.15

0.15

0.00

0.17

-0.17

-0.09

0.21

0.04

0.11

0.35

0.12

Source: Analyses based in EVS&WVS data from 2005. Table 10. Direct standardized effects of the explanatory model of the left-right dimension usage Direct Effects Education Salary per person Subjective class Interest in politics Information Usage of the dimension

Salary per Subjective person class 0.00 0.00

Interest in politics Information 0.00 0.00

Type of locality 0.42

Sex (feminine) -0.10

Age -0.30

Education 0.00

0.26

0.00

0.15

0.46

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.33

0.29

0.00

0.00

0.00

0.00

-0.22

0.05

0.15

0.00

0.07

0.00

0.00

0.08

-0.06

0.00

0.35

0.08

0.14

0.15

0.00

0.06

-0.07

-0.05

0.06

0.00

0.07

0.34

0.12

Source: Analyses based on EVS&WVS data from 2005. Table 11. Indirect standardized effects of the explanatory model of the left-right dimension usage Indirect effects Education Salary per person Subjective class Interest in politics Information Usage of the dimension

Salary per Subjective person class 0.00 0.00

Interest in politics 0.00

Type of locality 0.00

Sex (feminine) 0.00

Age

Education

0.00

0.00

0.20

-0.05

-0.14

0.00

0.00

0.00

0.00

0.00

0.27

-0.05

-0.10

0.13

0.00

0.00

0.00

0.00

Information 0.00

0.08

-0.02

-0.05

0.03

0.02

0.00

0.00

0.00

0.24

-0.08

-0.12

0.13

0.04

0.01

0.00

0.00

0.11

-0.10

-0.04

0.15

0.04

0.04

0.01

0.00

Source: Analyses based on EVS&WVS data from 2005

107

The Values of Romanians: 1993-2006. A sociological perspective Table 12. The evolution of self-placement within different categories of population Population Year

Average of the selfThe difference between the average placement of the category and the average of the (1=left; 10=right) population 1993 1998 1999 2005 1993 1998 1999 2005

Total population

5.7

5.3

5.8

6.0

0.0

0.0

0.0

0.0

Men

5.6

5.3

5.8

6.0

-0.1

0.0

0.0

0.0

Women

5.7

5.3

5.8

6.0

0.1

0.0

0.0

0.0

18-34 years

5.9

5.4

6.2

6.2

0.2

0.1

0.4

0.2

35-54 years

5.7

5.3

5.7

6.1

0.0

0.0

-0.2

0.1

5.5

5.2

5.6

5.6

-0.2

-0.1

-0.2

-0.4

5.5

5.1

6.3

5.5

-0.1

-0.3

0.5

-0.5

5.7

5.2

5.5

5.8

0.0

-0.1

-0.3

-0.2

5.7

5.3

5.9

6.1

0.0

-0.1

0.1

0.1

5.9

5.9

5.5

6.5

0.2

0.6

-0.3

0.5

Lower class

-

4.9

-

5.6

-

-0.4

-

-0.4

Labor class Down part of the middle class Upper part of the middle class Low salary

-

4.9

-

5.6

-

-0.4

-

-0.4

-

5.6

-

6.2

-

0.3

-

0.2

-

5.6

-

6.6

-

0.3

-

0.6

5.5

5.2

6.3

5.7

-0.1

-0.1

0.4

-0.3

Medium salary

5.7

5.3

5.6

5.9

0.0

0.0

-0.3

-0.1

55+ years At most comprehensive school Apprentices, unfinished college College, further education, unfinished faculty Faculty, M.A., PhD

High salary Rural-peripheral

5.8

5.6

5.8

6.2

0.1

0.3

-0.1

0.2

5.5

5.3

5.5

5.7

-0.2

0.0

-0.3

-0.3

Rural-center

5.5

5.0

6.4

5.9

-0.2

-0.3

0.6

-0.1

Small urban

5.7

5.2

5.6

6.1

0.0

-0.1

-0.2

0.1

Big urban

5.8

5.6

5.8

6.1

0.2

0.3

-0.1

0.1

Employed - high

5.9

5.8

5.7

6.6

0.2

0.4

-0.1

0.6

Employed - medium

5.6

5.2

5.8

6.2

-0.1

-0.1

0.0

0.2

Employed - low

5.8

4.9

5.5

5.8

0.1

-0.5

-0.3

-0.2 -0.4

Retired person

5.5

5.4

6.1

5.6

-0.2

0.1

0.3

Housewife, stay-at-home

5.7

5.7

6.3

6.1

0.0

0.4

0.5

0.1

Student/pupil

6.6

5.3

6.6

6.1

0.9

0.0

0.8

0.1

Unemployed Trade union’s member/professional association’s member Non-member of a trade union/professional association No confidence in trade unions and companies High confidence in trade unions and companies

4.0

5.6

5.7

6.2

-1.7

0.3

-0.2

0.2

5.9

5.5

5.7

6.7

0.2

0.2

-0.1

0.7

5.6

5.9

5.7

5.9

0.0

0.6

-0.1

-0.1

5.9

5.4

5.8

5.8

0.2

0.0

0.0

-0.2

5.5

5.2

5.9

6.1

-0.2

-0.1

0.1

0.1

108 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases Population Very high confidence in trade unions and companies Going to Church-once a year/more seldom/not at all Going to Church-at festivals Going to Church-once a month Going to Church-once a week Member of a religious organization Non-member of a religious organization Low confidence in Church High confidence in Church Very high confidence in Church

Average of the selfplacement (1=left; 10=right)

The difference between the average of the category and the average of the population

5.4

5.5

-

6.4

-0.3

0.1

-

0.4

5.5

5.2

5.9

6.3

-0.2

-0.2

0.1

0.3

5.7

5.4

5.9

6.2

0.0

0.1

0.1

0.2

5.8

5.3

5.8

5.7

0.1

0.0

0.0

-0.3

5.7

5.5

5.6

5.8

0.0

0.2

-0.2

-0.2

5.7

-

6.0

6.1

0.0

-

0.2

0.1

5.7

-

5.8

6.0

0.0

-

0.0

0.0

5.9

5.2

5.9

6.5

0.2

-0.1

0.1

0.5

5.7

5.4

5.6

5.7

0.0

0.1

-0.2

-0.3

5.5

5.3

5.9

6.0

-0.2

-0.1

0.1

0.0

Source: Analyses based on EVS&WVS data from 1993, 1998, 1999and 2005

109

The Values of Romanians: 1993-2006. A sociological perspective Table 13. Parties’ self-placement on the left-right axis (the average of the selfplacements of the party’s voters; 1=left; 10=right) Political formation

1993

1997

1998

1999

PNTCD

6.8

6.8

6.3

7.5

PNL

6.4

2005

2006

6.9

6.9

7.4

7.4

6.9

PD

5.8

5.7

5.7

6.9

6.4

CDR

6.3

6.1

6.7

6.4

4.4

4.5

PNL - PNTCD - AC - PE

7.1

DA Alliance FSN

5.3

FDSN

5.1

PDSR

4.7

3.9

PSDR ApR

5.1

4.2

3.3

4.6

4.8

5.8

PSD PUNR

5.3

4.8

5.0

5.0

PRM

4.6

5.3

4.2

5.1

5.7

5.3

UDMR

6.4

5.7

6.0

5.0

6.6

6.6

AC

6.1

PNL-Youngsters

6.6

PDAR

5.2

PSM

4.2

PNG

5.5

Source: Analyses based on BOP–FSD 2006; EVS&WVS 1993, 1997, 1998, 1999 and 2005

Table 14. The relative importance of different attitudes in explaining the selfpositioning on the left-right axis Item

1993

1997

1998

1999

2005

materialism vs. postmaterialism equality vs. freedom

2006

medie

0.11

-

0.00

0.04

0.10

0.00

-

-

0.00

-

0.04

0.01

0.18

0.07

equal vs. unequal salary

-

-

-

-

economy vs. environment differences among salaries: bad vs. good

-

-

0.04

-

0.00

-

0.00

0.00

-

0.02

0.30

0.05

0.09

0.16

0.15

0.25

0.17

property: state vs. private responsibility for the living conditions: state vs. individual competition: bad vs. good

0.16

0.19

0.13

0.15

0.11

0.02

0.13

0.05

0.15

0.07

0.14

0.27

0.00

0.11

0.06

0.03

0.23

0.07

0.07

0.07

0.09

success: luck and relations vs. work

0.03

0.10

0.04

-

0.07

0.04

0.05

0.08

0.14

0.07

-

0.21

0.26

0.15

-

-

-

-

0.11

0.18

0.15

richness rises: on others’ account vs. for everyone one’s way in life: destiny vs. selfdeterminacy

110 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases Item unemployed persons: right of refusal vs. any job public houses: state control vs. freedom

1993

1997

1998

1999

2005

2006

medie

0.05

0.05

-

0.06

-

-

0.05

-

-

-

0.10

-

-

0.10

-

-

-

0.11

-

-

0.11

responsible for the pension: state vs. individual responsible for the residence: state vs. individual major changes: attention vs. action

-

-

-

0.05

-

-

0.05

0.03

0.13

0.26

0.13

-

-

0.14

good ideas: old vs. new

0.04

0.16

0.06

-

-

-

0.09

Source: Analyses based on BOP-FSD data from 2006; EVS&WVS 1993, 1997, 1998, 1999 and 2005 *Pratt coefficients obtained by regression with categorical variables Table 15. Relative importance of common attitudes for different research made for explaining self-placement on the left-right axis Item differences among salaries: bad vs. good property: state vs. private the responsibility for the living conditions: state vs. individual competition: bad vs. good

1993

1997

1998

1999

2005

2006

average

0.61

0.26

0.20

0.31

0.26

0.48

0.35

0.29

0.39

0.22

0.13

0.17

0.14

0.22

0.03

0.24

0.27

0.41

0.45

0.11

0.25

0.07

0.11

0.31

0.16

0.12

0.27

0.17

Source: Analyses based on BOP-FSD data from 2006; EVS&WVS 1993, 1997, 1998, 1999 and 2005 *Pratt coefficients obtained by regression with categorical variables

Reference list Alina Mungiu-Pippidi (2002) Politica după comunism [Politics after communism], Bucureşti: Humanitas Arbuckle, L. James (2006) Amos 7.0 User’s Guide, SPSS Inc. Bartle, John (1998) Left-right position matters, but does social class? Causal models of the 1992 British General Election, British Journal of Political Science, vol. 28, nr. 3, pp. 501-529 Bădescu, Gabriel & Sum E. Paul (2005) The importance of left-right orientations in the new democracies, International Conference on „Elections and democratic governance”, Taipei Boy, Daniel. Jaffre, Jerome. Muxel, Anne (1997) Dissonances sur la dimension gauchedroite în Boy, Daniel şi Mayer, Norma (eds.) L’électeurs a ses raisons, Presses de la Foundation Nationale de Sciences Politiques Chiribucă, Dan (1996) Mentalitatea – factor de influenţă în tranziţia societăţii româneşti, [Mentality – an influence factor in Romanian society transition] in Studia Sociologie, nr. 1-2 Comşa, Mircea (2003) O analiză a ratei de răspuns la itemi în anchetele de opinie naţionale, [An analysis of the answering rate to items in national opinion surveys] in Sociologie Românească, nr. 3-4, pp. 56-72 111

The Values of Romanians: 1993-2006. A sociological perspective Comşa, Mircea (2006) Cultură, participare şi opţiuni politice [Culture, participation and political prefferences], în Sandu, D., Comşa, M., Rughiniş, C., Toth, A., Voicu, M., Voicu, B. Viaţa socială în România urbană [Social life in urban Romania], Iaşi: Polirom, pp. 137-199 Comşa, Mircea & Rotariu, Traian (coord.) (2005) Alegeri generale 2004. O perspectivă sociologică [2004 general elections. A sociological view], Cluj-Napoca: Eikon Corrie Potter (2001) Left-right self-placement in Western Europe. What responses and non-responses indicate, www.polisci.wisc.edu/~behavior/ papers/Potter2001.pdf Dalton, Russell (1988) Citizen Politics in Western Democracies. Public Opinion and Political Parties in the United States, Great Britain, West Germany, and France, NJ: Chatham House Publishers Dalton, Russell (2002) Citizen politics. Public opinion and political parties in advanced industrial democracies, NJ: Chatam House Publishers Dieter Fuchs & Dieter Rucht (1994) Support for New Social Movements in Five Western European Countries, in C. Rootes and H. Davis (eds.) A New Europe? Social Change and Political Transformation, London: University College London Press Eijk, van der Cees. Schmitt, Hermann. Binder, Tanja (2006) Left-Right Orientations and Party Choice, in Thomassen, Jacques (edt) The European Voter: A Comparative Study of Modern Democracies, Londra: Oxford University Press, pp. 167-191 Evans, Geoffrey & Whitefield, Stephen (1998) The Evolution of Left and right in PostSoviet Russia, Europe-Asia Studies, vol. 50, pp. 1023-1043 Evans, Geoffrey; Heath, Anthony; Lalljee, Mansur (1996) Measuring left-right and libertarian-authoritarian values in the British electorate, The British Journal of Sociology, vol. 47, no. 1, pp. 93-112 Evans, Jocelyn (2004) Voters&voting, Sage Publications Freire, Andre (2006) Bringing Social Identities Back In: The Social Anchors of Left– Right Orientation in Western Europe, International Political Science Review, vol 27, nr. 4, pp. 359-378 Gabel, J. Matthew & Anderson, J. Christopher (2001) Exploring the European Demos (or lack thereof): the Structure of Citizen Attitudes and the European Political Space, CEuS Working Paper no. 4 Hooghe, Marc (2003) Participation in Voluntary Associations and Value Indicators: The Effect of Current and Previous Participation Experiences, Nonprofit and Voluntary Sector Quarterly, vol. 32, nr. 1, pp. 47-69 Huber, D. John (1989) Values and Partisanship in Left-Right Orientations: Measuring Ideology, The European Journal of Political Research, vol. 17, pp. 599-621. Inglehart, Ronald & Abramson R. Paul (1994) Economic Security and Value Change, American Political Science Review, vol. 88, pp. 336-354 Inglehart, Ronald & Klingemann, Hans-Dieter (1976) Party Identification, Ideological Preference and the Left-Right Dimension among Western Mass Publics, în I. Budge H.-D. Klingemann, J. Bara, and E. Tenenbaum. (eds), Party Identification and Beyond: Representations of Voting and Party Competition, London: Wiley Inglehart, Ronald & Paul R. Abramson (1994) Economic Security and Value Change, în American Political Science Review, vol. 88, pp. 336-354. 112 corectura 2 28 iulie 2008

Ideological self-placement: identification, sophistication, bases Inglehart, Ronald (1990) Culture shift in advanced industrial society, NJ: Princeton University Press Inglehart, Ronald (1997) Modernization and Post-Modernization. Cultural, Economic and Political Change in 43 Societies, Princeton University Press Jacoby, William (2002) Liberal-conservative thinking in the American electorate, in Michael X. Delli Carpini, Leonie Huddy, Robert Y. Shapiro (Editors), Research in Micropolitics: Political Decision Making, Participation, and Deliberation, Volume Six. Greenwich, CT: JAI Press. Kitschelt, Herbert & Hellemans, Staf (1990) The left-right semantics and the new politics cleavage, Comparative Political Studies, vol. 23, nr. 2, pp. 210-238 Kitschelt, Herbert (1994) The Transformation of European Social Democracy, Cambridge University Press. Knutsen, Oddbjørn (1995a) Value Orientation, Political Conflicts and Left-Right Identification: A Comparative Study, European Journal of Political Science, nr. 28, pp. 63-93 Knutsen, Oddbjørn (1995b) The Impact of Old Politics and New Politics Value Orientations on Party Choice - A Comparative Study, Journal of Public Policy, nr. 15, pp. 1-63 Knutsen, Oddbjorn (1997) The partisan and the value-based component of the left-right self-placement: a comparative study, International Political Science Review, vol. 18, nr. 2, pp. 191-225 Knutsen, Oddbjorn (1998a) Europeans Move Towards The Center: A Comparative Longitudinal study of Left-Right Self-Placement in Western Europe, International Journal of Public Opinion Research, vol. 10, pp. 292-316 Knutsen, Oddbjorn (1998b) The strength of the partisan component of left-right identity. A comparative longitudinal study of left-right party polarization in eight West European countries, Party Politics, vol. 4, nr. 1, pp. 5-31 Kroh, Martin (2005) Surveying the Left-Right Dimension: The Choice of a Response Format, Discussion Paper 491, DIW Berlin – German Institute for Economic Research Socio-Economic Panel Study (SOEP) Krosnick, Jon A. & Matthew K. Berent (1990) Impact of Verbal Labeling of Response Alternatives and Branching on Attitude Measurement Reliability, American National Election Studies (ANES) Pilot Study Report, ftp://ftp.electionstudies.org/ftp/nes/bibliography/documents/nes002280.pdf Lambert, Ronald. Curtis, James. Brown, Steven. Kay, Barry (1986) In search of left/right beliefs in the Canadian electorate, Canadian Journal of Political Science, vol. 19, nr. 3, pp. 541-563 Lipset, Seymour M., Rokkan, Stein (1967) Cleavage Structures, Party Systems, and Voter Alignments in Party System and Voter Alignments, Londra: CollierMacmillan Mateju, Petr & Vlachova, Klara (1998) Values and electoral decisions in the Czech Republic, Communist and Post-Communist Studies, vol. 31, nr. 3, pp. 249-269 Meulman, Jacqueline & Heiser, Willem (2005) SPSS Categories 14.0, SPSS Inc. Meyers, Lawrence. Gamst, Glenn. Guarino, A. (2006) Applied multivariate research. Design and interpretation, Sage Publications Noelle-Neumann, Elisabeth (1998) A shift from the right to the left as an indicator of value change: A battle for the climate of opinion, International Journal of Public Opinion Research, vol. 10, nr. 4, pp. 317-334 113

The Values of Romanians: 1993-2006. A sociological perspective Page, Melanie. Braver, Sanford. MacKinnon, David (2003) Levine’s guide to SPSS analysis of variance, New Jersey: Lawrence Erlbaum Associates Inc. Rotariu, Traian & Iluţ, Petre (1997) Ancheta sociologică şi sondajul de opinie [Sociological survey and public opinion survey], Iaşi: Polirom Sandu, Dumitru (1996) Sociologia tranziţiei. Valori şi tipuri sociale în România [The sociology of transition. Values and social types in Romania], Bucureşti: Staff Sandu, Dumitru (1999) Spaţiul social al tranziţiei [The social space of transition], Iaşi: Polirom Sani, Giacomo and Sartori, Giovani (1983) Polarization, Fragmentation, and Competition in Western Democracies, in Western European Party Systems: Continuity and Change, Beverly Hills: Sage Todosijevic, Bojan (2004) The Hungarian voter: left-right dimension as a clue to policy preferences, International Political Science Review, vol. 25, nr. 4, pp. 411-433 Voicu, Bogdan (1999) Despre măsurarea intenţiei de vot în sondajele de opinie, [Measuring vote intention in opinion surveys] in Sociologie Românească, nr. 4, pp. 48-76 Zechmeister, Elizabeth (2006) What’s left and who’s right in Mexican politics? A Qmethod approach to understanding ideological labels in Mexican politics, Political Behavior, vol. 28, nr. 2, pp. 151-173 *** (2006) SPSS Base 15.0 User’s Guide, SPSS Inc.

Databases used Barometer of public opinion (BOP-FSD) (2003, 2006), surveys significant at a national level for adult uninstitutionalized population, www.osf.ro European Values Survey (EVS) (1999), series of comparative surveys significant at national level for uninstitutionalized adult population, http://www.europeanvalues.nl The Comparative Study of Electoral Systems (2003) CSES Module 1 Full release [dataset], Ann Arbor, MI: University of Michigan, Center for Political Studies [producer and distributor], series of comparative surveys significant at national level for uninstitutionalized adult population www.cses.org World Values Survey (WVS) (1993), series of comparative surveys significant at national level for uninstitutionalized adult population, www.worldvaluessurvey.org World Values Survey (WVS) (1997), series of comparative surveys significant at national level for uninstitutionalized adult population, www.worldvaluessurvey.org World Values Survey (WVS) (1998), series of comparative surveys significant at national level for uninstitutionalized adult population, www.worldvaluessurvey.org

114 corectura 2 28 iulie 2008

Institutional trust – victim of the postcommunist transition CLAUDIU D. TUFIŞ

To trust or not to trust? Regardless of the answer, this is not really a question, because all of us have trust. We are different only with respect to the level of trust and the objects of our trust. But all of us do trust, even if we are not completely aware of this. A simple example can offer evidence for this statement. Let’s follow Mr. Escu’s actions on Tuesday morning, between six and eight: at 6:00 AM the alarm of his electric clock woke him up. He took a shower and then he went out for a couple of minutes to buy a newspaper, which he read while having breakfast (eggs, ham, bread, and coffee). At 7:30 AM he headed to the bus station, stopping on his way to an ATM to pay the bill for his cell. Once at the bus station, he waited four minutes and then got into the bus that drove him to work. At 8:00 AM Mr. Escu arrived at his office. It was a regular weekday morning, just like any other, with an almost automatic behavior. A morning during which Mr. Escu had trusted at least eight times, both in other persons and, indirectly, in the state institutions. Even before waking up, he trusted that there will be no power failures during the night and that his alarm clock would wake him up in time. He also trusted that when he would take his shower he would have hot water to clean up. When he bought the newspaper, he had trust that his money would be accepted by the seller and that he would get the correct change (he put it in his wallet without even checking). He trusted that what he read in the newspaper was a correct description of the events that had happened the other day. While he was eating breakfast, he trusted that his food was not infected with harmful bacteria. When he used the ATM, he trusted the bank would transfer the money from his account into the account of the mobile phone company. Finally, when he got into the bus he trusted the driver, a completely unknown person, would drive him to his destination without having an accident. It’s difficult to imagine how Mr. Escu’s morning would have been if he had not trusted any of the situations 115

The Values of Romanians: 1993-2006. A sociological perspective

presented above. Mr. Escu’s morning suggests that trust is necessary for the proper functioning of a society. This applies even more in societies that are organized based on the principles of democracy, because democracy cannot survive in absence of confidence. Totalitarian or authoritarian regimes also need trust, but, in its absence, they can survive by repression, which is an inaccessible method for democratic regimes. The importance of trust is visible in the increasing interest for this issue in old and established democracies, where the decreasing levels of trust in the state’s institutions generated academic and public debates (such a debate was organized at Camp David under the aegis of the US presidency, in order to discuss the themes treated by Robert Putnam in Bowling Alone). The subject is all the more important in the new Central and Eastern European democracies: first, the institutional context of countries in transition is not stable, but rather in a continuous and complex transformation, with the purpose of developing a set of institutions that will fit the society’s characteristics. Secondly, the communist past proves to be extremely unfavorable to institutional trust: forty years of communism left as heritage a culture of mistrust in the state and in its institutions, which was built based on the pre-war culture. All these reasons justify any attempt to understand the mechanisms of institutional trust. Most of the studies of post-communist transitions focus on the role of institutional changes: the countries in transition are assessed according to the success they had in developing the set of democratic institutions necessary for creating a strong democracy. There is also a second dimension of transition, just as important: people’s values, attitudes, and beliefs – the basic components of political culture. When the elites cooperate, the formation of the new set of institutions of the democratic system is a fairly easy task. Post-communist transitions, however, are a complex phenomenon, one that only starts, but does not end, with the creation of a new institutional complex. The transition cannot be complete without changing citizens’ attitudes and behavior in their interactions with the new institutions, and this change is much more difficult than implementing new institutions: “many people anticipated that the relationship between citizens and the state would change immediately and dramatically, but this never happened. A lot of people continue to be afraid of the authorities and many officials, consciously or not, continue to behave like masters, or merely refuse to act as a result of their fear to assume any responsibilities” (Macovei, 1998). The transitions that do not manage to modify both the institutional and the cultural dimension are destined to fail: a democratic system that does not manage to gain its citizens’ confidence will stop, eventually, fulfilling its functions, and this can lead to democratic breakdown. Fortunately, these are long term predictions, meaning that state institutions have enough time at their disposal to prove their trustworthiness.

116 corectura 2 28 iulie 2008

Institutional trust – victim of the postcommunist transition

This chapter is dedicated to the cultural dimension of the postcommunist transition in Romania, trying to uncover whether Romanians have or do not have trust in their state’s institutions. Previous studies (Mishler and Rose 1997, Rose, Mishler and Haerpfer 1998, Ekiert and Kubik 1999, Mishler and Rose 2001, Howard 2003) show that all postcommunist countries are characterized by a deficit of trust, explained by the heritage left by the communist regimes, institutional inefficiency, and the short period of time that has passed since the fall of communism. The main goal of the analyses presented in this chapter is the identification of the factors that influence trust in the state’s institutions. I will not discuss here about the optimal level of institutional trust, and I will not argue that all citizens should trust the state’s institutions (the idea of total consensus, so dear to some politicians, is nothing but a romantic ideal that cannot be achieved in a complex society). On the contrary, a certain level of distrust is necessary for the health of a democracy, transforming citizens into guardians of democracy. In the analyses, I focus mainly on explaining the level of institutional trust in 2005. I also use data from 1993, 1997, and 1999, but only to present the evolution of institutional trust over time. In the next section I discuss a series of theoretical aspects relevant for trust, emphasizing the connections between trust and risk, trust and social capital, institutional trust and interpersonal trust, and trust and uncertainty. In this section, I will also provide a definition for the concept of trust. The chapter continues with the presentation of the theoretical model used in the analysis and with the presentation and discussion of the results.

Trust – theoretical aspects The postcommunist transitions in Central and Eastern Europe have renewed the interest for the role of trust in the process of democratic consolidation: “trust is linked to a series of dispositions which stand at the basis of democratic culture, including tolerance for pluralism and criticism” (Warren 1999: 9). Using a more general formulation, trust consists of “cognitive schemes and moral dispositions […] which are perceived by theoreticians as a variable upon which institutional viability depends” (Offe 1999: 42).

Trust and risk Most of the existing studies on trust can be grouped into two categories, depending on the theoretical approach they use. The first approach defines trust, from a rational choice perspective, as “introducing the risk in the decision to act” (Coleman 1990: 61). The decision to trust is based on a rational calculus that takes into account all available information. Analyzing the relationship between trust and risk, Luhmann sees trust as a risky investment whose main 117

The Values of Romanians: 1993-2006. A sociological perspective

goal is to reduce the complexity that characterizes modern societies (Luhmann 1979). Beck, in his discussion of modernity and risk, asks: “in modern societies, the social production of wealth is accompanied, systematically, by the social production of risk […] How can these risks and dangers, systematically produced as part of the modernization, be prevented, reduced, or channeled?” (Beck 1992: 19). Trust is the answer Beck offers to this question. Although the logic of this approach is attractive by the simplicity and high degree of generalization it offers, the rational choice theory suffers from a significant problem that has not been solved: people do not always act the way rational choice assumptions suggest they should act. “In the attempt to include trust in a perspective that explains human behavior in terms of maximizing individual utility, rational choice scholars end up by modifying and distorting the concept of trust in unrecognizable forms” (Ruscio 1999: 640). Experimentbased studies prove that empirical results contradict rational choice definitions of trust (Tyler and Degoey 1996). The second approach, political culture, defines trust as a combination of values, attitudes, and beliefs that are shared by the members of a community. Many theoretical and empirical studies define trust from this perspective, the most important including Putnam (1993), Fukuyama (1995), Misztal (1996), Inglehart (1997), and Sztompka (1999). The concept of risk is discussed in this perspective as well, but as an object of attitudes and not as a major component of rational calculus. Giddens, for instance, suggests that, in the absence of trust, individuals should analyze every detail of every interaction with the environment, which would lead to the individual’s impossibility to act: “trust, in this sense, is essential for a protection system that guards the individual in his contacts with day-to-day reality” (Giddens 1991: 3). The state’s total involvement in all the sectors of society during the communist regime resulted in the dramatic reduction of the risks associated with daily life. Under the new regime, democratic rules have required the withdrawal of the state from the society, letting the individual confront these risks alone. An example can clarify this idea: job security, characteristic of the planned economy, was replaced with the risk of losing one’s job during the transition to market economy. In a society that has become increasingly complex, individuals need trust in order to eliminate the impossibility to act Giddens was describing. Postcommunist societies have to develop new systems of cooperation among individuals and between individuals and the state, explaining, in part, the interest in trust as a research topic from the political culture perspective: communities depend on trust, and trust is culturally determined (Fukuyama 1995). The communist regime survived long enough to force a transition from trust to distrust (or, if one considers distrust to have its roots in the precommunist period, it survived long enough to consolidate it) and thus institutionalized a culture of generalized distrust (towards both the state’s institutions and individuals). 118 corectura 2 28 iulie 2008

Institutional trust – victim of the postcommunist transition

Is a reversed change possible? Can institutional trust be reconstructed? We know that cultural components are subtle, “they have the diffuse properties of belief systems […] which, anthropologists argue, spread with difficulty and suffering substantial changes” (Almond and Verba 1965: 3-4). The low levels of trust in different state institutions suggest that fifteen years have not been enough to transform the postcommunist citizen into a trusting citizen. The failure of this transformation, however, is hardly surprising if one remembers Putnam’s descriptions of civic traditions in Italy.

Trust and social capital Although the concept of social capital is not the topic of this chapter, I will briefly discuss its definitions, because all of them identify a significant relationship between trust and social capital1. Discussing the importance of social capital as individual resource, Coleman argued that the fate of social capital depends on how much trust one can have in the social environment (Coleman 1999: 20). Putnam understands trust as a component of social capital: “social capital refers here to features of social organizations, such as trust, norms, and networks, that can improve the efficiency of society by facilitating coordinated actions” (Putnam 1993: 167). Other authors offer similar definitions: “social capital is a capability that arises from the prevalence of trust in a society or in certain parts of it” (Fukuyama 1995: 17); “social capital is a culture of trust and tolerance, in which extensive networks of voluntary associations emerge” (Inglehart 1997: 188); “trust has important functions for wider communities within which it prevails. First of all, it encourages sociability, participation with others in various forms of association, and in this way enriches the network of interpersonal ties, enlarges the field of interactions, and allows for greater intimacy of interpersonal contacts” (Sztompka 1999: 105). If the idea of trust as one of the most important factors that facilitate the formation of relational social capital is fully accepted, less attention is given to the opposite relationship. Trust is necessary in order to expand social capital at higher levels: social capital exists, as a first level, within families and, more generally, within any type of organization built on kinship. The expansion to other types of organizations, which are based on other forms of “social glue”, requires the existence of trust. Yet, when social capital reaches a certain level, the relationship between trust and social capital becomes circular: trust promotes social capital and social capital leads to the formation of new trust relationships. Beyond this level the relationship between trust and social capital becomes a “virtuous circle”.

1

For a detailed discussion of the concept of social capital and of its applications in Central and Eastern Europe, see Mihaylova (2004). 119

The Values of Romanians: 1993-2006. A sociological perspective

Institutional trust and interpersonal trust Based on the object of trust, the literature distinguishes between institutional trust and interpersonal trust. The two types, however, are closely linked: “If your trust in the enforcement agency falters, you will not trust persons to fulfill their terms of agreement, and thus will not enter into that agreement [...] you will not trust the enforcement agency to do on balance what is expected of it. It is this interconnectedness which makes trust such a fragile commodity. If it erodes in any part of the mosaic, it brings down an awful lot with it” (Dasgupta 1998: 50). The police and the justice system are classical examples of such enforcement agencies. The relationship between institutional trust and interpersonal trust is a two-way relationship. Trust in an institution can be extended to the members of the institution and, by extension, to the group to which these members belong. For instance, trust in church can extend to trust in priests, so that when we meet a stranger and find out he/she is a priest, we may be willing to trust that person more than a person with a different occupation (a police officer for instance). At the same time, the relationship works in the opposite direction as well: low levels of trust in judges or district attorneys can generate low levels of trust in the judiciary.

Trust and uncertainty Seligman defines trust as “some sort of belief in the goodwill of the other, given the opaqueness of other’s intentions and calculations. The opaqueness rests precisely on that aspect of alter’s behavior that is beyond the calculable attributes of role fulfillment” (Seligman 1997: 43). The opaqueness referred to in this definition represents the risk element that is inherent to the concept of trust. Sztompka has a more general definition of trust: “a bet about the future contingent actions of others” (Sztompka 1999: 25). Both definitions regard trust as independent of the object of trust. Do they include, in this case, the state’s institutions as possible objects of trust? In both definitions uncertainty plays a significant role in determining the level of trust. But to talk about uncertainty with respect to the actions of the state’s institutions is somewhat problematic. Given that one of the main roles of the state is to serve its citizens, Seligman’s definition cannot be used, because citizens do not need to believe in the institutions’ goodwill; their goodwill is included in their own definition. The same problem affects Sztompka’s definition: we cannot make a bet about future contingent actions of state institutions, because there is no uncertainty related to their actions. The state’s institutions are required to react to any demand coming from its citizens, as long as the demand reaches the institutions through the pre-established access channels. The set of possible actions that the institution can take is restricted by internal and external regulations that the institution has to follow. Theoretically, the intentions and calculations of state institutions are characterized by high 120 corectura 2 28 iulie 2008

Institutional trust – victim of the postcommunist transition

visibility: the institutions’ functions, constraints, and procedures are available to any person that is interested in knowing them. Although these arguments suggest that the definitions presented above cannot be used in the study of institutional trust, this is a valid suggestion only with respect to the institutions of states with a democratic tradition. In postcommunist countries there are some institutional characteristics that generate a certain degree of uncertainty in the interactions between citizens and the state. This uncertainty may be caused by citizens having an incomplete understanding of the functions of the new institutions (the citizen that asks the president for a job is the classical example of a citizen that is confused about the president’s functions), by the short period of time since the new institutions were created (which makes them inefficient), or by the actions of the institution’s employees (corruption and lack of professionalism are just some of the possible sources of uncertainty in this case). The distinction between institutions and their incumbents has an important role in the analysis of trust. Hardin argued that institutional trust requires “a micro-level account of how government works at the macro level. This will largely be an account of rational expectations of what government and its agents are likely to do […] Citizens’ expectations must also be rational in the sense of depending on the rational commitments of officials. Rationally grounded trust in officials requires that the officials be responsive to popular needs and desires” (Hardin 1998: 12). People interact with the institutions of the state through two main forms: primary contact, when a person has a personal experience with the institution, and secondary contact, understood as a description of others’ personal experiences with the institution (often those who describe to a person their primary contacts are friends, relatives, or, at a more general level, the mass-media). The distinction between primary and secondary contacts is similar to Sztompka’s distinction between primary and secondary targets of trust (Sztompka, 1999: 41-51). The interactions between the people and the institutions are thus mediated by the incumbents of the institutions, and the evaluations of these interactions are extended from the incumbents to the institutions. Any contact with the police for instance is mediated through an officer from a police station. Since the officer is an employee of the police department, the citizen’s experience with the police will be interpreted in the light of the interaction with that officer. This generalization is based on the assumptions that the institution established a set of rules that have to be followed by all employees and that the institution will verify that its incumbents respect these rules. Thus, the interactions with the incumbents will be generalized to describe the interactions with the institutions. A negative experience with officer Escu will not make the citizen declare he/she does not

121

The Values of Romanians: 1993-2006. A sociological perspective

trust Escu; the citizen will be quick, however, to declare that they do not trust the police as an institution. Uncertainty is also related to the perceived efficiency of the institution. As Luhmann argued, “trust is only possible in a situation where the possible damage may be greater than the advantage you seek” (Luhmann, 1988: 98). In the case of institutional trust, the efficiency of the institutions is used in assessing the possible damage (or cost). To the extent that the citizens’ demands follow the official channels and satisfy all the requirements, one can be sure that the institutions of the state will engage in a series of activities triggered by one’s demand: there is no uncertainty related to the existence/nonexistence of a certain action. The uncertainty stems from the way in which the action will be performed by the institution (i.e. by the incumbents of the different roles of authority). Gambetta incorporates many of these elements in his definition of trust as „a particular level of the subjective probability with which and agent assesses that another agent or group of agents will perform a particular action, both before he can monitor such action and in a context in which it affects its own actions” (Gambetta 1998: 217). According to this definition, trust is a variable depending on certain personal and cultural factors that can be used to describe both interpersonal and institutional trust. The trust relation has three components: the truster, the trustee, and the event. The complexity of the interaction is reduced by the trust relation. The only element missing from this definition is the efficiency of the object of trust. Adding this element, I define trust as a context-dependent belief that an agent will perform a particular action that will correspond to the truster’s current expectations. People have different positions on the trust–distrust continuum, positions that could be explained by a series of personal and cultural factors that form together the set of internal determinants of trust. The position on this continuum, however, is not fixed. It can be influenced by context-dependent factors which form the external determinants of trust. These factors are the modifiers that make it possible so that a usually trusting person will not trust a certain agent in a certain context. The distinction between internal and external determinants has important methodological consequences. While internal determinants can be identified using survey data, external determinants can be uncovered only by a detailed description of each particular case (this approach is often used in case studies) or by using indicators that are measured at the macro-level. Since the analyses presented in this chapter are based on survey data, I will discuss here only internal determinants of trust, i.e. those corresponding to respondents’ characteristics.

122 corectura 2 28 iulie 2008

Institutional trust – victim of the postcommunist transition

Trust – analysis model The model I use in analysis starts from the definition of trust presented in the previous section and seeks to explain individual differences with respect to institutional trust.

Dependent variables The 2005 WVS/EVS survey measures (on a four point scale, from very little to very much) trust in 19 institutions that can be grouped into four categories: state institutions (presidency, parliament, cabinet, political parties, civil servants, police, courts, and city halls), civil society institutions (environmental NGOs, women NGOs, charitable NGOs, trade unions, newspapers, and television), traditional institutions (church and army), and international institutions (EU, UN, and NATO). I did not include in the analysis economic institutions, because they function on different principles and are interested, primarily, in their own profitability. Although newspapers and television are not, per se, civil society organizations, I included them in this group because their role in the relationship with the state is similar to the role of civil society institutions. I used this theoretical classification as a model for creating four dependent variables. In each of the four cases I verified through factor analysis and reliability analysis if the variables create a scale and, for easier interpretation, I created the dependent variables as additive scales. Thus, I use four dependent variables in my analyses: trust in state’s institutions2, trust in civil society institutions3, trust in traditional institutions4, and trust in international institutions5. The correlations between the four measures of institutional trust vary between 0.214 and 0.576. Given these correlations, it is possible that some of the dependent variables will have a common set of determinants. At the same time, some factors may have an effect only on some types of institutional trust and the size of the effects of the common factors may depend on the type of trust that I analyzed. What are the factors that may influence the level of institutional trust?

2

Factor analysis (ML – Oblimin) extracts a single factor that explains 45.8% of the variation. KMO = 0.864. Cronbach Alpha = 0.866. The variable ranges between 0 and 24. 3 Factor analysis (ML – Oblimin) extracts two factors, correlated at 0.371. KMO = 0.733. Cronbach Alpha = 0.779. The variable ranges between 0 and 18. 4 Trust in the army and trust in church are correlated at 0.401. The variable ranges between 0 and 6. 5 Factor analysis (ML – Oblimin) extracts a single factor that explains 73.3% of the variation. KMO = 0.738. Cronbach Alpha = 0.890. The variable ranges between 0 and 9. 123

The Values of Romanians: 1993-2006. A sociological perspective

Socio-economic characteristics A first set is represented by a person’s socio-economic characteristics: gender, age, education, income, ethnicity, religion, and size of locality. These are variables that define the initial position of a person on the trust – distrust continuum. Gender (dummy variable coded 0 for women and 1 for men) is a control variable: there are no reasons to expect significant differences between men and women with respect to institutional trust. Age (recoded in eight categories: under 21, 21-30, 31-40, 41-50, 51-60, 61-70, 71-80, and over 80) should have a significant positive effect on institutional trust. This effect is determined mainly by different attitudes toward authority and the state. Education (recoded in five categories: 4 grades and less, 5-8 grades, 9-11 grades, high school and post high school, and university or more) should have a significant negative effect. The more educated a person is, the higher his/her capacity to understand the institutions’ actions and the alternatives to these actions, which can lead to a more critical attitude towards the institutions. Income (measured as deciles of income per capita), ethnicity (dummy variable coded 1 for Romanian and 0 for other), and religion (dummy variable coded 1 for Orthodox and 0 for other) are used as a control variables. Ethnicity and religion indicate membership to the majority group in the population (Romanian and Orthodox versus other ethnicity or other religion). Trust in traditional institutions can be influenced positively by membership to the majority group, but in all other cases these variables should not have significant effects. The size of the locality of residence (variable with six categories: village, commune center, small city, medium city, large city, and very large city) should have a negative effect on trust: the larger the locality, the more frequent the interactions between citizens and the state, which could generate more critical attitudes towards the state’s institutions.

Values The second group of factors that might influence the level of institutional trust is represented by respondents’ values. This group includes three variables: religiosity, acceptance of deviant behavior, and interest for politics. Religiosity is an indicator of traditionalism (which is associated with an uncritical acceptance of authority) and should have a significant positive effect on institutional trust. The variable is composed of two indicators (subjective evaluation of religiosity and frequency of going to church) and it has the following categories: atheist, non-religious person, person with low religiosity (goes to church at most once a year), person with average religiosity (goes to church at most monthly), and person with high religiosity (goes to church at least once a week).

124 corectura 2 28 iulie 2008

Institutional trust – victim of the postcommunist transition

Acceptance of deviant behavior is an additive score that includes four variables of attitudes towards different forms of deviant behavior: demanding benefits one is not entitled to receive from the state, travelling by bus or train without paying the fare, not paying taxes if one has the chance of not paying them, and accepting bribes in order to fulfill one’s job responsibilities6. Accepting such types of behavior represents, in fact, an acceptance of corruption (in different forms), which Sztompka interprets as a perverse functional substitute for trust (Sztompka 2002: 6). This variable should have a significant negative effect on institutional trust. The third variable in this group measures interest for politics and is built as an additive score that includes interest for politics and the importance of politics in the respondent’s life7. Someone who is interested in politics is better informed, understands political events better, and probably evaluates political actors (including institutions) more carefully than a person that has no interest in politics. According to this interpretation, those interested in politics should have more trust in the state’s institutions.

Social position Institutional trust can also be influenced by a person’s position in the social environment. This group of factors includes two variables: interpersonal trust and access to relationships. The relationship variable shows a person’s availability of relational social capital (measured as the number of relationships a person can use to solve a problem in nine different situations: health problems, legal problems, city hall, police, getting credit, getting a job, business problems, problems in other countries, and county institutions) 8. The availability of these relationships and, especially, identifying them as a possible source for help indicate the acceptance of a pattern of interactions with institutions that is at odds with the rules that stand at the basis of the institutions. The fact that a person acknowledges having a relationship that can help with a city hall problem shows low trust in the ability of the city hall, as institution, to solve the problem. The higher the number of relationships a person has, the lower their trust in the state’s institutions should be. Since I have already discussed the relationship between institutional trust and interpersonal trust, I will only mention here that interpersonal trust (coded 0 for those who do not trust other people and 1 for those who trust others) should have a significant positive effect on institutional trust: a person 6

Factor analysis (ML – Oblimin) extracts a single factor that explains 64.7% of the variation. KMO = 0.790. Cronbach Alpha = 0.817. The variable ranges between 0 and 36. 7 The correlation between the two components is 0.526. The variable ranges between 0 and 6. 8 Factor analysis (ML – Oblimin) extracts a single factor that explains 42% of the variation. KMO = 0.873. Cronbach Alpha = 0.827. The variable ranges between 0 and 9. 125

The Values of Romanians: 1993-2006. A sociological perspective

who believes that most people can be trusted is a more trusting person by its nature.

Psychological factors The way a respondent evaluates control over his/her own life is an indicator of locus of control. The variable I use is a 10-point scale, ranging from 0 (complete lack of control over one’s own life) to 9 (complete control over personal life). People who feel they do not have control over their own life have an external locus of control, associated with unrealistic expectations from the state. If such a person cannot solve their own problems then it will expect the state, through its institutions, to solve these problems. Flooding victims are a good example in this case. Losing one’s house during a flood is interpreted as inevitable and when this actually happens people ask the state’s institutions (from the city hall to the president) to help them, without even thinking that not the state but people themselves should be responsible for having an insurance against such a risk. It is easy to guess how much trust these people have in the state’s institutions. People having an internal locus of control will first try to solve their problems by themselves and will ask for help only if they fail. It is the latter group that should have higher levels of institutional trust. The second variable in this group measures generalized satisfaction, as an additive score composed of three indicators (subjective evaluation of happiness, satisfaction with life, and satisfaction with how one lives) showing how satisfied a respondent is with their own life9. High levels of satisfaction should be associated with high levels of institutional trust.

Evaluations Institutional trust can also be influenced by how respondents evaluate the current situation (theirs or that of others) compared to the past and by how they evaluate their future situation. Starting from the literature, I include in this group four variables corresponding to the pocketbook and sociotropic theories (for more details, see Fiorina 1981, Kiewiet 1983, and Markus 1988). The four variables are: evaluations of the current personal situations compared to the previous year (retrospective pocketbook), evaluations of the personal situation next year compared to the current one (prospective pocketbook), evaluations of the current local situation compared to the previous year (local sociotropic), and evaluations of the current national situation compared to the previous year (national sociotropic). In all four cases positive evaluations should lead to higher levels of institutional trust, although it is unlikely that all four variables will have significant effects in the same model. 9

Factor analysis (ML – Oblimin) extracts a single factor that explains 69.5% of the variation. KMO = 0.703. Cronbach Alpha = 0.780. The variable ranges between 0 and 9.6.

126 corectura 2 28 iulie 2008

Institutional trust – victim of the postcommunist transition

The effect of the efficiency of the object of trust is captured by the last variable in the model, evaluations of cabinet’s activities in different areas (standard of living, public order, jobs, agriculture, privatization, health, education, housing, industry, and environmental protection) 10. Although all variables included in this index refer to the cabinet’s activities, it should be noted that the cabinet is the state’s main representative and that its actions in different areas are mediated by different institutions. In this case, this indicator can be interpreted as an evaluation of the state’s institutions. When people decide whether or not to trust these institutions, they take into account the previous results of the institutions’ actions. Positive results should be rewarded with more trust, while negative results should be penalized with a decrease of trust. Using Easton’s terms (Easton 1965, 1975), this variable is an indicator of specific support for the state’s institutions.

Treatment of missing data The treatment of the missing values may have significant effects on the inferences drawn from the results, ranging from incorrect estimates to selection bias. Until recently, the most common methods of dealing with incomplete data were case deletion (either listwise or pairwise) and single imputation (usually mean substitution or hot deck methods). For a detailed description of the disadvantages of listwise deletion by comparison to the advantages of multiple imputation (MI) methods, see King et al (2001). Recent computational advances have reduced the costs associated with the use of multiple imputation methods, so that they have become the new standard for dealing with the incomplete data problem. The most important academic journals in the fields of sociology and political science recommend the use of these methods. Similarly, the US Census Bureau uses multiple imputation methods in order to offer to those who are interested complete datasets. For technical details on MI methods, see Little and Rubin 1987, Rubin 1987, and Schafer 1997. For my analysis I used the Norm package (Schafer 2000) to obtain five complete datasets. In each of these the missing data were imputed using information offered by the variables in the dataset. The models are estimated in each of the five complete datasets and the results are combined using the formulas proposed by Rubin (1987), formulas that take into account both the variability that exists within each complete dataset and the variability that exists among the five datasets. This method offers two significant advantages: eliminating an important source of bias and using all the information that is available in the dataset.

10

Factor analysis (ML – Oblimin) extracts a single factor that explains 52.7% of the variation. KMO = 0.913. Cronbach Alpha = 0.900. The variable ranges between 0 and 30. 127

The Values of Romanians: 1993-2006. A sociological perspective

Results Figures 1-3 present the evolution of trust in the four types of institutions described above, between 1993 and 2005. The values in these figures represent the percentage of people who declared having much or very much trust in the institutions. What do these figures tell us about Romania’s institutions in 2005? There is a clear ranking of the four types of institutions: traditional institutions enjoy the highest level of trust (around 85%), followed by the international institutions (with an average of about 60%), the civil society institutions (45% average trust), and, in the last position, the state’s institutions (only 33% of the Romanians trust them). The results for trust in the state’s institutions are worrying: after functioning for 15 years, they still have not managed to convince the citizens to trust them. These results are consistent with Uslaner and Bădescu’s conclusions, who argue that it is exactly in those cases where trust is needed the most that trust is almost inexistent (Uslaner and Bădescu, 2003: 221). The fact that citizens have the highest level of trust in non-state institutions confirms Sztompka’s ideas (Sztompka 2002: 5-6). In fact, many of the functional substitutes of trust identified by Sztompka can be found in the Romanian society: the high level of trust in the church is an indicator of providentialism, high levels of trust in international institutions indicate the externalization of trust, while the high levels of trust in political actors such as Corneliu Vadim Tudor, Gigi Becali, or Traian Băsescu, actors who always play the role of a severe but correct pater familias, indicate the replacement of trust with paternalism. The 2005 data show that two of the main institutions of the state represent the limits of trust in the state’s institutions. At one end, more than half of the Romanians trust the presidency. At the other end, only 13% of the Romanians trust the political parties. Since data for trust in these institutions in the previous years are not available the evolution of trust cannot be estimated. In the case of the presidency, it is possible that the high level of trust recorded in 2005 was influenced, to a certain extent, by the high level of trust in its incumbent, Traian Băsescu. The same interpretation applies in the case of the city hall, which seems to benefit from a similarly high level of trust in 2005: the city hall is also a highly personalized institution, as a result of its identification with the mayor.

