Trust: Condition of action or condition of appraisal

Introduction

Discussion in the sociological literature tends to regard the significance of trust for understanding social relationships as perennial and assumes that trust is core in the formation of social relations and institutions. These ideas are often expressed through reference to Durkheim’s argument concerning the non-contractual basis of contract in trust (Durkheim, 2014: 158–168) and Simmel’s frequently quoted statement that ‘Without the general trust that people have in each other, society itself would disintegrate’ (Simmel, 2004: 178–179). It is less frequently appreciated, however, that whereas current sociological discussion is focused on inter-personal or social trust, Durkheim and Simmel refer to another sense of the term. For Durkheim, trust is a quality of organic solidarity and exists in the form of ‘purely moral rules’ that operate as a ‘network of obligation’ (Durkheim, 2014: 177–178) that is quite unlike the trust that an individual voluntarily accords to another. The context of Simmel’s statement, quoted above, is a discussion of the relationship between persons and money as a social artefact; the relevant sentence claims that the social effectiveness of money cannot be based on ‘rational proof or personal observation’ but rather must be founded on ‘trust’.

At the time Durkheim and Simmel wrote, at the end of the 19th and beginning of the 20th century, ‘trust’ typically referred to a form of agency relation, as when property is held in trust or an orphan is committed to trust. Indeed, the relationship indicated by Simmel would be better translated as ‘confidence’. In drawing on these and similar sources writers today confuse earlier notions of trust with current understandings, as outlined in the following section. This will be followed by a critical examination of the argument of embedded trust which shows, first, that contraventions of trust are not necessarily communicated to others, and second, broken trust does not necessarily put affected relationships at risk. Then follows exposition of the case of Chinese guanxi (social connections) in which there is high social solidarity in the absence of trust. Through these arguments and this case, the supposed significance of trust for social relationships is shown to be exaggerated. The research question pursued here, then, asks what role does trust play in inter-personal relations and in what institutional framework does trust operate. In particular, it will be shown that under conditions of precarious institutional maturity trust is not instrumental in realizing or maintaining social interaction or relationships, as typically assumed, but is principally an appraisal term of ego regarding alter.

Forms of trust

The concept of trust in the broadest sense has a long history, reflected in the ancient Greek words ‘pistis’ and ‘apista’ and the Latin words ‘fides’, ‘fido’ and ‘fiducia’ (Morgner, 2013: 511–515). Etymologically, the word ‘trust’ derives from the Old Norse term ‘traust’, meaning reliance, giving, to comfort (Weekly, 1921). This sense, arguably summarized in the notion of trust as a covenant, continues in 17th-century English, indicated as entreaties in the King James Bible to trust in God and to avoid trusting riches (Proverbs 11:28; Mark 10:24), Egyptian chariots (Isaiah 31:1; 36:9), (mortal) men (Jeremiah 17:5; 2 Corinthians 1:9) or one’s own beauty (Ezekiel 16:15). The idea of trust as a covenant is in fact continuous with common usage from the 14th century to the mid-20th century in which ‘trust’ refers to a form of principal–agent relation, frequently objectified as a Trust, in which one acts on behalf of another. In a paper first published in 1904 Maitland (2003) outlines the history of trust in England, referring also to developments in Germany, France and the United States, wholly in terms of trust as holdings of land and similar assets on behalf of another in the absence of legal contract but with the requirement of honesty and diligence on the part of the trustee (Maitland, 2003: 88, 95). Trust in this sense has not escaped sociological consideration as ‘impersonal trust’ (Shapiro, 1987), which must be distinguished from trust as a quality of individuals, rather than entities formed by them, exercised in their interactions with others. Indeed, the recentness of the idea of trust as a condition of inter-personal relations of cooperation emerges in the scholarly literature only after the 1950s.

