Trust and Food. A Theoretical Discussion

Trust and Food. A theoretical discussion Armando Salvatore and Roberta Sassatelli University of Bologna, Italy Consumer Trust in Food A European Study of the Social and Institutional Conditions for the Production of Trust1 Trust and Food. A theoretical discussion Armando Salvatore and Roberta Sassatelli University of Bologna, Italy Working Paper 1 Prepared by partner 2 November 2004 1 ‘Consumer Trust in Food. A European Study of the Social and Institutional Conditions for the Production of Trust’ has been funded by the European Commission, 5th Framework Programme, Quality of Life, Key Action 1, Contract no. QLK1-CT-2001-00291. Preface This report is a publication from Consumer Trust in Food. A European Study of the Social and Institutional Conditions for the Production of Trust. The TRUSTINFOOD project (2002-2004) is supported by the European Commission, Quality of Life and management of Living Resources Programme (QoL), Key Action 1 Food, Nutrition and Health (contract no. QLK1CT-2001-00291). Unni Kjærnes at The National Institute for Consumer Research (SIFO) is responsible for coordinating the project. On the basis of individual and institutional data, the study seeks to identify and analyse factors that determine trust in the food supply and in information sources. These factors include the roles of public authorities, consumer organisations, market actors, consumers, NGOs, etc. Representative surveys have been conducted in six countries, Denmark, Germany, Great Britain, Italy, Norway, and Portugal. Institutional studies have been carried out in the same countries and at the European level. By eventually bringing all these data together, we expect to achieve a systematic analysis of the institutional bases of consumer trust and distrust in food provision under varying conditions in contemporary Europe, including a critical analysis of alternative strategies for handling trust and distrust in the food system. More information and new publications are available on the project website: www.trustinfood.org. This paper draws on material which was initially presented in two research meetings held in Bologna and Copenhagen, in April and June 2002 respectively, as part of the project “Trust in food. A European study of the social and institutional conditions for the production of trust” funded by the European Commission Research Directorate-General "Quality of Life and Management of Living Resources", Key Action 1 on "Food, Nutrition and Health" (EUcontract number: QLK1-CT-2001-00291). All researchers, scientists in charge and advisors involved in the project are thanked for their comments. Special thanks are owed to Jasper Chalcraft for his useful suggestions and careful editing. Bologna, November, 2004 University of Bologna 2 Table of Contents PREMISE .................................................................................................................................................................4 INTRODUCTION ...................................................................................................................................................5 I. TRUST AND THE NODE OF “VALUE”: CLASSICAL APPROACHES IN SOCIAL AND POLITICAL THEORY: TRUST AS A PHILOSOPHICAL AND SOCIOLOGICAL PROBLEM .......................................8 II. TRUST AND CHOICE: RATIONAL CHOICE THEORY AND ITS COGNATES .................................14 III. TRUST AND DISTRUST: LUHMANN’S SYSTEMIC VIEW, AND THE BIFURCATION BETWEEN TRUST AND CONFIDENCE...............................................................................................................................18 IV. TRUST, RISK CULTURE, AND REFLEXIVITY ......................................................................................21 V. TRUST AS PRACTICE, HABIT AND ROUTINE........................................................................................23 VI. THE DISTINCTIVENESS OF TRUST VS. CONFIDENCE AND FAMILIARITY................................26 VII. THE INTERPLAY OF TRUST, CONFIDENCE AND FAMILIARITY WITHIN FOOD CONSUMPTION ARENAS: CONSUMERS’ CHOICES, MARKETS’ RESPONSES, AND THE RECONFIGURATION OF REGIMES OF TRUST ..........................................................................................30 CONCLUSION: A FEW ANALYTICAL NOTES FOR FURTHER DISCUSSION ......................................35 BIBLIOGRAPHY..................................................................................................................................................37 3 Premise The goal of this paper is to help develop a comprehensive theoretical basis for the empirical study of ‘trust’ and ‘distrust’ in food, by relating concepts such as routines, reflexivity and risk perception, which are crucial in the literature about food consumption. It is part of a larger strategy which includes also a wider discussion of the project, its methodology and its scope. Taken together these facilitate the development of a conceptual model that encompasses the various dimensions and relations of trust to be deployed in our project. This format itself reflects a view which has often been discussed among the partners in the project: both trust and distrust could be seen as constructions – historically grounded, concerted, and contested – which might facilitate better social worlds. To be sure, the project takes off from the recognition that trust has gained prominence in research concerned with risk management, especially in terms of how to cope with what is called a “crisis of confidence” (Stirling 1998). However it aims to consider both trust and distrust critically – that is in relation to their social determinants as grounded in specific social and institutional histories as well as in their positive and negative implications. We aim to provide an understanding of the conceptual dimension of trust in food, its relevance and its cross-cultural variation while offering some means to evaluate the possible ways forward for ameliorating European food systems and policies. In this sense, the question arises as to whether we should have a certain level of distrust in society or, in other terms: does (a measure of) distrust lead to better, safer and more democratic societies? In Europe the interest in the democratic management of risk has meant that food crises and trust deficit are no longer seen as social anomalies. Food issues, together with the application of technology to life and biology , are the dominant themes of contemporary public debates. According to one study, they represent more than 45% of the public debates taking place in the Western world (Marris and Joly 1999). This partly reflects the need to govern and control risk perception and trust mechanisms, but it is also arguably linked to the growing need to take seriously the different perspectives of the different actors involved. 4 Introduction Trust is a puzzling paradigm for social and political theory. It is puzzling because it challenges and yet thrives on individualistic views of the subject: half rational, half emotional, always singular and yet defined by its being amidst other individuals. It is puzzling because it draws attention to different aspects of the social bond, procedural and substantive, which may be conflicting. Finally, it puzzles also because it is experienced, conceived and modelled in different, sometimes irreconcilable, ways and yet it often derives its deeper value from a somewhat universalistic flair. Trust is, nevertheless, more and more thematised both within and without academic circles, often invoked as a default option to explain order or – when it is considered to be lacking – dis-order, both at the social and personal level. While the simple equation between trust and order (or its contrary) is dubious, it is quite probable that the current importance assigned to trust is the result of the failure of routine mechanisms of confidence. Yet, we should be wary of considering such failure as a one-dimensional, universal, seamless feature of “modernity” – as the case of the different responses to food scares across Europe well illustrate. The erosion of food confidence in contemporary Western countries has been widely debated in the literature on food (Fischler 1988; Gronow 1997; Maurer and Sobal 1995; Sellberg 1991; Szerszynski 1999; Wildalwski and Dake 1990). Risk theories address the specific characteristics of modern societies (cf. Beck 1992 [1986] and Giddens 1990 and 1991), or consider risk as an invariable feature of the ambivalence inherent in human food consumption (Fischler 1988), which might be aggravated by modern conditions of complexity, globalisation and massification (Beardsworth 1990 and 1995). All in all, they note the difference between consumers’ and experts’ perceptions. In the wider public the fears associated with food shortage have been replaced by preoccupations with food adulteration and food poisoning (Beardsworth and Keil 1997). In scientists' rankings of food hazards, microbial contamination occupies the top and pesticide residues and deliberate food additives the bottom. However, opinion surveys show that lay people are particularly uneasy about pesticide residues and preservatives (Hobam 1995; more widely on risk perception, trust and the environment see Douglas 1992; Szerszynki 1999 and Wildawsky and Dake 1990). While there might be a tendency to consider consumers’ worries as ‘unreasonable’, these do indeed appear to be grounded in concerns different from those of the experts, less specialist and more political. These concerns are probably better understood if we move from the vocabulary of risk to that of trust and distrust (which leads more easily into historical and sociological contextualisation). Present knowledge indicates that levels of trust vary considerably across Europe. Some of the differences can be explained by the absence or presence of food scandals, but the seriousness of such scandals does not seem to be inversely correlated with consumers’ level of trust. Not only do the levels of trust and distrust vary, but so also do the institutional arrangements, actors or symbols in which consumers place their trust, as well as responses to problematic events (Löfstedt and Frewer 1998). And while trust and distrust might appear to be obviously related to scares, how precisely they are linked is not quite clear. This paper starts from the idea that we need to problematise the notions of trust and distrust. We shall therefore ask: which kind of social relation is indicated with the idea of trust? Is it sufficient to speak of trust and distrust, or do we need parallel notions pointing out contiguous, but not identical social relationships that also impinge on relations of trust? Can we provide a single definition for all people, and all countries? How do historical and institutional features contribute to the definition of trust and to its practical implementation? As we shall see, the sociological classics (from Weber to Simmel) as well as the fathers of liberal thought (including Hobbes and Hume) have implicitly adopted a comparative and institutional 5 approach, suggesting that the historical, cultural and institutional components are important for accounting for differences between countries. They help us consider that the degree of trust in food may be commanded by confidence in public institutions at large. However, this is typically done with a strong emphasis on the agent, on his/her capacity to re-interpret reality and, in general, act meaningfully. Unlike rational choice theory – which is certainly a prominent stream of thought in the literature on trust – the sociological classics rarely define the model of the agent/actor and the rationale of action once and for all. Trust for the neo-classical mainstream economist is a lubricant of exchange, itself portrayed as a non-easily substitutable commodity which may not be produced very easily. Even for rational choice sociologists like James Coleman, trust should be related to purposive behaviour aiming at maximisation of utility under risk, with mutual trust emerging as a form of social capital which cuts down the cost of monitoring activities. Yet, as Hirschman (1984) suggested, if trust is a commodity, it is one which increases with use. As we shall see, trust and consumer choices are intertwined in complex ways which cannot be placed on a neat means-end sequence, nor comprehended without reference to the context where consumption takes place (see Sassatelli 2004a for a wider discussion). Furthermore, in our view, what is an actor, how action is defined and carried out is an object of analysis which varies together with trust and distrust. This entails that there is always a process of identity building in practical actions and that mechanisms of trust are partly related to this process. Incidentally the process of identity building is also strongly associated with the different actions and rhetorics that a number of relevant actors in the food system - and bearing on it ranging from cultural intermediaries, to marketing and advertising, to consumer protection organizations and new social moments, an so on- construct and promote to engage people qua food consumers (cfr. Sassatelli 1995; 2002a; 2004b). Rational choice theory has the merit of providing a clear-cut scheme for devising social policy, with neatly defined dimensions of trust that can be relatively easily operationalised and deployed in a comparative fashion. While our project starts from the idea that trust is a thoroughly historical and cultural phenomenon, we also want to compare different countries in their responses to food consumption and trust issues. Such comparison is made through attempting to organise our understanding of the multidimensional nature of trust (see conclusions). However, we do not want to lose sight of the differences and peculiarities of each institutional setting. To this end we shall bring back to institutional settings most conventional variables in risk analysis – i.e. who is trusted. We know that typically, while industry and government often lack public trust, consumer organisations, quality media and medical expertise are highly trusted, yet these are to be taken as dynamics which relate to particular cultural and social histories. A few elements from both the Luhmannian approach to trust and distrust and the Giddens-Beck thesis are useful here: both views place emphasis on the cognitive dimension (beliefs, definitions, frames), both emphasise systemic effects (outcomes independent from the aims of the actor) while portraying the actor as active or in the case of the Giddens-Beck thesis, reflexive. Reflexive individuals convinced of their personal responsibilities for looking after themselves (or consumers mobilising collectively) strengthen the importance of the institutional procedures in which people place confidence. (Again this feature is not specific to the case of food; it is rather an expression of a more general shift in the public’s attitudes towards authorities). At the same time, however, procedures also become more problematic (too abstract, too complex, too specialised to be understood and legitimised in an efficient way). Indeed, routines still play a large role in food consumption. This is why a different notion of “reflexivity” is also required. Trust within ethnomethodological approaches is not a strategy, it is a precondition of action: the orderliness of social life is grounded on the mutual trust people display towards one another, with the trustworthy person being the one who masters the discrepancy of prescribed attitudes with respect to reality in such a fashion as to maintain a public show of respect for them (Garfinkel 1963). 6 Garfinkel shows how taken-for granted assumptions which constitute everyday life attitudes produce actions that confirm the individual's expectations and how actors elaborate and stretch the existing rules in order to cover new events. Actors are not reflexive, yet their actions and discourses are, with participants in interaction attributing to one another a reflexive awareness of the normative accountability of their actions. In this sense, while normal social reality is a contingent ongoing accomplishment of competent social actors, routines are of essential importance emerging as they do out of the perceived normality of interpersonal events that members of a group seek to maintain through their adaptive activities and adjustments. If we take this approach seriously and combine it with Bourdieu’s and Douglas’ classic anthropological views (which consider that a consumer choice is not a choice about a good as such, but a choice about social position, identification and dis-identification, though this is to some extent misrecognised), we may appreciate how trust in food is related to the social structure. The value of a meal or a drink is deeply entangled with the many ways in which this very meal is a sign and a practice of trust and distrust, of how people position themselves in the world, associating with some and dissociating themselves from others. Still, this view does not allow for an analytical consideration of the multidimensionality of trust, of the different agents who might be involved in trust formations, and of the mediating role of institutions. If trust becomes an equivalent to the notion of social order, then where is trust? “What modernity has involved has been precisely the replacement of the criteria of familiarity with those of trust in and of the individual as an aspect of social unconditionality” (Seligman 1997: 91). The crucial tension throughout will be one between the (contractual) freedom of the social agent, the institutionalisation of rules for the market that place restrictions on such freedom, and the growing uncertainty surrounding the choices of consumers within food markets, an uncertainty that the associational pooling of the consumers as citizens is not able to govern, so that we are thrown back to examining the distinctive configurations of individuals' choices and institutional regulations. The root question to be developed from here will be: which institutional configurations of social arenas are apt to further (or discourage) trust among individuals, groups, and institutions? To provide analytical clarifications we will discuss the relation of trust, familiarity and confidence on the basis of Adam Seligman’s work. This helps to distinguish between a taken-for-granted (or factual, unreflective, operational) dimension of trust (i.e. confidence), and a more active, voluntary one (i.e. trust proper). The latter appears to gain prominence when confidence mechanisms do not apply, and puzzlement spreads out (mistrust). This view allows us to better define how distrust (the other possible answer to puzzlement) might be conducive to trust. The background of systems of confidence is provided by some kind of familiarity. By familiarity Seligman means a form of trust which is based on sameness and identification with a norm, an authority, a social arrangement, or a product (i.e. probably close to what others call tacit trust, that is related more to what we know from the past), while trust proper appears to entail difference among actors, and the need to bet on the future, the combination of which grounds negotiation. The process of re-building trust on the basis of familiarity, when confidence fails, can be further illuminated by introducing the notions of embedded vs. disembedded trust, which might be useful as a heuristic tool (rather than a descriptive or normative dichotomy) to help identify different configurations or regimes of trust in a comparative perspective (Sassatelli and Scott 2001). 