128 corectura 2 28 iulie 2008

Institutional trust – victim of the postcommunist transition Figure 1 Trust in the state’s institutions 100 City hall Police Public servants Courts

75

Cabinet Parliament 50

25

0 1993

1997

1999

2005

With the exception of the city hall11, in all other cases institutional trust decreases over time. It should also be noted that trust in the state’s central institutions (political parties, parliament, and the cabinet) is lower than trust in local institutions (police and courts) in all years included in the analysis. This difference can be explained, in part, by the fact that citizens have more direct contacts with the local institutions compared to the central institutions. If this is true, trust in central institutions is based how their activities are reported, either by other people or by the mass media (and, as we know, good news have a reduced chance of being published or broadcasted by the mass media12). In the case of civil society institutions, the 2005 data do not show significant differences between trust in NGOs and trust in the television: almost half of the population has trust in these institutions. The available data indicate the evolution of trust over time for only two institutions in this group. In the case of trade unions there is some variation over time, but the difference between average trust in 1993 and 2005 is not significant. In both years about 30% of the population has trust in trade unions. Trust in newspapers has 11

In this case data are available only for 1993 and 2005, and the two years are not fully comparable: in 2005 respondents were asked about trust in the city hall, while in 1993the question was referring to local authorities, which can include not only the city hall but also the prefecture or other local representatives of national institutions. 12 For more details on the role of mass media in reducing trust in the state’s institutions, see Cappella and Jamieson 1997. 129

The Values of Romanians: 1993-2006. A sociological perspective

increased over time, possibly as a result of the increase in the professionalism of the newspaper journalists. The evolution of Evenimentul Zilei is a perfect example in this context: if in 1993 the newspaper was publishing news on hens that were giving birth to live chicken, by 2005 Evenimentul Zilei was described as the leader of a small group of independent newspapers (Gallagher 2005). On the other hand, although trust in newspapers has increased over time, the number of people reading them has decreased. It should also be noted that television benefits from more trust than the newspapers although, given that the most important newspapers belong to mass media trusts that also own TV stations, the newspapers and TV stations belonging to the same owner should have a similar editorial approach. Figure 2 Trust in civil society institutions 100

Television Charitable NGOs Environmental NGOs Women NGOs

75

Newspapers Trade unions

50

25

0 1993

1997

1999

2005

The traditional institutions (the church and the army) are the most trusted institutions in Romania. More than four out of five Romanians trust these institutions, with a slight increase of trust in the church. Given the characteristics of these institutions (strict hierarchy, combined with respect for order in the case of the army and ownership of the absolute truth in the case of the church), excessive trust in them, coupled with low levels of trust in more democratic institutions, could be interpreted as a sign that Romanians have not developed yet a democratic culture. Although this interpretation may be true, trust in the traditional institutions is not the best indicator to support such a conclusion. For instance, 130 corectura 2 28 iulie 2008

Institutional trust – victim of the postcommunist transition

similar levels of trust in the church and the army can also be observed in the case of American citizens, who can be described in many ways, but not as lacking a democratic culture. In a modern society (I do not discuss here how modern the Romanian society is) the main function of the two institutions is symbolic; their image in the public’s mind is that of a citizens’ protector (physical protection in the case of the army and spiritual protection in the case of the church). Figure 3 Trust in traditional and international institutions 100

75

50

Church

25

Army UN EU NATO

0 1993

1997

1999

2005

The average level of trust in international institutions (EU, UN, and NATO) has increased over time, so that by 2005 three out of five Romanians have trust in them. It is interesting to note that although the international institutions enjoy high levels of trust, it is unlikely that a large proportion of the population has a detailed understanding of these institutions’ history, structure, and functions. Given that Romania has become both a NATO and EU member, it is possible that, over time, people will know more about these institutions. The international institutions receive a bonus of trust just because they are not Romanian institutions and another bonus because citizens usually do not have any contact with them, leading to their idealization. If one takes into account the experiences of other Central and Eastern European countries, it seems likely that trust in the EU will decrease in time. The democratic deficit and the 131

The Values of Romanians: 1993-2006. A sociological perspective

excessive bureaucracy are two of the most important problems the EU faces, at least in the European public’s perceptions. In the case of the new member countries, these problems are aggravated by the unrealistic expectations their citizens had during the pre-accession period. What is the general image offered by these results? A first conclusion shows that the most trusted institutions are those that are quite separated from what we usually understand by state: Romanians have the most trust in traditional institutions, followed by extra-state institutions, then in civil society institutions, and only then in the state’s institutions. Following the evolution of trust over time, the main conclusion is that the state’s institutions are the only ones that lose trust: they fail in convincing the public they are doing their job. This should not be a surprising conclusion, given that the mass media is reporting almost daily stories about the difficulties different institutions have in following the law. The examples vary from business deals between politicians and state institutions that cannot be explained by the rules of the market economy, to businessmen that are released from prison because they develop serious illnesses as soon as they are imprisoned, to tenders whose results are known before the decision is made, to different judicial decisions in identical trials, to custom officers that are able to build houses only from their civil servants income, to students that are able to buy the subjects for the baccalaureate exams, to patients that are left to die on the street because the doctors are sending them from one hospital to another, to the whole leadership of a political party supporting a colleague by joining him on his way to an interrogation in a corruption case, to MPs sleeping in their seats or reading newspapers during parliamentary debates, to a prime minister receiving a fabulous inheritance from an aunt that made some business deals that MBA holders dream about, to a president who insults journalists, and to another president who gets out of a pub, gets behind the wheel, and drives home to the presidential palace (and who also ends up insulting journalists). Whether these stories are true or not or whether they can be justified or not is less important because once the mass media has reported them, the public uses them to revise their opinions about the institutions that are involved. I have discussed so far the way institutional trust has evolved since 1993. What are the factors that influence the level of institutional trust? The answer I offer is based on the theoretical model presented at the beginning of the chapter. Tables 1-4 present the results of regression analyses having as dependent variables institutional trust in the four types of institutions13. 13

I estimated the models in SPSS 14.0.2. using the OLS (Ordinary Least Squares) regression. The starting model includes as independent variables only the socioeconomic characteristics of the respondents. For each group of variables I introduced in the analysis, I estimated a new model. The coefficients in the table are unstandardized regression coefficients (b). The last row in the table (R2 change) shows the increase in 132 corectura 2 28 iulie 2008

Institutional trust – victim of the postcommunist transition

The six groups of variables included in the last model in Table 1 explain, together, about 22% of the variation of trust in the state’s institutions. Each of the six groups contributes to explaining the level of trust, but evaluations of the activities of the state’s institutions have an effect that is twice as large as that of any of the remaining groups of independent variables. Among the group of socio-economic characteristics, gender, income, religion and ethnicity do not have significant effects. Age has a significant positive effect, suggesting that older people tend to trust the state’s institutions more. Education and size of locality have significant negative effects. Based on these characteristics, people can be placed at different positions on the trust – distrust continuum, having at one end older people, with low education, and living in rural areas (these people have high levels of trust in the state’s institutions) and, at the other end, young people, highly educated, and living in very large cities (people with low levels of trust in the state’s institutions). To a certain extent, the differences between these two types of respondents correspond to the distinction between traditionalism and modernity: the more traditional a person is, the higher his/her tendency is to respect authority. As I argued before, the initial position, which is determined by socioeconomic characteristics can be changed by other variables and this is exactly what the results presented here indicate. A person’s values have significant effect on their level of institutional trust. People with a permissive attitude toward deviant behavior (different forms of corruption) have less trust in the state’s institutions, suggesting that, indeed, corruption can replace trust in environments characterized by a generalized crisis of trust. People who are interested in politics tend to have more trust in the state’s institutions. A possible explanation for the positive effect of this variable consists of the fact that the lack of trust in institutions is partially determined by the lack of information regarding the institutions and their actions. The state’s institutions suffer from a significant image deficit, which can be explained by the citizens’ lack of interest and by mass media’s tendency to exaggerate the institutions’ real problems. It should be noted that religiosity, another indicator of traditionalism, loses its significance when the evaluation variables are introduced in the model.

the explanatory power of the current model compared to the previous model. The coefficients on this row are a measure of the importance of different groups of variables in explaining diffuse support for democracy. For instance, in Table 1, Model 6 differs from Model 5 by the variable of evaluations of cabinet’s activities. By adding this variable, compared to Model 5, the explanatory power of Model 6 increased with 7.3%, from 14.9% to 22.2%. 133

The Values of Romanians: 1993-2006. A sociological perspective Table 1 Trust in the state’s institutions (2005) Model 1

Model 2

Model 3

Model 4

Model 5

Model 6

Intercept

7.894***

6.876***

6.615***

5.157***

1.938*

1.567+

Male

0.334

0.178

0.155

0.170

0.173

0.208

Age

0.148*

0.111

0.083

0.148*

0.211**

0.167*

-0.214+

-0.316*

-0.398**

-0.363**

-0.291**

0.089+

0.093+

-0.050

-0.059

-0.076

Education Income

-0.348** 0.074

Size of locality

-0.289***

-0.269***

-0.229***

-0.214***

-0.209***

-0.222***

Religion: Orthodox

-0.557

-0.522

-0.465

-0.310

-0.302

-0.188

Ethnicity: Romanian

-0.133

-0.340

-0.135

-0.234

-0.173

-0.099

Religiosity

0.343**

Acceptance of corruption

-0.037*

Interest in politics

0.457***

0.360**

0.226+

0.219+

-0.049**

-0.044**

-0.042*

0.140 -0.057***

0.443***

0.380***

0.354***

0.285***

Relationships

0.157*

0.083

0.065

0.025

Interpersonal trust

1.754***

1.598***

1.337***

1.189***

Satisfaction

0.461***

0.298***

0.117+

Control over life

0.051

0.037

0.054

Future personal situation

0.571***

0.413**

Evaluation of national situation

0.833***

0.548***

Evaluation of institutions 2

0.277***

Adjusted R

2.7%

5.6%

8.3%

11.5%

14.9%

22.2%

R2 change

2.7%

2.9%

2.7%

3.2%

3.4%

7.3%

Notes:

The coefficients in the table are unstandardized regression coefficients. Significance levels: *** p < 0.001 ** p < 0.010 * p < 0.050 + p < 0.100.

The respondent’s position in the social space influences trust in the state’s institutions through the interpersonal trust variable. Trusting people also have a higher level of institutional trust, compared to those who consider that most people cannot be trusted. The availability of a network of relationships does not have a significant effect in the final model. Out of the psychological variables, only satisfaction has a significant effect, but this is marginal. The fact that its significance decreases when evaluation variables are taken into account suggests that satisfaction depends, in part, on these evaluations. The evaluation variables have strong positive effects. Positive evaluations of the future personal situation increase the level of trust in the 134 corectura 2 28 iulie 2008

Institutional trust – victim of the postcommunist transition

state’s institutions. Out of the three variables of retrospective evaluations (personal, local, and national), the one with the national referent has a significant positive effect: when taking into account last years’ changes in the decision to trust, people think of changes at the national level, rather than local or personal changes. The last variable in the model, evaluations of the cabinet’s actions in different areas, has the strongest effect on the dependent variable, being able to modify the level of trust by up to 30%. This result shows that trust in the state’s institutions depends mainly on institutional performance. It should be also noted that each of the six groups of independent variables brings its own contribution to explaining the level of trust in the state’s institutions. This is shown by significant differences between the multiple determination coefficients of the six models and by the fact that most of the variables maintain their significance in different models. In discussing the results of the other three dependent variables, I focus on differences from trust in the state’s institutions. Trust in civil society institutions has different determinants, as indicated by the results presented in Table 2. The final model has a smaller explanatory power than in the previous case (12.4%, compared to 22.2%) but the evaluations of the actions of the state’s institutions continue to be the most important predictor in the model. The size of the locality of residence is the only significant variable from the group of socio-economic characteristics, suggesting there are no significant differences between population sub-groups with respect to civil society institutions. Religiosity has only a marginally significant positive effect, which could probably be explained by the fact that charitable NGOs are among the civil society institutions included in the analysis. Acceptance of deviant behavior is no longer significant, but interest for politics has a significant positive effect on trust in civil society institutions as well. Interpersonal trust has a significant positive effect, but smaller than in the case of trust in the state’s institutions. Theoretically, given that interpersonal trust is closely related to the associative behavior that characterizes the civil society, this coefficient should have been more important in determining trust in civil society institutions. At the same time, previous studies have argued that regardless of the level of interpersonal trust, active participation in NGO activities (which could be interpreted as an effect of trusting these institutions) is reduced in former communist countries (see, for instance, Howard 2003 and Mihaylova 2004). The three evaluation variables continue having significant effects, but their coefficients are smaller than in the case of trust in the state’s institutions. This is not surprising, given the state – civil society distinction. Overall, the theoretical model explains to a lesser extent trust in civil society institutions, suggesting its possible incorrect specification for these types of institutions. Comparing the results, it can be seen that trust in civil society and trust in state institutions have different determinants. Since civil society is a separate domain 135

The Values of Romanians: 1993-2006. A sociological perspective

from the state, it is possible that trust in its institutions is influenced by cultural factors that are not included in the model. The weak performance of this model can also be explained by the fact that civil society in communist Romania was severely underdeveloped. If Poland had Solidarność, which managed to attract ten million members, if Czechoslovakia had Charta 77, which played a crucial role in the Velvet Revolution, if in Hungary the population mobilized against the communist regime in 1956, such movements were stopped in Romania before they could develop. Table 2 Trust in civil society institutions (2005)

Intercept

Model 1

Model 2

Model 3

Model 4

Model 5

Model 6

6.758***

5.656***

5.533***

4.937***

3.303***

3.078***

Male

-0.083

-0.190

-0.200

-0.195

-0.193

-0.172

Age

0.041

0.025

0.012

0.035

0.068

0.041

-0.008

-0.084

-0.097

-0.118

-0.100

-0.056

0.066

0.060

0.050

-0.118*

-0.113*

-0.110*

-0.118*

Education Income Size of locality

0.121**

0.119**

-0.152**

-0.138**

0.111**

Religion: Orthodox

0.038

0.093

0.120

0.175

0.178

0.247

Ethnicity: Romanian

0.062

-0.015

0.078

0.037

0.066

0.111

Religiosity

0.276**

0.284**

0.234*

0.230*

0.181+

Acceptance of corruption

0.015

0.009

0.012

0.012

0.004

Interest in politics

0.334***

0.327***

0.302***

0.288***

0.247***

Relationships

0.066

0.039

0.029

0.005

Interpersonal trust

0.853***

0.788***

0.655**

0.565**

Satisfaction

0.167***

0.083

-0.027

Control over life

0.031

0.023

0.034

Future personal situation

0.299*

0.202+

Evaluation of national situation

0.417***

0.243*

Evaluation of institutions 2

0.169***

Adjusted R

1.1%

3.8%

4.9%

5.8%

7.4%

12.4%

R2 change

1.1%

2.7%

1.1%

0.8%

1.6%

5.0%

Notes:

The coefficients in the table are unstandardized regression coefficients. Significance levels: *** p < 0.001 ** p < 0.010 * p < 0.050 + p < 0.100.

136 corectura 2 28 iulie 2008

Institutional trust – victim of the postcommunist transition

The results in Table 3 refer to trust in the church and the army. The final model explains approximately 21% of the variation of trust in traditional institutions, most of it being explained by only two groups of variables: socioeconomic characteristics and value (these two explain 19%). Table 3 Trust in traditional institutions (2005) Model 1

Model 2

Model 3

Model 4

Model 5

Model 6

3.447***

2.392***

2.383***

2.148***

2.185***

2.149***

0.031

0.029

0.031

0.031

0.034

0.107***

0.088***

0.085***

0.093***

0.091***

0.086***

Education

-0.121***

-0.134***

-0.127***

-0.135***

-0.134***

-0.127***

Income

-0.009

-0.007

-0.003

-0.017

-0.017

-0.018

Size of locality

-0.121***

-0.111***

-0.108***

-0.107***

-0.107***

-0.108***

Intercept Male

-0.053

Age

Religion: Orthodox

0.106

0.231*

0.228*

0.245*

0.248*

0.260*

Ethnicity: Romanian

0.580***

0.476***

0.462***

0.446***

0.449***

0.456***

0.365***

0.366***

0.348***

0.348***

0.341***

Religiosity Acceptance of corruption

-0.008+

Interest in politics

0.052**

-0.008+ 0.053**

-0.007

-0.007

-0.008+

0.044*

0.045*

0.038*

Relationships

-0.031+

-0.040*

-0.040*

-0.044*

Interpersonal trust

0.144*

0.119

0.124+

0.109

Satisfaction

0.052**

0.056**

0.038*

Control over life

0.019

0.020

0.021

Future personal situation

-0.028

-0.043

Evaluation of national situation

0.005

-0.023

Evaluation of institutions Adjusted R2 2

R change

Notes:

0.028*** 12.8%

18.9%

19.3%

20.0%

20.0%

20.9%

12.8%

6.2%

0.3%

0.7%

0.0%

0.9%

The coefficients in the table are unstandardized regression coefficients. Significance levels: *** p < 0.001 ** p < 0.010 * p < 0.050 + p < 0.100.

Just like in the model explaining trust in the state’s institutions, age, education, and the size of the locality have significant effects on trust: older people, with low levels of education, and who live in rural areas have more trust in the church and the army than young, educated urbanites. The most important difference is given by the coefficients for religion and ethnicity, which have 137

The Values of Romanians: 1993-2006. A sociological perspective

significant effects in this model. Romanians have more trust in the church, compared to people of other ethnicities. Similarly, Christian Orthodox respondents have more trust in traditional institutions than people belonging to other religions. If we consider the connections that exist in the public’s psyche between the idea of a Romanian state and the church and the army, as well as the fact that religion seems to be one of the main components included in the definition of being Romanian, then it is simple to understand why those belonging to the majority group trust these institutions more. The respondent’s religiosity also increases trust in the church and the army. The other variables included in the model either do not have significant effects or have effects that are statistically significant but that can be ignored. The final model in the analyses presented in Table 4 explains 13.6% of the variation of trust in international institutions. If in the previous models gender did not have a significant effect on the level of trust, in this case, compared to women, men have more trust in international institutions. It is possible this effect is resulting from the fact that men tend to be more interested in foreign affairs, obtaining, thus, more information about the international institutions, information that can increase their level of trust. The remaining socio-economic characteristics do not distinguish between different levels of trust in international institutions. Among the group of variables indicating the respondents’ values, interest for politics is the only one having a significant (positive) effect. The evaluation variables are the most important predictors in the model: the evaluations of the personal situation and of the national situation can change the level of trust by up to a point. Similarly, the evaluation of the cabinet’s activities can modify the level of trust by up to three points. Comparing the results presented in Tables 2 and 4, it can be observed that trust in civil society institutions and trust in international institutions have similar determinants (which was expected, given that the two dependent variables are correlated at 0.523).

138 corectura 2 28 iulie 2008

Institutional trust – victim of the postcommunist transition Table 4 Trust in international institutions (2005) Model 1

Model 2

Model 3

Model 4

Model 5

Model 6

Intercept

4.021***

3.581***

3.489***

2.856***

1.544**

1.408 **

Male

0.407***

0.309**

0.300**

0.304**

0.305**

0.318 **

Age

0.017

0.003

-0.009

0.006

0.035

0.019

Education

0.103+

0.050

0.049

0.027

0.039

0.066

Income

0.049+

0.050+

0.049+

0.017

0.012

0.005

Size of locality

-0.032

-0.023

-0.007

-0.004

-0.001

-0.006

Religion: Orthodox

-0.057

-0.055

-0.039

-0.004

-0.009

0.033

Ethnicity: Romanian

-0.072

-0.149

-0.101

-0.144

-0.128

-0.101

0.091

0.087

0.058

-0.011

-0.010

-0.015 +

Religiosity

0.125+

Acceptance of corruption

-0.010

Interest in politics

0.238***

0.132* -0.014 0.234***

Relationships

0.013

Interpersonal trust

0.715***

0.210*** -0.008

0.197*** -0.017

0.172 *** -0.031

0.646***

0.536***

Satisfaction

0.107**

0.036

Control over life

0.065*

0.058*

0.064 *

Future personal situation

0.281**

0.223 **

Evaluation of national situation

0.305***

0.201 **

Evaluation of institutions Adjusted R2 2

R change

Notes:

0.481 *** -0.031

0.102 *** 1.8%

4.4%

6.0%

7.3%

9.6%

13.6%

1.8%

2.7%

1.6%

1.3%

2.3%

3.9%

The coefficients in the table are unstandardized regression coefficients. Significance levels: *** p < 0.001 ** p < 0.010 * p < 0.050 + p < 0.100.

Conclusions As I have shown in this chapter, there are clear differences in institutional trust, depending on the type of institutions that are analyzed, both with respect to the level of trust and with respect to the determinants of trust. The 19 institutions included in the analysis can be grouped, theoretically, into four distinct categories, solution that is confirmed by the results: state institutions, civil society institutions, traditional institutions, and international institutions. Each of the four types has a series of characteristics that distinguish 139

The Values of Romanians: 1993-2006. A sociological perspective

it from the others. The state’s institutions are primarily institutions with attributions in the political and legal domains. Civil society institutions define themselves by opposition to the state. Traditional institutions symbolize the state and are characterized by strict hierarchies (for a similar interpretation of the church and the army, see Sandu 1999 and Voicu 2005). Finally, international institutions are external actors. These differences are reflected at the public level in different levels of institutional trust. Romanians have the highest levels of trust in traditional institutions, followed by international institutions, civil society institutions, and the state’s institutions. The state’s institutions are, at the same time, the only ones that lose trust over time. These results, combined, suggest a significant crisis of trust in the state. If this crisis continues for a sufficiently long period of time, it can generate problems for the democratic system in Romania. Trust in the state’s institutions can be interpreted as specific support for democracy. From this perspective, long periods of reduced specific support can contaminate diffuse support for democracy (for a detailed discussion of this perspective, see the chapter on democratization in this volume). Fortunately, the low levels of trust in the state’s institutions are strongly influenced by how people and mass media evaluate the institutions’ performance, suggesting that the improvement of institutional performance can lead to an increase of institutional trust. One should also keep in mind that the decrease of trust in the state’s institutions is not unique to Romania. Similar trends have been observed both in the former communist countries and in developed democracies from the Western world. There are, however, significant differences related to the causes of this decreasing trend. In the established democracies the decrease of trust in the state’s institutions is determined primarily by changes in the citizens’ expectations. Low levels of trust indicate, in these countries, the existence of a significant number of “critical citizens” (see Norris 1999) or “discontent democrats” (see Dalton 2005). In Romania (and, probably, in the other new democracies as well) the decrease of trust in the state’s institutions is determined primarily by the institutions’ inefficient functioning. Thus, while citizens of the established democracies have been satisfied with the institutions of the state and know they can function better, the experiences of the postcommunist citizens with the institutions of the state have generally been negative, leading to a skeptical attitude toward the institutions’ abilities. How will institutional trust evolve in Romania? The results in Table 1 show that an important part of the variation of trust in the state’s institutions is positively influenced by the citizens’ evaluations of their personal situation, of the national situation, and of institutional performance. During the last years, one could observe signs of improvement both in the state of the national economy and in the performance of the institutions. If these improvements will continue, then one can expect a reversal of the decreasing trend that has characterized trust in the state’s institutions starting with 1990 and the 140 corectura 2 28 iulie 2008

Institutional trust – victim of the postcommunist transition

elimination of one of the most important dangers to the Romanian democracy. The increase in institutional trust can lead to an increase in support for democracy. Traditional institutions enjoy at this moment such high levels of trust that further increases are unlikely. There are, however, two factors that can decrease trust in the church and the army. The Romanian mass media recently accused some of the clerics of being former collaborators of the secret police during the communist regime. To the extent these accusations will prove to be true and if there were many cases of collaboration, it is possible that trust in the church will decrease. With respect to the army, its importance in the collective psyche will decrease as a result of Romania becoming a NATO member (which implied transferring part of the army’s protection function from the Romanian army to NATO), and of eliminating the mandatory military service. These factors have the potential to decrease trust in the army. Among the group of international institutions, the European Union has become, once Romania joined the EU, as important for the Romanians’ life as the state’s institutions. Based on the experiences of other EU members, and given that a large proportion of the Romanian population had unrealistic expectations from the EU, it is very likely that trust in the EU will decrease (this can be accelerated by the almost instantaneous apparition of Romanian euroskeptics only days after Romania joined the EU). Returning to the title of this chapter, it can be argued that although institutional trust has certainly been a victim of the postcommunist transition, there are some signs that suggest that the victim may be getting better (but it is unlikely to observe a complete recovery).

References Almond, Gabriel, and Sidney Verba. 1965. The Civic Culture: Political Attitudes and Democracy in Five Nations. Boston: Little Brown. Beck, Ulrich. 1992. Risk Society: Towards a New Modernity. London: Sage Publications. Cappella, Joseph, and Kathleen Jamieson. 1997. Spiral of Cynicism: The Press and the Public Good. Oxford: Oxford University Press. Coleman, James Samuel. 1990. Foundations of Social Theory. Cambridge: Harvard University Press. Coleman, James Samuel. 1999. „Social Capital in the Creation of Human Capital.” In Social Capital: a Multifaceted Perspective, edited by Partha Dasgupta and Ismail Serageldin. Washington: World Bank. Dalton, Russell. 2005. „The Social Transformation of Trust in Government.” International Review of Sociology. 15 (1): 133-154.

141

The Values of Romanians: 1993-2006. A sociological perspective Dasgupta, Partha. 1988. „Trust as a Commodity.” In Trust: Making and Breaking Cooperative Relations, edited by Diego Gambetta. New York: Basil Blackwell. Easton, David. 1965. A Systems Analysis of Political Life. New York: Wiley. Easton, David. 1975. “A Re-Assessment of the Concept of Political Support.” British Journal of Political Science. 5 (4): 435-457. Ekiert, Grzegorz, and Jan Kubik. 1999. Rebellious Civil Society: Popular Protest and Democratic Consolidation in Poland, 1989-1993. Ann Arbor: University of Michigan Press. Fiorina, Morris. 1981. Retrospective Voting in American National Elections. New Haven: Yale University Press. Fukuyama, Francis. 1995. Trust: Social Virtues and the Creation of Prosperity. New York: Free Press. Gallagher, Tom. 2005. „Romania: Silencing the Media.” Transitions Online. Available online at http://www.projecttransitionaldemocracy.org/document.php?docid=844. Gambetta, Diego. 1988. „Can We Trust Trust?” In Trust: Making and Breaking Cooperative Relations, edited by Diego Gambetta. New York: Basil Blackwell. Giddens, Anthony. 1991. Modernity and Self-Identity: Self and Society in the Late Modern Age. Stanford: Stanford University Press. Hardin, Russell. 1998. „Trust in Government.” In Trust and Governance, edited by Valerie Braithwaite and Margaret Levi. New York: Russell Sage Foundation. Howard, Marc Morjé. 2003. The Weakness of Civil Society in Post-Communist Europe. Cambridge: Cambridge University Press. Inglehart, Ronald. 1997. Modernization and Postmodernization: Cultural, Economic, and Political Change in 43 Societies. Princeton: Princeton University Press. Kiewiet, Roderick. 1983. Macroeconomics & Micropolitics: The Electoral Effects of Economic Issues. Chicago: University of Chicago Press. King, Gary et al. 2001. „Analyzing Incomplete Political Science Data: An Alternative Algorithm for Multiple Imputation.” American Political Science Review 95 (1): 49-69. Little, Roderick, and Donald Rubin. 1987. Statistical Analysis with Missing Data. New York: Wiley & Sons. Luhmann, Niklas. 1979. Trust and Power: Two Works. New York: Wiley & Sons. Luhmann, Niklas. 1988. „Familiarity, Confidence, Trust.” In Trust: Making and Breaking Cooperative Relations, edited by Diego Gambetta. New York: Basil Blackwell. Macovei, Monica. 1998. „Legal Culture in Romania.” East European Constitutional Review. 7 (1). Markus, Gregory. 1988. „The Impact of Personal and National Economic Conditions on the Presidential Vote: A Pooled Cross-Sectional Analysis.” American Journal of Political Science. 32 (1): 137-154. Mihaylova, Dimitrina. 2004. Social Capital in Central and Eastern Europe: A Critical Assessment and Literature Review. Budapest: Center for Policy Studies Central European University.

142 corectura 2 28 iulie 2008

Institutional trust – victim of the postcommunist transition Mishler, William, and Richard Rose. 1997. „Trust, Distrust and Skepticism: Popular Evaluations of Civil and Political Institutions in Post-Communist Societies.” Journal of Politics 59 (2): 418-451. Mishler, William, and Richard Rose. 2001. „What Are the Origins of Political Trust? Testing Institutional and Cultural Theories in Post-Communist Societies.” Comparative Political Studies 34 (1): 30-62. Misztal, Barbara. 1996. Trust in Modern Societies: The Search for the Bases of Social Order. Cambridge: Polity Press. Norris, Pippa. 1999. Critical Citizens: Global Support for Democratic Government. Oxford: Oxford University Press. Offe, Claus. 1999. „How Can We Trust Our Fellow Citizens?” In Democracy and Trust, edited by Mark Warren. Cambridge: Cambridge University Press. Putnam, Robert. 1993. Making Democracy Work: Civic Traditions in Modern Italy. Princeton: Princeton University Press. Putnam, Robert. 1995. „Bowling Alone: America’s Declining Social Capital.” Journal of Democracy. 6 (1): 65-78. Rose, Richard, William Mishler, and Christian Haerpfer. 1998. Democracy and Its Alternatives: Understanding Post-Communist Societies. Baltimore: Johns Hopkins University Press. Rubin, Donald. 1987. Multiple Imputation for Nonresponse in Surveys. New York: Wiley & Sons. Ruscio, Kenneth. 1999. „Jay's Pirouette, or Why Political Trust Is Not the Same as Personal Trust.” Administration & Society 31 (5): 639-657. Sandu, Dumitru. 1999. Spaţiul social al tranziţiei. Iaşi: Polirom. Schafer, Joe. 1997. Analysis of Incomplete Multivariate Data. London: Chapman & Hall. Schafer, Joe. 2000. NORM: Multiple Imputation of Incomplete Multivariate Data under a Normal Model, Version 2.03. Software available at http://www.stat.psu.edu/~jls/misoftwa.html. Seligman, Adam. 1997. The Problem of Trust. Princeton: Princeton University Press. Sztompka, Piotr. 1999. Trust: A Sociological theory. Cambridge: Cambridge University Press. Sztompka, Piotr. 2002. Trust: A Cultural Resource. Available online at http://www.colbud.hu/honesty-trust/sztompka/pub01.pdf. Tyler, Tom, and Peter Degoey. 1996. „Trust in Organizational Authorities: The Influence of Motive Attributions on Willingness to Accept Decisions.” In Trust in Organizations: Frontiers of Theory and Research, edited by Roderick Kramer and Tom Tyler. London: Sage Publications. Uslaner, Eric, and Gabriel Bădescu. 2003. „Legacies and Conflicts: The Challenges to Social Capital in the Democratic Transitions.” In Social Capital and the Transition to Democracy, edited by Gabriel Bădescu and Eric Uslaner. London: Routledge. Voicu, Bogdan. 2005. Penuria Pseudo-Modernă a Postcomunismului Românesc: Resursele. Iaşi: Expert Projects. Warren, Mark. 1999. „Introduction.” In Democracy and Trust, edited by Mark Warren. Cambridge: Cambridge University Press.

143

Religiosity and Religious Revival during the Transition Period in Romania MĂLINA VOICU

Romania is one of the most religious countries in Europe. The existing literature provides many quantitative empirical evidences on this topic (Gheorghiu, 2003; Halman, Draulans, 2004; Muller, 2004; Pollack, 2001, 2003, 2004). Likewise, the researches focusing on the evolution of the religious field in post-communist countries indicate the existence of a religious revival phenomenon (Inglehart, Norris, 2004; Pollack, 2004; Froese, 2005; Kääriäinen, 1999). The uplifting of the restrictions imposed on religious practice by the communist regime, as well as the increasing insecurity regarding everyday life are factors contributing to the enhancement of religious revitalization. This paper aims at analyzing from a both longitudinal and transversal comparative perspective the degree of religiosity in Romania. To put it simply, we are trying to provide an answer to the following question: are Romanians indeed more religious than other Europeans, if we are to take religious belief and shared values as a reference point? By the same token, we may produce another question: if they are indeed more religious, why is it so? What exactly determines the differences in degree of religiosity of the various European people? A third target of this chapter would be to identify the causes leading to a religious revival in post-communist countries, especially in Romania. I shall, in a practical way, try to come up with an answer to this question: why are Romanians more religious today than they were a decade ago? This paper consists of four different sections. The first one is dedicated to theories that could help us in our search for an answer regarding the phenomenon determining the degree of religiosity in a people. The second section is a presentation of the indicators used in the analysis I have been carrying out in this chapter. The third part deals with data analysis and provides

144 corectura 2 28 iulie 2008

Religiosity and Religious Revival during the Transition Period in Romania

comments on the obtained results, while the final part is mainly a discussion on both results and their deriving implications.

Determinants of Religious Diversity in Eastern Europe Religious Europe may be viewed as a huge puzzle, containing points in which secularization has deeply invaded the social structure and areas in which religious continues to be an important part of people’s lives. The divisions are structured according to several lines. Geographically speaking secularized Northern Europe stands in contrast with the Southern one, which is highly religious. Catholic Europe has been influenced by secularization to a smaller extent than protestant Europe. For explaining the European cross-country differences of the secularization levels, one may employ various approaches. Besides the geographic position and dominant religion, the studies identified a series of other factors to consider. The classical secularization theory states that secularization appears as a natural outcome of the modernization and functional specialization processes, which are rooted in the high economic development level. (Chaves, 1994; Sommerville, 1998; Wilson, 2000). The more the society develops economically and develops towards industrialization and postindustrialization, the more functionally specialized the society becomes, as every domain produces its own functioning principles. This process leads to a diminishing social influence of religion, which loses its position as a dominant factor at a social level. Society tends to gradually separate from religion, becomes indifferent towards the latter and starts functioning as if “God didn’t exist” (Schlegel, 2005, p. 42). Nevertheless, Inglehart, Halman, Welzel (2003) consider that, although economic development stimulates socio-cultural change towards a unique direction, each society evolves in a particular, specific way. Socio-cultural and, implicitly, religious change is influenced by the level of economic development, but a series of other factors play an important part in this respect. These are political, social, technological and geographic factors, practically comprising the entire history of a population. In other words, economic development practically stimulates secularization, but at an equal development level we may encounter countries with a different degree of secularization. The differentiation is produced by the broader social and political context. On the other hand, the experience of contemporary societies does not entirely confirm the thesis according to which there should be a direct association between economy and religion. In this respect, a counterexample is provided by the Arabic countries of the Middle East, which are big producers of petroleum, with a considerable affluence level but a small degree of secularization and a smaller social development level, as compared to European 145

The Values of Romanians: 1993-2006. A sociological perspective

countries. The key to socio-cultural transformations does not reside as much in economic development as in a more extensive process of social development. The key of socio-cultural transformations does not reside as much in economic development, as in a more extended process of social change. What is important is not solely the economic growth, but the way in which it is transposed socially and moulded, in its turn, by society. The high level of material prosperity creates the premises of social and cultural change, but the change in values influences the economic conditions (Inglehart, Halman, Welzel, 2003). Which would be the immediate implication of this interdependence of economy and society for religiosity? A direct consequence is the emergence of secularization in strongly economic developed societies. But economy does not explain everything as we have shown above. According to Inglehart and Norris (2004) the relation between religiosity and economy is practically mediated by the feeling of human security and human vulnerability when faced with risks. People living in better off societies, where individuals are exposed to rather low social risks and where the social environment is quite predictable, are more exposed to secularization. In exchange, people living in poor societies, mainly concerned with agriculture and exposed to a less predictable environment will experience the need to subordinate to a religious authority that may govern their existence, shelter them from the risks they are exposed to and guarantee a better life, if not in this world, at least in the next. The surveys approaching this matter have demonstrated the existence of an association between the degree of economic development and religious values, proving that poor societies are usually described as being more religious (Pollack, 2003; Inglehart, Halman, Welzel, 2003). The same Inglehart (1971, 1990, 1997) shows that the experiences encountered in the early stages of socialization play an important part in the individuals’ orientation, as far as values are concerned and consequently also determine his/her degree of religiosity. People who are socialized during periods of economic and social insecurity will be more inclined towards order and subordination to authority. Those who are socialized in periods of economic and social welfare, as in the case of the large number of people born after World War II in Western Europe and North America are more inclined towards independence and self affirmation, post-materialist values which seldom maintain a good relationship with traditional religiosity. On the other hand, the change in the social context in which the mature individuals lives leads to the modifying of his/her value orientations. A period of economic and social instability leads to the re-orientation of the post-materialists towards materialistic values, like the subordination to an external authority, be it religious or not. In the ex-communist states of Central and Eastern Europe the decreasing social influence of religion had a double determination under the 146 corectura 2 28 iulie 2008

Religiosity and Religious Revival during the Transition Period in Romania

communist regime. On the one hand, modernization, urbanism, industrialization destroyed the traditional structures of society in which religion played an essential part. On the other hand, religious persecutions and the promotion of scientific atheism eliminated religion and religious institutions out of public life, at least. Nevertheless, there have been differences in the degree of religiosity of the people of the ex-communist countries, although the politics of atheism promotion was common in the entire area. The fall of the communist regime in the region determined the onset of deep social, political and economic changes. The renouncement of the centralized economy, the privatization and the passing to the market economy triggered a state of economic recession in the area. But, as in the cases presented above, the recession was not equally powerful in all states – some suffered more and some less from these transformations. Although having a noble aim, mainly the approach to democracy and free market economy, the transition period translated into unemployment, inflation, a decreasing life standard and material insecurity for many citizens of Eastern Europe. Moreover, the society of postcommunist states is a double risk society (Rinkevicius, 2000). On the one hand, we are dealing with the risks of the “classical disease”, like poverty, disease, unemployment, the acute lack of resources caused by economic transition. On the other hand, we may refer to a series of more “modern” risks, like environment pollution or nuclear energy. All of these create a growing degree of existential insecurity and may lie at the foundation of religious revitalization which was reported in Central and Eastern Europe after the fall of the communist regimes (Inglehart, Norris, 2004; Pollack, 2004; Froese, 2005; Kääriäinen, 1999). As a contradiction to the secularization thesis, the supply-side thesis, deriving from the economy, was elaborated. The adepts of this thesis (Finke, 1990; Finke & Stark, 1988; Iannaccone, 1991; Stark & Iannaccone, 1994) claim that the secularization thesis is contradicted by reality. Starting from the case of the United States, where the level of religiosity is a lot higher than the secularization theory would predict, the authors consider that economic development and modernization do not represent the major cause of religious decline in Europe. By contrasting the United States and Europe one may observe that while Northern America is characterized by both high religious diversity and also a high level of religiosity, in Europe the monopoly on the market of religious goods is associated with the strong secularization of society. The explanation of the difference between these two cases resides in the fact that while the American market of religious goods is free, in Europe traditional religions own the monopoly. According to the supply-side theory, a pluralist and free market of religious goods stimulates religious institutions to diversify their offers and to specialize. Thus, each person can find something that corresponds to his/her spiritual motivations and, consequently, adhere to a certain religion. To the 147

The Values of Romanians: 1993-2006. A sociological perspective

extent to which the state does not intervene and allows the market of religious goods to develop freely, the latter attracts more and more people towards religion. If the market is monopolized by one single dominant religion supported by the state, like in the Northern Protestant states, the offer is very limited, and those who do not find the elements that should satisfy their spiritual needs in the dominant religion tend to distance themselves from religion and church. The situation of Eastern countries during the communist regime was a special one from this point of view. On the one hand, the market of religious goods was blocked by the communist regime which has encouraged in some country’s religious monopolies of certain churches, like the case of the Orthodox Church in Romania or Russia. On the other hand, the communist government was attempting to impose its own religion – scientific atheism (Froese, 2004a, 2004b). The communist period represented a drawback for the religious market, but the liberalization that took place after 1990 opened the market for new religious institutions, so that individuals had the possibility of a wider choice. Nevertheless, Inglehart and Norris (2004) show that the analysis of empirical data does not support this hypothesis. A similar conclusion was reached by Halman and Draulans (2004) who tested the supply-side theory in the case of European countries. Beyond the effect of secularization and religious market liberalization, the uplifting of the restrictions imposed by the communist regime on religion had an expected impact on individual religiosity. The fall of the Marxist orientation regimes created the possibility of public manifestation of religiosity and opened the way for religious education and propaganda. The emergence of discussions on religious topics in the media, as well as the setting up of a religious educational system allowed the penetration, at the population level, of religious knowledge and created the premises for religious revival. Beyond the effects of economic decrease or of the liberalization of the market of religious goods, the uplifting of the restrictions imposed on religious practice may have had a contribution to the triggering of religious revitalization. By applying the facts presented above to the Romanian case, one may state that, taking into account the thesis of secularization under the influence of social development, one should expect Romania’s level of religiosity to be superior to that of strongly industrialized countries in Central and Eastern Europe, like the Czech Republic or Eastern Germany. If the level of religiosity depends on the level of social development, then Romania will be more religious than more developed countries and less religious than the less developed ones. However, one must keep in mind that human and social development does not just refer to more money, but to a more extended human capital. The analysis carried out in this chapter shall relate the level of religiosity not only to material welfare indicators, but also with those of human development. 148 corectura 2 28 iulie 2008

Religiosity and Religious Revival during the Transition Period in Romania

On the other hand, the social development impact on religious values theory states that a diminishing economic security and a growth of economic and social risks generates a change in values, so it can lead to religious revitalization. In Romania the transition period generated more poverty and existential insecurity than in countries like Eastern Germany and the Czech Republic. Under these circumstances one should expect the religious revival in Romania to be stronger as compared to one of the countries where the economic transformations generated less insecurity. Taking into account the supply-side theory, one should expect the religious revitalization experienced by this area and by Romania in particular to originate in the liberalization of the religious good’s market which took place after the fall of the communist regimes. By the same token, an association is expected to exist between religiosity and the level of religious diversity. In countries with an increased religious diversity the level of religiosity is expected to be higher, as compared to countries where there is a certain religious monopoly. At a first glance, Romania’s case contradicts this theory, as religious diversity is low enough (86% of the population is Orthodox) but the indicators of religious implication have high enough values. On the other hand, even if Romania does not confirm this hypothesis one should consider the association of the degree of religious revitalization of a country and religious diversity. Given this situation, religious revival may be the effect of the opening of the market of religious goods, without any influence of pluralism on the level of religiosity. Pollack (2001) shows that there are four factors influencing the degree of secularization of the post-communist societies: the degree of modernization, political oppression, dominant denomination and the approach to national ideas. Consequently, in the countries where the political oppression of religion was milder, churches managed to survive and maintain a certain ecclesiastic life. Religious values, too, where better kept by the population. Moreover, the surveys carried out after the fall of the communist regime showed that the religion that best survived communism was the Catholic one, supported by the international structure of the Catholic Church. (Pollack, 2001, 2003; Need, Evans, 2001; Froese, 2004a; Stark, 2001; Bruce, 2001). Except for the dimensions of political oppression, the type of dominant religion, the closeness of religion and the national ideas contributed to the preservation of a high degree of religiosity among the population, as religion was perceived as a keeper of national identity and a shelter from the bolshevik danger. This phenomenon took place in Poland, Lithuania and Romania – countries where the dominant religion played an important part in the formation of a national identity (Franklin Lytle, 1998; Gheorghe, 2004; Gillet, 2001; Bruce, 1999, 2001; Froese, 2004a). Coming back to the case of Romania, the situation of the Orthodox Church was a special one during the communist regime (Stark, 2001; Franklin 149

The Values of Romanians: 1993-2006. A sociological perspective

Lytle, 1998; Stan & Turcescu, 2000; Gillet, 2001; Meyendorff, 1996). Although oppressed by the political power, the Orthodox Church submitted to the government, following the byzantine tradition, and managed to overcome the communist experiment with smaller losses, as compared to other churches. Besides, Romanians often associate the idea of orthodoxy to that of nation, as Orthodox Christianity is strongly linked to the Romanian nation in the collective imagination (Gillet, 2001; Gheorghe, 2004; Stan, Turcescu, 2000; Franklin Lytle, 1998). Both the special situation of the Orthodox Church during communism and the association of orthodox religion to national ideas contributed, to a certain extent, to the preservation of a high level of religiosity in Romania, towards the end of the 20th century. Unfortunately, these two factors are difficult to translate in quantity measures that would allow their use as predictors of a statistics analysis. Nevertheless, we considered that they couldn’t be ignored when we attempted to explain why Romanians believe in God and grant a higher importance to religion than other Europeans. The analysis carried out in this chapter will follow the evaluation of religious value orientation in Romanians, as compared to other European people, as well as its variation in time. The analysis will attempt to establish which elements are the ones influencing the degree of religiosity of a people. We shall bare in mind a few contrasting hypotheses explaining the level of religiosity, mainly the association between religiosity and the level of social development, as well as the connection between the religious diversity of a country and the religious values shared by the population. By the same token, we shall try to find explanatory factors for the religious revitalization that was reported in Central and Eastern Europe, by using similar theoretical references.

The strategy for analysis and the employed indicators This paper employs a comparative analysis, by contrasting Romania and other European countries. On the other hand, we studied the evolution in terms of religiosity in Romania from a longitudinal perspective, thus trying to establish the variation of religious values at the level of the masses in two different moments in time. The analyses were run of the two waves of European Values Survey research, carried out in 1990 and 1999 / 2000. The first wave coveres 24 European societies, while the second provides information about 32 countries in Europe1. In order to identify the factors influencing the level of religiosity in different European countries, as well as the level of religious revitalization, a number of linear regressions were performed, which allow the determination of 1

See Halman (2001) for details about the research.

150 corectura 2 28 iulie 2008

Religiosity and Religious Revival during the Transition Period in Romania

the cumulative effect of the respective factors. We must mention the fact that the analyses included in the present chapter were carried out at the level of a unit, mainly, the country, the purpose being that of finding the elements that make Romanians more religious than other people, in general. We did not search for what influences a persons’ religious orientation, at an individual level. On the other hand, in the context of this paper the emphasis does not lie on explaining the degree of religiousness in other European people, but on the private case of Romania and the transversal comparison which serves us as a explanatory instrument. The data analysis implied the use of two types of variables: dependent variables, whose variation we want to explain and predictors, meaning variables with an explanatory role in the analysis schema. The dependent variables used were: religiosity and religious revival. Religiosity represents the latent dimension of religious belief. This is a reality we cannot measure directly. Religious values are translated, nevertheless, in attitudes and behaviors. Beyond the behaviorist indicators of religiosity, the attitudes exhibited towards the different components of the religious system (goods, persons, dogmas) allow us to evaluate the individuals’ religious values. Thus, the proper way of measuring religiosity resides in identifying the factors determining the common variation of the variables referring to the attitudes towards religious objects. The statistic method through which it can be done is represented by factorial analysis. The specific problem raised by comparative research in this case is related to the possibility of identifying a way of measuring religious belief that would be valid for all societies considered in the analysis. In order to come up with an indicator of religiosity that should be valid for all countries, irrespective of the moment in time when it was measured, we used a technique of data reducing, the type of factorial confirmatory analysis. Consequently, we built up a factorial model and carried out the analysis on the entire set of data, at one moment in time. Then, keeping the loadings of the factors at a constant level, that of the values resulted from the first model, we used the model for each country and calculated the index of fit, thus making sure that I was using the same measurement for religiosity in all countries and that I was measuring the same thing in all cases. I then repeated the analysis for different moments in time (1990 and 2000), making sure that the measurements were equivalent in the different years I had considered in the analysis. A table with the indicators of the fit of the model for each country included in the analysis at different moments in time can be found at the end of this chapter (Table 3).

151

The Values of Romanians: 1993-2006. A sociological perspective Diagram 1. The factorial structure of religiosity (factorial confirmatory analysis) e1

The Importance of I in personal t t lilife i iviata religion l ,80

e2

The Importance of God in personal lifel

,89

,58 e3

e4

Trust in the social implication of the Church

Religiosity

,78

Trust in religious ideas

RMSEA= 0,034 P test for close fit = 1,000; CFI=0,999, IFI=0,999

Source of data: EVS 1990-1993 and 1999-2000

Practically, the index of religiosity is constructed as a factorial score and it explains the variation of variables connected to the importance of religion in personal life, to the importance of God in personal life, to the trust in the intervention of the Church in social matters and to the belief in certain religious ideas. The importance of religion in personal life is measured by the answer to the question How important is religion in your life. The given answers are recorded on a scale of four levels. The importance of God is measured by the answer to the item How important is God in your life. The given answers are scaled from 1 to 10, where 1 means unimportant and 10 very important. The trust in the social implication of the Church represents a summative score built from the affirmative answers to the following questions: In general, do you believe the Church to offer the right answers to the individuals’ moral problems and needs / Problems of family life / Spiritual needs of people / Social problems of the country .The score takes values from 0 to 4, where 4 means maximal acceptance of the social implication of the church. The belief in religious ideas is also constructed as a summative score out of affirmative answers to the following questions: Do you believe that there is a God or not / Life after death / Hell / Heaven /Sin. The values of the score rank from 0 to 5, 5 meaning maximal acceptance of traditional religious beliefs. In essence, it is religiosity that influences the way in which an individual answers these questions. For instance, a high score of the religiosity variable indicates a high level of religiousness, while a low value indicates orientation towards secularization. The indexes presented in Diagram 1 shows that the model fits the data. Running the model on the full EVS 1990 data set, as well as on the EVS 2000 data, 152 corectura 2 28 iulie 2008

e5

Religiosity and Religious Revival during the Transition Period in Romania

proves its validity:, and indicates that the factorial structure is similar in both waves of the research. Also, separately using the model for each country included in the data sets for each year of research (1990 and 2000), it proved its adequacy in each case. Therefore, one may state that the model correctly indicates the degree of religiousness for each society, such that one may make both cross-country, as well as longitudinal comparisons. Also, the model proved to be adequate for the Romanian WVS 2005 data set, as the appendix shows. We must mention that the indicator has been rescaled and it takes values from 0 to 100, where 0 stands for the absence of religious belief and 100 is a maximal religious value orientation. Religious revival is measured by a number of indicators, each of them having a correspondent in the dimensions of religious involvement. Practically, three dimensions of religious revival where constantly kept in mind: belief, practice and affiliation. Consequently, religious revitalization was made operational by measuring the variation between the average of the index of religiosity for each country in 1990 and 2000, the variation of religious public practice in each country in 1990 and 2000 and the variation of religious nonaffiliation in each country in 1990 and 2000. The practice in public space was measured by using a standard indicator: attending church at least once a month. In order to study the dynamics and the dimension of religious affiliation we used a non-affiliation indicator, as from the data gathered previously we knew that the level of religious affiliation was very high in Romania. The predictors used in this chapter represent indicators for three series of different phenomena, corresponding to the research hypotheses we have presented above. In order to make a country’s level of social development operational we used three indicators. Starting from the idea that development does not solely refer to money and economic growth, three indicators were introduced in the analysis: the GDP for 2001, an indicator of economic development, the average of the higher educated population in the total of the population and the average life expectancy, the last two being indicators of the human capital stock of a society. Moreover, in order to measure the variations of the economic situation in a 10 year period I used the growth of the GDP between 1990 and 2000 as an indicator. Religious concentration was measured by borrowing an indicator from economic research– Herfindahl - Hirschman Index. It represents a widely used measure in the religious sociology survey, investigating the degree of monopoly on the religious market (see Inglehart, Norris, 2004; Halman, Draulans, 2004). In economy, this index is used to measure concentration on the economic market. In the case of the religious market, things are somewhat similar. The formula on which the index is based is 10 ∗ s i2 , where s i2 represents the

∑

percentage of a religious group out of the total population. When the market is in a monopoly situation, the value of the index tends to go towards 1000, while when the market is shared by several denominations, the value of the index 153

The Values of Romanians: 1993-2006. A sociological perspective

decreases. Small values of the index show the existence of a pluralism situation, while high values indicate the existence of monopoly. In the analysis carried out in this chapter the initiated value of the index was divided by 100 to facilitate the interpretation of the results. One last indicator refers to the change which occurred in postcommunist societies after 1990, a change that could have favored religious revitalization. An indicator of this change is the inclusion of the respective state to the group of post-communist countries. This indicator is a relatively weak one because it does not directly refer to the uplifting of the restrictions imposed under the communist regime, but it is a marker of the processes and evolutions which took place in these countries after 1990.