A Google Scholar search for articles on ‘trust’ by decade from 1900 revealed that up to 1950 the scholarly literature predominantly refers not to trust in inter-personal relations, as the term is widely understood today, but to trust as an entity, including corporate entities. This usage begins to change, however, from the 1950s through to the 1970s when a different understanding of trust emerges with the publications of social psychologists interested in inter-personal reliability or commitment and in development of such themes as trust and suspicion, trust and surveillance, trust and the F-scale and so on, possibly reflecting the guardedness of the Cold War period. During the following decade, 1970–1980, management researchers began to turn their attention to trust, borrowing the psychological meaning of the term. Trust becomes an established theme in the sociological literature only from the 1980s, encouraged by Luhmann’s (2017) essay, first published in English in 1979, and Barber’s (1983) short monograph, and marked by the reviews and expositions of Lewis and Weigert (1985) and Silver (1985) among others, and the multi-disciplinary collection edited by Gambetta (1988). The influence of social psychological research in the early sociological apprehension of inter-personal trust is evident in the fact that almost half of the references in Lewis and Weigert (1985) are to psychological publications. From this time a number of more thoroughly sociological monographs on trust begin to appear, the most notable include Fukuyama (1995), Misztal (1996), Seligman (1997), Sztompka (1999) and Möllering (2006). In addition to the relative recentness of sociological interest in trust is the growing intensity of that interest, described by one writer as an ‘explosion of social scientific work on trust’ (Shapiro, 2012: 99) which has affected not only sociology (Lewis and Weigert, 2012) but also management studies (Arnott, 2007) and political science (Citrin and Stoker, 2018; Levi and Stoker, 2000).

Expanding numbers of research outputs concerning inter-personal or social trust does not simply reflect changing scholarly taste. The increasing attention to trust over recent years has been taken as evidence of a new awareness of the role of trust in the consolidation and effectiveness of social relations. Indeed, it is implicit in some discussion that trust becomes of intellectually strategic concern only under certain conditions, as when Luhmann (2017) holds that the function of trust is to reduce social complexity, the latter being an historically emergent property of social systems. While different theorists draw upon distinctive vocabularies, there is arguably a convergence of ideas which suggest that in late modernity trust becomes important for the maintenance of social order, to paraphrase the subtitle of a key monograph (Misztal, 1996). It will be argued below that what might be described as ‘precarious institutional maturity’ gives rise to a set of experiences that are expressed through a vocabulary of trust, but in which trust is not necessarily constructive of social relationships, as widely assumed.

Before presenting the argument concerning the source of the sociological currency of inter-personal trust and its significance for apprehending the meaning of the term there are a number of additional factors that can be noted as relating to the particular nature of inter-personal trust and the recentness of sociological concentration on it. The idea that trust, as initially specified sociologically, ‘must be conceived as a property of collective units (ongoing dyads, groups, and collectivities) [rather than] of isolated individuals’ (Lewis and Weigert, 1985: 968; emphasis in original) is supported by the easy assimilation of trust into social capital, the latter being a transparently collective phenomenon. But even then trust in this sense necessarily remains the property of (non-isolated) individuals who engage in relations of trust with others. This form of trust operates in terms of dispositions, beliefs (or cognitions) and feelings (or affects and emotions) which are always properties of individual persons. A person who trusts another avoids isolation in the sense that trust is necessarily extended between persons. But inter-personal trust, unlike status or social standing, law and similar social phenomena, cannot be defined in terms of third-party and therefore collective involvement, facilitation or enforcement, even though the individual decision to trust has a context of extraneous and contingent relationships (Burt and Knez, 1995). Trust relations are always essentially dyadic, between two individuals, even if one of those ‘individuals’ is an abstract or collective entity in the form of an institution, a profession or a state, in which some form of ‘generalized’ trust is assumed, as ‘trust in abstract systems’ (Giddens, 1990: 83–88) or ‘system trust’ (Luhmann, 2017: 62–64, 99–100). The trust-giver him- or herself, however, is always an individual person.