7 I. Trust and the node of “value”: classical approaches in social and political theory: trust as a philosophical and sociological problem Trust has occupied several generations of political and moral philosophers and, since the 19th century, social scientists and in particular sociologists (see in particular Misztal 1996 and Seligman 1997 for the history of dealing with trust). The issue is intimately linked to the problem of explaining social order, or the “social bond”, at the transition to modern societies, that requires trust to reduce complexity and contingency, a function played in pre-modern societies by traditional arrangements and authorities (Sztompka 1999: 45), and more precisely by the role segmentation entailed by traditional social structure (Seligman 1997: 13-30). The issue of trust takes form gradually in a semantic field of reflection about structural transformations in industrial (and post-industrial) societies, within a process that gradually sets it apart from contiguous notions like faith, confidence, familiarity, loyalty, and solidarity, as well as role expectations, and professional reputation (cf. Seligman 1997: 44-74). For the scope of the project, the conceptual bifurcation between “trust” proper and “confidence” will be particularly crucial, but “familiarity” will also play a role. Among the classic theorists of trust, most prominent are Thomas Hobbes, John Locke, Adam Smith, and Alexis de Tocqueville. A caveat is due for the following brief review of their thought. Especially in the case of Hobbes and Locke, their notions of trust were still placed in a field of tension with the concept of faith (as faith in God), the latter being the less interesting among the family of notions at stake here, although the process through which trust differentiated semantically from faith was crucial to its genesis and to its classic theorists. For Machiavelli, the notion of trust was still enveloped in fides. “In contrast to the Latin meaning of fides, which implies the certainty of the remuneration, a certainty based most often on ascriptively defined loyalties, the unconditionality of trust is first and foremost an unconditionality in respect to alter’s response” (Seligman 1997: 44). The basic terrain is given by the changing structure of ego-alter relationships and the role which emerges for a third instance/actor, an “impartial observer” or an “invisible hand”. Hobbes justified Leviathan as the necessary artificial warrant of trust that prevents society from being a jungle, guaranteeing security and simplifying public life by eliminating issues of contention that threatened to disrupt the social bond (1750 [1640]). Hume provided the bottom-line of all subsequent versions of the rational-choice approach in his ‘philosophical egoism’ that demands that action be guided by how the actor judges its consequences: everything which counts can be expressed in consequential terms and each actor remains a bargain-hunter (Hollis 1998 and Sassatelli 2001b). Locke – whose view of trust has been at the root of several generations of liberal political philosophers – emphasised the importance of “promise” through which the social agent imposes obligations on himself, so creating the basis of trust within civil society (Dunn 1984). Smith added sympathy among human beings in aiding the mechanism through which out of their “natural” interests and passions a patterned order emerges that furthers common interest (the somewhat providential “invisible hand”: Smith 1759). Tocqueville went one step further in clearly stating that contract alone does not suffice, but that secondary associations are necessary to inculcate in social agents the discipline of self-imposed obligations (Tocqueville 1945). While Tocqueville’s theorising is the historical and theoretical antecedent to contemporary republican communitarians, in the Anglo-Saxon traditions a powerful utilitarian approach (in fact several variants thereof) crystallised, that ended up by somewhat taking for granted the social bond as residing in trust, since utilitarians emphasised that the prudential reasonableness of rational egoists warrants social equilibrium and common interest. In this sense, trust is not necessary. A particular string 8 within utilitarianism was Edward Spencer’s evolutionist theory, that stressed that co-operation is a product of an evolutionary learning process based on the principle of maximisation of collective utility. According to him, it is industrial society that furthers a heightened sense of co-operation through a selfimposed restraint, that makes coercion unnecessary and unwanted. The first theorist to operate an energetic shift from political and moral philosophy to a sociological approach was Ferdinand Tönnies, with his seminal distinction between Gemeinschaft and Gesellschaft. Elements of his characterisation of both are particularly interesting for our project. In the pre-modern Gesellschaft, co-operation and association is based on habit (that is in a first instance a sociologically much better focused concept than unqualified “tradition”), whereas in Gesellschaft the means-end rationality prevails on the basis of a quest for profit and power, facilitated by an ongoing manipulation of other peoples’ ends. The default form of trust is per se embedded in social relations based on habits reflecting a common will based on familiarity and friendship among social agents, therefore trust so defined is characteristic of Gemeinschaft, whereas Gesellschaft is bound to recreate trust artificially on the basis of instrumental rationality, which is unable to guarantee trust in a stable and systematic way. Manipulation and instrumentality only allow for surrogates of trust. The main form of this surrogate is (professional) reputation, a sort of impersonal (and, as we would say, disembedded) trust that results from the professionalisation of co-operative relationships. Experts are trusted because of their professional skills. Co-operation can thereby only be based on self-interest mediated by professionalisation. While so far Tönnies’ theory shares with the utilitarians the view that in modern societies co-operation can only be based on self-interested action, and thereby surrogate trust, he differs from them in stressing that the standards of reputation via professionalisation which uphold “impersonal trust” are intrinsically fragile, since not based on structures of mutual dependence in everyday life (Tönnies 2001 [1887]). We retain from his theory the double view of trust as either embedded, and based on habit, or disembedded, and based on expert’s reputation. To give a clue of Tönnies theory of Gesellschaft, let’s use a quote about money, that we can then contrast with Simmel’s and Luhmann’s views: “Society [Gesellschaft] produces an abstract version in the form of paper money, which it circulates by giving it a rate of exchange. This means that value is inherently the product of Society’s will. For Society is simply the embodiment of abstract reason – if we may imagine reason as being active and exercising will – and every reasonable being participates in it through his own thought” (p. 58). Emile Durkheim’s social theory blends Tönnies lucid diagnosis of the price to be paid through the rationalisation and abstraction of social relations with a proto-communitarian concern for reconstructing the social bond, or even social order, in modern societies, on the basis of a moral principle, which he tries however (unlike the political philosophers) to spell out in genuinely sociological – albeit also normativelyladen – terms. He therefore downgrades what in Tönnies was the force of habit based on mutual dependence to “mechanic solidarity”, and upgrades the system of rights and obligations linking social agents in moral society to an emerging principle of order – at the same time social and moral – reflecting an “organic solidarity”. The increasingly functional division of labour in modern societies is what inculcates in social agents a sense of the organic character of the social bond, through awakening an awareness that members of the social body depend on each other in an organic way. This patterned interdependence goes beyond the structures of mutual dependence in the much less differentiated Gemeinschaft of Tönnies. 9 This new form of interdependence and solidarity operates a passage from what we can term an enlightened self-interest to a genuine sociality facilitating the bracketing out of one’s self-interest for the sake of the collective interest, for whose representation the state is crucial, in its moral, more than coercive force, of representing the social body and the collective interest. Secondary associations are in this sense a necessary complement of state power, but not the ultimate source and warrant of organic solidarity (Durkheim 1984 [1893]). In Durkheim’s design, the state’s ritualised representation of collective interest is crucially complemented at a more grass-roots level by mass education and professional ethics. The notion of organic solidarity goes therefore beyond any prior classic or utilitarian definition of trust as tied to contract. Durkheim’s view radically opposed the view that the constitution of society could be explained on the basis of a social contract. His rebuttal of contractualism is more radical than Toqueville’s. “[I]t is not only outside of the sphere of contractual relationships, but also on the interplay between these relationships themselves that social action is to be felt. For in a contract not everything is contractual. The only undertakings worthy of the name are those that are desired by individuals, whose sole origin is this free act of the will ... Wherever a contract exists, it is submitted to a regulatory force that is imposed by society and not by individuals: it is a force that becomes ever more weighty and complex” (Durkheim 1984 [1893]: 158). It has been remarked that what makes Durkheim a leading sociological theorist of trust, the one who provided the truly sociological vocabulary and argument to conceive of trust as crucial to modern societies, is his argument that “in a well-integrated order, one can trust others to be honest and truthtelling, and to respect their promises, because of the shared morality and conformity to norms” (Misztal 1996: 46). However, by theorising about the sublimation of the personality of each member of society into a selfless commitment to an organic collective good, and a corresponding taming of narrow-minded self-interest, in terms of mutuality the root of trust is suppressed at the outset. Trust is, again, not really needed. However, Durkheim’s theory prefigures a functional view of trust as a social device for complexity-reduction, almost an intelligent ‘trick’ of the social system, in that the moral rules of organic solidarity make the social environment more predictable through inculcating in individuals a self-restraint that aids the crystallisation of habits. The nation and its identity – along with its core symbols and rituals - is determinant in this mechanism, and in this Durkheimian perspective there is ample room for redefining civil society, as the US social theorists Robert Bellah (1991) and Jeffrey Alexander (1998) have done, two authors who have tried – along the Durkheimian path – to escape the dilemma between liberalism and communitarianism through grounding the autonomy of the agent on the social subjectivity entailed by organic civic ties within society. Georg Simmel operated a clear shift back from solidarity to exchange, as the basic form of modern social relationships, "a sacrifice in return for a gain". In a move that could still be contained within an utilitarian logic, Simmel added that it is exactly the sacrifice that inculcates a sense of reciprocity, which is required for social relationships. But what makes him go beyond the utilitarian logic, is a view of exchange as per se creating the social bond, via the incorporation of a notion of generalised trust. While Simmel starts with a definition of money that is similar to Tönnies’ (“Money belongs to t[he] category of reified social functions. The function of exchange, as a direct interaction between individuals, becomes crystallised in the form of money as an independent structure”: 1978 [1900/1907]: 175), he goes on saying that “[w]ithout the general trust that people have in each other, society itself would disintegrate, for very few relationships are based entirely on what is known with certainty about another person, and very few relationships would endure if trust were not as strong as, or stronger than, rational proof or personal observation. In the same way, money transactions would collapse without trust” (Simmel 1978 [1900/1907]: 178-9). 10 Money itself depends on specific institutional configurations: the abstraction from infinitesimal exchanges into the institutional objectivity of money entails that exchange has become something other than a private process between two individuals (Sassatelli 2001b; 2004a). Yet, if trust becomes a glue (a synthetic force) within society, it is one that is not really prior to even the forms of exchange mediated by money as the most abstract and reified medium of exchange, but an essential component of it. What makes trust necessary is that a calculated assessment of probabilities about the behaviour of the social agents with whom one interacts is hardly possible. This calculating component of the assessment has to be supplemented by some sort of familiarity with the object of trust that is based on partial knowledge, on “something that is both less and more than knowledge”, “a weak form of inductive knowledge” that contains a “further element of social-psychological quasi-religious faith” (Simmel 1978 [1900/1907]: 179). Here, as in Durkheim, the genuinely sociological element of structural uncertainty surfaces and imposes itself as the main determinant of a notion of “generalised trust” that is again by Simmel best revealed by the possession of money: “The sense of personal security that the possession of money gives is perhaps the most concentrated and pointed form and manifestation of confidence in the socio-political organization of order.” (ibid.). The shift from “trust” to “confidence” when trying to indicate the most abstract, system-oriented and generalised form of trust here prefigures later developments spelled out by Luhmann. Whereas in Durkheim the other is always the general other, in Simmel, even if trust operates in a generalised form, the other is still basically the other of a specific social relationship. Though generalised, Simmel’s trust encompasses simultaneously an interpersonal and a systemic component. Money best exemplifies the generalised characteristic of trust, even if it is necessarily embedded in interpersonal relationships. As with money, mechanisms of trust make the mutuality of dependence between individuals impersonal. In Simmel, unlike in Tönnies, the impersonality of the general medium is reinterpreted positively, thanks to the fact that it brings freedom from personal dependence. However, the interpersonal character of relationships is preserved, and even upgraded through valuing the inner personality of the agent that operates a choice. On the one hand, social relationships are supported by the upgraded personality of the agent, on the other, they are diluted by the impersonality of the medium. Trust reveals the same ambivalence. Simmel’s view of the ambivalence of social relationships and trust within them can also be turned into an ambiguity and weakness of his own theory, based on the softness of how he characterises choice, as resulting by default from his characterisations of “sociation” (Vegesellschaftung), i.e. without invoking a singular theory of social action. Against this weakness, Max Weber stressed that action has to be characterised again by defining its purposive dimension, to be understood in terms of meaning as tied to a specific context of value. Value is context bound, yet it is part and parcel of the logic of social action, and cannot be derived from general forms and media of social cohesion within modern society. This even applies to the process of erosion of value-rational actions. This erosion could only originate from a particular interpretation and performance of the calling by puritan entrepreneurs, and that allowed for the transition from faith to trust as essential to modern capitalism. For instance, even in the mutual adjustment of interests facilitated by trust that characterise social and economic relationships within Gesellschaft, interactions are still formed on the basis of a (however diluted) purposive sense of value. As against the sharp dichotomisations found in Tönnies, Durkheim and Simmel, Weber acknowledged that the communal and the associative models of social relationships necessarily concur in shaping social action and interactions in modern societies. In other words, the dimension of “value” and “value rationality” is never erased from “choice”, and this, as we will see, can be applied to how we view the relationship between the individual consumer, the voluntary associations, and the market and state institutions intervening within food consumption arenas. Weber’s assessment is an aid to complexify the way we conceptualise this arena. 11 Thus, it is mainly through thematising the necessary legitimacy – in terms of value – of social arrangements that Weber raised the problem of trust: trust mainly resides in the validity of rules enshrined in the legal-rational basis of associative relationships, thereby taking the form of general trust. However, the processes of rationalisation of the economy and the bureaucracy show a tendency towards enacting legitimacy by renouncing – or at least eroding – trust. There is no final word from Weber in evaluating this trend, that exposes the danger that the manager, the expert and the bureaucrat might at any moment be seen as divested of authority proper, and therefore not worth trusting. Weber’s main legacy is that, while Simmel maintained – against Durkheim – that general trust must also be embedded in social relationships, he stressed that general trust alone provides a form that is necessary to social action, but not sufficient (especially in the long run) if not supported by a communal form of trust that is still basically that of Tönnies’ Gemeinschaft. As we will see in the last sections of this paper, Weberian sociologists are very sensitive to the risk of a sudden collapse of generalised trust, in those social contexts where this form of trust happens to absorb all others. In contrast to Durkheim, whose sociology of organic solidarity was inevitably normative by conception, Weber comes close to a sociological view that shows us the extent to which a normativity of values necessarily enters the realm of even self-interested, rationally instrumental social relationships. Instead of a normative consensus assumed by default while theorising about modern society, we face here a pluralism, or rather a clash of values and even of forms of social action: all of them have to respond to the generalised forms of legitimacy that are provided by capitalism, bureaucracy, and – we may add – representative democracy (which also includes lobbying and pressure groups). Even the way in which different forms of social action respond to these general constraints may vary greatly according to the values incorporated in the various forms of social action, and still be legitimate. Common to all “classic” theorising about trust, is a tendency to incorporate a growing institutional property, the necessity to capture impersonal and largely abstract systems of allegiance and co-operation as opposed to the personal, daily and often face-to-face quality of social ties in pre-modern communal settings. What differs in different theories is the assessment of this institutional property. A form of trust that is third to interpersonal and generalised-systemic trust, and that we can term “personal” trust, is rather identified as tied to pre-modern forms of social organisations based on charisma and local power. This third form, basically neglected by post-Weberian social theory, might resurface – as taught by Weber – as intrinsic to interpersonal trust and mingled with its associative form. Nowhere is there a “pure” civil society that can dispense of “authority”. Or is there one? Mainly in its systemic, but also in its interpersonal form (see below more on this), the emerging form of trust in modernity has become a key-indicator of social capital, of “civic and political culture” and “civil society” (Bellah 1991, Seligman 1992 and 1997) as the locus where mutual obligations are formed, cultivated and under normal circumstances kept. Trust has been the object of attention of the literature concerned with social capital, as the main ingredients of a dense associational life furthering co-operation, and where trust is both the condition and the product of a virtuous circle revolving around the accumulation of social capital (from Almond and Verba 1980 to Putnam 1993 and Alexander 1998). Trust in this sense is a systemic quality of associational life that is nurtured by social relations, and which feeds back on the whole system positively. In other words, civic life itself is a social fabric of trust. Such a notion brings together Tocquevillian and Durkheimian suggestions. This strand also crosscuts communitarian theorising, oriented towards re-evaluating trust as a collective resource furthering co-operation and mutual obligations (cf. Walzer 1983, Kymlicka 1989). This strand of literature might be used to analyse emerging circuits of food production and consumption based on co-operative enterprises and creating new ties of trust from below, but through using the legitimacy of official certifications/labels of the origins of their product. 12 Weber’s view of overlapping or conflicting forms of social action that can eat up very quickly the default measure of trust that is incorporated in the institutions and procedures of the economy and of the political system (and which some theorists prefer to call “confidence”, as distinct from trust proper: see below) leaves an acid after-taste to those who search for a view of trust writ large, pre-packaged for suitable forms of institutionalisation. Modern trust conceived after Weber’s lesson is, from the outset, volatile. Before guaranteeing formalised protocols of co-operation and regulation, trust has to enter the most routine-like aspects of social interactions and transactions in everyday life. The possibility itself of grassroots communication depends on a minimal level of trust, as evinced in Habermas’ theory of communicative action (Habermas 1984 [1980]). On the back of all “classical” approaches to trust by political philosophers and social theorists, reiterated and developed by some contemporary authors, we might observe the resilience of key anthropological notions; like those provided by Mauss’ theory of the gift that tries to spell out the roots of normative notions of social exchange (Mauss 1970) and to show why sheer calculation cannot account for gift-giving, and how a basic element of trust is built into the exchange through the expectation that the gift will be reciprocated in the long run. This perspective helps us to see how viewing trust as the root of cooperation is an after-the-fact kind of theorising, somewhat internal to an already-defined, packaged notion of modern (civil) society and its social agents, whose rationality is basically acknowledged only after they are educated into an ethic of co-operation (see also Lewis and Weigert 1985 for this critique). While Mauss has been mainly associated with Durkheim, it rather immunises us against viewing any form of trust and co-operation as organically tied to “higher” (more modern or rational) forms of social life. The main lesson drawn from the classics is that, at the very least, the notion of trust should be differentiated into different types of trust corresponding to various types of social relations, which are in turn tied to different kinds of obligations. In other words, at issue is not so much the necessity of purifying the notion of trust from any intrinsic normativity, but rather to see how different types of social relationships and attendant forms of trust produce different kinds of normativity and rationality, whereby a certain type of normativity (or what Foucault called governmentality) is indispensable to a corresponding type of rationality. In the contemporary reception of classics trust appears (especially when linked to social capital) as the conceptual banner of a transitional stage of societies, and of world society after the fall of the Soviet block re-launched the power of co-operation against what was increasingly perceived as a useless and even dangerous rhetoric of conflict. So, we should be aware of a limitation intrinsic to a concept that regained popularity in an epoch of transition, of a rise in neo-liberal recipes and weak neo-social democratic/new labour solutions, an epoch that appears to now be at an end because those recipes of solution no longer work. In this agony of a conceptual mode, we start to see a resurfacing readiness of several political and scholarly tendencies to accept again that conflict is a notion basic to societies, and even more to world society. As far as the EU is concerned, we don’t know yet how its underlying theorem of governance and participation – of its search for balances between integration and the preservation of regional identities, between standardised/consensual and local/fragmented forms of trust – is going to be affected by this reorientation. What is certain is that in a EU perspective, as in the present project, the conceptual demands on the notion of trust are clearly tilted towards evidencing its quality and potential as a “public good” to be consciously engineered, more than as a simply self-regulatory device of the social system. 