Helmut Anheier and Jeremy Kendall

Interpersonal trust and voluntary associations: examining three approaches

ABSTRACT The relationship between interpersonal trust and membership in voluntary associations is a persistent research nding in sociology. What is more, the notion of trust has become a central issue in current social science theorizing covering such diverse approaches as transaction costs economics or cognitive sociology. In different ways and for different purposes, these approaches address the role of voluntar y organizations, although, as this paper argues, much of this thinking remains sketchy and underdeveloped. Against an empirical portrait of this relationship, the purpose of this paper is to assess such theorizing. We rst set out to explicate major approaches to trust in economics, sociology and political science, using the non-pro t or voluntary organization as a focal point. We then examine the various approaches in terms of their strengths and weaknesses, and, nally, identify key areas for theoretical development. In particular, we point to the social movement literature, the social psychology of trust, and recent thinking about civil society.

KEYWORDS:

Trust; voluntar y associations; non-pro t sector; social capital

INTRODUCTION

The notion of trust is one of the most topical issues in current social science. In many western countries, this higher pro le may stem, in part, from the seeming erosion of popular trust held in institutions like the government, the media, the churches or the family, frequently documented in public opinion polls in recent years. For example, general interpersonal trust levels in both the USA and Britain were lower in the 1990s than they were in the 1980s, as is con dence in government, the press or in large corporations (see Pharr and Putnam, 2000; Inglehart 1999). What is more, the experience of wide-spread popular distrust in the political and economic systems of Eastern and Central Europe, and their ultimate breakdown in the late 1980s, may have contributed to a general sentiment that
British Journal of Sociology Vol. No. 53 Issue No. 3 (September 2002) pp. 343362 2002 London School of Economics and Political Science ISSN 0007-1315 print/1468-4446 online Published by Routledge Journals, Taylor & Francis Ltd on behalf of the LSE DOI: 10.1080/0007131022000000545

344

Helmut Anheier and Jeremy Kendall

trust is a fragile element of modern societies (Beck 1992). Although we could add many more examples from political scandals to fraudulent business deals and professional malpractice contemporary cultural diagnoses suggests that trust is a problematic element of the modern zeitgeist (Habermas 1985). Three widely cited recent publications have further increased the attention social scientists pay to trust. In contrast to the problematic notion of trust in cultural discourse, these books try to show how the vibrancy and developmental potential of society is rooted in everyday mechanisms that generate and maintain trust. In Making Democracy Work, Putnam et al. (1993) suggests that dense networks of voluntar y associations are the main explanation for northern Italys economic progress over the countr ys southern parts. In Bowling Alone, Putnam (2000) looks at participation in voluntar y associations in the USA and argues that a dramatic decline in both membership rates and other forms of civil engagement led to lower levels of trust in society and, consequently, to general increases in social ills such as crime. Making a more general argument in his book Trust, Fukuyama (1995) shows that differences in economic success among the USA, Germany or Japan are predicated on reservoirs of sociability and social trust, which, in turn, depend on some kind of associational infrastructure. Such recent formulations tend to be suggestive of a signi cant relationship between voluntary associations and trust be it in Putnams analysis of Italy, his diagnosis of current US culture, or Fukuyamas study of major industrial economies, or other current work on social capital (Dasgupta and Serageldin 1999; Halpern 1999). According to this thinking, voluntary associations form part of the social infrastructure of society that makes the generation of trust possible, and that at least makes it easier for trust relations and trusting attitudes to develop and to re-enforce themselves within the population. Indeed, the relationship between interpersonal trust and membership in voluntar y associations is a persistent research nding in sociology, particularly in population surveys. For example, the 19992000 wave of the European Values Survey1 shows for 28 of the 32 countries participating in this effort a positive and signi cant relationship between the number of associational memberships2 held an interpersonal trust.3 Irrespective of the level of general interpersonal trust in a countr y, the survey results suggest that members of voluntary associations have higher trust levels than non-members, and that these differences widen as af liation rates increase. The summary results from the European Values Survey reveals a striking pattern, as Table I shows. Respondents reporting three or more memberships were twice as likely to state that they trust people than those holding no memberships. Overall, there is almost a linear relationship between increases in membership and the likelihood of trusting people. Nor is this nding limited to the speci c question about interpersonal

Interpersonal trust and voluntary associations

TABLE I:

345

Interpersonal trust by membership in voluntary associations (%)

Number of memberships held None One Two Three or more % % % % 23 77 18,661 (100%) 30 70 9,114 (100%) 40 60 4,056 (100%) 51 49 4,930 (100%)

Statements about trust:

Most people can be trusted Cannot be too careful Total N = 36,261

Source: European Values Survey 2000.

trust used in the European Values Survey. For example, in the USA, a similar pattern emerges in relation to the question Do you think that most people would try to take advantage of you if they got a chance, or would you say that most people try to be fair4 Results show that ever y second (46 per cent) respondent with no memberships felt that people would tr y to take advantage, as opposed to ever y third (37 per cent) for those with three memberships, and nearly ever y fourth (29 per cent) for those ve and more memberships. Vice versa, 70 per cent of respondents with ve or more memberships feel that people tend to be fair, compared to only 54 per cent for those with no memberships.5 These persistent ndings do indeed suggest a general sociological pattern, and with it three closely related questions that need to be addressed within and across countries First, what factors explain differences in af liation rates? This question refers to the set of independent variables such as religion or social status that increase or decrease the likelihood of joining, thereby generating speci c membership patterns in the population (McPherson et al. 2001). Second, what are the wider consequences of higher trust levels generated by associational membership? This brings in focus the range of outcome variables like economic performance, political stability or con ict, or measures of social cohesion and quality of life (Dekker and Van den Broek 1998). Finally, the question that will be the primar y focus of this article: how do we explain the intermediary step of actual trust generation? This mechanism appears as the crucial link in the implied causal chain. In other words, what is it about voluntary associations that facilitates trust? The role of voluntar y associations as the intervening factor in trust generation is, however, largely left unspeci ed by current sociological thinking. Putnam (2000: 1359) and others like Fukuyama (1995) and Dasgupta and Serageldin (1999) tend to remain silent on how and under what conditions voluntary associations generate and preserve trust for their members and more widely for society as a whole.