The individual aspect of trust indicated in the previous paragraph, then, resonates with incipient social conditions which are reflected in recent developments. In broad terms, a growing research focus on trust coincides with the rise of rational choice theory, in which individual choice, including the choice to trust, is treated as a determinate factor (Coleman, 1990; Goldthorpe, 2000). While sociologists typically avoid regarding trust as amenable to calculative reasoning, trust research in contiguous disciplines encourage this perspective (Hardin, 2006). In any event, rational evaluations and decisions to trust are each necessarily undertaken by individuals. The upsurge in research on trust also coincides, more substantively, with a decline in institutional participation and commitment that arguably renders social life increasingly personalized, privatized and individualized (McPherson et al., 2006; Putnam, 2000; see also Lawler et al., 2009). This is the other side of an endemic decline in social trust (see O’Neill, 2002). These trends converge with another, connected with it in various ways, namely neoliberalism as late modern policy form. Trust, then, as a means through which individuals connect with others, is convergent with these other late modern methodological, epistemological and ideological trends. In spite of the insistence that trust ‘must be conceived as a property of collective units’, as quoted above, the concept of social trust refers to relations in which an individual, through their decision to trust, is the pivot and anchor of the implied connection. Only persons as individual actors can provide or reciprocate trust in the sense understood in the current literature, or be the objects of trust, provided ‘artificial individuals’ are included within the scope of the term. The implications of this observation will be expanded below.

Embedded trust and its obverse

If the provision and reciprocation of trust, and its object, are restricted to persons as individual actors, then a characterization of trust requires a statement of the properties of trust as they relate to the actions of individuals. For someone to trust another (including ‘artificial persons’ in the form of a profession, a state or some other collectivity) it is necessary that the trust-giver expects that the person they trust can be relied on to satisfy their needs or serve their interests. This is more than an expectation of benign intent and typically includes a sense that dependence on the other’s capacities or actions through trust will yield some benefit. Internal to this proposition is a second attribute of inter-personal trust, namely the idea that trust is a belief or a feeling. The point here is not to arbitrate on these alternatives regarding the subjective composition of trust, as cognition (belief) or affect (feeling or emotion), but to appreciate that trust is always based on something other than knowledge or certainty. This connects with a third characteristic of inter-personal trust, namely that it is always future-orientated. Trust is not simply a belief or feeling about another person but especially about what that person is likely to do in the future.

If trust facilitates relations between persons by providing a sense of assured expectation that one will not be subject to incompetent support or betrayal, there is a correlative expectation that an absence of trust would jeopardize cooperation between the persons involved and their relationship. Corresponding to this idea is a more positive supposition that given the prevalence of social relations it can be assumed that trust is self-enforcing, that in the provision of trust in relations between persons are factors inhibiting breaches of trust. These ideas have become central in sociological understandings of trust through the argument of embeddedness, an early version of which holds that embeddedness in social exchange relations generates and maintains inter-personal trust. This notion derives from what Blau regards as an elemental quality of trust, namely its propensity for self-enforcement (Blau, 2017: 28; see also 29, 89, 98–99, 133–136, 141–142):

Since there is no way to assure an appropriate return for a favor, social exchange requires trusting others to discharge their obligations [and while social exchange] may originate in pure self-interest [it generates] trust in social relations through their recurrent and gradually expanding character. (Blau, 2017: 94)

In this case ego’s trust of alter derives from alter’s demonstration of their trustworthiness by returning a favour received. In this account, then, trust both emerges from exchange and is required for it. Blau’s prescient account of trust has been highly influential (Coleman, 1990; Lin, 2011).