13 Main lines of argumentation • The “classics” do not obviously make a “school”. However, the main commonality is that while they were intent on theorising (modern) society and what holds it together, they were also directly or indirectly concerned with defining social agency and the social agent; in this sense, from Hume to Weber, a concern for the agent is prominent. • The agent is defined in its simultaneous relation to modern institutions of the market, capitalism, the state and the world of voluntary associations. • The emergence – or the demise – of trust within modern societies is thereby defined as part of the emergence of a new world of institutions mediating social relationships. Trust – or the absence thereof – is discussed as an intrinsic part of these new forms of abstract mediations. II. Trust and choice: rational choice theory and its cognates The starting platform of a rational choice approach is to be found in a convenient reduction of the social agent’s motivation to a behaviour as a rational egoist prudently calculating the outcomes of his choice in terms of utility. From Hume through Smith to a vast array of utilitarians, the genealogy of this rational actor goes deep into classic theories of trust, and restates its concern for how public goods are produced while starting from the actions and choices of selfishly calculating social agents. The complication and sophistication introduced by German social science through Tönnies, Simmel and Weber are largely bracketed out. Means-end rationality is the sole vector, and goods of efficiency the exclusive goal of social action. Moreover, the theory of the illustrious predecessors of Anglo-Saxon (as well as Scottish) political philosophy is purified from any residual reference to natural law, and a full investment into economists’ definition of the social agent and his motivation builds the starting platform, integrated by a consideration of what trust contributes to the agent’s calculations and decisions. The underlying view of trust is one of a necessary lubricant of co-operation (Arrow 1972), especially, but not only, among economic actors, and therefore as a commodity sui generis (since it can neither be priced and purchased, neither so easily substituted: Dasgupta 1988) and, finally, as a public good (especially Gambetta 1988a, but also, before him, Arrow 1974 and Hirsch 1977). It is apparent that even within the rational choice approach that streamlines the cluster of motivations that ground trust and distrust, we meet a range of arguments and conceptualisations which oscillate between a focus on interpersonal trust and an emphasis on systemic trust. This is particularly evident in the essay of Gambetta that concludes the famous collection edited by him and based on an interdisciplinary seminar held at King’s College Cambridge in 1985-86, an essay that also wrestles with the latest theory of trust by Luhmann – which we deal with separately in the next section – and with analyses of trust that are historically and culturally situated (as in Southern Italy and in Muslim societies). Building both on Dasgupta’s typical rational choice approach, and on Luhmann’s, Gambetta defines trust in the following way: “trust (or symmetrically, distrust) is a particular level of the subjective probability with which an agent assesses that another agent or group of agents will perform a particular action, both before he can monitor such action (or independently of his capacity ever to be able to monitor it) and in a context in which it affects his own action (Gambetta 1988a: 217). But it is interesting that he recognises that “[t]he problem ... is essentially one of communication: even if people have perfectly adequate motives for cooperation they still need to know about each other’s motives and to trust each other, or at least the effectiveness of their motives” (p. 216). 14 The bottom line of trust is – and this makes Gambetta’s approach a sophisticated version of rational choice, but still one solidly rooted in it – that “it is a matter ... not just of feasible alternatives, but also of interest, of the relative attraction of the feasible alternatives, the degree of risk and the sanctions they involve” (p. 222). From a broader perspective, trust is here the key to overcoming the atomistic individualism’s aporiae in explaining social order merely on the basis of the actions of “rational egoists”, without however changing the basic assumption about the rational-calculating nature of the agent. Trust faces – and tries to solve – the problem of free-riding and the opaqueness of social actors to one another. Opaqueness is here the flip side of complexity. From the viewpoint of our project, it is important to remark that this opaqueness concerns the professional expert as much as the ordinary citizen (Sztompka 1999: 13). The leading theorist of trust within a sociological rational choice approach is James Coleman. His approach is rigorously centred on the actor’s choice, and bases trust entirely on the interpersonal/relational dimension (Coleman 1990). As the classics did, he first refocuses on a basic feature of trust that is prior to the characterisation of the agent, i.e. uncertainty about other agents’ motivations. Soon after, however, Coleman has to resort to neo-classical economics in reducing goals to interests and the agent’s rationality to the rationale of matching his expected gains against the anticipated losses as related to one particular course of action, a process whose outcome is “choice”. Similarly to Luhmann (see below), trust is correlated to the risk entailed in a given situation requiring a choice of maximisation of utility. The decision to trust is then rational if the expected gains surpass the predictable losses. However, trust is necessarily mutual, and as such it builds the basis of a particular form of “social capital”. In order to be mutual, trust rests on explicit bargaining among actors, supported not only by the attempt to match interests, but also by an agreement on sanctions. The basic form of sanction is, however, calculated mistrust, which hinders or reduces exchange. We reach here the opposite pole of Durkheim, since trust is here entirely conscious to the actor, and negotiated. In order to be effective, this type of trust must be embedded in homogeneous communities or networks. The corresponding normative sanctions are therefore a sort of extension of self-interest with other means. This type of trust hardly captures Durkheim’s society, or Tönnies Gesellschaft. This rational choice approach to trust is further refined by theorising trust not as a lubricant, but as a non-Durkheimian precondition of co-operation, supplementing contractual obligations, and in some cases substituting them (Gambetta 1988a and 1988b). Correspondingly, the view is more refined than in Coleman: “Cooperation requires trust in the sense that dependent parties need some degree of assurance that non-dependent parties will not defect: ... this need not take the form of ‘thick’ trust, individually based belief, but it must take some form or other” (Williams 1988: 8). In the contributions in the volume edited by Gambetta, more latitude is allowed to actors’ consideration of other actors’ motivations. “The next step ... is to claim .. that people will not trust others enough to bring about co-operation unless their assurance is to some extent well based: that is to say, unless people are also in general motivated, one way or another, not to defect if they are in a non-dependent position” (ib.). In a neo-Humean move, the introductory essay by Williams (a philosopher) complexifies the picture – compared with Coleman – of the basis of trust and the agent’s corresponding constraints, so reaching the intermediary conclusion that if “no egoistic motivation by itself will do”, and “non-egoistic micro-motivation will not do by itself”, one should ask “what kinds of non-egoistic micro-motivations there are, and which of them might relate coherently (more coherently than national-level [Durkheimian] thick trust could) to both the egoistic micro-motivations and the large impersonal structures of a modern state”. And the answer can only relate to a “given historically shaped society” (pp. 12-13). From there the discussion was open as to how to elaborate a view of trust as public good, and a role was allowed for the mediation of “moral codes” in posing limits to defection or unreliability (Dasgupta 1988). Trust emerges here as a pre-commitment, through which “we can impose some restraint on 15 ourselves and thus restrict the extent to which others have to worry about our trustworthyness” (Gambetta 1988a: 221). While this argument first looks like reintroducing Durkheim from the window after having expelled him through the door, the aid to reach this conclusion is evidence drawn from the theory of repeated games under uncertainty, that injects some sophistication in the view of the choices of rational actors over time. Repetition of games facilitates envisioning a long-term projection of self-interest, especially if there are conditions of successful communication. However, the theory stumbles here in its claim to see in trust a pre-condition of co-operation. Repeated games rather show that trust stabilises, but does not engender, co-operation. So we are back to the “lubricant” view of trust, now in a longer term projection. This function is further downgraded by John Elster to viewing trust as being a by-product of cooperation, and therefore at best improving the system, but never solving any basic problem related to the functioning of the system (Elster and Moene 1988; Elster 1989). The merit of these rational choice theorists is to ultimately do away with one-dimensional assumptions about the rationality of the agent of neo-classical economists, though this was their own point of departure; they effect this shift with arguments largely internal to the rational choice approach, but developed through the analysis of an historically and culturally diverse material. They show that “the assumption about unbounded, cost-free rationality is rarely met and that the uncertainty and ambiguity of many social situations result in interpretations of information by individuals in line with their preconceptions.” (Misztal 1996: 83). The resulting need to re-problematise and complexify the notion of the social agent, whilst starting from considering trust “within reason”, has been tackled by the philosopher Martin Hollis, in his stated goal to search for a form of reason flexible enough to account for trust resulting from every-day interactions and choices, but also undiluted enough as to distil a reasonable form of trust from other forms that, being unreasonable, cannot be considered as contributing to the social bond in a stable way (Sassatelli 2001b). But the Humean bottom line resists all attempts (from utilitarianism to revealed preferences) at refining the underlying model of rationality, and leaves trust-within-reason ungrounded. It seems therefore that promises in a modern world of transactions can only be uttered in a ‘consequentialist’ language, that is eminently Lockean (and, not to be forgotten, the ultimate and non-substitutable ‘hard’ consequences invoked by Locke are God’s sanctions: Dunn 1988). So that raising the problem of trust-within-reason (human reason) requires going beyond such consequentialism. The basic objection of Hollis, reflecting on Hume’s own question, is the following: “in a psychology where ‘reason alone cannot be a motive to any action of the will’ how can such a remedy bind us in foro interno?” (Hollis 1998: 65). It is the Humean anthropology of the social agent, and its legacy; in other words, it is the question itself that is probably formulated in the wrong way (sociologically). The short-cut of Kantian reason (as operating, for instance, in Simmel’s notion of autonomous personality, but also in Durkheim’s view of a transcendental – as opposed to transcendent – commitment of the agent to organic solidarity) is no substitute for an anthropological model based on instrumental rationality, since it relies too much on a moral obligation (of a Prussian-Lutheran kind) and does not help to ground trust-within-reason. The Kantian option is circular, in that it makes trust a token of rationality, rather than the bond of society emerging out of social relationships, and thus prefigures issues of governmentality (or the internalisation of self-limitation and self-control) that the rational choice approach stubbornly expunges from consideration. Hollis looks for a “third way” between Hume and Kant: capturing "the conditional character of everyday trust" and asking "which social relations generate trust-within-reason" (Hollis 1998: 105). It is symptomatic that Hollis does not provide a solution with purely theoretical tools. His main merit is to reformulate the question in an uncompromising way. 16 As we have seen, some rational choice theorists have come close to discovering that a basic deficit of their approach, and the underlying view of the social agent, was an inadequate consideration of the issue of communication. The problem is that it is highly questionable to try to solve this issue in any neo-Humean way. Communication is a much more complex issue than access to information, and entails the basic sociological dilemma of the double contingency. Without delving into such abstract issues, we have to admit that efficient communication cannot be warranted by rational procedures by way of a fiat, but requires a cumulative and tortuous learning processes. Trust has to be learned through subsequent stages, by trial and error. But the farthest level of sophistication reached by rational choice theory on this path is still self-centred, and relates to the necessity of actors to build their own reputation and bet on others’ reputation. On the other hand, through this way the theorem of double contingency is only solved within compact and homogeneous groups, something that often favours mafia-like or clientelistic social formations and a corresponding use of reputation and trust which are far away both from trust-within-reason and from the trust idyll of Tönnies’ Gemeinschaft (Gambetta 1988b, Roniger 1992). So we are in a dilemma where trust-cum-reputation is either encapsulated in tight mechanisms of monitoring within small groups, or is diluted – as Tönnies remarked – in the fragile and anonymous mechanisms of expertise. However, it is also possible to hold the more optimistic view of trust-cumreputation as the necessary survival of a form of “personal trust” that is recuperated due to the scarcity of information, and to the corresponding incentive to invest (also via engaging a learning process) in a kind cumulative evidence for trusting and distrusting, which would not stand a rigorous rational calculation. We will see, however, that in order to hold such a view we will first take into consideration the notion of systemic trust provided by Luhmann, and later re-examine comparative (and largely historical) approaches to various processes of trust-formation and crisis. The simultaneous tackling of the issue of the agent and the import of communication within interactions could not occur within the limits of rational choice theory. A partial advance that started from within these limits was achieved by Ostrom, who showed “how individuals imperfectly but persistently try to follow the norms and engage in mutual monitoring whilst building institutional arrangements in many different empirical situations” (Ostrom 1990). But this position seems to rely on the ex post facto realisation that people do indeed co-operate and trust each other under certain conditions. A further step, done by Barnes, leads squarely into theories of collective action, but stops at countering the view of the selfish agent by opposing it to a bastardised Aristotelian view: “the key to understanding collective action lies in the existence of mutual symbolic sanctioning considered as an aspect of communicative interaction, that is normal and natural to us social beings” (Barnes 1992: 263). A theoretically even less convincing way out has been the attempt by some rational choice theorists to add norms and beliefs, affection and love, to self-interest in the motivational prism of social agents. While having the merit of taking care of the micro-macro linkage, and providing a nice theoretical rebuttal of the one-sidedness of macro-functionalist (in fact macro-normative) approaches, the rational choice approach stumbles on the refusal to complexify the motivational prisms of social agents and only insists on – however varied – articulations of self-interest and of means-end rationality. Even more, as a contrast, we are reminded of Weber’s point that the dynamics of trust and distrust are also – or mainly – generated by sharing or clashing conceptions of the social goods, i.e. of the ends themselves of social action, which are often embedded in habits and practices, and in the underlying imbalances of power among social groups. Correspondingly, parallel to different notions of the goods, different modalities of trusting and distrusting are also enacted, provided that a general normative and legal framework is warranted via the enforcing of regimes of contract and by bureaucratic steering. As opposed to practice, 17 what can be retained from the rational choice approach is a view of trust not as a structural dimension of social relations, but as an ‘event’ related to the specific context within which the agent faces a variety of options and courses of action. Main lines of argumentation • The rational choice approach is grounded in a convenient reduction of the social agent’s motivation to a selfish and prudential calculation of the outcomes of his choice in terms of utility maximization, only constrained by information limits. • Trust is seen as a lubricant of co-operation, either as a form of rational calculation or as an externality. • Rational egoism is defined in strictly philosophical terms, i.e. altruistic motives might be included in the framework provided they are conceptualised as part of perceived individual utility. III. Trust and distrust: Luhmann’s systemic view, and the bifurcation between trust and confidence From within the Durkheimian-functionalist heritage, Talcott Parsons had theorised trust as a basic form of symbolic legitimation of the social system, that is cognitively internalised by social agents who learn to practice solidarity and ‘other-orientation’ mainly through their participation in secondary associations. While this process of trust-building is viewed as aided by social control, Parsons saw monitoring as a bottom-up phenomenon, i.e. as a tool for controlling those in power. This disregard for unequal power and heterogeneous sets of motivations among social agents, inherited from Durkheim, was transmitted to Parsons’ most famous pupil, Niklas Luhmann, who however championed a much more sophisticated and diversified theory of trust in society. This theory has been able to engage rational choice theory and appears today as the most frequently mentioned conceptual tool in debates about trust within food consumption arenas. He provides us with the most radically system-oriented definition of trust, keeps Parsons’ cognitivist bias, but eschews all culturalist orientations present in Parsons. First, Luhmann delineates a difference between familiarity and trust, that will play a crucial role in the scheme that we are going to develop. “Familiarity ... makes it possible to entertain relatively reliable expectations and, as a consequence, to contain the remaining elements of risk as well ... As against this, there is the future orientation of trust. Of course, trust is only possible in a familiar world ... But rather than being just an inference from the past, trust goes beyond the information it receives and risks defining the future ... In trusting, one engages in action as though there were only certain possibilities in the future” (Luhmann 1979 [1968/1973]: 19-20). Trust is exactly what helps social relationships to move beyond elementary forms (p. 88), since it "reduces social complexity by going beyond available information and generalising expectation of behaviour in that it replaces missing information with an internally guaranteed security. It thus remains dependent on other reduction mechanisms developed in parallel with it, for example those of law, of organization and, of course, those of language, but cannot, however, be reduced to them" (p. 93). The process is accompanied by a further crucial feature: that through the increasing 18 diversification of patterns of familiarity and unfamiliarity danger is replaced by risk, so that a risk-taking rationality is required, and trust is exactly what social agents put to work in order to govern this kind of rationality. Generalised trust is essential to the four main general media within society, that is truth, money, power and love (the “unlikely quartet”, according to Poggi’s introduction to the translation of Luhmann’s work on trust) which do not erase choice, but ‘contain’ it in an increasingly complex social world. Trust is thereby at the same time a property of the social system, and a crucial operator within the rationality of the social agents. “The basis of all trust is the presentation of the individual self as a social identity which builds itself up through interaction and which corresponds to its environment ... Whoever wants to win trust must take part in social life and be in a position to build the expectations of others into his own selfpresentation” (Luhmann 1979 [1968/1973]: p. 62). Luhmann’s