Religiosity and religious revival in post-communist Romania The data referring to religiosity confirm the results of the previous surveys. In Romania, religiosity has very high values, situating Romania among the first European countries, alongside Malta, Poland and Ireland, all of these being predominantly Catholic countries. According to the data in Figure 0.1 the medium values of religiosity are close to maximum in Romania, just like in Malta, Poland and Ireland, indicating a predominance of religious values and a low degree of secularization. Practically, Romania proves to be the most religious Orthodox country and, alongside Poland, the most religious excommunist country. According to the WVS 2005 data, Romania displays an even higher religiosity score, close to the one registered in 2000 for Malta. At the opposite pole we find the Czech Republic, Eastern Germany and Estonia, alongside the Northern countries, Denmark and Sweden. The data support the idea of profound secularization of Protestant societies, either Eastern or Western, as well as that of the preservation of religious values in Catholic societies. Nevertheless, the predominant denomination does not entirely explain the variation of the level of religious belief as long as mainly Catholic countries, like France, are among the champions of secularization, while the mainly Orthodox Romania is at the opposite pole. If we choose to look at them from the perspective of time evolution in religious values, the data in Figure 2 indicate the emergence of a phenomenon of religious revival especially in East European countries, like Bulgaria, Romania, Latvia and Slovakia, while secularization spectacularly advanced in a space where religious values had lost much of their territory – Eastern Germany. While Romania and Bulgaria are the champions of religious growth in the Europe of the last decade of the 20th century, Eastern Germany stands opposite to them. In most European states religious values stagnate, as no spectacular evolutions are being recorded. Could this be the explanation of this religious renewal in Romania and Bulgaria? Or could it be religious 154 corectura 2 28 iulie 2008

Religiosity and Religious Revival during the Transition Period in Romania

liberalization, increasing existential risks or increasing religious diversity? We shall look for the answer to these questions in the following section. Figure 1 Average level of religiosity in Europe, in 2000 Eastern Germany Czech Republic Denmark Estonia Sweden France Netherlands Slovenia United Kingdom Bulgaria Hungary Luxembourg Belgium Russia West Germany Latvia Spain Finland Belarus Iceland Ucraine Austria Lituania Slovakia Greece Croatia Italy Portugal Irland Poland Romania2000 Romania 2005 Malta 0

10

20

30

40

50

60

70

80

90

Source: EVS 1999-2000, WVS 2005 Figure 2 The variation of the average level of religiosity between 1990 and 2000 in Europe

East Germany Great Britain Irelanda Spain Hungary France Poland Netherlands Icelanda Slovenia Denmark Austria Belgium Czech Republic West Germany Finland Sweden Italy Slovakia Portugal Latvia Romania Bulgaria -10

Secularizationm

-5

Religious revival

0

5

10

15

Source: EVS 1990-1993, 1999-2000

155

The Values of Romanians: 1993-2006. A sociological perspective Figure 3 Variation of average level of religiosity by cohorts, between 1990 and 1999 in Romania 90

80

70

60

50

40

30 1993 2000

20

2005 10

0

born after 1984

born between born between born between born between born between born between 1974 - 83 1964 -73 1954-63 1944-53 1934-43 1924-33

born before 1923

Date: EVS 1993, 1999

If we change the register and strictly refer to Romania, the analysis of religious variation on units indicates an almost linear variation among religiosity and age, especially for the data of 1990. Practically, the level of religious belief grows as people grow older. The previous result supports the theory of age effect on religious belief, meaning that as people turn older they tend to get more religious. Comparatively, in 2000 an important growth of religiosity was noticed, especially in groups socialized during the communist period. For the youngest cohort, consisting of people born after 1974, the variation between the average of religiosity in 1990 and 1999 is at approximately 20 points, while for the other units it is at approximately 10 points. We must mention the fact that for the older cohorts, who have lived a longer period of time outside the communist regime, religious belief is preserved at the same level for both moments in time, the indexes having the highest possible values. As compared to the early 1990s, the 2005 data shows a different picture of the Romanian society. In 1993, the religiosity level increased with age. However, in 2005, the intergenerational differences are smaller, even if the older cohorts remain more religious than the younger ones. On average, the society is more religious in 2005 than it used to be in 1993 or 1999, the most spectacular increase being recorded for the younger cohorts. For the ones born in 1974-1983, the religiosity indicator is 30 points higher in 2005 as compared to 1993 and 1999. Similarly, those born before 1984 have a very high level of 156 corectura 2 28 iulie 2008

Religiosity and Religious Revival during the Transition Period in Romania

religious belief. They were only 5 years old or less in 1989, and were practically socialized in an environment allowing the free manifestation of religiousness. All data converge shows that Romania religiosity is not declining, but on the contrary, it tends to increase. The data referring to the non-affiliation gives other evidence suggesting that the Romanian society is strongly oriented towards religion. The percentage of those declaring non-affiliation significantly decreased between 1993 and 2005, nowadays almost everyone declaring to belong to a religious denomination (see Figure 5). As compared to other European societies, only Malta displayed a higher level of affiliation in 2000 (Halman, 2001). However, since Romania orthodoxy used to represent an important role in defining and preserving national identity, the high level of religious affiliation also comes as a consequence of expressing this identity. Figura 5. Variation of the percent of non-affiliated persons in Romania between 1993 and 2005 7%

1993: 6% 6%

5%

4%

3%

1997: 2% 2%

1999: 2%

2003: 1% 1%

2005: 0,4%

0%

Source: EVS 1993, 1999, WVS 1997, 2005, BOP 2003

From a longitudinal perspective, religious affiliation increases between 1993 and 2000, the percentage of the non-affiliated ones decreased by 4 percentage points. A similar phenomena, labelled as religious revitalization (Pollack, 2003), take place almost everywhere in the former communist societies (see Figure 6). Within the region, Bulgaria and the three Baltic states display the higher religious “re-affiliation”. Poland and Eastern Germany do not display significant differences. While in the Czech Republic, secularization continues in the 1990s. In the Western part of the continent, with a few 157

The Values of Romanians: 1993-2006. A sociological perspective

exceptions (the UK, Portugal, Austria), the rest of the societies experienced secularization, smoothly decreasing the religious affiliation the dominant tendency being the dominant process. One may notice that, overall, from the point of view of religious affiliation, the 1990s were marked by religious revival in the Eastern Europe and a continuation of secularization in the Western countries. Figura 6. The variation of the percentage of the non-affiliated persons in Romania between 1993 and 2005 Bulgaria United Kingdom Latvia Lithuania Portugal Estonia Slovacia Romania Austria Hungary Poland Finland East Germany Denmark Iceland Italy West Germany Slovenia France Belgium Spain Irealnd Netherlands Sweden Czech Republic

Religious revival

Secularization

-40

-35

-30

-25

-20

-15

-10

-5

0

5

10

Source: EVS 1990-1993, 1999-2001. The figures represent the difference between levels of non-affiliation in 2000 and 1993.

In Romania, the religious revival had different paces during the 1990s. There is a sharp increase of church frequenting immediately after the communist breakdown, but after the mid-90s, the increase stopped, and no significant change occurred (see Figure 7). During the whole period, the religious practice increased from 30% to 45%. The initial increase, as well as the later stagnation suggest that the latent religiosity, freed from the communist constrains, became manifest at the beginning of the transition period. The “religiosity stock” showed out, reaching a maximum, and then it remained at this equilibrium point.

158 corectura 2 28 iulie 2008

Religiosity and Religious Revival during the Transition Period in Romania Figure 7. The variation of religious practice in Romania between 1993 and 2005 50%

45%

1999 46%

1997 45%

40%

2005 45%

2003 43%

35%

30%

25%

1993 31%

20%

15%

10%

5%

0% 1992

1994

1996

1998

2000

2002

2004

2006

Source: EVS 1993, 1999, WVS 1997, 2005, BOP 2003 The figures represent the percentage of those going to church at least once a month.

Figure 8. The variation of public religious practice between 1990 and 2000 Irland Eastern Germany Belgium United Kingdom Poland Netherlands Hungary France Spain Slovenia Austria Sweden Czech Republic West Gemany Denmark Italy Iceland Finland Portugal Latvia Slovakia Bulgaria Romania -25

Secularization

Religious revival

-20

-15

-10

-5

0

5

10

15

20

Source: EVS 1990-1993, 1999-2001. The figures represent the difference between the percentage of those going to church at least once a month, in 2000 and 1990. 159

The Values of Romanians: 1993-2006. A sociological perspective

During the decade which succeeded the communist breakdown, across Europe, Romania had the highest increase of religious practice, as measured by the percentage of those declaring to go to church at least once a month. With an increase of the respective index with 15 percentage points between 1993 and 1999, Romania may be labelled as the European ‘champion’ of religious revival at the end of the twentieth century. Bulgaria, Slovakia and Latvia are the other societies which experienced significant scores in the 2000s EVS wave as compared to the 1990 one. For the former soviet countries like Russia, Ukraine and Belarus, there is no data to allow longitudinal comparisons. At the opposite pole, Ireland displayed higher secularization, with 15 percentage points less in 2000 as compared to 1990, when considering the percentage of those going to church at least once a month. Among the former communist societies, Eastern Germany and Poland, the register dropped of 5 percentage points. The former DDR continues the tendency of decreasing the social impact of religion, probably as a part of the converging process with the West Germany. In Poland, the Catholic Church played an important role during communism. When communism ended, this opposition naturally disappeared, as well the incentive to publicly contest the regime by going to church.

Religious revival: economic decrease or religious diversity? In order to identify the factors influencing the level of religiosity of a country we ran some regression models that included different sets of predictors. When we only used as variables the degree of religious diversity and the Gross Domestic Product, the results of the regression indicated a double determination of the general level of religiosity in a country (Model 1 of Table 1). However, the relation indicated by the data between religious diversity and the degree of religiosity was reversed as compared to the one predicted by the theory of offer and demand. The lower the diversity was, the more the belief spreads among the population. This strong and highly significant relationship can be encountered in all three models of regression in Table 1. Thus, the empirical data show that religious pluralism is the one associated to secularization and it does not generate religiosity, a fact supported by Berger (1969) and Bruce (1999). According to the cited authors pluralism facilitates contact with other religions and questions the authority of only one supernatural being, a fact which weakens the religious feeling. Another aspect tested by the regression models is the effect of the level of development on the degree of religiosity. In the first model, which is only included as predictor for the GDP, the regression highlights a significant, but lower effect of the GDP on the degree of religiosity. According to this model, in countries with a lower level of economic development religious values are more spread out among the population. But if we include the average of the higher educated population in the model (Model 2 of Table 1) the effect of the GDP 160 corectura 2 28 iulie 2008

Religiosity and Religious Revival during the Transition Period in Romania

disappears but the model includes the average of the higher educated population as a significant predictor. Consequently, the effect of the level of economic development was not a real one, but it was in fact determined by the degree of education of the population which determines and is, at its turn, determined by the degree of economic development. In the third model we have replaced the GDP from set of predictors with the average life expectancy (the two variables cannot be simultaneously introduced in the regression for reasons of multicollinearity2), preserving the average of the higher educated population and religious diversity in the model. In this case, too, the average of the higher educated population is kept in the regression model as a significant predictor of the degree social development, while the average life expectancy is eliminated from the model. We may conclude that a populations’ level of religiosity depends on its degree of development, but the main effect is not that of the economic component of development, but the human capital, expressed by means of the educational capital. Table 1. Linear regression, dependent variable: religiosity in Europe 2000

Constant Religious concentration GDP 2001 Percent of higher educated population Average life expectancy

Model 3

Model 1 B β 0,124

Model 2 B β 0,538

B 1,548

β

1,533

0,63***

1,253

0,51***

1,270

0,52***

-0,017

-0,32*

-0,012

-0,22

-1,695

-0,31*

-0,020

-0,37*

-1,539

-0,15

R R2 Adjusted R2

0,67 0,45 0,41

0,73 0,59 0,47

0,71 0,50 0,45

***p<0,001; **p<0,01; *p<0,05 Data source: EVS 1999-2000

Regarding the three models as a whole, one may notice that religious diversity is the predictor with the highest effect on the degree of religiosity of a population. The degree of human development also plays an important part, but the level of economic development does not have any impact when it is being controlled for the effects of the educational stock. The models invalidate the 2

For all of the three models multi-colinearity was tested by Tolerance. For all predictors included in the presented models Tolerance did not drop below 0,600, a fact which does not indicate the multi-colinearity of the predictors. 161

The Values of Romanians: 1993-2006. A sociological perspective

supply-side theory and support the secularization thesis. However, secularization is not determined economically, but by the general level of human development. Figure 4 Religiosity by religious concentration in European countries in 2000 Malta

Romania 1,50 Poland Italy

Irelanda Portugal Croatia Greece

Slovakia

Religiosity

1,00

Lithuania

Ukraine

Icelanda Austria

Belarus Latvia 0,50

Finland

West Germany Luxembourg Hungary Slovenia Netherlands

Spain

France Sweden Czech Republic

Denmark Estonia

0,00

R Sq Linear = 0,349

East Germany

0,20

0,40

0,60

Religious concentration

Data: EVS 1999-2000

162 corectura 2 28 iulie 2008

0,80

1,00

Religiosity and Religious Revival during the Transition Period in Romania Figure 5 Religiosity by percent of population with university degree in European countries in 2000 2,00 R Sq Linear = 0,293

Malta

Romania 1,50 Poland Irelanda

Religiosity

Portugal

Croatia

Italy 1,00

Greece

Slovakia

Lithuania

Austria Ukraine

Hungary

0,50

Slovenia

Icelanda

Belarus

Spain Russia Luxembourg

Bulgaria

Finland Latvia

Netherlands France Sweden

Czech Republic

Denmark Estonia

0,00 10,00

20,00

30,00

40,00

percent of population with university degree

Data: EVS 1999-2000

Starting from the results of the three regressions we may discuss the causes determining the existence of such a high level of religiosity in Romania. We have seen that a population’s level of religiosity is the result of a combination between the degree of religious pluralism and the stock of educational capital. By considering Romania’s case, the respective model fitted the existing situation pretty well. The graphics presented in Figure 4 and Figure 5 better illustrate what the results of the regression have revealed. The high degree of religiosity in Romanians can by explained by the fact that the religious market is almost a monopoly of the Orthodox Church and by the fact that Romanians are among the European people with the lowest level of higher education. Figure 4 presents the variation of religiosity according to religious diversity. On the graphic, one may notice both the strong association of the two variables and Romania’s position, in the right upper corner, next to Malta, Poland, Ireland and Portugal, all states with one single denomination dominating the religious market. Moreover, Figure 5 illustrates the relationship between religiosity and the average of the higher educated population. Again, Romania is situated next to Malta and Poland, in the upper left corner, among 163

The Values of Romanians: 1993-2006. A sociological perspective

countries with high religiosity and a low degree of higher educated population. 2005 WVS data support the above noticed tendency, showing a further increase of religiosity, in a society with low religious diversity and few higher educated people. According to the presented results, the answer to the question, “Why are Romanians among the most religious Europeans?” can be found in the combination between the small religious diversity and the low average of higher educated populations. These are the two factors which contribute decisively to the explanation of the high level of religiosity in the Romanian population. Table 2 Linear regressions – dependant variables: the variation of religiosity between 1990 and 2000, the variation of public religious practice between 1990 and 2000, the variation of non-affiliation between 1990 and 2000. The variation of religiosity 1990 2000

The variation of religious practice 1990 - 2000

B 0,327

B 7,715

β

(Constant) GDP growth 1990-2000

-0,078

0,955***

-2,587

-0,689**

-0,007

-0,379

-0,304

-0,336

0,036 -0,102

0,041 -0,318

2,581 -1,382

0,063 -0,094

Higher educated population Religious concentration Ex-communist country R R2 Adjusted R2

β

0,782 0,612 0,515

0,635 0,403 0,254

The variation of religious nonaffiliation 19902000 B β -2,388 3,774 0,707** 0,019

0,014

-6,719 -2,427

-0,113 -0,114 0,736 0,541 0,433

***p<0,001; **p<0,01 Data source: EVS 1993, 1999-2000.

By using a similar set of predictors to the one I have used to explain the variations in the degree of religiosity, I have investigated the determinants of religious revival which occurs mostly in ex-communist countries. The difference from the previous models lied in the fact that the GDP used as indicator for the level of economic growth was replaced by the GDP growth 1990-2000. While the first indicator referred to an existing state at a certain moment in time, the second one reflected the transformations that had taken place along a decade in the economic situation of the countries included in the analysis. Moreover, I have also introduced a variable indicating the belonging to the ex-communist countries in the set of predictors. By using as predictors the growth of the GDP 1990-2000, the average of higher educated population, religious diversity and the belonging to the group of ex-communist countries I 164 corectura 2 28 iulie 2008

Religiosity and Religious Revival during the Transition Period in Romania

applied three models of regression, having as predicted variables the variation of the degree of religiosity, the variation of religious public practice and the variation of religious non-affiliation. The results of the three linear regressions lead to similar conclusions, as was shown in Table 2. In all of the three cases the only predictor significantly contributing to the explanation of the variations of religious implications is the variation of the GDP. According to the results, the decrease of the GDP generates an increase of both religious faiths and religious practice, as well as a decrease of non-affiliation, thus confirming the thesis of the secularization and of the impact the economically insecure periods of time have on the religious phenomenon. The religious revitalization that has been noticed in the last decade originates in the growth of social insecurity on the background of the changes taking place especially in societies making the transition from command to market economy. The other predictors - the degree of religious diversity, the percentage of the higher educated population, and being a former communist country - have no significant effect for any of the dimensions of religious revival. The data validate the hypothesis of secularization and invalidate the theory of the effect that religious diversity has on the growth of religious involvement. In the case of Romania, the high level of religious revitalization is justified by the strong economic decrease experienced during the transition period and measured by the decrease of the GDP. The lowering of life standards and the increase of social risks generated by the economic recession led to a growth of religious belief which, in this context, appears as a factor of stability, of ordering life promising, eventually, a better future, like Inglehart and Norris (2004) show. The ‘90s were, for Romania, years of stronger economic recession than for other ex-communist countries and this fact accounts for the high level of religious revitalization

Conclusions At the beginning of the millennium Romania is one of the most religious countries in Europe, both from the point of view of religious practice, as well as from that of belief and shared values. In a Europe that turns its face from religion more and more, Romanians are not only among the most faithful inhabitants of the continent, but the quantitative data demonstrate a religious revival on several dimensions of religious involvement. The ‘90s brought up a religious renewal in Romania and this fact had no effect on the uplifting of restrictions the communist regime had imposed on religious practice or of the increasing offer on the religious goods market, as one may think at a first glance. For European countries the degree of religiosity is the result of the combination of two factors: the level of human development and the absence of religious pluralism. What is important is not as much the level of economic 165

The Values of Romanians: 1993-2006. A sociological perspective

development, but the educational stock of the population. The more educated the population of a country is, the lower religious belief becomes. Education makes the surrounding world more predictable, easier to understand, and it reduces the risks to which the individual is exposed, securing a better life for him/her. On the other hand, the access to education takes away the “magic” of the world, university people being more inclined to interpret the world rationally and to know it through the logic of science. All of these reduce the orientation towards a religious vision and decrease religiosity. On the other hand, the data of the analysis do not support the supplyside theory. On the contrary, it looks, as in the case of Europe, religious diversity strongly influences the degree of religiosity, but the relation is reversed. Reduced diversity encourages religious belief. The lack of direct contact with other religions strengthens the main religion. The monopoly is actually an advantage, instead of a disadvantage. One should bear in mind that the level of religious faith is influenced by a combination of two factors and not just by religious diversity. Indeed, Northern countries are under the conditions of a religious monopoly, but they have a higher education level, which explains the low religiosity. In the case of Romania, the high level of religious faith originates in both the situation of a religious semi-monopoly, in which the Orthodox Church occupies the biggest share of the market, as well as in the small average of the higher educated population. Why are the Romanians among the most religious Europeans? The answer proved to be simple: because we have a very strong Church and a reduced human capital. The religious revival reported in postcommunist societies is generated by the augmentation of risks and existential insecurity. On the background of difficulties generated by economic recession, an important number of people turned to religion so as to maintain their inner balance. Romania suffered quite a lot during the transition period, as it had to outgrow a moment of profound economic recession. This fact accounts, to a big extent, the dimensions that religious revitalization was acquired in our country. In this context the posterior evolutions of religious involvement and of faith in Romania are very interesting. On the basis of the obtained results one should expect that a better economic situation should bring about a long term stagnation of religious revival and even a decrease of faith and religious practice. But these are predictable only in the long run because values in general and, implicitly, religious values, change slowly. The future researches may validate or invalidate this prediction and should show us whether we are heading towards secular Europe or we are developing a strongly religious influenced culture.

166 corectura 2 28 iulie 2008

Religiosity and Religious Revival during the Transition Period in Romania

Annex Table 3 Indexes of fit of the factorial model in Diagram 1 for the countries included in the analysis. Country Austria Belgium Bulgaria The Czech Republic Croatia Denmark Estonia France Eastern Germany Western Germany Greece Ireland Iceland Italy Latvia Lithuania Luxemburg Malta Great Britain Holland Poland Portugal Romania Russia Slovakia Slovenia Spain Sweden Ukraine Hungary Romania 2005

RMSEA 1 0,088 0,055 0,092

1990

2000

CFI2 0,993 0,997 0,991

RMSEA 0,063 0,057 0,070

CFI 0,996 0,997 0,995

0,102

0,991

0,054

0,997

0,177

0,968

0,046 0,057 0,053

0,998 0,997 0,998

0,195

0,973

0,086 0,255

0,994 0,904

0,074 0,100 0,193 0,100 0,100

0,995 0,991 0,975 0,992 0,991

0,082 0,056 0,095 0,165

0,994 0,997 0,992 0,972

0,123 0,986 RMSEA=0,108

0,146 0,983 0,095 0,992 0,142 0,980 0,063 0,996 0,121 0,987 0,111 0,990 0,122 0,986 0,184 0,974 0,160 0,978 0,120 0,998 0,089 0,993 0,145 0,982 0,097 0,991 0,145 0,924 0,082 0,993 0,083 0,994 0,180 0,977 0,181 0,974 0,122 0,960 0,086 0,993 0,092 0,994 0,056 0,997 0,081 0,994 0,134 0,981 0,079 0,994 0,075 0,995 CFI=0,997

Note: 1. The closer the value of RMSEA is to 0,000, the more adequate it is to the empirical data. Given the dimensions of the samples and the number of variables in the model, indexes smaller than 0.100 can be considered as valid. 2. The closer the value of the CFI is to 1.000, the more adequate the model is to the empirical data. Given the dimensions of the samples and the number of variables in the model, indexes are higher than 0.950 which indicate a satisfactory fit of the model. 167

The Values of Romanians: 1993-2006. A sociological perspective

References Berger, Peter (1969) - The Sacred Cannopy: Elements of a Sociological Theory of Religion. New York, Anchor Books Bruce, Steve (1999) – “Modernization, Religious Diversity and Rational Choice in Eastern Europe”, Religion, State and Society, vol. 27 (3/4), p. 265 – 276 Bruce, Steve (2001) – “The Supply-side Model of Religion: The Nordic and Baltic States”, Journal for Scientific Study of Religion, vol. 40 (1), p. 32 – 47 Chaves, Mark (1994) – “Secularization as Declining Religious Authority”, Social Forces, vol. 72(3), p. 749 – 774 Finke, Roger (1990) – “Religious de-regulation: origins and consequences”, Journal of Church and State, vol. 32, p. 609 – 626 Finke, Roger, Stark, Rodney (1988) – “Evaluating the Evidences: religious Economies and Sacred Canopies”, American Sociological Review, vol. 54, p.1054 – 1065 Franklin Lytle, Paula (1998) – “Religion and Politics in Eastern Europe”, Remet, Sabrina (ed..) Eastern Europe: Politics, Culture and Society since 1989, Bloomington, Indianapolis, Indiana University Press, , p. 304 – 329 Froese, Paul (2004a) – “After Atheism: An Analysis of Religious Monopolies in the Post-communist World”, Sociology of Religion, vol. 65 (1), p. 57 – 75 Froese, Paul (2004b) – “Forced Secularization in Soviet Russia: Why an Atheistic Monopoly Failed”, Journal for Scientific Study of Religion, vol. 43 (1), p. 35 – 50 Froese, Paul (2005) – “Secular Czechs and Devoted Slovaks: Explaining religious differences”, Review of Religious Research, vol. 46 (3), p. 269 - 283 Gheorghe, Manuela (2004) – Religion and social change in Romania [Religie şi schimbare socială în România], Iaşi, Axis Publisher House Gheorghiu, Elena Iulia (2003) – „Religiosity and Christians in Post-communist Romania” [„Religiozitate şi creştinism în România postcomunistă”], Romanian Sociology [Sociologie Românească], nr. 3, p. 102 – 121 Gillet, Olivier (2001) – Religion and Nationalism. The Ideology of Romanian Orthodox Church under the Communist Regime [Religie si naţionalism. Ideologia Bisericii Ortodoxe Române sub religmul communist], Bucharest, Altfel Company Halman, Loek (2001) The European Values Study: The Third Wave. Tilburg, WORK, Tilburg University Halman, Loek, Draulans, Veerle (2004)- “Religious beliefs and practices in contemporary Europe”, in Arts, Wil, Halman, Loek (eds.) - European Values at the Turn of the Millennium, Boston: Brill, pp. 283-316 Iannaccone, Laurence (1991) – “The Consequences of Religious Market Structure”, Rationality and Society, vol. 3, p. 156 – 177 Inglehart, Ronald (1971) – “The Silent Revolution in Europe: Intergenerational Change in Post-Industrial Societies”, American Political Science Review, vol. 65, pp. 991-1017 Inglehart, Ronald (1990) – Culture Shift in Advanced Industrial Societies, Princeton University Press Inglehart, Ronald (1997) – Modernization and Post-Modernization. Cultural, Economic and Political Change in 43 Societies, Princeton University Press

168 corectura 2 28 iulie 2008

Religiosity and Religious Revival during the Transition Period in Romania Inglehart, Ronald, Halman, Loek, Welzel, Christian (2003) – “Introduction”, Inglehart & others (eds.), Human Beliefs and Values: A Cross-cultural Sourcebook based on 1999-2000 Values Survey, Mexico, Siglo XXI Editores, Inglehart, Ronald, Norris, Pippa (2004) - Sacred and Secular, Cambridge, Cambridge University Press Kääriäinen, Kimmo (1999) – „Religiousness in Russia after the collapse of Communism”, Social Compass, vol. 46 (1), p. 35 – 46 Meyendorff, Jean (1996) – Orthodox Church Yesterday and Today [Biserica Ortodoxă ieri şi azi]. Bucharest, Anastasia Publisher House Müller, Olaf (2004) – “Religiosity in Central and Eastern Europe: Results from the PCE Survey”, în Jerolimov Dinka, Zrinscak Sinisa, Borowik Irena (eds.) Religion and Pattern of Social Transformation, Zagreb: Institute for Social Research, pp. 61 - 78 Need, Ariana, Evans, Geoffrey (2001) – “Analyzing Patterns of Religious Participation in Post-communist Eastern Europe”, British Journal of Sociology, Vol. 52 (2), pp. 229 - 248 Pollack, Detlef (2001) – “Modification in the Religious Field of Central and Eastern Europe”, European Societies, vol. 3, p. 135 – 165 Pollack, Detlef (2003) – “Religiousness Inside and Outside the Church in Selected Postcommunist countries of Central and Eastern Europe”, Social Compass, vol. 50 (3), p. 321-334 Pollack, Detlef (2004) – “Institutionalized and Subjective Religiosity in Former Communist Countries of Central and Eastern Europe”, în Jerolimov Dinka, Zrinscak Sinisa, Borowik Irena (eds.) Religion and Pattern of Social Transformation. Zagreb: Institute for Social Research, pp. 79 – 88 Rinkevicius, Leonardas (2000) – “Public Risk Perception in a ‘Double – Risk’ Society: The Case of Ignalian Nuclear Power Plant in Lithuania”, Innovation, Vol. 13 (3), pag. 279 – 289 Schlegel, Jean-Louis (2005) – The God’s Law against Human Liberty. Intergism and Fundamentalism, [Legea Domnului contra libertăţii oamenilor. Integrisme şi fundamentalisme], Bucharest, Runa Publisher House Sommerville, John (1998) – “Secular Society / Religious People: Our Tacit Rules for Using the Term Secularization”, Journal for Scientific Study of Religion, vol. 37 (2), p. 249 – 253 Stan, Laviana, Turcescu, Lucian (2000) – “The Romanian Orthodox Church and Postcommunist Democratization”, Europe – Asia Studies, vol. 52 (8), p. 1467 – 1488 Stark, Rodney, Iannaccone, Laurence (1994) – “A Supply-side Reinterpretation of ‘Secularization’ in Europe”, Journal for Scientific Study of Religion, vol. 33 (3), p. 230 – 252 Stark, Rodney (2001) – “God, Rituals and Moral Order”, Journal for Scientific Study of Religion, vol. 40 (4), p. 619 – 636 Voicu, Mălina. 2007. România religioasă: pe valul european sau în urma lui?, Iaşi: Institutul European Wilson, Bryan (2000) – Religion from Sociological Perspective [Religia din perspectivă sociologică], Bucharest, Trei Publisher House.

169

Family values in Romania and in Europe RALUCA POPESCU

This study aims at analyzing changing family values in Romania, from a comparative perspective with the European countries. The study starts from the question whether Romanians have or do not have a different value pattern compared to other European countries and tries to distinguish their common and different aspects. How important is the family in the Romanian society in a European context, which is the place that it occupies in the individual’s life, how much is the marriage valorised in comparison with other family models, how tolerant are Romanians related to different aspects regarding family and sexuality? These are several of the questions this study tries to answer to. An important goal is identifying the determinant factors of the analysed value models and drawing an explanatory model for the specific value configuration. The conceptual framework of the analyses is the traditional-modernpostmodern trajectory. The work tries to identify where Romania is placed on this axis, taking into account more criteria: the importance of the institution of family in society, the propensity towards alternative family models, the woman’s participation in the labour market and tolerance. The study is structured in two parts. The first part, with a more descriptive character, illustrates the main changes that have taken place at a demographic level and in the family lifestyles. The second section focuses on the comparative data analysis discussing the results we obtained. A transversal comparative analysis on the EVS/WVS data in the 1999-2001 wave was approached at the European level. Romania’s situation was analysed both transversally and longitudinally with data from three available EVS/WVS waves (waves 1990-1993, 1999-2001, 2005). The main conclusions of the analysis are presented in the final part, discussing the way in which European countries are grouped depending on the value orientations we analysed and what are the specific features of the Romanian profile. 170 corectura 2 28 iulie 2008

Family values in Romania and in Europe

The family between change and stability Researchers, political actors and public opinion in general seem to favour the idea that family has suffered deep transformations for the last decades. At the same time, the family continues to represent a fundamental institution in the society, a fact for which in public debates the problem of “family decline” is regarded with more and more concern. The typical image of the family, assumed also by the political and scientific discourse, is that of an institution preserving the national values and traditions, being relatively independent of the socio-economic context and having a high inertial stability (Ghebrea 2000). Without diminishing its importance as a social institution, the changes in the last decades in the sphere of family entitle the analyses to consider that, on the contrary, family does not represent a conservative institution, but an institution adapted to the transformations at the level of the society: “Family is now trying to get rid of the «glory» of conservatism, of the merit of being the keeper of the national values, rather becoming «the barometer» of the social changes, crossing a visible democratization, secularization and liberalization process” (Mihăilescu 2000, p.17). The family seems more and more integrated in the dynamics of the society, all the more conditioned by economic and social changes, influencing the overall evolution in its turn. The changes in the last decades in the Western society have raised the idea that we are facing a new post-industrial, post-materialistic civilisation. New lifestyles, a philosophy of freedom and experimentalism, a new phase of consumerism, a controlled hedonism, and other ways of spending leisure time appear. In the sphere of family, the individual interests become more important than the expectations the society had from the family, the alternative models to the classic nuclear family turn out to be more and more spread out, the functions of the family are redefined, the roles and statuses are facing a process of democratizations and last, but not least, the family values are changing.

Changes at a demographic level Demographic evolutions have kept alive the alarmist discourses about the family crisis in the contemporary society. More and more persons do not marry or postpone the moment of their marriage. The marriage rate1 finds a continuous decrease in all the European countries, with an average of all the 27 countries of the European Union, only 4.88 marriages per thousand population in 2005 (compared to 5.31‰, as it was recorded in 1994). In the context of delaying marriage and spreading the consensual unions, the average age at the

1

The source for all the macro-social data used in this chapter is from the Eurostat, Statistical Office of the European Communities (http://epp.eurostat.ec.europa.eu/). 171

The Values of Romanians: 1993-2006. A sociological perspective

first marriage came to be almost 30. In Sweden it is 30.5 years for women and 32.9 years for men. The pattern of early and universal marriage is still maintained in Romania: the marriage rate is among the highest (6.56‰), in a slight ascending tendency in the last years (compared to the lowest value of 5.85‰ in 2001) and the average age at the first marriage is among the lowest in Europe, of 24.1 years for women and 27.5 years for men. Given the fact that less and less people marry or delay the marriage, the divorce stayed relatively constant in most of the countries, the average in the EU countries being of 2 divorces per marriage. In Romania, in the last decades, the value is relatively constant, at approximately 1.5 divorces per marriage. Still, the researches that went further than simply comparing the statistical data, demonstrate that we have to look with restraint at the pessimistic predictions about the disintegration of the family and the disappearance of marriage as a life contract. The unfavourable comparisons between present and past can be misleading. The stability of the family may be a myth rather than a historical reality. The researchers suggest that, centuries ago, the rates of the successive marriages after the decease of one of the partners (in the context of a high mortality rate) may be at the same height as the current ones successive to divorce (Parkinson, 1993, p.12). Besides, the number of marriages that ended with informal separations cannot be compared with the nowaday’s figures, because at that time it was not recorded. The average duration of a current marriage is comparable to the span of 100 years ago. Due to woman’s emancipation and to her participation in the labour market, in the context of recent value changes, more and more couples postpone the moment of childbearing or they even renounce to have children. The decrease of fertility under the replacement level, probably represents the most important problem. Most often, the number of children that a woman can have is reduced by postponing the moment of childbearing and, thus, by the shortening the fertility period of time. The fertility at the level of EU countries is at 1.52 children per woman, and in Romania of 1.32 children per woman. The average age at the first child’s birth is 30 years old as well. In most countries, this age is lower than the marriage age, in the context of the multiplication of births outside the marriage. One third of the European children are born outside the marriages and in some societies more than half of the new-born children are in this situation (Island – 6%, Estonia – 58%, Sweden – 55%, Norway – 51%). In Romania there is recorded average value of almost 30%.

Changes in family lifestyles Important changes have also happened in what concerns the family lifestyle. Single-parent families have become widespread. In the Baltic States approximately one out of four families is a single-parent family (in Latvia, even one out of three), high percentages, up to 20%, being common to more countries 172 corectura 2 28 iulie 2008

Family values in Romania and in Europe

in Eastern Europe (Czech Republic, Poland, Slovenia, Slovakia). Among the Western European countries, most single-parent families are in Ireland and Great Britain (approximately 16% of the families). Romania records an average value at a European level (13.4%). On the other hand, the reconstituted family (the second marriage for two divorced persons who have children) is more and more spread, the same with other types of arrangements: homosexual marriages, groups of persons who decide to live together for financial reasons, serial marriages, conglomerates of marriages, etc. Though, the attachment for a stable relationship remains high, as shown by further considerations. More and more couples prefer the unofficial cohabitation instead of the legal marriage. In Denmark, one out of five families is of this type. Values above 10% of the total number of unions are also recorded in Finland, Norway, Estonia, Holland, Great Britain, or Austria. In Romania, the consensual unions are not spread out in the same way, representing 6.5% of the total number of households, a higher value compared to some Eastern countries (Poland, The Czech Republic, Slovakia, Lithuania) or Mediterranean countries (Italy, Spain, Greece or Portugal). In this new type of union, couple relations define by what Anthony Giddens calls “pure relationship”: “a social relation that was initiated only for itself, for what any person can obtain from an relationship shared with the other person, and which continues to the extent in which both partners consider it brings them enough satisfaction in order to be supported by each one of them” (Giddens, 2000, p.59). By the pure relationship type, this is not meant by only love relationships; this is also valid for all the other relations: parents – children, relatives, and friends. All of these become relationships that develop permanently and in which the way of referring to the other is discovered “on the fly”. Alternative strategies of child raising appear and intimacy replaces the parental authority. The parents are rather the child’s friends, his/her protectors, confidants and advisers. In Romania, all these value changes in the family lifestyle or in family relationships are less visible, because the transformation process is at the beginning. As we will see in the next sections, these options/choices are rather scarce.

The importance of the family The place the family occupies in the individual’s life Compared to the other European societies, in Romania, the importance of family records an average value. In the case of the countries with the highest scores, more than 90% of the population declares that the family is important or very important. The group formed of these societies is heterogeneous from a geographic point of view and seems to not respect a religious or historical173

The Values of Romanians: 1993-2006. A sociological perspective

political criterion. In the cathegory with highest scores we find countries from former Yugoslavian space, more traditionalist countries like Turkey, Albania, Malta, more conservative catholic countries like Iceland, Great Britain or former socialist countries like Hungary. Levels comparable to the ones of Romania can be found in Austria, Belgium, Norway, Slovakia, Denmark, Finland, France, Spain (with values higher than 85%). The lowest scores are registered in some socialist countries, especially in the former Soviet Union – Russia, Belarus and the Baltic States – the only ones that, with the exception of Germany, record values below 80%. It is interesting to notice that in most of the countries it is considered that the importance conferred to family should rise in the future, especially in the countries which, at the moment, do not grant a very big importance to family. Besides this, some other authors (Mihăilescu, 2000) noticed that “the wish for the rise in the importance conferred to family is stronger in the countries in which the changes in family models were deeper than in the societies where these changes had a smaller role (...)” and the explanation could refer to: “becoming aware of the difficulties produced by family changes and the wish to strengthen the nuclear family or, on the contrary, to increase the devotion to the new forms of family” (p.24). Analysing the experience of the Western societies and, especially, of the Northern societies, it seems that the individualization tendency rises up to a certain extent. Inglehart (1997) also considers that postmodern value orientation is characterized, besides the spreading of new values and lifestyles, also by the revalorization of tradition. “The signs of a continuous individualization are counterbalanced by a rise in the importance of the traditional values over the family structure, especially in some of the most individualized states (...). We have got the impression that the pendulum of individualization moves backwards, after the people had been confronted with the consequences of the ideas that went too far” (Van der Akker, Halman, de Moor, 1993, p. 109). The devotion to the family stays high and it is expected to keep its values or to rise in the future. The importance offered to family does not differ significantly according to the individuals’ religiosity, such that for the greatest majority of the ones who do not go to religious services, the family does not represent the most important domain in their life. We can consider that the family represents a universal human value, a fundamental component of the human existence. In Romania, the attachment to family maintained a high level, being at the same time the aspect of life that offers the biggest satisfaction to the individuals. In the three waves, the importance conferred to the family kept high constant values. There are not significant differences depending on sex, education, job or the residential area. Still, the individuals’ age and marital status influence the importance conferred to the family. For the youngsters, the family is more important than for the old persons, as for the married persons or for the ones who live in consensual unions compared to the persons who 174 corectura 2 28 iulie 2008

Family values in Romania and in Europe

experienced the dissolution of their own family – divorced, separated persons or widows and widowers. A high importance is paid by the unmarried persons, but in this case the family they relate to is their family of origin, not their own one. Figure 1 The importance conferred to the family in the European countries (in the present and in the future)

Source: EVS/WVS 1999-2001, for Romania WVS 2005. The vertical axis represents the percentage of the ones who believe that the family will have a bigger importance in the future. The horizontal axis represents the percentage of the ones who declare that the family is “very important” or “important”.

Compared to the other important aspects of the life, the described pattern is one in which, as it was expected, the family occupies the first place, being followed by work and religion; friends and leisure time also play an important role. Political life is rather without importance in most individuals’ perception. 175

The Values of Romanians: 1993-2006. A sociological perspective

As Figure 2 suggests, the importance conferred by Romanians to different domains of life is constant along time, with the exception of religion. This one, found in an ascending evolution, in 2005 reaches a very close level to the importance offered to work, a domain which, traditionally, is placed on the second place in Romania, after the family. Contrary to the general tendency in Western Europe, in Romania the religiosity degree is ascending (see the chapter on religion in the present volume). The religiosity in Romania represents an explanatory component even for the importance of family. The majority of the individuals in Romania (70%) consider that the church offers answers to the problems of family life, besides the spiritual, moral and social ones. All the socio-demographic criteria trigger significant differences: the faith in solving one’s problems with help from the church is the highest with women, with persons that are more than 55 years old, with the persons with a low level of education, with the ones who live in the rural areas and the ones who experimented unhappy family events (divorced, widows or widowers). Figure 2 The importance of different domains in the individual’s life 100 90

1993

1999

2005

80 70 60 50 40 30 20 10 0

Family

Work

Religion

Friends

Leissure

Politics

Note: The figures represent the cumulative percentage of the ones who answered important and very important. Source: EVS/WVS Romania 1993, 1999, 2005.

The pattern of the importance of different domains in the individual’s life is common with most of the European states. Pollini (2006) identifies 2 different models: “structure A”, where the family is placed on the first place and 176 corectura 2 28 iulie 2008

Family values in Romania and in Europe

work on the second, a category to which the majority of the European countries belong and “structure B”, where the family is on the first place and friends, on the second. To the latter category, belong the states with postmaterialist values (Inglehart 1997), for which friends and leisure time play a more important role than work: Germany, Finland, Holland, Great Britain, North Ireland, Denmark and Switzerland. In the perspective of postmaterialist values, friends become the same important as relatives. Friendships can manage to do things the family failed to do; it represents a kind of “family that you construct”, “a family of choice” (Hardyment, 2000, p.87). Granovetter (1974) demonstrated that friends, neighbours, acquaintances play a very important role in the individual’s life and even in the successful functioning of the family (Broderick, 1988, p.298), ensuring crucial contact with the public sphere.

The importance of marriage: is marriage an outdated institution? Anthony Giddens (1992) considers that the popularity of the marriage’s institution (against the spreading of the alternative cohabitation patterns) is an indicator of “the search for a complete relation”. Modernity is characterized by “institutional reflexivity”, a process of permanent re-evaluation/reconsideration of the social institutions. As in the case of the other spheres of the social life, individuals redefine their needs, their wishes, and their expectations in the sphere of personal, intimate relations continuously. For the great majority of individuals in the European countries, marriage is still valorised representing an important social institution. The highest agreement with the statement according to which marriage is an outdated institution is encountered in France, Luxembourg, Belgium and the lowest is in Malta, Island, Turkey and Albania. Romania records a relatively low value in comparison with the other European countries included in the analysis. The historical experience and the cultural factor seem to be responsible for the differences among countries. Countries with a larger spreading of the post-materialist values do not assign the same significances to marriage, its role of a fundamental institution in the society being reconsidered. On the contrary, former socialist countries (to a smaller extent the ones in the former Soviet Union) keep a high attachment to this institution. In Romania, for the greatest majority of individuals, marriage does not represent an outdated institution, but the percentage of those who agree with this statement is growing. The agreement with this statement is higher in the case of youngsters and of those who live in alternative family arrangements to the legally constituted unions: consensual unions, divorced persons.

177

The Values of Romanians: 1993-2006. A sociological perspective Figure 3. The share of population who declare that the marriage is an outdated institution

16 14 12 10 8 6 4 2 0

12,5

14,3

8,6

1993

1999

2005

Source: EVS/WVS Romania 1993, 1999, 2005

Surprisingly, marriage is considered an outdated institution especially by the persons with a lower level of education (22% of the ones without education in comparison with 10% of the ones with a university degree) or by the ones in the rural area (17% in comparison with 12% of the ones in the urban area). In this respect, the analyses have to be made carefully, because/as there can be identified different ways to approach marriage values. The decreasing importance of the institution of marriage for the persons with a low level of education living in rural areas cannot be explained by their less traditional value orientations. The analyses focusing on poverty or on vulnerable groups in the Romanian society (Zamfir ed., 1999, Stănculescu ed., 2004) draw the attention on the consolidation of the family disorganization processes in certain poor communities with low human capital, from the suburbs of big cities or from the rural area. Survey data, confirmed by the census data, emphasize the fact that many of the alternative patterns of families (single-parent families, reconstituted families, consensual unions) come from poor communities or from the rural area, rather being of a traditional type. Consequently, we can estimate that for a part of the population considered in the analysis, the agreement with the statement according to which marriage is an outdated institution represents rather an indicator of adaptation to poor living conditions and of some processes of social disorganization, more than an indicator of some important value changes. Despite a slight decline of the institution of family, numerous researches showed that the ratio of relatively traditional family behaviours stayed very high. People are more individualistic in words than in facts and, 178 corectura 2 28 iulie 2008

Family values in Romania and in Europe

especially, they are permissive in what concerns the others than in what concerns themselves: “Accepting the rise in the number of alternative relations does not mean that people themselves will get involved in these kinds of relations. Marriage is not considered indispensable anymore, but, at the same time, marriage is still preferred (…); what changed in these areas of family life was a rise of tolerance for the behaviours considered unacceptable before, but not also a rise in the active involvement in such behaviours” (Elster, Halman, de Moor, 1993, p.14).

The importance of marriage: tolerance for different aspects regarding marriage and sexuality The highest tolerance for divorce can be found in Northern states (values around 7 on a scale from 1 to 10). Romania has one of the lowest values, together with countries with a more conservative culture like Malta, Turkey or Albania. The negative values (average values lower than 5, which is the average value on the scale) belong generally to Eastern countries (with the exception of Ireland). In comparison with other aspects discussed, related to family and intimate relations, tolerance for divorce is higher (average value 5.5), followed by the tolerance for abortion (average 4.6), homosexuality (average 4) and casual sex (average 3.1). The lowest tolerance is the one for prostitution (average 2.7) and adultery (average 2.6). Comparing data for every country, two categories seem to stand out: one of the countries in which the postmodern values prevail and another one of the former communist and conservative countries. In the first category, it can be appreciated that the tolerance for divorce is not specific, but is an expression of a general tolerance, a common value in the relation to the analysed aspects. Still, a high tolerance for divorce or abortion is found in many of the former socialist countries, as a consequence of the fact that this kind of behaviour is quite common in these countries. The states of the former Soviet Union have a pattern of a high rate of divorce (among the highest values in Europe) that explains the high tolerance for it. In the context of the restrictive pro-natalist policy in the socialist period and on the background of the missing family planning education, the practice of abortion was widespread in the majority of these countries. In what concerns the level of tolerance for some other phenomena (for example, homosexuality), it is very low for these countries: with the exception of the Czech Republic, Slovakia and Slovenia, all the other former socialist countries, together with Malta and Turkey record average values lower than 3 (on a 1 to 10 scale). Surprisingly, these countries are less intolerant for prostitution or adultery in comparison with many of the Western European states. For example, Northern states show the highest values towards homosexuality, divorce, abortion and among the lowest values towards the casual sex and adultery. 179

The Values of Romanians: 1993-2006. A sociological perspective

Inglehart (1997) considers that the postmodern value orientation can be measured by three directions: the rejection of modernity, the revalorization of tradition, the spreading of new values and lifestyles. Consequently, we can explain how the countries with postmaterialist values reject to a larger extent practices like adultery, considering them unjustified choices. In modernity, casual sex or adultery represented ways of adapting to external social pressures (the impossibility of separation, “saving the appearances”). Given the fact that in the postmodern society these constraints have disappeared, behaviours of this type are no longer reasonable. In this context, the revalorization of tradition should be understood as the enforcement of the options based on principles, the morals of a -controlled hedonism. Consequently, the postmodern value orientation is not characterized by an increase in undifferentiated tolerance, but for assumed individual choices, for responsible life options. Romania presents low values at all the analysed aspects, placed next to Malta, Turkey, and Albania which are among the most intolerant countries. The religious factor explains a part of intolerance that these countries show, being the countries with the highest level of religiosity in Europe. The low tolerance is also explained by other factors such as the low level of education (the share of the population with higher education is low in the European context), the high percentage of people who live in rural area where values of traditional type or low level of trust in people prevail. The level of trust is high only in the case of family (97%) or relatives (80%), less in the case of neighbours (50%) or friends, acquaintances (56%) and in the case of unknown people it is quite low (12%). The reduced relational social capital, rather of „bonding”2 type (based on family and inside the affiliation group relationships) also represents an important explanation of the Romanians’ intolerance (B.Voicu, 2005).

2

In Robert’s Putnam terms regarding the two types of social capital: “bonding” (relationships with the people like you, inside the affiliation group) and “bridging” (useful relationships with individuals with different social statuses, outside the affiliation group, with an important role in the individual fulfillment and social development).

180 corectura 2 28 iulie 2008

Family values in Romania and in Europe Table 1 Tolerance for aspects of family life and sexuality, on socio-demographic criteria How justifiable is:

sex

age

education

area of residence Total

Divorce

Abortion

Prostitution

man

4.3

3.3

2.2

Homo sexuality 2.0

woman

4.2

3.3

2.1

2.1

18-24 years old

4.6

3.5

2.6

2.7

25-34 years old

4.6

3.5

2.3

2.3

35-54 years old

4.6

3.7

2.3

2.2

55 or more years old

3.6

2.7

1.7

1.6

Elementary (up to 8 grades) Secondary education (vocational school, college) Higher education (university degree) rural

3.5

2.5

1.6

1.5

4.4

3.5

2.2

2.2

5.6

4.5

3.1

3.2

3.6

2.7

1.8

1.7

urban

4.7

3.7

2.4

2.4

4.2

3.3

2.1

2.1

Note: average values, on a scale from 1– never justifiable to 10 – always justifiable Source: WVS Romania, 2005

Tolerance for divorce, abortion or homosexuality is different depending on individuals’ socio-demographic characteristics. The tolerance varies significantly depending on age, level of education and residence area. The highest values are registered with youngsters, people with higher education and with persons who live in urban areas. The differences according to sex are not statistically significant. It has to be noticed that all the analysed aspects record values closer to the negative pole, only tolerate divorce to people with higher education recording a value over 5. The specific tolerance for each phenomenon is an expression of the general tolerance, as there is a significant association between variables3.

3

The correlation coefficients are: divorce – abortion = 0,707/divorce – prostitution = 0,458/divorce – homosexuality = 0,385/prostitution homosexuality =0,686/prostitution – abortion = 0,541/abortion – homosexuality = 0,485. All of them are significant at the level p<0,0005. 181

The Values of Romanians: 1993-2006. A sociological perspective

Alternative lifestyles: single mothers Baumann (2003) shows that if in the modern period the identity was based on stability, in postmodernity, individuals search for means of avoiding determinism and not restraining their opportunities or „the modern obsession of gaining in all areas and not sacrificing any of the possibilities” (Béjin, 1998, p.178). The characteristics of the postmodern life strategy are not stable, the rules of the game change permanently. Alternative styles of family life – singleparent families, consensual unions, homosexual couples, etc. are more and more spread. Marriage still remains a stable value and, as we showed, the majority consider that the importance of family should increase in the future. Figure 4. Woman as a single parent (Romania) 1993

50%

49%

45%

1999

2005

48%

40% 35%

38%

38% 34%

30%

28%

25%

30%

20%

22%

15%

14%

10% 5% 0%

approve

dissaprove

it depends

Source: WVS/EVS Romania 1993, 1999, 2005

Generally, single mothers are largely accepted in the society. In the attempt of classification, the political and historical criteria seem to not function anymore, high values being registered both in the Northern countries and in the former socialist countries. The lowest agreement (below 20%) is found in traditional countries such as Turkey and Albania or in more conservative catholic countries like Malta. Still, clear categories cannot be established, for example Spain’s positioning among the most open countries to this alternative pattern of family and Sweden’s positioning among the most reserved countries are somewhat surprising. The percentage of single mothers in those societies, the problems associated with this type of family or the type of social policy adopted, provide possible explanations. Sweden, for example, has developed a social policy that includes a lot of support benefits for single mothers. 182 corectura 2 28 iulie 2008

Family values in Romania and in Europe

The countries with a higher percentage of single-parent families (usually ruled by women), like Great Britain or Estonia seem to manifest a rather reserved attitude towards this pattern (the higher percentage is taken by the answer, “it depends on the situation”). Still, in Latvia for example, we find a percentage of single mothers, 30% in the total number of families, but also a high agreement of this family type (above 55%). We can estimate that, instead, contextual variables related to social policy (“positive discrimination” how much are the single-parent families advantaged by the social benefits) or to problems associated with this type of family in the respective society can form a reserved attitude. In Romania, the acceptance of these families is relatively high in comparison with the other European countries (approximately half of the respondents agree). According to the census in 2002, 11, 4% of the households are of this type, which represents an average value in a European context. Along time, a tendency of polarization of the answers can be noticed – the percentage of those who answer “it depends on the situation” has decreased and the share of people who disapprove has increased, the percentage of the ones who approve being rather constant in the last years. The approval of single mothers is higher among young people (57% of the ones younger than 24, in comparison with 38% of the ones older than 65), among persons with a high level of education (64% of the ones with higher education in comparison with 24% of the ones without education), in urban areas (57% in comparison with 36% in rural areas), but none of these socio-demographic characteristics induce statistically significant differences. The marital status also represents a criterion of differentiation, the acceptance being higher among the persons who are not married or who live in informal unions, who are divorced or separated in comparison with married persons or with widows and widowers.

Alternative lifestyles: consensual couples Consensual unions represent a cohabitation pattern more and more spread in the contemporary societies, symbolizing a new type of intimate relationship, called by Giddens (1992) “pure relationship”, conceiving it as a relationship existing only by itself, without any external constraints. Refusing to become formal, the couple tries to develop a completely self-sufficient relationship. The vulnerability of this relationship is what makes it valuable in the partners’ eyes. It offers a large range of choices, the partners are free to set up rules to protect the couple from dissolution – regarding, for example, the fidelity or the distribution of the household responsibilities, rules that do not have any external support. The success of the relationship depends on the result of compatibility or of the common understanding attempts.

