Trust:The Moral Importance of an Emotional Attitude 

---

Practical Philosophy November 2000 Volume 3.3 Pages 45-54

Jessica Miller

Annette Baier has recently pointed to the ‘strange silence’ of mainstream moral philosophy on the topic of trust.  In my view, this neglect has a very definite source: the tendency to assume that trustworthiness is the real moral good, and that trust is at best a necessary evil in a world of less-than-fully-trustworthy people.  Russell Hardin makes a representative remark when he writes that ‘trust by itself... constitutes nothing. … without [trustworthiness], there is no value in trust’ (Hardin, 1996: 28-9).  And, noting that trust implies vulnerability and is unavoidably risky, Partha Dasgupta asserts that, ‘The problem of trust would of course not arise if we were all hopelessly moral’ (Dasgupta, 1988: 53). Of course, these theorists are correct in pointing to the connection between trust and trustworthiness. No one would deny that to continue to trust in the face of blatant and ongoing betrayal is foolish.  But the tendency to view the value of trust as parasitic upon, and thus less than, that of trustworthiness obscures the ways in which trust itself can be morally valuable.  In this essay, I first define trust as a kind of emotional attitude.  I argue that this account is superior to a competing account, which I call the ‘strategic view’.  I then argue that trust can be morally valuable in four areas of ethical life: moral development, moral identity, moral perception and judgement, and living a good life.

According to the ‘strategic view’, trust has very little, if anything, to do with emotions.[1] On this view, cooperation is necessary for social life, and trust is necessary for any co-operative scheme to get off the ground.  Trust is manifested in strategic choices about action which are based on a truster’s beliefs or expectations about the potential trustee.  Specifically, person A trusts person B if A believes or expects that it will be in B’s interest to be trustworthy in the appropriate manner at the appropriate time.  Such beliefs are translated into an assessment of whether trust in a particular case is too risky given the potential rewards.  In sum, trust reduces the potential uncertainties of dealing with others to something better than random chance. There are several problems with this strategic understanding of trust. First, few people would recognise themselves on this model, which ignores the fact that conscious monitoring of transactions might actually destroy some relationships.  Unless something has already gone disastrously wrong, a spouse, lover, parent, or friend (to name just a few examples) will not tend to think in terms of what he or she might gain from staying in the relationship versus what he or she might gain from leaving it. In fact, there is something contradictory in the view that one person trusts another because she judges (consciously or not) that he is motivated to be trustworthy. Suppose the two are spouses.  The strategic view tells us that a husband is motivated not to betray this trust (say, by being unfaithful) by the thought that their committed marital relationship is of value to him.  But if this is true, then it is not the case that fulfilling his wife’s trust is his primary concern.  Rather, it is second to his concern to stay in a committed marital relationship with her.  But, what is a committed marital relationship in which fulfilling trust obligations is not of primary concern? Put more broadly, it seems to me that I do not really have good reason to trust someone who is solely motivated by self-interest, because should his own interests change, so will his preferences, and therefore his choices.

Second, a very real problem is the strategic view’s unwillingness to acknowledge the importance of the past for present acts of trust.  Most proponents of the strategic view assume that two rational agents, given the same information about the other players and identical preference rankings, should trust to the same degree.  This excludes the possibility that variations in the personal histories of different agents can affect the way that agents perceive co-operative ventures.  But surely, a history of betrayed trust (in other relationships) will have some impact on an agent’s ability to trust in new situations. The third major problem with the strategic view of trust is that it leaves out entirely any emotional component of trust.  That there is an affective component of trust becomes apparent with a moment’s reflection: trusting and distrusting often do not feel the same.  This has been observed by Annette Baier, who notes that ‘trust has a special feel, most easily acknowledged when it is missed, say, when one moves from a friendly ‘safe’ neighbourhood to a tense insecure one’ (Baier, 1994: 132).  Carolyn Gratton, researching the ‘lived experience’ of trust, found that people tend to describe trust in physical terms: sudden breaches of trust and distrust are described as bodily ‘stiffness’, ‘alertness’, or tenseness’, while trusting is described as ‘lightness’, ‘relaxation’, or ‘calmness’ (Gratton, 1973: 281).  And Susan Brison notes that the loss of self-trust that accompanies trauma can be manifested in heightened bodily awareness or, in extreme cases, eating disorders (Brison, 1997: 18).

