4   Norms of good citizenship
Bas Denters, Oscar Gabriel, and Mariano Torcal
Introduction
Normative political theorists have argued about good citizenship for centuries. From Aristotle to Alexis de Tocqueville and Walter Bagehot, civic virtues such as rationality, moral obligation to pursue the common good, social engagement, and political activism have been interpreted as prerequisites of a good society and a good polity (see, for example, Almond 1980; Walzer 1989). There is, however, still no generally acknowledged, uncontested model of good citizenship, even after centuries of philosophical and academic debates.
In recent years the citizenship debate, previously a purely philosophical and academic issue, has evolved into an important topic of the public discourse. Once representative democracy was established and the basic political and civil rights conquered, citizens began to question their role and their involvement, as well as their capacity to influence and control governments and public policies. The debate on the quality of those democracies and citizenship has begun. Tins discourse was driven by the process of societal modernisation that has resulted in greater individual autonomy and provided more and more citizens with the intellectual resources to define their place in society and in the democratic system. Moreover, globalisation and transnational migration have added to the diversity of outlooks on socio-political life in modern European societies. At the same time, we hear pleas in the public debate to restore people's responsibility. Willi the decline of the traditional welfare state, citizens are expected to become more actively involved in resolving their personal problems as well as engaging themselves in the production of goods and services that were previously provided by the authorities. These expectations are based on a conception of citizenship in which people should not only be concerned with their private affairs, but should feel a moral obligation to contribute to the promotion of the common good.
Whether or not contemporary citizens are willing and able to take on such responsibilities is a matter of debate (see, for example, Klages 2000a, 2000b). Social scientists have expressed divergent ideas regarding the
Norms of good citizenship 89
potential consequences of modernisation. On the one hand some analysts have emphasised the emancipative effects. Ronald Inglehart, for example, has pointed out that the value change that accompanies modernisation will provide the basis for a new, more participatory, type of social and political order and citizenship. On die other hand, scholars have indicated that people's willingness to accept traditional norms such as compliance, discipline, and thinking of others may be declining in an age of increasing individualisation (for example, Noelle-Neumann 1978). Likewise, Putnam (1995b, 2000) has stated concerns about the disappearance of the 'Civic Generation', and Munkler (1997: 153) has pointed out that, due to the recent social, political, and economic changes, many people fear that 'good citizenship' might become extinct (see also Walzer 1991).
In this chapter, we will examine how mass publics in western and eastern Europe perceive good citizenship. We will start with the assumption that a pluralism of civic norms, entailing a coexistence of different conceptions of good citizenship, is characteristic for modern societies (Sniderman et al. 1996). In this contribution we will focus on three notions of citizenship (similarly to Conover and Searing 2002; Rose and Pettersen 2002): a traditional elitist model (core norm, law-abidingness); a liberal model (core norm, deliberation); and a communitarian model (core norm, solidarity). Pluralism of civic norms, however, does not necessarily imply cultural fragmentation. The coexistence of norms may take many different forms. At the individual level, citizens might endorse either a particular 'pure' conception of citizenship (for example, a predominantly communitarian view) or mixture of norms stemming from different conceptions. At the societal level, normative pluralism might imply anything ranging from widespread consensus over a particular pattern of norms (either one of die 'pure' types or a particular mixture) to the prevalence of dissent between advocates of divergent normative conceptions of citizenship. Empirical research of the type we have conducted is expected to show how European mass publics perceive citizenship.
In this chapter we will address two main questions. First, we will look into the consequences of value pluralism. We will start this exploration by examining whether, in the hearts and minds of the members of European mass publics, the various notions of citizenship have been incorporated in a more or less integrated belief system or whether these notions constitute 'pure' - mutually exclusive and competing - philosophies. In the latter case of cultural fragmentation, we should find no or only weak positive correlations between the different individual value orientations and we should be able to classify most individuals as 'believers' in one or another 'pure' belief system. In the second section, we shall deal with such individual level questions. We can, however, also observe such patterns of integration and fragmentation at the aggregate level. In the third section, we will therefore ask similar questions about the attitudinal patterns that are characteristic for each of our nations. On the basis of such analyses we
90   Denters, Gabriel, and Torcal
will be able to answer the question as to whether value pluralism results in cultural fragmentation, where at the individual and/or the aggregate level we find evidence of fragmented belief systems.
After these descriptive analyses we will turn to a second question that pertains to one of the main threads in this book: what is the democratic relevance of social involvement and social trust? In the final section of this chapter, we will look into the question as to whether particular civic norms go hand-in-hand with social involvement and social trust.
Concepts and the structure of individual orientations
Concepts
Before we begin our empirical investigation it is necessary to determine the key components of the various concepts of citizenship typical of the western tradition. According to Conover et al. (1991: 805; italics added), 'citizenship is a fundamental identity that helps situate the individual in society'. The status of the citizen is defined in two fundamental relations: the relation between die individual and the other members of his/her society; and the relation between the individual and his/her institutions of government (e.g. Prior el al. 1995: 5-6). Notions of citizenship specify the principles according to which these two relations should be conceived in order to realise a good society.
In their Civic Culture, Almond and Verba (1963: 337f.) presented a model of citizenship integrating two different and partially contradicting elements of the citizen's relation to the political authorities. In order to meet the requirements of democratic politics, citizens should be politically interested, active, self-conscious, and critical, but at the same time loyal, trusting, and deferential., On the one hand, political authorities need discretion in order to perform their task of making and implementing authoritative decisions in an effective way. On the other hand, the more they are trusted and the more people behave as loyal subjects instead of self-conscious citizens, the higher the risk of abuses of political power by officeholders. Thus, in addition to loyalty, a critical sense and activism are required for effective democratic checks on political power. Clearly, the first segment of the civic role is rooted in a traditional-elitist view of democracy. Regarding the critical activist component of citizenship, the literature provides a number of different views.