183

The Values of Romanians: 1993-2006. A sociological perspective

Consensual couples are not spread out in Romania. For the first time, in the 2002 Census4 there were 828000 registered persons (only 3.9% of the total population), 3.2% in the urban areas and 4.6% the in rural areas of people who are living in such unions. The couples of this type represent 6.5% of all the households. Only half of the ones who live in consensual unions are aged 20-34, a category in which, in Western societies, cohabitations are the most numerous, (hence, the name of “juvenile cohabitation” under which the phenomenon is also known). The data in the WVS survey (2005) estimate an even lower percentage of these ones, 2.8% of the investigated population declaring that they live as if they were married. Analysing the socio-demographic characteristics of this population (taking also into account that in the sample there are few cases, so we cannot draw any conclusion), we can identify surprising tendencies. Being an indicator of postmodern family lifestyles, the higher percentage of the consensual unions in the rural areas or among the ones with a low level of education (almost half have an elementary education) is surprising. An important part of the population (a quarter) is represented by the Roma people and they have low incomes (one third place in the first income quintile). By age categories, only half of the persons who live in consensual unions are younger than 35. The profile of the sample, similar to the one described by the Census data, illustrates a heterogeneous composition of the population who lives in unions in Romania. In urban areas and especially in big cities, consensual informal unions are more frequent with youngsters, the ones with a high level of education, unmarried or divorced persons and with more pronounced postmaterialist value orientations. In the rural areas we encounter cohabitations between persons of different ages, rather single or widows and widowers, from Roma traditional communities (they follow the pattern of the “marriage without certificate”, established on the norms of the community and informally recognized, but illegalized). Besides, the preference for unofficial family arrangements is frequently found in poor communities, confronted with social disorganisation processes. Therefore, we can estimate that the persons who live in consensual unions in Romania form a heterogeneous class, which tends to polarize in two different categories: one for which the cohabitation represents a contextual option, a result of the adaptation to a problematic situation (family problems, widowhood, the lack of a dwelling, the lack of incomes, disorganization) and another one for which the consensual union represents a life choice, the expression of a postmodern value orientation. It is difficult to approximate the share of the two categories in the total number of unions of this type. Even if the involvement in such relationships is not so spread out, the tolerance for the unmarried couples who live together is high. Only 15% 4

According to the “Census of Population and Dwellings”, National Institute for Statistics, 2002.

184 corectura 2 28 iulie 2008

Family values in Romania and in Europe

mentioned that they would not like to have such a couple among their neighbours. Education induces significant differences, as it was expected; social distance is more reduced in the case of the people with higher education (9%), in comparison with the people without education (23%). The marital status also influences the tolerance towards consensual unions. The persons who live in unions, unmarried persons, are more tolerant in comparison with the persons who are married, divorced, separated or widowed. Young persons are more tolerant, but the difference from the other age categories is not statistically significant. The affection for/devotion to a stable relationship remains vey high. In most of the European countries, the majority consider that a stable relationship is necessary in order to be happy. As in the case of the perception of the importance of the institution of marriage, former communist countries have the highest values (above 80%). In Western countries and in Southern Europe (Greece, Portugal, France and Italy) the belief in the ideal of a long-lasting relationship is more encountered. The countries where the postmaterialist values prevail (the Northern countries, Great Britain) record the lowest values (below 40%). In Holland only one fifth of the population considers that a stable relationship is necessary for one’s own happiness. Comparatively, Romania presents one of the highest agreements with this statement (approximately 85%). Complementary, it is interesting to notice that among all the analysed aspects regarding the tolerance, the strongest intolerance is the one for adultery, the norm of the fidelity in the couple staying the same important as in the past. In the sphere of intimate relationships, Europeans generally share the ideal of a stable, self-sufficient relationship based on fidelity (given the fact that more and more of the external constraints disappeared).

Roles and statuses in the family Considering that a housewife is fulfilled just as as a wife who works, represents an indicator difficult to analyse. Malta, with more conservative value orientations, together with Finland5, where post-materialist value orientations are spread, also records the highest acceptance of this lifestyle (87%, respectively 80%). A high agreement (more than 70% of the investigated population) is found in several former socialist countries such as Lithuania, the Czech Republic, Moldavia, Bosnia or Slovakia and in more traditional countries like Turkey. In the Western countries, the agreement is relatively high 5

Finland confronts with a special situation – the profound economic crisis in the '90s led to a strong recoil of the society to values of a traditional type.The family and the church experienced a spectacular rise of their importance as social institutions in the context of a return to traditional values (see the observations in the last chapter of this work). 185

The Values of Romanians: 1993-2006. A sociological perspective

(approximately 60%) in Iceland, France, Belgium, Great Britain, Spain but lower in Northern countries like Denmark, Holland, Sweden. Romania is part of the category of the countries with a low agreement (below 50%) together with Croatia, Bulgaria, Albania, Greece, Latvia, and Germany. Different factors provide the explanations: the low share of working women, the woman’s involvement in housework despite the employment on the labour market, the woman’s image in the society or even contextual factors that belong to the economic and political situation in every country. Generally speaking, every country with a high rate of occupancy tends to manifest a lower agreement with this statement. Malta and Turkey as well as conservative countries, record the lowest rates of women’s employment (34%, respectively 24%) and they also show the highest agreement with this statement. The woman’s employment does not seem to be the only determinant factor. The explanatory model also has to take into account other contextual factors regarding the economic situation or the type of social policy. For the former socialist countries, the agreement with this statement represents rather an expression of a culture of woman’s involvement both in the labour market and household. For the Northern states, massive woman’s employment in the labour market, gender equality value represents the main explanation. In Romania, the agreement with the statement, according to which to be a housewife can contribute to the woman’s personal fulfilment to the same extent as the choice of having a career, is quite low and its findings have a descending tendency. Housewives are less and less valorised. Analyses of gender values (M.Voicu, 2004; Voicu &Voicu, 2002) show that Romania shares the common pattern of the majority of the former communist countries, characterized by the massive participation of women in the labour market, doubled by a large value support, similar to Northern countries (lower in comparison with Sweden, but higher in comparison with Finland and Denmark) and significantly higher than in Western societies. On the other hand, the equality between women and men inside the household is supported by fewer Romanians than in most of the Western European countries and even Eastern European countries: women were responsible for the household duties to a larger extent. The classical division of work inside the household is maintained and the woman has, in fact, two jobs: one on the labour market and another one at home. The culture of the woman’s involvement both in the labour market and in the household explains this situation only partially. Woman’s exceeding responsibilities inside the household represents, at the same time, an indicator of the low standard of living (R. Popescu, 2002).

186 corectura 2 28 iulie 2008

Family values in Romania and in Europe Figure 5 The Romanians’ agreement with the statement: “Being a housewife is just as fulfilling as a paid work” agree 50%

strongly agree

45% 40% 35%

32%

34%

30% 25%

25% 20% 15% 10%

16%

14%

10%

5% 0%

1993

1999

2005

Source: WVS Romania, 2005

There are no significant differences between women and men in this respect. However education strongly influences the level of aspirations and the individuals’ value orientations. Housewives are less valorised by the persons with higher education (only 23% of the ones with a university degree agreed with this statement in comparison with 40% of the ones with primary education). At the same time, the acceptance of the housewife pattern is widespread in rural areas, where the woman’s employment on the labour market is much lower and where the traditional value orientations are more frequent.

Satisfaction with the family life The family is one of the domains that confers the greatest satisfactions to the individuals, 93% of people being satisfied and very satisfied with their family life. Women have a lower satisfaction level, as a consequence of the fact that they confront to a larger extent with the household duties. Satisfaction with the family relations tends to decrease along with age, according to the family life cycle theory. The area of residence is also a criterion of significant discrimination. The individuals in the urban areas are more satisfied with their family relations in comparison with the ones in the rural areas. The high level of education also involves high satisfaction (an almost double percentage – 72% of the ones with a university degree declared they were very happy in comparison with 37.6% among the ones who have completed elementary school). The fear of the aggravation of family life conditions is more reduced, the percentage of worried people decreasing with more than one fifth in 187

The Values of Romanians: 1993-2006. A sociological perspective

comparison with the percentage recorded in the 1999 wave. Despite this alleviation, given the maintenance of the economic uncertainty, the worries for the family life conditions remain widespread– approximately 80% of the people answered that they are afraid or very afraid. The fears are more frequent for the categories with higher responsibility in ensuring the economic function of the family: for women, who traditionally are confronted with more household duties, for the categories of age more active on the labour market, for the persons living in urban areas and for the persons with a lower level of education, more vulnerable during economic changes.

European patterns of family value orientations Considering all the analyzed variables, Europe remains a heterogeneous space of values, with several categories and distinctive profiles, from conservative countries to individualistic and tolerant ones. The majority of exsocialist countries are grouped in a distinctive class as well as the countries driven by strong religious beliefs. Figure 6 The classification of European societies according to their value orientations regarding family

"libertine" (9) conservative (2) rejecting single mothers (8) tolerant & inidvidualistic (4) former socialist (14)

Source: EVS/WVS 1999-2001

188 corectura 2 28 iulie 2008

Family values in Romania and in Europe

A classification of the family values pattern existing in the European societies can contribute to a better understanding of Romania’s position in the continent from the family value’s point of view. We selected 10 of the indicators described in the previous sections, presented in Table 2. Then we grouped the European societies depending on the distances among them given by the average values of the indicators. The cluster 6 analysis shows the fact that European countries group in five distinctive classes. Each one of these classes is characterized in Table 2 by the average value that countries record for each one of the indicators. It has to be noticed that the labels we used and the characteristics of each class, as well, are relative to the Europe assembly and the classifications are not absolute. Table 2 Average values of the variables depending on each class Conservative countries Number of the countries in each cluster Family is very important Family has to have a greater importance in the future Single mother without a stable relationship A stable relationship is necessary to be happy A housekeeper wife is as fulfilled as a working wife Marriage is and outdated institution Tolerance for divorce Tolerance for adultery Tolerance for homosexuality Tolerance for abortion

Types of societies Tolerant and Countries which “Libertine” individualistic reject single countries countries mothers

Ex-socialist countries

2

4

9

9

14

9,6

8,7

8,7

8,5

8,3

9,8

8,3

9,1

9,0

8,8

5,1

7,2

4,1

6,5

6,9

7,4

5,7

6,9

7,2

7,9

4,8

6,1

5,9

5,7

5,8

0,7

1,6

1,6

2,2

1,7

3,1 1,1 2,1 1,9

7,1 2,2 7,2 6,2

5,1 2,6 4,1 4,0

6,1 2,9 5,3 5,2

4,9 2,7 2,5 4,2

Note: The items were calculated on a scale from 1 to 10. The figures in the table represent averages of the indicators on the rows by the cluster of countries (on the columns). The higher a value is, the stronger the average support for the statement in the cluster is. The maximum values on each row are marked with bold and minimum values are marked with italics and they are also underlined.

6

We used both the Hierarchical Cluster Analysis and K-means Cluster from SPSS. The calculation method of the distances between cases (countries) was the square Euclidian distance. For measuring distances between clusters, Ward and BAVERAGE were used, the result confirming the stability of the solution with 5 clusters (adjusted Rand index=0.78). The solution explains (ETA2k=) 64% from the variation of the analyzed items. 189

The Values of Romanians: 1993-2006. A sociological perspective

The five categories of societies are briefly described as it follows: 1.

2.

3.

4.

5.

The first category is the one of the conservative countries where Malta and Turkey are grouped together. They offer high importance to the family and develop conservative attitudes for the all analysed items: general intolerance, disagreement with the alternative family lifestyles, and support for the woman involved in housekeeping activities. Individualistic and tolerant countries – Denmark, Island, Holland, Sweden – display the highest level of tolerance (except adultery), support alternative family lifestyles (single mothers), support woman employment, give less importance to family in the future and show less support for a stable relationship. Postmaterialistic values are widespread. The countries that reject the single mothers – Albania, Estonia, Germany, Ireland, Italy, Poland, Slovakia and Great Britain – have average values on all the dimensions, but they disapprove single-mother families. The percentage of the families of this type (with the exception of Albania and Italy) are among the highest in Europe and the rejection of single mothers is probably explained by the problems associated with this type of family in the respective society. Libertine countries – Austria, Belgium, the Czech Republic, Finland, France, Greece, Luxembourg, Slovenia, Spain, and North Ireland – believe to the largest extent that the institution of marriage is outdated and have a lower intolerance for adultery and average values for the other dimensions. The block of the former socialist countries groups the majority of the former communist countries, with the exception of Estonia, Poland and Slovakia (that follow the pattern of the countries that reject single mothers) and of the Czech Republic and Slovenia (which are grouped in the class of the “libertine” countries). It also includes Portugal, with a totalitarian past and with a lower development level in comparison with the rest of the Western countries. These countries consider family less important compared to the other groups, but they value more stable relationships as a central element for a happy life. They have a high agreement with single mother families but have a low general tolerance (the lowest values except the conservative group).

Romania is part of the group of former socialist countries, but values for certain indicators make it similar with the group of the conservative countries, describing a pattern where the family represents a fundamental institution in the society, whose importance should increase in the future.

190 corectura 2 28 iulie 2008

Family values in Romania and in Europe

Conclusions The family represents the aspect of life that offers the highest satisfaction to the individuals and is followed by work and religion. On the background of economic improvements in the last years, the fear of aggravation of life conditions of the family record a descending tendency. Despite this alleviation, because of the economic uncertainty maintaining, the worries regarding the family life conditions are widespread, being more frequent at the categories that have a higher responsibility in assuring the economic function of the family (among women, who are traditionally concerned with more household duties, for the economically active age categories), for the people in urban areas and for persons with a lower level of education, more vulnerable during economic changes. Marriage does not represent an outdated institution, but the share of the people agreeing with this statement is rising, especially in the youngsters’ case, but also for a distinct category of the population with low human capital. On the background of the consolidation of some processes of family disorganization in certain poor communities, considering marriage an outdated institution must be interpreted as an indicator of adjustment to poor living conditions and social disorganization, rather than an expression of some value changes. The devotion to a stable relationship remains very high in comparison with the other European countries. The alternative family patterns – single mothers or consensual couples – are not widespread and their significance can be contextually different. The cohabiting persons in Romania form a heterogeneous class, which tends to polarize in two distinct categories: one for which cohabitation represents more a contextual choice, the result of adjustment to a problematic situation and another one for which a consensual union represents a life choice, the expression of a postmodern value orientation, being difficult to approximate the percentage of the two categories among the overall consensual unions. The agreement with the statement that being a housewife can be as fulfilling for a woman as having a career is quite low and finds itself in a descending tendency. The housewife is less and less valorised, especially by youngsters and persons with higher education. Romania presents among the lowest values at all the aspects concerned with tolerance, a fact which makes it similar to the conservative countries. The most open and tolerant attitude, with post-materialistic family values is characteristic rather to a minority: young people in the urban areas, with higher education and average incomes. The religious factor, the low level of education for the overall population , the high share of the rural areas in which the traditional values prevail, low level of trust in people and the scarce relational social capital represent explanations for the Romanians’ conservative attitude. The family value pattern is changing, but the postmodern tendencies are weak, characteristic rather to a minority. The importance of family remains very high, placing it on the first position in the values hierarchy and representing the domain that confers the greatest satisfaction to individuals. 191

The Values of Romanians: 1993-2006. A sociological perspective

References Béjin, André, 1998. „Căsătoria extraconjugală de astăzi”, in Phillipe Ariès and André Béjin (eds.)., Sexualităţi occidentale, Antet Bauman Z., 2003. Liquid Love: On the Frailty of Human Bonds, Cambridge: Polity Press Elster, Peter; Loek Halman and Ruud de Moor, 1993. „Value Shift in Western Societies”, in Peter Elster, Loek Halman and Ruud de Moor (ed.)., The Individualizing Society. Value Change in Europe and North America, Tilburg Unversity Press Ghebrea, Georgeta, 2000. Regim social-politic şi viaţă privată (Familia şi politica familială în România), The Publishing House of the University in Bucharest. Giddens, Anthony, 2000 (1992). Transformarea intimităţii. Sexualitatea, dragostea şi erotismul în societăţile moderne, Antet Inglehart R, 1997. Modernization and Postmodernization. Cultural, Economic and Political Change in 43 Societies, Princeton: Princetone University Press Hardyment, Christina, 2000. Viitorul familiei, Bucureşti: Editura Ştiinţifică Mihăilescu, Ioan, 2000, „De la familie la familii”, in Ioan Mihăilescu (ed.)., Un deceniu de traziţie. Situaţia copilului şi a familiei în România, Bucharest: UNICEF Nikodem K, P. Aracic, 2005., „The family in Transformation” in Josip Baloban (ed). In Search of Identity. A Comparative Study of Values: Croatia and Europe, Zagreb, Golden marketing – Tehnicka knjiga Parkinson, Lisa, 1993. Separarea, divorţul şi familia, Alternative Publishing House, Bucharest Popescu, Raluca, 2002. „Situaţia familiei şi a copilului în societatea românească” in Ioan Mărginean şi Ana Balaşa (eds). Calitatea Vieţii în România, , Expert Stănculescu Manuela, Ionica Berevoescu (eds), 2004. Sărac lipit, caut altă viaţă, Polirom Voicu, Bogdan, 2005., Penuria Pseudo-Modernă a Postcomunismului Românesc, Expert Projects Voicu Mălina, Bogdan Voicu. 2002. Gender values dynamics. Towards a common European pattern?, Romanian Journal of Sociology, XIII (2002), 1-2, p. 42-63. Voicu, Mălina, 2004. “Women Work and Family Life: Value Patterns and Policy Making”, in Will Arts and Loek Halman (editors). European Values at the Turn of the Millennium, Leiden: Brill Van den Akker, Piet; Loek Halman; Ruud de Moor, 1993., „Primary Relations in Western Societies”, in Peter Elster, Loek Halman and Ruud de Moor (editors)., The Individualizing Society. Value Change in Europe and North America, Tilburg Unversity Press Zamfir, Cătălin (ed.), 1999. Politici sociale în România: 1990 - 1998, Bucureşti:Expert

192 corectura 2 28 iulie 2008

Social status and child-rearing values1 PAULA A. TUFIŞ

Child-rearing values are often associated in sociological approaches with individuals’ locations within the social class system or their social status (see, for example: Duvall, 1946; Inkeles, 1960; Kohn, 1963, 1969; Kohn et al., 1983; Alwin, 1989; Kohn et al., 1990; Spade, 1991). Kluckhohn (1951, p.395) defines a value as “a conception, explicit or implicit, distinctive of an individual or characteristic of a group, of the desirable which influences the selection from available modes, means, and ends of action”. Based on this definition, Kohn (1963, p. 471; 1969, p. 7) considers child-rearing values as parents’ conceptions of what is desirable for their children. The link between parental values and social class is mainly explained by the differences in socio-economic profiles (described by education, occupation, and earnings) characterizing each social class. These differences in socio-economic profiles determine differences between classes in terms of living conditions and lifestyles, which, in turn, translate into class-specific world-views and class-specific values. This assumption is the basis for formulating the hypothesis that different social classes adopt different child-rearing attitudes, values, and practices, which may determine differences in children’s personalities and their educational and occupational career trajectories. Much of the Western child-rearing values literature has focused on differences in child-rearing values between middle class and working class parents, in other words, the first link in the causal chain described above. This chapter will also focus on this link by first presenting some of the theories advanced in the literature in order to explain the relationship between social class or social status and child-rearing values. Based on these theories, in the second part of the chapter, I will propose a model for studying some of the 1

I would like to thank Duane Alwin, Dumitru Sandu, and Claudiu Tufiş for their valuable suggestions and comments on earlier drafts of this chapter. 193

The Values of Romanians: 1993-2006. A sociological perspective

mechanisms that could explain this relationship. The main starting point for the model is Kohn’s and his colleagues’ work (Kohn, 1963; Pearlin and Kohn, 1966; Kohn, 1969; Slomczynski et al., 1981; Kohn et al., 1983; Kohn et al., 1986; Kohn et al., 1990). In the third part of this chapter, I will detail the methodology used for the statistical analyses and then present the estimated empirical model results, using the 2005 Romanian data from the World Values Survey (WVS). Since some of the theoretical assumptions present in the childrearing values literature may be generalized to a variety of national and cultural contexts, while others may be sensitive to differences between national contexts2, the present empirical analysis is mainly an exploratory one, aiming to describe relationships in the present Romanian context. I will conclude by reviewing some of the more important results of the analysis presented in this chapter and discuss possible implications of these results. In the U.S. context, empirical results (see, for example: Kohn, 1963; 1969) suggest that working class parents place more value on their children’s ability to conform to a set of externally imposed rules, while middle class parents focus more on developing their children’s ability for self-control and on the development of a set of internal behavior guidelines. Parental values3 reflect parents’ ambitions for their children and their conceptions about the most efficient strategies to fulfill these ambitions. Two social status components (education and occupation) play an important part in the formation of parents’ views on what constitutes an efficient strategy, through their effects on intellectual flexibility and occupational autonomy (Kohn, 1969; Alwin, 1989). First, based on the assumption that higher schooling levels focus more on developing abstract and critical thinking, people with higher educational attainments will tend to examine externally imposed rules and consequences of following these rules in specific situations more closely. Kohn (1963) associates higher educational levels with greater intellectual flexibility, Wright and Wright (1976) with enlightening social experiences (such as reading, traveling, or being active in various associations and organizations), Alwin (1984) with openness to the idea to think for oneself, and Bowles and Gintis (1976) find the dichotomy between conformity and autonomy precisely in the organization of the schooling experience (the first school years versus medium and high levels of schooling). The link between education and parental values is thus explained by the degree of intellectual flexibility. 2

The conclusions regarding the relationship between social class or social status and child-rearing values in the US are confirmed in other national contexts as well (Inkeles, 1960; Pearlin and Kohn, 1966; Kohn, 1969). Most of the analyses in the child-rearing values literature refer to Western capitalist countries, but the relationship was also confirmed to exist in Poland (Slomczynski et al., 1981; Kohn et al., 1986), Japan (Kohn et al., 1990) and China (Xiao, 2000b). 3 In this chapter, I use the terms “child-rearing values” and “parental values” interchangeably. 194 corectura 2 28 iulie 2008

Social status and child-rearing values

Secondly, based on the assumption that occupations with higher social status are characterized by different behavior, social interaction, and promotion guidelines than occupations with lower social status, the occupational status will have an impact on individuals’ values, strategies for success, and worldviews. While members of middle or upper class occupations are confronted with occupational situations that tend to require independent action, members of working class occupations are probably exposed more often to occupational situations in which conformity with externally imposed rules is essential. Kohn (1963, p. 476) argues that the main difference between middle class and working class occupations is that the former imply the manipulation of interpersonal relationships, ideas, and symbols, while the latter involve the manipulation of objects. The second important difference in his view is that standardization and supervision levels are higher in working class occupations, compared to middle class occupations. A third difference, in his opinion, is that problems tend to be solved through collective action in working class occupations, and through individual action in middle class occupations. Inkeles (1960) also links the type of occupation to values, experiences, and perceptions. He associates working class occupations with greater pressures to conform. In this case, the degree of occupational autonomy is one of the main mechanisms mediating the relationship between occupation and parental values. The parents’ financial situation has a weaker impact on child-rearing values, compared to those of education and occupation (Kohn, 1969, p. 137; Spade, 1991, p. 349). However, taking into account that the financial situation is correlated with education and occupation, the link between income and parental values may be explained in terms of mechanisms linked to education and occupation such as those discussed above. In studies that conceptualize social stratification in terms of discrete social classes, the middle class (characterized by higher educational and occupational levels and better financial situations compared to the working class) tends to place more value on developing children’s autonomy, selfcontrol, and their reliance on a set of internal rules for guiding behavior. The working class tends to place more value on conformity and adherence to externally imposed rules in children’s socialization. Moving on to a conception of social stratification based on social gradations, the previously mentioned hypotheses can be generalized in terms of social status and parental values: the higher one’s social position, the more inclined one will be to focus on the development of children’s autonomy and the less inclined to raise children in the spirit of conformity and obedience. The choice between social class and social gradations conceptualizations of social stratification is largely determined by the author’s theoretical preferences4. 4

See Kohn et al. (1990) for a detailed discussion of differences between social class conceptualizations and social gradations conceptualizations in studies on child rearing values. 195

The Values of Romanians: 1993-2006. A sociological perspective

For example, Kohn (1963; 1969; Kohn et al., 1990) uses the social class approach and compares middle class parental values to working class parental values. However, he points out that no social class is homogenous and parental values vary within each social class, according to social status gradations within each class (Kohn, 1963, p. 472). Other studies (Slomczynski et al., 1981; Alwin and Jackson, 1982; Alwin, 1986; Luster et al., 1989) employ the social gradations approach. In this case, the comparison between two or more discrete groups is no longer possible and the research problem is reformulated in terms of the magnitude of the effect of social status on child-rearing values. Different authors label the theoretical dimensions that appear in the classification of parental values differently, but there is a high degree of overlap among classifications. Duvall (1946) differentiates between developmental parental values (children should be healthy and happy, should be able to share and cooperate with others, should love their parents and confide in them, should be eager to learn, should be capable of taking care of themselves and also capable of handling different situations) and traditional parental values (the children should be clean and neat, should obey and respect their parents, should please their parents, should take care of their own things, should be religious, reliable and dependable, and they should also help out around the house). Lenski (1961) contrasts parental preferences for intellectual autonomy (the child should think for him/herself) and preferences for intellectual heteronomy (the child should obey others). The classification adopted by Kohn (1969) groups parental values into values that indicate self-control or self-direction (children should show consideration towards others, should be interested in how and why things happen, should be responsible and should have self-control) and conformity (they should be well-mannered, clean and neat, they should get good marks at school, be honest and obedient). Xiao (2000a), using WVS data for the United States, groups independence, perseverance, and imagination as measures of parental preference for autonomy, and obedience, good manners, and religious faith as measures of parental preference for conformity. The same author uses a slightly different operationalization of parental preference for conformity in China’s case – the indicators he uses are obedience, hard work, and thrift (Xiao, 2000b). The two types of parental values, conventionally labeled autonomy (or self-control or self-direction) and conformity (or obedience) are treated either as the extremes of the same dimension (for example, see Schaefer, 1959; Kohn, 1969; Kohn et al., 1990, Hagenaars et al., 2003), or as separate dimensions (for example, see: Alwin, 1986, 1989; Luster et al., 1989; Xiao, 2000a, 2000b). Generally, the choice is determined partly theoretically and partly empirically. Based on results from several exploratory analyses on the 2005 Romanian WVS data, the two types of parental values are conceptualized here as different dimensions.

196 corectura 2 28 iulie 2008

Social status and child-rearing values

In the United States’ case, empirical data supports the hypothesis of a gradual decrease over time in parental preferences for conformity, in concert with an increase in preferences for children’s autonomy and self-control (Duvall, 1946; Wright and Wright, 1976; Alwin, 1984, 1986, 1990). Parents’ wish to distance themselves from the way they themselves were raised, combined with the influence of experts’ advice on child-rearing advertised by the media (Bronfenbrenner, 1958; Wright and Wright, 1976; Alwin, 1984) may determine a decrease in the association between social status and parental values, caused by the “conversion” of some social segments to the styles and parental values that emphasize the child’s self-control. Although the influence of experts’ advice is likely to take effect first among the highly educated social strata, over time, the child-rearing techniques promoted by experts and their associated values may permeate the other social strata. In addition, overtime, changes in family size and structure may also contribute to the change in the relationship between social status and parental values (Alwin, 1984). Recent studies show that, even in the presence of these changes, the impact of social status on parental values persists and the magnitude of the relationship is still substantial (see, for example: Kohn et al., 1990; Spade, 1991; Xiao, 2000a).

The role of child-rearing values in social reproduction The arguments presented above, according to which parents develop child-rearing strategies based on their own occupational and life experience, do not necessarily imply that this process is exclusively rational, and/or deliberate. Parents might internalize behavior guidelines and success strategies to such an extent that they unwarily and unintentionally come to instill them in their children (Kohn, 1963). Actually, for parents with modest social positions, who aspire to higher social positions for their children, the rational strategy would be to adopt the parental values that characterize the targeted social position. From this point of view, a study on the impact of parental aspirations on parental values could offer interesting conclusions in this field of study. On the other hand, it is possible that parents do not necessarily desire their children to be upwardly mobile as much as to avoid downward social mobility (Inkeles, 1960; Breen and Goldthrope, 1997). In this case, the process through which parents teach their children status-specific values acquires a stronger rational action component. At the same time, the existence of a connection between the parent’s social position and the values parents try to instill in their children (hypothetically, these values are precisely the values the parents themselves adhere to) does not imply social immobility or the perpetuation of poverty and social advantage. Kohn (1969, p.51) notices that the social class impact on parental values is not impressive in terms of its magnitude, but rather in terms of its consistency (the relationship persists across a variety of items that 197

The Values of Romanians: 1993-2006. A sociological perspective

measure parental values and across a variety of national and temporal contexts). In and of itself, the fact that parents, consciously or unconsciously, prepare their children for the same type of occupation and the same type of social position as theirs – in the same way schools play a role in the reproduction of inequalities (Bowles and Gintis, 1976) – does not determine a high degree of social reproduction, as long as parental values are not the only factor influencing the child’s final social position. It is undeniable that social origins do have an impact on the child’s final social position – not only through the transmission of parental values, but also due to financial, cultural, and social capital investments in the child that the origin family can afford (see, for example, Blau et al., 1978; Lareau, 1987; Coleman, 1988). However, there are resources that are not directly connected to social origins that intervene in the social mobility process. For example, the Wisconsin models of status attainment demonstrate the importance of sociopsychological factors (e.g., the child’s mental ability, influences of mentors and significant others, self-conception, or the experience of school failure or success) in mediating the impact of social origins on social destinations (Sewell et al., 1970; Sewell and Hauser, 1980; Sewell et al., 2001). Pearlin and Kohn (1966) believe that this is not exclusively a process through which parents prepare their children for the same type of occupation as theirs. It is more likely that parents, during their occupational careers, come to fully subscribe to the values, attitudes, and behaviors that are deemed acceptable within their occupational group and are determined by characteristics of their occupation. Consequently, parents believe that it is important and beneficial for their children to adopt the same values, attitudes, and behaviors. Furthermore, as Wright and Wright (1976) notice, the degree to which socialization is successful (the extent to which children come to adopt the values their parents deem important) is largely unknown (the empirical results are contradictory). On one hand, several studies have found small correlations between parental values and children’s values (for example, Whitbeck and Gecas, 1988). On the other hand, a study examining the process of intergenerational transmission of values, using data for the United States and Poland (Kohn et al., 1986), reported moderate to high correlations (ranging from 0.37 to 0.59) between parents’ valuation of self-direction and children’s valuation of self-direction.

Socio-demographic characteristics and parental values It is possible that the relationship between social status and parental values is largely explained by a series of parental socio-demographic characteristics. Social status is correlated with socio-demographic variables such as sex, age, family structure, and residential environment. In turn, these variables are correlated with parental values. Given these circumstances, part of 198 corectura 2 28 iulie 2008

Social status and child-rearing values

the association between social status and parental values may be attributable to parents’ socio-demographic profiles. Parental values may differ according to the child’s gender, but also according to the parent’s gender (Kohn, 1969; Alwin and Jackson, 1982; Spade, 1991; Xiao, 2000a). As long as social behavior norms differ according to gender, it is likely that parents will adopt one set of values when raising their daughters and a different set of values when raising their sons. For example, in traditional societies there is a clear separation between the expected behavior for men and the expected behavior for women (Dreyer and Wells, 1966, p.83). In modern societies, these differences in gender specific expected behaviors are less significant, but they persist. Consequently, it is likely that the separation between parental values for daughters and parental values for sons will also persist. Furthermore, each of the two parents may take on different roles in child-rearing, leading to differences between mothers and fathers in childrearing values5. Traditionally, mothers tend to value children’s autonomy more than fathers do (Xiao, 2000a), and while fathers are responsible for control and discipline in parent-child relationships, mothers tend to take on responsibilities related to the child’s emotional support and creative development, although the roles can differ according to cultural context, social class, and the specific parent-child pairing (mother-daughter, mother-son, father-daughter, or fatherson) (Kohn, 1963). Even though the data used for the analyses in this material does not allow for the identification of the child’s sex, the parent’s sex is known, and allows us to test hypotheses regarding differences between women’s and men’s parental values. The length of parental experience (measured by the present age of the first-born child) can also contribute to the explanation of the type of adopted parental values (Duvall, 1946; McNally et al., 1991). On one hand, Duvall (1946) finds that mothers with grown up children tend to value conformity more than young mothers with small children. She attributes this to the fact that as children grow up, their interactions outside the family circle increase in frequency and both children and parents are under increased pressure to conform to social standards. On the other hand, it is possible that new parents will adopt a more authoritarian child-rearing style and the associated values, given that they are confronted with a new situation and might be overly protective of the child. As children grow up and enter their teenage years, it is precisely their increased interactions outside the family circle that limit the parents’ possibilities for close supervision and increase the child’s desire to be independent. Furthermore, parents may change their parental styles and childrearing values according to the child’s cognitive development level, de5

It is difficult to determine the causal directions in the relationships among parental roles, parental styles, and parental values. It is possible that these dimensions influence each other, at least during the initial period of parenthood. 199

The Values of Romanians: 1993-2006. A sociological perspective

emphasizing conformity values as the child’s cognitive development level increases (Roberts et al., 1984; McNally et al., 1991). Besides the fact that the direction of the effect of parental experience on child-rearing values is uncertain from a theoretical point of view, with crosssectional data (such as the data used in this material), the effect of parental age (a variable highly correlated with the length of parental experience) on childrearing values captures not only age effects but also cohort effects (Alwin, Hofer et al., 2004; Alwin, McCammon et al., 2004). Due to long-term changes in family structure and functions, in acceptable social behavior norms, and in mean educational levels (among other factors), older cohorts are likely to be more fervent supporters of conformity values than younger cohorts. Unfortunately, the data used here does not include a measure of the first-born’s age. The parent’s age can be used as a proxy for the length of parental experience. However, in this case, the effects of parental experience, age, and cohort will be combined in a single coefficient. Family structure (operationalized by marital status and number of children) may also have an effect on parental values. First, it is expected that there will be differences between parental values adopted by parents in intact families and parental values in single-parent families, with single parent families focusing more on conformity values (Xiao, 2000a). Second, it is likely that there will be differences in values according to whether the single parent has never been married, is divorced, or widowed (Alwin, 1984), but the small number of cases in these categories does not allow for the estimation of separate effects for each. The number of children may influence child-rearing styles and values, since maintaining order in families with more children can be more problematic, determining a stronger orientation towards conformity values (Xiao, 2000b). Child-rearing values may also vary according to residential area. The differences between urban and rural areas in traditions, opportunity structures, and occupational and life experiences produce not only differences in childrearing values, but also differences in other value orientations. Furthermore, the intensity of the relationship between social status and parental values might be different in the two residential areas.

Mechanisms mediating the relationship between social status and parental values Kohn’s work suggests that one of the most important mechanisms mediating the relationship between social status and parental values is linked to characteristics of occupations (especially the degree of occupational autonomy, operationalized by the degree of complexity, supervision, and routinization of work). A series of preliminary analyses on the Romanian data show that 200 corectura 2 28 iulie 2008

Social status and child-rearing values

occupational autonomy does not have an independent contribution to the explanation of parental values when controlling for social status6. This result replicates Xiao’s (2000a; 2000b) findings for China and the United States. The author attributes the lack of effect of occupational autonomy on child-rearing values to various changes that took place over time in the workplace environment. In the Romanian case, either the lack of effect is due to deficiencies in the occupational autonomy measure, or the effect of occupational autonomy on parental values diminished over time. In either case, occupational autonomy in the current Romanian context is not an important mediating mechanism in the relationship between social status and child-rearing values. Consequently, the empirical analysis will focus on other possible mechanisms that might mediate this relationship. In addition to occupational autonomy, Kohn (1969; Kohn et al., 1990) proposes as mediating mechanisms a series of dimensions linked to psychological functioning (anxiety and fatalism, psychological well-being, intellectual flexibility) and parental orientations to society, work, and self (authoritarian conservatism, morality standards, trust, receptiveness to change, conformity, value placed on extrinsic job characteristics, self-confidence). Due to data constraints, the present material will only explore the second type of mediation mechanisms proposed by Kohn: parental orientations to society, work, and self. The theoretical model employed here (see Figure 1) broadens the scope of this mechanism and includes seven parental orientations in various areas: orientations to self, work, life, society, and religion. These orientations are chosen based on the hypothesis that they are partially determined by the individual’s social status and they indicate different worldviews and different life experiences, which, in turn, have an influence on parental values.

6

The analyses (results are not presented here) explored the effects of the degree of freedom of decision at the workplace on parental values. Although this indicator is different from the ones used by Kohn and it captures a limited facet of occupational autonomy, it may be considered as an indicator measuring occupational autonomy. Although the correlation between this indicator and social status is moderately high and statistically significant at any conventional significance level (r = 0.433), the bivariate correlations between this indicator and the two types of parental values are small and not statistically significant (r= 0.008 for autonomy and r = -0.053 for conformity). In the multivariate analyses, the occupational autonomy indicator does not have an independent contribution to the explanation of parental values and its addition to the model does not diminish the effects of social status on parental values. It is possible that a more detailed measurement model for occupational autonomy, similar to the one used by Kohn, might produce different results. However, the data used here does not contain the necessary indicators. 201

The Values of Romanians: 1993-2006. A sociological perspective Figure 1. Mechanisms explaining the relationship between social status and parental values – theoretical model

It is assumed here that in the causal relationship between general value orientations and the more specific child-rearing values, the former determine the latter. This particular causal order is based on the assumption that general value orientations are formed before individuals reach adulthood (during childhood and youth) and they are relatively stable after that (Inglehart, 1977), while childrearing values become stable only when an adult has become a parent. In this new role, the adult defines his/her child-rearing values so that they are congruent with his/her value orientations in other areas. From this point of view, child-rearing values are not new values that individuals adopt once they become parents, but rather re-evaluations and adjustments of other value orientations to the specific situations arising in the interactions between parents and their children. First, it is expected that the type of personality will have an impact on parental values. Extrovert personalities are compatible with autonomy values and incompatible with conformity values (Roccas et al., 2002). Based on the assumption that, generally, parents consider their own values and own characteristics as desirable for their children as well, extroverted parents will tend to focus on the development of their children’s self-control ability. At the same time, parents with this type of personality will tend to give little importance to conformity values in child-rearing. If there is a significant association between social status and extroverted personality, then the type of 202 corectura 2 28 iulie 2008

Social status and child-rearing values

personality will mediate the relationship between social status and parental values (if the type of personality has an independent effect, over and above that of social status, on parental values). Moving on to orientations to work, the hypothesis that drives the inclusion of these orientations in the theoretical model is that preferences for certain types of jobs are determined by social status and that these preferences can offer additional information about the type of adopted parental values. The explanation for the existence of a relationship between orientations toward work and parental values follows the same reasoning as in the case of the relationship between occupational autonomy and parental values. Individuals in social positions with higher social status are employed in jobs with a higher degree of occupational autonomy, but they also tend to place more importance on occupational autonomy among job characteristics, in comparison to individuals with lower social status (Kohn, 1969). Consequently, job characteristics such as opportunities to use initiative, the feeling that you can achieve something, and the degree of responsibility, are probably more important for those with higher social status (Inkeles, 1960). In addition, because financial aspects are less problematic for people in high social status positions, non-financial aspects of a job become more salient for these categories of individuals (Kohn, 1969). Consequently, these individuals place greater importance both on occupational autonomy, but also on other job characteristics such as the degree to which the job is interesting and the degree to which it meets one’s abilities. Due to the requirements of a job with high non-financial advantages (initiative, responsibility, the pressure to get results and be successful), it is expected that preferences for these types of jobs will be compatible with value orientations focused on autonomy and self-control rather than on conformity. Under these conditions, it is possible that workplace preferences are even more strongly linked to child-rearing values than occupational autonomy is. It is expected that satisfaction with different areas of life is partially determined by socio-economic status, and, in turn, the degree of satisfaction influences the strategies and values adopted in child-rearing. Although subjective evaluations of satisfaction with life and of material and psychological well-being are relatively unrelated to educational levels in some national contexts, they are generally positively associated with absolute and relative positions in the financial hierarchy and with occupational positions (Inkeles, 1960; Fernandez and Kulik, 1981; Veenhoven, 1995; Easterlin, 2001). Veenhoven’s (1995) analyses suggest that in less affluent countries, the relationships between education and income on one hand, and satisfaction on the other hand are stronger than those in affluent countries. Considering this, it is expected that social status is positively correlated with satisfaction in the Romanian context. Independently of the effect of social status, the evaluation of global satisfaction with life can have an effect on parental values, to the extent that the 203

The Values of Romanians: 1993-2006. A sociological perspective

degree of satisfaction with life reflects differences in life experiences and conceptions on life. The theoretical model adopted here assumes that people with positive life experiences and optimistic outlooks on life, characterized by higher levels of satisfaction, will tend to focus more on the development of their children’s self-control abilities and less on children’s conformity. In contrast, a fatalistic outlook on life is probably negatively correlated with social status and the valuation of children’s autonomy. A fatalistic orientation is largely determined by occupational situations individuals encounter during their occupational careers and by the way they were raised in their families of origin (Inkeles, 1960; Kohn, 1969). Thus, one’s social origins and one’s social status are associated with varying perceptions of the world’s degree of complexity and of the degree of control over life (Inkeles, 1960), which, in turn, may influence parental values. Two orientations to society are included in the model: receptiveness to change and morality criteria. Kohn (1969) associates advantaged positions in the social structure and valuation of children’s autonomy with greater receptiveness to change and more pronounced tendencies to obey the spirit of the law rather than the letter of the law. Last, but not least, religiosity intervenes in the relationship between social status and parental values due to its connection with the tendency to obey the dictates of an external authority (Inkeles, 1960). Consequently, religiosity is expected to be negatively associated with social status and the valuation of children’s autonomy, and positively associated with child-rearing values focused on conformity. Tendencies of religious parents to place more value on children’s conformity might also be explained by the congruence of conformity values with religious doctrine (Xiao, 2000a).

Among the seven orientations discussed above, it is likely that the last two (criteria of morality and religiosity) will intervene in the relationship between social status and parental values in a different way from the others (extrovert orientation, importance of non-financial aspects of a job, satisfaction with life, control over one’s own life, and receptiveness to change). While the first five orientations are hypothetically positively related both to social status and to child-rearing autonomy values, but negatively related to child-rearing conformity values, the last two orientations are likely negatively related to social status and autonomy and positively related to conformity values.

204 corectura 2 28 iulie 2008

Social status and child-rearing values

Data and methodology Model and Sample Using the 2005 Romanian WVS data, the empirical model7 studies the relationship between social status and two types of child-rearing values: autonomy and conformity. The main goal of the analysis is to explain the association between social status and parental values. Two types of explanatory mechanisms are proposed here: the socio-demographic profile and respondents’ orientations to various domains. The socio-demographic profile acts as a control mechanism in the model, while the respondents’ orientations act as mediating mechanisms. I employ the distinction between control mechanisms and mediating mechanisms in order to differentiate between mechanisms involving exogenous variables and mechanisms involving endogenous variables. The socio-demographic variables are exogenous variables in this model – they are associated with social status, but not causally related to it. The term “mediation” is reserved for the case in which a variable intervenes in the causal relationship between two other variables – it is caused by an independent variable, and causes in turn another dependent variable (Baron and Kenny, 1986). Using this terminology, respondents’ orientations are considered as mediating mechanisms. Both types of mechanisms serve the same purpose – that of explaining the relationship between social status and child-rearing values. The secondary goal of the analysis is to explore differences between respondents living in urban and rural residential areas, both in the degree of association between social status and parental values and in the degree to which this relationship is explained by the control and mediating mechanisms. Preliminary analyses (results are not presented here), suggest that both sex and residential area may interact with the other variables in the model. The sample size does not allow for the estimation of the model in the presence of interactions of the other variables both with sex and with residential area (a four-group model on urban men, urban women, rural men, and rural women). Since the interactions with residential area are stronger than those with sex, I have chosen to estimate a model that only takes into account the interactions of the other variables with residential area. In order to explore these differences, 7

The models are estimated using FIML (Full Information Maximum Likelihood). This method estimates model coefficients in the presence of incomplete data. The estimation algorithm uses the information from the observed portions of the data in the presence of an unlimited number of missing data patterns, and the estimated parameters include information about the mean and variance of missing portions of a variable, given the known information from the observed portions of other variables (Wothke, 2000). FIML produces efficient and unbiased estimators when the data are missing at random, and several simulations have demonstrated that the algorithm performs well even with non-ignorable missing data (Wothke, 2000). 205

The Values of Romanians: 1993-2006. A sociological perspective

the model is simultaneously estimated for the two groups defined by residential area8. The structure of the final model is based on the theoretical model diagram (see Figure 1), and includes social status and the socio-demographic variables as intercorrelated, exogenous variables. All the exogenous variables influence the variables measuring the respondent’s orientations. No causal structure is imposed among the variables measuring orientations. Instead, in order to take into account the associations existing among these variables that are not explained by the influences accounted for by this model, the structural errors associated with these variables are intercorrelated (Maruyama, 1998, p. 188)9. Both the exogenous variables and the mediating variables cause the final two dependent variables of interest: valuation of autonomy and valuation of conformity in child-rearing. In a similar way, the part of the association between these variables, unaccounted for by the other variables included in the model, is represented in a non-causal way, through the correlation between the structural errors associated with the valuation of autonomy and the valuation of conformity10. In order to explore both the individual and the combined impact of the two explanatory mechanisms on the relationship between social status and parental values, a sequence of models is presented. The first model estimates the effect of social status on the two types of parental values, without accounting for any other influences. The second model estimates the effect of social status on parental values in the presence of controls for socio-demographic variables. The third model estimates the same effect in the presence of mediating variables, and the final model includes both the control mechanisms and the mediating mechanisms in the estimation of this effect. It is expected that the effect of social status on parental values (in absolute value) will diminish once control mechanisms and mediating mechanisms are taken into account. 8

The models are estimated using AMOS 7.0. The simultaneous group analysis implies the estimation of a single model for both groups, resulting in a unique set of parameters (covariances, variances, regression coefficients, and R2 coefficients) for each group. Estimating a simultaneous group model instead of a separate model for each group, results in more efficient parameter estimates and allows for statistical tests of parameter differences between groups (Arbuckle, 2006). 9 Although the causal ordering of these variables might be theoretically determined (for example the personality type influences preferences for certain job characteristics), in this model, where these variables only play a mediating role, the causal ordering does not offer additional information on the relationship between social status and child-rearing values. 10 In this case, the modeling choice reflects the idea that the two types of values do not influence each other, but that they covary. For example, because of accumulated life experiences, a person adopts parental values that focus on conformity and simultaneously rejects parental values that focus on autonomy. Alternatively, a person may adopt a combination of the two types of values, although it is likely that a person that scores high on the autonomy dimension will have low scores on the conformity dimension. 206 corectura 2 28 iulie 2008

Social status and child-rearing values

Furthermore, in order to explore the effects of socio-demographic variables and the effects of orientations on child-rearing values in more detail, two additional models are estimated. These two models do not account for effects of social status, in order to diminish collinearity effects. The fact that social status has moderate to high correlations both with the socio-demographic variables and with the mediating variables creates problems with the partitioning of common variance in the dependent variables between predictors (Baron and Kenny, 1986, p. 1177; Maruyama, 1998, p. 62). Consequently, the first of these models only includes effects of socio-demographic variables on parental values, and the second of these models only includes effects of orientations on parental values. Although all respondents in the sample, regardless of their parental status, answered the questions referring to child-rearing values, the analysis presented here will focus on the subsample of parents, because the issue of adopted parental values is more relevant for this subsample. At the same time, it is likely that parental values are more crystallized within this group.

The measurement of parental values The final endogenous variables in the analysis (valuation of autonomy and valuation of conformity in child-rearing) were constructed as scales, using items that have been associated with these dimensions in the child-rearing values literature: independence, imagination, and perseverance as indicators of autonomy and hard work, obedience, religious faith, and thrift as indicators of conformity (Duvall, 1946; Lenski, 1961; Kohn, 1969; Xiao, 2000a, 2000b; Hagenaars et al., 2003). The respondents were shown a list of 10 qualities which children could be encouraged to learn at home, and they were asked to choose up to five qualities they consider especially important. The chosen qualities were coded 1, and the qualities that were not mentioned were coded 0. For the analyses presented here, the respondents that did not provide an answer, and those that chose more than five qualities were excluded from the analysis (39 cases from the subsample of parents). Choosing more than five characteristics is considered invalid, since it disregards research instructions. Furthermore, this situation suggests that the respondent found it hard to differentiate among the ten proposed values (Xiao, 2000a). Cluster analysis11 was employed in order to decide whether it is more appropriate to operationalize child-rearing values in terms of a single dimension or two dimensions, and in order to check whether the empirical grouping of parental value items is compatible with theoretical expectations.