definition, though largely original, shows thereby interesting links not only with classic social theory and with rational choice, but also with the sociology of risk and especially with the sociology of practice that we will present in the following sections (in particular with Goffman). This particular positioning of Luhmann within the theory of trust makes him much like the spider in the web. Trying to translate Luhmann’s jargon into a language shared by rational choice theorists, we could say that people trust on the assumption that other people also trust – and here we would still have an anticipation of reciprocity - but also – in a Goffmanian twist - based on the appearance that other people trust, everything being, so to speak, in order (Sassatelli 2002). Trust is a complexity-reduction device inherent in the web of social interactions (that for Luhmann is simply the social system) and not reducible to the sum of all social relationships. Almost a self-fulfilling prophecy. Unlike Durkheim, Luhmann radicalises Parsons’ view of the symbolic functioning of trust mechanisms, and stresses that it is through representing integration that social symbols perform an integrative function. Unlike Parsons and those Parsonsian authors (like Alexander) who sought a cross-fertilisation with the theory of civil society, he is rigorous in eschewing any dimension of Gemeinschaft from trust. Since trust rests on the appearance – or rather the reiterated collective-ritual performance – that others will also trust, the real cement is not ‘pure trust’, but, rests on a peculiar mechanism of systemic reflexivity, a sort of ‘trust in trust’. This corresponds well with Luhmann’s view that the social system always depends on the reiteration of its observation by an agent who is third to the myriad interactions taking place in the system. Such representational reflexivity grounds trust in trust as one of the main glues of Die Gesellschaft der Gesellschaft (the title of his last work epitomising his theory of society). The result is that any trust is ultimately trust in not one but the abstract system. The bifurcation of trust between embedded and disembedded trust does not make sense for Luhmann. Likewise, interpersonal trust has no autonomy, it is always a manifestation of a generalised systemic trust. The merit of Luhmann’s theory is that it takes care of the micro-macro link, by showing how furthering trust in micro-interactions contributes to the systemic dimension of trust. In a later move, in an article published in the volume edited by Gambetta, and where the parameters of discussion were set by rational choice theorists, Luhmann tried to improve on the receptivity of his approach towards the micromacro link and made a step to accommodate rational choice. There he redefined trust in abstract systems as “confidence”, and considered “trust” proper as more specifically related to the choice contained in one course of action. The latter (trust proper) is unique in giving the system the fresh input without which confidence would finally die out. “Confidence in the system and trust in partners are different attitudes with respect to alternatives, but they may influence each other; and in particular, a decline in confidence or an increasing difficulty in finding situations and partners which warrant trust may unleash deteriorating effects which diminish the range of activities available to the system (Luhmann 1988: 99). “Lack of confidence 19 and the need for trust may form a vicious circle. A system ... requires trust as an input condition. Without trust it cannot stimulate supportive activities in situations of uncertainty or risk”. (p. 103). This view does indeed dynamise Luhmann’s approach, even more so since the next move is to take account of distrust, i.e. the withdrawal of trust, as an additional element that functions to rescue the system from the erosion of confidence. In other words, distrust is not disruptive, but also a reasoned choice providing crucial input to the system. It is important that the system allows for some form of institutionalisation of distrust, so that this is not perceived as a challenge and a blind threat, but feeds into impersonal-systemic mechanisms (though, in this sense, it should act as ‘dis-confidence’). We see that Luhmann’s approach is highly formalised and so refrains from reintroducing any moral element in the definition of trust, a step that even some rational choice theorists are often tempted to make. The only ‘good’ to which trust contributes is the maintenance of the system, under the assumption that it operates under increasing complexity. This would make obsolete both the Durkheimian-Parsonsian and the Tocquevillian views of trust. On the other hand, Luhmann considers rational choice as finally missing the point in demanding that trust be a crucial intervening variable in a model for operating rational choice via calculation. For Luhmann this is clearly a performance that trust, as a property of the system and not of the agent, cannot deliver. A possible correction of Luhmann’s approach, that would make it useful for the comparative study of concrete configurations of trust, would be to move from a much too generic reference to the social system, towards discrete institutional configurations. “Trust cannot be fully understood and studied without the examination of institutions as repositories of a legacy of values and without addressing a practical issue of how far human beings’ concepts of duties and obligations are influenced by the societal institutions which organize ways in which people are bound together” (Misztal 1996: 25). The merit of Luhmann’s work is therefore in helping foreground aspects of the theory of trust that are typical of complex systems. These aspects have been further elucidated by other authors and approaches. “Trust is particularly relevant in conditions of ignorance or uncertainty with respect to unknown or unknowable actions of others” (Earle and Cvetkovich 1995: 38). This is the main link to risk as a structural condition of social life. Or, even more radically, “trust is a bet about the future contingent actions of others” (Sztompka 1999: 25; cf. also Coleman 1990: 99). Though being true primarily of the realm of interpersonal trust (Good 1988), this condition that is inherent in increasingly complex and differentiated social systems affects the way the intersection of multiple and overlapping relations of trusts (or corresponding “bets”) create a web of calculated anticipations. This is the condition of the emergence of shared standards of anticipation of consequences, which are produced and maintained at the intersection of social control and self-control. Trust is a substitute of control and monitoring (Dasgupta 1988), but is structurally different, in that while control is intended as continuous, and continually providing the background information for steering measures and decisions, trust incorporates in itself a move to take a decision (cf. Luhmann 1979 [1968/1973]: 24-25). In other words, for producing those standards of anticipation of consequences of each other’s behaviour that is essential to the functioning of social systems via communication and co-operation, one needs not only control but also trust. A corollary of this is that trust is neither fully in the hands of the agent, nor fully incorporated in the background – largely institutionalised – mechanisms of “confidence”, but is a crucial link between the agent and the system. What Luhmann also helps to point out is that the opposite of trust, distrust, “is also a bet, but a negative bet. It involves negative expectations about the actions of others (of their harmful, vicious, detrimental actions towards myself), and it involves negative, defensive commitment (avoiding, escaping, distancing myself, refusing actions, taking protective measures against those I mistrust).” (Sztompka 1999: 26). Distrust is therefore not a token of passivity, since it also involves choice and action. For analytical 20 purposes, “mistrust” can also be useful, defined as “a temporary intermediate phase in the dynamics of trust-building, or of trust-depletion. Mistrust is either a former trust destroyed, or former distrust healed” (Sztompka 1999: 26-27). Mistrust might be a concept susceptible of operationalisation in indicating a state of suspension of the attribution of trust; i.e. of betting on somebody or something, before moving on in the direction of either trust or distrust. As suggested by Unni Kjaernes, Luhmann’s view of complexity reduction cannot be equated with a passive role of consumer’s trust exercised within consumers’ choices: “Our comprehension and interpretation of reality entails a reduction of complexity, with implications for feelings of inner safety. This can take place by refusing to consider the impressions that may threaten trust (in situations when the object of trust is indispensable) – or the opposite, by differentiating the trust relations. This latter option requires substitutes to the object of distrust. Having felt strongly that it is safe to eat meat in general, new information on risk may be neglected in order to avoid anxiety. If we know the options, we may differentiate types of meat, for example, by preferring meat from a certain butcher or supplier. Alternatively, we can choose to abandon meat altogether (Kjaernes 1999a). However, system trust is certainly not adequate in accounting for a fully active trust of the agent/(consumer), for which a theory of reflexivity is required. Main lines of argumentation • The introduction of a difference between confidence (based on a web of shared anticipations of each other’s behaviour) and trust (which is a voluntary engagement and calculation). • Emphasis on a systemic level, wherein trust acts as a complexity reduction device feeding into the cycle of production and reproduction of confidence, and particularly crucial when confidence is eroded (though a parallel collapse of trust and confidence is also likely). • Distrust (the active and conscious withdrawal of trust) is not necessary disruptive and can be institutionalised in the system. IV. Trust, risk culture, and reflexivity In a polar relationship with Luhmann’s systemic theory of trust, while sharing with it some key-notions – like the notion of risk within increasingly complex societies – we face Giddens and Beck’s theory of trust within modern society. This strand of theory envisions the development of disembedding mechanisms and corresponding individualisation mediated by a reflexive appropriation of knowledge. In this context, as with Luhmann, trust is the solution to keep under control the risk that results from disembedding processes, and the related separation of time and space (Giddens 1990). The bottom line of the discussion of trust in Giddens is strictly related to the identity construction of the individual: “Giddens, drawing mainly on contributions from social psychologists and symbolic interactionists, has discussed the basic role of trust in a caring family climate during early socialisation and the evolution of personal history ... ” (Kjaernes and Dulsrud 1998: 3). As against the embedded, “local”, “traditional”, “natural”, and “personal” forms of trust based on kinship and religion (Giddens seems to equate these four attributes almost by default, or to consider 21 them as largely overlapping: though in one case he delivered a coherent definition of “tradition” that would deserve to be separately discussed: Giddens 1994a), the modern context relies on individuals who, without any traditional authority, have to self-reflexively reconstruct their own identities, or rather to self-position themselves in a complex social world where risk, unlike danger, is not natural, but socially constructed. “Risk society” is a designation first used by Beck (1992 [1986]): “In the risk society, the recognition of the unpredictability of the threats provoked by techno-industrial development necessitates self-reflection on the foundations of social cohesion and the examination of prevailing conventions and foundations of rationality” (Beck 1994: 8). While Giddens starts – like Tönnies – from an extremely dual view of social relationships, of Gesellschaft completely erasing Gemeinschat, unlike Tönnies he does not see trust in modern societies as fragile, because solely based on the reputation of expertise. First, for Giddens expertise is itself sustained by reflexivity. Second, by synthesising Durkheim, Simmel, and a simplified view of Weber, he sees symbolic media (including money) as also concurring in securing trust. But most important of all, and this is his original contribution, a third dimension is highlighted as securing trust, and this is the process of deliberate, self-conscious re-embedding of trust in personal relationships. In this way, trust bifurcates in trust in abstract systems (“faceless commitment”) and trust in persons (“facework commitment”). One is left with the impression that the latter is an existential arena of reflexivity and affection that makes the former form of trust, the really social one, both systemically functional and individually bearable. Here we meet a practical-habitual dimension of trust that is better developed by other approaches explicitly centred on practice. Giddens does not go beyond emphasising that the result of the training of anxiety reduction through facework commitment is a “practical consciousness” serving “routine or tact, seen as a cluster of rules which regulate activities in ways that make them predictable and orderly” (Misztal 1996: 92). Giddens, however, insists on seeing routines and practices as resting on wellfunctioning abstract systems, and here we are back to a view that is very similar to Luhmann’s. Moreover – and this sets him apart both from Luhmann and the theorists of trust as habit and practice that we are going to examine in the next section – Giddens insists in thematising the construction of trust as part of the process of building a social self as a reflexive project, a view that probably suits at least the self-image of educated urban middle-class/middle-aged populations of at least some (especially post-Reformation “secularised”) Western societies (for a similar critique see Warde 1994). The bottom-line is that a considerable burden of responsibility is placed on the individual for deciding what and whom to trust and so taming risk, while not being an expert. If applied to food consumption systems, it has been remarked that “Giddens’ proposition would actually imply that the consumers’ awareness of risk on food issues and turbulence in the food market is a modern phenomenon. “And that a reflexive process expressed as a consumer concern about the food provision process is discordant with pre-modern ways of life, ... [and] that ... the idea of reflexiveness ... involves a certain degree of irreversibility. Once reflexivity, and hence suspicion, is introduced to our attitude towards others or an issue, there is no way of return. The challenge of high modern societies, according to Giddens, is to maintain legitimacy by responding with continuing opening up processes and ‘active trust’ ” (Kjaernes and Dulsrud 1998: 8). But it is exactly the postulation of this “active trust” based on a strictly individual form of reflexivity that is problematic. Not only does it presuppose a distinctive, and probably sociologically rare form of mature modern self (who “has turned egoism into a virtue, making what for Durkheim was a pathological problem into a highly positive form of conduct”: Warde 1994: 889), but it also entails “that identity-value has entirely supplanted use-value and exchange-value in consumer decision-making” (p. 891). 22 Kjaernes and Duslsrud have suggested a collective form of reflexivity, which is not necessarily provided by the cumulative result of individual reflexivities à-la-Giddens-and-Beck. Collective reflexivity – and corresponding social action, as mediated by consumer associations for example – would produce a “tacit trust”, as the one that is engendered by “the extensive demand for labelling of products ... concerning the mode of production, country of origin, etc. ... When such demands are turned down, for example in the case of genetically modified foods, this is met by distrust among many consumers. Thus there seems to be a tension between the (quest for) reflexivity and the need of routines” (Kjaernes and Dulsrud 1998: 9). This proposition summizes two elements: to look for reflexivity in the associational and institutional mechanisms that produce reasoned trust and distrust, and to pay attention to the routine components of market action, that can also incorporate large amounts of collective reflexivity. One step further, Giddens and Beck’s overemphasis on individualisation makes the public – and often conflictual – mechanisms of adjudication of trust and distrust across different groups and social classes scarcely intelligible, since public disputes are clearly dependent on elaborate strategies and corresponding forms of leadership (based on power games, imbalances, and conflicts) that their theory does not take pains to deal with. Main lines of argumentation • Building on a theory of reflexive individualisation, we get a picture of modernity as characterised by relatively isolated actors who face ever more remote risks; traditional forms of trust are said to be forever lost. • Experts (including risk evaluators) are called upon to clarify the terms within which different risk choices have to be taken by reflexive individuals. • Local knowledge is re-embedded information filtered through individual reflexivity, but derived from specialised knowledge. V. Trust as practice, habit and routine In contrast to all previous views which stress trust as an individual or collective good furthering social cooperation, a fifth strand of theory takes a step back from such a pre-emptive positive evaluation, and attempts to ground trust on a more fundamental level of social practice. Not by chance this approach is first represented among phenomenologists and ethnomethodologists. “Trust, understood as a specific type of habitus, allows us to account for the fact that social agents perceive the social world as stable” (Misztal 1996: 11). This approach presents an interesting analogy to what Bourdieu says of habitus as “mak[ing] coherence and necessity out of accident and contingency” (Bourdieu 1977 [1972]: 87). We will see how from there on the notion of trust might be able not only to re-capture dimensions of power imbalances and conflict that all previous approaches obscured, but also aspects of social life and social agency that dual views like those opposing traditional and modern social worlds, or Gemeinschaft and Gesellschaft, downplayed or even conceptually suppressed. Through this last strand there is also the chance to recapture a bit of Weber’s complex view of society and trust. In fact we will see that the two concomitant neglected dimensions of power imbalances and the reconceptualisation of the social agent on the basis of habitual forms of trust, can be fruitfully linked to each other. 