In a liberal interpretation (ICymlicka and Norman 1995: 297-8), general virtues such as law-abidingness and loyalty should go hand-in-hand with critical and deliberative values: that is, political virtues such as the ability to question authority and a willingness to engage in die public discourse.-Within liberal theory there is some debate about the normative weight of this component. Whereas Kymlicka and Norman (1995) stress the importance of public engagement and die development of a critical attitude
Norms of good citizenship 91
towards the authorities, someone like Walzer (1995: 158-9) presents a different view: 'To live well is not to make political decisions or beautiful objects; it is to make personal choices ... And tile market within which the choices are made ... largely dispenses with politics; it requires at most a minimal state'.
Whereas the previous notions of citizenship focus on the relation of individuals with their government, communitarians emphasise the importance of members of die community towards one another. Voluntary associations of citizens are perceived as the breeding ground of civic virtues. In his discussion of the communitarian tradition, Putnam (1993: 86-91; Conover and Searing 2002) highlights die importance of orientations like solidarity, community participation, and tolerance. Such principles are reflected in a communitarian notion of good citizenship, that specifies norms regarding the individual and collective actions perceived as legitimate and the rights and duties of citizens as 'good member? of a society, 'entitled to whatever prerogatives and encumbered with whatever responsibilities are attached to membership* (Walzer 1989: 211). The communitarian view stresses the notion of active participation in social life as the core element of good citizenship (Walzer 1989: 216-18) and civic virtues such as solidarity and being socially active for the good of the society (Kymlicka and Norman 1995: 294-7). According to Conover et al. (1991: 802), communitarians favour a model of citizens:
who are not so much autonomous individuals making private choices, as social and political people whose lives are intertwined ... Such 'communal' citizens share with their neighbours' common traditions and understandings, which form the basis for their public pursuit of a common good ... Communal citizen identities are 'thick' - such citizens not only have the right to participate in public affairs but also are expected to do so actively for the community's sake and for their own ... Individual rights tend to drift to the background where they are regarded as contextually defined. Duties, by contrast, are brought forward because they involve responsibilities that are to be welcomed rather than shunned.
Finally, the participatory model emphasises active political participation as the key element of citizenship which 'enlarges the minds of individuals, familiarises them with interests which lie beyond the immediacy of personal circumstance and environment, and encourages them to acknowledge that public concerns are the proper ones to which they should pay attention' (Oldfield 1990:184; cit. in Kymlicka and Norman 1995: 292-3). According to Kymlicka and Norman, 'civic republicanism' is a variant of the participatory model, conceiving participation as a value in itself rather than as a means of civic education. The rationale behind this view is that playing an active role in social and political life is morally superior and
92   Denters, Gabriel, and Torcal
more rewarding than restricting oneself to the pursuit of private pleasures. Walzer (1995: 155) assesses active participation in a collective setting as the common denominator of a participatory interpretation of citizenship: 'To live well is to be politically active, working with our fellow citizens, collectively determining our common destiny ... for the work itself, in which our highest capacities as rational and moral agents find expression'. The participatory view of citizenship was very popular in the late 1960s and early 1970s (Habermas 1961; Pateman 1970), but is still endorsed by theorists like Barber (1984) and Mansbridge (1980).
This short review of the most prominent notions of good citizenship reveals several tilings. First, civic virtues consist of rather abstract principles which cannot be easily operationalised for the purpose of empirical analyses. Second, when it comes to specific political norms or virtues, we cannot overlook a considerable overlap among competing conceptions of good citizenship. Law-abidingness, for example, is shared by traditional elitists, liberals, and communitarians; and solidarity is common to the advocates of the participatory model and communitarians.
The norms of goad citizenship: structures of citizen views
Only recently have there been attempts to link normative theories of good society and good citizenship to empirical research (for example, Staub 1989; Conover et al. 1991; Nie et al. 1996; Gabriel et al. 2002: 68-96; Rose and Pettersen 2002). The CID surveys contain a set of items tapping various components of civic orientations, thus enabling an empirical study of the ideas of good citizenship prevailing in various European countries. These items were a sub-set of a broader battery of 16 items of a Swedish survey in which four sub-dimensions of citizenship were identified (Peters-son el al. 1998).3 On the basis of our limited sub-set of items we were not able to repUcate the exact results obtained in the Swedish study. Therefore, we had to focus on only three of the four sub-dimensions: law-abidingness, critical and deliberative principles, and solidarity.
With regard to our theoretical discussion in the previous section, we would expect to End three distinct but related dimensions of citizenship rights and duties. In order to examine whether or not this theoretical distinction would be empirically valid in the countries in our sample, we conducted an exploratory principal component analysis of six items that, according to the criterion of face validity, could be considered as representations of good citizenship.4 Theoretically, we would expect to find either a coherent one-dimensional 'syndrome' of good citizenship or, alternatively, several distinct, but interrelated, sub-dimensions referring to the key norms of solidarity, critical and deliberative principles, and law-abidingness (Kymlicka and Norman 1995: 294-6; Rose and Pettersen 2002).
First of all, die loadings on the un-rotated first factor (results not reported) show that there is a common notion of good citizenship
Norms of good citizenship 93
underlying the items selected for analysis. However, a closer examination of the data included in Table 4.1 suggests a more differentiated pattern with three factors standing for the theoretically expected sub-dimensions of the idea of good citizenship. Three of the sub-dimensions stipulated by Rose and Pettersen (2002) were identified in die factor analysis: law-abidingness (not evade taxes, obeying laws), solidarity (solidarity, think of others), and critical and deliberative principles (form own opinion, be self-critical). Although these sub-dimensions will be treated separately in subsequent analyses, both tile initial, un-rotated, factor solution and the correlations among the three factors underline that diey belong to a common syndrome.