11

Because the items are binary, I have used a type of cluster analysis which takes this into account: hierarchical cluster for binary variables, with clustering on variables, using the between-groups linkage clustering method and the Phi 4-point dissimilarity measure. 207

The Values of Romanians: 1993-2006. A sociological perspective

The results of this analysis are presented separately for urban and rural residential areas in Figure 2. The theoretical expectations regarding the grouping of items measuring child-rearing values are confirmed in both residential areas: independence, imagination, and perseverance are grouped into one cluster, and obedience, thrift, religious faith, and hard work are grouped into another cluster. In order to construct the parental values scales12, I used the arithmetic mean of the items belonging to a cluster. Due to the fact that each of the items can take a value of 0 (the respondent did not mention the quality) or 1 (the respondent mentioned the quality), the two resulting parental values scales have the same amplitude and range from 0 (none of the qualities on that parental values dimension was mentioned) to 1 (all of the qualities on that parental values dimension were mentioned). The choice to measure the valuation of autonomy as a separate dimension from the valuation of conformity is preferred here also because this distinction allows for a more detailed description of the relationships between social status and child-rearing values. Otherwise, parental preferences for children’s autonomy or conformity are essentially the extreme opposites along a single dimension of child-rearing values13. It is possible, however, that the impact of social status on one type of parental values is different from the impact on the other type of parental values. Figure 2. The clustering of items measuring child-rearing values URBAN

0

5

10

15

20

25 RURAL

Thrift

Independence

Relig.faith

Imagination

Hard work

Perseverance

Obedience

Hard work

Independence

Obedience

Perseverance

Relig.faith

Imagination

Thrift

12

0

5

10

15

20

25

As before, because of the binary nature of the items, I opted for scales instead of using a latent variable in this case, given the fact that the indicators do not conform to the multivariate normality assumption imposed by the estimation method used here (maximum likelihood) (Arbuckle, 2006, p. 39). 13 The correlations (in absolute values) between the two measures for the two types of parental values on one hand, and a unique measure of parental values (computed by subtracting the number of qualities indicating a preference for conformity from the number of qualities indicating a preference for autonomy) are very strong (approximately 0.86). 208 corectura 2 28 iulie 2008

Social status and child-rearing values

Determinants of parental values Social status was constructed as a latent variable14 with three observed indicators: education, occupation, and income. Education is measured by categories indicating progressively higher levels of schooling, ranging from 1 (no schooling) to 14 (M.A. degree or Ph.D. degree) 15. Occupation is measured using a prestige scale (Standard International Occupational Prestige Scale SIOPS) developed by Treiman (1977) and updated by Ganzeboom and Treiman (2003)16. On this scale, high scores indicate high levels of prestige. In order to capture the relative positions of individuals in the financial resources hierarchy, incomes are measured in deciles of household incomes per capita. The control variables are all observed variables, and include the respondent’s age, marital status (single versus the reference category: married), number of children, and sex (man versus the reference category: woman). Mediating variables are either observed variables (control over one’s own life, openness to scientific and technological advances, importance of God), latent variables (extrovert orientation, satisfaction, morality/legality criteria), and scales (valuation of non-financial aspects of a job). Control over one’s own life is measured by the reactions, on a scale from 1 (everything is determined by fate) to 10 (people shape their fate themselves) to the statement: “Some people believe that individuals can decide their own destiny, while others think that it is impossible to escape a predetermined fate”. Openness to technological and scientific advances is measured by the agreement, on a scale from 1 (completely disagree) to 10 (completely agree), with the statement: “Because of science and technology, 14

For each of the latent variables used in the model, an exploratory factor analysis was carried out as a preliminary step to the analysis (results are not presented here). Based on these results, confirmatory factor models were constructed for each of the latent variables (see details on model goodness of fit for these confirmatory factor analysis models in Table A-1). 15 This variable and all other ordinal level variables in these analyses are treated as interval level variables. 16 In order to transform the variable measuring occupation from its original version containing partially ordered categories into an interval level variable it is necessary to recode it. The recoding scheme is based on recoding tables developed by Ganzeboom and Treiman (2003). Groups with no occupation and “other” occupations are treated differently, according to which category they belong to: students, housewives, and unemployed respondents are allocated prestige code 0, while retired respondents and those with “other” occupations (for whom no details were recorded regarding their last occupation, respectively their present occupation) are treated as incomplete data, to be taken into account by the FIML estimation procedure. I opted for treating the last two categories of respondents in this way in order to avoid excluding them from the analysis. The assumption here is that each of the respondents in these two categories has an associated occupational prestige code (be it one derived from a past occupation or one derived from a present occupation). 209

The Values of Romanians: 1993-2006. A sociological perspective

there will be more opportunities for the next generation”. In order to measure religiosity, I have opted for the use of a single indicator, linked to one of the dimensions of religiosity (religious faith) 17: importance of God in respondents’ lives, ranging from 1 (not at all important) to 10 (very important). Extrovert orientation is a latent dimension, operationalized by answers to items from the portrait-values battery: “To what extent are you similar to a person for which it is important18: (a) to have a good time; to spoil oneself, (b) to be rich; to have a lot of money and expensive things, (c) to have adventures and to take risks; to have an exciting life, (d) to be successful; to have people recognize one’s achievements”. This dimension captures two types of values with congruent motivations19: selfenhancement and openness to change (Schwartz et al., 2001). Both types of values are associated with extroverted personalities (Roccas et al., 2002). Satisfaction is also a latent dimension, measured by items of satisfaction with various domains: satisfaction with one’s life as a whole (the scale ranges from 1 completely dissatisfied to 10 completely satisfied), satisfaction with the state of one’s own health, with the amount of money one has, and with one’s way of living20 (the scales for these items range from 1 not at all satisfied to 4 very satisfied), and a global happiness measure (“Taking all things together, would you say you are very happy (4), rather happy (3), not very happy (2), or not at all happy (1)”). The latent variable measuring criteria of morality/legality is constructed using items grading (on a scale from 1 always justifiable to 10 never justifiable) several behaviors that violate morality and legal standards: claiming government benefits to which you are not entitled, someone accepting a bribe in the course of their duties, cheating on taxes if you have a chance, avoiding a fare on public transport. The preference for non-financial aspects of a job is constructed as an additive scale, based on a set of dichotomous items. Respondents were asked to 17

Preliminary analyses explored the effects of an additional indicator of religious practice (churchgoing frequency). However, the indicator does not bring additional contributions to the explanation of child-rearing values when controlling for the religious faith measure. 18 This is a retranslation of the item as it was included in the Romanian questionnaire. The original WVS item stated: “Now I will briefly describe some people. Please indicate for each description whether that person is very much like you, like you, somewhat like you, not like you, or not at all like you”. 19 According to Schwartz et al. (2001), the 10 items included in this battery of questions describe a circular value space, in which adjacent values are compatible. As the distance on the circle’s diameter increases, the values become incompatible. In the circular structure described by the 10 values included in the portrait-values battery, the four items chosen here to operationalize extrovert orientation have adjacent positions. 20 Satisfaction with the state of one’s own health, with the amount of money one has, and with one’s way of living are country-specific items, not included in the original WVS questionnaire. 210 corectura 2 28 iulie 2008

Social status and child-rearing values

evaluate the importance (0 not important, 1 important) of the following job characteristics: a job that is interesting, an opportunity to use initiative, a job in which you feel you can achieve something, a responsible job, and a job that meets one’s abilities21. In the newly created scale, each respondent received a score equal to the number of non-financial aspects of a job that he/she thought were important.

Results The examination of the descriptive statistics for the subsample under study, separately for urban (N=762) and rural (N=596) residential areas, suggest that respondents living in urban areas are characterized by higher mean scores on all three indicators of social status (the results are not presented here). Respondents living in urban areas are also younger and tend to have fewer children than respondents in rural areas. For the mediating variables, with the exception of morality and religiosity indicators (which theoretically were expected to display a different behavior from the remaining mediating variables), urban residents have consistently higher mean scores than rural residents. Generally, among the indicators used to measure the final dependent variables, items measuring the valuation of autonomy in childrearing are preferred by higher percentages of parents living in urban areas, while items measuring the valuation of conformity in child-rearing are preferred by higher percentages of people living in rural areas (with the exception of ‘thrift’). In general, social status is associated with all of the control variables and all of the mediating variables – correlation coefficients are statistically significant, except for the correlation with morality criteria in urban areas (the results are not presented here). Older parents, single parents, and those with more children tend to have a lower social status in all three samples (total, urban, and rural). Men tend to have higher social statuses than women do, and in the entire sample of parents, social status scores are higher in urban areas compared to rural areas. Parents with a higher social status tend to have extroverted personalities, and to prefer jobs characterized by non-financial advantages. They also have higher levels of satisfaction. Finally, they have the feeling that they can control their own lives, and they are more receptive to changes brought by technological and scientific advances, these results offering support to theoretical expectations. Parents with lower social status tend to be more religious and more inclined to reject amoral and illegal forms of behavior. As expected, social status is positively associated with the valuation of autonomy in child-rearing and negatively associated with the valuation of conformity in child-rearing, and these associations are statistically significant. The 21

These items have been included in the 2005 Romanian WVS, but they are from a battery of items from the European Values Survey. 211

The Values of Romanians: 1993-2006. A sociological perspective

correlations are low to moderate (-0.159 and -0.306 for relationships with the valuation of conformity and 0.273 and 0.315 for relationships with the valuation of autonomy). The association between social status and the valuation of autonomy is relatively stronger in urban areas (the covariance is 0.191, compared with 0.139 for the rural areas). The differences between urban and rural residential areas are more pronounced in the case of the association between social status and valuation of conformity (the covariance is -0.175 for urban areas and -0.069 for rural areas). In general, associations between valuation of autonomy in child-rearing on one hand and socio-demographic and mediating variables on the other hand, go in opposite directions in comparison to associations between valuation of conformity in child-rearing and the socio-demographic and mediating variables. This suggests that the socio-demographic and orientation profiles of parents who adopt the two types of values are polar opposites. The associations between the valuation of autonomy in children and socio-demographic and orientation variables have the same signs as the associations between social status and socio-demographic and orientation variables. In the case of the valuation of conformity in children, the directions are reversed: associations between the valuation of conformity and the other variables go in an opposite direction from associations between social status and the other variables. The two parental values dimensions are significantly correlated, with moderate to high negative correlations (approximately -0.450). This suggests that the two types of values are dissonant, but that, at the same time, the parental value space can include a combination of the two types of values. The directions of associations suggest that each of the proposed sociodemographic and orientation variables can act as explanatory mechanisms of the relationship between social status and parental values. Parents with higher social status tend to value autonomy and reject conformity in children, but, at the same time, each social status position is associated with a certain sociodemographic and orientation profile, a profile that, in turn, may influence the type of adopted parental values. The questions that motivate the analyses that follow are: (a) whether this profile has an independent effect on child-rearing values, over and above the effect of social status, and (b) to what degree the social status effect on parental values can be explained by taking into account the characteristics of these profiles. In order to answer the first question, the discussion of results will focus on the statistical significance of the effects of control variables and orientation variables on parental values when controlling for social status. In order to answer the second question, the discussion will focus on comparing social status effects on parental values in models that do not control for socio-demographic and orientation profiles effects and in models that include these controls. Table 1 through Table 4 present the main interest effects in multivariate models, and contain both unstandardized and standardized coefficients, in order to facilitate both comparisons between the two residential groups and an exploration of the intensity of effects.

212 corectura 2 28 iulie 2008

0.08

b p ß 0.030 *** ( 0.283)

Model 1 p

Model 2

0.07

ß

-0.003 *** (-0.152) -0.040 (-0.059) -0.033 *** (-0.141) 0.047 * ( 0.085)

b

ß

Model 4

Model 6

Social status and child-rearing values

Model 5

-0.009 0.005

(-0.018) -0.016 ( 0.054) 0.005

(-0.005)

(-0.031) ( 0.048)

p

Model 3

ß ( 0.166) (-0.111) (-0.044) (-0.053) ( 0.026) ( 0.073)

b p ß 0.021 *** ( 0.215) -0.002 ** (-0.114) -0.035 (-0.051) -0.015 (-0.063) 0.025 ( 0.046)

0.028 * ( 0.110)

0.017 ** ( 0.106) ( 0.084) ( 0.064)

(-0.006) -0.001

b p ß b p 0.023 ** ( 0.199) 0.018 * -0.002 ** -0.030 -0.012 0.014 0.019 0.033 ** ( 0.126)

0.017 ** ( 0.105) 0.042 + 0.006 +

-0.001

0.007 ( 0.038) 0.008 ( 0.040) -0.010 + (-0.063) -0.008 (-0.052) 0.11 0.13

(-0.002) 0.005 ( 0.026) -0.013 * (-0.087) 0.08

0.000

0.11

0.023 *** ( 0.143)

b

Table 1. Regression coefficients in the prediction of the valuation of autonomy in child-rearing for urban residents

Social status Age Single Number of children Man Extrovert orientation Non-financial job aspects Satisfaction Control over life Openness to S&T advances Criteria of morality Importance of God R2

Notes: *** p < 0.001, ** p < 0.01, * p < 0.05, + p < 0.1

213

The Values of Romanians: 1993-2006. A sociological perspective

Model 1 b p ß 0.030 *** ( 0.273)

0.07

Model 2 p

0.03

ß

-0.002 ** (-0.121) -0.008 (-0.012) -0.015 * (-0.094) 0.008 ( 0.017)

b

Model 3 p

ß

ß ( 0.224) (-0.036) ( 0.010) (-0.045) (-0.024)

Model 4 b p 0.022 *** -0.001 0.007 -0.007 -0.011

Model 5

Model 6

b p ß b p ß 0.023 ** ( 0.197) 0.021 * ( 0.194) 0.000 (-0.026) 0.023 ( 0.037) -0.003 (-0.020) -0.028 (-0.059) 0.022 + ( 0.100)

0.020 0.003

0.008

( 0.020)

( 0.045) ( 0.037)

( 0.062)

0.002

0.026 0.004

0.008

( 0.019)

( 0.058) ( 0.051)

( 0.061)

0.021 + ( 0.097)

0.002

( 0.066) 0.069 ** ( 0.155) 0.003 ( 0.033)

0.009

( 0.043)

(-0.036) -0.005 (-0.030) (-0.010) -0.004 (-0.020) 0.11 0.11

0.005

-0.006 -0.002

(-0.044) (-0.045) 0.08

0.07

-0.007 -0.009

0.023 + ( 0.102)

b

Table 2. Regression coefficients in the prediction of the valuation of autonomy in child-rearing for rural residents

Social status Age Single Number of children Man Extrovert orientation Non-financial job aspects Satisfaction Control over life Openness to S&T advances Criteria of morality Importance of God R2

Note: *** p < 0.001, ** p < 0.01, * p < 0.05, + p < 0.1

214 corectura 2 28 iulie 2008

Social status and child-rearing values

(-0.074)

0.003 0.013

-0.011

(-0.029) -0.003

( 0.018) ( 0.030)

0.003 0.013

(-0.048) -0.004

0.003 0.003

(-0.009)

(-0.032)

( 0.020) ( 0.031)

(-0.016)

( 0.012) ( 0.007)

Model 3 p ß

-0.016

(-0.037) (-0.116)

-0.002

(-0.005) -0.001

b

-0.005 -0.050 *

(-0.045)

-0.001

( 0.040) (-0.031)

-0.004

(-0.012)

0.008 -0.014

-0.001

(-0.074)

0.013 ** ( 0.102) 0.12

(-0.066) -0.012 +

0.013 ** ( 0.100) 0.11

-0.011 0.10

(-0.048) 0.018 *** ( 0.136) 0.06

-0.008

Model 4 Model 5 Model 6 b p ß b p ß b p ß -0.022 *** (-0.263) -0.028 *** (-0.288) -0.026 *** (-0.280) 0.001 * ( 0.083) 0.002 * ( 0.101) -0.014 (-0.024) -0.020 (-0.033)

Table 3. Regression coefficients in the prediction of the valuation of conformity in child-rearing for urban residents Model 1 Model 2 b p ß b p ß Social status -0.028 *** (-0.306) Age 0.002 *** ( 0.130) Single -0.008 (-0.014) Number of 0.027 *** ( 0.135) children Man -0.037 * (-0.080) Extrovert orientation Non-financial job aspects Satisfaction Control over life Openness to S&T advances Criteria of morality Importance of God R2 0.09 0.05 Note: *** p < 0.001, ** p < 0.01, * p < 0.05, + p < 0.1

215

The Values of Romanians: 1993-2006. A sociological perspective

Model 3 p ß

( 0.089)

b

( 0.078)

( 0.060) (-0.040) (-0.028)

0.017

0.015

( 0.056) 0.007 (-0.049) -0.015 (-0.010) -0.002

( 0.074)

0.014

0.006 -0.019 -0.001

-0.009 * (-0.096) -0.009 * (-0.098) 0.010 ( 0.071) 0.010 ( 0.067) 0.007 ( 0.041) 0.006 ( 0.037) 0.04 0.06

0.006 ( 0.053) -0.045 * (-0.118) 0.000 (-0.007) -0.010 * (-0.111) 0.011 ( 0.076) 0.011 ( 0.063) 0.04

0.04

Model 4 Model 5 Model 6 b p ß b p ß b p ß -0.008 + (-0.095) -0.012 (-0.123) -0.007 (-0.071) 0.001 + ( 0.081) 0.001 + ( 0.088) -0.008 (-0.016) -0.005 (-0.010) 0.012 * ( 0.086) 0.013 * ( 0.091) 0.002 ( 0.006) 0.002 ( 0.005)

Table 4. Regression coefficients in the prediction of the valuation of conformity in child-rearing for rural residents Model 1 Model 2 b p ß b p ß Social status -0.015 ** (-0.159) Age 0.001 ** ( 0.118) Single -0.003 (-0.006) Number of children 0.015 * ( 0.107) Man -0.005 (-0.011) Extrovert orientation Non-financial job aspects Satisfaction Control over life Openness to S&T advances Criteria of morality Importance of God R2 0.03 0.03 Note: *** p < 0.001, ** p < 0.01, * p < 0.05, + p < 0.1

216 corectura 2 28 iulie 2008

Social status and child-rearing values

The first model estimates the total association between social status and child-rearing values. The results of this model are similar to the previously discussed results regarding the bivariate relationships between variables, except that model 1 simultaneously estimates the influence of social status on both dimensions of parental values. In addition, model 1 provides values of the unstandardized regression coefficients (b coefficients). The model is estimated only in order to obtain estimates of regression coefficients that are comparable with results from the other multivariate models. Model 2 estimates effects of control variables on parental values without accounting for the effect of social status. Controlling for the other socio-demographic variables’ influences, age and the number of children have statistically significant influences on parental values, both in the urban and in the rural area (negative effects on the valuation of autonomy and positive effects on the valuation of conformity in child-rearing). Furthermore, among urban residents, sex is also a statistically significant predictor of parental values (men, compared to women, are more inclined to value children’s autonomy and less inclined to value children’s conformity). The result contradicts the theoretical expectations based on results from other national contexts. It is possible that past social norms of acceptable behavior, during the period in which the parents in this sample were developing their value spaces, but also more recent social norms of behavior, during the period when parents in this sample try to develop the values of their children, place a greater importance on conformity for girls. Given that the effect of sex disappears when social status is controlled for (see model 4), it is also possible that differences in parental values between men and women are due to differences in socio-economic status between the two groups. Marital status is not a statistically significant predictor of parental values, neither among urban residents, nor among rural residents, when the other sociodemographic characteristics are controlled for (see Table 1 through Table 4). Model 3 presents the effects of parents’ orientations on child-rearing values, without accounting for social status influences. In the prediction of the valuation of autonomy, among the seven orientation variables included here, two (openness to scientific and technological advances and criteria of morality) do not have statistically significant effects when the other orientations are controlled for (see Table 1). In contrast to effects among urban residents, in rural areas, most of the orientation variables have small and statistically nonsignificant effects on the valuation of autonomy in child-rearing (among the orientation variables, only satisfaction and extrovert orientation have statistically significant effects – see Table 2). Similarly, in the prediction of the valuation of conformity, the majority of the effects of orientation variables are small and statistically non-significant, this time both among urban and rural residents (see Table 3 and Table 4). The only exceptions are the effects of satisfaction and of the importance of God among urban residents, and the effects of satisfaction and openness to scientific and technological advances 217

The Values of Romanians: 1993-2006. A sociological perspective

among rural residents. These results suggest that although in the bivariate approach all of the parent’s orientations are statistically significantly associated with parental values, there is some overlap among their effects on parental values in the multivariate approach. Only in urban areas and only in the prediction of the valuation of autonomy in children, all of the orientation variables that were statistically significantly associated with that parental value dimension maintain their statistical significance in the multivariate analysis. Models 4 and 5 estimate the effects of social status on parental values, in the presence of control variables, respectively orientation variables. In these models, due to the collinearity of predictors, a literal interpretation of the coefficients’ statistical significances is problematic. In general, the effects of control variables and orientation variables diminish and become statistically non-significant in comparison to the results of the previous two models discussed above. Age remains a consistent and statistically significant predictor of parental values (with the exception of its relationship with the valuation of autonomy in rural areas). Among rural residents, the number of children continues to be a statistically significant predictor of the valuation of conformity in child-rearing. In model 5, with several exceptions, orientation variables remain consistently linked to the two parental values dimensions. However, only a few of these effects retain their statistical significance: for the prediction of autonomy child-rearing values, among urban residents – extrovert orientation, the valuation of non-financial aspects of jobs, and the importance of God, and among rural residents – extrovert orientation; for the prediction of conformity child-rearing values, among urban residents – the importance of God, and among rural residents – openness to scientific and technological advances (see Table 1 through Table 4). However, the main goal of these models is to estimate the impact of social status on parental values. Compared to the total association estimated in model 1, controlling for the socio-demographic profile reduces the impact of social status on valuation of autonomy by 30% among urban residents and by 27% among rural residents, and on valuation of conformity by 21% among urban residents and by 47% among rural residents (percentages based on unstandardized coefficients). Controlling for the effects of orientation variables also produces a decrease in the impact of social status on parental values in all cases, except for the relationship with the valuation of conformity in urban areas (in this case, probably due to a higher degree of collinearity among predictors, some of the regression coefficients bounce through 0 and this influences the results). The decrease in the impact of social status on autonomy is of 23%, both in the urban and the rural areas, and the decrease in the impact on conformity in rural areas is of 20%. The final model estimates the impact of social status on parental values in the presence of the combined effects of control mechanisms and mediation mechanisms. Together, the two types of variables explain 40% of the effect of 218 corectura 2 28 iulie 2008

Social status and child-rearing values

social status on valuation of autonomy among urban residents, and 30% of the effect among rural residents. The decrease in the effect of social status on the valuation of conformity in urban areas is modest (7%), however this result is once again influenced by some coefficients bouncing through 0 (especially the coefficient for criteria of morality, recording a marginally statistically significant effect in the opposite direction from the expected direction). In rural areas, the effect of social status on conformity is reduced by 53% (see Tables 1 through 4). These results suggest that both the socio-demographic profile variables and the orientation variables contribute to the explanation of the relationship between social status and parental values. The two profiles combined (sociodemographic profile and orientation profile) explain a greater percentage of the relationship between social status and parental values than each profile in part, suggesting that each profile has individual contributions to the explanation of the relationship, although there is some degree of overlap. In all cases analyzed here, the impact of social status on child-rearing values is reduced when controlling for the variables that have been chosen as explanatory mechanisms and, in the case of the relationship between social status and conformity in rural areas, the effect becomes statistically non-significant. In the final model, among the control variables, age continues to have a statistically significant effect, independent of the effect of social status on parental values (with the exception of the prediction of the valuation of autonomy in rural areas). Additionally, the number of children remains a marginally significant predictor of the valuation of conformity among rural residents. Among the proposed mediating mechanisms, only a few have statistically significant relationships both with social status (see Table A-3) and with parental values in the model that controls for the complete set of influences (see Table1 through Table 4): in the prediction of the valuation of autonomy, in urban areas – the valuation of non-financial aspects of a job, and in rural areas – extrovert orientation; and in the prediction of the valuation of conformity, in urban areas – the importance of God1, and in rural areas – openness to scientific and technological advances. These variables on which social status has statistically significant effects and which, in turn, have statistically significant effects on parental values constitute the main pathways that mediate the effect of social status on parental values.

1

Additionally, in this case, criteria of morality are statistically significantly related both to social status and to the valuation of conformity. This mediating variable was excluded from the enumeration because the sign of its effect on the valuation of conformity is reversed in comparison to the theoretically expected sign. Furthermore, the bivariate correlation between criteria of morality and the valuation of conformity is not statistically significant. 219

The Values of Romanians: 1993-2006. A sociological perspective

In order to explore differences between urban and rural areas in the processes that control and mediate the relationship between social status and child-rearing values, I also estimated tests for the equality of structural relationships between urban and rural areas for the initial model and the final model (see Table A-5). These tests show that in the final model, there are statistically significant differences between urban and rural residents in the relationships among the exogenous variables (the associations between social status and control variables) and in the effects of control variables on the mediating variables. There are no statistically significant differences between urban and rural residents in the direct effect of social status on the valuation of autonomy. However, the social status effect on the valuation of conformity is significantly higher (in absolute value) in urban areas than in rural areas (see Table 3, Table 4, and Table A-5). This is mainly because in urban areas the total association from the initial model between social status and the valuation of conformity in children is significantly higher than in rural areas (see Table 3, Table 4, and Table A-5). A second reason is that the control and mediating mechanisms have a diminished explanatory power in the case of urban residents when compared to rural residents. Conclusions Using a theoretical model proposed by Kohn and his colleagues (Kohn, 1963; Pearlin and Kohn, 1966; Kohn, 1969; Slomczynski et al., 1981; Kohn et al., 1983; Kohn et al., 1986; Kohn et al., 1990) and the 2005 Romanian WVS data, I have examined the relationship between parents’ social status and their valuation of autonomy or conformity in child-rearing. The two types of parental values distinguish between two different child-rearing strategies. The valuation of autonomy in children denotes an increased focus on the construction of an internal set of behavior guidelines, while the valuation of conformity in children denotes an increased focus on obeying externally imposed rules. Although the two child-rearing value orientations are largely discordant, this does not mean that parents cannot adopt a combination of the two parental value types. In general, however, the higher the valuation of autonomy in children, the lower the valuation of conformity. The analyses presented here show that there is indeed a significant relationship between social status and parental values: in comparison to parents in lower social status positions, those with higher social status tend to focus more on developing their children’s autonomy and they tend to de-emphasize conformity values in child-rearing. The intensity of these associations and the analysis results regarding the relationships among social status, parental values, and socio-demographic and orientation profiles reiterate Kohn’s (1969) observation that the relationships between social status and parental values are not impressive in regard to their intensity, but because of their consistency. In 220 corectura 2 28 iulie 2008

Social status and child-rearing values

the analysis presented in this chapter, the correlations between social status and parental values are small to moderate (approximately ranging between 0.16 and 0.30 in absolute values), but the relationships between these two variables, and the relationships of these two variables with other variables describing the socio-demographic and orientation profiles are consistent with the theoretical expectations. The existence of a relationship between social status and parental values, even if this relationship is small to moderate in intensity, raises the problem of the intergenerational inheritance of social advantages and disadvantages. If parents instill in their children values that characterize their own social status position, then the process of socialization is tantamount to preparing the child for the same social status and the same occupational position as the parents’ position. Under these circumstances, it is important to understand which are the mechanisms that allow social status to influence parental values. In this chapter, I have taken into consideration two types of mechanisms that could explain part of the relationship existing between social status and parental values: socio-demographic profiles and orientation profiles. First, a part of the total association between social status and parental values may be due to social status differences between groups that are characterized by different socio-demographic profiles and to the relationships that may exist between the socio-demographic profile and parental values. The results of this analysis suggest that men, urban residents, young people, married persons, and those with fewer children tend to have a higher social status, and at the same time, they tend to value autonomy and reject conformity in children. In the multivariate models presented in this chapter, these socio-demographic characteristics explain between 20% and 40% of the total association between social status and parental values. Secondly, another part of the total association between social status and parental values may be due to the indirect effects of social status on parental values. The model in this chapter examined the role of parental orientations to different domains as mechanisms that mediate the relationship between social status and parental values. The orientations that were examined here ranged from domains that are more concrete to domains that are more abstract: orientations to self, work, life, society, and religion. These are some of the mechanisms that can intervene in the relationship between social status and parental values. However, other important mechanisms could not be examined here. For example, the degree of occupational autonomy and mechanisms linked to intellectual flexibility and psychological well-being may act as mediators of the relationship between social status and parental values, intervening in the model that was proposed here between social status and the orientation variables. Among the examined mediating variables, the analysis results suggest that social status is positively associated with extrovert orientation, the 221

The Values of Romanians: 1993-2006. A sociological perspective

valuation of non-financial job advantages, the degree of satisfaction and control over life, and openness to scientific and technological advances, and negatively associated with intolerance towards amoral or illegal behavior and with religiosity. The direction of the association between social status and the examined orientations is paralleled in the associations between parents’ orientations and the valuation of autonomy in children, and is reversed in the associations between parents’ orientations and the valuation of conformity in children. In the multivariate models, the orientation variables explain approximately 20% of the total association between social status and parental values, with the exception of the prediction of the valuation of conformity among urban residents, where the mediation effects act in opposite directions and cancel each other out. Together, the control mechanisms and the mediating mechanisms explain one third to a half of the effect of social status on parental values. Once again, in this case, the relationship between social status and the valuation of conformity in urban areas is an exception, given the fact that some of the indirect effects cancel each other out. Among the socio-demographic variables, age is the most important control mechanism. Either because older parents have had more experience (both life and parental experience), or because effects are due to differences between cohorts, this category of parents tends to place more value on conformity in children and to de-emphasize autonomy values. An explanation that conceptualizes age as duration of life may be that as people grow older, the accumulation of life experiences determines successive revisions of the hierarchy of values, with the focus moving from the valuation of autonomy toward the valuation of conformity. An explanation based on the length of parental experience may be that advanced by Duvall (1946), according to which, as children grow older and start to interact with people from outside the family circle, parents come under increased pressures to conform to social standards. Explanations based on cohort effects may invoke differences in the political, economic, social, and cultural contexts marking the life experiences of each cohort. For example, for older cohorts in this sample, a larger part of their socialization was spent under the communist regime, a context in which conformity was extremely important. Orientation variables also contribute to the explanation of the relationship between social status and parental values. In each of the four situations analyzed here (two types of parental values in two residential areas), a different orientation emerges as the most important mediating factor. Among the mediators in the relationship between social status and the valuation of autonomy, the orientations to domains that are more concrete are more important (extrovert orientation among rural residents and the valuation of nonfinancial job aspects among urban residents). In contrast, among the mediators in the relationship between social status and the valuation of conformity, the orientations to domains that are more abstract are more important (openness to 222 corectura 2 28 iulie 2008

Social status and child-rearing values

scientific and technological advances among rural residents and religiosity among urban residents). Under these conditions, it is possible that the valuation of autonomy is mainly determined by a person’s orientations to self and the immediate surrounding environment, while the valuation of conformity is mainly determined by a person’s orientations to domains that transcend the self and the immediate surrounding environment. Since autonomy values imply a focus on an internal set of behavior guidelines, it is reasonable that orientations to domains closer to the self emerge as the most important ones. In parallel, the fact that the valuation of conformity focuses on obedience of externally imposed rules justifies the stronger link between the valuation of conformity and domains that transcend the self. Even in the presence of the various explanatory mechanisms used here, the direct impact of social status on parental values persists and amounts to approximately half of the total association between these two variables. The only case in which the direct effect of social status is reduced to a degree where it is no longer statistically significant, is the case of the prediction of the valuation of conformity among rural residents, where the total association between social status and this type of parental value was small to begin with.

Appendix The comparability of results between urban and rural areas In order to ensure the comparability between estimated parameters for urban and rural residents, for all the latent variables the measurement weights were constrained to be equal across the two groups (Maruyama, 1998, p. 261). For two of the four latent variables used here (criteria of morality and satisfaction), these constraints do not have a statistically significant effect on the model goodness of fit in comparison to the unconstrained model (see Table A-1). For the other latent dimensions (social status and extrovert orientation), imposing these constraints on the model produces a significantly worse model fit, according to the Δχ2 test.

223

The Values of Romanians: 1993-2006. A sociological perspective Table A-1 Tests of measurement model invariance across residential areas

Unconstrained model χ2 χ2 χ2 RMSEA RMSEA (p) (df) (p) 0.00 0 -------

Social status Extrovert 52.09 4 0.000 0.094 orientation Criteria of 14.25 4 0.007 0.043 morality Satisfaction 32.38 10 0.000 0.041

Equal measurement weights model χ2 χ2 χ2 RMSEA RMSEA (p) (df) (p) 6.97 2 0.031 0.043 0.563

Invariance test Δχ2 Δχ2 Δχ2 (df) (p) 6.97 2 0.031

0.001

64.88 7 0.000 0.078

0.003

12.80 3

0.005

0.623

15.78 7 0.027 0.030

0.944

1.54 3

0.674

0.821

32.67 14 0.003 0.031

0.987

0.29 4

0.990

Notes: A separate model is estimated for each latent variable. The unconstrained model for the social status factor is just-identified.

Due to the fact that χ2 goodness of fit tests and Δχ2 model comparison tests are sensitive to sample size and test an implausible hypothesis of perfect model fit (Cochran, 1952; Jöreskog, 1969; Arbuckle, 2006), an alternative measure of model goodness of fit (RMSEA) is also examined. Browne and Cudeck (1993) and Arbuckle (2006) suggest that an RMSEA value of 0.05 or less indicates a good model fit and values between 0.05 and 0.08 indicate a model with a reasonable approximation error. For the two latent factors for which the Δχ2 test suggested that measurement weights equality constraints produce a significantly worse fit (social status and extrovert orientation), the alternative model fit test (RMSEA) suggests that imposing the constraints produces reasonably fitting models, even if the model fit is significantly reduced in comparison with the unconstrained model. Furthermore, I considered that the comparability of results takes precedence over the model fit. Consequently, the sacrifices in terms of the model fit are acceptable. Additional results In the final model (model 6), all of the indicators have statistically significant loadings on the factors they measure, and all standardized factor weights are greater than 0.5, suggesting that the indicators are strongly linked to the dimensions they measure, providing an empirical justification (in addition to the theoretical justifications) to include them as measures of the latent variables (see Table A-2).

224 corectura 2 28 iulie 2008

Social status and child-rearing values Table A-2. Factor weights (model 6) Urban b

ß

Social status Education 1.000 (0 .783) Occupation 6.532 ( 0.737) Income 0.645 ( 0.630) Extrovert orientation Have a good time 1.000 ( 0.794) Be rich 0.909 ( 0.705) Take risks 0.898 ( 0.724) Be successful 0.907 ( 0.697) Criteria of morality (it is not justifiable to ...) Claim undeserved benefits 1.000 ( 0.722) Accept bribes 0.574 ( 0.625) Cheat on taxes 1.253 ( 0.724) Avoid a fare on public transport 1.101 ( 0.832) Satisfaction Happiness 1.000 ( 0.738) Satisfaction with life 3.098 ( 0.714) Satisfaction with health 0.877 ( 0.566) Satisfaction with amount of money 0.958 ( 0.649) Satisfaction with way of living 0.909 ( 0.706)

Rural p

R2

b

ß

p

R2

--- 0.61 *** 0.54 *** 0.40

1.000 ( 0.784) --- 0.61 6.532 ( 0.676) *** 0.46 0.645 ( 0.529) *** 0.28

--*** *** ***

0.63 0.50 0.52 0.49

1.000 0.909 0.898 0.907

( 0.801) --( 0.723) *** ( 0.722) *** ( 0.701) ***

0.64 0.52 0.52 0.49

--*** *** ***

0.52 0.39 0.52 0.69

1.000 0.574 1.253 1.101

( 0.705) --( 0.691) *** ( 0.766) *** ( 0.895) ***

0.50 0.48 0.59 0.80

--*** *** *** ***

0.55 0.51 0.32 0.42 0.50

1.000 3.098 0.877 0.958 0.909

( 0.712) --( 0.700) *** ( 0.549) *** ( 0.645) *** ( 0.674) ***

0.51 0.49 0.30 0.42 0.46

Note: --- fixed coefficients; unstandardized factor weights in the measurement model are constrained to be equal across residential areas; *** p < 0.001, ** p < 0.01, * p < 0.05, + p < 0.1

The effects of social status on the mediating variables in the final model are all in the expected direction (see Table A-3). Both among urban residents and among rural residents, social status has a positive impact on extrovert orientation, the valuation of non-financial aspects of a job, satisfaction, the degree of control over life, and openness to scientific and technological advances, and a negative effect on criteria of morality and religiosity. In general, the effects are statistically significant, with a single exception (the effect on the valuation of non-financial job aspects among rural residents). The effect of social status on satisfaction is one of the strongest effects in the model (β=0.623 in urban areas and β=0.478 in rural areas).

225

The Values of Romanians: 1993-2006. A sociological perspective Table A-3. Regression coefficients in the prediction of mediating variables (model 6) Extrovert orientation Urban b

p

Rural ß

b

p

Social status 0.131 *** ( 0.310) 0.089 * Age -0.023 *** (-0.317) -0.007 + Single 0.028 ( 0.010) -0.182 No. children 0.072 + ( 0.078) -0.021 Man R2

0.147 +

Non-financial job benefits

( 0.068)

0.196 +

0.23

Urban ß

( 0.179)

b

p

0.231 *** ( 0.348)

0.41

0.002

( 0.019) -0.001

(-0.005)

(-0.019) -0.107

(-0.022)

(-0.029)

0.026

( 0.089)

-0.270 *

( 0.018) -0.097 + (-0.077)

Rural

Urban

Single

-0.353 + No. children -0.078 Man 0.040 R2

ß

b

ß

0.03

b

p

Rural ß

b

p

ß

0.172 ** ( 0.156)

0.247 ** ( 0.188)

( 0,000)

0.001

0.009

( 0.049)

(-0,152)

-0.439

(-0.063) -0.379

(-0.050)

(-0,068)

-0.032

(-0.013)

(-0,019)

p

( 0.043) -0.006

( 0.006)

0.679 ** ( 0.121)

0.149 + ( 0.076) 0.406

0.06

( 0.070) 0.05

Criteria of morality Urban ß

(-0.042)

b

p

-0.055 +

Rural ß

b

p

ß

(-0.096) -0.080 + (-0.125)

0.013 ** ( 0.133)

0.000

(-0.001)

(-0.068)

0.724 ** ( 0.127)

-0.213

(-0.058)

0.221

( 0.059)

(-0.043)

0.107

( 0.073)

-0.120 *

(-0.095)

0.046

( 0.048)

( 0.010) -0.027

(-0.006)

0.069

( 0.024)

0.073

( 0.026)

0.04

0.03

Importance of God Urban

Rural ß

b

p

ß

(-0.102) -0.066 +

(-0.123)

( 0.030)

0.006 +

( 0.080)

0.178 ( 0.040) -0.063 No. children 0.088 ( 0.057) 0.008 Man -0.770 *** (-0.216) -0.265 *

(-0.020)

Single

R2

( 0.045)

( 0,478)

0.193 ** ( 0.195)

0.02

b p Social status -0.072 * Age 0.004

0.168

Control over life

Rural

( 0.100)

(-0.080)

0.11

Openness to S&T advances b p Social status 0.082 + Age 0.006

ß ( 0.103)

-0.078

0.32

Urban

p

0.087

(-0.108)

0.10

b p ß b p Social status 0.134 *** ( 0.623) 0,116 *** Age -0.004 ** (-0.118) 0,000 Single -0.183 *** (-0.134) -0,213 ** No. children 0.057 ** ( 0.120) -0,025 Man -0.107 ** (-0.098) -0,020 R2

b

(-0.063)

Satisfaction Urban

Rural ß

0.08

( 0.010) (-0.111)

0.05

Note: *** p < 0.001, ** p < 0.01, * p < 0.05, + p < 0.1

226 corectura 2 28 iulie 2008

0.03

Unconstrained model χ2 χ2 RMSEA χ2 (df) (p) RMSEA (p) 22.40 10 0.013 0.030 0.974 22.40 10 0.013 0.030 0.974

Equal structural weights χ2 χ2 RMSEA χ2 (df) (p) RMSEA (p) 1445.57 458 0.000 0.040 1.000 1439.54 455 0.000 0.040 1.000 1495.04 476 0.000 0.040 1.000 1428.95 449 0.000 0.040 1.000 1437.86 452 0.000 0.040 1.000 1433.00 455 0.000 0.040 1.000 1432.85 449 0.000 0.040 1.000 1430.03 452 0.000 0.040 1.000 1436.76 455 0.000 0.040 1.000

Invariance test Δχ2 Δχ2 Δχ2 (df) (p) 16.70 10 0.081 10.68 7 0.153 66.17 28 0.000 0.08 1 0.781 8.99 4 0.061 4.13 7 0.765 3.98 1 0.046 1.16 4 0.884 7.89 7 0.342

Invariance test Δχ2 Δχ2 Δχ2 (df) (p) 0.00 1 0.996 4.46 1 0.035

Social status and child-rearing values

Equal structural weight χ2 χ2 RMSEA χ2 (df) (p) RMSEA (p) 22.40 11 0.021 0.028 0.989 26.86 11 0.005 0.033 0.965

Table A-4. Structural weights equivalence tests across residential areas (model 1)

Social status → Autonomy Social status → Conformity

Unconstrained model χ2 χ2 RMSEA χ2 (df) (p) RMSEA (p) 1428.87 448 0.000 0.040 1.000 1428.87 448 0.000 0.040 1.000 1428.87 448 0.000 0.040 1.000 1428.87 448 0.000 0.040 1.000 1428.87 448 0.000 0.040 1.000 1428.87 448 0.000 0.040 1.000 1428.87 448 0.000 0.040 1.000 1428.87 448 0.000 0.040 1.000 1428.87 448 0.000 0.040 1.000

Table A-5. Structural weights equivalence tests across residential areas (model 6)

Social status ↔ Control variables Social status → Mediating variables Control variables → Mediating variables Social status → Autonomy Control variables → Autonomy Mediating variables → Autonomy Social status → Conformity Control variables → Conformity Mediating variables → Conformity

Notes: ↔ covariances, → causal relationships; control variables: age, marital status, number of children, sex; mediating variables: extrovert orientation, non-financial job aspects, satisfaction, control over life, openness to scientific and technological advances, criteria of morality, importance of God.

227

The Values of Romanians: 1993-2006. A sociological perspective

References Alwin, D. F. (1984). Trends in Parental Socialization Values: Detroit, 1958-1983. The American Journal of Sociology, 90 (2), 359-382. Alwin, D. F. (1986). Religion and Parental Child-Rearing Orientations: Evidence of a Catholic-Protestant Convergence. The American Journal of Sociology, 92 (2), 412-440. Alwin, D. F. (1989). Social Stratification, Conditions of Work, and Parental Socialization Values. In N. Eisenberg, J. Reykowski and E. Staub (Eds.), Social and moral values: individual and societal perspectives (pp. 327-346). Hillsdale, NJ: L. Erlbaum Associates. Alwin, D. F. (1990). Historical Changes in Parental Orientations to Children. In P. A. Adler and P. Adler (Eds.), Sociological Studies of Child Development (Vol. 3, pp. 65-86). Greenwich: Jai Press Inc. Alwin, D. F., Hofer, S. M. and McCammon, R. J. (2005). Modeling the Effects of Time: Integrating Demographic and Developmental Perspectives. In R. H. Binstock and L. K. George (Eds.), Handbook of Aging and the Social Sciences. Sixth Edition (pp. 20-38). New York: Academic Press. Alwin, D. F. and Jackson, D. J. (1982). The Statistical Analysis of Kohn's Measures of Parental Values. In K. G. Joreskog and H. O. A. Wold (Eds.), Systems under indirect observation: causality, structure, prediction (Part I, pp. 197-223). Amsterdam ; New York: North-Holland. Alwin, D. F., McCammon, R. J. and Hofer, S. M. (2006). Studying Boomer Cohorts within a Demographic and Developmental Context : Conceptual and Methodological Issues. In S. K. Whitbourne and S. L. Willis (Eds.), The Baby Boomers Grow Up: Contemporary Perspectives on Midlife (pp.45-71). Mahwah, N.J.: Lawrence Erlbaum Associates Inc. Arbuckle, J. L. (2006). Amos 7.0 User's Guide. Chicago, IL.: SPSS Inc. Baron, R. M. and Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of personality and social psychology, 51 (6), 11731182. Blau, P. M., Duncan, O. D. and Tyree, A. (1978). The American occupational structure. New York: Free Press. Bowles, S. and Gintis, H. (1976). Schooling in capitalist America: educational reform and the contradictions of economic life. New York: Basic Books. Breen, R. and Goldthorpe, J. H. (1997). Explaining Educational Differentials: Toward a Formal Rational Action Theory. In D. B. Grusky (Ed.), Social Stratification: Class, Race, and Gender in Sociological Perspective (2001 ed., pp. 459-470). Boulder, CO: Westview Press. Bronfenbrenner, U. (1958). Socialization and Social Class through Time and Space. In E. E. Maccoby, T. M. Newcomb and E. L. Hartley (Eds.), Readings in Social Psychology. New York: Henry Holt & Co. Browne, M. W. and Cudeck, R. (1993). Alternative Ways of Assessing Model Fit. In K. Bollen and J. S. Long (Eds.), Testing Structural Equation Models (pp. 311359). New York: Plenum Press. 228 corectura 2 28 iulie 2008

Social status and child-rearing values Cochran, W. G. (1952). The χ2 test of goodness of fit. Annals of Mathematical Statistics, 23 (3), 315-345. Coleman, J. S. (1988). Social Capital in the Creation of Human Capital. The American Journal of Sociology, 94, s95-s120. Dreyer, A. S. and Wells, M. B. (1966). Parental Values, Parental Control, and Creativity in Young Children. Journal of Marriage and the Family, 28 (1), 83-88. Duvall, E. M. (1946). Conceptions of Parenthood. The American Journal of Sociology, 52 (3), 193-203. Easterlin, R. A. (2001). Income and Happiness: Towards a Unified Theory. The Economic Journal, 111 (473), 465-484. Fernandez, R. M. and Kulik, J. C. (1981). A Multilevel Model of Life Satisfaction: Effects of Individual Characteristics and Neighborhood Composition. American Sociological Review, 46 (6), 840-850. Ganzeboom, H. B. G. and Treiman, D. J. (2003). Three Internationally Standardized Measures for Comparative Research on Occupational Status. In J. H. P. Hoffmeyer-Zlotnik and C. Wolf (Eds.), Advances in cross-national comparison: a European working book for demographic and socio-economic variables (pp. 159-193). New York: Kluwer Academic/Plenum Publishers. Hagenaars, J., Halman, L. and Moors, G. (2003). Exploring Europe’s Basic Values Map. In W. Arts, J. Hagenaars and L. Halman (Eds.), The Cultural Diversity of European Unity: Findings, Explanations, and Reflections from the European Values Study (pp. 23-58). Leiden: Brill. Inglehart, R. (1977). The Silent Revolution: Changing Values and Political Styles among Western Publics. Princeton, N.J.: Princeton University Press. Inkeles, A. (1960). Industrial Man: The Relation of Status to Experience, Perception, and Value. The American Journal of Sociology, 66 (1), 1-31. Jöreskog, K. G. (1969). A general approach to confirmatory maximum likelihood factor analysis. Psychometrika, 34 (2), 183-202. Kluckhohn, C. (1951). Values and Value-Orientations in the Theory of Action. In T. Parsons and E. A. Shils (Eds.), Toward a general theory of action (pp. 388433). Cambridge: Harvard University Press. Kohn, M. L. (1963). Social Class and Parent-Child Relationships: An Interpretation. The American Journal of Sociology, 68 (4), 471-480. Kohn, M. L. (1969). Class and conformity: a study in values. Homewood, Ill.: Dorsey Press. Kohn, M. L., Naoi, A., Schoenbach, C., Schooler, C. and Slomczynski, K. M. (1990). Position in the Class Structure and Psychological Functioning in the United States, Japan, and Poland. The American Journal of Sociology, 95 (4), 9641008. Kohn, M. L., Schooler, C., Miller, J., Miller, K. A., Schoenbach, C. and Schoenberg, R. (1983). Work and Personality: An Inquiry Into the Impact of Social Stratification. Norwood, New Jersey: Ablex Publishing Corporation. Kohn, M. L., Slomczynski, K. M. and Schoenbach, C. (1986). Social Stratification and the Transmission of Values in the Family: A Cross-National Assessment. Sociological Forum, 1 (1), 73-102. Lareau, A. (1987). Social Class Differences in Family-School Relationships: The Importance of Cultural Capital. Sociology of Education, 60 (2), 73-85. 229

The Values of Romanians: 1993-2006. A sociological perspective Lenski, G. E. (1961). The religious factor: a sociological study of religion's impact on politics, economics, and family life (1st ed.). Garden City, N.Y.: Doubleday. Luster, T., Rhoades, K. and Haas, B. (1989). The Relation between Parental Values and Parenting Behavior: A Test of the Kohn Hypothesis. Journal of Marriage and the Family, 51 (1), 139-147. Maruyama, G. (1998). Basics of structural equation modeling. Thousand Oaks, Calif.: Sage Publications. McNally, S., Eisenberg, N. and Harris, J. D. (1991). Consistency and Change in Maternal Child-Rearing Practices and Values: A Longitudinal Study. Child Development, 62 (1), 190-198. Pearlin, L. I. and Kohn, M. L. (1966). Social Class, Occupation, and Parental Values: A Cross-National Study. American Sociological Review, 31 (4), 466-479. Roberts, G. C., Block, J. H. and Block, J. (1984). Continuity and Change in Parents' Child-Rearing Practices. Child Development, 55 (2), 586-597. Roccas, S., Sagiv, L., Schwartz, S. H. and Knafo, A. (2002). The big five personality factors and personal values. Personality and Social Psychology Bulletin, 28 (6), 789. Schaefer, E. S. (1959). A Circumplex Model for Maternal-Behavior. Journal of Abnormal and Social Psychology, 59 (2), 226-235. Schwartz, S. H., Melech, G., Lehmann, A., Burgess, S., Harris, M. and Owens, V. (2001). Extending the Cross-Cultural Validity of the Theory of Basic Human Values with a Different Method of Measurement. Journal of Cross-Cultural Psychology, 32 (5), 519-542. Sewell, W. H., Haller, A. O. and Ohlendorf, G. W. (1970). The Educational and Early Occupational Status Attainment Process: Replication and Revision. American Sociological Review, 35 (6), 1014-1027. Sewell, W. H., Haller, A. O. and Portes, A. (2001). The Educational and Early Occupational Attainment Process. In D. B. Grusky (Ed.), Social stratification: class, race, and gender in sociological perspective (2nd ed., pp. 410-420). Boulder, Colo.: Westview Press. Sewell, W. H. and Hauser, R. M. (1980). The Wisconsin Longitudinal Study of Social and Pshychological Factors in Aspirations and Achievements. Research in sociology of education and socialization, 1, 59-99. Slomczynski, K. M., Miller, J. and Kohn, M. L. (1981). Stratification, Work, and Values: A Polish-United States Comparison. American Sociological Review, 46 (6), 720-744. Spade, J. Z. (1991). Occupational Structure and Men's and Women's Parental Values. Journal of Family Issues, 12 (3), 343. Treiman, D. J. (1977). Occupational prestige in comparative perspective. New York: Academic Press. Veenhoven, R. (1995). The Study of Life Satisfaction. In W. E. Saris, R. Veenhoven, A. C. Scherpenzeel and B. Bunting (Eds.), A Comparative Study of Satisfaction with Life in Europe (pp. 11-48). Budapest: Eötvös University Press. Whitbeck, L. B. and Gecas, V. (1988). Value Attributions and Value Transmission between Parents and Children. Journal of Marriage and the Family, 50 (3), 829-840. Wothke, W. (2000). Longitudinal and multi-group modeling with missing data. Accessed March 26, 2007, 230 corectura 2 28 iulie 2008

Social status and child-rearing values http://www.smallwaters.com/whitepapers/longmiss/Longitudinal%20and%20 multi-group%20 modeling%20with%20missing%20data.pdf Wright, J. D. and Wright, S. R. (1976). Social Class and Parental Values for Children: A Partial Replication and Extension of the Kohn Thesis. American Sociological Review, 41 (3), 527-537. Xiao, H. (2000a). Class, Gender, and Parental Values in the 1990s. Gender and Society, 14 (6), 785-803. Xiao, H. (2000b). Structure of Child-Rearing Values in Urban China. Sociological Perspectives, 43 (3), 457-471.

231

Identity and axiological profile: value identifications for Romanian young people HORAŢIU RUSU

Identity is a fascinating subject, always in the limelight, which was and is still approached in many ways, both in the scientific and in the public, media or political agenda. In this respect, we may, for example, think about “Face Off”, a movie directed by John Woo, which in 1997 achieved huge success with the public1. The entire plot of the movie focuses on the reciprocal substitution of the identities of the main characters. Sean Archer, an FBI agent, trying to baffle a terrorist plan, takes Castor Troy’s look, a famous criminal, who was caught by the police and found in a coma. Unfortunately, things do not happen as they were secretly planned by the FBI; Castor comes to his senses and, in his turn, takes hold of Archer’s look. Of course, the “role plays” they interpret are imperfect, so the people with whom they interact begin to ask themselves: “Who is this Castor?”, respectively “Who is this Archer?”. Another example, which circulates more in the political environment, is known under as the The Duck Test. In 1950, during the Cold War, Richard Patterson, The United States’ ambassador in Guatemala, accusing the local government that it was communist, used the following allegory: “...suppose you see a bird walking around in a farm yard. This bird wears no label that says ‘duck’. But the bird certainly looks like a duck. Also, he goes to the pond and you notice he swims like a duck. Then he opens his beak and quacks like a duck. Well, by this time you have probably reached the conclusion that the bird is a duck, whether he’s wearing a label or not” (Immerman, 1982:102). The question that lies behind this argument is: “How can we know that a certain bird is a duck?”. 1

The movie is among the first 10 most successful movies, measured according to income in the USA and also it was in the 13th position worldwide in the same year.

232 corectura 2 28 iulie 2008

Identity and axiological profile: value identifications for Romanian young people

“Who or what is X?” and “How can we establish who X is or what X is?” are maybe the first and the most important questions that lay behind the reflections researched by any person interested in the domains of identities, be they individual – as in the first example – or collective – as in the second example. Apparently simple, the answers I am searching for cannot be grasped before finding out the theoretical solution to the question, “What is identity?” i.e., how the title also suggests, “What is identity at a collective level?”. Keeping this in mind, I will also raise the question “How can we establish an identity profile at a collective level?” and, only after this, I will search for answers for the main question, the one that organizes this entire chapter: “What is the Romanian youngsters’ identity profile, from an axiological point of view? This chain of questions transposes in the structure of the chapter in the following way: an initial part where I make an incursion in the universe of the theories of collective identities, followed by a brief review of several contemporary conceptual models of the axiology of social domain; a second part dedicated to the description of the methodology and the used indicators; a third part dedicated to the analyses and data presentation; finally, the last part where I outline some conclusions. Also, here should be mentioned that by young people I mean the persons between 18 and 35 years old.