The argument concerning the self-enforcing nature of trust is more fully developed in Granovetter’s (1985) classic account of the embeddedness of trust in social relations: through repeated exchanges participants in social relations acquire information about each other so that should opportunistic behaviour occur it would be exposed, generating in others prosocial conduct through a fear of acquiring a reputation for unreliability or untrustworthiness with consequent loss of future transaction opportunities (Granovetter, 1985: 487–493). Trust thus arises out of socially embedded relations and those relations are in turn encouraged by trust. Granovetter (1985: 491) is aware, however, that ‘distrust, opportunism, and disorder are by no means absent … [and] while social relations may indeed often be a necessary condition for trust and trustworthy behavior, they are not sufficient to guarantee these and may even provide occasion and means for malfeasance and conflict on a scale larger than in their absence’. The acknowledgement here, that while social relations may be a necessary condition for trust they are not sufficient to guarantee it and may be a source of malfeasance, encourages critical reflection on the argument that underpins a great deal of the discussion of trust in social science literatures. Indeed, the confidence trick is a significant category of malfeasance that requires inter-personal trust (Goffman, 1952). Even in the more agreeable world of everyday interactions not only is trust likely to be more precarious than is typically assumed in the literature, but the inhibitions on betrayal of trust, that Blau, Granovetter and others suggest are internal to social relationships, are much less reliable than these writers assume; additionally, and even more telling, the failure of trust through betrayal is less significant for the relationships affected than is typically believed.

It is widely accepted that the provision of trust, like all investments, entails an element of risk. It must further be acknowledged, however, that as a form of commitment there is also in trust a potential for cognitive distortion and misinterpretation of relational information that is higher than would otherwise be the case. The idea, then, that information concerning risk to trust is available through social embeddedness cannot be taken at face value. Irrespective of motivational effects, such as wishful thinking, situations of uncertainty, including those in which trust is relied upon, inherently distort information processing (Tversky and Kahneman, 1974). In the provision of trust, additionally, wishful thinking is typically high. One who is committed to a trust relation is predisposed to its success and contrary signals are likely to be misread through confirmation bias or optimism bias (Duttle, 2016; Forgas, 2014). Indeed, over-estimation of the reliability of a partner in whom trust has been given is a typical pattern in such relations and an obvious factor underlying alter’s wilful betrayal of ego. Not even self-acknowledgement of broken trust necessarily leads to disclosure. Broken trust indicates judgement errors concerning the behaviour of the transgressor but also errors in the judgement of the trustor (Barbalet, 2009: 376). In exposing breaches of trust one also reveals self-limitations, and there are obvious inhibitions on doing so in embedded social contexts. This is a factor of embeddedness that contravenes the embeddedness argument.

The consequences of acknowledged breaches of trust are therefore not necessarily communication of broken trust and withdrawal from and therefore cessation of the relationship in which trust is regarded as playing a constructive role. One possible reaction to a breach of trust is an endeavour to re-establish it, to mend the broken relationship. Whether or not this prospect is practicable or desired, it does indicate a more general quality of social relationships, namely that there are bases for their continuing even when trust is broken or absent. Participation in the majority of social relationships does not require trust and while familiarity with others through sustained interaction may give rise to a sense of trust, a subsequent loss of trust will not necessarily disrupt the relationship in question. There are a number of reasons for this. A feature of embedded relationships is power disparity between participants. The experience of a betrayal of trust on the part of a subordinate is unlikely to lead to withdrawal from the relationship because of a lack of independent resources, fear of retaliation for doing so or desire to maintain rather than withdraw from current benefits. Irrespective of power inequalities, social relationships in employment, civil society and families entail varying degrees of multiple interdependence so that withdrawal from one particular relation has negative consequences for many others; a breach of trust may be tolerated in order to avoid jeopardizing other relationships because their loss is likely to be regarded as much more undesirable than the distaste of interacting with an untrustworthy person.

Social relations are subject to and generate practical interdependencies and are therefore compelling irrespective of trust considerations. A betrayal of trust in many instances will not disrupt the affected relationship; rather, it is typically significant on a plane of social reality quite distinct from the framework and content of ongoing social relations. Broken trust invariably provokes re-evaluations of self (as injured, foolish, vulnerable) and other (as distasteful, unreliable, dishonest). This revision in understanding the consequences of broken trust leads to reconsideration of the notion of trust itself. But before undertaking such reconsideration it is appropriate to consider another type of situation, in which trust is absent.