23 The idea of identifying a distinctive approach to trust as habit or habitus is first to be attributed to Barbara Misztal (1996). We are here following in her footsteps (though not uncritically), but we also rely on the notion of “tacit trust” evidenced above from the work of Kjaernes and Duslrud as specifically related to food consumption systems. However, Misztal’s proposition comes at a huge expense of downplaying the distinctiveness (and the potential usefulness) of each of the main social theorists addressed. Not only is Bourdieu’s sociology more complex (and correspondingly fragile) than the one of phenomenologists and ethnomethodologists, but we cannot even speak of “Goffman and Garfinkel” as we do with “Giddens and Beck”. Thus, what follows has to discount a certain simplification. In Garfinkel (1963), as well as in Goffman (1970), trust is defined as manifest in social actors’ expectations of “normality” which are upheld by rule-following behaviour. This is the key to make the social world predictable and readable, and also stable and reliable. Based on Schultz’s notion of the “reciprocity of perspectives’ among social actors, in Garfinkel the emphasis is on the constraining power of rather unreflective, taken-for-granted frameworks of common reference-construction. As against structural objectivism, the resulting routine structures of social life are a delicate product of a permanent tension between structures and subjective perceptions. We cannot understand the social agent without analysing habitus, that is the product of this dialectic tension, and that delimits the ensemble of durable and transposable dispositions which produce and reproduce social structures (Bourdieu 1977 [1972]). Through everyday practices, the habitus re-enacts the past in the present, and relays it to the future. At a first level of conceptualisation, habitus is a more precise notion than “tradition”, which is used by sociologists in a derogatory sense, or, inversely, by romanticising it (the two extreme cases being well represented by Tönnies and Giddens among the authors so far considered), in both cases denoting the taken-for-granted, automatic and basically unreflective character of practices. It is not by chance that Bourdieu first developed his notion of habitus and practice in his investigations of Kabylian society, a kind of theoretically informed socio-anthropological study for which any pre-emptive dualism of tradition and modernity does not make much sense. The merit of habitus is in striking a balance between the atomistic status of the agent in post-Humean social theory and the overdeterminism of the notion itself of social structure. The system of dispositions is at the same time inherited from the past (including the individual past) but open to the selective feed-back of new experience. In this context, and even if Bourdieu hardly makes an explicit mention of trust in his original theory, we can see trust as a specific manifestation of habitus, “understood as a system of lasting and transposable dispositions which, integrating past experience, functions at every moment as a matrix of perceptions, appreciation, and actions, and makes possible the achievement of infinitely diversified tasks, thanks to analogical transfers of schemes permitting the solution of similarly shaped problems” (Bourdieu 1977 [1972]: 82-83). We see that trust as habitus is not unreflective as assumed in the mainstream sociological characterisation of habit, custom, and tradition as constraining or destroying the individual’s authenticity, subjectivity and openness. This also marks a difference with Garfinkel’s view of how individuals, in their rule-following behaviour, confirm and adapt rules through techniques of normalisation. Here the space for reflexivity is not absent, but rather thin, in that the reflective potential of social action is rarely retrieved. This potential is only activated when an interpretative break-down occurs as a consequence of an abrupt and unexpected breach of rules, rituals, common-sense assumptions, and finally trust in other actors who are assumed to share the same assumptions as a matter of social and moral necessity. Goffman (1972) lays more stress on the work done to reinstate the ritual order of interaction through “facework”, a notion that implies tactful and sometimes even inventive accommodation, but hardly a reflexivity that might feed-back on the identity and the motivational prism of the agent. Interpretation is rather intended by Goffman in a 24 theatrical sense, as interpretation of one’s own role in a given situation governed by a script that, as a set of rules, does not cover every detail and leaves open spaces to adaptive performance. Trust is here an important accompanying dimension in the production of etiquette and civility and in the negotiation of public space (see also Goffman 1963), but it is neither a structuring factor of social interactions, nor their main product. Stability and reciprocity of mutual expectations is the form that trust takes in this kind of theorising. We would need to overstretch the points made by Garfinkel and Goffman to see in their views of trust also the production of “shared meaning” and therefore the legitimisation of authority. To which extent is this form of trust merely a matter of presentation: a ceremony and a rhetoric which fixes identities, bonds and worldviews? Is trust a modality governing choice via the self-positioning gestures and rituals of a world of face engagements of ultimately atomised selves, or via the habitualised practices and the distinctive tastes structuring a world of socio-cultural stratification? Can the elaboration of a notion of “tacit trust” be a main conceptual gain from within this strand of social thought that we have juxtaposed to the Giddens-Beck’s approach? Kjaernes and Dulsrud have suggested to use tacit trust in relation to Swidler’s theory of “culture in action”. “A number of processes influence whether trust and distrust remain tacit, are made explicit or are once again rendered tacit (as a result of routinisation). In [Swidler’s] theory there are clear parallels to the concept of tacitness and reflexivity in her terms ‘common sense’ vs. ‘highly articulated, self-conscious ideologies’, at least at a collective level. However, for her it is not a question of tradition vs. modernity ... Following this argument, both individually and collectively there is a dynamic between tacitness and reflexivity which also applies to trust, but where tacitness is the dominating mode” (Kjaernes and Dulsrud 1998: 8). Trying to distill a useful way to conceive of trust as a form of habit/habitus, Misztal suggests the following tripartition: “social habits of conduct or routinised practices; mental habits or background/taken-for-granted assumptions; and ceremonial habits or rituals” (Misztal 1996: 104). In all three forms, habit helps pattern everyday life, and operates therefore (as trust does) as a reducer of complexity and therefore as an aid to deliberation. “Simplifying social complexity by taking a risk (the function of trust) and simplifying social complexity by dependence on familiar practices (habit) can be seen as part of the same process” (p. 106). Here we see a convergence with Luhmann’s view of trust, where he observes that “one trusts if one assumes that this behaviour will fit in meaningfully with one’s own pattern of life; one distrusts if one reckons that this will not be the case” (Luhmann 1979 [1968/1973]: 72). Our own habits are, inversely, what facilitates other people to trust, in a Luhmannian sense. On the other hand trust-as-habit also entails a binary and potentially exclusionary function of distinguishing between familiar and unfamiliar, them and us. While there is some potential of operationalisation of this notion (see all following sections), we should avoid, however, trivialising this view into postulating an ontological security – as Giddens does – that disguises unpredictability through a feeling of normality as the default condition of trusting. Overemphasising this condition prevents us seeing the aspect of social reality that is shaped by the competition and clash between different and partly incompatible conceptions of social goods and values and of ways of organising related social practices, which might become institutionalised to various degrees. Here it is possible to link the competitive dimension with the reflexivity that is also entailed in the notion of habitus. In the necessary work to redefine the agent-in-context that is necessary for a flexible but coherent view of trust, we should also avoid the opposite danger of romanticising agency vis-à-vis structure (Collins 1992). The possible key to a balanced and conceptually enriching view of trust as habit that tries to incorporate the somewhat dissonant Weberian element (as compared to all strands on trust here examined) is that the undoubted routinisation of social life that creates patterns of familiarity and unfamiliarity is not 25 contrasted with the sheer – and sociologically obscure – spontaneity and originality of agency, but with taking into full account Weber’s view of charisma in all its distinctive implications. These cannot be reduced, as Bourdieu does, to “the symbolic violence [of] the transfiguration of relations of domination ... into affective relations” (Bourdieu 1998 [1994]: 102). Charisma is a possible conceptual key to the understanding and classification of what is mainly identified as “local systems of trust” as related to “home” and oikòs. The next move would then be to relocate them in the wider undertaking of a comparative sociology of types and configurations of trust, and here the Weberian legacy and comparative sociology of Eisenstadt and some of the scholars who worked with him (like Seligman 1997, and Roniger 1992) might be very useful in linking the “order of practices” to the “(cohesive) framework of reference” of social relations. Main lines of argumentation • A variety of contemporary theories which might be subsumed under the so-called “theory of practice” underline that trust is to be conceived as embedded in practice, prudential and yet only partially “rational”. • While acting, people rely on trust and at the same time construct trust. • Trust is not easily separated from social order and identity building. VI. The distinctiveness of trust vs. confidence and familiarity The reconceptualisation of trust vis-à-vis confidence and familiarity by Adam Seligman might be particularly useful in that it goes to the root of a series of aporiae that we found in virtually all authors and schools so far examined. In particular it has the merit to develop Luhmann’s later distinction between confidence and trust in a way that relates it immediately to exchange, the functioning of markets, and their regulation, and more in general to issues of social organisation and specific institutional configurations. Seligman starts out from reflecting on classical formulations of the issue of trust. “Trust, as both a solution to and an articulation of a specific interactional problem, is tied to a particular idea of the self that we identify, most broadly, with modern social formations” (Seligman 1997: 8). For solving an interaction problem that he admits as increasingly centred on “risk”, it is imperative to avoid “the Scylla of rationalchoice perspectives on trust (which are often but extended studies on the conditions of confidence in any interaction) and the Charybdis of a normative perspective (which would apotheosize trust as the conscience collective of society ...)” (ibid.). While accepting an orientation to problem-solving and conflict management in a society made of interconnected networks between citizens, household units, voluntary organisations, institutions, and obviously markets, he states that trust as the notion grounding a pre-contractual element in social arrangements (“the necessity of rules regulating markets and governing the workings of contract that are themselves not subject to contract”) becomes problematic exactly because of its relation to modes of social organisation and institutional configurations. The bifurcation of the notion into “trust” proper and “confidence” as suggested by Luhmann is crucial, as it is also rooted in popular perceptions (Vertrauen ist gut, Sicherheit noch besser). Seligman first tackles Giddens’ tripartition of trust into “trust in persons, trust in institutionalized personal ties, and trust in abstract systems” (p. 18). Seligman sees the first as 26 undertheorised, and the second and the third as scarcely differentiated by Giddens, for whom “abstract systems themselves (whether of money or ‘expert knowledge’) are, in essence, no more than generalized modes of exchange and, as such, a form (albeit highly differentiated) of institutionalized personal relations” (ibid.). A more appropriate starting point is identified by Seligman in Luhmann’s distinction between “confidence in the proper operation of a system which is, in essence, reliance on the proper workings of general media of communication” and trust as another kind of expectation, yet also generalised, and addressed to the general other as showing a pattern of action – based on freedom – coherent with the personality traits displayed by him/her (p. 19). While Seligman suspects that confidence in the later Luhmann (the one who interacted with Gambetta) corresponds to what the earlier Luhmann defined as “system trust”, one might have a different impression, i.e. that while the later Luhmann defined confidence as different from trust, this latter trust remained basically “system trust”. The objection, as also suggested by Kjaernes, is whether this would allow for some agential form of “system distrust” parallel to the “disconfidence” that Luhmann hypothesises as the phenomenon of the creeping (less-than-active) withdrawal of confidence. In identifying an agential form of distrust that makes sense from the viewpoint of system theory, we cannot rely – for the reasons explained in the two previous sections – on a form of individual reflexivity à la Giddens and Beck. However, Luhmann’s distinctive definition of trust as addressed to the socially available and visible other suggests a convergence with Goffman in showing the trust that really matters to the maintenance and reproduction of systems of social relations, while confidence represents instead the take-for-granted dimension of the system’s functioning, or the systems’ operating code (p. 20). This latter dimension corresponds to what other theorists (like Parsons, P. Blau, Baber, and Shapiro) have more simply and simplistically defined as “trust” (“The very process of iteration, the memory of previous encounters, and the construction of a reputation are all functional equivalents of what we are calling confidence in the system”: p. 26). Seligman suggests that this convergence on a unified notion of trust, that he, like Luhmann, rejects, has something of the Humean paradigm, and of its weaknesses, trust being “an artifice of society, but one of a very peculiar nature, tied as it is to a fundamental ambiguity of human condition” (p. 22). Seligman finds a way out of this paradigm by reinterpreting the analysis of role-taking as a process delivered by the symbolic-interactionist school in a tendentially – though not entirely – Luhmannian way, “as emerging out of interaction, as less determined by systemic constraints, as essentially more negotiable, more a function of reciprocity between role incumbents” (p. 23), and this is exactly what makes trust apart from confidence: “trust enters into social interaction in the interstices of the system, or at system limit, when .... systemically defined role expectations are no longer viable” (p. 25). This sketch of a definition of where trust might lie can be very useful for the operationalisation of the concept of trust as referred to food consumption systems, and to overcome the current dichotomisations between “universalistic” and “local”, or between “modern” and “traditional” references of systems of food production and distribution, as we will see in the following section. It might be this distinct variable of an “interstitial” trust that makes a difference in consumers’ behaviour within one particular setting. Interestingly, trust so defined is crucial for modern social systems, but is not exclusive to them. Also in pre-modern settings, trust was located at the borders of the system, and became crucial in liminal stages of system breakdown and transformation. The distinctiveness of these modern systems is that the liminal is no longer a stage but is perpetuated in the system, which permanently needs this type of interstitial trust. In a certain sense, a breakdown of confidence (as in the case of food scares) cannot be restored by directly rebuilding confidence in the system, but by reconstructing this liminal and interstitial trust! 27 It is basically the relation of tension between institutionalised confidence and interstitial trust that characterises a certain type and regime of trust, as exemplified in the following section that addresses the case of food consumption. The historical example used by Seligman, by reference to Alan Silver, is related to the emergence of “friendship” as a distinctively modern form of social relation based on the emerging form of trust by authors of the eighteenth century Scottish Enlightenment: “Friendship ... emerged in the early modern era as a realm of interaction distinct from the motives of self-interest and rational calculation ... in contrast and explicit rejection of those types of behaviour identified with aristocratic court society, where the calculation of every word and gesture was necessary to achieve success” (p. 32). One step forward, as noted by Silver, Adam Smith established “the moral basis of commercial society in the association of private individuals meeting in a social space not shaped by institutional constraints” (Silver 1997: 15). Common to thinkers from Hume through the Scottish moralists to Smith – and in spite of their differences – was an awareness that side by side with the emergence and crystallisation of market exchange, contract law, and impersonal institutions (the “invisible hand”), a new form of social relation emerged that could not be reduced to any of them, yet was in a sense their condition: trust. In a clear rebuttal of Tönnies, Seligman states that “the ‘trust’ that is so often seen to bind members of tribal, peasant or other types of premodern societies is not trust at all but confidence in a very particular mode of social organization based on ascriptive categories” (p. 37), while within modern social relations “the greater indeterminacy and the greater negotiability of role expectations lead to the greater possibility for the development of trust as a form of social relations” (p. 39). In this sense, trust is never purely “personal” nor “systemic”, but emerges “in the interstices of systemically defined role expectations - that very lability that is built into modern systems of confidence and which allows the emergence of trust in a manner that was not possible in other forms of social organization” (p. 40). By being so far defined as other than both faith and confidence, trust is tentatively defined as “some sort of belief in the goodwill of the other, given the opaqueness of other’s intentions and calculations” (p. 43), therefore “a recognition of alter’s agency ... in modern social formations with their increased division of labor and system differentiation” (p. 62-63), by which we are back to Luhmann’s definition – with the little qualification that trust is less a property of the system than a necessity of the system’s limits. And here, as in Luhmann, trust’s relation to risk becomes apparent: “Trust is not only a means of negotiating risk, it implies risk ... the risk implied is precisely that which is inherent in alter’s realization of agency” (p. 63). “Trust cannot be demanded, only offered and accepted”. Therefore it can never be edified via institutional means or through a concertation among social actors, markets, associations, and public authorities, but emerges through particular configurations or regimes that depend on the distinctive ways through which trust is produced at the limits of the system. What the “limits of the system” are – so it seems – concerns to a great extent the relation between trust and “familiarity” in a given system, a relation that might then be crucial for the characterisation of different trust relations. “Familiarity ... is ... the establishment (real, or often, fictitious) of that generic human bond rooted not in difference, but in sameness or identity. Trust, by contrast, recognizes difference including the different possible bases for alter’s strong evaluations and ‘trusts’ that these will limit alter’s freedom, channelling the use of other’s agency in ways not inimicable to our own interests or desires”, up to the point that “where familiarity (real or assumed) is sufficient (and often is), we are spared from taking the next step into trust” (pp. 69-70). Durkheim’s mistake was to construct organic solidarity not on trust but on a “universal otherhood” based on an upgrading and universalisation of familiarity, whose philosophical underpinning could only be a “transcendental self” of a Kantian type, whose unfeasibility for justifying trust has been stressed by Martin Hollis as well (1998). This Dukheimian-Kantian pseudo-solution completely eludes the risk implied in trust as based on the recognition of discrete, not universal forms of otherness. 28 Therefore, a crisis of confidence within a given system can be restored – as shown by ethnomethodological investigations – by either recovering elements of familiarity, or by constructing or reconstructing trust on a novel basis. Here trust and familiarity are either mutually exclusive, or can act in different combinations. It is these different combinations that characterise different settings and societies. Trust is the form that more purely corresponds to contract law and self-regulating markets, but where these are not able to cover all of “exchange”, familiarity might play a correspondingly large role. Gemeinschaft and Gesellschaft are both components of modern social system, and here Luhmann is weak in identifying the social system entirely with Gesellschaft. Considering confidence as given and “circulating” in every social system, what makes a difference among various regimes is the relationship between familiarity, as the routine mechanism for maintaining confidence (therefore coming close to habitus), and trust as located on the system limits or interstices. The paradox is that with the growing complexity of social systems, the importance of the system’s limits, and therefore of trust, increases. Confidence requires increasingly more trust to perpetuate itself. Trust is there where social actors actively trust, and this makes trust ever more important but inherently fragile. Yet, as we will see in the example of food consumption, there is the risk that by focusing on trust we exaggerate its importance to the detriment of the enduringly crucial relation between trust and familiarity. This might be useful to the discussion of conceptual short-cuts which result from reducing trust to neat dichotomies such as traditional-modern, local-universalistic, embeddeddisembedded and the like. Luhmann might help us a bit again here, where he says that “the relative emphasis on familiarity, confidence, or trust is not simply prescribed by social structures or cultural imperatives. To a large extent this remains a matter of definition, and particularly with respect to confidence and trust one can choose to see the relation – [e.g.] the decision to see the doctor – as either unavoidable confidence in the medical system or a matter of risky choice. Belonging to the same family of self-assurances, familiarity, confidence and trust seem to depend on each other and are, at the same time, capable of replacing each other to a certain extent” (Luhmann 1988: 101). Parallel to that, even Giddens’ view of active trust as part of the reflexive self of post-traditional societies can be seen as boiling down to the disappearance of “final authorities”, but not of authorities as such: “since there are no super-experts to turn to, risk calculation has to include the risk of which experts are consulted, or whose authority is to be taken as binding” (Giddens 1984: 87). This entails that “local knowledge is re-embedded information derived from abstract systems of one type or another. There is not a one-way movement towards specialisation” (p. 88). Main lines of argumentation • The reformulation of the distinction between trust, confidence and familiarity (originally made by Luhmann) permits the conceptualisation of confidence as a mechanism inherent in all social systems irrespective of their degree of differentiation and complexity. • Trust emerges in the interstices of the system through incongruences due to lack of consolidated role expectations, role duplications or overlapping, and renewed (perpetual) differentiation of sub-systems. • How familiarity is being built and how it is related to trust proper is among the key elements which we may consider as distinguishing different configurations of trust in different societies. 