346

Helmut Anheier and Jeremy Kendall

Parallel and independent to the more sociological research mentioned so far, a separate effort based on the economics of information and transaction costs, closely allied to legal scholarship, emerged in the 1980s, and has gathered momentum since then.6 In this approach the concept of trust plays a central role in explaining the existence and behaviour of non-pro t organizations (see Rose-Ackerman 1996). Yet this line of economic thinking and the current sociological interest in trust have not been brought together, and the two elds remain Balkanized from each other. This would, therefore, seem to be a timely juncture at which to review both perspectives, and to examine the contribution non-pro t theory can make to the study of trust speci cally. To this end, we rst review major approaches to trust in economics, sociology and political science, while using the non-pro t or voluntary organization as a focal reference point throughout. We then assess the various approaches in terms of their strengths and weaknesses, and, nally, identify key areas for theoretical advancement in an effort to enrich current theorizing.

THREE APPROACHES

To make our task manageable, we initially focus on the dominant formulations available in the literature from economics/law, sociology, and political science, particularly from the 1980s onwards, when non-pro t theories began to take shape.7 We do not attempt to be comprehensive, given the vast and disparate literatures involved, but rather concentrate on three approaches which are particularly prominent for the task at hand: contract failure theor y in economics, the concept of trust in cognitive sociology, and the study of social capital (see Table II). Legal-Economic Approaches While economists have tackled trust in a variety of ways (Anheier and Kendall 2000), two approaches have probably exerted most in uence. Following rational choice logic, non-pro t organizations are seen as a response to the demands of individual agents seeking solutions to information problems in particular industries in free market economies.8 Hansmann (1980), combining legal scholarship with economic reasoning, was the rst to develop the argument most fully. His starting point was that health care, social services and some other policy elds are inherently characterized by information asymmetries. The role of non-pro t organizations as major providers of such services in the USA was hypothesized to stem from their greater trustworthiness over pro t-seeking enterprises. This, Hansmann avers, is predicated upon one particular feature of their constitutional design: the non-distribution constraint, i.e., the prohibition of distributing pro t to owners as income. This constraints acts as a crude but effective signal to consumers about the motives, intentions and

TABLE II: Role of voluntar y organizations Key strengths Parsimonious; theoretically grounded; trust substitutions; trust as medium VOs are a Individual actors Thin trust response to and their utility market failure; functions Institutionallynon-distribution based trust constraint signals trustworthiness Thick trust Characteristicbased trust VOs make use of Institutions, symbolic reproutines, resentations of individuals pre-existing trust; institutionalized enactment of trust routines Interstitial networks linking institutions, and citizens Thin trust Process-based trust VOs as incubators of values, civic attitudes, and styles of organizing Actors and units of analysis Applies primarily to:

Three approaches to the study of trust

Key weaknesses Preferences are given; relationship between trust and risk unclear in its consequences

Approach

Central reference

Core de nition

Economics

Market transactions

Trust is an ef cient mechanism to economize on transaction costs

Interpersonal trust and voluntary associations

Sociology

Social order

Taken-forgranted, socially constructed reliability; a priori assumption of trust

Origin of trust; different forms of trust

Static; lack of parsimony; no substitutions for trust speci ed

Social Capital (Political science)

Networks of social ties

Trust as social capital, as civil virtue

Dynamic; focus on unanticipated consequences of trust; crosscultural applicability

Unclear on vertical and non face-to-case trust; power relations neglected

347

348

Helmut Anheier and Jeremy Kendall

behaviours of suppliers. This signal offers a sort of guarantee that nonpro t organizations, unlike businesses driven by the imperative of maximizing shareholder value, lack incentives to cut corners on quality or otherwise take advantage of the user vulnerability. Thus, non-pro t organizations are more immune against moral hazards than for-pro t rms would be under similar circumstances because of a particular aspect of their legal structure. Ben-Ner and Van Hommissen (1993) go further, stressing positively the control rights of stakeholders, rather than merely the absence of incentives to opportunism. They view voluntary organizations as coalitions of stakeholders who have a strong simultaneous interest in both the demand and the supply side of a particular service. If founders remain in control of the organization and have a direct stake in output, incentives to cut corners on quality are not only reduced because of the non-distribution constraint, but because to do so would be to in ict self-harm (Ben-Ner and Gui 1993). Thus, trust is protected twice, from the supply and the demand side. In sum, the short answer economists give to the higher trust levels among members of voluntar y associations would rest on the greater trustworthiness of these organizations and, by implication, the lower exposure of members to moral hazard problems. Voluntary associations shield members and clients from bad experiences, and provide opportunities for trust-building and trust-enhancing experiences. From a comparative perspective, economists see the non-pro t form as one way of either generating the trust which the presence of otherwise high transactions costs and goods with experience characteristics renders necessar y, or substituting for it. Other possibilities include formal structures or institutions involving direct control (Krashinsky 1986; Ben-Ner and Van Hoomissen, 1993), public ownership or strict state regulation and inspection (Hansmann 1981, 1996), professional control or self-regulation (Krashinsky 1986), or co-operatives (Ben-Ner 1994; Barbetta 1997). Without departing from rational choice logic, the transaction cost approach can be criticized for a number of reasons. While there is some evidence that voluntary organizations behave differently from for-pro t rms in a fashion that is broadly consistent with economic theory, the observed relationships could also be explained by other factors (Galaskiewics and Bielefeld 2002). More importantly, however, the theory fails to take account of different forms of trust identi ed by sociologists, and these differences can matter more than signals of trust like the non-distribution constraint. For example, the contrast between the deferential trust and the more active and re exive forms of trust is not addressed by the theor y. Deferential trust is associated with traditional, hierarchical institutions (prototypical would be welfare services run by Trinitarian churches), and re exive trust more with participatory organizations and social movements (Giddens 1990; Beck 1992). Moreover, the distinction between thick (based on closely knit ties of family and friendship) and thin (professional networks, acquaintances) trust is also left unspeci ed (Putnam 2000: 136),