An absence of trust: The case of guanxi

Discussion above of the currency of trust and consequences of breaches of trust are not sufficient to situate the historical recentness of inter-personal trust. In order to advance the argument concerning the nature of social trust in late modernity it is necessary to consider a case of the absence of inter-personal trust through brief examination of the Chinese mode of relationship in guanxi. Guanxi is a form of asymmetrical exchange of favours between persons on the basis of enduring sentimental ties in which enhancement of public reputation or ‘face’ is the aspirational outcome (Barbalet, 2017; Hwang, 1987; Lin, 2001). In Chinese culture malfeasance tends to be dealt with by attempting to return disrupted relations between persons to harmony not by correcting wrongs with rights but through compromise and preservation of face, including for the wrongdoer. In these circumstances courtesy is more important than verisimilitude, and not telling the truth is not necessarily regarded as lying (Blum, 2007; Fu et al., 2008). An underlying requirement of trust relations, on the other hand, is strict truthfulness, the absence of which can be readily taken as evidence of untrustworthiness (Offe, 1999: 73–74). A significant difference between Chinese culture on the one hand and West European and American cultures on the other, responsible for the different understandings and evaluations of strict truthfulness, is the hierarchical nature of social relations and the determination of social obligation through role requirements in China, often summarized as renqing (Barbalet, 2018: 940–941; Ruan, 2017: 97–100), and the cultural significance of horizontal equivalence between persons in Europe and America (Fei, 1992: 62–76; Hamilton, 1990). In spite of routine acknowledgement of the above it is of particular interest that the majority of social science discussions of guanxi insist that it operates in terms of trust between guanxi participants. It can be noted in passing that this may be taken as evidence of the dominance of the ‘trust discourse’ in the social sciences today, and also of its potential distortion of certain categories of relationships through exaggeration of application.

In social science discussions guanxi is widely contrasted with more formal instruments of interaction and exchange, such as legal contract, and in these comparisons guanxi is almost always characterized as a trust-based relation (Boisot and Child, 1996; Lovett et al., 1999; Tong and Yong, 1998). Trust is not only regarded as the basis of guanxi but guanxi is also held to generate trust (Lee and Dawes, 2005; Smart, 1993). A generally agreed feature of trust, though, is that it is given by one person to another as a matter of individual choice and that compulsion can have no role in trusting another (Baier, 1986: 244–245, 247; Luhmann, 2017: 45) and therefore that it is ‘not possible to demand the trust of others’ (Luhmann, 2017: 47). In Chinese society, on the other hand, and in guanxi relations in particular, close personal monitoring, pervasive hierarchy-based dependence and role obligation together mean that inter-personal trust as it is normally understood on the basis of Baier and Luhmann’s criteria and others mentioned earlier is not possible. Arguments which may implicitly acknowledge these limitations but nevertheless refer to ‘trust’ in guanxi understand it in special terms, ‘not the same as the general Western conception of trust … [but as] trust in lian (moral character) … established through observation of social norms, or li, and based on renqing’ (Bedford, 2011: 155). This is similar to the idea that trust is a form of personal loyalty (zhong), a form of Confucian virtue (Ruan, 2017: 156), as well as an associated notion, the ‘normative structure of pao’ (rule-based reciprocity) which ‘implies … particularistic trust’ (Luo, 2005: 455). Such arguments for an indigenous basis of particularized trust may contrast it with (a low incidence of) generalized trust (Huhe, 2014). This latter is generally seen to derive from the invasiveness of official powers and the high incidence of corruption. Personal trust in business communities is also low given the prevalence of family involvement in business, so that non-family members tend to be regarded with suspicion or distrust (Ermisch and Gambetta, 2010; Whyte, 1996: 3–4). Pervasive book-keeping practices, including the routine provision of inaccurate reporting of transactions to business partners (Kao, 1996: 66; Wank, 1999: 73, note 4; Zhang, 2014: 25, 97), also lead to the absence of trust in business communities. While inter-personal trust and social trust are thus arguably absent from Chinese society there is, nevertheless, high social solidarity, especially between guanxi participants.