Theories on collective identities or “How can we establish that a certain bird is a duck?” Besides the fact that it is a very common word in everyday life, no matter the age, the sex, the profession, the ethnicity or the race, identity is one of the most ubiquitous concepts of the literature concerned with social sciences today, to which a lot of definitions were given, and which probably encompasses a great number of differences. Thus, we find it quite frequently in a large panel of combinations: cultural identity, national identity, ethnic identity, social identity, collective identity, categorical identity and so on (see Smith, 1991; Gellner, 1983; Bauman, 1992; Calhoun, 1994; Tilly, 1996; Jenkins, 1996, etc.). Sometimes the spheres of these notions overlap to a larger extent, some other times to a smaller extent or even not at all, sometimes they are considered interchangeable and sometimes they exclude each other. These differences or similarities, that originate in the positions assumed by the scientists of identity on ontological (between realism2 and

2

It is a position named by some scientists called holism, or naturalism, or structuralism, integrated or currently known in the social theories of identity and culture under the generic name of essentialism and which postulates the existence of some superstructures, certainties, essences, realities independent of individuals and above 233

The Values of Romanians: 1993-2006. A sociological perspective

relativism3), epistemological (between objectivism and subjectivism) and methodological (between quantification of the causal relations and interpretation or description) continuums, find themselves, in different forms or formulations, in the theories of identity. Presently, the differences, oppositions or bipolarities that were evoked previously are transposed in a quasiunanimous way – see: Sayer (1997), Bacová (1998), Woodward (2001), Brubaker, Loveman and Stamatov, (2004), etc. – in the terms of essentialism and constructivism (for other typologies related especially to the theories of national and ethnical identity, see: Smith (1999: 319), Ozkirimli, (2000), Dungaciu, (2004: 26-31). Of course that, as Sayer (1997) also suggests, there are many variants of essentialism and many critics of essentialism, each one of them modifying along with the context and the problem that is discussed. We meet the “extreme” essentialist claiming that everything is full of substance, and “moderate” essentialisms accepting that, while some things have substance, some other do not. What unifies them all is the convergence to the assumption that “objects” exist by themselves and also the fact they possess certain essential properties that make them a certain kind, rather than the other. As there are many types of essentialisms, there are also many types of constructivism. The area of the theories that can be comprised under this term is at least the same large and various as in the case of essentialism. However, what really unifies them are the basic assumptions on social reality, which is constructed, changeable and multiple; things do not exist by themselves, but only as categories of constructs generated subjectively, as our representations. Thus, constructivism reunites those perspectives that support the idea of some contextual, processual, instrumental, changing or invented4 identities, and essentialism reunites those perspectives that advance or start from the idea of the existence of some fixed, static, with pronounced determinist character, affective, homogenizing5 identities. All these points of view are accompanied by speeches about culture, cultural features, structures, borders, categories, power, actors, memories, images, emotions, cognitions, etc. The main axes them and still reproduced by them; the identity appears as existence in itself, as a priori reality, as a structure. 3 It is a position that could be named either non-realist, or nominalist, or even individualist. Currently, in the social theories on identity, it is known under the generic name of constructivism and postulates that the existence of reality is produced by individual consciences; identity emerges as a product of the self, as an independent representation in and through conscience. 4 Theories known under names like: instrumentalism, situationalism, cognitivism, modernism, inventionism or even postmodernism tend to be included in this group. The same happens for authors like: Barth, Mitchell, Brass, Horowitz, Calhoun, Anderson, Jenkins, etc. 5 For example, theories like primordialism, perrenialism, sociobiologism and authors like Shils, van den Berghe, Geertz, Isaac, Grosby, Connor, etc. tend to be included in this group. 234 corectura 2 28 iulie 2008

Identity and axiological profile: value identifications for Romanian young people

around which any debate on identity gravitates are: culture - heredity, structure action, collective - individual, affectivity -cognition, emic - ethical. If we were to imagine that we found in a room where there are also several research fellows in disagreement with the status of identity, we surely could select some fragments from their dialogues, as follows: “it is natural…”; “it is invented…”; “it is an elites’ instrument”; “it is hereditary, it is in our genes”; “it is a matter of pure cognition”; “it is a matter of emotions and feelings”; “it is a category created by others”, “it is a deeply subjective belief, it exists only in our imagination”. The consequence of this so-called “fight” – existing ever since the world began, between Eleatics and Heraclitians6, as some people would say – is that most of the discussions are captive in an apparently sisyphic attempt to solve the problem in an unequivocal way: is identity inherited, essential, perennial or is it permanently socially built and re-built, fluid and imagined? The answer to this question can be given if we leave aside both the extreme essentialist (determinist) position – which confines the idea of the free will and transforms the individual in “cultural dopes”, the way Garfinkel (1967:68) called it – and the extreme constructivist one (postmodern) – which, in fact, deconstructs the social reality and, in the worst case, inducts solipsism. The simpler the answer may seem, the more paradoxical it is: identity is, at the same time, essential and constructed. Arguments to support this idea can be found both on the ground of philosophy, social psychology and sociology and on the ground of the theories of identity. An example in this sense, familiar to identity logicians, is known under the name of “Theseus ship”. Theseus was an Athenian hero, who defeated the Minotaur, thus saving the lives of the young Athenians, sacrificed once every nine months in the Labyrinth in Crete. They say that the ship on which he traveled to the island of Crete and the way back was kept by Athenians in his honor and at the same time as a symbol of the victory over Minos, the Cretan king. Once a year, Athenians used to organize a parade and sail on Theseus’ ship. Of course, one may suppose that, over time, parts of the ship eroded and it was necessary to replace them with some new parts, shaped exactly the same way and built exactly from the same materials as the old ones. Over time, it is normal to suppose that every part of the ship was gradually replaced. If at the certain Tk moment in time, all parts came to be replaced, the following question could be raised: Is Theseus’ ship fundamentally the same one? The answer to this question can be given if we search for it not in philosophy, or in logic, but in sociology. The solution is provided by the wellknown Thomas theorem, claiming that a situation defined as real becomes real 6

Eleatics (Xenophanes of Colofon, Parmenides of Elea, Zeno of Elea) supported the uniqueness and immutability of world and existence, while Heraclides of Ephesus favored the idea of continuous change, very suggestively expressed by his aphorism: Panta Rhei. 235

The Values of Romanians: 1993-2006. A sociological perspective

in its consequences (Thomas and Thomas, 1928: 572). Transposed in the language of the previous example, the theorem says that what makes the ship the same is the people’s faith or the fact that people perceive and identify it as being the real one and act accordingly towards it, organizing parades. To express it in a different way, even though, from a qualitative point of view, the “object” changes its properties (essences), its significance is the only one that matters. It is the same ship because this is what people (the Athenians) believe and think, and one of the main rules of the researcher interested in the social domain is never to neglect what people say, because, in fact, this is what really interests him. Consequently, not the formal logic, but the social logic is the one that allows the use the concept of identity and the study of its change at the same time. The idea that all the categorical constructs are essential is also supported by Allport (1954) – in the domain of social psychology – who shows that (discussing the problem of prejudices, of the sources of discrimination in inter-group’s relations) people do not only have the tendency to think in categorical terms, but they also have the tendency to look at social categories as being natural and to develop beliefs in their essential character. In the field of the theories of identity, the same idea is found from Smith7, who states that, in fact, “the social construction”, seen as a type of categorization, does not necessarily mean a negation of the reality of identity, but an underlining of its ambivalent status: of a real (essential) product and at the same time8 of an imagined (constructed) product. Connor finds on the same coordinates when he says that “Identity does not draw its sustenance from facts but from perceptions; not from chronological/factual history but from sentient/felt history”, in other words, identity is there, it exists because it is perceived as being there; it is perceived as being “eternal” and “above time” and “not facts but perceptions of facts that undergird attitudes and behavior” (Connor,, 2004: 45). Finally, theories from other fields of sociology lead to the same idea. Among them, it could be mentioned the theory of structuration, Giddens’ idea of duality of the social structures and agents. There is a “structural” reality more or less fluid where the social actor performs – either we see it as a construction, as a great narrative, as a pre-assigned thing or as a gift (Bădescu, 2002: xxvii): there are institutions or typical ways of acting, there are symbols or memories and there are “legitimized” interpretations given to them, etc. – but, at the same

7

In the framework of a debate with Gellner, named “The nation: real or imagined?” (see http://members.tripod.com/GellnerPage/Warwick0.html) 8 We may also notice as a result of a thorough study that most of the theories of identity contain this implicit idea, the ambivalence of identity: being at the same time state and process. In other words, identities “are constructions” (structures or essences) and “they are built” (processes or categorizations) at the same time, as it can be seen at Durkheim (1964), Weber (1968) or from Shils (1957) to Hall (1991). 236 corectura 2 28 iulie 2008

Identity and axiological profile: value identifications for Romanian young people

time, they exists only to the extent in which they are produced, reproduced and recognized by social actors. Transposed in the field of identity, this idea can sound in the following way: identity exists at a collective level to the extent in which it is reproduced at an individual level. Thus, the problem of the acknowledgment of reality, of the uniqueness, stability and efficiency of identity raises in the register of its actualization, its production and re-production by social9 action, be it discourse, attitude or behavior of social actors; in other words, it becomes effective or real when it has implications for the social actors and he/she identifies with it. Thus it can be said, starting from this idea, that collective identity is the identification with the others through similar “elements” or characteristics. Hence, it results the methodology I use in the present paper. I do not search for “discovering” the identity of a collectivity, of a collective entity a priori assumed, but I decide for the search of common aspects of individual identities, i.e. for the elements of common identification – be they values, memories, images, processes, or feelings – of the social actors. In other words, not the reified constructions themselves, but the process of identification with them is the one that gives the reality of a collective identity (see also Barth (1969) for details regarding the relation between categorization, (ethnical) identity and (cultural) elements). In other words, on the collective-individual axis, the landmark of the empirical study is represented by the individual, and the collective identity will be an emergence of the common identifications. Thus, this is the answer which shows “how can we establish an identity profile at a collective level” and which allows me to pass on to the next question “Which identitary profile, at an axiological level, do young people in Romania have?” The spheres of identity are, as it could be inferred until now, numerous and they have their referential, depending on the examined theories: from common images, collective memories, values or ideologies10. In this paper, as the title also suggests, I try to investigate only one of them, namely the axiological dimension of the collective identity of the young people. In other words, what I will present in the following sections is subordinated by the idea of the study of values with which Romanian young people identify today.

The axiological dimension of identity Values, as we already know, constitute a kind of abstract, absolute landmarks that cannot be observed straightway, and that are approached specifically by every collectivity, community, ethnicity, nation through their 9

Or “feeling”, as Bădescu says (2002:xxxiii). See any of the theories of identities from Durkheim (1964) to Weber (1978) or Cooley (1909), from Geertz (1963) to Shils (1957) to Barth (1969), Horowitz (1975), Anderson (1983), Brass (1991), Smith (1999) and others. 10

237

The Values of Romanians: 1993-2006. A sociological perspective

own culture, during their history. Values add an important tinge, probably the most important, to the portrait of “who is or what is”. One of the definitions given to the value to which many research fellows refer is the one from Kluckhohn: value is “a conception, explicit or implicit, distinctive of an individual or characteristic of a group, of the desirable which influences the selection from available modes, means and ends of action” (Kluckhohn, 1951: 395). They appear as elements of the symbolic, latent universe of reality and as a support of the level of updates. “The achievement” of values thus represents one of the reasons of all the individuals’ actions, namely the entire social life: “the system of value standards when institutionalized in the social systems and internalized in personalities guides the action” (Parsons and Shils, 1992: 41). As values orientate people’s actions and options, but they cannot be studied directly, their analysis implies the investigation of the level of actualizations, i.e. the level of attitudes, knowledge and individuals’ behavior. There are three theoretical and empirical models most invoked by social scientists today11 in the study of the axiological dimension: Inglehart’s model, Schwatz’s model and Hofstede’s model. The analyses I will make further on, uses, as a background, especially the first two models. That is why I will present only their main ideas synthetically. Inglehart’s model proposes a hierarchical distinction between two types of values: materialist and postmaterialist and, at the same time, it predicts an evolution of orientations at the level of societies from the first to the second type. Theoretically, this distinction bases not only on Weber’s thesis of secularization, but also on Maslow’s (1943) thesis of the motivational pyramid. In a practical way, Inglehart draws his hypothesis on the observation of the solutioning of the crisis of the basic resources after World War II and on the orientation of the political systems to solving other types of problems. His analyses indicate the existence of a connection between the interest in satisfying some basic needs (social security, economic prosperity, and so on), identified as being materialist and the low economic development of a society, and the interest in satisfying some superior, intellectual, esthetical needs (selfachievement, preoccupation for the environment, and so on) identified as being postmaterialist, and a strong economic development (Inglehart, 1997). Schwartz proposes a universal typology of values. The structural organization (the arrangement, the pattern) of values according to this model is given by the compatibility or incompatibility between them. From the point of view of the theoretical construction, Schwartz draws on Rokeach’s model (1973) which shows that the differences among individuals are given rather by the importance they confer to certain values to the detriment of others than by their presence or absence in the individuals’ orientation system. The ten basic values, about which Schwartz says can be identified in all the societies, are: self 11

See, for example, Arts et al. (2003); Vinken et al. (2004); Ester et al. (2006); Ramos (2006).

238 corectura 2 28 iulie 2008

Identity and axiological profile: value identifications for Romanian young people

direction, stimulation, hedonism, self-achievement, power, security, conformism, tradition, benevolence, and universalism. They are structured in two dimensions: openness to change versus conservation and self-enhancement versus self-transcendence (Schwartz, 1994, 2006). Hofstede’s model proposes five dimensions of culture: social distance, the avoidance of uncertainty, individualism versus collectivism, masculinity versus femininity and the orientation to the future (for details, see Hofstede, 2001). As already announced, I will not refer to this model, given the fact that I do not have data that would allow me to create the indicators proposed by it. Before passing to the second part of the work, where I will describe the methodology and the items used, I would like to mention again that the entire analysis is dedicated only to the axiological dimension of identity, values being, as I showed in the previous section, a part of the collective identity.

Indicators and methodology The empirical analysis I propose is, by and large, a descriptive one, following especially to point out the structure, stability or the dynamics of value identifications from the perspective of the longitudinal available data – the data bases I will use are EVS 1993, EVS 1999 and WVS 200512. In order to measure the values, I will use a series of indexes built on the basis of the factorial analyses. The structure and the stability of values, where the analyses imply the determination of some factors, will be verified with the help of the correlation between the factors obtained in each wave and an index created on the basis of the saturations of the factors obtained in 2005, and the dynamics of the identifications with the help of ANOVA analysis on the basis of the indexes I obtained. Besides this, I will try to underline the possible differences among various categories of young people; in other words I will examine the game of overlapping between collective and categorical. Of course, it is not my purpose to be exhaustive in what concerns the axiological universe, a reason for which I will stop at the investigation of value identifications that allow longitudinal13 comparisons. Thus, the first step is to find those values (that is results of the common identifications or factors, in terms of the statistical analysis) which could help to outline an axiological profile whose stability could be studied. Above the conditions imposed by the availability of the data, I took into account, when I chose the factors, not only the theories regarding the value systems evoked previously, but also a taxonomy resulted from Arts’ et al. analyses (2003: 23-48). They identify, on the basis of 12

The youngsters’ sub-samples taken into account, corresponding to the age interval 18-35 years old include 399 subjects in 1993, 369 subjects in 1999, respectively 486 in 2005. 13 on the basis of the items that can be found in all the three questionnaire and data bases EVS/WVS, 1993, 1999, 2005. 239

The Values of Romanians: 1993-2006. A sociological perspective

EVS data, two dimensions of the axiological universe (similar to both at a social and individual level): a dimension that would represent the social-liberal orientation (whose key features, related to postmodernism, are individual liberty and autonomy14) and a dimension that would represent the normative orientation (among whose features will be: the support for the traditional family, religiosity, the strong appreciation for the social norms and so on, things that are related to materialism). Thus, I endeavored that the values I used in the analysis to be theoretically distributed on a conservatism/normativism – openness to change axis. More precisely, I considered the support for: religiosity, the idea of authority versus autonomy, intolerance (towards deviant behaviors), gender (in) equity (as values of a conservative type) and alternative family models, equity, permissiveness, the liberal model of a society versus the egalitarian model of a society (as values related to the dimension of openness to change)15. Different studies show that religiosity is not only understood, but also measured in different ways (see Bădescu, 2002; Halman and Petterson 2006; Voicu 2006; Rughiniş 2006). The most frequently explored dimensions of religiosity are the faith and the religious practice. In my analysis I propose an index that includes, besides the dimensions mentioned above, a dimension of self-perception and another one of the positioning towards the church. Thus, the index consists of other indexes that measure separately the importance of religion and also God’s importance, the Christian ethos and the importance attributed to the institution of church, an indicator that measures the religious behavior and another indicator that measures the self-perception of religiosity (see the annex). Permissiveness is measured on the basis of the acceptance of some behaviors that transcend the borders of “normality”: homosexual behavior, prostitution, abortion, euthanasia, divorce. The index that I proposed is also present in some other analyses, either as a simple index, or encompassing two dimensions of permissiveness: a personal and a sexual one (for details, see for example Arts et al., 2003). Tolerance is generally understood as the acceptance of some behaviors or attitudes with which we do not agree or which we do not like (Medrano & Rother, 2006). The intolerance towards deviant people is determined in my analysis on the basis of the non-acceptance as neighbors of some persons belonging to some groups traditionally considered deviant: alcoholic people, homosexuals and drug dependents. The index I will name “authority” from now on appears frequently in the literature under the name of authority versus autonomy and has various 14

This “type of individualism” – as the authors say – should not be interpreted in terms of egoism, narcissism, hedonism or ethical relativism; it does not lack a spirit of community but is socially committed; ”– Arts et al. (2003:31). 15 The practical way in which these indicators are constructed is described in the annex. 240 corectura 2 28 iulie 2008

Identity and axiological profile: value identifications for Romanian young people

modalities of construction. The different variants actually have as background the Weberian theory of the rationalization of societies. On this basis, Inglehart16 (1997:390) proposes that the value should be given by the subtraction between the sum of two indicators that denote the spirit of “Protestantism” (perseverance/determination and thrift) and the sum of two indicators that denote the preference for the traditional facet (religiosity and obedience). For example, Schwartz (1992) creates two indexes, one of conformism and the other one of independence which, in the model he advances, appear at the counter-poles of the axis. Conformism is an additive index created on the basis of an item that measures the importance conferred to the freedom of choice and of another item that measures the importance conferred to creativity. Here, I decided for a more elaborated variant, after the model used by Arts et al., too (2003: 53). The index is of an additive type, the items that refer to the support for authority, namely hard work, thrift, religious faith, obedience are summed up, and the ones that refer to the support for autonomy, namely independence, responsibility feeling, perseverance/ determination and imagination were subtracted. The other values included in the analysis (the acceptance of alternative family models, gender inequity, equity, acceptance/non-acceptance of the model of egalitarian society) are based on the answers to a single question.

What kind of value identifications can we find from young people? First of all, in this section, I will try to see how the eight values included in the analysis group together over time (from 1993 to 2005), and I will also see if there is a preference for certain type of values. Theoretically speaking, as was expected, collective identifications with the values enumerated above are made on two different dimensions: conservatism and openness to change. These two dimensions (easily visible in Figure 1) were practically obtained on the basis of the cluster17 analysis. Intolerance, gender inequity, support for the authority and religiosity group together in a cluster that reflects the conservatism, while the support for the equity, for the alternative family models, and the liberal society model group together in a cluster that reflects openness to change. The dendograms in Figure 1 present the analyses results for all the three waves. 16

For the abbreviated variant of achievement index. Although I tested the way values group with the multidimensional scaling method, obtaining similar results, I preferred to present in this paper the results obtained using the cluster analysis method, due to the fact that the graphical representation is more intuitive. The cluster analysis used here is a hierarchical one, with clustering on variables, Ward method; the variables were standardized. The solution is stable, similar results are obtained using the Baverage method. 17

241

The Values of Romanians: 1993-2006. A sociological perspective

Figure 1. The clustering of the values on the dimension conservatism/openness to change in the dynamics 1993-2005

242 corectura 2 28 iulie 2008

Identity and axiological profile: value identifications for Romanian young people

Note: The inclusion of the identification with the liberal/egalitarian society model in the cluster that refers to the conservative dimension is due to an abrupt rearrangement on the liberalism/egalitarianism “axis”. In 1993 the average value is 7,02, a fact that indicates a strong support for liberalism, and in 1999 it is 3,86 indicating a strong support for egalitarianism; in 2005 it comes back to 5,14.

243

The Values of Romanians: 1993-2006. A sociological perspective

The data for 2005 allow me, besides the analyses above (created having as theoretical reference Inglehart’s, Arts’ et al., models), to resort to some of the afferent indicators of Schwartz’s model (see the annex). In Schwartz’s original version, values are measured according to the answers given to 21 questions. Each index of the ten fundamental values is built with 2 individual indicators, except for universalism, built with 3 indicators. WVS/EVS questionnaire from 2005 includes only 10 items (from all the 21), one for each value. Further on, I intend to analyze on the same theoretical axis – conservatism-openness to change – the intensity of the value identification, with the help of Schwartz-type available items. According to Schwartz (2006), on this axis, self-direction and stimulation (i.e. openness to change) are opposed to tradition, conformity and security (i.e. conservatism). This theoretical assumption (for the Romanian society) of the two poles’ segregation is also confirmed by the data obtained through cluster18 analysis (Figure 2).

Figure 2. The clustering of the values on the conservatism/openness to change axis in 2005 on the basis of Schwartz items

18 The choice for the presentation of the cluster analysis results is motivated by the same principle that I have presented above, that is the possibility of facile visual interpretation. I used a hierarchical-type cluster analysis with clustering on variables,, i.e. the Ward method. The solution is stable, and similar results are obtained when using the Baverage method. Similar results are also obtained if the multidimensional scalar method is used (the coefficient that measures the agreement to data Normalizes raw stress = 0,005 and the similarity coefficient Tucker = 0, 99). The same answer is obtained with the same two dimensions if the factorial analysis method is being used.

244 corectura 2 28 iulie 2008

Identity and axiological profile: value identifications for Romanian young people

The hierarchy of value identifications that gives us a hint regarding the pole that prevails, the “most powerful” pole, is practically established based on the average values of each item (Figure 3): the lower the average is (according to the scale – see the annex), the more intense the identification with the respective value is. Figure 3. The hierarchy of value identifications (Schwart’s items)

Tradition (2.47)

Security (2.52)

Conformism (2.55)

Self direction (2.61)

Stimulation (3.89)

Note: Each of the five dimensions can vary from a theoretical minimum value of 6 to a theoretical maximum value of 1.

It can be noticed that, although the differences are not always statistically19 significant, the three values assimilated to the conservative dimension tend to situate in the superior part of the “pyramid”, and the other two (openness to change type) tend to place at the pyramid basis. This “hierarchy” allow me to say that in 2005 the young people preferred to identify themselves with values belonging to the category of conservative ones.

Young people’s values identification: which way? As it was noticed above, using a synchronic analysis, the identifications with conservative values seem to prevail over the identifications regarding the 19

The significant differences for p=0,000 are situated among tradition-stimulation, security-stimulation and conformism- stimulation, and for p<0, 1 between security-self direction and tradition-self direction. 245

The Values of Romanians: 1993-2006. A sociological perspective

openness to change. Further, I will follow the evolution of this relation (conservatism - openness to change) from a longitudinal perspective. The period 1993-2005 offers the possibility to see which way youngsters’ preferences went, and even though I do not intend to make predictions, it leaves open the opportunity to advance some hypotheses. To analyze the dynamics of the phenomenon, the changes in the youngsters’ values, identification were looked at, first of all, at a global level and then in particular to see if there are differences, depending on some categories (the place where youngsters live, sex and age). Table 1 synthesizes the results of the analyses made at a global level. Table 1. The dynamics of Romanian youngsters’ value identifications Year Values Authority Religiosity Gender inequity Intolerance Unegalitarian society model Permissiveness Alternative family model Equity

1999 versus 1993 È È

2005 versus 1999 ∼ È

2005 versus 1993 È È

∼ ∼

∼ ∼

∼ ∼

Ê

È

Ê

∼ ∼ ∼

Ê Ê Ê

Ê Ê ∼

Note: the sign È or Ê reflects a significant intensification or a decrease (to p<0.05) of the identification with the mentioned values in comparison with the reference year; the sign ~ indicates the fact there are no significant changes in comparison with the years when the reference is made. (The results are based on the ANOVA type analysis; for gender inequality, intolerance, inequality society model, the used test is Bonferroni, and for all the others, Tamhane).

The global image obtained form the data presented here, indicates that from 1993 to 1999 we assist to an intensification tendency of the identification with the so-called conservative values (the religiosity and the preference for the authority increase), and a stability tendency or even diminishing tendency of the identification with values that would indicate openness to change (the support for an unequal society model). The observation is similar for the period 19992005 (religiosity is continually increasing while the support for the alternative family model, equity and permissiveness is continually decreasing) except for the identification with the idea of social inequality. The dynamics is more striking in the larger period 1993-2005 (religiosity and preference for the authority are significantly increased while the support for the model of an unegalitarian society and the model of the alternative family and permissiveness 246 corectura 2 28 iulie 2008

Identity and axiological profile: value identifications for Romanian young people

are significantly lowered). In a practical way, the result can be compared with the facts indicated by the hierarchy of values built above and depending on Schwartz’s model items. If the analysis based on Schwartz’s items indicates the fact that conservative values are situated in 2005 in the superior part of the identification “pyramid”, Table 1 shows that the young people in 2005 identify themselves more than in 1993 or 1999 with conservative values and less with those values that reflect openness to change. Table 2 presents the results of the analyses made on the categories of young people from different residential environments. The interpretation of this table can be made both diachronically and synchronically. I am not going to make an individual presentation, at the level of the indicator, but I intend to make the “translation” globally, depending on the two axes: conservatism openness to change axis and rural - urban axis. Table 2. Value identification of the young people in the urban and rural areas Comparison

Rural in comparison with Urban

Urban

Rural

1999 2005 2005 1999 2005 2005 Value orientations 1993 1999 2005 versus versus versus versus versus versus 1993 1999 1993 1993 1999 1993 Authority È È È È ∼ ∼ ∼ ∼ ∼ Religiosity È È È È È Ê ∼ ∼ ∼ Gender inequity È È È ∼ ∼ ∼ ∼ ∼ ∼ Intolerance ∼ ∼ ∼ ∼ ∼ ∼ ∼ ∼ ∼ Unegalitarian Ê Ê È Ê Ê Ê ∼ ∼ ∼ society model Permissiveness Ê Ê Ê Ê Ê ∼ ∼ ∼ ∼ Alternative society Ê Ê Ê ∼ ∼ ∼ ∼ ∼ ∼ model Equity Ê ∼ ∼ ∼ ∼ ∼ ∼ ∼ ∼ Notes: 1. the signs È or Ê indicate a significant positive difference and a significant negative difference respectively (p≤0,05) of the identification with the values enumerated in comparison with the year and the residence environment to which the reference is made; 2. the signs Ì or Ê have the same meaning only for p≤0,10; 3. the sign ~ indicates the fact that there are no differences of identification between residence environments or significant changes regarding the years to which the reference is made. 4. The results are based on the ANOVA analysis (For authority and the unegalitarian society, the test I used was Bonferroni and, for all the others, Tamhane)

247

The Values of Romanians: 1993-2006. A sociological perspective Examples of interpreting: 1. The first cell in the upper left corner indicates the fact that the young people living in the urban and rural areas were found in 1993 on the same positions regarding the orientation to a society based on authority. 2. The first cell in the upper right corner indicates the fact that the preference for the authority of the young people living in the rural areas is significantly increased in 2005 than in 1993.

Synchronically, that is from the point of view of the comparisons of the identifications in the same year among youngsters from different residential environments (first three columns of the Table 2), the analysis reveals the fact that in 1993 the differences between the two dimensions, conservatism versus openness to change, were given especially by the inequality of gender and permissivity. In 1999 the differences become sharper. On one hand, to the two values already mentioned, authority and religiosity are added, and on the other hand, the support for the alternative family model and permissiveness are added. In 2005, the split of rural youngsters from urban youngsters in what concerns the type of value identification is maintained on the conservatism dimension referring to authority and religiosity, and it becomes total on the openness to change dimension. From a diachronic point of view (the last six columns of Table 2), in rural environments as well as in urban ones, the identification with the conservative values becomes more powerful in time or it remains at the same levels as in 1993. On the openness to change dimension the situation is somehow reversed, the identification tending to diminish or to remain the same as in 1993. If we look at the relation rural-urban dynamically, we could say that rural youngsters identify more intensively with conservative values, while urban youngsters identify especially with values regarding the openness to change. Metaphorically speaking we could imagine a train that runs, let’s say, from the West to the East (towards conservatism). On the way, some youngsters get off the train (those who have surpassed the age interval of 18-35 years), some others get in (those that are 18 years old). Those that remain on the train and those that get in, make the crowded areas in the train move to its margins. The change is not chaotic, but indicates a polarization: youngsters from rural areas start moving rather to the beginning of the train (towards conservatism), while those from the urban areas seem to rather prefer the last wagons of the train. Further, I will pass to the presentation of the results of the analyses made on categories of youngsters organized depending on the sex variable. Table 3 below synthesizes these results.

248 corectura 2 28 iulie 2008

Identity and axiological profile: value identifications for Romanian young people

Table 3. Youngsters’ value identifications depending on gender

Comparison Feminine versus Masculine Feminine masculine 2005 2005 2005 1999 2005 1999 1993 1999 2005 versus versus versus versus versus versus Value orientations 1993 1993 1999 1993 1999 1993 Authority È È ∼ ∼ ∼ ∼ ∼ ∼ ∼ Religiosity È È È È È ∼ ∼ ∼ ∼ Gender inequity Ê Ê ∼ ∼ ∼ ∼ ∼ ∼ ∼ Intolerance ∼ ∼ ∼ ∼ ∼ ∼ ∼ ∼ ∼ Unegalitarian society model Ê È Ê Ê È Ê ∼ ∼ ∼ Permissiveness Ê ∼ ∼ ∼ ∼ ∼ ∼ ∼ ∼ Alternative family model Ê ∼ ∼ ∼ ∼ ∼ ∼ ∼ ∼ Equity ∼ ∼ ∼ ∼ ∼ ∼ ∼ ∼ ∼ Notes: 1. The signs È or Ê reflect a significantly positive difference, respectively a significantly negative difference (p≤0,05) of the identification with the values enumerated in comparison with the year and the residence environment, the reference is made; 2. The sign ~ indicates the fact that there are no differences of identification among residential environments and there are no significant changes in comparison with the years and the reference years (the results are based on the ANOVA-type analysis; for the authority and inequality model the used test is Bonferron and for all the others is Tamhane). Example of interpreting: The first cell in the right indicates the fact that the preference for authority increases significantly among women in 2005 and 1993.

249

The Values of Romanians: 1993-2006. A sociological perspective

Although globally it seems that gender does not make evident any difference between women and men on the dimension regarding the openness to change, and only a few differences on the conservative dimension there are a group of value orientations that attract attention to them; that is religiosity, authority, and gender inequality. Synchronically analyzed, it is seen that women are more religious in 1999 and more oriented towards gender equality than men were in 1993 and 2005. Dynamically it can be seen that in 1999 and 2005 women are more religious and identify especially with the authority in comparison with the men` s situation and to their own situation in the precedent years. The analyses (ANOVA) made on age categories (18-23, 24-29, 30-35) did not reveal significant differences in the value identifications either synchronically or diachronically.

Conclusions The main question I have followed throughout the text was: „What is the identity profile of Romanian young people, in the axiological level?” In search of the answer I started from the idea of a theoretical definition of identity and the search for its analysis methodology. I showed that the best method of study is that of collective identifications. In a strict matter of speaking, the identitary profile was not investigated in its entirety, but only its axiological dimension. Thus, I was trying to show how these identifications are structured and what their dynamics are during the 1993-2005 period of time. I discovered, as was expected based on the contemporary theories of social axiology, that the values to which the youngsters adhere to are systematically grouped into two clusters: one indicating openness to change and the other showing a conservative attitude. The values that have been identified as belonging to the first cluster are: support for alternative family models, equity, permissiveness and a liberal social model, whereas the ones belonging to the second cluster are: religiosity, the idea of authority, intolerance and support for gender inequality. The synchronic analysis, in which I built a hierarchy based on items of the S. Schwartz model, has showed that in 2005, the top of youngsters’ preferences tends to be more conservative. In other words, I have noticed that young people identify more with values labeled as conservatory (tradition, conformity, security) rather than with those of openness to change (autonomy and stimulation). After a diachronic analysis of identifications with these categories of values, I realized that, generally speaking, a relatively constant growth can be observed (from 1993 to 1999 and from 1999 to 2005) in support of conservative values and a decrease in values related to openness to change. While for the differentiation between men and women, the collective-categorical interplay 250 corectura 2 28 iulie 2008

Identity and axiological profile: value identifications for Romanian young people

does not show great differences, for the rural-urban it shows a tendency of differentiation that is more and more pronounced concerning identifications, as young people in the rural area become significantly more conservative and less open to change over time. Thus, in the 1993-2005 dynamic, in the axiological profile of young people, the most significant characteristic that seems to be revealed along with the passage of time is the “conservative” trait (in which values like religiosity and authority play the main role) while the characteristic of “openness to change” is less pronounced and seems to keep becoming so. Although I did not intend to analyze all of the causes that determine these tendencies in the value identifications of young people, it is reasonable to assume that they are part of the more general picture of social changes that the Romanian society is experiencing20. If the post-revolutionary “euphoria” from the beginning of the 90s has left a strong mark on youngsters’ options towards change and liberalism, the reign of the transition that brought with it a growth in the social risks, a diminishing of perspectives, a decline in trust and hope and a heightening of pessimism (Rusu, & Bălăşoiu, 2005) makes these options weaker. It is still expected that, on an average term, as the Romanian society becomes more stable and modern, more integrated in the European complex, as the culture of mistrust (Sztompka, 1996, 1998) will vanish, young people (especially those in the urban areas) will reorient themselves and will identify more with the values that belong to the dimensions of openness, change and social liberalism.

20

For details concerning these changes related to various problems discussed by Romanian authors from different perspectives, see Sandu (1999); Mărginean (1999); Zamfir (1999); Bădescu (2002); Iluţ (2004) Voicu (2005a, 2005b); Voicu (2005); Comşa (2006); Vlăsceau (2007) etc. 251

The Values of Romanians: 1993-2006. A sociological perspective

Annex Part A – The used indexes in cluster and ANOVA analysis and in their way of construction

1. RELIGIOSITY The steps in the construction process of the index are as it follows: Step 1. The calculation of the index of God’s importance and religion: Communalities

Saturations

Variables

1993

1999

2005

1993

1999

2005

Believes in God

0,453

0,374

0,305

0,673

0,612

0,552

Religion is important

0,430

0,295

0,312

0,656

0,543

0,558

God is important

0,853

0,971

0, 790

0,923

0,985

0,889

Explained variation

58%

55%

47%

KMO

0,664

0,609

0,625

Correlation* with the Index ** “Religion is important”

0,997

0,984

1,000

Note:

* All of the correlations are significant for p=0,000 **The index is computed based on the saturation of the 2005 factor. The used method of extraction: Principal Axis Factoring. The used indexes are: Will you please tell us how important is religion in your life? (4 point scale); Do you believe in God? (Yes/No); How important is God in your life? (10 points scale). The cases of a refusal to the answer (I will not answer) and in the case of indecision (I do not know) were treated as missing values. Step 2. The computation of the index of Christian ethos Communalities

Saturations

Variables I believe in the existence of … life after death

1993

1999

2005

1993

1999

2005

0,568

0,506

0,508

0,754

0,711

0,713

I believe in the existence of … Hell

0,718

0,820

0,956

0,847

0,905

0,978

I believe in the existence of … Heaven

0,875

0,881

0,806

0,935

0,939

0,898

I believe in the existence of … Sin

0,294

0,337

0,359

0,542

0,580

0,599

The explained variation

61%

54%

66%

KMO

0,771

0,774

0,771

0,987

1,000

The correlation* Christian ethos

with

the

Note:

Index**

0,984

* All of the correlations are significant for p=0,000 **The index is computed based on the saturation of the 2005 factor. The used method of extraction: Principal Axis Factoring. The four analyzed items are dichotomist (the possible answers are Yes/No). The cases of a refusal to answer (I will not answer that) and in the case of indecision (I do not know) were treated as missing values. 252 corectura 2 28 iulie 2008

Identity and axiological profile: value identifications for Romanian young people Step 3. The calculation of the index of the importance of the Church on the account of the “universal” explanations it gives Communalities Variables The Church offers answers to moral problems The Church offers answers to the problems of family life The Church offers answers to social problems The explained variation

Saturations

1993

1999

2005

1993

1999

2005

0,389

0,550

0,542

0,624

0,742

0,736

0,756

0,747

0,804

0,869

0,864

0,897

0,257

0,274

0,395

0,507

0,523

0,629

47%

52%

58%

KMO

0,627

0,643

0,673

The correlation* with the Index** Religion is important

0,947

0,968

1,000

Note:

* All of the correlations are significant for p=0,000 **The index is computed based on the saturation of the 2005 factor. The used method of extraction: Principal Axis Factoring. The four analyzed items are dichotomist (the possible answers are Yes/No). The cases of a refusal to answer (I will not answer) and in the case of indecision (I do not know) were treated as missing values. Step 4. The calculation of the index of Religiosity Communalities

Saturations

1993

1999

2005

1993

1999

2005

Index – the Church, a Universal solution

0,276

0,368

229

0,525

0,607

0,479

Index – Christian ethos

0,465

0,412

0,332

0,682

0,642

0,576

Index – religion is important

0,691

0,786

0,561

0,831

0,887

0,749

the self perception of religiosity

0,435

0,399

0,359

0,660

0,632

0,599

religious behavior

0,421

0,312

0,267

0,649

0,559

0,517

The explained variation

46%

46%

35%

KMO

0, 817

0, 768

0, 765

The correlation* with the Index** of Religiosity

0,985

0,995

1,000

Variables

Note:

* All of the correlations are significant for p=0,000 **The index is computed based on the saturation of the 2005 factor. The used method of extraction: Principal Axis Factoring. The self-perception of religiosity is measured based on answers to the question: Independent of the fact that you go to Church or not, would you say you are: a religious/non religious person/an atheist (in the analysis, the answers were transformed into dummy variables where religious person received the value of 1 and the answers non-religious person and atheist were given the value of 0). Religious behavior is decided by the answer to the question: Do you take some moments of prayer, meditation or contemplation or something like that? (with the possible answers of Yes/No). The cases of a refusal to answer (I will not answer) and in the case of indecision (I do not know) were treated as missing values. 253

The Values of Romanians: 1993-2006. A sociological perspective

2. PERMISIVITY Variables

Communalities

Saturations

1993

1999

2005

1993

1999

2005

how justified is homosexuality

0,283

0,403

0,461

0,532

0,635

0,679

how justified is prostitution

0,383

0,383

0,586

0,619

0,619

0,766

how justified is abortion

0,546

0,519

0,565

0,739

0,720

0,751

how justified is divorce

0,486

0,657

0,583

0,697

0,810

0,763

how justified is euthanasia

0,255

0,337

0,453

0,505

0,580

0,673

The explained variation

39%

46%

53%

KMO

0, 676

0, 803

0, 771

The correlation* with the Index** of Permisivity

0,990

0,989

1,000

Note:

* All of the correlations are significant for p=0,000 **The index is computed based on the saturation of the 2005 factor. The used method of extraction: Principal Axis Factoring. The answers to the used items were measured on a scale of 1 to 10, where 1 means not at all justified and 10 means totally justified. The cases of a refusal to answer (I will not answer) and in the case of indecision (I do not know) were treated as missing values.

3. INTOLERANCE Communalities

Saturations

Variables

1993

1999

2005

1993

1999

2005

does not accept drug addicts as neighbors

0,650

0,689

0,577

0,806

0,830

0,760

does not accept homosexuals as neighbors

0,480

0,509

0,267

0,693

0,713

0,516

does not accept heavy drinkers as neighbors

0,308

0,315

0,439

0,555

0,562

0,662

The explained variation

48%

50%

43%

KMO

0,657

0,659

0,644

The correlation* with the Index** of Intolerance

0,984

0,983

1,000

Note:

* All of the correlations are significant for p=0,000 **The index is computed based on the saturation of the 2005 factor. The used method of extraction: Principal Axis Factoring. The three analyzed items are dichotomist (the possible answers are Mentioned/Unmentioned). The cases of a refusal to answer (I will not answer) and in the case of indecision (I do not know) were treated as missing values.

The name of the index is given by the fact that most of the young people (as the following table shows) are against the acceptance of these categories of individuals as their neighbors.

254 corectura 2 28 iulie 2008

Identity and axiological profile: value identifications for Romanian young people 1993

1999 Accepted

Accepted

66%

34%

73%

27%

33%

56%

44%

52%

48%

26%

75%

25%

66%

34%

Accepted

69%

31%

Homosexuals

67%

Heavy drinkers

74%

Drug addicts

2005 Not accepted

Not accepted

Not accepted

4. AUTHORITY The individuals in the sample group were asked to specify which are the most important things children could learn at home. A maximum of five choices were allowed, from a list including among others: Independence, Feeling of responsibility, Imagination, Perseverance, Hard work, Thrift, saving money and things, Religious faith and Obedience. Each person received a point if choosing any of the four attributes that were not underlined and were deducted a point if choosing any of the first four attributes (the underlined ones).

5. ALTERNATIVE FAMILY MODEL This orientation is measured based on the acceptance of the idea that women can be successful single parents. The used question was: if a woman wants to have a child as a single parent, but she doesn’t want to have a stable relationship with a man, would you approve or disapprove that she should have the child? The possible answers were: I approve, I disapprove and it depends. The variable was later made dichotomist as it follows: 1 = I approve, 0 = I disapprove; it depends answers as well as indecision and the refusal to answer were considered missing values. More than half of the respondents (74% in 1993, 75% in 1999, respectively 62% in 2005) declare that they agree with the idea that a woman can raise her child alone, outside of a traditional family model.

6. GENDER INEQUITY (in the area of work) This orientation is measured based on the approval/disapproval of the affirmation: “When jobs are scarce…men should have more right to a job than women” (scale of 3 points). The variable was then encoded to the dichotomist 1 = I agree, 0 = I disagree; it does not matter answers as well as indecision and the refusal to answer were considered missing values. More than half of the respondents (61% in 1993, 61% in 1999, respectively 61% in 2005) are against gender equality in the workplace.

7. EQUITY (in the workplace) This is measured based on the accord with the difference in the wages of two individuals with the same organizational status, but of different role performances. The used question is: Imagine two secretaries of the same age, 255

The Values of Romanians: 1993-2006. A sociological perspective

doing practically the same job. One finds out the other earns much more than she does. Still, the better paid secretary is faster, more efficient and more reliable in her job. In your opinion is it fair or not that a secretary earn more than the other? (Possible answers: fair/unfair)

8. THE MODEL OF EGALITARIAN/LIBERAL SOCIETY This is measured based on the accord with the idea of equality of income. The used question is: What is your opinion related to the following affirmations, using a scale of 1 to 10, 1 meaning that you agree entirely with the affirmation in the left and 10 meaning that you agree entirely with the affirmation in the right. You may, of course chose any intermediary number to nuance your response:

1

2

3

4

5

Incomes should be made more equal

6

7

8

9

10

There should be larger income differences as incentives for individual effort

PART B – The items in the Schwartz scale The subjects were asked to answer the following set of questions: To what extent do you act like to a person for whom the following The indicator used are important... in the text tradition is important, to follow the customs handed down by tradition one’s religion or family. living in secure surroundings; to avoid anything that might be security dangerous. to always behave as we should; to avoid doing anything conformity people would say is wrong to think up new ideas and be creative; to do things one’s own autonomy way; adventure and taking risks; to have an exciting life. stimulation to be rich; to have a lot of money and expensive things. power to have a good time; to look after themselves. hedonism to help those around them; to enjoy looking after their well-being. benevolence to be successful; to be recognized by other people.

self-fulfillment

to be concerned with the environment and to protect nature.

universalism

The possible answers are scaled from 1 to 6, where 1 means “entirely” and 6 means “not at all”. The first 5 used items (bolded items) were included in our analysis as well. Each item determines the measure of the subjects’ identification with an imaginary person. 256 corectura 2 28 iulie 2008

Identity and axiological profile: value identifications for Romanian young people

References Allport, Gordon. (1954) The nature of prejudice, Reading, MA: Addison-Wesley. Arts, Wil. Hagenaars, Jacques & Halman, Loek. (eds.) (2003). The Cultural Diversity of European Unity. Findings, Explanations and Reflections from the European Values Study, Leiden: Brill Anderson, Benedict. (1991) Imagined communities: Reflections on the Origins and Spread of Nationalism, London: Verso. Barth, Fredrik. (1981) Ethnic group and boundaries. in Kuper, Adam (ed.) Selected essays of Fredrik Barth, Vol. I: Process and form in social life. London: Routledge and Kegan Paul. Bacova, Viera. (1998) The construction of national identity – on primordialism and instrumentalism. Human Affairs, 8:1, pp. 29-43. Bauman, Zygmunt. (1992) Soil, Blood, and Identity, Sociological Review, 40:4, pp. 675– 701; Bădescu, Ilie. (2002) Noologia. Bucharest: Valahia Publishing House. Bell, Duncan S. A. (2003) Mythscapes: memory, mythology, and national Identity, in British Journal of Sociology. 54:1, pp. 63-81. Berger, Peter L. and Thomas, Luckmann. (1966) The social construction of reality. New York: Doubleday & Co. Bourdieu, Pierre. (1986) Economia bunurilor simbolice, Bucharest: Meridiane Publishing House. Brass, Paul. (1991) Ethnicity and Nationalism – theory and comparison. New Delhi [etc]: Sage Publications. Brubaker Rogers. and Cooper, Frederick. (2000) Beyond “identity”. in Theory and Society, No. 29:1, pp. 1-47. Brubaker, Rogers. Mara, Loveman and Peter Stamatov. (2004) Ethnicity as Cognition. Theory and Society. 33:1, pp. 31-64 Calhoun, Craig. (1997) Nationalism. Minneapolis: University of Minnesota Press. Calhoun, Craig. (1994) Social Theory and the Politics of Identity, in Calhoun, C. (ed.) Social Theory and the Politics of Identity. Oxford:Blackwell. Cerulo, Karen. A. (1997) Identity construction: new issues, new directions. Annual Review of Sociology. 23, pp. 385-409 Cooley, Charles H. (1909) Social organization. New York: Charles Scribner's Sons. Connor, Walker. (1994) Ethno-nationalism: The quest for understanding. Princeton, N.J.: Princeton University Press. Connor, Walker. (2004) The timelessness of nations. in Nations and Nationalism 10:1/2, pp. 35–47. Comşa, Mircea. (2006) Stiluri de viaţă în România după 1989. Cluj-Napoca: Cluj University Press. Durkheim, Emile. (1964) The division of labor in society. New York: Free Press. [1895] Durkheim. Emile. (1983) Le suicide. Paris, P.U.F. [1897] Dungaciu, Dan. (2004) Naţiunea şi provocările (post)modernităţii. Bucharest: Tritonic Publishing House. Eller, Jack. D. and Coughlan Reed. M. (1993) The poverty of primordialism: the demystification of ethnic. in Ethnic and Racial Studies, 16:2, pp.183-202. Erikson, Erik. (1968) Identity: Youth and Crisis, New York:Horton. 257

The Values of Romanians: 1993-2006. A sociological perspective Ester, Peter. Braun, Michael & Mohler Peter. (2006) Globalization, Value Change and Generations. Boston:Brill Garfinkel, Harold (1967): Studies in ethnomethodology. Englewood Cliffs: Prentice Hall. Giddens, Anthony (1979) Central problems in social theory. London: Macmillian. Giddens, Anthony (1991) Modernity and Self-identity: Self and Society in the Late Modern Age, Cambridge: Polity Press. Gellner, Ernest. (1983) Nations and Nationalism. Oxford:Basil Blackwell Publisher. Geertz, Clifford. (1975) The Interpretation of Cultures. Selected Essays. New York: Basic Books. Geertz, Clifford. (1963) The integrative revolution, in C. Geertz, Old Societies and New States. New York: Free Press. Grosby, Steven. (1994) The verdict of history: the inexpungeable tie of primordiality – a response to Eller and Coughlan. in Ethnic and Racial Studies. 17:1, pp. 164-71. Hall, Stuart. (1991) Old and New Identities, in King, A. D.(ed.) Culture, globalisation and the world system. London: Macmillan. Halman, Loek & Pettersson Thorleif (2006) A decline of religious values? in Ester, Peter. Braun, Michael & Mohler Peter. Globalization, Value Change and Generations. Boston:Brill pp. 31-60. Hay, Colin. (2002) Political Analysis. Houndmills: Palgrave. Hofstede, Geert. (2001) Culture’s consequences. London:Sage. Horrowitz, Donald L. (1975) Ethnic identity. in Glazer, Nathan and Moynihan, Daniel P., Ethnicity. Theory and Experience. Cambridge: Harvard University Press. pp.111-140 Inglehart, Ronald, (1997). Modernization and Post-Modernization. Cultural, Economic and Political Change in 43 Societies. Princeton: Princeton University Press. Iluţ, Petru. (2004) Valori, atitudini şi comportamente sociale, Iasi: Polirom Publishing House. Immerman, Richard H. (1982) The CIA in Guatemala: The Foreign Policy of Intervention. Austin: University of Texas Press. Isaac, Harold. (1975) Basig group identity. In Glazer, Nathan and Moynihan, Daniel P. (eds.) Ethnicity. Theory and Experience. Cambridge: Harvard University Press, pp.29-52. Jenkins, Richard. (1996) Social Identity, London, Routledge Kellner, Douglas. (1992) Constructing postmodern identities. In Modernity and Identiy. Lash, Scott&Friedman Jonathan (eds). Oxford [etc.]: Blackwell Publishers. Kluckhohn, Clyde. (1951) Values and Value Orientations in the Theory of Action, in T. Parsons and E. A. Shils (Eds.) Toward a general theory of action Cambridge: Harvard University Press. pp. 388-433. Mărginean, Ioan (1999) Suportul social pentru democraţie [The social support for democracy] în Sociologie Românească., 2, pp.3-18. Maslow, Abraham (1943) A Theory of Human Motivation in Psychological Review. 50:4, pp. 370-96. Mitchell, James Clyde (1974) Perceptions of ethnicity and ethnic behavior: an empirical exploration. in Cohen, Abner (ed.) Urban Ethnicity, London: Tavistock Publications. pp. 1-35 Ozkirimli, Umut. (2000) Theories of Nationalism: A Critical Introduction. New York: St Martin’s Press. Rokkan, Stein and Derek Urwin. 258 corectura 2 28 iulie 2008

Identity and axiological profile: value identifications for Romanian young people Parsons, Talcott and Shills, Edward. (1992) Values and social systems. in Alexander, Jeffery C. and Seidman Steven. [eds.] Culture and Society. Cambridge: Cambridge University Press. Ramos, Alice (2006) Social values dynamics and socio-economic development. In Portuguese Journal of Social Science, 5:1, pp.35-64 Rother, Nina & Medrano, Juan Diez. (2006) Is the West becoming more tolerant? in Ester, Peter. Braun, Michael & Mohler Peter. Globalization, Value Change and Generations. Boston:Brill, pp. 151-178. Rokeach, Milton (1973) The Nature of Human Values. New York: The Free Press Rughiniş, Cosima (2006) Religiozitate şi scepticism religios. in D. Sandu (coord.) Viaţa socială in România urbană. Iasi:Polirom Publishing House, pp.123-132. Rusu, Horaţiu şi Bălăşoiu, Raluca. (2005) The pessimistic society. in Rusu, Horatiu & Voicu, Bogdan (eds). Perspectives on the european postcommunist societies. Sibiu: Psihomedia. Sandu, Dumitru. (1999) Spaţiul social al tranziţiei. Iasi: Polirom Publishing House. Sayer, Andrew (1997) Essentialism, social constructionism and beyond. The Sociological Review. 45:3, pp. 453–87 Shils, Edward (1957) Primordial, personal, sacred and civil ties: some particular observations on the relationships of sociological research and theory. In The British Journal of Sociology. 8:2, pp. 130 – 45. Smith, A. D. (1991) National identity. London, Penguin Books. Smith, Anthony. D. (1999) Myths and memories of the nation.Oxford: Oxford University Press. Thomas Willliam.I. &. Thomas Dorothy.S (1928). The Child in America: Behavior Problems and Programs. New York: Alfred A. Knopf. Tilly, Charles. (1996) Citizenship, Identity and Social History, in Tilly C. (ed.) Citizenship, Identity and Social History, Cambridge: Cambridge University Press, pp. 1– 17 Van den Berghe, Pierre L.(1995) Does Race Matter? in Nations and Nationalism. 1:3, pp. 356-68 Vinken, Henk, Soeters Joseph & Ester Peter (2004) Comparing cultures: dimensions of culture in a comparative perspective. Leiden: Koninklijke Brill NV. Voicu, Bogdan. (2005a) Penuria Pseudo-Modernă a Postcomunismului Românesc. Volumul I. Schimbarea socială şi acţiunile indivizilor, Iasi: Expert Projects. Voicu, Bogdan. (2005b) Penuria Pseudo-Modernă a Postcomunismului Românesc. Volumul II. Resursele, Iasi: Expert Projects. Voicu, Mălina (2005) Ce fel de bunăstare îşi doresc românii? Despre legitimitatea politicilor sociale in România. Iasi: Expert Projects. Voicu, Mălina (2006) Valori and comportamente religioase in spaţiul urban românesc: o abordare longitudinală. in Sandu D. (coord.) Viaţa socială in România urbană. Iasi: Polirom. pp.103-119. Vlăsceanu, Lazăr (2007) Sociologie şi modernitate: tranziţii spre modernitatea reflexivă. Iasi: Polirom Publishing House. Schwartz, Shalom H. (1992) Universals in the content and structure of values: Theory and empirical tests in 20 countries. In M. Zanna (Ed.), Advances in experimental social psychology (Vol. 25) New York: Academic Press. pp. 1-65.