Interpersonal trust and voluntary associations

349

as is Zuckers (1986) typology (reviewed below). In all of these distinctions, trust plays different roles and rests on different motivations and value bases. Yet these differences are not recognized by legal-economic theor y where trust is basically a functional element facilitating transactions that would otherwise be more complicated and less ef cient. Heavy emphasis is placed on formal legal structures and in the non-pro t sector case the non-distribution constraint (Hansmann 1996), or members control rights (Ben-Ner 1994) as ultimately the originator and sustainer of trust. Sociological Approaches Sociology takes a different approach to the notion of trust, to a large extent re ecting its treatment of rationality as a variable, rather than as simply axiomatic as in rational choice theories. Sociological approaches to trust emphasize the ability to take for granted the relevant motivations and behaviours of others. It is a pre-rational, pre-existing trust that Durkheim (1933) observed underlying all contracts: behind every contract is a host of tacit agreements, and while they are not formally speci ed, they are nevertheless assumed to hold in contractual arrangements. As Giddens (1990: 33) suggests, the notion of trust as presumed reliability, is ver y different from the rationalistic, or risky conceptions of trust in rational choice approaches. From a sociological perspective, such an instrumental calculation reduces trust misleadingly to a matter of risk assessment, and misses the point, since all trust, is in a certain sense blind trust (op. cit.: 33). The sociological recasting of trust as a taken-for-granted belief distinct from risk assessment has important implications for the relationship between trust and voluntary organizations. Trust is seen not as a matter of legally framed transactions between contracting parties but as generated by social structures not reducible to instrumental individual actions, and supported by a normative infrastructure. The notion of trust as a taken-forgranted assumption of reliability takes on an ethnomethodological perspective. For example, the analysis of Anheier and Romo (1992) suggests some types of voluntar y associations, particularly religious ones, are well positioned to draw upon pre-existing trust that is unlikely to be questioned in the course of transactions. In seeing members through extended periods of taste formation and socialization, non-pro t organizations shape, develop and re-enforce deep-seated convictions, values and beliefs. Thus, as socialization agents, they implicitly provide guidelines on whom to trust, whom not to trust, and in what circumstances. Moreover, religious institutions in particular are strategically located along major events during the life course (birth, marriage, severe illnesses, death) which they underscore with routinized and stylized rites such as sacraments that instil and reinforce trust.9 Thus, the sociological counterpart to rational choice approaches sees voluntar y associations as a vehicle for the enactment of pre-existing, socially embedded trust and as a re-enforcement mechanism of trust through

350

Helmut Anheier and Jeremy Kendall

organizational rituals and cultures. Yet voluntary associations are not alone in performing this function. Now, unlike economists, sociologists tend to look not to other, formal institutions or sectors but to less formal forms of associating like kinship, friendship circles, and business contacts as their point of departure. Indeed, voluntary associations for sociologists are not even among the prime engines of trust generation: these are the family and other agents of primary socialization. Voluntary associations are important, of course, for trust maintenance, but they are second-order institutions when it comes to the generation of trust as a taken-for-granted assumption of reliability. Informal social structures are thus axiomatically prior to individuals decision making choices as between formal modes of organization. Not surprisingly, given such reasoning, few sociologist establish a link between trust and voluntary associations that is as direct and explicit as the rational choice approach in economics. Tonkiss and Passey (1999) are in a minority when they argue that trust forms the basis for voluntary association itself, and suggest, following Luhmann (1988), that trust rests on shared values in contrast to the interest-based notion of con dence, which involves contractual relations among participants. In fact, sociologists have moved away from treating trust as one singular phenomenon and have begun to examine different dimensions and facets of trust in differing contexts and circumstances. Putnams (2000) distinction between thick and thin trust is important in this regard. Thick trust is much closer to the pre-existing notion of trust ethnomethodologists speak about. It is embedded in highly personal relations that usually form the densest part of an extended network of family and friendship ties. By contrast, thin trust or social trust, based on everyday contacts, professional and acquaintance networks, involves a much greater number of ties that form less dense relations. Behind thin trust or trustworthiness is, according to Putnam (2000), the notion of the generalized other and the standing decision to give most people even those one does not know from direct experience the bene t of the doubt (Rahn and Transue 1998: 543). Voluntary associations, therefore, help bring about the conditions under which the generalized other can develop simply by giving members more opportunities for positive experiences with others under the controlled circumstances of shared interest. More generally, sociological theorizing sheds useful light on the distinction between thick and thin trust by showing how the meaning and enactment of trust varies according to geographical and temporal context. Zucker (1986) sees a general shift from particularistic ( = thick) trust based on individual characteristics to trust based on process and experience, and then to more generalized institutional trust as societies develop, i.e., thin trust. Similarly, Beck (1992), Giddens (1990) and Seligman (1998) suggest that the re exive, thinner trust in modern societies is different from the thicker but essentially fatalistic trust in earlier times. Zuckers classi cation (1986) helps re ne these distinctions: Characteristic-based trust is tied to a person, depending on characteristics such as background or ethnicity.