Rather than ‘trust’ the effective bond between persons in guanxi relationships is what might be described as ‘sincerity’, ‘integrity’, ‘credibility’, ‘reputation’ or possibly ‘trustworthiness’, captured by the Chinese term xinyong (Ruan, 2017: 154–145; Tong and Yong, 1998: 85). Indicators of reliability are displayed through habituated behaviour which expresses role obligation through signals of probity; these are achieved through repeated and close contact and other bases of familiarity designed to reassure the other of the dependability of the person entering a guanxi exchange or continuing in a guanxi relationship. Trustworthiness in this sense, as distinct from trust (see Hardin, 1993: 512–513; 1996; Tullberg, 2008), is here premised on a social perception of reliability expressed as social reputation or face. The distinction between trust and trustworthiness is frequently ignored, and while confusion between the two is almost a constant in discussion of both trust and guanxi it is in the study of guanxi that the importance of the distinction becomes especially clear. It was mentioned above that successful guanxi exchanges enhance the standing or reputation of participants, that is to say, in their practice of guanxi participants gain face (Ho, 1976; Hwang, 1987; Qi, 2014: 143–164). Face and reputation stand as proxies for reliability or ‘trustworthiness’ in guanxi relations.

A point to notice in this brief account is that social reputation, and especially its regulatory mechanism in face, does not operate in terms of dyadic relations, as with of trust, but in triadic relations in which reputation and therefore ‘trustworthiness’ is a function of the public visibility or third-party judgements of performance of expectations regarding guanxi decorum and adherence to guanxi norms (Barbalet, 2014: 63–64). In his discussion of relations within a Taiwanese business community DeGlopper (1995: 205–206) indicates that the ‘firm’s most valuable asset is hsin-yung [xinyong] … a reputation for meeting one’s obligations’. The achievement of this reputational trustworthiness, he goes on to say, is in the fact that ‘All transactions take place before an audience or chorus of nei-hang-ren [fellow businessmen], who continually observe and comment on each other’s doings’ (DeGlopper, 1995: 206; see also Hamilton and Zheng, 1992: 27). That guanxi exchanges necessarily involve third-party observation through which the currency of reputation is maintained and sanctions against possible defection from agreements are executed indicates a triadic form unlike the dyadic structure of trust relations (Barbalet, 2014). Assurance in guanxi derives not from inter-personal trust but from public or third-party scrutiny in which successful adherence to the norms and expectations of participation (renqing) leads to enhancement of social reputation or the gaining of face, and defection or incompetence in maintaining the decorum or norms of guanxi leads to loss of reputation or loss of face, and therefore likely exclusion from future guanxi exchanges. Knowledge concerning adherence to these norms derives from mutual surveillance and close monitoring.

While the assurance mechanism of guanxi is triadic, in the initial formation of a guanxi relation two persons relate to each other dyadically, often disclosing personal information as evidence of sincerity (Barbalet, 2014: 60–61; Lin, 2001: 158). It is possible that these practices of self-disclosure may be seen to be similar to those that are regarded by some writers as generative of what has been called ‘swift trust’ (Meyerson et al., 1996; Robert et al., 2009) or ‘fast trust’ (Blomqvist, 2005; Perks and Halliday, 2003). In the present case, however, the notion of trust – swift, fast or otherwise – as a basis of guanxi cooperation is misplaced. This is because the intimate bonding practices of guanxi, while convivial also possess an underlying coercive element that is the obverse of trust. This latter factor is not pernicious, as in blackmail, but only because the covert potential threat is mutual rather than asymmetrical. These bonding practices are close to those of sworn brotherhood (jiebai xiongdi), entailing not only secrecy but self-interest dressed as group loyalty. The basis of cooperation in these cases, then, is not conceivably based on trust but is in fact its opposite.