The general pattern described so far is prevalent (with relatively minor deviations) in most, but not all, countries. If we take into account that there was only a limited number of items available we can say that the three-factor solution worked out reasonably well. The results for Switzerland and Slovenia reflect the theoretically expected pattern of loadings of the six items on the three factors. Moreover, with minor deviations this pattern was also found in Sweden, Denmark, Norway, West and East Germany, and Russia (where tile item 'being self-critical' had a double loading on both the critical and deliberative principles and the solidarity factor), the Netherlands (where the item 'show solidarity' had a double loading on the critical and deliberative principles and the solidarity factor), and Romania (where 'being self-critical' loaded on the solidarity dimension but, contrary to our expectations, not strongly on the critical and deliberative principles factor). Stronger deviations were found in Spain (where not only the item 'being self-critical' had a double loading on bodi the critical and deliberative principles and the solidarity factor; but where unexpectedly 'show solidarity' loaded on die law-abidingness factor) and in Moldova (where 'being self-critical' loaded on the solidarity dimension but not strongly on the critical and deliberative principles factor, and notably, 'form own opinion' produced a double loading on the critical and deliberative principles factor and the law-abidingness factor). The strongest deviations occurred in Portugal, where it was impossible to replicate the general pattern described so far.
In conclusion, however, the pattern of loadings of the six items on the underlying dimensions of citizenship provides empirical support for the theoretical expectations. 'Being self-critical' appears to be the only item not neatly fitting into the expected pattern due to its frequent double loadings. It also becomes obvious that this item is associated with both communitarian (solidarity) and liberal interpretations of citizenship (critical and deliberative principles). Although this finding is not in line with our expectations, we should point out that it matches Putnam's notion of good citizenship that entails communitarian as well as liberal principles. In almost all remaining instances, the items roughly show the loadings we would have expected from a theoretical point of view (Putnam 1993:
94  Denters, Gabriel, and Tarcal
Table4.1 Patterns of good citizenship: results of principal component analysis per country"
Factor and country	Think of	Show	Obey laws	Not evade	Form own	Be self-
	others	solidarity		taxes	opinion	critical
Solidarity					-0.06	0.34
Denmark	0.66	0.54	-0.02	0.11		
East Germany	0.84	0.49	0.02	0.03	-0.09	0.33
Moldova	0.85	0.46	-0.00	0.02	0.45	0.74
The Netherlands	0.62	0.36	-0.00	0.03	-0.13	0.24
Norway	0.78	0.45	-0.01	0.03	0.00	0.49
Portugal	0.78	0.12	-0.07	0.06	0.02	0.73
Romania	0.89	0.37	0.04	-0.03	0.04	0.59
Russia	0.91	0.37	0.00	-0.02	-0.03	0.42
Slovenia	0.78	0.49	0.01	-0.01	-0.10	0.13
Spain	0.89	0.27	-0.02	-0.03	-0.08	0.36
Sweden	0.63	0.44	0.02	0.10	-0.06	0.40
Switzerland	0.66	0.50	0.02	0.06	-0.09	0.23
West Germany	0.86	0.59	-0.03	0.06	-0.06	0.38
Critical and deliberative principles						0.36
Denmark	-0.04	0.00	0.11	-0.04	0.59	
East Germany	-0.07	0.18	0.19	-0.14	0.90	0.55
Moldova	-0.14	0.06	-0.02	-0.00	0.39	0.12
The Netherlands	-0.03	0.31	0.04	-0.03	0.76	0.56
Norway	-0.14	0.16	0.01	0.00	0.62	0.31
Portugal	-0.15	0.09	-0.12	0.07	0.45	0.14
Romania	-0.16	0.18	0.01	0.00	0.66	0.24
Russia	-0.09	0.20	0.01	-0.03	0.75	0.36
Slovenia	-0.07	0.20	0.00	-0.02	0.86	0.58
Spain	-0.10	0.15	0.02	-0.01	0.76	0.37
Sweden	-0.09	0.08	0.08	0.01	0.62	0.41
Switzerland	-0.07	0.19	-0.03	0.06	0.62	0.55
West Germany	-0.13	0.17	0.07	-0.04	0.76	0.47
Laia-abidingness			0.73		0.13	-0.06
Denmark	0.04	0.11		0.65		
East Germany	0.01	0.16	0.64	0.74	0.09	-0.09
Moldova	0.01	0.22	0.90	0.72	0.10	-0.06
The Netherlands	0.09	0.03	0.74	0.71	0.08	-0.03
Norway	0.06	0.12	0.70	0.74	0.06	-0.05
Portugal	0.06	0.40	0.90	0.62	0.45	-0.04
Romania	0.04	0.09	0.69	0.76	0.11	-0.06
Russia	-0.01	0.13	0.77	0.81	0.02	0.04
Slovenia	0.01	0.07	0.78	0.85	0.04	-0.01
Spain	0.07	0.30	0.83	0.77	0.10	-0.04
Sweden	0.06	0.18	0.72	0.77	0.13	-0.18
Switzerland	0.07	0.04	0.79	0.61	0.13	-0.09
West Germany	0.06	0.04	0.79	0.78	0.10	-0.07
Note
a Entries arc (actor loadings (pattern matrix after oblimin rotation). Loadings equal to or higher than 0.30 are in hold. In cases where all loadings with absolute values greater than 0.30 an one component are negative (i.e. in East Germany for solidarity and in the Netherlands, Norway, Russia, and Switzerland for lausibidingness), the signs of the loadings have been reversed for ease of interpretation.
Norms of goad citizenship 95
86-91}. On the basis of these results, we will differentiate between the three aforementioned notions of citizenship in the remaining sections of this chapter.
As was expected, the correlations among the three factors in all countries turned out to be moderately strong.5 This indicates that, in the hearts and minds of larger parts of the national mass publics, the various notions of citizenship go hand-in-hand. Apparently, citizens do not consider the norms implied in the conceptions of citizenship incompatible. Another test for the fragmentation hypothesis is to consider how many people adhere to an integrated conception of citizenship, in which all three values (criticism and deliberation, solidarity, and law-abidingness) are considered important. Our analyses (not reported in a table) indicate that an absolute majority of the citizens in the countries of our survey confirm the importance of all three clusters of citizenship values. The percentage of people that endorse all three values simultaneously is highest in Denmark (80 per cent). But even in the lowest-ranking countries - Switzerland and Romania (55 per cent) - a comfortable majority of the population consider all three citizen values as important. Like the results of die factor analysis, this suggests that, at the level of individual citizens, there is no widespread cultural fragmentation.