Interpersonal trust and voluntary associations

351

It rests on similarity in culture, values and behaviour. Process-based trust is tied to past or present exchanges as in reputation or gift exchanges and involves an incremental process of building trust and presupposes a degree of stability and reliability in mutual expectations (Zucker 1986: 60); and, nally, institutionally-based trust, i.e. Luhmanns con dence, which is based on institutions such as certi cations, form characteristics, or legal constraints. The importance of the four types of trusts changes with the development from pre-modern to modern societies. For Seligman (1998) and Zucker (1986), modern societies generate more institutional trust but lower levels of characteristic-based trust created through family and friends. Processbased trust, too, is more dif cult to generate as expectations are more varied and levels of heterogeneity in societies have increased signi cantly (see Lane 1998: 1112). As a result, and combined with weaker sanction mechanisms, modern society may have lower levels of predictability (see Beck 1992).10 These arguments are relevant for theorizing the linkages between voluntar y associations and trust. As organizations preserve to some extent the conditions present at their establishment (Stinchcombe 1965), the composition of organizational forms in a particular eld may re ect a mixture of these forms of relations.11 Typically, voluntar y associations include forms originating in the pre-modern, traditional era; organizations mobilized around a particular actors background, ethnic or religious characteristics; and entities, which essentially represent the institutionalization of friendship networks and bonds of civility which cut across ascriptive categories. In sum, sociological theorizing on trust lacks the parsimony of rational choice approaches; does not address the choice between formal sectors (for-pro t, voluntar y, government); and has not put forth strict arguments for establishing a link between macro-level characteristics of trust embedded in culture and shared values, and its micro-level level manifestations in modern societies. The main achievement of sociological theor y in this eld has been to identify different forms of trust, and to link these to distinct institutional settings. For the purposes of this analysis, the link between process-based and institution-based trust is critical. First, processbased trust is facilitated by voluntary organizations through encounters and transactions among members pre-sorted by similarity of interest; and, second, institution-based trust is signalled by form characteristics such as charitable non-pro t status, democratic decision-making, or mission statements. Social Capital Approaches Perhaps ironically, more recent accounts ultimately rest on arguments from economic sociology, in particular Colemans (1990: chapter 12; 1993) rational choice approach. Social capital refers to networks of relations linking individuals and organizations that involve some element of trust. As

352

Helmut Anheier and Jeremy Kendall

such networks and contacts are accumulated over time, social capital is seen as a resource established not only for short-term economic gains but also with the longer-term view of status competition and strategic in uence seeking in mind (Bourdieu 1979). Voluntary organizations become instruments for the formation of social capital. As participatory organizations they facilitate social connections and co-operation, and by virtue of repeated interactions they engender trust among members. In carrying Colemans understanding forward, Putnam (Putnam et al. 1993) suggests that voluntar y organizations are distinctive institutions in their capacity to function as repositories for all the other sources of social capital too: obligations and expectations, information potential and norms and sanctions (op. cit.: 89; 2000, chapters 3 and 8). As such, they are characterized as incubators of civic virtue, which Brennan de ned as the dual attribute of a capacity to discern the true public interest and a motivation to act as the public interest, so discerned, requires (1997: 259). As Putnam himself puts it, virtuous citizens are helpful, respectful, and trustful towards one another, even when they differ on matters of substance (Putnam et al. 1993: 889). Indeed, a neo-Tocquevillian perspective is emerging that sees a strong and vibrant civil society characterized by a social infrastructure of dense networks of face-to-face relationships that cross-cut existing social cleavages such as race, ethnicity, class, sexual orientation, and gender that will underpin strong and responsive democratic government (Edwards, Foley and Diani 2001: 17). Norms of reciprocity, citizenship, and trust are embodied in networks of civic associations. Moreover, Sirianni and Friedland (2001) argue that these inter-personal and inter-associational networks are key sources of social, cultural and political innovation in countries like the USA. In this vein, Putnam argues that democratic, horizontal (i.e., nonexclusionary) voluntar y associations characterized by high level of face-toface interaction are involved in a virtuous circle with trust. This bridging social capital is closely related to thin trust, as opposed to the splitting social capital of thick trust. Bridging capital is formed by instilling habits of co-operation, solidarity and public-spiritedness, and it presents inculcate skills for social, economic and political activities of all kinds. Moreover, bridging capital prevents factionalism through crosscutting ties qua association membership. In sum, membership in voluntary associations creates thin trust and helps preserve and reinforce both thin and thick trust, thereby explaining the higher trust levels among members seen in Table I. What are the shortcomings of this approach? There is the dif culty with its use of a functional de nition: by implicitly de ning social capital in terms of its consequences, it appears to have a circular or tautological character. Putnams modi cation of Colemans approach, on the other hand, avoids these dif culties but at a price. The problems arising from con ating norms and values, networks, and their consequences lead to a loss of conceptual clarity (Newton 1997: 583).12

Interpersonal trust and voluntary associations

353

Second, the concept of trust appears at one level diffuse and overgeneralized, yet at another uni-dimensional and highly speci c, characterizing trust as an emergent property of face-to-face interactions. In particular the emphasis on local horizontal associations fails to engage with the modern, re exive kinds of trust sociologists have identi ed. National social movements, which in the USA, UK, Sweden or Germany have been important cultivators of more modern forms of trust are much less central in Putnams thinking (Putnam et al. 1993; Putnam 2000). Yet, exploiting modern technological opportunities, these have developed mediated and abstracted relations, building symbolic communities which do not entail fact-to-face relationships, but nevertheless provide a social resource for those involved (Minkoff 1997; Keane: 1998). More generally, social capital approaches, at least as currently operationalized, rarely put the role of associations in the context of, or in comparison with, other ways of creating and maintaining trust. Neither the formal institutions upon which economists have focused, nor the informal relations which have preoccupied sociologists are accounted for; nor do they consider the role of the broader voluntary sector with its inheritance of more vertical organising principles. For example, in Making Democracy Work, Putnam (1993) relied on indicators of the densities of participation in cultural and recreational associations, all of which are implicitly assumed to involve continuous face-to-face interaction. He therefore ignores around two thirds of the Italian non-pro t sector, even using a de nition exclusive of political parties, trade unions and religious organizations (Barbetta 1997: 17677). In other words, social capital approaches focus on a rather narrow range of voluntar y associations.13

CONCLUSION: TOWARDS A SYNTHESIS?