It might be objected that affects cannot be integral to trust since trust does not always feel like something.  In response, it should first be noted that such feelings need not be experienced as bodily sensations. As Justin Oakley points out, joy and disappointment are examples of emotions which may not be accompanied by bodily affects, and which we often describe in terms which imply psychic feelings, such as feeling ‘up’, or ‘down’. Moreover, it is possible to be joyful or disappointed without being even psychically aware of it.  Oakley calls this sort of un-felt feeling a ‘psychic affect’, and describes it as a kind of ‘mental tone’ which ‘permeates our perceptions, desires and actions in ways which we are not always aware of’ (Oakley, 1993: 11).  Psychic affects are also characteristic of enduring emotions, such as love or grief.

By cashing out the affective component of some emotions in terms of a perception-guiding mental tone or colour, Oakley echoes other theorists, such Ronald de Sousa and Amelie Oksenberg Rorty, who contend that emotions are made up in part by intentional habits of salience, organisation, and interpretation, which, while not beliefs in propositional form, can give rise to beliefs.  Similarly, trust is a way of seeing that guides our attention, colours our perceptions, and thus gives rise to certain beliefs and precludes entertaining others about the one trusted.  Trust involves dispositions to have certain affects in the presence of the trusted other.  Typically, it is characterised by a sense of security which screens out negative interpretations of motives and behaviour.  This explains why interactions with the trusted other, although by definition fraught with risk, are not experienced as risky.[2] It may seem odd, in light of the fact that I have defined trust as a kind of emotion, that I include a cognitive component.  However, many philosophers, including Rorty, de Sousa, and Oakley, agree that emotions have some sort of cognitive content.  This agreement arises partly from the difficulties in trying to characterise different emotions solely on the basis of how they feel.  But there is no compelling reason why the cognition in question must be a proposition, in the form of either a belief or evaluation, rather than a mere apprehension or imagining.

The affective component of trust is some kind of feeling of security, hopefulness or optimism.  But what are we secure, hopeful, or optimistic about? Following Karen Jones and Trudy Govier, I will characterise this as a feeling of security in the benevolence and competence of the trustee in a particular sphere of interaction.  In normal cases, this sphere of interaction will be quite small for strangers, and progressively larger as relationships become more intimate.  Correspondingly, the basis of the trustee’s benevolence might well vary with the sphere of interaction: we might expect the benevolence of a stranger to be based on a general goodwill towards humanity, the benevolence of a boss to be based on an appreciation of our diligence and skill, and that of an intimate to be based on fondness or love of us in particular.

The competence required for a trust relationship is of two interrelated sorts.  One we might call ‘technical competence’, namely the ability to do what is expected by the truster, whether to fix the pipes, to put the children to bed at 7:00pm, or to repair the torn ligament.  The second kind of competence underlies, and in some trust relationships is inseparable, from the first: this is an ability to understand what is expected by the truster.  As Annette Baier has claimed, trust always involves discretionary powers.  Knowing what one is trusted to do (or not do) often requires paying attention to the cues - explicit and implicit - offered by the truster.

On this account, a truster has a special sort of expectation: she expects that her trust will have a certain motivational force on the one trusted, such that he will be directly moved (when appropriate) by the thought that his actions affect someone who is vulnerable to them within a given sphere of interaction. The expectations characteristic of trust are closely tied to particular emotional responses in the face of disappointment.  The feelings of betrayal, anger, and indignation which attend breakdowns in trust are evidence for the claim that trust involves an expectation that one’s trust will have motivational force on the trustee.  In fact, the truster’s readiness to feel betrayal in itself accounts for at least some of the motivational force.