The distribution of the ideas of good citizenship
If we now turn to the aggregate level of analysis, we can ask to what extent the citizens of particular national political communities differ in their beliefs about good citizenship. This question is of interest for two reasons. First, within a political community there should be a considerable degree of consensus over key norms and values of citizenship (as a core element of a system's political culture). From this perspective, it is important to see whether, in modern societies, such a consensus still exists (Sniderman el at. 1996; Tetlock 2002). Second, many observers have claimed that the level of support for civic norms is cause for concern. In the western world, former civic norms are supposed to be eroding, while the post-communist central and eastern European societies apparently still lack a civil society tradition and are facing problems in creating a civic culture (Putnam 1993 and 2000; further references may be found in Stolle 2003).
Notwithstanding such rhetoric of crisis, die three sets of civic norms are surprisingly widely endorsed in all societies under observation (Table 4.2) .B On a scale that runs from zero to one, the support scores for law-abidingness vary between 0.71 (in the Netherlands, West Germany, Switzerland, and East Germany) and 0.89 (in Romania). For critical and deliberative principles, the scores vary between 0.70 (in Portugal) and 0.83 (in Denmark and Sweden); and for solidarit)', between 0.63 (in West Germany) and 0.75 (in Romania).7 These consistently high scores indicate
96   Denters, Gabriel, and Torcal
Table 4.2 Average support for different citizen values"
Country Citizenship values
	Solidarity	Law-abidingness	Critical and deliberative principles
Romania	0.75	0.S9	0.77
Sweden	0.71	0.84	0.83
Denmark	0.73	0.80	0.83
Slovenia	0.71	0.78	0.83
Norway	0.73	0.76	0.79
The Netherlands	0.71	0.73	0.80
Russia	0.64	0.S1	0.78
Moldova	0.69	0.S1	0.73
Spain	0.74	0.75	0.72
Switzerland	0.64	0.72	0.79
East Germany	0.64	0.71	0.79
West Germany	0.63	0.71	0.77
Portugal	0.70	0.73	0.70
Total	0.70	0.77	0.77
Note
a Scales range fram 0 to 1. Countries are ordered by the average support For all diree values.
that in all countries the public considers the three value orientations as almost equally important components of civicness.
Nevertheless, the level of support varies For these sub-dimensions. Public support for the norms of law-abidingness and for criticism and deliberadon is clearly even more widespread than the support for solidarity. This is not surprising, since, as Rose and Pettersen (2002) rightly argued, the former two values and norms are also part of a liberal interpretation of citizenship, whereas solidarity is a distinctive element of a 'thick', communitarian interpretation of citizenship.
From a comparative perspective, there are no clearly denned clusters of countries where the public embraces a particular concept of good citizenship (for example, a predominantly liberal conception). Support for solidarity is most widespread in countries as different as Romania and Slovenia, on the one hand, and Denmark, Norway, and Sweden, on the other. For law-abidingness, Romania, Russia, and Moldova rank highest alongside Denmark and Sweden. Likewise, the value of critical and deliberative attitudes is broadly supported not only in Sweden, Denmark, and the Netherlands, but also in Slovenia and East Germany. This indicates that levels of support for civic norms in some of die post-communist countries, particularly in Romania and Slovenia, are almost as high as in some of the traditional western European democracies like Sweden and Denmark. At the odier end of the scale, some long-standing democracies in the west (notably, Switzerland and West Germany) rank relatively low
Norms of good citizenship 97
on both the solidarity and the law-abidingness scales. Given the vast differences in the political history over die last half century of these countries, diese results are striking.
However, we should not overlook the still prevalent differences between the west and die east. If we establish a rank-order of the three components of citizenship per country, we find an interesting picture. First, there is a cluster of nations (die regular printed countries in Table 4.2) in which the public gives the highest priority to critical and deliberative principles, followed by law-abidingness and solidarity. This rank-order can be found in all traditional western European democracies with the sole exception of Sweden. Two of the most westernised post-communist publics, East Germany and Slovenia, join this group. A second group, where law-abidingness ranks first and critical and deliberative values second, consists of three post-communist nations (Russia, Romania, and Moldova, plus Sweden, which are italicised in Table 4.2). Finally, there is a third pattern which is to be found in the two southern European democracies where the three different notions of citizenship are almost equally widespread and law abidingness prevails only slighdy (bold printed in Table 4.2). It is tempting to interpret this pattern in terms of die various national political traditions. The position of Sweden, however, clearly points to the necessity of caution in interpreting such patterns.
Civic norms, social involvement, and social trust Tlieoretical considerations
We now turn to our second question: what are the relations between various components of social capital and civic norms? In the previous section we have already indicated that people's involvement in social networks (and ensuing social trust) may also play a key role in civic norms. The name of Alexis de Tocqueville is inextricably bound to this line of reasoning. In his analysis of nineteenth century American democracy, he pointed to the important role of voluntary associations in the development of democratic virtues. In a de Tocquevillean spirit, both the theory of neo-pluralism and research on political culture have perceived voluntary associations as training grounds for democracy (Kornhauser 1960; Almond and Verba 1963; for a concise summary, see Stolle 2003). More recendy, the role of voluntary associations in the dissemination of civic virtues was underlined in the debate on social capital from a similar point of view." In Robert Putnam's (1993: 89-91) words:
The norms and values of the civic community are embodied in, and reinforced by, distinctive social structures and practices ... Internally, associations instil in their members habits of cooperation, solidarity, and public spiritedness ... Externally, what twentieth century political
98   Benters, Gabriel, and Torcal
scientists have called interest articulation and interest aggregation are enhanced by a dense network of secondary associations ... According to this thesis, a dense network of secondary associations both embodies and contributes to effective social collaboration. Thus ... in a civic community of associations of like-minded equals contribute to effective democratic governance.'