Each approach shows strengths and weaknesses, as summarized in Table II. For example, parsimony is a key strength of the economic approach, together with its ability to conceptualize the non-pro t form as one institutional solution to situations that call for trust. The weaknesses are largely its exclusive emphasis of micro-economic phenomena, neglecting connections to macro developments; insensitivity to trust as a phenomenon whose meaning and functioning varies over time and space; and its assumption of the existence of a stable and well enforced legal framework. The sociological approach to voluntar y organizations and trust has remained underdeveloped as a general theory, but has added useful distinctions and classi cations of trust. Finally, the social capital approach shows the functional importance of trust for society, but ultimately addresses trust too implicitly, and rests on a contingent, indirect relationship between the creation of trust and the creation of social capital. What theoretical modi cations and innovations are needed to preserve the explanatory power of each while overcoming their respective

354

Helmut Anheier and Jeremy Kendall

limitations? 14 We suggest that three other strands of social science thinking may well provide some of the missing links and connective tissue needed for cross-fertilization among the three approaches (Figure I). We will discuss each case in turn, beginning with social capital. This approach, with its focus on networks, personal ties and civic virtue, is a mixture of sociological, political science and rational choice thinking. Voluntar y associations are portrayed as important incubators of the values and attitudes needed in modern societies. While we find much to recommend in the social capital approach, including its imagination in connecting micro phenomena to macro outcomes, its major weaknesses are a fixation on local, geographically defined communities, and a failure to specify clearly how micro and macro aspects are linked to one another. We suggest that these limitations can be overcome if the social capital approach is conceptually nested in the concept of a wider civil society. For political scientists like Keane (1998: 5), civil society is a complex and dynamic ensemble of legally protected non-governmental institutions that tend to be non-violent, self-organizing, self-re exive, and permanently in tension with each other and with the state (1998: 6). This tension-ridden and con ictual associational infrastructure creates opportunities and mechanisms for the generation of trust among some citizens as either individuals or by virtue of their membership in organizations. These opportunities in the form of social inclusion and participation in extra-familial networks, in turn, create social capital, which become a major factor in social mobility at the individual level (Bourdieu, 1979) and for economic

Transaction costs

Social psychology

Civil society

Social routines and values

Social movements

Social capital

FIGURE I:

New approaches to the study of trust

Interpersonal trust and voluntary associations

355

advancement of entire population segments more generally (DiMaggio and Mohr 1985). Such opportunities may lead to the creation and maintenance of trust under two circumstances: rst, if forms of social inclusion, participation and capital formation enforce beliefs in the basic legitimacy of the social order and the political system as rightful expressions of fundamental values; and second, if they strengthen con dence in the operation of society as a reliable and predictable system. The central point is that the relationship between trust and social capital is highly conditional, i.e., dependent on the structure of civil society and the legitimacy of the political system, and indirect, i.e., mediated through processes like social inclusion and participation. This is one reason why a strong and positive relationship between trust and membership in voluntar y associations failed to materialize in countries such as the Ukraine and Belarus (see footnote 1), where the legitimacy of the social and political order is questioned in fundamental ways. Other examples are, based on results from the World Values Survey 199597, Nigeria, at that time under militar y dictatorship; Brazil, suffering from immense problems of social inequities and exclusion; or Bosnia, which experienced severe ethnic strife and the beginnings of a full scale civil war when the survey was carried out. If the link between trust and voluntary associations were essentially a re ection of their legal form as implied by legal-economic theorists, then the relationship between membership and trust would be predicated on the availability of a mature legal system to enforce the non-distribution constraint and enforce appropriate regulation. In countries with high scores in this regard (see Salamon and Toepler 2000), we would expect a correlation to emerge between membership and trust; but this should be absent in countries where that system has remained under-developed. However, a civil society perspective insists on setting this relationship in a wider context, accounting for the political character and in particular, the systemic legitimacy of each social setting. Under this view, if and only if there is general con dence in the system, greater trust will be associated with voluntary association membership, whether or not they happen to have developed non-pro t legal formats in a mature way. Whereas the relationship between trust and social capital at the macrolevel is indirect and remains somewhat abstract even if set in the wider context of civil society, the link between trust and opportunities for social engagement qua voluntar y associations may be more direct at the individual level. It is in this context that cognitive psychology and the social movement literature offer useful insights. Psychologists have explored how motivations and actions depend in part upon language and labels. Taylor-Gooby (1997) reviews the literature on market-like game situations, and concludes that markets do not immediately impose a crude rationality on their participants (op.cit.: 10). At the same time, people appear to learn rapidly, and adapt their behaviour according to the moves of other players and what they perceive to be the

356

Helmut Anheier and Jeremy Kendall

norms of the situation. They can learn to act pro-socially as well as sel shly, depending on the cues that the rules of the game and their co-players behaviour provides, a situation characterized by real fragility. So, whether trust can ourish depends upon how the relevant actors perceive their environment. Social psychologists have tried to compare across different situations and context conditions, with the competitive market as just one possibility. The results support the proposition that language and presentation signi cantly shape the likelihood of social co-operation.15 Situations in which the rules of the game are respected and mutually understood, and where collective action is required and achievable in co-operation rather than zero-sum competition, trust may be both created and sustained. This nding ts with the economic understandings of voluntary associations reviewed above, and the imager y of caring, altruism and empathy in social policy discourse.16 Moreover, the social movement literature provides some clues as to why there may be more opportunities for encouraging collective action in some types of voluntary organization than in other social settings. McAdam et al. (1988) and others suggest that two crucial factors in uence the success or failure of social movements: resource mobilization and framing. The theory of resource mobilization assumes that movements can most ef ciently enlist nancial and human resources for their purposes if they develop along existing social structures (Neidhardt and Rucht 1993; Stevenson and Greenberg 2000). For example, movements develop within and between existing organizations such as church groups in the former Communist regimes of Eastern Europe. Frequently, social movements form large networks linking informal groups with more formal associations to create what McAdam et al. (1988) label a movement industr y capable of spanning numerous non-pro t advocacy organizations and service providers over time, e.g., the womens movement or the environmental movement. Movement success requires cognitive and cultural translation of the movements intentions and objectives in a sustained way (Passy and Giugni, 2000). Both have to be readily integrated into the patterns of interpretation and meanings of members and supporters without creating major cognitive dissonance or fundamental contradictions. Particularly important is the process of framing (Goffman 1974; Snow et al., 1986), which describes the routine use of interpretive schemas to recognize, integrate and evaluate facts, actions, statements and programmes. These frames allow the integration of a broad spectrum of experiences, events and themes to a more or less coherent and meaningful cognitive structure. Members of organizations rooted in communities based on either geographic proximity, mutual interest, or shared values would have a competitive advantage over outsiders in knowing the relevant mindsets, life situations, aspirations and problems. This makes the establishment of