29 VII. The interplay of trust, confidence and familiarity within food consumption arenas: Consumers’ choices, markets’ responses, and the reconfiguration of regimes of trust From the viewpoint of general trust theory, the consumption of food reflects a distinctive type of social practice – and a corresponding specific entry point into the problematic of trust. This practice puts in evidence both the habitual character of trust, and a particular strong link to issues of identity formation and the maintenance of familiarity, due to the plain fact – that is both material and symbolic - that food becomes part of the consumer’s body on a daily basis: man ist was man ißt (Sassatelli and Scott 2001). The location of the origin of the food commodity is a particularly powerful tool of identification or disidentification for the consumer, much more than for other (i.e. non-food) items of consumption, since eating, the act of consuming food, defines an arena of social interaction traditionally “characterised by congeniality and caring”, but also structurally “associated with fundamental gravity and deeply rooted anxiety” (Kjaernes 1999a: 3-4?). Moreover, it has been argued that trust might or might not be essential to co-operation – or important for it to different degrees – so that we could ask “why bother about trust” if co-operation can be achieved by other means (Gambetta 1988: 224; Elster and Moene 1988); in the case of food consumption cooperation without trust is not an option, and trust is relevant on its own. Due to these peculiarities, it would be worth asking (especially if one is interested in engaging with Luhmann’s definition of trust as developed by Seligman) if food consumption defines a distinctive field or arena, or even a “subsystem” within society, in that the type of trust/distrust operational box that operates within it rests on specific symbolic clusters that mediate the attribution and the generalisation (or focalisation) of trust. This is why trust as related to food provides one of the most interesting fields for the comparative study of configurations of trust relations, also or especially in a diversified European framework marked by the harmonisation of markets and divergent practices of consumption, modes of trust construction and defiance, and corresponding mechanisms of identification impacting on the structures and the boundaries of food markets. These differences cannot be located solely in the variety of national agro-alimentary practices, but have to take account of evolving and still largely non-crystallised consumers’ practices, which are affected by both habitual patterns of choice, and by the impact of cyclical food scares, as filtered through media representations, whose specific impact on the making of food scares is still largely unexplored. The latter represent another element of the distinctiveness of trust as related to food: the overproportional dependence on media communication – whilst everyday politics still certainly plays a role – within a public space dominated by symbolic references that can be manipulated by big (national and European) as well as small (local and regional) media. The whole issue of the reconfiguration of citizenship rights as related to food consumption is thereby affected not only by the trade-off between the model of the consumer-citizen influencing purchasing patterns vs. the citizen-consumer participating in setting up political agendas for furthering specific local, regional or national circuits of food production and commercialisation; but also in relation to the sensitiveness of either model (or their combination) to the shocks affecting trust through periodic food scares, whose intensity or frequency has no equivalent in other political or social arenas of trust as related to institutions, professional roles, and markets. The latter issue can also be formulated as the question why risk has appeared so far as the most appropriate category to tackle the problem, so that formulating the issue 30 in terms of trust vs. risk requires a deep work of conceptual reconstruction and its verification through accurate empirical case studies. This is why we have to turn first to the strand of literature known under the label of “risk analysis” (i.e. the analysis of risk as perceived, or simply of “risk perception”). While originating in a “psychometric paradigm” concerned with analysing reactions to (especially technological) hazards, this school has on the way anticipated some theoretical debates, including the one on “risk society”. However, this risk research approach reveals a self-propelling dynamic immunising it against a radical conceptual self-criticism and revision. Basically, “risk perception and risk-taking behaviour can be characterised ... as judgements under uncertainty” (Löfstedt and Frewer 1998: 4). This recognition creates a first, though tiny, conceptual intersection with the discussions reviewed in this paper. On the other hand, the main representatives of risk research admit that their approach “has been criticized for its lack of integration in the social and cultural context, and its impact on the institutional sphere” (ib.). And we read in the outline of the trust-in-food project that “the reason why we ground this study on the term 'trust' and not 'risk perception', is precisely because we want to put stronger emphasis on social, historical and present institutional determinants of safe foods, that is how the food supply sector and the food policy regimes work.” The question then is how to manage the basic divergence with the need to cope with the tiny intersection? An increasing confrontation with several disciplines and strands of research has been present within risk perception studies. How deep the conceptual engagement has been is a different question. “Risk research, since the empirical work performed by Starr in 1969, had been largely dominated by psychologists who were familiar with the normative methodology favoured by the psychometric paradigm, and had strong backgrounds in cognitive psychology. While the input into the area from anthropology continued, students with other disciplinary background began to contribute to the area. Social psychology, political science, sociology and geography now started to shape the directions that research into risk were taking. As the field progressed, and as more scientists (especially in the United States) became involved in various research programmes, the multidisciplinarity of the area began to become more apparent and also more integrated” (pp. 5-6). Risk researchers engaged with such divergent theoretical approaches as Putnam’s view of social capital and civic engagement and Douglas and Wildavski’s (and Thompson’s) “cultural theory”, in search for the institutional determinants of riskrelated behaviour that had eluded their efforts in its earlier phases. However, the engagement with “cultural theory” precipitated risk analysis in the verification of four to five typified individual attitudes to risk. The ensuing empirical verification studies showed an acute deficit of the theory of “risk and culture” in terms of its operational potential, and re-exposed the need for risk analysis to capture institutional factors. More fruitful has been the ramification of risk analysis into studying the role of media in the social amplification – or attenuation – of risk perception. “From a cognitive psychological perspective, simple but frequently repeated messages are likely to result in the formation of an ‘availability heuristic’ or establishment of an internal rule that signifies potential danger” (p. 10). However, this subfield of “risk communication analysis” has not delivered any strong insight on the role of the media in the “social construction of risk”, and it is not by chance that the recognition of the weakness of this area of investigation is what has led risk analysts into engaging with the notion of “trust”: “one of the most likely explanations for the failures of risk communication initiatives is that reactions to risk communication are not only influenced by message content (i.e. what is communicated about risks (and indeed benefits) of particular hazards) but also by trust in those responsible for providing the information” (p. 11). The role played by the structure of communication itself had to be taken into consideration, with different media reaching different people in different ways. 31 Again, however, it appears that in this approach trust is taken into account in the form of the ‘risk to erode trust in decision-makers and experts’, so that the issue of trust appears too quickly short-circuited into the much too immediate policy-making constraint that has so far facilitated the flourishing of risk analysis: “Research indicates that the public believe that governments work closely with industry, and the latter may be seen as possessing a vested interest in putting forward a particular point of view. This in turn causes public distrust in regulation and legislative controls ... Inappropriate handling of crises may lead to further loss of trust in the risk regulators” (p. 12). Since this diagnosis is given theoretical depth by Beck, it is not surprising that the proponents of risk analysis find comfort in the Giddens/Beck approach, and here it is quite apparent that discharging responsibilities on individual reflexivity is a theoretical solution to institutional quandaries: “The individual must [my emphasis] establish his or her social networks and establish trust with the members of these networks, and with a series of experts and policy makers who maintain present-day order” (p. 14). Therefore, the interest in trust slips into studying the “credibility” and “accountability” of policy makers and experts from the viewpoint of the general public. A parallel strand of theoretical discharge sought by the risk analysts is therefore the social psychological literature, especially those works that “have extensively studied the area of persuasion and the impact of information and communication on attitude change”, and in particular the so-called “elaboration likelihood model” (ELM), applied in particular in the US to the Food and Drug Administration and to the Environmental Protection Agency. Frewer and others tend now to criticise the approach, since “source characteristics are seen to align with public perceptions about the hazard, rather than these perceptions being determined by trust in the source itself” (p. 16). At the very least, it seems that risk perception analysis goes full circle in the search for a theoretical fundament (or relief) to its strand of empirical studies, and that the orientation to trust does not provide a stable shore of relief to them. However, they admit that “the issue of trust still merits a great deal of further empirical investigation” and are even ready to hypothesise a positive function of distrust: “ Does distrust lead to better, safer, less hazardous societies?” And, in a final Putnamian twist: “Are societies that are more ‘civic’ also more likely to trust regulators and policy makers?” (p. 19). One might say that the reasoned subordination of issues of risk to issues of trust in the project “trustin-food”, and the attending EU-framework, makes the notion of trust a sort of public good that is by necessity an integral part of the research question. The first problem arises by meeting systems (or rather regimes) that mainly incorporate this public good in a universalistic and highly institutionalised or institutionalisable form, vs. those regimes that relate this public good to local or regional circuits of production and commercialisation of food often also identified as “traditional”. In the universalistic kind of regime, as exemplified by the UK, “the consumer is ascribed both the ability to make informed choices for (usually) him/herself and his/her family and the responsibility for the consequences of those choices ... [T]he role of the government is regulatory and advisory; i.e. largely restricted to provide a pluralistic forum to insure the adequacy of the labelling that is to inform those choices. If ingredients labelling constituted one side of the strategy of disembedded trust, reliance on expert scientific advice constitutes the other.” (Sassatelli and Scott 2001: 220). While such a highly universalistic approach to a mainly disembedded form of trust certainly has deeper roots in the political history and culture of a given society, it is possible to see it also as a response to the high incidence of food scares, ranging, in the British case, from salmonella in eggs to BSE. In other words, the institutionalising, disembedding and universalising type of regime can be interpreted as a response to a generalised sense of a loss of control over the chains of production, distribution, and not least preparation of food, that can no longer be recuperated piecemeal, but through the imposition of higher standards that are mediated between the responsibility of public authorities, those of the producers, and not least those of the consumers. 32 In a certain sense, this is a regime that already presupposes a kind of reflective social agent and consumer-citizen of the kind exemplified by Giddens’ theory of trust and reflexivity. This type of regime seems – so far – not able to integrate or mobilise consumption patterns oriented towards organic products into the model of the distribution of reflective responsibility. However fast-growing – as again in the case of the UK– organic markets are not embedded in a circuit linking the political promotion of specific supply systems and the bottom-up movement of recreating trust relationships on a local basis. The indicator of this lack of embeddedness is in several cases reflected in symbolic references vehicled by the media which are often related to tokens of locality and resistance to homogenisation, as exemplified by the possible – but lacking within this regime – emphasis on flavours and smells, in favour of alternative but universalistic patterns of consumption, identity and advocacy, such as those which refer to vegetarianism and animal rights. These are alternative versions of the same pattern of confidence in expert knowledge (Sassatelli and Scott 2001). The counter-regime of trust stressing the embeddedness of trust within local traditions starts from a model of consumers’ identity emphasising the uniqueness of every locale and the related brands of food, as well as the practices – many of which are of a ritual nature – attending this primacy of the locale. Exemplified by the Italian case, and reflected and promoted by popular TV programs like Linea Verde on RAI-1 and conducted by a popular journalist and media personality (now the director of RAI1), this regime of embeddedness depends on two dimensions of homogenisation and standardisation that are expected to protect and promote the “nature” (in fact the perception) of embeddedness, without contradicting its perceived rooting in local tradition: first, the creation of increasingly EU-recognised local or regional certifications of origin of food produced by farms of a specific locale (but in fact often extending beyond the historical boundaries of a region with which one specific product is identified, as in the case of the consorzio mozzarella campana), and second, the symbolic mediation through a massive, often concerted intervention of national and local media, recreating a “national” approach to food and trust out of bits of regional and local circuits of production, but whose distribution circuits often range well beyond, and in some cases are the source of supply of market niches of high priced, luxury food. Far from being “traditional” (in any of the senses elucidated by Giddens), we see that the model might well cover disparate circuits of distribution and marketing. Furthermore, this type of regime also emphasises efficiency, transparency, and a shared responsibility between producers, consumers, and regulating authorities, and this is what makes it also dependent on a universalistic model of disembedded standards. This tendency is not only promoted by institutions, but it is also furthered by the growing awareness that locality does not always produce embedded trust by default, but sometimes engenders highly criminal forms of law-breaking, as in the recent cases disclosing the existence of butcheries run by mafia-related organisations in some regions of Southern Italy, suspected of illegally commercialising infected beef (cf. also Berg 2000). To summarise, “it is to the regional peasant tradition that Italians are invited to resort for confidence-enhancing strategies. Regions are indeed crucial, both at the level of rhetoric and the level of practices (regulation, production, exchange and consumption): it would be very difficult to understand the Italian reality without taking into account regional variation and, simultaneously, such variation makes Italy into a complex case.” (Sassatelli and Scott 2001: 227). This complexity is a blend of “localised trust”, the crystallisation of EU opportunity structures of labelling and standardisation of local and regional products, and the symbolic and discursive mediation through media orchestration. But one main aspect of how this complex case affects trust relationship is that traceability (furthering trust through transparency as a key resource for the responsible and reflexive consumer) can never be reduced to locality (localised trust based 33 on familiarity with the origin), though this is exactly what the discourse tries to suggest. This reconstructed sense of localised trust substitutes labelling for familiarity. However, the success of labelling depends on representing familiarity. Is the resource of trust (its being a public good) really localised trust as opposed to generalised confidence in the system, or an adequate representation and standardisation of trust that translates well into generalised confidence in the system? Is there any residual (positive) role of “familiarity”, and could this be (remember the definitions – and related distinctions - given by Seligman) a better word for “localised trust”? Is familiarity still available at all in a pure, unmediated form, or only as mediated by big and small media? Certainly, trust is not just the result of a balance of ignorance and information, but depends on communication and representation – which includes the symbolic element of representing closeness, similarity, identity and, of course, familiarity. It would be a dangerous interpretative short-cut to see this counter-model as “a form of backlash to modernity toward traditionalism, whereby society is less complex and trust relations are easy to formulate” (Löfstedt and Frewer 1998: 14), a diagnosis that one could draw from the fact that “risk society” makes government lose control of regulatory structures of risk (Beck 1992 [1986]). More in general, it has been suggested that “analyses stemming from the Beck-Giddens reflexive modernity thesis have largely missed the point. There we find support for a theory of homogenising individualisation ... On the contrary, we have seen that each individual still makes his or her choices in the contexts of regulated markets” (Sassatelli and Scott 2001: 136; cf. also Almas 1999), and this certainly applies to both the UK and the Italian regimes of trust. A one-dimensional view of choice as related to consumer’s active trust and distrust towards food commodities is a hindrance. We need a more plastic view of choice incorporating a richer view of the agent than provided by rational choice theory and cognates, the notion of the reflexive agent, or Luhmann’s subsumption of the agent into the system’s mechanism. Habitus as a notion relating routine and active improvisation is the most interesting of all such notions, in that it also incorporates a dimension of “discipline” of the consumer (cf. Warde 1994). In order to understand trust regimes, we would then also need to relate habitus as rooted in practice to familiarity as increasingly subject to symbolic and discursive mediation (and manipulation). It thus emerges that trust might become an interestingly operative notion only in the reconstructed form of trust regime (for further discussion, see Sassatelli and Scott 2001), within which a whole family of concepts revolving around trust plays a role. Those which we have tentatively put to work in this latter section are but a small sample, that should be enriched with lessons learned not only from some of the most en vogue schools producing theories of trust, but also from some of the “classics”, especially Tönnies and Weber. Whilst we would like to capitalise on Weber’s sociology, we should at the very least be aware: a) of the inherent limits of the legitimacy that the modern institutions of the market and the state might gain for themselves, and that this weakens any discussion of “confidence and trust” as a package to ensure the system’s stability; b) of the modern tendency that however we conceptualise consumers’ “choice”, the issue of “value” cannot be eliminated from it, and the ensuing “polytheism of values” characteristic of modern societies, according to which consumers’ choices will permanently realign and be subject to shifts and sometimes subversions, will also relativise any strategy of research or policy recommendation targeting a package notion of trust-cum-confidence targeting the functionality of the system – or of the food consumption arena – as a whole (cf. Gronow 1997). 