The effects of social involvement may be both direcdy and indirectly linked to social trust. Putnam sees social trust as a 'beneficial side-effect' of social involvement (for example, Putnam 1993: 173-4; see also, Offe 1999; Gabriel et al. 2002; Stolle 2002). Without social trust, communitarian citizenship is inconceivable: 'Interpersonal trust is probably the moral orientation that most needs to be diffused among people if republican society is to be maintained' (Poggi as quoted approvingly by Putnam 1993: 89). Therefore, support for a communitarian or republican interpretation of citizenship can be regarded as a product of voluntary activity and of social trust. However, the causal order among those variables is open to question. On the one hand, we may assume that social involvement is die determinant of civic norms and values. On the other, it is equally plausible that becoming active in voluntary associations and developing feelings of trust to one's fellow citizens can be regarded as typical attitudes and behaviours of people strongly endorsing civic norms and values.
Based on these general assumptions, organisational involvement, social trust, and civic values and norms should be positively related. Therefore it should not come as a surprise that, in Robert Putnam's (1993: 86-91) account of civic virtues, solidarity, trust, and associational involvement are considered as parts of a general syndrome, which Putnam named 'social capital'. Accordingly, people strongly involved in the voluntary sector and people who trust their fellow citizens should support civic norms and values more strongly than others - non-joiners and non-trusters (Gabriel et at 2002; Stolle 2003).
As indicated in the previous chapter of this book, however, these general formulations need to be specified. The type of involvement (mere membership, being active, and having friends) in organisations, for example, may vary. Different types of associational engagement may have different effects upon the acquisition of civic virtues. Our general assumption is that membership matters as such, but.that a closer involvement (either having personal friends in such organisations or active membership and voluntary work) has a greater effect in terms of civic virtues (Gabriel andKunz 2000; Stolle 2003)."
Empirical results
In order to test this assumption, we will now turn to the data. Following the underlying research question, we will treat the three dimensions of
Norms of good citizenship 99
citizenship as dependent variables, while we include four aspects of people's social capital in the analysis as explanatory factors. The included aspects of social capital are social trust and three aspects of organisational involvement: membership, activity, and friendship ties within organisations.1" These explanatory variables were simultaneously included in a multiple regression model (model 1). In a second step, additional variables (education, age, church affiliation, left-right self-placement, and television-exposure) that, in previous research, have often been considered as relevant predictors of civic virtues were introduced (model 2).11 This second step is aimed at getting a more complete picture of the impact of social involvement and social trust on civic values and norms while other relevant factors are held constant12
In the following pages, we will show the results of separate analyses for the three dimensions of citizenship, that is, law-abidingness, solidarity, and critical and deliberative principles. The focus here is not on the impact of each of die four aspects of social capital separately, but on two more general questions. How important are the four aspects of social capital taken together in explaining people's conceptions of good citizenship (model 1)? And how relevant are these aspects of-social capital when we take alternative explanations into account (model 2)? In a subsequent section we will turn to the more detailed question about which elements of social capital are important in shaping people's civic attitudes.
How relevant is social capital to the various components of citizenship ?
According to the results presented before, citizenship is a multifaceted concept, including the traditional norms of law-abidingness, solidarity, and criticism and deliberation. In this section we will try to clarify to what degree the various components of citizenship are linked to die four important aspects of social capital.
As shown in Table 4.3, critical and deliberative principles are only weakly related to social capital in most societies under observation. In no single country was it possible to explain more than 10 per cent of the variance by invoking interpersonal trust and social involvement. If we are even more lenient and take 5 per cent as our criterion, the four aspects of social capital only seem to have an impact in East Germany, Norway, Romania, and West Germany. As a first conclusion, it seems reasonable to state that social capital is not particularly relevant to promote citizens' critical and deliberative virtues. Including variables such as education, media use, age (year of birdi), and so on, does not greatly change die situation. More than 10 per cent of die variance is explained in Romania and West Germany, but, in general, the explanatory power remains poor.
Roughly die same applies to law-abidingness, where the results for
15 co o pi m w w cn oi 01 m m oo.ooppppppppp o'o' 000000" 00000
* * ********** as **********
1—i-—lOpr-jOOOOOi-jOtM
o'o'o" 0000000000
* *
* *
* *
.. * * * * *
* * * *
OOOOt—*opoopppi-j
O OOOOOOOOOOOO
In tH H O !Ti I" t>j (X) O tO 1m ^ O
CO 13 tD Di IT) IT) PU-' J> ^ U]
ooooopppppp*-^p
OOOO* OOOOOOO'OO'
*   * *
* *
»   » 3f
******** OCQlT-MiniDMOHCOlMWO
ooot—■ —• o q q o o h h h ooo'oooo'o'o'oooo"
***** * * * * ft
ft * * * * * * *
* *
* * ***** Trsr-jv
ooooooooooopo o'o'ooooooodooo
I
o o o p p p p p p o p o p oooo'oooooo'oo'o
I
* *
OOOpppT-Jppppp'-H
000 000 a 00* 0000
** **** **** ** **** **** * * ^ * * *
^ooLinntooHcooHO
COOOOcOlflcOlCli—1 o       C CI CO
o o o o o p p 0 p p p p p 000000 00 000 o'o"
■a
S S
u E rt 3
u 6 ""
— .3
1 E
d oj > oj      E?n rj     *-      C i, ri
S 8 °^ S    § I S-g.g|jB
•a
Norms ofgood citizenship 101
niodel 1 meet the 5 per cent criterion in the Netherlands, Norway, and Sweden. Contrary to the findings on criticism and deliberation, introducing additional variables (particularly age and religion) leads to a substantia) increase of explained variance. For model 2, five countries pass die 10 per cent hurdle (Switzerland, the Netherlands, Norway, Sweden, and West Germany), and six more meet the 5 per cent criterion (Denmark, East Germany, Spain, Portugal, Romania, and Russia). Only in Moldova and Slovenia does model 2 fail to meet the 5 per cent criterion.
Finally, die baseline social capital model accounts for more than 10 per cent of die variance in the support for solidarity in West Germany and Norway; in Denmark, this model explains more than 5 per cent. Again the introduction of control variables results in an increase in the explanatory power. In five countries the second model explains more dian 10 per cent (Denmark, East Germany, Norway, Sweden, and West Germany), and in Jive more countries die 5 per cent hurdle is surpassed (Switzerland, Spain, Portugal, Romania, and Russia).