Interpersonal trust and voluntary associations

357

frame resonance (Snow et al. 1986) easier to achieve. Being one of us and speaking the right language by using the right terms and labels in addressing members is a necessar y condition in generating interpersonal trust and a readiness for collective action. Yet what are the mechanisms that link trust, organization, and frame resonance? Here we turn to social psychology, in particular work on intrinsic motivation (Deci and Ryan 1985).17 Social psychology recognizes the importance of voluntarism per se, and argues that, to a considerable degree, voluntary activity is undertaken irrespective of external or extrinsic rewards. Policies which seek to introduce the latter, or cast external judgment on those involved, may be dangerous if they communicate to the participants that they cannot be trusted to undertake the activity adequately if it is pursued voluntarily. This is so because individuals perceiving external interventions as controlling can feel overjusti ed unless they downgrade their own intrinsic motivation. This dilemma is made worse if the introduction of external control does not carr y with it acknowledgement of the intrinsic motivation. Then, the person can feel that his competence is not appreciated which leads to an impaired self-esteem, resulting in reduced intrinsic motivation (Frey 1997: 1045).18 Finally, the social psychology may have particular relevance to one of the core elds in which voluntary associations are very prominent sports and leisure but for which economic trust-based theories appear to be of limited relevance. Deci and Ryan (1985) summarize how . . . the few surveys that have explored the reasons why people engage in sports suggest that for the average amateur, intrinsic factors are dominant in their motivation (op. cit.: 314). This clearly involves people participating neither to provide trust goods to other people in need, nor engage in civil virtues, but voluntarily interacting socially with others in a way that brings its own, internally experienced rewards. This, it is argued, is because of the experience of self-control and freedom from the pressures of other contexts that it brings. These activities are generally engaged in freely, and afford one the chance to stretch capacity and build skills . . . and they represent a possibility for recovering the self-esteem that is lost in the work lives of many individuals (Deci and Ryan 1985: 3134, our emphases). Thus, this literature suggests that economic arguments that focus on trust as an attribute of goods and services are limited in their applicability to voluntar y action as a whole, since they allow little room for the intrinsic motivations. Trust generated under such circumstances may then be an unintended byproduct of sel sh behaviour. Like the civil society approach, this line of argument rejects the legaleconomic suggestion that trust and membership will follow from setting in place well developed legal arrangements for policing organizations, and suggests that the relationship is fundamentally contingent. However, the focus now is at a more micro level, and suggests that the link will hold if

358

Helmut Anheier and Jeremy Kendall

and only if membership is experienced as freely chosen, and having a genuinely voluntary character which imparts a sense of control on the member.

CONCLUSION

For the three approaches to the study of trust and the role of voluntary organizations, we plead for a research strategy that takes into account some of the missing concepts we identi ed above: frame resonance in social movement theory, intrinsic motivation from social psychology, and the embeddedness of trust relations in a wider civil society perspective. Speci cally (Table II, Figure I), for the relationship between economics and the social capital approach we suggest consideration of recent work on civil society (Cohen and Arato 1992; Keane 1998; Edwards et al. 2001). These studies offer an important avenue to better understand the relationship between trust substitutions and trust shifts on the one hand, and the concepts of power and legitimacy associated with accumulated social capital on the other. For sociology and the social capital approach, we point to the social movement literature (Snow et al. 1986; McAdam et al. 2001) to look at how organizations, networks, values and motivations interact with trust in various types of communities. Finally, for the relationship between economics and sociology in particular, we argue that recent work on motivation in social psychology can shed light on how language and subjective experiences can be important in understanding how trust is associated with, and nurtured by, voluntar y associations. (Date accepted: April 2002) Helmut Anheier and Jeremy Kendall Centre for Civil Society The London School of Economics and Political Science

NOTES

1. The EVS (see Halman 2001) covers next to Britain the following countries: France, Germany, Austria, Italy, Spain, the Netherlands, Belgium, Denmark, Sweden, Iceland, Northern Ireland, Ireland, Estonia, Latvia, Lithuania, Poland, Czech Republic, Slovakia, Hungar y, Romania, Bulgaria, Croatia, Greece, Russia, Malta, Luxembourg, Slovenia, Ukraine, Belarus, and the USA. The countries where the positive relationship between trust and memberships in voluntar y associations

either does not exist or is weak are: Romania, Russia, Ukraine and Belarus. 2. This includes memberships in health and social welfare associations, religious/ church organizations, education, arts, music or cultural associations, trade unions and professional associations, local community groups and social clubs, environmental and human rights groups, youth clubs, womens groups, political parties, peace groups, sports and recreational clubs, among others.

Interpersonal trust and voluntary associations

3. Measured by the following questions: Generally speaking, would you say that most people can be trusted or that you need to be very careful when dealing with people? (Inglehart et al. 1998). 4. World Values Survey 2000. USA Survey, conducted by Gallop for Virginia Hodgkinson, Helmut K. Anheier, and Ronald Ingehart. 5. See also Putnams analysis of trust in the USA (2000: 139). 6. See special issue of the journal Voluntas (Volume 8, Issue 2, 1997) for an overview of economic theorizing in the eld of non-pro t organizations. 7. Hence, classical thinkers such as de Tocqueville in political science or Durkheim in sociology, who had a great deal to say on the issue of trust and associations are only implicitly presented in as much as they have been ingredients of these recent formulations. 8. Arrows (1963) seminal paper had originally drawn attention to the role of trust in reducing uncertainty in healthcare, a commodity characterized by chronic informational problems. 9. This reasoning may also help explain the fragility of non-pro t organizations (see Powell and Friedkin, 1987: 8488). Pre-existing trust, once violated, is difficult to re-establish. Beyond certain thresholds, relatively modest trust violations may trigger far-reaching and unanticipated effects, as scandals in charities and religious organizations amply testify. 10. Seligman (1998) goes one step further and argues that traditional societies did not rely extensively on trust (explicitly pace Durkheim, 1933: 6), but rather on confidence. Economic and social relations were embedded in extensive systems of kinship that involved the possibility of recourse to extensive formal and informal sanctions. This system served to foster predictability, which in turn limited the need for recourse to trust. For Seligman, trust is relevant precisely in those situations characterized by the absence of predictability and the absence of familiarity. Following Beck (1992), he suggests that modernity is more unpredictable than previous eras because there is no underlying sense or terms of familiarity or sameness (p. 4).