The absence of trust in Chinese guanxi and Chinese social relations in general is largely a consequence of the continuing significance of intergenerational obligation in Chinese family life and social structure (Qi, 2015, 2018; Whyte, 1997; Yeh et al., 2013). In Europe, on the other hand, from the 9th century extended kinship structures ceased to underlie economic, social and legal forms and relations, and non-kin cooperation and impersonal exchanges were supported by developments that reinforced a generalized morality and a formalized infrastructure of enacted codes and regulations (Greif and Tabellini, 2010). These developments were completed in England by the 16th century (Stone, 1975). In China, on the other hand, kinship solidarity and clan structure were the primary source of public goods, including education, welfare and public safety; economic and civic cooperation was sustained by family obligation and reputation (Hamilton, 1990). Dispute resolution did not enforce an abstract morality or legal framework, as in Europe, but arbitrated compromise. Rather than an interdependence of ascriptive relations based on role obligation, persons in Western societies experience themselves as autonomous and independent from others, constrained only by the rules and requirements of the institutions that provide the basis of social being in the absence of socially determinative intergenerational kin relations.

The institutional context of trust

It can be concluded from the preceding discussion that the institutional framework in which inter-personal trust operates is characteristic of European and American economic, legal and familial refinement and absent in China, where a different institutional framework operates. Particular institutions permit and effectively encourage trust between otherwise independent and autonomous persons because they provide sets of general rules and norms which support a number of qualities that underscore a sense of collective belonging, as noted by Offe (1999: 70–76), and at the same time, and just as importantly, provide sanctions against defection from trust and protection for those whose vulnerability is exploited by untrustworthy compatriots in the form of legally based guarantees or means of compensation in the event of breaches of trust. Persons in West European and American societies feel that they might trust another because of an institutionally inculcated set of socially diffuse universal rules and norms as well as institutionally provided sanctions and mechanisms of assurance, so that while trust always involves some risk it will not be reckless to risk trust in these circumstances.

The argument here, then, supports the proposition that while the notion of trust offers little for understanding social relationships in China it is central to social relationships in West Europe and America. This is because the persistence of kinship-based norms and intergenerational family obligation in China means that the preconditions for trust, namely that it is prior to obligation and based on an unimpeded choice to trust, do not obtain; at the same time these preconditions are defining of the social conditions that characterize West European and American social life. In these latter persons are independent of each other while at the same time subordinate to institutions emanating universal norms and societally based safeguards. The institutional framework through which inter-personal trust is possible is seldom addressed in the literature, but its qualities mentioned here, including the instruments of legal protection that reduce the consequences of risk for trust-givers, have been present in varying degrees since the late 18th century with the rise of liberal democratic polities. Given the relatively long historical presence of these conditions, the broad institutional argument itself, then, does not explain the recent social science interest in trust which is taken here to reflect a distinctive and novel discourse concerning inter-personal trust in everyday experience and the social science exploration of that experience. An argument addressing the rise of this discourse requires a periodization of social being in liberal democratic polities in which operate the social conditions of trust relations.

The pre-history of the importance of trust for understanding inter-personal or social relationships is provided in a classic account which shows how customary society is subverted through the emergence and development of national markets (Polanyi, 2001). According to Polanyi, persons in customary societies are interdependent members of kin-based collectives. The extension of market prerogatives in this context undermines customary relationships by ultimately constituting persons as self-sufficient and independent entities. The political corollary of this development is a state sovereignty which relates to individual persons as citizens defined by their political rights and freedoms. These dual processes of market and political individuation converge in constituting the experience of self as essentially isolated from other selves. In these circumstances cooperation is necessarily based on relations of trust between persons who otherwise would have no foundation for the formation of relations with or obligation to others. This general situation can be divided schematically into two distinct sequential phases.

In the first phase of the history of what might be described as individuated society the self is not only experienced as an independent entity but regarded by participants and observers alike as rational in being both the proprietor of his or her own capacities and an able manager of his or her own interests. The sense of rationality here, and the effectiveness of (calculative) consciousness as its mechanism, requires that persons have a sense of their ability to exert a meaningful influence on processes to which they are subjected. This prospect prevails when the scale of organizations in the society is commensurate with effective engagement on the part of individuals. In general terms, such a situation exists historically even after the collapse of craftwork and the rise of a factory system when ownership continues to provide managerial discretion and when local economies are not predominantly subject to global forces. Cooperation between individuals based on trust relations under these circumstances is experienced by participants to derive from their conscious calibration of interests; the sense of trust is here secondary to those interests and elided as a means of satisfying them. It is in these terms that at this time trust is objectified as either a covenant relation or a form of agency as trusteeship, as discussed above.