34 Conclusion: a few analytical notes for further discussion In Europe, in particular, the voices of experts have increasingly been called upon to counter the erosion of consumers’ confidence following a succession of different food crises. If food scandals have paved the way for a problematisation of food consumption, new technologies (like ‘gene technology’) have been portrayed as confronting the consumer with ‘alien’ foods, even in countries like Italy where food crises had been quickly and quietly absorbed. Furthermore, “European societies are both facing the effects of increasing globalization in the food trade and undergoing a process of harmonisation and integration. Under these conditions the link between everyday practices and community has become increasingly problematic, ‘traditional’ and tacit forms of trust may no longer be sufficient” (Sassatelli and Scott 2001: 225). In this context, as we have suggested, theories of trust might be put to use in order to help define different types of trust regimes or configurations. We may consider that the notion of trust regime is able to account for different combinations of embedded and disembedded forms of trust relations. Certainly, when talking about disembeddedness we shall distance ourselves from the Giddens/Beck approach to individualisation, and take the view that economic action is by definition ‘embedded’ (Granovetter 1985) and that discourses on food consumption and safety are still rooted in social relations (Dreyer 1999). Yet these actions and relations take on very different shapes. The notion of trust regime may thus help us to understand qualitative variations which survey data may struggle with: put simply, to understand how trust varies across different populations we need to have a better sense of how trust is characterised in different countries. Considering the notion of the trust regime as a possible heuristic tool to discover and systematize different institutional configurations which produce different responses in terms of trust, distrust and risk perception, we have firstly to take onboard that it consists of heterogeneous elements. Trust regimes are characterised by specific connections between various actors (producers, distributors, consumers and regulating authorities), institutionalised rules (such as laws and regulations), technological systems (such as farming know-how and packing technology), material culture (such as farming equipment, processing plants, distribution and transport facilities), as well as specific notions and beliefs about nature, food and actors in the food chain. If every society is characterised by relatively stable attitudes towards food, then a trust regime may be considered as the relatively stable and coherent – albeit contested – set of definitions of, and institutional approaches to, the food safety that is typical of a given society. Trust regimes appear to work as a ‘frame’ through which actors in the food system identify, interpret and process new information - which in turn feeds back on the initial ‘frame’ (Goffman 1974; see also Sassatelli and Scott 2001: 238 and ff.) - as well as policy-making. The feedback process is, in its turn, an important object of study. Indeed, it is at this level that, under pressures of different kinds, the institutional conditions of trust may be altered to a critical point. Trust regimes may vary in terms of their capacity to metabolise pressures and of their overall resilience. A comparison of their characteristics should therefore enable us to identify those institutional features which are most conducive to consumer trust in food. This notion might be taken as presupposing a basic analogy of approaches to food safety and underlying arrangements among very different food items, something that we may want to leave to demonstration (especially because we deal with two very specific food items, tomatoes and beef). To address this possibility we shall consider that a trust regime might be characterised by the specific position that each item occupies in the overall food system. 35 The notion of the frame deployed here shall not be considered in exclusively cognitive terms: a frame is a set of dispositions orienting practices – according to Goffman’s definition. This emphasis of frames must be related to a research strategy placing a strong emphasis on specific actors and their roles, as well as on the dimensions of trust which are institutionally supported. Relevant institutional conditions of trust regimes to be studied include the regulatory framework (legislation, organisation, implementation), pressure groups, and the vertical and horizontal structure and distribution of power in the food market. These conditions provide the necessary frame for trust or distrust to develop and it is important to determine the extent to which they impact upon consumers’ perceptions. We thereby need to place emphasis on the role of different actors. Indeed, trust might be placed in: a) governmental authorities and their generalised control functions; b) commercial organisations like manufacturers and supermarkets; c) local authorities and local chains of supply; d) traditional relationships between small businesses and their customers; e) consumer movements; f) food experts and gastronomic manuals. These are not mutually exclusive sources of trust, and a task of the project might be to empirically ground the notion of trust regime considering how different trust regimes might be the result of special combinations of trust-sources. A focus on the role of different agents might appear too static if only ‘who is trusted’ and ‘how much’ are taken into consideration. We need to include some dynamic element, considering the historical role that each actor has played in each country. The negotiation process that takes place between public and private actors at different levels concerning participation and responsibilities in procedures for control, documentation, standard-setting, labelling and communication with the consumers might be an appropriate focus of analysis to provide the model with the necessary dynamic. Also, the responses of both private and public actors to food crises, and the corresponding degrees of openness to market innovation are fundamental. They appear to touch on various dimensions of trust. As we have seen, there are several ways of distinguishing dimensions of trust, which we may call: a) ontological (what kind of trust are we dealing with; here we find (socio-theoretical) distinctions between trust, faith, belief, confidence, etc.); b) phenomenological (how trust is experienced; here we find (socio-psychological) distinctions between cognitive and emotional dimensions of trust, etc.; c) sociological (how trust is socially organised; here we find our distinction between embedded and disembedded trust, generalised or personalised (focalised according to Roniger 1992) trust, etc. While we shall consider (and problematise) both the ontologically and phenomenologically defined dimensions of trust, the sociologically defined dimension of trust is the most relevant for our project (especially for the institutional study). A discussion of the sociologically defined dimension of trust shall of course be preceded by a focus on the extent to which trust is thematised (and how) in relation to food issues in each country (and across different sectors of the population) – i.e. how much can be explained or otherwise in terms of trust. However, it is the sociological dimension which shall take centre stage for the analysis of trust regimes. A first formulation of the articulation of this dimension could be seen as follows: a) political: the extent to which (and how) food issues are treated as a private rather than a public matter; (a special sub-category of this might be, for example, whether the public is conceived as a generalised other or as a community of citizens/consumers developing and putting to work Luhmann/Seligman’s distinction between familiarity and trust); b) organisational: the extent to which trust is more or less generalised and formalised rather than being itself subject to specific groups, preferences or distinctions; (here we may adjust and develop the classics’ insights, as well as Luhmann’s, Roniger’s, and Giddens’ and Beck’s ideas; c) relational: the extent to which (and how) people place trust in different actors; (here the interactionist and ethnomethodological approaches allow for an understanding of the role of negotiation, discourse and self-presentation). 36 Bibliography Aaker, D. & Day, G. (eds.) 1982. Consumerism: Search for the Consumer Interest. New York: Macmillan. Adam, B. 2000. The Temporal Gaze: The Challenge for Social Theory in the Context of GM Food, British Journal of Sociology, 51 (1): 125-42. Alexander, J.C. (ed.) 1998. Real civil societies. London: Sage. Almås, R. 1997. "Food scandals in risk society. Producers, consumers and experts in late modernity," in Social construction of safe food. Health, ethics and safety in late modernity. Workshop report, Vikhammer (Trondheim), April 14-16 1997, R. Almås, (ed.), Trondheim: Centre for Rural Research, pp. 3-16. Almås, R. 1997. Social construction of safe food. Health, ethics and safety in late modernity. Workshop Report, Vikhammer (Trondheim), April 14-16 1997. Trondheim: Centre for Rural Research. Almås, R. 1999. "Food trust, ethics and safety in risk society", Sociological Research Online, 4 (3):1-9; http://www.socresonline.org.uk/socresonline/4/3/almas.html Almås, R. & Nygård, B. 1995. New biotechnologies: attitudes, social movements and regulation. A comparative analyses of a European discourse. Trondheim: Centre for Rural Research. Almond, G. and Verba, S. 1980 (eds.). The civic culture revisited. Boston: Little Brown. Angelillo, I. F., Viggiani, N. M. A., Rizzo, L., & Bianco, A. 2000. "Food handlers and foodborne diseases: Knowledge, attitudes, and reported behavior in Italy", Journal of Food Protection, 63 (3): 381-385. Arrow, K.J. 1972. “Gift and exchanges”, Philosophy and Public Affairs, 1 (4): 343-62. Arrow, K.J. 1974. The limits of organization. New York: Norton. Askegaard, S. & Madsen, T. K. 1995. European food cultures: An exploratory analysis of food related preferences and behaviour in European regions. MAPP Working Paper no 26, Århus. Axelrod, R. 1984. The Evolution of Co-operation. New York: Basic Books. Barling, D. and Lang, T. 1999. European Public Policy on Genetic Modification of Agricultural Products and Food, European Journal of Public Health, 9 (3): 163-5. Barnes, B. 1992. “Status groups and collective action” Sociology, 26 (2): 259-270. Barber, B. 1983. The logic and limits of trust. New Brunswick, New Jersey: Rutgers University Press. Baum, F. 1999. "Food, social capital and public health: exploring the links," in Eating into the Future. The First Australian Conference on Food, Health and the Environment, 11-13 April. Conference Proceedings. Adelaide: Eat Well SA. Beardsworth, A.D. 1990. ‘Trans-science and moral panics: understanding food scares’, British Food Journal, Vol. 92, No. 5, pp 11-16. Beardsworth, A.D. 1995. ‘The Management of Food Ambivalence: Erosion and Reconstruction?’ in Maurer, D. and Sobal, J. (eds) Eating Agendas: Food and Nutrition as Social Problems, New York: Aldine de Gruyter. 37 Beardsworth, A. & Keil, T. 1997. Sociology on the Menu. London: Routledge. Beck, U. 1992 [1986]. Risk society: towards a new modernity. London: Sage. Beck, U. 1994. “The reinvention of politics: towards a theory of reflexive modernization”, in Reflexive modernization. Politics, tradition and aesthetics in the modern social order, U. Beck, A. Giddens, & S. Lash, eds., Polity Press, Cambridge, pp. 1-55. Belasco, W. 1993. Appetite for Change, Ithaca: Cornell University Press. Bell, D. & Valentine, G. 1997. Consuming Geographies. We are where we eat. Routledge, London, New York. Bellah, R. 1991. The good society. New York: Knopf. BEPCAG - Biotechnology and the European Public Concerted Action Group. 1997. Europe is ambivalent about biotechnology, Nature, 387: 845-7. Berg, L. 2000. Trust in Food in Europe. National Institute for Consumer Research (SIFO, Norway), Working Paper n. 15, September. Berg, L. & Kjærnes, U. 1999. 'Risk reducing' or 'complexity reducing' consumers? A Nordic survey of consumer trust and distrust in food. Paper in the ESA subgroup "The Sociology of consumption", 18-21 August. Berg, L. & Kjærnes, U. 2001. "Consumers and food safety policies in the Nordic countries," in Eating patterns. A day in the lives of Nordic peoples. Report No 7, U. Kjærnes, (ed.). Lysaker: The National Institute for Consumer Research, pp. 235-254. Berg, L. 2000a. Naiv tillit? Forbrukernes tillit til mat i Norge, Belgia og England. [Naive trust? Consumer trust in food in Norway, Belgium and the United Kingdom.] Notat No 2. Lysaker: The National Institute for Consumer Research. Berg, L. 2000b. Tillit til mat i kugalskapens tid. En komparativ kartlegging, med fokus på forbrukertillit og matsikkerhet i Norge, England og Belgia. [Trust in Food in the Age of Mad Cow's Disease] SIFO Report no.6. Lysaker: The National Institute for Consumer Research. Berg, L. 2000c. Trust in Food in Europe. focus on Consumer Trust in Norway, England and Belgium. Working Paper No 15. Lysaker: The National Institute for Consumer Research. Bjørkum, E., Lien, M., & Kjærnes, U. 1997, Kjøtt, forbrukeroppfatninger og offentlig debatt. En litteraturstudie. [Meat, consumer opinions and public debate. A literature survey]. SIFO Arbeidsrapport nr.3. Lysaker : Statens institutt for forbruksforskning,. Bonny, S. 2000, "Les consommateurs, l'agriculture, la qualité et la sécurité des aliment: une analyse du questionnement des consommateurs et des réponses apportées", Productions Animales, 13 (5): 287-301. Bourdieu, P. 1977 [1972]. Outline of a theory of practice. Cambridge: Cambridge University Press. Bourdieu, P. 1984 [1979]. Distinction. London: Routledge. Bourdieu, P. 1998 [1994]. Practical reason. On the theory of action. Cambridge : Polity Press. Buttel, F. 1999. Agricultural Biotechnology. Its Recent Evolution and Implications for Agrofood Political Economy, Sociological Research Online, 4 (3): http://www.socresonline.org.uk/socresonline/4/3/buttel.html 38 Bredahl, L., Grünert, K. G., & Frewer, L. 1998. Consumer attitudes and decision-making with regard to genetically engineered food products - a review of the literature and a presentation of models for future research. Working Paper no 52 MAPP. Århus: Center for markedsovervågning, vurdering og -bearbejdning til fødevaresektoren. Bredahl, L. 2001. "Determinants of consumer attitudes and purchase intentions with regard to genetically modified foods - results of a cross-national survey", Journal of Consumer Policy, 24: 23-61. Brewer, M. S., Sprouls, G. K., & Russon, C. 1994. "Consumer attitudes toward food safety issues", Journal of Food Safety, vol. 14, no. 1, pp. 63-76. Keywords: consumers/consumer attitudes/attitudes/food/safety/United States/trust Brillat-Savarin, A. 1822. La physiologie du goùt, It.tr. La fisiologia del gusto, Milano: Bur, 1985. Brom, F.W.A. 2000. Food, Consumer Concerns, and Trust: Food Ethics for a Globalizing Market, Journal of Agricultural and Environmental Ethics, 12 (2): 127-39. Bruhn, C. M. 1997. "Consumer concerns: Motivating to action", Emerging Infectious Diseases, 3 (4): 511-515. Brunsø, K. & Grünert, K. G. 1998. "Cross-cultural similarities and differences in shopping for food", Journal of Business Research, 42 (2): 145-150. Buzby, J. C. & Ready, R. C. 1996. "Do consumers trust food safety information?", Food Review,19 (1): 46-49. Cahill, J. 1998. "Food safety control", Consumer Choice, 22: 324-327. CCE 1994. La politica del consumatore nella Comunità Europea, Regione Toscana, Comitato Regionale Consumatori e Utenti, Firenze. Christoffersen, S. A. 2000. "Tillit og konvensjon i K.E.Løgstrups etikk [Trust and convention in the ethics of K.E. Løgstrup.]", Sosiologi i dag, 30 (3): 49-66. Cole, R. L. 1973. "Toward a model of political trust: A causal analysis", American Journal of Political Science, 17: 809-817. Coleman, J. S. 1988. “Social Capital in the Creation of Human Capital” American Journal of Sociology (supplement) 94: 95-120. Coleman, J. C. 1990. Foundations of social theory. Cambridge MA: The Belknap Press of Harvard University. Coleman, J.S. 1993. “The Rational Reconstruction of Society” American Sociological Review 58: 1-15. Collins, R. 1992. “The romanticism of agency/structure versus the analysis of micro/macro” Current Sociology, 40: 77-97. Collins, R. 1993. “Heroizing and deheroizing Weber” Theory and Society, 22: 861-70. Commandeur, P. et al. 1996. Public Debate and Regulation of Biotechnology in Europe, Monitor, 26, March: http://www.pscw.uva.nl/monitor/2601.htm Consumers Union 28/12/99 Top Ten Consumer Needs for the Year 2000, http://www.consumersunion.org/finance/toptendc100.htm 39 Consumers Union 6/11/99 Summary of Public Opinion Surveys Related to Labelling of Genetically Engineered Foods, http://www.consumersunion.org/food/summpollny699.htm Consumers Union, April 1998 Why We Need Labelling of Genetically Engineered Food, http://www.consumersunion.org/food/whywenny798.htm Csyzewski, M. 1994. "Reflexivity of Actors versus Reflexivity of Accounts", Theory, Culture and Society, 11: 161-8. Danielsen, K. & Serck-Hansen, C. 2001. "Om salg, tillit og emosjonelt arbeid i forsikring. [On marketing, trust and meotional work in insurance.]", Sosiologisk tidsskrift, vol. 9, no. 4, pp. 328-349. Dasgupta, P. 1988. "Trust as commodity," in Trust: making and breaking cooperative relations, D. Gambetta, (ed.). New York and Oxford: Basil Blackwell, pp. 49-72. Davies, S. & Todd, S. 1996. The need for an independent food agency. Consumer Policy Review 6 (3): 82-86. Douglas, M. 1966. Purity and Danger. London: Allen and Unwin. Douglas, M. 1992. Risk and Blame: Essays in Cultural Theory. London: Routledge. Douglas, M. 1996. Thought styles. Critical essays on good taste. London: Sage. Douglas, M. & Wildavsky, A. 1982. Risk and culture. An essay on the selection of technological and environmental dangers. Berkeley: University of California Press. Douglas, M. & Isherwood, B. 1978. The World of Goods. London: Routledge. Dreyer, M. 1999 “Biotechnological communication and he socio-cultural embededdness of Economic Actors”, in O’Mahony, cit. Dunn, J. 1984. “The concept of ‘trust’ in the politics of John Locke” in Philosophy in history R. Rorty et al. (eds.). Cambridge: Cambridge University Press, pp. 279-301. Dunn, J. 1988. “Trust and Political Agency” in Trust: making and breaking cooperative relations, D. Gambetta, (ed.). New York and Oxford: Basil Blackwell, pp. 73-93. Dulsrud, A. 1998. Trust in asymmetrical dyadic relationships. Paper to the EGOS 14th Colloquium. 'Stretching the boundaries of organisation studies into the next millenium. Subtheme 13: Trust and Control in Organisational Relations. Durant, J. et al. (eds) 1998. Biotechnology in the Public Sphere. London: Science Museum. Durkheim, E. 1984 [1893]. The division of labor in society. New York: The Free Press. Durkheim, E. 1995 [1912]. The Elementary Forms of Religious Life. London: Routledge. Earle, T. & Cvetkovich, G.T. 1995. Social trust: toward a cosmopolitan society. New York: Prager Eisenstadt, S. N. 1995, Power, trust, and meaning. Essays on sociological theory and analysis. Chicago & London: The University of Chicago Press. Elster, J. 1989, The cement of society. Cambridge: Cambridge University Press. Elster, J. & Moene, K. (eds.). 1988. Alternatives to capitalism. Cambridge: Cambridge University Press. 40 Esping-Andersen, G. 1999. Social foundations of postindustrial economies. Oxford: Oxford University Press. Esping-Andersen, G. 1990. The three worlds of welfare capitalism. Cambridge: Polity Press. Fine, B. & Leopold, E. 1993. The World of Consumption. London: Routledge. Fischler, C. 1988. “Food, Self and Identity”, Social Science Information, 27 (2): 275-92. Fisman, R. & Khanna, T. 1999. "Is trust a historical residue? Information flows and trust levels", Journal of Economic Behavior & Organization, 38: 79-92. Flynn, A., Marsden, T., & Harrison, M. 1999. "The regulation of food in Britain in the 1990s", Policy and Politics, 27 (4): 435-447. Foucault, M. 1978. 'Governmentality' in G. Burchell and C. Gordon, (eds.) (1991) The Foucault Effect: Studies in Governmentality. Hempstead: Simon and Schuster. Frankel, S. H. 1977. Money: Two Philosphies. The Conflict of Trust and Authority. Oxford: Basil Blackwell. Frewer, L. 1998. "Consumer perceptions and novel food acceptance", Outlook on Agriculture, 27 (3): 153-156. Frewer, L. 1998, "Public perceptions of genetically modified foods", Pesticide Outlook, October:11-15. Frewer, L. 1999. "Risk perception, social trust, and public participation in strategic decision making: implications for emerging technologies", Ambio, 28 (6): 569-574. Frewer, L.J. et al. 1994. Biotechnology and Food Production: Knowledge and Perceived Risk, British Food Journal, 96 (9): 26-32. Frewer, L. J., Howard, C., & Shepherd, R. 1997. "Public concerns in the United Kingdom about general and specific applications of genetic engineering: risk, benefit, and ethics", Science, Technology and Human Values, 22 (1): 98-124. Frewer, L. J., Howard, C., Hedderley, D., & Shepherd, R. 1996. "What determines trust in information about food-related risks? Underlying psychological constructs", Risk analysis, 16 (4): 473-485. Frewer, L. J., Howard, C., Hedderley, D., & Shepherd, R. 1997. "Consumer attitudes towards different food-processing technologies used in cheese production - the influence of consumer benefit", Food Quality and Preference, 8 (4): 271-280. Frewer, L. J., Howard, C., Hedderley, D., & Shepherd, R. 1997. "The elaboration likelihood model and communication about food risks", Risk analysis, 17 (6): 759-770. Frewer, L. J., Howard, C., Hedderley, D., & Shepherd, R. 1998. "Methodological approaches to assessing risk perceptions associated with food-related hazards", Risk analysis, 18 (1): 95-102. Frewer, L. J., Howard, C., Hedderley, D., & Shepherd, R. 1999. "Reactions to information about genetic engineering: impact of source characteristics, perceived personal relevance, and persuasiveness", Public Understanding of Science, 8 (1): 35-50. Fuchs, D. & Klingemann, H.-D. 1995. "Citizens and the state: A changing relationship?," in Citizens and the state, H.-D. Klingemann & D. Fuchs, (eds.). Oxford: Oxford University Press, pp. 1-24. Fukuyama, F. 1995. Trust. The social virtues and the creation of prosperity. London: Penguin Books. 41 Gambetta, D. 1988a. "Can we trust trust?," in Trust. Making and breaking cooperative relations, D. Gambetta, (ed.). New York, Oxford: Basil Blackwell, pp. 213-237. Gambetta, D. 1988b. “Mafia: the price of distrust”, in Trust. Making and breaking cooperative relations, D. Gambetta, (ed.). New York, Oxford: Basil Blackwell, pp. 158-175. Garfinkel, H. 1963. A conception of, and experiments with 'trust' as a condition of stable, concerted actions, in O.J. Harvey (ed) Motivation and Social Interaction. New York: The Ronald Press, pp. 187-238. Giddens, A. 1990. The consequences of modernity. Oxford: Polity Press. Giddens, A. 1991. Modernity and self-identity. Self and society in the late modern age. Cambridge: Polity Press. Giddens, A. 1994a. “Living in a post-traditional society”, in Reflexive modernization. Politics, tradition and aestethics in the modern social order, U. Beck, A. Giddens, & S. Lash, (eds.). Cambridge: Polity Press, pp. 56-109. Giddens, A. 1994b. "Replies and critiques. Risk, trust, reflexivity," in Reflexive modernization. Politics, tradition and aestethics in the modern social order, U. Beck, A. Giddens, & S. Lash, (eds.). Cambridge: Polity Press, pp. 184-197. Gilg, A. W. & Battershill, M. 1998. "Quality farm food in Europe: a possible alternative to the industrialised food market and to current agri-environmental policies: lessons from France", Food Policy, 23 (1): 25-40. Goffman, E. 1970. Strategic interaction. Oxford: Basil Blackwell. Goffman, E. 1972. Interaction rituals. Essays on face to face behaviour. Harmondsworth: Penguin Books. Goffman, E. 1974. Frame Analysis. New York: Harper and Row. Goncalves, M. E. 2000. "The importance of being European: The science and politics of BSE in Portugal", Science, Technology and Human Values, 25 (4): 417-448. Good, D. 1988. “Individuals, interpersonal relations, and trust”, in Trust. Making and breaking cooperative relations, D. Gambetta, (ed.). New York, Oxford: Basil Blackwell, pp. 31-48. Goyens, M. 1992. 