359

11. As well as re ecting prevailing conditions, non-pro t organizations are also often created precisely as vehicles with which to oppose them, as the social movement literature shows. This is discussed in the next section. 12. The view of voluntary associations as consensual and con ict-free has been challenged by Edwards and Foley (1997), among others, by pointing to problems of exclusivity, particularism, and the promotion of values associated with intolerance and prejudice. 13. Putnams most recent study (2000) of the USA is somewhat broader than the earlier Italian analysis in its empirical reach, but the theoretical conceptualization of trust at the core of this study still privileges local face-to-face interaction, downplaying other spatial possibilities and communicative modes. 14. We do not advocate replacing the three approaches by a common perspective this would be unwise since each approach serves important and wider roles in their respective disciplines, seeking to answer the different stylized questions that each pose. 15. An example would be Ross and Samuels (1993) demonstration that groups responded ver y differently in an experimental setting in playing an otherwise identical game depending on whether or not it was labelled the Wall Street Game or the Community Game. Importantly, this effect was found to be far more important than the students personalities in determining whether or not they acted co-operatively. 16. It is also consistent with welfare mix arguments which contrast the instrumental logic of the market place with the norms and traditions of personal obligations which characterise the community and simultaneously emphasize how many voluntary organizations are in uenced by their rootedness in the latter as opposed to the former (Evers 1995). 17. Within this tradition of psychology, self-determination is a quality of human functioning that involves the experience of choice . . . [it is] the capacity to choose and to have those choices, rather than reinforcement contingencies, drives or any other forces or pressures, be the

360
determinants of ones action . . . [these] typically have the bene t of developing competences (Deci and Ryan 1985: 38). In making these choices freely and pursuing the concomitant goals, it would appear that participants achieve a sensation of selfexpression and self-fulfilment that is valued for its own sake. 18. However, this literature also recognizes that external intervention can have the opposite effect of nurturing intrinsic motivation, and hence individuals sense of self-esteem: this can be the case if the attempt to steer voluntar y activity is accompanied by positive feedback, and involving intervention which is perceived to be supportive of participants autonomy rather than attempting to control them.

Helmut Anheier and Jeremy Kendall

Bourdieu, P. 1979 La distinction: critique sociale du judgement, Les Editions de Minuit, Paris. Brennan, G. 1997 Selection and the Currency of Reward, in R. E. Goodin (ed.) The Theor y of Institutional Design, Cambridge University Press, Cambridge. Cohen, J. L. and Arato, A. 1992 Civil Society and Political Theor y, Harvard University Press, Cambridge, MA. Coleman, J. 1990 Foundations of Social Theor y, Harvard University Press, Cambridge, MA. 1993 A Rational Choice Perspective on Economic Sociology, in N. J. Smelser and R. Swedberg (eds) The Handbook of Economic Sociology, Princeton University Press, Princeton. Dasgupta, P. and Serageldin I. (eds) 1999 Social Capital. A Multifacted Perspective, The World Bank, Washington, DC. Deci, E. and Ryan R. 1985 Intrinsic Motivation and Self Determination in Human Behavior, Plenum Press, New York. Dekker, P. and Van den Broek A. 1998. Civil society in comparative perspective: involvement in voluntar y associations in North America and Western Europe, Voluntas 9(1): 1138. DiMaggio, P. and Mohr, P. 1985 Cultural capital, educational attainment, and marital selection, American Journal of Sociology 90 (6): 123161. Durkheim E. 1933 The Division of Labor in Society, Free Press, New York. Edwards, B. and Foley, M. 1997 Social Capital, Civil Society and Contemporary Democracy American Behavioral Scientist, Special Issue 40(5): 550678. Edwards, B., Foley M. and Diani, M. 2001 Beyond Tocqueville: Civil Society and the Social Capital Debate in Comparative Perspective, University Press of New England, Hanover, New Hampshire. European Values Survey 19992000 CD Prepared for data workshop on 19992000 wave of European value Survey, Zentralarchiv, Cologne. Evers, A. 1995 Part of the Welfare Mix: The third Sector as an Intermediate area between Market Economy, State and Community, Voluntas 6(2): 15982. Frey, B. 1997 A Constitution for Knaves crowds out Civic Virtues, The Economic Journal 107: 104353.

REFERENCES

Anheier, H. K. and Kendall, J. 2000 Trust and voluntary organizations: Three theoretical approaches, Civil Society Working Paper No 5, Centre for Civil Society, London School of Economics, London. Anheier, H. K. and Romo, F. P. 1992 The Philanthropic Transaction. Paper presented at the 1992 Convention of the American Sociological Association, Pittsburgh. Arrow, K. 1963 Uncertainty and the Welfare Economics of Medical Care, American Economic Review 53: 94173. Barbetta. P. 1997 The Non-pro t Sector in Italy, Manchester University Press, Manchester. Beck, U. 1992 Risk Society: Towards a New Modernity (Theory, Culture, and Society Series), Sage, Thousand Oaks. Ben-Ner, A. 1994 Who bene ts from the Non-profit Sector? Reforming Law and Public Policy Towards Non-pro t Organizations, Yale Law Journal 104(3): 73162. Ben-Ner, A. and Gui, B. (eds) 1993 The Non-pro t Sector in the Mixed Economy, The University of Michigan Press, Ann Arbor. Ben-Ner, A. and Van Hommissen, T. 1993 Non-profit Organizations in the Mixed Economy: A Demand and Supply Analysis, in A. Ben-Ner and B. Gui (eds) The Nonpro t Sector in the Mixed Economy, The University of Michigan Press, Ann Arbor.