On this basis we can draw several conclusions. First, social capital in its own right is only a weak predictor of people's support for civic values; this is true for all three dimensions of civic values. Second, our analyses indicate that, especially for law-abidingness and solidarity, the introduction of control variables improves the explanatory power of the model. Third, we can conclude that, with regard to the latter two dimensions, die models perform best in some (though not in necessarily all) of die established western democracies.
Wliat factors affect civicness?
In the previous section we have seen that adding control variables results in a substantial increase in the explained variance of the regression models (this is the case for all three dependent variables, though to a varying degree). Table 4.3, however, did not tell us much about the causal effects of the various components of social capital and die odier variables in the analysis. We do not know which - if any - of the respective components are responsible for die total effect of social capital. Neither do we know whedier the effects of the social capital factors remain significant after the introduction of the control variables. In order to simplify the analysis we have focused on a composite index that measures people's overall support for the diree distinct, though correlated, components of good citizenship.,a The higher the score on tiiis composite measure, die more supportive people are of various components of good citizenship.
102   Denters, Gabriel, and Torcal
EXPLANATORY POWER.
If we consider the evidence from Table 4.4, we see that, in terms of the proportion of explained variance, the results for model 1 (social capital components only) are far from impressive. In Norway and in West Germany, more than 10 per cent variance is explained and in four more countries (Denmark, East Germany, the Netherlands, and Sweden) the percentage of explained variance exceeds 5 per cent. When we introduce control variables (model 2) the explanatory power of the model improves; in some countries substandally (e.g. in Switzerland), in others less so (e.g. in Moldova and Slovenia). In conclusion, the extended model explains more than 10 per cent of variance in eight countries (Switzerland, Denmark, East Germany, die Netherlands, Norway, Romania, Sweden, and West Germany). The countries where the model performs less well are all located in southern (Spain and Portugal) and eastern Europe (Moldova, Russia, and Slovenia). The combinadon of these results for models 1 and 2 suggest that social capital (in its own right and in combination with control variables) is a rather poor predictor of support for civic virtues in some (though not all: see the cases of East Germany and Romania) of die countries in eastern Europe and to a lesser extent in southern Europe.
THE EFFECTS OF SOCIAL CAPITAL
When we turn to the impact of various components of social capital, our analyses show that, of the four components of social capital under study, social trust turns out to be not only the strongest, but also the most consistent factor influencing people's overall support for civic values. This is true for bodi model 1 (without controls) and for model 2 (with controls). Only in Moldova and Slovenia does trust fail to have an impact on the dependent variable. Moreover, if compared to the effects of the other aspects of social capital, social trust is by far the strongest determinant of support for civic values. Trust turns out to have a relatively strong effect (&0.10 in both models 1 and 2) in all the established western European democracies and in Portugal. In the other countries, the effects of trust are weaker (i.e. East Germany, Spain, Romania, and Russia) or absent (i.e. Moldova and Slovenia). This general pattern of results is obtained for both models (eidier with or without control variables). This indicates that factors like education and age (year of birth) do not eliminate the effect of social trust. These findings suggest that, at least to some extent, Putnam (2000: 137) was right when he claimed that 'trusting citizens are good citizens'.
If we turn to organisational involvement (either sheer membership, having friends, or active participation), we have already seen that the direct effects of these three components of social capital are eclipsed in their importance by die effects of social trust This is not to say, however,
Norms of good citizenship 103
fliat forms of organisational involvement are widiout any impact In nine of the 13 countries we found statistically significant (but consistently weak) effects of one or more of the involvement variables, even after we controlled the effects of other variables. The exceptions are Moldova, Romania, Russia, and Sweden. Moreover, we observed diat mere membership is only occasionally of importance (i.e. in the Netherlands, Norway, and Slovenia). Only when associations become a more-or-less important part of one's life (either by active participation or by having personal friends in associations), does membership gain some importance for people's support for civic virtues. We find such effects in Switzerland, Denmark, East Germany, Spain, the Netherlands, Portugal, Slovenia, and West Germany. In almost all instances, such effects survive after the introduction of control variables. This clearly implies that not just being a member in organisations counts, but that associations are only effectual if members are embedded in some dense personal networks of the voluntary sector. However, we should realise that, in terms of their direct effects on people's support for civic values, the importance of associational involvement is limited.
ADDITIONAL FACTORS
Social trust and involvement are not necessarily the only factors contributing to the support of civic norms. Some authors have pointed to the role of education (Nie et al. 1996), and odiers emphasise the importance of age, and the effects of (excessive) exposure to television (Putnam 2000: 216-76). Religion and left-right placement are also considered to be important factors (Putnam 2000). If we inspect die explained variance for model 2 in Table 4.4 (and in Table 4.3), we immediately see that, even after the introduction of additional factors, the improvement of the regression model in terms of its explained variance is only modest.
Examining the effects for each of diese additional factors separately, we first find a consistent and relatively strong (in nine cases the coefficient exceeds 0.10) age (year of birth) effect, with the older age cohorts more strongly supporting civic norms dian the younger ones. Second, the widespread assumption that education fosters support of norms of good citizenship values is not generally confirmed. Only in five instances (Switzerland, die Netherlands, Portugal, Romania, and West Germany) are the better educated indeed most supportive of civism. Third, contrary to Putnam's hypothesis, frequent exposure to television is not a systematic impediment for people's support of civic values. Only in Switzerland, Sweden, and West Germany is there some support for this hypothesis. In the other countries we either find no effect or even a positive effect of television exposure (Moldova, the Netherlands, and Romania). Fourth, church attendance is a consistent, though a rather weak, explanatory factor. Church attendants tend to be relatively supportive of civic values.