In a second phase of the historical development of institutionalized society, while experience of self as an independent being continues, the sense of possessing a capacity to exert an influence on market and state administrative processes is diminished and becomes remote as bureaucratization and globalization advance. In these circumstances the notion of the self undergoes qualified reconceptualization. The self in this phase of development is orientated not principally to consciously managing external forces but is experienced as an arena in which the individual’s subjective faculties require self-management. At this historical juncture, then, a sense of the rational control of external factors yields to a felt need to adjust psychic reactions to circumstantial limitations (Barbalet, 2001: 172–174). Self-awareness of trust is now experienced as a belief or feeling about the security and consequences of inter-personal relations; it emerges in these circumstances as a primary focus of cooperative relations and it is in these terms that it becomes an object of social science enquiry. This situation results from organizational growth towards global dimensions that is experienced by individuals as ‘precarious institutional maturity’; it includes a sense of increased insecurity to the effect that inter-personal trust is assumed to be necessary for social life even though trust is understood to be in itself precarious or fallible.

The salience of inter-personal trust as a substance of direct experience is heightened in this second phase of individuated sociality under conditions of precarious institutional maturity; the advent of a social science research concentration on inter-personal trust reflects the experiential concerns of persons as they navigate their relations with strangers in pursuit of future benefits. The effectiveness of trust, however, in realizing opportunities in social relations is largely over-estimated when only the sense of the importance of trust for social participants is the basis of scholarly narrative. When the specific institutional context, which supports the possibility of trust relations and informs their complexion, is left out of the analysis then the relational significance of trust becomes exaggerated. As a social phenomenon inter-personal trust is more a reactive attitude to precarious institutional maturity than a means of relational efficiency. It has been shown that relationships are typically not interrupted by failures of trust. The efficacy of trust in social relationships more completely operates on the subjective plane of attribution of motive or value to self and other. When trust is understood in this light the relatively muted consequence of trust, whole as well as broken, for the actual practice of ongoing and future relationships will be appreciated, for it has long been understood that motive and value have only indirect and arguably indifferent relations with actual behaviour (Bardi and Schwartz, 2003; Hechter, 1992; Mills, 1940).

Conclusion

The important insight, that trust ‘varies in relation to the structure of society’ (Morgner, 2013: 510), is too frequently ignored. The present article examines inter-personal trust in terms of its relative recentness in sociological research and the tendency to exaggerate the relevance of trust for the stability and security of social relationships. The term ‘trust’ has a long history; distinctive specific meanings of the term operate in different historical periods. In this article it is shown that up to the middle of the last century trust is typically understood as a form of covenant and also as an agency relation of trusteeship. Sociological research on trust, on the other hand, beginning in the late 1970s, tends to conceptualize trust as a quality of inter-personal relationships. While it is widely held that these relationships readily generate information regarding breaches of trust and that should such breaches occur the threat of withdrawal from them is high, it is shown above that these assumptions are less credible than the literature suggests. It is also shown that trust is not the only means whereby inter-personal relations are secured, as in the case of Chinese guanxi in which trust is functionally absent even though widely assumed to be central.

These findings encourage the argument presented here that trust requires a specific form of institutional context which mirrors social individuation, the situation in which individuals are self-sufficient and in that sense isolated from each other, and in which trust in another is a means of relational engagement in overcoming separation between strangers. The article goes on to show that the conditions of social individuation can be understood to obtain in terms of two distinct phases and that a sense of inter-personal trust is particularly relevant to social actors in the second phase when trust is experienced as a felt motive for action or a value in any given action. In these terms trust is a condition of appraisal rather than a basis of action. This novel characterization of inter-personal trust accounts for both its high social salience and its low behavioural significance.