Interpersonal trust and voluntary associations

Fukuyama, F. 1995 Trust, Free Press, New York. Galaskiewicz, J. and Bielefeld W., and (forthcoming) 2002 The behaviour of organizations, in H. K. Anheier and A. Ben-Ner (eds) The Study of Non-pro t Enterprise: Theories and Approaches, Plenum/Kluwer Academic Press, New York. Giddens, A. 1990 The Consequences of Modernity, Stanford University Press, Stanford. Goffman, E. 1974 Frame Analysis: An Essay on the Organization of Experience, Harvard University Press, Cambridge, MA. Habermas, J. 1985 Die neue Unuebersichtlichkeit, Suhrkamp, Frankfurt. Halman, L. 2001 The European Values Study: A Third Wave. Source book of the 1999/2000 European Values Study Surveys, EVS, WORC, Tilburg University, Tilburg. Halpern, D. 1999 Social Capital. The New Golden Goose?, Institute for Public Policy Research, London. Hansmann, H. 1980 The Role of NonPro t Enterprise, Yale Law Journal 89(5): 835901. Hansmann, H. 1981 Non-profit Enterprise in the Performing Arts, Bell Journal of Economics 12: 34161. Hansmann, H. 1996 The Ownership of Enterprise, The Belknap Press of Harvard University Press, London. Inglehart, R. 1999 Postmodernization, Authority and Democracy, in P. Norris (ed.) Critical Citizens: Global Support for Democratic Government, Oxford University Press, Oxford. Inglehart, R., Basaet, M. and Moreno, A. 1998. Human Values and Beliefs: A CrossCultural Source book. Political, Religious, Sexual and Economic Norms in 43 Societies. Findings from the 19901993 World Values Survey. University of Michigan Press, Ann Arbor, Michigan Judge, K. 1984 Caring for Pro t: Private Residential Homes for Elderly People: The Establishment of Family Enterprises, Discussion Paper 311, PSSRU, University of Kent at Canterbur y. Keane, J. 1998 Civil Society, Polity Press, Cambridge. Krashinsky, M. 1986 Transaction Costs and a Theor y of Non-profit Organizations, in S. Rose-Ackerman (ed.) The

361

Economics of Non-pro t Institutions, Oxford University Press, Oxford. Lane, C. 1998 Theories and Issues in the Study of Trust, in C. Lane and R. Bachmann, (eds) Trust within and between Organisations: Conceptual and Empirical Applications , Oxford University Press, Oxford. Luhmann, N. 1988 Familiarity, Confidence, Trust: Problems and Perscriptives, in D. Gambetta (ed.) Trust: Making and Breaking Cooperative Relations, Basil Blackwell, Oxford. McAdam, D., McCarthy, J. D. and Zald, M. N. 1988 Social Movements, in J. N. Smelser (ed.) The Handbook of Sociology, Sage Publications, Newbury Park, CA. McAdam, D., Tarrow S. and Tilly C. 2001 Dynamics of Contention, Cambridge University Press, Cambridge. McPherson, M., Smith-Lovin, L. and Cook J. M. 2001 Birds of a feather: Homophily in social networks, Annual Review of Sociology 27: 41544. Minkoff, D. 1997 Producing Social Capital: National Social Movements and Civil Society, American Behavioral Scientist 40(5): 60619. Neidhardt, F. and Rucht, D. 1993 Auf dem Weg in die Bewegungsgesellschaft? ber die Stabilisierbarkeit sozialer Bewegungen, Soziale Welt 44: 30526. Newton, K. 1997 Social Capital and Democracy, American Behavioral Scientist 40(5) 57586. Passy, F. and Giugni, M. and 2000 Lifespheres, networks, and sustained participation in social movements: a phenomenological approach to political commitment, Sociological Forum 15(1): 11744. Pharr, S. J. and Putnam, R. (eds) 2000 Disaffected Democracies: Whats Troubling the Trilateral Countries?, Princeton University Press, Princeton, NJ. Powell, W. W. and Friedkin, R. 1987 Organizational Change in Non-profit Organizations, in W. W. Powell (ed.) The Non-pro t Sector: A Research Handbook, Yale University Press, New Haven, CT. Putnam, R. 2000 Bowling Alone: The Collapse and Renewal of American Community, Simon and Schuster, New York. Putnam, R. with Leonardi, R. and Nanetti, R. 1993 Making Democracy Work: Civic

362
Traditions in Modern Italy, Princeton University Press, Princeton, NJ. Rahn, W. and Transue, J. 1998 Social and Trust and Value Change: The Decline of Social Capital among American Youth 19761995, Political Psychology 19: 54556. Rose-Ackerman, S. 1996 Altruism, Nonpro ts, and Economic Theory, Journal of Economic Literature 34(2): 70128. Ross, L. and Samuels, S. 1993 The Predictive Power of Personal Reputation versus Labels and Construals in the Prisoners dilemma Game, unpublished manuscript, Stanford University. Salamon, L. M. and Toepler, S. 2000 The in uence of the legal environment on the development of the non-profit sector, Center for Civil Society Studies Working Paper No. 17, Center for Civil Society Studies, Institute for Policy Studies, Johns Hopkins University, Baltimore. Seligman, A. 1998 Keynote speech at 4th Conference on Researching the Voluntar y Sector, Loughborough University, National Council of Voluntary Organizations. Sirianni, C. and Friedland, L. 2001 Civic Innovation in America: Community Empowerment, Public Policy, and the Movement for Civic

Helmut Anheier and Jeremy Kendall

Renewal, University of California Press, Berkeley, California. Snow, D. A., Rochford, E., Worden, S. and Benford, R. 1986 Frame Alignment Processes, Micromobilization and Movement Participation, American Sociological Review 51: 46481. Stevenson, W. B. and Greenberg, D. 2000 Agency and social networks: strategies of action in a social structure of position, opposition, and opportunity, Administrative Science Quarterly 45(4): 65178. Stinchcombe, A. 1965 Social Structure and Organization, in J. G. March, (ed.) Handbook of Organizations, Rand McNally, Chicago. Taylor-Gooby, P. 1997 Markets and Motives: Trust and Egoism in Welfare Markets, Journal of Social Policy 28(1): 97114. Tonkiss, F and Passey, A. 1999 Trust, Confidence and Voluntar y Organizations: Better Values and Institutions, Sociology 33(2): 25774. Zucker, L. 1986 Production of Trust: Institutional Sources of Economic Structure 19401920, in S. Bacherach, (ed.) Research in Organizational Behavior, Jai Press, Greenwich, CT.