Table 4.4 Effects of four components of social capital (model 1) and of these components plus control variables (model 2} on overall support for civic values"
Country Components of social capital and model---■
Control variables
Constant R2
Social      Member in   Friends in    Active in     Age Education TV
trust        organisation organisation organisation (reversed exposure
scale)
Church Left—right attendance placement
AR-
model 2-model 1
Denmark
Model 1       0.19*** 0.00
Model2       0.15*** -0.01
East Germany
Model 1       0.08*** 0.00
Model 2       0.06*** 0.01
Moldova
Model 1 0.02 Model 2 -0.04
0.03 -0.01
The Netherlands
Model 1       0.13*** 0.00
Model 2       0.10*** 0.02**
Norway Model 1 Model 2
Portugal Model 1 Model 2
0.28*** 0.02***
0.24*** 0.02**
0.13*** 0.00
0.15*** 0.00
0.01 0.01*
0.05*** 0.05**
-0.05 -0.05
0.05*** 0.01*
0.01 0.01
0.05*** 0.04***
0.01 0.00
0.03* 0.02
0.04 0.05
-0.00 0.01
0.01 0.01
-0.00 -0.02
-0.15*** 0.01
-0.16*** -0.07*
0.00 0.02
-0.10*** 0.10***
-0.11*** 0.00
-0.08* 0.14***
- - - 0.64*** 0.071 - . 0.04 0.06*** -0.06*** 0.78*** 0.111 0.040
- - - 0.65*** 0.076 --0.03 0.14*** -0.10*** 0.81*** 0.147 0.071
0.20** 0.02
0.73*** 0.000 -0.07**  0.68**   0.013 0.013
- - _         0.63*** 0.057 -0.04*    0.05*** -0.06*** 0.68*** 0.103 0.046
- -         0.55*** 0.135 -
0.00     0.08*** -0.10*** 0.67*** 0.189 0.054
- - 0.62*** 0.047 -0.11      0.08*** -0.06**  0.51*** 0.083 0.036
Romania Model 1 Model 2
Russia Model 1 Model 2
Spain Model 1 Model 2
Slovenia Model 1 Model 2
Sweden Model 1 Model 2
0.10*** 0.03 0.09*** 0.04
0.10*** 0.09***
0.07*** 0.07***
0.00 0.03
0.17*** 0.12***
Switzerland
Model 1 0.14*** Model 2 0.11***
West Germany Model 1 0.15*** Model 2 0.16***
-0.01 0.02
0.01 0.00
-0,02 -0.04**
0.01 -0.00
-0.00 -0.01
-0.02 0.00
0.02 0.00
0.02 0.02
0.03*** 0.04***
0.02 0.00
0.02** 0.01
0.01 0.01
0.07*** 0.07***
0.02 0.02
0.02 0.01
0.02** 0.02**
0.01 0.04**
0.00 0.00
0.03*** 0.02*
0.02* 0.00
-        - - - 0.76*** 0.047 -
-0.08      0.18***      0.20*** 0.05* 0.05**   0.65*** 0.116 0.069
- - - - - 0.70*** 0.017 _
-0.22*** 0.01 0.07      0.04* 0.05**   0.77*** 0.054 0.037
- - - - - 0.69*** 0.034 -
-O.08*** -0.01        0.02      0.03***      0.02      0.73*** 0.066 0.032
-0.10*** 0.10 0.14 0.04*
0.77*** -0.00 --0.03   0.79*** 0.029 0.029
- - - - 0.66*** 0.080 -
-0.12*** 0.03        -0.18*** 0.05***    -0.08*** 0.80*** 0.127 0.047
- - - -         0.61*** 0.039 --0,24*** 0.07*** -0.19*** 0.08*** -0.10*** 0.77*** 0.133 0.094
- - -         - -         0.58*** 0.143 --0.12*** 0.04** -0.21+** 0.06*** -0.07*** 0.67*** 0.203 0.060
Nate
a Entries are adjusted R"s. Levels of statistical significance: *** = 0.01, ** = 0.05, * = 0.I.
106   Denlers, Gabriel, and Torcal
Such an effect emerges in all die nations under observation - widi die exception of Moldova. Finally, in all western European democracies, East Germany, and Portugal, people's identification with a leftist position on an ideological left-right scale is positively related to support of civic values." In Moldova, Romania, and Russia, we find the reverse effect: positions on die right are associated with support for civic values.
In addition to evidence for the important role of social capital, particularly trust and active participation or integration in voluntary associations, and in the shaping of civic attitudes, we have found diree odier major factors affecting civic attitudes: age, religion, and ideological orientation. Most of these results are well in line with Putnam's assumptions about die roots of good citizenship.
Conclusions
In this chapter we focused on two main questions. First, we asked whether the pluralism of social norms that is characteristic for modern societies results in cultural fragmentation. In answering this question we looked both at the structure of individual civic attitudes and at cross-national patterns of citizenship. After these descriptive analyses we turned to our second question that pertains to one of the main threads in this book: what is die democratic relevance of social involvement and social trust?
With regard to die first question, our findings suggest that moral pluralism has not resulted in cultural fragmentation. We found that, in the hearts and minds of citizens, the support for various citizenship norms are closely linked and that, in all countries analysed, widespread support for the norms of law-abidingness, criticism and deliberation, and solidarity exists. Moreover, we found tiiat in each of our countries the majority of citizens internalised a fully-integrated concept of citizenship, based on simultaneous strong support for each of the three basic components of citizenship. These findings confirm the analyses of Klages (2000a and 2000b), who argued that a synthesis of traditional orientations like rule conformity and modern norms and values (for example, political activism) is typical for a contemporary civic society.
With regard to our second question, we find that social trust and forms of associational involvement make a modest contribution to die shaping of civic norms in many of die countries under analysis. If we look at the overall results, reported in Table 4.4, and the results for the sub-dimensions of solidarity and law-abidingness, it is clear that the relative impact of the social capital factors (and social trust in particular) in building support for civic values is greatest in various established western democracies. This finding is in line with the 'Tocquevillean model of civic virtue' (Newton 1999a): social involvement and its 'beneficial side-effect', are related to public support for solidarity.
Norms of good citizenship 107
From diis perspective, the results for some of the eastern European countries are, however, somewhat puzzling. In the third section of this chapter we discovered that in Slovenia and Romania, for example, a large majority of the public supports norms like solidarity, law-abidingness, and criticism and deliberation. Such a result was not expected, given the clearly different social and political history of the respective nations during die second half of the twentieth century. In such societies the key mechanism of die 'Tocquevillean model', a flourishing civil society with a wide variety of voluntary organisations, is not yet fully developed.