259

The Values of Romanians: 1993-2006. A sociological perspective Schwartz, Shalom H. (2006) Basic Human Values: Theory, Methods, and Applications. http://dpms.csd.auth.gr/emplak/Schwartzpaper.pdf accessed on the 19th of February, 2006. Smith, Anthony. D. (1991) National identity. London: Penguin Books. Sztompka, Piotr. (1996) Trust and emerging Democracy. International Sociology 11:1, 37 – 62. Sztompka, Piotr. (1998) Trust, Distrust and Two Paradoxes of Democracy. European Journal of Social Theory. 1:1, pp.19- 32. Weber, Max. (1968) Economy and society. New York: Bedminster Press. Weber, Max [1905] (1994) Etica protestantă şi spiritul capitalismului. Bucharest: Humanitas Publishing House. Woodward, Kathryn. (1997) Identity and difference. London: Sage. Zamfir, Cătălin. (ed) (1999) Politici sociale in România: 1990 - 1998. Bucharest: Expert Publishing House.

260 corectura 2 28 iulie 2008

Between tradition and postmodernity? a dynamic of value orientations in Romania: 1993-20051 BOGDAN VOICU

The radical change in the social structure, which started in 1989, opened the question of the need to change the patterns of thinking and behaving, as a prerequisite of the new social order. For many persons, the mere institutional change, with its two main features – marketization and democratization – seemed sufficient. As I showed somewhere else (B. Voicu, 2005a: 43-45), in the public space as well as that in the academic world, the debates initially focused mainly on these two transitions, referring to the political and the economical organization2. These two elements actually pointed out three main changes: the political sphere concerned both the internal reorganization – i.e. democratization – and the reorganization of the foreign policy. The fourth domain of change became visible a little later, and included the transformations of the social structure. Even later, at the end of the 1990s, the transition literature started to include some more references to the changes in value orientations3. This chapter intends to describe the dynamics of social values within a few key domains, with the aim to analyse if the Romanian society, as a whole, 1

Parts of this work were carried out at the European Data Laboratory for Comparative Social Research (ZA-EUROLAB). Access to the ZA-EUROLAB was supported by the European Community under the “Structuring the European Research Area” specific programme Research Infrastructures Action in the 6th Framework Programme, (project 026142 RITA, ZA-EUROLAB 01-06-MV). 2 On the duality of the transition, see the article by Claudiu Tufiş about the support for democracy and market economy, included in this volume. 3 Such approaches existed starting with the early 90s (see, for instance, Sztompka, 1993), but, as compared to the whole transitology literature, represented rather isolated manifestations until the end of the respective decade. 261

The Values of Romanians: 1993-2006. A sociological perspective

changed or not towards cultural modernity during the postcommunist period up to the present days. I do not intend to go into a complex analysis, but only to make a succinct presentation of the main tendencies. The general background in which the discussion is built is that of the value modernization and postmodernization processes. I have dedicated a short presentation to these one in the first section of the chapter. Then I focus on certain domains (work, family, gender, tolerance, religiosity, attitudes towards democracy etc.), discussing the dynamics of Romanians’ value orientations after 1990 from a comparative perspective. The other European countries are the referential. However, I avoid the global comparison with the “European average”, which is not meaningful in many of the analyzed areas, given the fact that the cultural diversity in the Old Continent can be very high. I rather prefer to place Romania in relation to the major tendencies observed in regions where more homogenous cultural patterns can be noticed: Northern Europe, Western Europe, Southern Europe and the excommunist area. Sometimes, I include in the analysis all the European countries, using graphs and maps to exemplify the differences across Europe, and the evolution of Romania through this mosaic. Other times, being constrained by the available data4, I compare Romania with typical societies for these social frameworks: Sweden for the North, Italy for the South, Germany for Continental Western Europe and Poland for the ex-communist space. Firstly, I discuss the observed tendencies with regard to the attitudes towards the environment, the ethos of labor, religious attitudes and the attitude towards democracy, family, gender equality, tolerance, etc. For some of these domains, the present volume already includes extended analysis. In these cases, I preferred to restrain the presentation to just a few general conclusions, focusing rather on the areas not covered in the other chapters. The order in which I approach the domains is somewhat random, not following a particular rule, as my preference in this respect is to present the apparently unrelated images of the different fields of daily social reality, and in the end, to further merge these pieces in the last section, which is dedicated to the empirical analyses of modernity as a whole. In this last section, I built a complex index regarding modernity and I describe Romania’s position in Europe with relation to this synthesis index. I also sketch some integrative explanations for the changes that I notice for each of the analyzed domains. Additionally, I search for an answer for the question of homogeneity in Romania, trying to identify those social groups that differ to a large extent from the average, individualizing themselves as distinct social entities. In the end, I exploit some of the results of the literature dedicated to the theme of transition, trying to speculate on the tendencies the Romanian society 4

At the time of writing, the aggregate WVS 2005-2006 data set have not yet included all the European countries, but only a few.

262 corectura 2 28 iulie 2008

Between tradition and postmodernity?

may undergo on an average run, with respect to the evolution of value orientations towards modernity, traditionalism or post-modernity.

On modernization and post-modernization The theories of modernization and post-modernization represent the main theoretical framework that I employ for explaining the tendencies that are to be presented in this chapter. Far from intending to be exhaustive, I summarize some of the features of these theories. The existing literature abounds with syntheses and seminal woks, both with respect to the modernization process (Sztompka, 1993; Kumar, 1999; Inglehart, 1997; Hapfercamp and Smelser, eds., 1992; Hall, 1996, etc.), as well as for the postmodernization one (Giddens, 1990; Beck, 1992; Kumar, 1995; Lash, 1990; Inglehart, 1997, 2000; Beck, Giddens, Lash, 1992; etc.). In the last few years, Romanian sociologists have also produced many works in which the theories of modernization and post-modernization are presented (Roth, 2002; Sandu, 1996; Vlăsceanu, 2001, 2007; B. Voicu, 1999, 2001, 2005b; Chiribucă, 2004; etc. ) In its common meaning, modernization is understood through its visible processes and it is associated with technical progress, with industrialization and urbanization. However, the process of social change that is presently named modernization is much more profound. Industrialization, urbanization and the development of infrastructure are only a few of its surface manifestations, reflecting a more general tendency towards a rationalization of the way of life, of thought and of the relation with a natural and social environment, about which the social representations are beginning to be looked at as controllable. The technological and scientific progresses are factors that have been the catalysts of this process of rationalization. They have contributed to providing a certain certainty to satisfying the basic needs (food, health, personal and social security, etc.) that have freed both individuals and collectivities from these needs, allowing them to focus more on improving the social organization, and later (in post-modernity), on the individual development and self-achievement. The spreading of the pragmatic usage of scientific knowledge was, in its turn, conditioned by the presence of a culture that would allow rational, secular approaches of the natural phenomena in the detriment of more traditionalist explanations. The failure of Galileo Galilei or Giordano Bruno in receiving social recognition during their lifetimes is eloquent in this sense. For many projects that Leonardo da Vinci had, the process of going from theory to practice came far after his time, largely not on account of lacking material or financial resources but mostly because of the reluctance to following new paths and to develop new ways of doing things. Building mainly on the works of Weber (1978 [1922], 1995 [1920]), Parsons (1964, 1971) and Inkeles (1996 [1968], 1974), social sciences have created a relatively consensual image with regard to the meaning of 263

The Values of Romanians: 1993-2006. A sociological perspective

modernization. Modern societies differ from the traditional ones by the preeminence of the rationality as compared to the traditional-religious explanation. Traditionally, objects and facts are the way they are because that is the “way that things are”, “this is how gods decided”, “this was the fate”, “things are like they have always been” or because “this is what the local lord decided”. In modernity, societies look for explanations based on rational knowledge rather than on the supernatural explanation or random decision of an individual or a group of individuals invested with some inherited authority. Ascribed authority is therefore replaced by the epistemic, rational knowledge. Voting as a manifestation of the people’s will, the development of bureaucracy as an expert system in the administration of the city are rational ways of social organization. To the individual level, the manifestations of modernity are openness to change, autonomy, abandonment of the passive attitude, and planning, and participation in city life. In opposition to the traditional patterns, the modern individual is less normative, accepts the existence of some other different people (with regard to ethnicity, race and, more moderately, to lifestyle), is more willing to risk trying new things, sets up long term objectives, renouncing to live only in the present, accepts and begins to develop interactions outside his/her primary groups, is at least interested in the finality of the decisional processes, if not in the way of structuring the political process itself. Risk is a key element of this change. Traditional societies are affected by material incertitude: the basic needs are met with great difficulty, as the efforts of everyday life are focused mainly on lowering this kind of uncertainty. People and collectivities cannot afford to ask questions about the meaning of life or their ability to obtain knowledge; they need a world that is as predictable as it can possibly be, a world that would allow them to survive with small resources at their disposal. With the growth of material security, individuals and collectivities can afford to give up the safety of the traditional explanation which fully describes the world, making a choice for an incomplete explanation, based on rational knowledge. The migration towards the cities, as well as the development of transport, tourism and international migration of the workforce leads to a growth in the frequency of interactions with different individuals with less predictable behavior, a fact that determines another growth of the axiological uncertainty regarding the actions of the people in the immediate social environment.

264 corectura 2 28 iulie 2008

Between tradition and postmodernity? Figure 1. Social change and the evolution of risks Axiological uncertainty

T

M

r

ivi ex efl

ty

Tolerance diversity

Rationalization secularization

Di st r u c a st in pa c it the y o ex f r pla el n a igi to on ry

Religion; supernatural forces

The level of “not knowing anything”

PM • globalization • dev. of technology • increasing prosperity

• conquest of ext. markets • dev. of technology • welfare services

Material security

Note: The figure is reproduced from B. Voicu, 2005b, p. 166.

Several processes mark an additional growth of the axiological uncertainty. Modernity is still very much related to normativeness. Science, holding absolute truth, is a science which does not admit alternatives. However, gradually this also starts to change. A considerable number of disciplines, led by social sciences and medicine as well, or the more traditional exact sciences like physics, biology or astronomy have revealed to the public the diversity in explaining paradigms that govern them. Alternative explanations on subjects like the growth of life on Earth, the evolution of species and the causes and treatments of diseases, the evolution of the Cosmos and planets, etc. is today the substance of the main magazines which popularize science. The social acceptance of the multi-paradigmatic character of the scientific explanation is part of a broader process of accepting diversity as a social norm (at the beginning of the third millennium, political correctness implies generalized tolerance) as well as reflexive analysis, permanently questioning of any type of knowledge and the consequences of technological progress. The debate over the consequences of the development for the environment would not have been possible without concern for long-term planning (that appeared in the modern period), without giving up normativeness, and without the large growth in tolerance not only with regard to people’s diversity, but also in different ways of living. All of these processes have lead to a new type of social equilibrium, labeled in several different ways: late modernity, post-modernity, post-material society, reflexive modernity, risk society, postindustrial society, etc. This last label – postindustrial society – derives from the new structure of the labor 265

The Values of Romanians: 1993-2006. A sociological perspective

market. If agriculture used to dominate the traditional society and the industry was preponderant in the modern period, in late modernity, the technological progress highly increases work productivity, and this, in conjunction with the relative wealth and the extremely diversified individual needs 5, transforms the service sector in the main employer. Once the burden of the basic needs is eliminated, with more free time at their disposal (due to the same high productivity, in conjunction with the life expectancy on the rise), individuals start to be more and more motivated by the impulse to satisfy their superior needs. The need for self-expression and selfaffirmation manifests through the pursuit of knowledge, through a higher focus on leisure and leisure quality, through experimenting new roles and new situations, through hedonism, etc. Post-modernization is far from being a completely different process than modernization. It rather represents a natural continuation of the modern processes. Some of the tendencies seen in modernization (the growth in the levels of tolerance, rationalization, etc) further develop in late modernity. Other transformations are new, marking deviation from the initial path of social change. Figure 1 symbolically suggests this. This presentation of the processes of modernization and postmodernization is nevertheless reductionist. The extent and the objective of this material (the comparative description of the dynamic of values in Romania) determines me to describe only these general tendencies, without going into details, without mentioning different criticisms and variations of the theory. More, the above evolutionist story is not valid as such for any society. It broadly describes the main trajectories of European societies, particularly the Western ones. The Eastern societies have followed similar paths, with the syncope of the communist experiment. Communism, although a modern project by many up of its facets and goals, has lead to a rather fake modernity or pseudo-modernity. Sztompka (1993: 137)6 describes the respective society as a product of the top-down modernization process, manifested however only in certain areas of social life, and larded with many traditionalist relics, some of which (like authoritarianism, the lack of autonomy, dependence) were imposed by the modernization instance (the state, the communist elite) and were adorned with symbols that, in fact, were just imitations of the Western modernity (like the institution of voting with regards to political elections7). Social values were

5

The high productivity of work in the industrial and agriculture areas reduces the need for workforce in these sectors. The diversification of needs is associated with a high demand for different services (see next paragraphs), stimulating the massive employment in the service sector. 6 For similar argumentations, see Winiecki, 1988; Voicu, 2001, 2005b; Chiribucă, 2004; etc. 7 In communist Romania for example, elections were organized every 4 years, but the real results of the vote were never shared with the public. The official handouts always 266 corectura 2 28 iulie 2008

Between tradition and postmodernity?

in part modern, and in part pre-modern. The contemporary process of transformation of the ex-communist European space has been labeled by Sztompka as neo-modernization. The goal is to get closer to Western postmodernity, in a kind of convergent transformation of one of the facets regarding the multiple modernities mentioned in the introduction of this volume. The modern-traditional opposition is the one relevant in contemporary Romania too (Voicu, 2005b; Sandu, 1996), even though some incipient tendencies towards a reflexive modernity are present as well (Vlăsceanu, 2007; Voicu, 2005b). On the other hand, the periods of social instability and those of economic recession generate the regrouping of the value orientations towards materialism and traditional values (Inglehart, 1997; Inglehart, Baker, 2000). This is the case of post-communist societies, whose neo-modern transition is in fact confronted with the refocus on traditional values. In the sections that follow, the discussion is centered on the dynamics of the value orientations of Romanians in a few areas of social life, between 1993 and 2005, in comparison with other European societies. The goal is purely descriptive and the explanations do not constitute anything else but context elements. The final objective is to describe the dynamics of Romania on the traditional-modern cultural axis in the period beginning with the fall of the communist regime and ending with the integration in the European Union. I treat Romania as a whole, and only towards the end of the chapter do I discuss some of the existing differences between social groups. The selection of the analyzed domains was determined by the available information from the EVS/WVS questionnaires. I briefly discuss the value orientations regarding the protection of the environment, tolerance and normativism, work ethos, religion and the preference for authority or autonomy, for democracy or authoritarianism, family, gender relations, and post-materialism. The order in which I discuss these domains is random, as they are complementary to each other and of equal importance, relating to an extremely complex social reality. Their selection was not random, but determined by relevance and data availability: I preferred those domains for which comparative information is mandatory available for Romania as well as for most of the European countries, for at least two of the three analyzed timeframes (1990-1993, 1999-2001, respectively 2005-2006).

indicated the expected winners. In all probability, the votes were not even counted or centralized. 267

The Values of Romanians: 1993-2006. A sociological perspective

Attitudes towards environment protection The reflexivity of the late modernity brings an increasing interest for the effects of human activities on the natural environment. In principle, the industrial production brings a higher comfort level and higher quality of life for human beings. On the other hand, industrial processes are harmful for the environment. Their side effect is pollution, associated to the production process, but also to the non-biodegradable waste products, a typical acquisition of the industrial modern era. The traditional society did not produce almost any rubbish: there were no PET bottles, no plastic bags, soda cans, tin foil, broken plastic toys or worn out polyester clothes. Glass, paper, bricks, food waste are more or less “natural”, in the sense that they can become, through a process of natural degradation, a part of the natural environment from which they were created. The concerns regarding the protection of the environment are completely lacking from the traditional society, which has rarely been confronted with these risks, but at the same time has been involved in the struggle for the survival of the human species. The modern societies have explored the existing technical potential to the maximum to solve their needs of security, food or comfort. Only the postmodern reflexivity brings ecological concerns to the foreground. Today, almost everywhere in Europe, but mainly in the Western countries, one of the dilemmas that animate the public (and ideological) debate opposes economical growth to the need of protection for the environment. In the last few decades, the world’s consensus over the need to ensure the durability of the development process has been stimulated by these kinds of concerns. Looking at the EVS/WVS wave of 1999-2001, Romania found its place among the European countries whose citizens were the least concerned with environmental protection. For example, the respondents of the respective survey were asked about their approval or disapproval of the affirmation: “I would give part of my income if I were certain that the money would be used to prevent environmental pollution”. The percent of Romanians who answered yes to this question8 (44%) is similar to the percentage obtained in France, Great Britain, Estonia, Austria and Hungary, and higher than the percentages obtained in Germany (32%), which still was under the pressure of the need to preserve jobs at any cost in order to sustain the price of the Eastern integration, Lithuania (23%) – under the influences of the discussions over the closing of the Ignalina nuclear power plant, one of the main employers for the Russian speaking population, and Turkey (19%) – still a highly traditional country. The rest of the

8

They agreed “in a very large extent” or “in a large extent” with the question (on a scale from 1 to 4).

268 corectura 2 28 iulie 2008

Between tradition and postmodernity?

European countries displayed a higher support with regard to this question, up to 80% in the Northern countries, the ex-Yugoslav states, Holland and Greece. The analysis of an aggregate index tapping for the support for the environment pollution9 confirms Romania’s presence in the group of the countries with low interest in ecological issues. One should also note that, between 1991 and 1999, all of the European countries for which there is available data, experienced falls in the levels of the respective index, with a sharper decrease in the post-communist societies. It is likely that Romania (for which there are no such data available for the 1993 wave) knew a similar evolution. The data from the 2005 wave indicate the same tendencies: a drop of the index values for Romania, Poland, Slovenia and Sweden, and stability in Finland and Italy. The decrease in support for the protection of the environment is connected to two types of explanations. The first is related to the immediate impact of the ecologist measures, to the closing of some polluting industrial units, to the growth in production costs (with consequences over final prices and the slowing of wage increases). The second one, in Romania and in the rest of the post-communist area, is connected to economic recession and, as we have shown, to the refocusing of values in the traditionalist direction. This process is taking place in a society in which the attention is anyway focused on the need for growth, for reducing the development gap that separates it from the western societies, a society were people are mainly concerned with the satisfaction of basic needs, a society that tends anyway to ignore any orientations towards the higher needs.

Between tolerance and normativism Tolerance, accepting the others, is a trait specific to post-modernity. Knowledge leads to understanding, and alterity does not represent a risk any more. Opposed to this, traditionalism and modernity are normative, hardly accepting the attitudes and behaviors substantially different from those of the majority. The value’s surveys include a few items that allow a longitudinal analysis in the evolution of tolerance. In the beginning of the 1990s, Romania was one of the European countries that were characterized by a low level of

9

The index was calculated as a factor score, using the presented variable and two more 4-point scales, including the following: “I would agree to an increase in taxes if the extra money is used to prevent environmental pollution” and “The Government has to reduce environmental pollution but it should not cost me any money”. I have tested (using Amos 4), the stability of the factor structure on the data from the 1990-1993, 1999-2002, 2005-2006 EVS/WVS waves. 269

The Values of Romanians: 1993-2006. A sociological perspective

tolerance. Like most of the ex-communist countries (except for Slovenia10 and Poland11), in 1993, Romanians did not accept as their neighbors any of the groups deviant from the majority norm. Alcoholics, homosexuals, drug users were all rejected in the same degree, by three quarters of the population. People with AIDS were also categorically rejected by two thirds of Romanians. In most of the European countries, ethnic and religious intolerance is lower than the intolerance for the deviant groups. In 1993, a third of Romanians did not want a person of another race as their neighbor; this percentage was comparable with that of the individuals who did not want immigrants or foreign workers as their neighbors. Again, as with all the former communist countries 12, Romania was, at the beginning of the 1990s, one of the less tolerant European societies. The same behavior could be observed with regard to the rejection of Jews (a third of the population did not want a Jew as a neighbor) and gypsies (three fourths of the population rejecting them). In comparison to 1990, in 1999, most of the European countries displayed lower levels of intolerance (Figure 2). The same situation happened in Romania as well, and the tendency of the decreasing of intolerance manifested all the way to 2005. The WVS wave of 2005-2006 indicates a decrease to nearly 50% of the levels registered in 1990, with regard to people with AIDS, with the figures reaching 33% in the present day. The levels of intolerance towards alcoholics (66%), homosexuals (59%) and drug users (74%) are still high. Similarly, the percentages of intolerance towards immigrants and other races dropped to 16% and 18%. In all of these cases, the levels of intolerance registered in 1997 and 1999 are between the values of 1993 and those of 2005.

10

Slovenia is in a special situation, due to the fact that the level of development was higher even during the communist period, as well as to the higher interaction with the Western world. This probably explains the substantially higher level of tolerance registered by the 1990-1993 EVS/WVS wave. 11 For Poland, data was collected in 1990, when the Polish society was still marked by the wave of generalized enthusiasm due to the end of the communist regime. This state of mind may be the source of a higher level of tolerance (Grzymała-Kazłowska, 2004: 156). Bulgaria, as well as the Baltic countries, displayed in the early 90s a lower tolerance than Romania, and a much lower one as compared to Poland. 12 The exception is Poland (see the previous note). Slovenia, still involved in the very active events with ethnical connotations that marked the independence from Yugoslavia, showed in 1990 high levels of ethnical and religious intolerance. 270 corectura 2 28 iulie 2008

Between tradition and postmodernity? Figure 2. Intolerance in Romania in a European context (1999, 2005) and in dynamics (1993-2005) FR

7

6

HU

ethnical & racial intolerance

1999-2001 2005-2006 RO-93 RO-99

5

RO-2005 4 TR RO-93

3

BG RO-99

RU AL

HRPL

2

BE GR FRES

1

LU

0 2

IE

IT

CZ

FI IT

GB NL AT DK DE PTDE GB IC SE AD NL SE 3

4

BA SK EE LI MK SI BY PL UA N.Irl MT FI RU LA RO-2005

SI

MD

YU

5

6

7

8

9

intolerance towards deviant groups Sources: EVS/WVS 1993, 1999-2001, 2005-2006. Note. The subjects have been asked if they would like to have as a neighbor a person from different social groups. The intolerance index towards deviant groups sum up the number of groups rejected as neighbors by each respondent; from the following four possible choices: heavy drinkers, drug addicts, people who have AIDS, and homosexuals. Ethnic and racial intolerance is related to the following two groups: individuals of a different race, respectively immigrants/foreign workers. Each of the two indexes was later transformed so that it would vary from 0 (maximum tolerance) to 10 (maximum intolerance).

The comparison with the rest of Europe is difficult to make for 2005, because at the moment when this material was written, it was available data only for a few countries. Figure 2 allows for the observation of the same gap in 271

The Values of Romanians: 1993-2006. A sociological perspective

relation to Western Europe13, but also of the similarity with the evolution of Poland between 1999 and 2005. Additionally, if Romania and Poland register significant growths of tolerance between 1999 and 2005, the Western countries tend to stagnate at the same values as in 1999, while societies like Holland, Finland, Sweden, Germany, and Great Britain become slightly less tolerant towards the deviant groups. The intolerance towards gypsies, although in a continuous decrease in the post-communist period14, is still high: in 2005, 48% of the Romanian population did not want gipsy neighbors. This percentage is close to that registered in Slovenia (39%) and far lower than the one in Italy (74%). Overall, in Romania, the postcommunist period marked an increase of tolerance towards the others. Normativism is still dominant, but interactions with the other life styles and diversity increased the knowledge that the average Romanian has about the alterity, and facilitates a better understanding and acceptance. However, intolerance remains high as compared to most of the Western societies.

Work ethos For the communist regimes, work was the central declarative value. People got their wealth from work, and the complete occupation of the workforce was an essential goal of these regimes. In Romania, people did not have only the right, but also the duty to work. The right to work is more of a modern conceptual product. In the primitive societies, not working was equivalent to not surviving. The existence of the individual, of the collective and, in the end, of the human species, was conditioned by the participation of all in the productive activities, the only ones that brought food and minimal security in the face of nature’s challenges. Only in certain stages of modernity, characterized by unemployment and apparent supra-population, the right to work became an element that took a prominent place in the public debate, nowadays being guaranteed to all citizens. The abundance of jobs was over. More than this, the work of the individual was no longer essential for the survival of the rest. On the other hand, normativism implied an equalitarian conception and the need that each individual participate in the workforce. People were represented as identical beings, with needs and pleasures that could not be anything else but identical, so that all had to work in the same degree, so that they would not differ from the rest.

13

The violent ethnic incidents of 2005-2006 make France an atypical case from this point of view. 14 See M.Voicu, 2007. 272 corectura 2 28 iulie 2008

Between tradition and postmodernity?

Gradually, the acceptance of different lifestyles as ways of expression of the individual personality has led to different conceptions of work. The technologic progress, which led to an always increasing work productivity, has allowed a decrease in the time dedicated to work. The postmodern emphasis on self-expression contributes to a mix of value orientations that gave work a smaller place in the time budget, as compared to the past. Table 1. Attitudes towards work in Romania and a few other European countries Romania

Poland

Italy

Germany

Sweden

% „approval”* 1999 2005 1999 2005 1999 2005 1999 2005 1999 2005 To fully develop your talents, you need to have a job

71% 75% 91% 87% 91% 87% 72% 61% 50% 40%

It is humiliating to receive money without having to work for it

63% 59% 61% 58% 61% 58% 66% 58% 39% 31%

People who don’t work turn lazy

78% 80% 76% 74% 76% 74% 74% 72% 36% 39%

Work is a duty towards society

74% 66% 71% 66% 71% 66% 66% 68% 58% 61%

Work should always come first, even if it means less spare time

77% 72% 61% 59% 61% 59% 49% 47% 29% 36%

Sources: EVS/WVS 1999-2001, 2005-2006. * Each of the five items are accepted answers on a 5-point scale, expressing the agreement with the given affirmation. The figures in the table indicate, for each country, the percentages of those persons who declared that they “agree strongly” or “agree” to the respective affirmation.

The data reflects these tendencies. Table 1 suggests that work is more important in countries which have a lower economic output than those societies which are more traditionalist from a cultural perspective. Romania, Poland and Italy are in a visible contrast with Sweden. For these countries, particularly for Romania, work is a given without which life cannot be imagined. The position of Italy indicates the fact that the communist tradition is not necessary the one which determines this attitude in Romania and Poland. A more traditional cultural pattern might be the explanation for the higher support for the salience of work. An index of work ethos15, synthetically exploiting the information from Table 1, shows that most of the European societies experienced relative stability with regard to work ethos. The index displays are at similar levels in 15

The index is built as a factor score, based on the five items from Table 1. Indecision and the refusal to answer (“don’t know” and “no answer”) were recoded as mid-scale. The stability of the factor structure has been tested using Structral Equation Modeling. 273

The Values of Romanians: 1993-2006. A sociological perspective

1999 and 2005, and ordering the countries according to the respective values is likely to lead to similar rankings for the two timeframes. In 1999, across Europe, only in Turkey did the people give work more importance that in Romania16, while a few other societies (Albany, Bulgaria, Hungary, Macedonia and Poland) were similar to the Romanian space. The Mediterranean countries, as well as the ex-soviet ones, displayed average levels, while in the West and North, the place of work was not as central. In 2005, all of the European countries for which there was available data17, including Poland, registered a significantly lower level of work ethos than Romania. Traditionalism is one of the possible explanations for the high Romanian attachment toward work. One should also notice that this also reflects in the behaviors: data from 2003 shows that across the EU member or candidate states, except for Turkey, considering those people who had a job, the Romanians were working the highest number of hours per week (about 50), at more than 10 hours per week more than to the EU average (Voicu, 2006b).

Religiosity One of the chapters included in this book discusses in great detail the salience of religious belief in Romania. Mălina Voicu, the author of the chapter, makes the argument that Romanians are some of the most religious Europeans, and that their religious belief grew significantly between 1993 and 1999, and since then has been somewhat constant. I will not reiterate this data here, but I will mention three anecdotes, all connected to the religious practice where Romania apparently finds an average value in comparison to the rest of the European countries. In February 2007, there was a meeting of the main EVS investigators of each European country, in Sibiu. One of the present Dutch sociologists was very surprised by the very high number of the churches in the city. Only a short time before, I was witness to a discussion among some of my students, about the religiosity of their generation. One of the participants, a young woman, defined herself and acted as being not so religious in comparison with the rest. Accidentally, it happened that later I drove her home and I noticed that, when we passed by a church, she crossed herself. This gesture has reawakened an older observation in me: I have never visited a country where people would cross themselves when passing by a church as often as in Romania.

16

All the significance tests of this chapter considers p≤0,05. Mentioned in the introduction of this volume: Romania, Italy, Poland, France, Slovenia, Finland, Sweden, Great Britain, Germany, Holland, Andorra, and Russia. 17

274 corectura 2 28 iulie 2008

Between tradition and postmodernity?

Finally, after visiting Malta for a couple of weeks, I noticed the presence, in all public transportation (bus, boat, etc.), at least one religious icon or crucifix. I later realized that, in Romania as well, a religious icon can be found in these vehicles, around the driver seat, which is used as a sort of private area within the public space of the bus18. More than this, as much as I have noticed, in Romania, the houses that do not display any kind of religious icon are extremely rare. A search over the complex world of Internet may provide other hints of the Romanian religious practice. People may share pictures using the peer-to-peer networks over the Internet. I have searched through such pictures, depicting interiors, using the Romanian DC++ servers, were most of the users are Romanians. In almost all the sets of the photos I was able to locate at least a religious icon displayed on the walls or on the furniture. This means that such symbols are to be found in almost all the dwellings from which I was able to see pictures using at random this particular type of Internet sharing. One should also note that these dwellings are not selected at random. In order to work efficiently, the DC++ networks require broadband internet access, which has higher transfer rates, but is more expensive. People who can afford and have access to broadband connection, usually have a higher education, and better incomes. This is the category which is likely to be more secular in beliefs and religious practices than the rest of the population. Considering that almost all households from this more secular status group display religious icons on their walls, it is likely that the same happens when considering the houses where more religious people live. The three indicators – a high number of churches, crossing when passing a church, and the high concentration of religious icons – are not measured through the usual public opinion surveys. They are not the result of systematic observations, but converge with the survey data to indicate a traditional level of religiosity, this time converted in a type of religious practice, which might be at least as important as the standard measurement like the frequency of going to church, the usage of talismans, etc.

Traditional order or autonomy? In the initial section of this chapter, I argued that modernity, and particularly late modernity, are linked to a growth in the autonomy of the individual, and to the renouncing to authoritarianism. Two types of indicators tapping for orientation towards authoritarianism can be computed in all of the EVS/WVS waves. The first one is related to the opinions about what children should learn in the family. If one desires that the main attributes to be learned 18

Until recently, each bus driver has been assigned to only one car and each bus was assigned to maximum of two drivers. 275

The Values of Romanians: 1993-2006. A sociological perspective

are independence, responsibility, imagination and perseverance, this indicates a value orientation towards autonomy and independence. At the opposite pole there is the preference towards the learning of religious faith, obedience, thrift and hard working19. The second index may be computed starting from the expectations that people do have when looking toward the future. These are rather hopes and reflect the current values of the respective individuals. If, for in the near future, one expects changes that bring more emphasis on family life and “greater respect for authority” this indicates a high probability for the respective respondent to consider the traditional order as a central element of the social structure. The respective expectations are manifestations of the current latent orientations towards the reduction of individual autonomy in favor of the traditional social control. Table 2. Qualities which children can be encouraged to learn at home, considered important by the Romanian EVS/WVS samples in 1993, 1999, and, respectively, 2005 1993 1999 2005 Hard work

71%

82%

83%

Religious faith

43%

59%

63%

Thrift

37%

31%

53%

Obedience

19%

19%

17%

Imagination

17%

14%

18%

Independence

24%

30%

29%

Determination, perseverance

40%

19%

30%

Feeling of responsibility

56%

62%

69%

Source: EVS/WVS. Note: The figures represent the percentages of individuals who indicated the respective quality as being important, out of the total sample for the respective year.

For Romania, the 1993 EVS/WVS wave did not included items related to expectations for the future. However, the questions regarding the main qualities which children can be encouraged to learn at home are available and indicate a society that is significantly less traditionalist than Poland or Malta and similar to Hungary, Northern Ireland, Portugal, Russia, Slovakia and Turkey. All the other European societies are significantly less traditionalist. Hard working and religious faith were, in 1993, close to those of perseverance 19

See Hagenaars and others, 2003 for a similar index, labeled in the same way. A detailed analysis of the parental values of socialization, starting with the same set of items, is proposed by Paula Tufiş in this book. 276 corectura 2 28 iulie 2008

Between tradition and postmodernity?

and responsibility, whereas in 2005 they become prevalent. This is immediately reflected in the international comparisons as well: in 1999 only Poland, Turkey and Portugal are as traditionalist as Romania, considering the parental values; in 2005, all of the countries for the which data is available (including Poland) are significantly less traditionalist than Romania (see Figure 3). In fact, looking at the 1999 data, Romania remains at the same levels registered in 1999 too, whereas in Poland, for instance, the orientation towards authoritarianism diminishes in favor of the orientation towards autonomy. With regard to the expectations from the future, the orientation towards order slightly decreases from 1999 to 2005. In 1999, on this second authoritarianism scale, Romania was one of the most traditionalist European countries. Only Malta surpassed it, while Ireland, Portugal and Turkey had about the same levels. In 2005, the data show two Western societies, France and Great Britain, to be significantly more than Romania in search of the increase of the traditional-type order (see Figure 3). Considering the absolute numbers, Romania remains one of the few European countries that considers that the basis of education for children should consist rather from elements related to order and work than to creativity and independence (see the vertical axis on the graph). On the other hand, with regard to the expectations for the future, all of the analyzed societies except Sweden seem to wish to return to some of the more traditional order values. This tendency is more salient in the case of the ex-communist countries and some Western countries which, like France, have experienced a social integration crises. Analyzing the dynamics of the two indexes reveals rather opposite trends across Europe. For both abovementioned dimensions, between 1990 and 1999, as well as between 1999 and 2005, the ex-communist countries had the tendency to weaken the orientations towards the traditional order, moving in the direction of autonomy. However, some of the Western countries display different patterns. Most of the societies continued the process of strengthening the value orientations that put an emphasis on autonomy as a main element in the education of new generations. Only few societies (Germany, France, Finland, Holland and Belgium) display a different trend, as they have experienced in the 1990-1999 period a growth in latent orientations towards authoritarianism. For the 1999-2005 timeframe, among the countries for which data is available, Finland, France and Holland have continued the respective tendency of increasing authoritarianism against autonomy, while Great Britain and Sweden joined the same pattern, Italy does not display any significant change, while Germany, probably by solving the problem of Eastern immigration, experiences changes towards autonomy.

277

The Values of Romanians: 1993-2006. A sociological perspective Figure 3. Value orientations towards authoritarianism in Romania and in Europe (1999, 2005)

Sources: EVS/WVS 1999-2001, 2005-2006. Note. „Children should learn rather authority than autonomy”: the interviewed subjects were asked to choose which five “qualities which children can be encouraged to learn at home” are most important: hard work, obedience, religious faith, thrift, tolerance and respect for others, uunselfishness, perseverance, feelings of responsibility, imagination or independence. The index scores one point for each of the first 4 and decreases one point for each of the last 4. The factor analysis confirms the validity of the index. „Expects more emphasis on family and a greater respect for authority”: the summative index adds a point for each positive opinion towards of the two variables describing “various changes in our way of life that might take place in the near future”: “more emphasis on family” and “greater respect for authority” Both indexes were transformed in this graphic so that they would vary from 0 to 10.

278 corectura 2 28 iulie 2008

Between tradition and postmodernity?

All these tendencies, contradictory at first sight, draw the image of a changing world in search of a new social equilibrium. Changing the orientation towards autonomy tends to modify its meaning when societies are confronted with additional tensions (given, for example, from the need to integrate immigrants) and seems to become stable where there is economic growth.

Democracy of authoritarianism? The same discussion around authoritarianism is common for the latent orientations towards the type of organizing society considered as being the best. One’s position on the democracy-authoritarianism axis is directly linked to the opposition between modernity and traditionalism. Modernity brought a social organization that wanted to be rational, where societies are ruled by everyone’s participation, most often by voting on both the legislative and executive decisions. At the opposite side, the authoritarian leadership of a single person, of military or technocratic instances20 is a way of replacing the democratic principle of representation with the holding of power by a person or a group of persons. These kinds of alternatives constitute traditional ways of organizing the society. In a chapter dedicated to democratic orientations, included in this current book, Claudiu Tufiş discusses in detail the undoubted preference of Romanians for democracy, stronger and stronger since 1993 until the present day. A nuance has to be noticed: the meaning of democracy may vary. Romania is still one of the few European societies in which the model of an authoritarian leader “who does not have to bother with parliament and elections” receives the approval of the majority (see Table 3). Even if it does not gather such great support, a military regime would have far more partisans in Romania than in most of the other European countries. Also, the governments made of “experts” are highly supported as well as in most of the ex-communist countries.

20

The governments of “experts” contain, in essence, a rational attitude, induced by a division of labor that considers the specialization of jobs and professions. The individuals who are specialized to express opinions about the development would be, in this context, entitled to lead the society. But this invalidates another characteristic of the democratic rationality, referring the legitimating of the political decisions through the vote of the majority, which has to be convinced of the rationality of each of the political decisions and has to participate in the consultation process. Implicitly, the technocratic rule is an authoritarian one, not fully considering the individual freedoms and autonomy. 279

The Values of Romanians: 1993-2006. A sociological perspective Table 3: Attitudes regarding the political organization in Romania and in other European countries: 1999-2005 % subjects who answered that the Romania Poland Italy Germany Sweden following ways of governing a country are 1999 2005 1999 2005 1999 2005 1999 2005 1999 2005 ”very good” or “ fairly good”* Having a strong leader who does not have to bother with parliament and 57% 66% 20% 27% 15% 14% 15% 13% 21% 18% elections Having experts, not government, make decisions according to what they think is 74% 60% 77% 73% 46% 44% 50% 45% 39% 35% best for the country Having the army rule the country

23% 17% 15% 19%

Having a democratic political system

75% 83% 73% 73% 92% 92% 91% 92% 95% 97%

4%

6%

2%

3%

7%

5%

Data source: EVS/WVS 1999-2001, 2005-2007. * The exact question was: “I’m going to describe various types of political systems and ask what you think about each as a way of governing this country. For each one, would you say it is a very good, fairly good, fairly bad or very bad way of governing this country?”

I have computed an aggregate index of the orientation towards democracy21, taking into account the answers to all the four questions in table 3. It indicates that, in comparison with almost all the European countries, in 1999, Romanians were significantly more supportive for the authoritarian alternatives. Moldavia, Macedonia and Turkey showed similar levels, while the other societies displayed a significantly higher support for democracy. The same index, when calculated for the countries available in the 2005-2007 WVS wave, places Romania at the same level as in 1999.

Family, marriage and gender relations In another chapter of this book, Raluca Popescu shows that a large majority of the Romanian society considers family to be “very important”. There are no significant changes to be recorded in the three considered EVS/WVS waves: both in 1993, 1999, and 2005, about 85% of Romanians considered that family is “very important”. This places Romanians around the European average, most of the other European societies displaying similar scores. All over Europe, only a minority of citizens believe that marriage is an outdated practice. Romanians show no exception. Only 8% agreed with this 21

The index is calculated as a factor score, based on the original variables, with the indecision (‘Don’t know’) and missing values (not answering) being recoded into the scale mid-point. The stability of the factor structure in 2005 compared to 1999 has been tested using SEM (Amos). 280 corectura 2 28 iulie 2008

Between tradition and postmodernity?

statement in 1993, 11% in 1999 and 14% in 2005. The trend is similar to that of Europe, with more and more Europeans, especially among the Western ones, people seeing marriage as an institution from the past. It has to be noticed that, in many other societies, in this respect, as in most of the above described ones, the traditionalist orientations are less present than in Romania. For instance, in 1999, in France, Belgium and Luxembourg, the percentage of those who defined marriage as outdated was as high as a third of the population (the highest among the considered European countries). The tendency of the European societies is to maintain family as a very important, if not essential element of the social structure. Its form seems to be undergoing a changing process, and marriage, as a founding event, slightly diminishes its role. This implies only a reserved attitude towards formalizing it in front of a representative of a public authority or a priest, which is considered less and less necessary. In this static image, Romania finds a place rather in the more conservative half of the European countries. However, the cross-European variation in this respect is relatively small, as compared to the other dimensions that are being analyzed in this chapter. There is a domain of social life, that is traditionally linked to the structure of family life, and which values seem to show a more pronounced dynamic. This is the way of structuring the gender relations. Some analysis of the values in this domain (M. Voicu, 2004; Voicu & Voicu, 2002) shows that in 1999, Romania shared a common pattern with the other ex-communist countries: The presence of women in the marketplace was a natural reality, with the value orientations of the population supporting it at the same levels as in the North of the continent (less than in Sweden, but more than in Finland or Denmark) and significantly more than in the Western societies. On the other hand, gender equality inside the household was supported in 1999 by less Romanians than in most of the Western, Northern, Southern and even Eastern countries: women were, to a large extent, the only ones in charge of the household duties. Other analyses (B.Voicu, 2006; Voicu, Voicu, Strapkova, 2006), this time referring to behaviors and not to values, allow some nuances: Romania shows one of the more equalitarian models in Europe in relation to the time spent by each gender doing domestic work, with the exception of raising children, which is, almost exclusively a women’s responsibility. In all of the three reference waves (1993, 1999, and 2005), the EVS/WVS data allow observing the evolution over time of only two types of values: if women should be part of the workforce22, and to what extend should the labor market be equally open to men and women23. 22

Measured through a single item, namely the disapproval with the following statement: “Being a housewife is just as fulfilling as working for pay”. The possible answers were given on a 4 point scale: strongly agree, agree, disagree, strongly disagree. 23 The orientation towards that particular value is tested through a single item as well: “When jobs are scarce, men have more right to a job than women”, accepting the answers: “Agree”, “Disagree”, “Neither”. 281

The Values of Romanians: 1993-2006. A sociological perspective Figure 4. Gender values in the labor market: Romania in a European context

Sources: EVS/WVS 1990-1993, 1999-2001, 2005-2006. On the horizontal axes is represented the percentage of those who disagree with the statement “When jobs are scarce, men have more right to a job than women”. The vertical axe displays the percent of those who disagree with the statement “Being a housewife is just as fulfilling as working for pay”.

Compared to other societies, the Romanian citizens strongly support the presence of women in the labor market in all of the three considered moments: 1990-1993, 1999-2000 and 2005-2006. There are no differences in the support levels registered in 1993 and 1999, as half of the sample group answers in favor of women participation to the paid workforce. In 2005, this figure grows 282 corectura 2 28 iulie 2008

Between tradition and postmodernity?

significantly to 65%, far superior to all of the societies for which the data is available (Figure 4). Three factors probably contribute to the explanation of these phenomena. One is in close connection with work ethos, work being one of the central values of the traditionalist society, as we have already made the argument in one of the preceding sections. The second explanation may find its source in poverty: to reach a minimal living standard, only one salary is not enough. The third factor is related to the ideology of equality in the workplace, which was intensely promoted during communism and adopted as a powerful value orientation. With respect to the second dimension, things are different. Romania is constantly placed among the traditionalist countries when looking at the equality between men and women: if there are only a small number of jobs, men should get an advantage. This does not affect the values of support for the presence of women in the labor market, but, as a consequence, jobs with higher responsibility must be occupied by men24. Such an explanation is consistent with the higher responsibilities that Romanians confer to women in what concerns the process of child care and the raising of children. The direct consequence over family is, in fact, representing it as a hierarchical unit, in which the man is the one who decides25 everything.

Post-materialism – at the end of the tunnel? In Romania, the discussion about post-materialism most often attracts perplexed smiles. Both the society and the individual are concerned rather with material safety than with beauty and self-fulfillment. Using the classical Inglehart scale, in its shorter version26, only 8% of Romanians could be characterized as post-materialist in 1993, 7% in 1999 and 5% in 2005. The number of materialists and the percentage of those with mixed orientation

24

The scarcity of jobs implies the choice between committing the responsibility of earnings to women or to men. The preference for inequality works here as an indicator to give large responsibilities to men. 25 For some data and references about this subject, see Vlăsceanu, 2007: 205-2006. 26 The respondents were asked to indicate, from a list of four items which two were the most important: “There is a lot of talk these days about what the aims of this country should be for the next ten years. […]. If you had to choose, which of the things on this card would you say is most important? 1. Maintaining order in the nation, 2. Giving people more say in important government decisions, 3. Fighting rising prices, 4. Protecting freedom of speech”. Individuals who chose items 1 and 3 are labeled as materialist, and those who chose items 2 and 4 are labeled as post materialist. Any other combination leads to the labeling of that individual as having a mixed orientation, between materialism and post-materialism. 283

The Values of Romanians: 1993-2006. A sociological perspective

maintained the same over the entire period, and practically equal (45-48%, respectively 47-48%). Consequently, Romania is among the European societies with a high level of materialism, next to most of the ex-communist countries, from which it does not differ very much. Poland, for example, in 2005 had 7% postmaterialists, the same as in 1999. In 1999, Lithuania had 6% postmaterialists, Hungary had 2% and the Czech Republic had 10%. The Western societies and particularly the Northern ones displayed much higher levels: in 2005, in Germany, 20% of the population showed post-materialist orientations; in Italy there were 19%; in Sweden - 23%; in England - 24%; and in Holland - 19%. The structure of the population indicates for Romania a majority of materialists and individuals with mixed orientation, both groups with equal percentages. The pattern is common in most of the ex-communist societies, except for the Czech Republic, Slovenia, Poland and the Baltic countries, were, like in the West, those with mixed orientations form the larger group. On the whole, the levels of post-materialism decreased from West to East, and from North to South, with Romania belonging more to the group of countries orientated towards materialism. It is still possible that, under the impact of value contagiousness created by the effect of the massive circulatory migration, of economic growth and of access to university studies, the countries that today are oriented towards materialism will see a decrease in this orientation, making room for other value orientations guided towards superior needs.

Romania 2005: (still) at the beginning of modernity A few technical considerations Most of the quantitative attempts to simplify the interpretation and explanation of the changes from traditionalism to modernity and the cultural processes related to postmodernization have tried to identify sets of two major factors, each depicting one of the processes. Inglehart27 discusses about a bidimensional space of the social values, defined by the secular-religious polarity, and by the opposition between the orientation towards values of survival and that towards values of self-expression, of personal development. Similarly, in the analysis proposed by Hagernaars et al. (2003), two factors are identified, labeled as “normative-religious”, respectively “autonomy – socio-liberalism”. Both Inglehart and Hagernaars et al. use factor analysis to investigate the same data sets, using almost the same items. The two spaces that they construct are similar, and the labels associated to the two sets of axes suggest 27

Inglehart (1997), Inglehart and Baker (2000), Inglehart and Welzel (2005).