'Consumer protection in a single European market: What challenge for the European agenda', Common Market Law Review, 29: 71-92. Granovetter, M. 1985. Economic Action and Social Structure: The Problem of Embeddedness, American Journal of Sociology, 91: 481-510. Grant, W. 1992. Models of interest intermediation and policy formation applied to an internationally comparative study of the dairy industry. European Journal of Political Research 21 (1-2): 5368. Grimen, H. 1996. Tillit og rutinar. [Trust and routines.] Notat 9618 LOS-senteret. Bergen: University of Bergen. Grobe, D., Douthitt, R., & Zepeda, L. 1999, "A model of consumers' risk perceptions toward recombinant bovine growth hormon (rbGH): The impact of risk characteristics", Risk analysis, 19 (4): 661-674. 42 Gronow, J. 1997. The risks and dangers of eating: trust and legitimacy. Paper presented at the 2nd Conference of the European Association Research Group 'Sociology of Consumption', Essex, UK, 27-29 August 1997. Gronow, J. and Warde, A. (eds). 2001. Ordinary Consumption. London: Harwood. Groth II, E. 1991. Communicating with consumers about food safety and risk issues. Food Technology , May: 248-255. Grünert, K. G., Baadsgaard, A., Larsen, H. H., & Madsen, T. K. 1996. Market orientation in food and agriculture. Boston, Dordrect, London: Kluwer Academic Publishers. Gulbrandsen, T. 2001. "Makt og tillit. [Power and trust]", Sosiologisk tidsskrift, 9 (4): 297-327. Gulbrandsen, T. 2000. "Om tillit. [On trust]", Sosiologi i dag, 30 (3): 67-95. Gumbrecht, H. U. 2001. "How is our future contingent? Reading Luhmann against Luhmann", Theory, Culture & Society, 18 (1): 49-58. Guseva, A. and Rona-Tas, A. 2001. 'Uncertainty, risk and trust: Russian and Amarican credit card markets compared', American Sociological Review, 66: 623-646 Habermas, J. 1984 [1980]. The theory of communicative action. Boston: Beacon Press. Habermas, J. 1998. 'Die postnationale Konstellation und die Zukunft der Demokratie' in Die postnationale Konstellation. Frankfurt a.M: Suhrkamp. Halvorsen, T. 1995. Profesjonar, makt og tillit. [Professions, power and trust.] Notat nr.25 Institutt for administrasjon og organisasjonsvitenskap. Bergen: Universitetet i Bergen. Henson, S. 1997. Changes in food consumption patterns in Europe: are we moving to a common European diet? Paper presented at 'Social construction of safe food. Health, ethics and safety in late modernity.' International Workshop, Trondheim April 14-16, 1997. Henson, S. 1996. "Consumer willingness to pay for reductions in the risk of food poisoning in the UK", Journal of Agricultural Economics, 47 (3): 403-420. Henson, S. & Caswell, J. 1999. "Food safety regulation: an overview of contemporary issues", Food Policy, 24 (6): 589-604. Herrmann, R. O., Sterngold, A., & Warland, R. H. 1998. "Comparing alternative question forms for assessing consumer concerns", Journal of Consumer Affairs, 32 (1): 13-29. Herrmann, R. O. 1996. "Misinterpreting expressions of consumer concern about food safety", Consumer Interests Annual, 42: 311-314. Hirsch, F. 1977. Social limits to growth. London: Routledge. Herrmann, R. O., Warland, R. H., & Sterngold, A. 1997. "Who reacts to food safety scares?: Examining the alar crisis", Agribusiness, 13 (5): 511-520. Hirschman, A. O. 1970. Exit, voice and loyalty. Responses to decline in firms, organizations and states. Cambridge, MA: Harvard University Press, Hirschman, A. O. 1984. "Against Parsimony: Three Easy Ways of Complicating some Categories of Economic Discourse", American Economic Review, 74: 88-96. Hisman, R. & Khanna, T. 1999. "Is trust a historical residue? Information flows and trust levels", Journal of Economic Behaviour & Organization, 38: 79-92. 43 Hoban, T.J. 1995. “The Construction of Food Biotechnology as a Social Issue” in D. Maurer and J. Sobel (eds.) cit. Hobbes, T. 1750 [1640]. “Human Nature, or the Fundamental Elements of Policy”, in The English works of Thomas Hobbes. Aalen: Scientia Verlag. Hollis, M. 1998. Trust Within Reason, Cambridge: Cambridge University Press. Inglehart, R. 1990. Cultural shift in advanced industrial society. Princeton, New Jersey: Princeton University Press,. Inglehart, R. 1997. "Postmaterialist values and the erosion of institutional authority," in Why people don't trust government, J. S. Jr. Nye, P. D. Zelikow, & D. C. King, (eds.). Cambridge MA, London: Harvard University Press, pp. 217-236. Ingold, T. 1994. "From trust to domination. An alternative history of human-animal relations," in Animals and human society. Changing perspectives, A. Manning & J. Serpell, (eds.). London, New York: Routledge, pp. 1-22. Jacobsen, E. 1999. "Mangfoldet er uroens mor". Endrede institusjonelle forutsetninger for tillit i verdikjeden for matvarer. ["Diversification is the source of unrest". Shifting institutional conditions for trust in the food chain.] Arbeidsnotat nr.11. Lysaker: The National Institute for Consumer Research. James, A. 1993. “Eating Green(s): Discourses of Organic Food”, in K. Milton (ed) Environmentalism. London: Routledge. Johnson, J. C. & Griffith, D. C. 1996. "Pollution, food safety and the distribution of knowledge", Human Ecology, 24 (1): 87-108. Jukes, D. 2000. "The role of science in international food standards", Food Control, 11 (3): 181-194. Jussaume, R. A. & Higgins, L. 1998. "Attitudes towards food safety and the environment: A comparison of consumers in Japan and the U.S.", Rural Sociology, 63 (3): 394-411. Kaase, M. 1999. "Interpersonal trust, political trust and non-institutionalised political participation in Western Europe", West European Politics, 22 (3): 1-21. Kees, B. & Phillips, B. 1994. Nothing sacred. Journalism, politics and public trust in a tell-all age. Nashville: The Freedom Forum First Amendment Center at Vanderbilt University. Kjærnes, U. & Dulsrud, A. 1998. Consumption and mechanisms of trust. Presented at The ESA subgroup 'The Sociology of Consumption', University of Milan, Sept 16-17, 1998. Kjærnes, U. 1999a. "Food risks and trust relations", Sosiologisk tidsskrift, 7 (4): 265-284. Kjærnes, U. 1999b. "Two parallel, but very different Norwegian debates: the cases of BSE and scrapie," in Food Quality: Regional and National Aspects in Contemporary Europe, S. Jansson & A. Murcott, (eds.). Stockholm: Södertörns Högskola. Kjærnes, U. 2001. "A Nordic survey of consumer trust and distrust in food," in Sociology of consumption, L. Voltchkova, J. Gronow, & V. Minina, (eds.). St. Petersburg: The Library of Contemporary Sociology of Journal of Sociology and Social Anthropology, pp. 193-215. Kymlicka, W. 1989. Liberalism, Community and Culture. Oxford: Oxford University Press. 44 Lacy, W. B., Busch, L., & Lacy, L. R. 1991. "Public perceptions of agricultural biotechnology," in Agricultural biotechnology. Issues and choices, B. R. Baumgardt & M. A. Marshall, (eds.). West Lafayette: Purdue University. Lagerspetz, O. 1996. The tacit demand. A study of trust. Åbo: Filosofiska Institutionen. Lagerspetz, O. 1998. Trust: The Tacit Demand. Dodrecht: Kluwer Academic Publishers. Lanska, D. J. 1998. "The Mad Cow problem in the UK: Risk perceptions, risk management, and health policy development", Journal of Public Health Policy, 19 (2): 160-183. Latouche, K., Rainelli, P., & Vermersch, D. 1998. "Food safety issues and the BSE scare: some lessons from the French case", Food Policy, 23 (5): 347-356. Latouche, K. et al. 1998. Food Safety Issues and the BSE Scare: Some Lessons from the French Case, Food Policy, 23 (5): 347-56. Leidwein, A. 1996. 'Österreiche Agrapolitik nach dem Beitritt zur EU' in A. Khol, G. Ofner and A. Stirnmann (eds.) Österreichisches Jarhbuch für Politik 1995, Vienna: Verlag für Geschichte und Politik. Levidow, L. et al. 2000. Genetically Modified Crops in the European Union, Journal of Risk Research, 3 (3): 189-208. Levidow, L. and Carr, S. 2000. UK: Precautionary Commercialization?, Journal of Risk Research, 3 (3): 261-70. Lewis, J.D. & Weigert, A. 1985. “Trust as a social reality” Social Forces, 63: 967-85. Lien, M. 1997. "Nation, technology and nature. Constructions of safe food in the market," in Social construction of safe food. Health, ethics and safety in late modernity. Report no.5, R. Almås, (ed.), Trondheim: Centre for Rural Research, pp. 39-52. Light, D. W. & Schuller, A. 1986. Political values and health care: The German experience. Cambridge, Massachusetts: The MIT Press. Listhaug, O. & Wiberg, M. 1995. "Confidence in political and private institutions," in Citizens and the state, H.-D. Klingemann & D. Fuchs, eds., Oxford: Oxford University Press, pp. 298-322. Liu, S., Huang, J.-C., & Brown, G. L. 1998. "Information and risk perception: A dynamic adjustment process", Risk analysis, 18 (6): 689-699. Löfstedt, R.E. & Frewer, L. 1998. “Introduction”, in The Earthscan Reader in Risk and Modern Society, London: Earthscan Publications, pp. 3-27. Lohmann, U. 1986. "Structure and values of state and society," in Political values and health care: The German experience, D. W. Light & A. Schuller, (eds.), Cambridge, Massachusetts: The MIT Press, pp. 27-52. Lovell, D. W. 2001. "Trust and the politics of postcommunism", Communist and Post-Communist Studies, 34 (1): 27-38. Luhmann, N. 1988. "Familiarity, confidence, trust: problems and alternatives," in Trust. Making and Breaking Cooperative Relations, D. Gambetta, (ed.), New York: Basil Blackwell, pp. 94-108. Luhmann, N. 1979 [1968/1973]. Trust and power: two works by Niklas Luhmann (with Introduction by Gianfranco Poggi). Chichester: John Wiley & Sons. 45 MacKenzie, D. R. 1987. "Risk and the public's acceptance of genetic engineering," in Public perceptions of biotechnology, L. R. Batra & W. Klassen, (eds.), Bethesda: Agricultural Research Institute. Mahon, B. E. 1999. "Consequences in Georgia of a nationwide outbreak of Salmonella infections: What you don't know might hurt you", American Journal of Public Health, 89 (1): 31-35. Maney, A. & Plutzer, E. 1996. "Scientific information, elite attitudes and the public debate over food safety", Policy Studies Journal, 24 (1): 42-56. Marsden, T., Flynn, A., & Harrison, M. 2000. Consuming interests. The social provision of foods. London: UCL Press. Marris, C. & Joly, P.B. 1999.”La gouvernance technocratique par consultation? » in «Risque et démocratie», Les Cahiers de la sécurité intérieure, 38: 97,124. Maurer, D. & Sobel, S. (eds.) 1995. Eating Agendas. New York: Alatine de Gruyter. Mauss, M. 1970. The gift: forms and functions of exchange in archaic societies. London: Cohen and West. McDonald, A.-L. & Langford, I. 1998. Influences on behaviour towards food-related health risks: a pilot study. CSERGE Working Paper PA 98-03 Centre for Social and Economic Research on the Global Environment, University of East Anglia and University College London, London. McIntosh, W. A., Acuff, G. R., Christensen, L. B., & Hale, D. 1994. "Public perceptions of food safety", Social Science Journal , 31 (3): 285-292. McIntosh, W. A. 1996. Sociologies of food and nutrition. London, New York: Plenum Press. McKee, M. 1999. "'Trust me, I'm an expert.' Why expert advisory groups must change (editorial)", European Journal of Public Health, 9 (3): 161-163. Mennell, S. 1985. All Manners of Food: Eating and Taste in England and France from the Middle Ages to the Present. Oxford: Blackwell. Mepham, B. (ed.) 1996. Food ethics. Professional ethics.London, New York: Routledge. Mepham, B. 1996. Ethical Analysis of Food Biotechnologies, in B. Mepham (ed.) Food Ethics, Professional ethics. London: Routledge. Micklitz, H.-W. 1992. "The international dimension of product safety," in Product safety and control processes in the European Community, M. Fallon & F. Maniet, (eds.), Brussels: Centre de Droit de la Consommation. Miller, A. H. & Listhaug, O. 1998. "Policy preferences and political distrust: A comparison of Norway, Sweden and the United States", Scandinavian Political Studies, 21 (2): 161-187. Miller, D. & Reilly, J. 1995. "Making an issue of food safety: The media, pressure groups, and the public sphere," in Eating agendas. Food and nutrition as social problems, D. Maurer & J. Sobal, (eds.), New York: Aldine de Gruyter, pp. 305-336. Misztal, B.A. 1996. Trust in modern societies: the search for the bases of social order. Cambridge: Polity Press. Mitchell, V.W. and Greatorex, M. 1990. Consumer perceived risk in the UK Food Market, British Food Journal, 92 (2): 16-22. Mjøset, L. 2000. "Stein Rokkan's thick comparisons", Acta Sociologica, 43: 381-397. 46 Molm, L. D., Takahashi, N., & Peterson, G. 2000. "Risk and trust in social exchange: An experimental test of a classical proposition", American Journal of Sociology, 105 (5): 1396-1427. Mortensen, N. 1999. "Introduktion. [Introduction]," in Tillid - en mekanisme til reduktion af social kompleksitet [Trust - a mechanism for reducing social complexity], N. Luhmann, (ed.). Copenhagen: Hans Reitzels Forlag, pp. 7-27. Mossel, D. A. A., Weenk, G. H., Morris, G. P., & Struijk, C. B. 1998. "Identification, assessment and management of food-related microbiological hazards: historical, fundamental and psycho-social essentials", International Journal of Food Microbiology, 40: 211-243. Murcott, A. 1986. "We eat what we are: beliefs about food and eating", Journal of Social Medicine, 23: 79-84. Nerlich, B. et als 1999. The Influence of Popular Cultural Imagery on Public Attitudes Towards Cloning, Sociological Research Online, 4 (3). Nottingham, S. 1998. Eat Your Genes. London: Zed Books. Nygård, B. & Storstad, O. 1998. "De-globalization of food markets? Consumer perceptions of safe food: the case of Norway", Sociologia Ruralis, 38 (1): 35-53. Offe, C. 1996. Trust and knowledge, rules and decisions., Washington DC: Georgetown University Press. O’Mahony, P. (ed.) 1999. Nature, Risk and Responsibility. Discourses on Biotechnology. London: Macmillan. Ostrom, E. 1990. Governing commons: the evolution of institutions for collective action. Cambridge: Cambridge University Press,. Palmer, C. 1998. From Theory to Practice. Experiencing the Nation in Everyday Life, Journal of Material Culture, 3 (2): 175-200. Pellizzoni, L. 2001. 'Democracy and the governance of uncertainty. The case of agricultural gene technologies' Journal of Hazardous Materials 86: 205-222. Pellizzoni, L. & Ungaro, D. 2000. "Technological risk, participation and deliberation. Some results from three Italian case studies", Journal of Hazardous Materials, 78: 261-280. Potter, J. 1988. "Consumerism and the public sector: how does the coat fit?", Public Administration, 66: 149-65. Powell, D. A. 2000. "Food safety and the consumer - perils of poor risk communication", Canadian Journal of Animal Science, 80 (3): 393-404. Przeworski, A. & Teune, H. 1970. The logic of comparative social inquiry. New York: John Wiley. Putnam, R. D. 1993. Making democracy work. Civic traditions in modern Italy. Princeton, New Jersey: Princeton University Press,. Reilly, J. & Miller, D. 1997. "Scaremonger or scapegoat? The role of the media in the emergence of food as a social issue," in Food, health and identity, P. Caplan, (ed.). London, New York: Routledge, pp. 234-251. Rexha, N. & Miyamoto, T. 2000. "Relationship-building su.pplier behaviors for winning customer trust. New insights from case research with nine leading Japanese manufacturers," in Developments 47 in Australasian marketing, R. B. McNaughton, (ed.). Stamford, Connecticut: JAI Press, pp. 7794. Roniger, L. 1992. La fiducia nelle societa’ contemporanee. Un approccio comparativo. Messina: Rubbettino. Rousseau, D. M. 1998, "Roderick M. Kramer and Tom R. Tyler, eds.: Trust in organizations: Frontiers of theory and research", Administrative Science Quarterly, 43 (1): 186-188. Ruckenbauer, P. 1998. 'Gentechnik -- Glaubenkrieg oder Zukunftschange?', in A. Kohl, G. Ofner and A. Stirnemann (eds.) Österreichisches Jarhbuch für Politik 1997. Vienna: Verlag für Geschichte und Politik. Sassatelli, R. 1995. Power Balances in the Consumption Sphere: Reconsidering Consumer Protection Organizations, European University Institute Working Paper SPS, n.95/5. Sassatelli, R. 2001a. "Tamed Hedonism: Choice, Routines and Deviant Pleasures", in J. Gronow and A. Warde (eds.) Ordinary Consumption. London: Routledge. Sassatelli, R. 2001b. Trust, Choice and Routines: Putting the Consumer on Trial in Critical Review of International Social and Political Philosophy, 4, 4, 84-105. Sassatelli, R. 2004a. Consumo, cultura e società, Bologna: il Mulino(forthcoming in English as Consumer Culture. History, Theory and Politics, London: Sage). Sassatelli, R. 2004b.“The Political Morality of Food. Discourses, contestation and alternative consumption” in M. Harvey, A. MeMeckin, e A.Warde (a cura di) Qualities of Food. Alternative Theoretical and Empirical Approaches, Manchester: Manchester University Press. Sassatelli, R. and Scott, A. 2001. "Wider Markets, Novel Food and Trust Regimes", European Societies, 3 (2): 231-63. Scroggins, C. D. 1993. "Consumer confidence in food safety - a matter of perspective", Food, Drug, Cosmetic Law Journal, 48 (3): 463-472. Seligman, A. B. 1992. The idea of civil society. New York: Free Press. Seligman, A. B. 1997. The problem of trust. Princeton: Princeton University Press. Sellerberg, A.-M. 1991. "In food we trust? Vitally necessary confidence - and unfamiliar ways of attaining it," in Palatable worlds. Sociocultural food studies, E. L. Fürst et al., (eds.). Oslo: Solum forlag, pp. 193-202. Shapiro, S. P. 1987. "The social control of impersonal trust", American Journal of Sociology, 93 (3): 623-658. Shaw, A. 1999 What are they doing to our Food? Public Concerns about food in the UK, Sociological Research Online, 4 (3): http://www.socresonline.org.uk/socresonline/4/3/shaw.html Silver, A. 1997. “Two different sorts of commerce”, in Public and Private in thought and in practice: perspectives on a grand dichotomy, J. Weintraub and K. Kumar (eds.). Chicago: University of Chicago Press. Simmel, G. 1978 [1900/1907]. Philosophy of Money, 2nd enlarged edition. London: Routledge. Simmons, P. 1997. The Construction of Environmental Risks, in P. Sulkunen et al. (eds.) Constructing the New Consumer Society. London: Macmillan. 48 Smith, M. E. & Kim, C.S. 1998. Managing or Mis-Managing the Genetic Revolution? International Biotechnology Policy and the Scientific Community, paper presented at the Annual Meeting of the American Political Science Association, Boston. Szerszynki, B. 1999. “Risk and Trust”, Environmental Values, 8: 239-52. Siegrist, M. 2000, "The influence of trust and perceptions of risks and benefits on the acceptance og gene technology", Risk analysis, 20 (2): 195-203. Silverglade, B. A. 1998. "The impact of international trade agreements on U.S. food safety and labeling standards", Food And Drug Law Journal, 53 (3): 537-541. Sjöberg, L. 2001. "Limits of knowledge and the limited importance of trust", Risk analysis, 21 (1): 189198. Smith, A. 1759. The theory of moral sentiments. London: Penguin (2000). Smith, M. J. 1991. "From policy community to issue network: Salmonella in eggs and the new politics of food.", Public Administration, 69: 235-255. Smith, N. C. 1990. Morality and the market. Consumer pressure for corporate accountability, London, New York: Routledge. Soper, J. C. 1994. Political structures and interest group activism: A comparison of the British and American pro-life movements. Social Science Journal 31 (3): 319-334. Stirling, A. 1998. “Risk at a turning point?”, Journal of Risk Research, 1 (2): 97-109. Swidler, A. 1986. "Culture in action: symbols and strategies", American Sociological Review, 51 (2): 273-286. Szerszynski, B. 1999. "Risk and trust: the performative dimension", Environmental Values, 8 (2): 239252. Sztompka, P. 1997. Trust, distrust and the paradox of democracy. Paper P97-003 Berlin: Wissenschaftzentrum Berlin für Sozialwissenschaft. Sztompka, P. 1999. Trust. A sociological theory. Cambridge: Cambridge University Press. Tacke, V. 2001. "BSE as an organizational construction: a case study on the globalization of risk", British Journal of Sociology, 52 (2): 293-312. Tocqueville, A. de. 1945. Democracy in America, Vols. I and 2. New York: Knopf. Tönnies, F. 2001 [1887]. Community and Civil Society, (ed.) J. Harris. Cambridge: Cambridge University Press. Togersen, H. and Seifert, F. 2000. Austria: Preacautionary Blockage of Agricultural Biotechnology, Journal of Risk Research, 3 (3): 209-17. Ungar, S. 2001. "Moral panic versus the risk society: the implications of the changing sites of social anxiety", British Journal of Sociology, 52 (2): 271-291. Uslaner, E. M. 1999. "Trust but verify: social capital and moral behavior", Social Science Information, 38 (1): 29-55. Uslaner, E. M. 2000. "Producing and consuming trust", Political Science Quarterly, 115 (4): 569-590. 49 Walgenbach, P. 1998. The production of distrust by a means of producing trust. Presented at the 14th EGOS Colloquium "Stretching the Boundaries of Organisation Studies into the next Millenium", July 9-11, 1998, Maastricht. Walzer, M. 1983. Spheres of Justice. New York: Basic Books. Wandel, M. 1994. "Understanding consumer concern about food-related health risks", British Food Journal, 96 (7): 35-40. Warde, A. 1994. “Consumption, identity-formation and uncertianty”, Sociology, 28 (4): 877-898. Warde, A. 1997. Consumption, Food and Taste. London: Sage. Wildavsky, A. 1994. "Why Self-interest means less outside of Social Context", Journal of Theoretical Politics, 62: 131-59. Wildawsky, A. and Dake, K. 1990. “Theories of Risk Perception: Who Fears What and Why?”, Daedalus, 119 (4): 41-60. Williams, B. 1988. “Formal structures and social reality”, in in Trust: making and breaking cooperative relations, D. Gambetta, (ed.). New York and Oxford: Basil Blackwell, pp. 3-13. 50