However, as Newton (1999a) pointed out, diere are odier mechanisms that may provide social trust and civic virtues. These are mechanisms of social control in close-knit communities, or alternative institutions like the mass-media or schools; a set of mechanisms that may serve as vehicles for political socialisation. Both these mechanisms may provide part of an explanation for the eastern European findings. Whether or not this is actually the case is a matter for further investigation. If such speculations have some validity, diey dramatically highlight the vulnerability of die cultural basis of democratisation in at least some of die eastern European countries. Ironically, in the coming years the process of modernisation and the renunciation of the cultural heritage of former communism may well erode public support for some of die norms and values that are at die core of democratic citizenship in western-style democracies. Since the post-communist nations in central and eastern Europe are still on the road to die formation of social capital, they will still need a considerable amount of time before having established alternative mechanisms for securing continuing public support for key values and norms of democratic citizenship.
Notes
1 In his analysis of the values of modern societies, Klages (1984) depicts a similar pattern by contrasting conformist (hierarchy, obedience, achievement orientation) to non conformist (self-actualisation, participation, equality) values.
2 In addition to such virtues, these authors also emphasise the relevance of other values: social virtues (open-mindedness and independence), and economic virtues (work ethic, and capacity to delay self-gratification).
3 The Swedish researchers labelled these dimensions participation, deliberation, solidarity, and law-abidingness. This division is closely related to the fourfold division between political participation, critical and deliberative values, solidarity, and law-abidingness that we identified in the discussion of normative theories of citizenship.
4 Accordingly, three factors were pre-selected before rotating the initial solution. We were well aware that we were actually violating the rule of having at least three items per factor because of our limited number of items. Therefore we should treat die results of the analysis as illustrative radier than as conclusive.
5 The correlations between these factors were -0.36 (solidarity with law-abidingness), -0.42 (law-abidingness with critical and deliberative principles), and 0.44 (solidarity with critical and deliberative values). Negative correlations
108   Dealers, Gabriel, and Torcal
are due to different signs of the factor loading coefficients. Interpreted in a substantial sense, the three concepts of citizenship are positively correlated.
6 For this purpose we have computed three indices, one Tor each of the dimensions resulting from the previously reported factor analyses. Each index therefore is based on the sura score of two items. All items were rescaled so as to have a minimum of 0 and a maximum of 1. These three indices will also be used for subsequent analyses.
7 This is also confirmed by the relatively modest standard deviations for the mean support rates. For law-abidingness, with a mean score of 0.70, the standard deviation was 0.21; for criticism and deliberation, the mean was 0.77 and the standard deviation 0.18. And finally for-solidarity, the mean was 0.70 and the standard deviation 0.20. If we look at comparable figures per country, we get similar results.
8 See Coleman 1988; Staub 19S9; Putnam 1993, 2000; Verba el al. 1995; Feldman and Steenbergen 1996; Warren 1999a, 2001; Claibourn and Martin 2000; Dekker and Uslaner 2001; Gabriel et al 2002; Hooghe and Stolle 2003b; and Stolte 2003.
9 In this chapter we will not analyse the potential effect of people's involvement in different types of organisations. Although previous work (for example, Putnam 1993; Foley and Edwards 1996; Cohen 1999; Putnam 2000: 31-115; Stolle and Rochon 2001; Stolle 2003) has shown that not all types of organisations will have the same impact on the transmission of civic virtues, a detailed analysis of this topic would be beyond the scope of this chapter.
10 Information on operationalisation is provided in Chapter 2 for social trust and in Chapter 3, note 13, for the measures of organisational involvement
11 This may be seen as problematic, because there is likely to be reciprocal causality between the dependent and the independent variables. For our purposes, however, it is not necessary to get an exact estimate of the relative strengths of the reciprocal causal effects; therefore, we can ignore this complicating factor.
12 Therefore, the results of the extended models will be examined only:
1 to see if controls affect the impact of our two key explanatory variables;
2 to establish the relative explanatory power of these two factors vis-á-vis the control variables;
3 to gauge the gains in additional explanatory power by introducing these additional factors.
13 The use of this composite index is justified by both the relative strength of the correlations between the three factors and by the basic similarities in Ure analytical results reported in Table 4.4. The composite index is computed as the average score of the three separate indices.
14 A low score indicates a position on the far left; a high score indicates a position on the far right
5   Political and social tolerance
Jose Manuel Leite Viegas
Introduction
In the final quarter of the twentieth century, a considerable number of countries made the transition from totalitarian or authoritarian regimes to democratic government, particularly in southern Europe, eastern Europe, and Latin America. While the introduction of democracy does not automatically imply respect for the rules and the law that the constitutions stipulate, comparative empirical studies show that support for democracy has risen among the citizens of these new democracies (Klingemann 1999; Norris 1999b) as has the acceptance of basic democratic values (Dalton 1994; Thomassen 1995).
At first glance, we might think that tolerance would be included in these trends. After all, political tolerance springs naturally from the egalitarian and pluralistic principles of liberal democracy: respect for individual liberties, equality before the law, acceptance of diverse opinions and lifestyles, rejection of arbitrary conduct in complying with the law, and the endorsement of minority rights.
, The problem, however, is in fact somewhat more complex. Democratic pluralism is based on the recognition that modern societies are heterogeneous and that social differentiation may entail positive results in terms of social and political progress. But the majority principle, another rale of democracy, may also jeopardise respect for the opinion and behaviour of less popular or minority groups in society, a risk that Alexis de Tocqueville
([1835, 1840] 2000) clearly emphasised. It was recognised early on that political intolerance is not restricted to non-democratic societies. It represents a serious problem, which has been analysed empirically in contemporary democracies, particularly in the postwar period.
Until the work of Sullivan et al. (1982), empirical analyses concurred with the thesis that political tolerance had tended to increase in democratic countries in the postwar period, mainly due to higher levels of education among their citizens (Stouffer 1955; Davis 1975; Nunn el al. 1978). Later studies confirm this tendency. They also confirm the importance of education and date of birth, both of which are of course associated with