284 corectura 2 28 iulie 2008

Between tradition and postmodernity?

that they partially overlap. The first axis opposes in essence the modern and the traditional value orientations. The second contrasts orientations specific to the late modernity to the modern ones. Both analyses similarly order the European countries on the two axes. The relations between the two pairs of factors and other indicators (those of economical growth or of the democratic orientation for example) are empirically consistent for both pairs and satisfy diverse theoretical expectations. In spite of all this, I would say that both approaches have the disadvantage of treating the two dimensions as independent from one another. In other words, the modernization and post-modernization processes would be completely different and unconnected. It is true that none of the mentioned authors presents the analyzed axes as dimensions of modernity or post modernity. However, as I have already mentioned, the way of structuring these factors makes them overlap with the two processes. On the other hand, the whole literature dedicated to postmodernity28, implies continuity between the two processes, as it has been already argued in the first section of this chapter. Hence, the expectation that it is possible to build a model which considers two factors of modernization and post-modernization that are not orthogonal, but would accept to correlate29. Such can be the one depicted in Figure 11 of the Annex30. The two described factors from that model cover the modernity – traditionalism and modernity – post-modernity polarities. The goodness of fit indexes shows that the empirical data supports both the existence of two factors and of the covariance relation that ties them. Several important precautions should be considered before drawing any conclusions about the theoretical consistency of the proposed model. First, it must be noticed that a reduced number of indicators has been included, as opposed to the two studies mentioned before. I tried to reduce the number of variables, and to include in the model only those ones existing for the three EVS-WVS waves which I am using, in order to be able to compute scores that would allow the analysis of the dynamics of Romanians’ orientations towards modernity, in comparison with other European societies. The problem is that the three waves do not repeat all the items that might have been considered in order to have a complete model to describe the modernization and postmodernization 28

No matter whether it uses different labels, such as late modernity, reflexive modernity, risk society, postindustrial society, etc. 29 The option to extract orthogonal factors, rotated so that the dimensions’ interpretation is simplified, is without any doubt a legitimate choice, at least when the goal is the analysis of major tendencies of structuring the field of social values. 30 I have proposed another model that accepts correlations between the two factors in B. Voicu, 2005b: p. 192. The model that I propose in this paper is a simplification and an improvement of the previous one, also using only items that might be found in the 19901993, 1999-2001 and 2005-2006 waves of the value surveys. 285

The Values of Romanians: 1993-2006. A sociological perspective

processes. Moreover, in some of the waves, for certain countries, not all the relevant items are available. However, the high level of correlations between the variables assures that, even if a smaller number of indicators are included, similar factors will be obtained31. Apart from the limited availability of the data, there is a technical obstacle that requires cautious interpretation of the results. For analyzing the data, I have employed Amos 4, which did not allow weighting systems to be used. However, subsequent validation of the results, through analyzing the relations with other variables (external validation) indicates a high level of consistency for the extracted factors. Another problem could be represented by the not very high levels of some of the saturations observed through the analysis. A relatively recent discussion (Jagodinsky, 2003) brings into focus the need to better explain each of the analyzed value orientations. Jagodinsky pleads for reporting models with higher levels of saturation. His argument is based on a potential criticism regarding the inclusion of some of the attitudes and opinions measured through questionnaires as being manifestations of more general value orientations, like the orientation towards modernity. A low saturation coefficient indicates that the respective values and attitudes would be only loosely explained by the respective value orientations. This might implicitly require a better specification of the models. However, particularly in the case of the individual analysis, the errors of measurement are higher and may reduce the saturations. More than this, as Jagodinsky shows, the models do not necessarily have to include complete explanations for each variable. In the case of the models presented in the current paper, as the analysis refers to the individual level and the sign of each of the observed relations has a strong theoretical support, the model can be considered valid, even if some of the saturations are lower. Besides all these, the main problem of such a model is, in my opinion, the nonlinearity of the relation between the two major processes: modernization and post-modernization. The two processes come from one another and are related to each other, but, as it is illustrated in Figure 1, the link between them is not and cannot be one of direct proportionality. Postmodernization begins when the modernity is not yet complete. At that moment, the speed of modernization tends to decrease. Rationality has already spread in most of the social groups and patterns of thinking, and suddenly, spectacular changes are unlikely to occur. Any increase in the modernization level can not be sharp. On the other hand, the postmodernization process is just at the beginning. Consequently, it may have enormous growth potential, and the pace of the accompanying changes can be relatively high. At least induced by the value contagion due to the international migration and to globalization, elements of postmodernization

31

See for this Inglehart, Welzel, 2005.

286 corectura 2 28 iulie 2008

Between tradition and postmodernity?

may appear in the traditional societies as well where modernization has a higher speed. The relation is therefore complex, not following the logic of linear or at least exponential dependency. This makes the relation very difficult to model, particularly considering the current limitations of the structural equation modeling software. A possible solution for this problem could be the independent analysis of each identified factor. This is the option I will use next. Figures 12 and 13 (from the Annex) show the structure of the factors of traditionalism and postmodernism. These latent dimensions are the same with the ones from the previous model; only this time they are independently analyzed. The factor structure is the same in all in the three analyzed timeframes (Table 4, Annex). For certain EVS/WVS waves, the questionnaires miss few of the considered items. Therefore, in order to be able to include most of the European societies in the comparative analyses, I chosen to work with the reduced models which are described in Figures 5 and 6. The respective models continue to have acceptable goodness of fit indexes and conceptual consistency. Figure 5. The factor structure of traditionalism – the reduced model (2005-2006) ,27 intolerance towards 'deviant groups'

e1

ethnic-racial intolerance

e2

,21 e3

,44

,42

TRADITIONALISM

Religiousity

,28 Materialism e5

,64 e10

-,24 e18

,13

,09 e16

authority, nu autonimy

national pride

e19

Family=important

Women to work

Goodness of fit indexes: IFI=0,996; CFI=0,996; RMSEA=0,042. The coefficients on the arrows are standardized estimates, and are significant at p=0,05.. Data Source: WVS 2005-2007, only the European countries that had available data in October 2006 (see text), N=15034.

287

The Values of Romanians: 1993-2006. A sociological perspective

Figure 6. The factor structure of postmodernism – the reduced model (2005-2006)

e10

authority, not authonomy

Women=Men at the job

-,59

e17

,42 e6

,27

POST-MODERN

Postmaterialism

,25 Leisure=important

,57

,47 Homosexuals=OK

,43

Divorce=OK

,40

e15

Abortion=OK

,59

e12

,42

e13

e14

Goodness of fit indexes: IFI=0,999; CFI=0,999; RMSEA=0,025. The coefficients on the arrows are standardized estimates, and are significant at p=0,05. Data Source: WVS 2005-2007, only the European countries that had available data in October 2006 (see text), N=15034.

The factor scores computed using the reduced models show extremely high correlations (over 0.9) with their “complete” versions, which indicate their relevance. Even more, they positively correlate with the orthogonal factors extracted by Inglehart, at about 0.532. In the following sections, I will use these factors estimated through the reduced models to describe Romania as compared to other European societies, and to investigate the differences between different status groups within the Romanian society.

32

The factors are practically the same, but they are not subjected to an orthogonal rotation. 288 corectura 2 28 iulie 2008

Between tradition and postmodernity?

A map of Europe The communist regime33, in its continuous search for legitimacy, strongly promoted the idea of Romania as a modern state, relatively well developed, with a highly educated and efficient workforce. The completely enslaved press, the results in sport, the results of the Romanian pupils in the international school Olympics, the promoting of the social group of writers as exponents of the modernization levels were ways meant to convince the population of how modern, high performing and developed Romania was. Meanwhile, the political regime proposed an autocratic social structure, based on traditionalist logic. The economical processes were governed by nonrational rules as well. How much, where and under what kind of conditions one should produce were not the results of any kind of efficiency calculi, or rational planning, but the answers to political commands often meant to convince the population that the society was truly functional and could, without any difficulty, sustain the wellbeing of its citizens. In a Romania of severe penury, as it was during the 80s, the government was involved in an obvious conspicuous consumption: extremely expensive spectacles to praise the party and its leader, financial help for underdeveloped societies, but with the same role for consolidating the status of the leader, strong investments to a few top performs in sports or education, to underline the society’s capacity to have better results than any other country. The visible support elements of this kind of discourse were the modernizations of the infrastructure and housing. Although incomplete, the post-War urban development represented a step forwards as compared to the problems of a society almost completely lacking in public utilities, infrastructure, urban transportation, running water and pluming, etc, as was Romania between the World Wars. The model proposed by the communist rulers was apparently modern (especially through the technological differences and the changes in infrastructure compared to the previous period), but still profoundly traditional through the social organization, the hierarchical structure of human relations, the generalized mistrust, the tendency for conspicuous consumption that would be eventually transferred at the individual level as well, etc. I have sketched in the previous sections, links between this recent past and the value orientations in various domains, discussing the positioning of Romania as compared to other European countries. Using the aggregate indexes of modernism, respectively postmodernism, one may analyze how contemporary Romanian society, in its more general cultural traits, is different or similar to other societies, and how it has changed over the past 15 years. 33

The following paragraphs synthesize a perspective of the way the Romanian communist regime functioned; this perspective is based on the analysis of vast literature described in detail in B. Voicu, 2005a. 289

The Values of Romanians: 1993-2006. A sociological perspective

The map from Figure 7 describes the differences between the European countries, concerning the orientation towards traditionalism. This is about opposing the more religious, more normative, more materialist, more orientated towards the rejection of women from the job market, to those who have tendencies toward secularization, autonomy, gender equality, etc. Figure 7. Cultural modernity across Europe, 1999-2001*

Note: The darker a country is colored, the higher the average level of modernity is. The countries colored in light grey are more traditional than the others. Data sources: EVS/WVS 1999-2001. *For Norway and Great Britain I have used data from the 1990-1993 wave. For Cyprus and Andorra I have used data from the 20052006 wave. For the European countries colored in white (Turkey, Georgia, Armenia, Azerbaijan, Switzerland, Croatia) no complete data were available.

Romania is placed in the group of the more traditional countries, together with most of the ex- communist states. The tendency towards traditionalism grows from East to West and from South to North. This holds true for the growth of postmodernism as well (not shown results). The correlations between the level of modernity of the European societies and the levels of economic development, the degree of democracy and 290 corectura 2 28 iulie 2008

Between tradition and postmodernity?

the education indicators are extremely high34, confirming the theoretical expectations. In 2005, Romania’s level of modernity is similar to that of 1999 and that of 1993. The tendency seems to be one of small variation above and below the same value. To be more exact, between 1993 and 1999 a small increase in traditionalism seems to be registered, and may be explained by the economical recession and social instability of transition. After 1999, a contrary tendency seems to be registering, but with a low intensity as well. In the case of the European countries for which data is available for the 2005 wave, from both East or West, a tendency of increasing modernity can be seen, both between 1990 and 1999, as well as between 1999 and 2005. The differences are in most of cases not significant, but their meanings, the fact that these slight increases are registered in all the analyzed societies, represent a sign of tendencies from the modernization process to continue. The fact that Romania registers an inverse variation during the 1993-1999 period may be most probably explained by the atypical transition period, particularly slow as compared to the other former communist societies. On the other hand, this relates to the low levels of cultural modernity that characterized the Romanian society in the beginning of the 1990s. The indicator that was built and discussed here has similar levels for Romania in 1993 with the ones registered in 19901991 for Malta, Poland, Hungary and Russia. In the 1990-1993 EVS/WVS wave, this is the group of European countries that are less modern (data is missing for most of the ex-soviet and ex-Yugoslavian countries). In 1999, the group Romania belongs to is almost the same, as only The Moldavian Republic is added (for which there is not data available for the early 1990s). The only one difference is the fact that, while all of the other countries from this group show transformations towards modernity, Romania goes the opposite way35. This reduced level of initial modernity probably acts as a break in the way of reform, and is closely connected to a more accentuated conservatism and 34

I have considered the level of PIB/inhabitant, the percentages of those with tertiary education within the 25-64 year old population, and Freedom House Civil Liberties Score. 35 Finland knows a similar trajectory in the 90s: The economic boom of the 1980s came together with a fast cultural modernization. The recession of the 1990s, worsened by the break of the USSR with whom Finland had a large part of its commercial relations, is marked by the return to more traditional values. The unrealistic hopes induced by the previous growth are shattered, and the Finns started to fear the bankruptcy of the society and the state as a whole. Unemployment grew during the 1990s, contributing also to the decrease of trust in the institutions of the democratic order (Listhaug, 2003) and to the re-orientation towards the traditional ones. The re-stabilization in the second part of the 1990s and the reprised growth both meant a new orientation towards modernity (I thank Sakari Nurmela, from the TNS Gallup Finland, for his help in clearing this explanation.). 291

The Values of Romanians: 1993-2006. A sociological perspective

to a more reduced predisposition to risks. Hence it derives a slowing of the transition and the initial preference for a gradual change. On the other hand, the transition was accompanied by persistent poverty, an increase in unemployment and an understanding of the low economic development levels of the country. All of these uncertainties can be added, related to the ways in which someone can manage in this new society. All of these factors are a mix that determines a return to traditionalism. After 2000, the tendencies of economic growth, the adaptation to new system, the certainty of European integration, all of them start to allow the return to the modernization process. Let also note that the above mentioned hierarchies are relative. The map from the Figure 7 displays the position of each European country only in comparison to the other societies in the Old Continent. Traditionalism and modernity are not normative labels, as there are no clear guidelines according to which one can say a country is traditionalist or modern. Romania is more traditionalist than most of the European countries, but this does not mean that it is necessarily traditionalist in itself. If we look, for example, at most of the African countries included in EVS/WVS data set, the modernity-traditionalist indicator shows that these societies are less modern than Romania, but, following the same logic as above, this does not mean that Romania is a modern country. Considering the positioning related to these different reference systems (Europe and Africa), as well as the theoretical and empirical arguments from the previous sections, I would say that Romania is still at the beginning of its cultural modernization process. Some dimensions indicate an accentuated modernization (the value orientation related to gender equality in the labor market, for example). Others (religiosity, for instance) indicate strong traditionalism. The ensemble of Romania’s values brings the country rather near incipient modernity than to the late one. The differences that I have described as existing in different domains are induced mainly by the type of the modernization project that the Romanian society experienced. The communist experiment was an incomplete modernization, leading to the current pseudomodernity.

Romania 2005: differences between groups In this society which, in its whole, oscillates between tradition and modernity, being closer to cultural traditionalism, there are groups that share distinct cultural patterns. Nothing about this is surprising. In every society there are value polarities, usually opposing the young and the old generations (Inglehart 1990, 1997; Ester et al, 1994), but also structured around interregional differences (Sandu, 1996, 1999), or according to different other social statuses (depending on gender, ethnicity, religion, education, etc). These are the differences presented in this section. 292 corectura 2 28 iulie 2008

Between tradition and postmodernity? Figure 8. Comparing the cultural modernity levels across different status groups in Romania (2005) Bucureşti

unfinished university or college

Bucureşti & Ilfov

town: 200.000+ inhab.

Banat Crişana-Maramureş

town: 1-200.000 inhab.

Oltenia Moldova Valahia Transilvania

Dobrogea

postgrad. (MA, Ph.D.) university/B.A.

town: 30-100.000 inhab. town: less than 30.000 inhab. village: commune center peripheral village

employee high school

part-time employee

post-high school unfinished highschool VET apprentince school

man Orthodox Protestant (Roman)Catholic

lower secondary incomplete lower sec.

primary school

Romanian Hungarian Gipsy woman

self-employed

unemployed housewife retired

GreekCatholic

Note: The graphs are similarly scaled and represent the average values of the modernity-traditionalism index described in the text, for the different status groups in Romania indicated in the image, according to the WVS 2005 data.

Figure 8 allows for a few rough comparisons. There are some groups who are clearly more modern than the rest of the population. Romanians from larger cities and the ones with higher education display a higher likelihood to be more modern. At the opposite side, there are the small villages and people with less education. The size of the village or city and the level of education are the characteristics that determine the more contrasting differences. A large city has its own dynamics, a lifestyle that implies overstepping from the extended family, the contact with different individuals, renouncing to the traditional social control, a larger individual autonomy. Similarly, higher education allows individuals to better understand and predict the social environment and the way in which this is organized and enable them to easier cope with higher levels of axiological uncertainty.

293

The Values of Romanians: 1993-2006. A sociological perspective

MODERNITY

Figure 9. Modernity and traditionalism in Romania, in comparison to other European countries

Sweden 2005

Germany 2005 Finland 2005 Slovenia 2005 Italy 2005 Bulgaria 1999

Poland 2005 Romania 2005 village: commune center 2005

Rep. Moldova 1999 Romania 1993 Romania 1999

TRADITIONALISM

Bucureşti 2005

Note: The figures in the graph are the averages of the modernity-traditionalism index described in the text, in conformity to the WVS 2005-2006 and EVS/WVS 1999-2002 data.

Bucharest, an urban conglomerate larger than anywhere in the country, but also with a population twice as much educated and wealthy36, has a modernity level significantly higher than any other region of the country. The historical provinces, in their whole, do not differ in the levels of modernity, as the only significant difference is the one between Banat and Dobrogea. However, these regions are not very homogeneous, or at least not as homogeneous as the socio-cultural areas37. Considering this sort of grouping, 36

Considering the percentage of tertiary graduates in the active population, respectively the average wages and the GDP/capita. 37 Defined by Sandu (1999), the cultural areas display homogenous groupings of counties when looking at the development levels, groupings that in fact are smaller segmentations of the historical regions. For example, Walachia is divided as follows: the Northern counties are grouped together (Argeş, Dâmboviţa, Prahova), the poor 294 corectura 2 28 iulie 2008

Between tradition and postmodernity?

the picture changes. For instance, Bucharest does not significantly differ from Northern Walachia or Crişana; in Moldavia the Galaţi-Iaşi grouping are significantly more modern than the rest of the region; in Walachia, the Northern counties are significantly more modern than the ones from the south; in Transylvania, the Sibiu-Braşov grouping is more modern than CovasnaHarghita and Bistriţa-Sălaj. However, some caution is necessary when interpreting these results, given the relatively small number of cases and the small geographic dispersion of the sampling points in such small geographic areas. Ethnicity and religion do not generate significant differences. Romanian, Hungarians and gypsies do not register differences in their levels of modernity, with the same caution related to the small number of individuals who have declared to belong to the last mentioned group. Orthodox, Catholics and Protestant persons are not different as well, the only important opposition being the one between Orthodox’s and Greco-Catholics. The latter ones, with relatively many located in smaller communities in the rural area, tend to be more traditionalist. Some differences are noticeable when considering the employment status. The occupationally active individuals (employed, self-employed/ employers) are less traditionalist than the inactive ones (retired, housewives). Similarly, women are significantly less modern than men. I have already shown that, in comparison to other European countries, the degree of modernity in Romania is relatively low. The more modern groups of the society make no exception. Bucharest, where the average cultural modernity is the highest across the country, is significantly below the European average. High school graduates, as well as the inhabitants of cities with a population larger than 200.000, place at an average level similar to that of Poland, one of the societies less oriented towards modernity when considering the European societies as a whole. Along with education and the size of the locality, age is the third sociodemographic factor that determines the important variations within Romania. The younger the individuals, the more modern they are. The 20-29 years old cohort registered in 2005 a similar level to that of Poland as well, while the age group of 30-39 years old gravitated around the Romanian country average. All of the other age groups were significantly more traditional.

counties of the South form a second group (Teleorman, Giurgiu, Călăraşi, Ialomiţa), and the counties of Buzău and Brăila form the third group. 295

The Values of Romanians: 1993-2006. A sociological perspective Figure 10. The inter-generational differences in cultural modernity in Romania: 1993, 1999, and 2005

1993 1993

modernity level

1999 1999

2005

2005

2005 1993 1999

before 1924

1924-1933

1934-1943

1944-1953

1954-1963

1964-1973

1974-1983

after 1984

year of birth

Source: EVS/WVS 1993, 1999, WVS 2005.

Figure 10 also suggests that Romania respects the same patterns that are to be found elsewhere with respect to the relation between age and cultural modernity. The inter-generational differences are not induced purely by age: older people are not more traditionalist solely because they are older. The differences are in fact structurally related to the intrinsic differences between generations. The today older generations tend to be less traditionalist than the previous generations were at their age. For instance, the individuals who were born between 1944 and 1953, were of ages 50-60 in 2005, being a little more modern than the previous generation (individuals born between 1934 and 1943) at the same age (meaning in 1993). On the whole, all generations knew a withdraw process towards traditionalism in 1999 compared to 1993. This is more evident at the young cohorts, even if they continued to maintain a higher level of modernity than the elderly. Moreover, these young cohorts (individuals that were less than 30 years old in 1993 and less than 42 in 2005), continued the process of retraditionalization after 1999 as well, when the rest of the population tended to orient toward cultural modernity more than in the 90s. The explanations are multiple. They may come from the higher openness to new of the young generations. In the 1990s, the “new” has largely meant the rediscovery of identities and ways of doing things dating back to a rather old period of history, namely the between-Wars period, a time marked by cultural traditionalism. The repudiation of the recent communist history has led to the need to search for the 296 corectura 2 28 iulie 2008

Between tradition and postmodernity?

legitimacy of other “Golden Ages”38. Taking over values from this period, in which Great Romania had reached its maximum of economic, social, politic and artistic development, appeared as a natural consequence. On the other hand, these cohorts cover the bulk of a generation who has the conscience to have provoked a fundamental societal change in 1989, through revolution. Therefore, it is natural that they would develop stronger feelings of belonging in relation with the rediscovered values and ways of doing, no matter if the respective values would be traditional or not. More, this is a generation that has rapidly conquered important positions in the social structure39, and that tries to preserve the new order, consolidating its own power. Both arguments lead to the hypothesis of a growth in traditionalism even after 1999, for these younger strata of society, being consistent with what the empirical data show.

Conclusion. The fifth transition: from communism to postmodernity? Initially, the transition has been teleologically defined, as a process to organize the communist societies on a capitalist base. The goal of the transformation process was the western capitalism, as a successful model in the efficiently structuring the institutional framework, perceived as being the best suited to the development levels of its time. The transitology literature, and the debates in the public space have focused mainly on four dimensions of change: 1. restructuring of the economy through marketization and privatization; 2. internal redefinition of the state through democracy; 3. external redefinition of the state, through the redirecting the external relations and alliances and (in the case of USSR, Czechoslovakia, Yugoslavia, GDR) the adjustment of borders; 4. changes of the social structure, implying the increase of inequality and also the re-conversion of capitals, entrepreneurship etc.40 The discussion about cultural change and about values did not have a central place in the debate, even though it sometimes appeared in subsidiary. On one hand, communism had been always seen as a modernization process whose flaws had only affected the institutional frameworks, but not the essence of the type of social cohesion and representations about the world and life. On the 38

The communist propaganda used to label the Ceauşescu’s rule as the „Golden Age”. See numerous young business men owning profitable companies, the large number of successful professionals, with better situations than older fellow citizens with the same educational levels and ascribed statuses. In the political space the respective generation held in 2005 numerous important positions, including a few ministers, the presidency of the Chamber of Deputies, several mayor portfolios, etc. 40 See Sawka (1999), Kuzio (2001), Pickel (2002), Pickavance (1999), Stark and Bruszt (2002), Sztompka (1999), B. Voicu (2005a: 43-47). 39

297

The Values of Romanians: 1993-2006. A sociological perspective

other hand, in the beginning of the 90s, the thinking about development was still paying tribute to the normativism of unique models, and to institutionalism, considering that the forms are the sufficient prerequisite to regenerate content and to ensure an acceptable human and social development. The discussion about cultural differences, either inherited from the pre-communist period (B. Voicu, 2005a) or generated by communism as a cultural trauma (Sztompka, 2000), eventually represented a constant presence in the analysis of postcommunist transformations, becoming an element of interest; a process which I would label as being the one from the fifth transition (B. Voicu, 2005a: 43-47). Eastern Europe, as I have already shown in this article, proved to be less modernized than the West, and even less oriented towards modern values. Romania is no exception, being one of the more traditionalist societies in Europe. Without doubt, the process of cultural neo-modernization is still contextually relevant for the problem of transition. The initial teleology of the post-communist transition is in this sense affected by two obstacles that are difficult to overcome. First, the West does not offer any unique model of social organization. An immediate example for this variability may come from considering the differences between the welfare regime of the North, the continental conservatism, and the Mediterranean model of the South41. However, there are several common traits of European societies, expressing a pattern that could be, principally speaking, the target of the transition processes. Rationalization of social order, planning, democracy and equality are all principles which were stipulated through the project for a European Constitution, and found their sources in the modern orientations of value. Western modernity is a modernity in changing, a late modernity, still unstable. The cultural changes are accompanied by essential changes in the model of societal organization. Using the previous example of welfare regimes, let’s note the fundamental changes that affect their essence, through the decentralization, the privatization of social services, the pregnant orientation towards new domains like single parent families, gender equality, solving the problem of the inverse relation between the volume of active and inactive individuals (see Giddens, 1998; Esping-Anderson, 2006 etc.). Particularly this transition of the Western modernity towards postmodernity became the second obstacle in taking the West as a model for the post-communist transition. A variety of questions raised for the formercommunist societies, al putting the difficult task to find the most adequate answers. What should be the first goal, building modern or postmodern 41

I refer to the models of providing welfare centered on action of the state (in the North), of the community (in central Europe), of family and kinship (in the South). See Esping-Andersen (1990), Liebfried (1993), and Fitzpatrick (2005) regarding quadruple typology (triple at Esping-Andersen) of the welfare regimes in Europe. 298 corectura 2 28 iulie 2008

Between tradition and postmodernity?

capitalism? Is postmodernity a viable project? Can it be accomplished in the absence of the completion of modernity? Are the postcommunist societies ready to face the risks specific to the modern societies (material uncertainty), as well as those specific to the post-modern societies (axiological uncertainty)42? I have argued that postcommunist pseudo-modernity implies the coexistence of modern and traditional values, and the East European societies are in a full modernization process, while post-modernization tendencies are more or rather incidental. Romania is even a little more culturally traditional than the rest of the Eastern societies. I have also argued that Romania, as a whole, displays a relative homogeneity of values. Most of the status groups do not differ significantly from one another. Only individuals with higher education (approximately 10% of the population) and individuals living in Bucharest (10% of the population as well) are, on average, significantly more modern than the rest, but even these groups are placed relatively far from the averages of modernity in the rest of the European countries. Several important factors may generate change through contagiousness of values. Globalization and circulatory migration favor the interaction of the Romanians with the Western values. The tendency to adopt the Western models becomes explicit for the more modern social groups. For the rest, it manifests itself through the imitation of consuming patterns. In their turn, economic growth, generational change, higher and higher access to university education43 and a continuation of the urbanization process are catalyst factors for the modernization process. In the following years, Romania will probably know a relatively fast process of cultural modernization. Simultaneously, but slower in the beginning, a post-modernization process will affect the life-styles, individualizing and diversifying them. Religious values and respect for traditional hierarchies will be the most resilient to change, particularly due to the fact that the most exposed groups to the tendencies of modernization and post-modernization are placed in social positions that allow usage and control of these hierarchies for own purposes. On the other hand, some groups, like the circulatory migrants and their social networks will tend to act in a way to change the cultural preeminence of the traditional hierarchies. Another important factor could be the form taken by the migration waves towards Romania. Taking into account their content and the type of population that will come to Romania, these waves can mean a retreat towards more traditional values, or, the other way around, they can mean an emulation of modern and late modern values. For example, attracting a large mass of 42

Starting from Beck’s discussion on risk society, Rinkevicius (2000) argues that the post-communist countries are societies of double risk, combining the risks of their current modernity with the ones produced by the Western post-modernity. 43 Still very low in comparison to the European average (B.Voicu, 2005b). 299

The Values of Romanians: 1993-2006. A sociological perspective

expats as experts for the Romanian-based companies or branches of international companies can induce modernization effects, intensifying the contact with modern or postmodern values. Opposed to this, waves of lower qualified and, maybe, more traditional labor force can generate the adverse effect, especially in what concerns intolerance. Major national or a global economic crisis can determine fast and long term returns to traditional values, in the conditions in which the value orientations associated to modernity are not yet strongly consolidated in the Romanian structure of the culture. This can happen in the case of important increases of intolerance levels in the Western societies, with an explicit or diffuse direction towards emigrants, including Romanian emigrants. It is very unlikely that other stagnations of post-modernization in the West44 would significantly affect the value changes in Romania. As I have already shown, modernization is much more relevant in Romania than post-modernization. This makes the variations of post-modernization in other societies less likely to affect our own modernization process by contagiousness, even if the respective societies are the ones relevant for Romanians as reference group. Concluding, one may expect a continuation of the post-communist transitions, even after the integration into the European Union45. In fact, they manifest as processes of normal social change, processes which are continuous by their nature.

44

Using the same incomplete set of data of WVS 2005-2006 as I exploit in this chapter, Welzel (2006) observes,, a series of Western societies, decreases in the orientations towards values which can be assimilated to post-modernity (like the survival/selfexpression dimension proposed by Inglehart). 45 In the literature, accession to the European Union is sometimes defined as a conventional terminus point for the end of postcommunist transitions (Lavigne, 2000). 300 corectura 2 28 iulie 2008

Between tradition and postmodernity?

Annex Figure 11. A two-factor model of analysis for the social value (1999) ,29 intolerance towards 'deviant groups'

e1

ethno-racial intolerance

e2

-,25

e8

clear good-evil guidelines

-,29

-,45

e10

-,10

-,11

authority, not autonimy

Religious faith

e3

Materialism

-,26

-,44

,03

e20

,41

TRADITIONAL

-,32

predicts authoritarianism e9

-,47

supports democracy

national pride

e19

-,50

,19

Women to have job

Family=important

e18

-,09

,11

,06

e16

,01

e11

e17

-,20

Women=Men at workplace

Marriage=outdated

-,08

e5

,26

e6

-,35 Postmaterialism

-,23

POST-MODERN

Ecological concerns

-,09 -,76

-,63

Homosexuals=OK

-,57

Divorce=OK

,03 e12

,04

Abortion=OK

e7

-,19

Leisure=important

,46 e13

e14

e15

Source: EVS/WVS 1999-2001. N=52258. All the coefficients have been allowed to vary freely, with the exception of the covariance between the two factors (traditionalism and post-modernity) which was fixed at 0.5. Goodness of fit indexes: IFI=0,988; CFI=0,988; RMSEA=0,048. The coefficients on the arrows are standardized estimates, and are significant at p=0,05. 301

The Values of Romanians: 1993-2006. A sociological perspective Notes on the indicators used in Figure 11: ¾

The two indicators of intolerance, the orientation towards supporting democracy, and the orientation towards support for environment protection (ecological concerns) are built as factor scores, being described in the text.

¾

Tolerance towards homosexuals, divorce and abortion (“homosexuals=OK”, “divorce=OK”, “abortion=OK”) represent indexes of approving each of the three phenomena as being justified. The indexes are scaled from 1 (always justified) to 10 (never justified). Refused to answer and indecision (NA and DK codes) were recoded as scale midpoint.

¾ Religiosity is an index of religious faith, computed as a factor score, and explaining the answers to questions that measured: The intrinsic importance of religion: - How important is God in your life? (10 point scale), - How important is religion in your life? (4 point scale). Refused to answer and indecision (NA and DK codes) were not considered in the analysis. The importance of the Church in the public space: Generally speaking, do you think that your church is/the churches are giving, in your country, adequate answers to… The moral problems and needs of the individual; The problems of family life; People's spiritual needs; The social problems facing our country today? The four variables accepted only answers of yes and no, and they were recoded as dummy variables. I tested the stability of the factor structure for each of the EVS/WVS waves in the analysis, using as criteria IFI, CFI and RMSEA. An additional covariance was accepted between the first two indicators.

46

¾

The importance of free time and the importance of family were measured through 4 point scales. Refused to answer and indecision (NA and DK codes) were recoded as scale midpoint.

¾

The “clear good-evil guidelines” is a dummy variable, showing the support for the statement “There are absolutely clear guidelines about what is good and evil. These always apply to everyone, whatever the circumstances.” against the alternative “There can never be absolutely clear guidelines about what is good and evil. What is good and evil depends entirely upon the circumstances at the time” or indiference between the two.

¾

“National pride” represents the answers to the question: „How proud are you are you to be a Romanian citizen?46” on a 4 points scale.

¾

“Woman should work”, “Women=Men at the workplace”, “Marriage=outdated” are built on the basis of a single item each, according to

For each society, the name of that particular country was mentioned.

302 corectura 2 28 iulie 2008

Between tradition and postmodernity? the description in the section dedicated to family and gender relationships in the text. ¾

“Predicts authoritarianism” and “Authority, not autonomy” are the two factor scores described in Figure 3.

¾

Materialism and postmaterialism are indexes derived from the standard scale of postmaterialism by Inglehart (the one with four items), also described in the text.

Figure 12. The factor structure of traditionalism (2005) ,27 intolerance towards "deviant groups"

e1

ethnic-racial intolerance

e2

,20

e8

clear good-evil guidelines

predict authoritarianism

,25

supporting democracy

,41

-,50

TRADITIONAL

,29

e10

,59

e3

Materialism

,38 -,09 ,15

authority, not autonomy

Religious faith

,48

e9

,00

e20

e5

,08

national pride

Women should work

Family=important e19

e18

-,20

-,08 e11

e16

,01

mariage=outdated

Source: WVS 2005-2007, only the European countries that have available data in October 2006 (see the text). N=15034. Goodness of fit indexes: IFI=0,987; CFI=0,987; RMSEA=0,064. The coefficients on the arrows are standardized estimates, and are significant at p=0,05.

303

The Values of Romanians: 1993-2006. A sociological perspective Figure 13. The factor structure of postmodernism (2005) e10 e16 e11

,00

authority, nu autonomy

e17

e6

Women=Men at workplace

Marriage=outdated

-,59 ,16

,00

,42 Postmaterialism

,28

POST-MODERN ,14 ,56

,48

Homosexuals=OK

,39

,25

,44

Divorce=OK

Environmental concens

Abortion=OK

e7

Leisure=important

,58

e12

,41

e13

e14

e15

Source: WVS 2005-2007, only the European countries that have available data in October 2006 (see the text). N=15034. Goodness of fit indexes: IFI=0,997; CFI=0,997; RMSEA=0,035. The coefficients on the arrows are standardized estimates, and are significant at p=0,05.

304 corectura 2 28 iulie 2008

Between tradition and postmodernity? Table 4. Goodness of fit for the various models that were ran Goodness of fit EVS/WVS Model Version data set IFI CFI RMSEA PCLOSE complete Traditionalism reduced

1999-2001

0,996

0,996

0,038

1,000

2005-2006

0,994

0,994

0,041

1,000

1990-1993

0,996

0,996

0,039

1,000

1999-2001

0,998

0,998

0,035

1,000

2005-2006

0,996

0,996

0,042

1,000

1990-1993

0,997

0,997

0,036

1,000

complete 1999-2001

0,999

0,999

0,026

1,000

2005-2006

0,999

0,999

0,025

1,000

1990-1993

0,997

0,997

0,031

1,000

1999-2001

0,998

0,998

0,032

1,000

2005-2006

0,997

0,997

0,035

1,000

Postmodernity reduced

Note: The factor structure of the complete models is presented in Figures 12 and 13. For the reduced models, the factor structures are presented in Figures 5 and 6. In all of the cases, I imposed the regression coefficients to take the estimated values for the models built on the 1999-2001 data. For the 1990-1993 EVS/WVS wave, the complete model of traditionalism could not be run because of the missing data.

References Arts, Wil, Jacques Hagenaars, Loek Halman, editors. 2003. The Cultural Diversity of European Unity. Findings, Explanations and Reflections from the European Values Study, Leiden: Brill. Beck, Ulrich (1992). Risk society : towards a new modernity. London, Sage. Beck, Ulrich, Anthony Giddens and Scott Lash. (1994). Reflexive Modernity. Politics, tradition and aesthetics in the modern social order, Cambridge: Polity Press. Chiribucă, Dan. 2004. Tranziţia postcomunistă şi reconstrucţia modernităţii în România, Dacia, Eikon, Cluj-Napoca. Esping-Andersen, Gøsta. 1990. The Three Worlds of Welfare Capitalism, Cambridge: Polity Press. Esping-Anderson, Gøsta, with Duncan Gallie, Anton Hemerijck, John Myles. Why We Need a New Welfare State. Oxford: Oxford University Press Ester, Peter, Loek Halman, Ruud de Moor, 1994. The Individualizing Society. Value Change in Europe and North America, Tilburg University Press. Fitzpatrick, Tony, Jo Campling. 2005. New theories of welfare. Basingstoke, Palgrave Macmillan. Giddens, Anthony, 1990. Consequences of modernity, Polity Press, Cambridge. Grzymała-Kazłowska, Aleksandra. 2004. Three Dimensions of Tolerance in Poland and in Europe, in Aleksandra Jasińska-Kania, Mirosława Marody, eds., Poles among Europeans, Warsaw: Widawnictvo Naukowe Skolar, p. 152-172. 305

The Values of Romanians: 1993-2006. A sociological perspective Hagenaars, Jacques, Loek Halman, Guy Moors. 2003. Exploring Europe’s basic values map, in Arts and others., eds., p. 23-66. Hall, Stuart, David Held, Dan Hubert and Kenneth Thompson, editors, 1996. Modernity. An Introduction to Modern Societies, Blackwell Publishers, Cambridge. Haperkamf, Jurgen, Neil Smelser, editors, 1992. Social Change and Modernity, University of California Press, Berkeley · Los Angeles · Oxford. Inglehart, Ronald, 1990. Culture Shift in Advanced Industrial Societies, Princeton University Press. Inglehart, Ronald, 1997. Modernization and Post-Modernization. Cultural, Economic and Political Change in 43 Societies, Princeton University Press. Inglehart, Ronald, Wayne E. Baker, 2000. Modernization, Cultural Change, and the Persistence of Traditional Values, American Sociological Review, vol. 65, February, p, 19-51. Inkeles, Alex, 1996 [1969]. Making Man Modern: On the Causes and Consequences of Individual Change in Six Developing Countries, American Journal of Sociology, 75, 2 (September, 1969): 208-225, also reproduced in Alex Inkeles and Masamichi Sasaki (editors), Comparing Nations and Cultures. Readings in a Cross-Disciplinary Perspective, Prentice Hall, p. 572-585. Inkeles, Alex, David Smith, 1974. Becoming modern. Individual Change in Six Developing Countries, Harvard University Press, Cambridge (Massachuttes). Jagodzinski, Wolfgang. 2004. Methodological problems of value research, in Vinken and others, eds.: p. 97-121. Kumar, Krishan, 1995. From Post-Industrial to Post-Modern Society. New Theories of the Contemporary World, Blackwell, Oxford (UK) & Cambridge (USA). Kumar, Krishan, 1999. Modernization & Industrialization, Encyclopedia Britannica, vol. 24, 15th Edition, p. 255-266, reproduced in Malcolm Waters, editor – Modernity. Critical Concepts. Volume I: Modernization, Routledge, London and New York, 1999, p. 72-104 [first edition: 1990]. Kuzio, Taras, 2001. Transition in Post-Communist States: Triple or Quadruple?, Politics, 21 (3): 168-177. Lash, Scott, 1990. Sociology of Postmodernism, Routledge, New York and London. Lavigne, Marie, 2000. Ten years of transition: a review article, Communist and Postcommunist Studies 33: 475-483. Leibfried, Stephan. 1993. Towards a European Welfare State? On Integrating Poverty Regimes into the European Community, in Catherine Jones, ed., New Perspective on Welfare State in Europe, Routledge, p. 133 – 153 Listhaug, Ola, 2003. Confidence in Governmental Institutions and Strength of Civic Morality in Europe: A Comparison of Trends in New and Old Democracies 1981-1999, presented at Integration & Fragmentation. European Values at the Turn of the Millennium, Tilburg University, March 26-27, 2003 Parsons, Talcott. (1964). Evolutionary Universals in Society, American Sociological Review, vol. 29, no. 3, p. 339-357 Parsons, Talcott. (1971). The system of modern societies. Englewood Cliffs, N.J., 1971 Pickel, A, 2002. Transformation theory: scientific of political?, Communist and PostCommunist Studies, 35, p. 105-114. Pickvance, Christopher G., 1999. Democratization and the Decline of Social Movements : the Effects of Regime Change On Collective Action in Eastern 306 corectura 2 28 iulie 2008

Between tradition and postmodernity? Europe, Southern Europe, Latin America, Sociology, vol. 33, no. 2 (May): 353-372. Rinkevicius, Leonardas, 2000. Public Risk Perceptions in a ‘Double-Risk’ Society: The Case of the Ignalina Nuclear Power Plant in Lithuania, Innovation. The European Journal of Social Sciences, vol.13, no. 3: 279-289. Roth, Andrei. 2002. Modernitate si modernizare socială, Polirom, Iasi. Sandu, Dumitru, 1996. Sociologia tranziţiei. Valori şi tipuri sociale in România, Staff Publishing House, Bucharest. Sandu, Dumitru, 1999. Spaţiul social al tranziţiei, Polirom, Iasi. Sawka, Richard, 1999. Postcommunism, Open University Press. Stark, David, Laszlo Bruszt, 2002. Transformarea politicii şi a proprietăţii in Europa de Est, Ziua Publishing House, Bucharest [first edition, written in English: 1998]. Sztompka, Piotr, 1993. The Sociology of Social Change, Blackwell, Oxford (US) and Cambridge (UK). Sztompka, Piotr, 1999. The cultural core of post-communist transformations, in Thomas P. Boje, Bart van Steenberg, Sylvia Walby – European Societies. Fusion or Fission?, Routledge, London and New York: 205-214. Sztompka, Piotr, 2000. Cultural Trauma. The Other Face of Social Change, European Journal of Social Theory, 3(4): 449-466. Vinken, Henk, Soeters, Joseph, Peter Ester, eds. 2004. Comparing Cultures: Dimensions of Culture in a Comparative Perspective, Leiden: Brill. Vlăsceanu, Lazăr, 2001. Politică şi dezvoltare. România încotro?, Trei Publishing House, Bucharest. Vlăsceanu, Lazăr. 2007. Sociologie şi modernitate. Tranziţii spre modernitatea reflexivă, Iasi: Polirom. Voicu, Bogdan, 1999. Modernitatea între tradiţie şi postmodernism, [Modernity between Tradition and Postmodernism], Revista de Cercetări Sociale, nr. 34/1999, 36-59. Voicu, Bogdan, 2001. România pseudo-modernă [Pseudo-modern Romania], Sociologie Românească, 1-4/2001: 36-59. Voicu, Bogdan. 2005. Penuria Pseudo-Modernă a Postcomunismului Românesc. Volumul II. Resursele, Iasi: Expert Projects. Voicu, Bogdan. 2005a. Penuria Pseudo-Modernă a Postcomunismului Românesc. Volumul I. Schimbarea socială şi acţiunile indivizilor, Iasi: Expert Projects. Voicu, Bogdan. 2005b. Penuria Pseudo-Modernă a Postcomunismului Românesc. Volumul II. Resursele, Iasi: Expert Projects. Voicu, Bogdan. 2006a. Household and family relations, in Ioan Mărginean, coord., First European Quality of Life Survey: Quality of Life in Bulgaria and Romania, European Foundation for Improving Living and Working Conditions, Luxembourg: Office for Official Publications of the European Communities, http://www.eurofound.eu.int/pubdocs/2006/67/en/1/ef0667en.pdf: 35-42. Voicu, Bogdan. 2006b. Work and life balance, in Ioan Mărginean, ed., First European Quality of Life Survey: Quality of Life in Bulgaria and Romania, European Foundation for Improving Living and Working Conditions, Luxembourg: Office for Official Publications of the European Communities, http://www.eurofound.eu.int/pubdocs/2006/67/en/1/ef0667en.pdf: 43-48. 307

The Values of Romanians: 1993-2006. A sociological perspective Voicu, Mălina, 2004. “Women Work and Family Life: Value Patterns and Policy Making”, in Will Arts and Loek Halman (editors). European Values at the Turn of the Millennium, Leiden: Brill Voicu, Mălina, Bogdan Voicu, Katarina Strapcova. 2006. Housework and gender inequality across Europe, IRISS working papers, http://www.ceps.lu/iriss/wps.cfm Voicu, Mălina, Bogdan Voicu. 2002. „Gender values dynamics: Towards a common European pattern?”. Romanian Journal of Sociology, nr. 1-2 Voicu, Mălina. 2007. Toleranţă şi discriminare percepută, in Gabriel Bădescu et al.., Barometrul Incluziunii Romilor, Bucharest: Open Society Foundation, p. 55-64. Weber, Max, 1978 [1922]. Economy and Society, University of California Press, Berkely, Los Angeles, London. Weber, Max, 1995 [1920]. Etica protestantă şi spiritul capitalist, Humanitas Publishing House, Bucharest. Welzel, Christian. 2006. Tracing the Magnitude and the Direction of Global Value Change: 1981-2006, presented at World Values Conference. Society, Politics and Values: 1981-2006, Istanbul, 3-4 November 2006. Winiecki, Jan, 1988. The Distorted World of Soviet-Type Economies, Routledge, London.

308 corectura 2 28 iulie 2008

Appendix: List of Romanian Political Parties & Map of the regions

Romanian Political Parties The main Romanian parties, referred in different parts of the book are described bellow: •

•

•

PSD (The Social-Democratic Party) was formerly named the Romanian Party of Social Democracy (PDSR) and the Democratic Front of National Salvation (FDSN). FDSN separated in 1991 from the National Salvation Front (FSN), where it has been constituted the conservative wing. Until recently, the party was dominated by the figure of the former president Ion Iliescu, who is still an important member of the PSD (President of Honour). Since many of its founding members, particularly Ion Iliescu, were parts or somehow related to the communist nomenclature, PSD was and still is labeled (and accused) as being neo-communist. PD (Democratic Party) is the party of the current president Traian Băsescu. Until recently, the party claimed a social-democratic ideology, but in 2005, after coming to power, they decided to join the family of popular parties. PD and PSD have common routes, since PD is the continuator of the National Salvation Front (FSN), established in January 1990. Nowadays, the party fully depends on the decisions of the president Traian Băsescu (according to the Romanian constitution, when Băsescu become president, he was forced to resign from PD, becoming the “mediator” of the political debate). His idea to have a two-party political system, involving the merging of PD and PNL was one of the sparks for broking the government alliance between the respective parties. PD absorbed PLD in December 2007, forming PD-L (The Democrat-Liberal Party), but they have actually changed nothing but the name. PNL (the National Liberal Party) was reestablished in 1990, 40 years after its dismissal under the communist rule. Several crumblings and re309

The Values of Romanians: 1993-2006. A sociological perspective

•

• •

•

•

•

unifications mark the past 15 years of PNL History, different wings leaving and eventually re-joining the party. However, after 2000 a relative stability is noticeable. PLD (The Liberal-Democrat Party) is the only wing who left PNL after the reunification of all the liberal fractions. PLD supports Băsescu, the current president, this being the major difference to the PNL. The name of the party reminds the intention of Băsescu’s supporters to merge the liberal and the democrat party. PLD have finally merged into the PD, to form the PD-L (The Democrat-Liberal Party) in December 2007. PRM (the Great Romania Party) is a nationalist party, dependent on the image of its founder and president, Corneliu Vadim Tudor, who was the runner up in the 2000 presidential elections. PNG (the New Generation Party) claim to be a representative of ChristianDemocracy. However, the ideology is not important, the party being practically bought in 2003 by its billionaire president Gigi Becali, a former shepherd better known as owner of the football club Steaua Bucureşti. Becali promotes populist positions, with some religious, anti-Semite, nationalistic, and homophobic nuances. UDMR (the Democratic Union of Hungarians from Romania) is an ethnic party, collecting votes almost exclusively from the Hungarian minority, located mostly in Transylvania. They are also part of the European Popular Party. PNŢ-CD (the Christian and Democratic National Party of Peasants) was established re-established in December 1989 after the communist interdiction to exist. They had been in power in 1996-2000, but have lost popularity and have never managed to accede in the parliament since. PC (the Conservative Party, organized around the mass-media mogul Dan Voiculescu) and PIN (The National Initiative Party, a small group who left PD in early 2005, after a conflict with Traian Băsescu) are two other small parties (1-2% of the voters), both with unclear ideology.

The Map of Romania and the location of the main regions referred in the text The 41 smaller regions displayed on the next page map are the Romanian counties, corresponding to the NUTS3 statistical level. The larger regions (Wallachia, Oltenia, Dobrogea, Moldova, Transylvania, Banat, Crişana & Maramureş) roughly overlap the NUTS2 regions. They are also known as “historical provinces”: 310 corectura 2 28 iulie 2008

Between tradition and postmodernity?

• • • •

•

Wallachia and Oltenia use to have the same ruler starting with the 12th century and they merged with Moldova in 1859 to form Romania. Dobrogea, ruled by the Otoman Empire, joined after the The Russo-Turkish War of 1877–1878, known by Romanians as the Independence War. Crişana and Maramureş had the same rule as Transylvania starting with the 12th century, and all three provinces were incorporated with the Hungarian Kingdom, joining Romania after the First World War (1918). Banat knew Hungarian rule from the 10th to the 15th century, being incorporated into the Otomans (16-17th century), then becoming a separate province of Habsburg Monarchy, along with Southern Crişana (Arad County), within very fluctuant borders. In 1918, after a short independence, and two months of Serbian occupation, the Banat Republic joined Romania The Northern parts of the Suceava County (Northern Moldova) were annexed by the Habsburg Empire in the 18th century, to form the province of Bukovina. In 1918, Bucovina joined Romania, but most of the territory was annexed in 1939 to the Soviet Union, being included in the Soviet Republic of Ukraine, along with the Southern Basarabia (Nothern Basarabia is the nowadays Republic of Moldova, another part of the former Moldavian Principate with the capital in Suceava, then in Iaşi). TRANSYLVANIA Botoşani Botoşani Maramureş Maramureş Maramureş

Mare Satu Satu Satu Mare

Suceava Sucea Suceava va

CRIŞANA-MARAMUREŞ

MOLDOVA

Bistriţa Bistriţa Bistriţa Năsăud Năsăud Năsă ud

Sălaj Săla Sălajj Bihor Bihor Bihor

Iaşi Iaşi Iaşi Neamţ Neamţ Nea mţ

Cluj Cluj Mureş Mureş Mureş

Va Vaslui Vaslui slui

Bacău Bacău Bacău

Harghita Harghita Harghita

Arad Arad Alba Alba Alba

Covasna Covasna Covasna

Sibiu Sibiu Timiş Timiş Timiş

BANAT

Hune Hunedoara Hunedoara doara

Galaţi Galaţi Galaţi

WALACHIA

Buzău Buzău Buzău

Caraş Caraş Ca raş Severin Severin

Vâlce Vâlcea Vâlcea a

Gorj Gorj

Brăila Brăila Brăila

Prahova Prahova

Argeş Argeş

Tulcea Tulcea

Dâmboviţa Dâmboviţa Dâmboviţa

Mehedinţi Mehedinţi Mehedinţi

OLTENIA

Vrancea Vrancea

Braşov Braşov Braşov

şti Bucure Bucureşti Bucureşti & Ilfov Ilfov & Ilfov

Ialomiţa Ialomiţa Ialomiţa Călăra Călăraşi Călăraşi şi

Olt Olt Dolj Dolj

Giurgiu Giurgiu

DOBROGEA Constanţa Constanţa Constanţa

Teleorman Teleorma Teleorman n

BUCHAREST & ILFOV

311

The Values of Romanians: 1993-2006. A sociological perspective

About authors Bogdan Voicu Ph.D. in Sociology (University of Bucharest), principal research fellow with the Research Institute for Quality of Life of the Romanian Academy of Science, associate lecturer at the University of Bucharest. Recent books: The PseudoModern Penury of the Romanian Postcommunism (Expert Projects, 2005), Romanian village on its way to Europe (2006, Polirom, editor), Social Sciences perspectives on the European postcommunist societies (Psihomedia, 2005, editor), European Integration from East to East. Civil society and ethnic minorities in a changing world (Psihomedia, 2004, editor), Globalization, Integration, and Social Development in Central and Eastern Europe (Psihomedia, 2003, editor). Mălina Voicu Ph.D. in Sociology (University of Bucharest), principal research fellow with the Research Institute for Quality of Life of the Romanian Academy of Science. Member of the Executive Committee of the European Values Study. Recent books: Religious Romania. On the European tide or behind it? (Institutul European, 2007), Romanian village on its way to Europe (2006, Polirom, editor), What type of welfare do Romanians want? On the legitimacy of the social policies in Romania (Expert Projects, 2005), Social life in urban Romania (co-author, Polirom, 2006). Mircea Comşa Ph.D. in Sociology (Babeş-Bolyai University of Cluj), research director with Metro Media Transilvania, associate proffesor at the Babeş-Bolyai University of Cluj UBB. Recent books: Lifestyles in Romania after 1989 (Presa Universitară Clujeană, 2006), Social life in urban Romania (co-author, Polirom, 2006), General elections 2004. A sociological perspective (Eikon, 2005, editor). Raluca Popescu Ph.D. in Sociology (University of Bucharest), principal research fellow with the Research Institute for Quality of Life of the Romanian Academy of Science. She is preparing a book on family in Romania.

312 corectura 2 28 iulie 2008

About authors

Horaţiu Rusu Ph.D. in Sociology (University of Bucharest), lecturer with the Lucian Blaga University of Sibiu. Recent books: Social Sciences perspectives on the European postcommunist societies (Psihomedia, 2005, editor), European Integration from East to East. Civil society and ethnic minorities in a changing world (Psihomedia, 2004, editor), Globalization, Integration, and Social Development in Central and Eastern Europe (Psihomedia, 2003, editor). Claudiu D. Tufiş Ph.D. in Political Science (Pennsylvania State University), principal research fellow with the Research Institute for Quality of Life of the Romanian Academy of Science. Recent books: The Romanian Higher Education System (Afir, 2007, co-author), Roma Life Strategies: People and Communities (forthcoming, editor). Paula A. Tufiş Ph.D. in Sociology (Pennsylvania State University), principal research fellow with the Research Institute for Quality of Life of the Romanian Academy of Science.

313