This emotional component of trust is important for making sense of how it is possible to believe that someone is trustworthy but not to trust him. This is a major reason why trust is not best understood as a judgement. To demonstrate this very point Karen Jones has constructed a situation in which she distrusts a salesman who has been recommended by a friend.  Jones believes that the salesman is trustworthy, yet cannot help but have the attitude of distrust towards him.  She asks us to assume both that she cannot explain why she finds him untrustworthy and that she is, without realising it, a terrible judge of trustworthiness (Jones, 1996: 24).  Her view of the salesman as untrustworthy, therefore, has not been ‘formed by a reliable process’ and is actually undermined by another belief.  Moreover, her attitude does not ‘track the truth across some range of counterfactual circumstances’ (Jones, 1996:24).  Thus, argues Jones, her distrustful attitude is unjustified according to both internalist and externalist accounts.  Yet, if what is at stake in this case is very valuable, we might still want to say her distrust is justified.  Because trust is a complex emotional attitude, and not a belief, the way is open to make this claim.

A second case in which beliefs and attitudes may not coincide is discussed by Trudy Govier in the context of self-trust.  Govier notes that while anyone might doubt her abilities at one time or another, ‘core self-distrust’ can threaten one’s very identity, by preventing the development of self-esteem and self-respect (Govier, 1993:109).[3]  Yet, a person who suffers from core self-distrust can feel competent in several areas.  But if self-trust is a matter of beliefs, we seem forced to say that she is irrational because she both believes and doesn’t believe that she is competent and trustworthy.  However, if we allow that the cognitions which are a part of trust can take the form of felt apprehensions or imaginings, as opposed to beliefs, we are able to say that this person has a self-image, a felt way of seeing herself, which resists the reinterpretation suggested by specific beliefs, such as beliefs that she is a person of worth or is well-qualified for the position or is trustworthy  

One might think that if trust is an affective attitude, then it is useless to try to specify conditions for justified trusting.  I agree that it is impossible to come up with a set of necessary and sufficient conditions for justified trust, but not because trust is an affective attitude: as noted above, emotions usually have some cognitive content which does, however roughly, track the world.  This is why it makes sense to speak of unreasonable anger or grief.  Rather, the problem is that several factors, which can combine in different ways, determine whether or not trust, in any given case, is justified.

One factor, most apparent in cases of strangers trusting other strangers, is the presence or lack of a climate of trust.  Baier notes that in some countries, such as her native New Zealand, doors are left unlocked and children are not discouraged from accepting strangers’ gifts, while in the United States candy from strangers is rarely and suspiciously accepted (Baier, 1994: 191).  Recently, a native of Denmark was charged with child neglect for leaving her infant in a stroller outside of a New York restaurant while eating lunch inside.  While many New Yorkers found the woman unreasonably trusting of passers-by, she successfully defended her actions by noting that this was common custom in Denmark.  Of course, differences in trust climate can be felt when moving from city to city, or even neighbourhood to neighbourhood.  We also speak of climates of trust in organisations, firms and corporations.  During wars, under totalitarian regimes, or in times of great social upheaval, the rational stance towards strangers might be one of distrust.  A pervasive enough climate of distrust can destabilise trust among acquaintances, friends, and intimates, as well.  Trust involves expectations, and what is reasonably expected of not only strangers, but friends, professionals, politicians, spouses, and other social groups can vary.

A second variable involves the content of the trust, and the consequences of its betrayal.  Baier has suggested that all trusting can be viewed as a kind of entrusting, where A trusts B with valued thing C (Baier, 1994: 102).  Although she has been criticised for this model, on the grounds that it makes trusting seem too much like something that can be willed (Jones, 1996: 17), Baier’s proposal does focus attention on the sense in which trust implies a vulnerable good.  Much more is at stake in leaving one’s child with a stranger than in asking a stranger to hold one’s place in line. Betrayal in one case is merely irritating, in another, tragic.  Thus, the nature and importance of the shared good make a difference in evaluating the reasonableness of trust.

Baier’s formulation also helpfully draws attention to the sense in which trust, even among strangers, is relational.  This suggests that a third factor is the particular relationship between or among the trusting and trusted parties.  The duration and nature of the relationship will in part determine whether trust is justified, and will specify both what is being entrusted, and the scope of the trustee’s discretionary powers.  Of course, in a loose sense (which I cannot explore here) social groups can have a history of  ‘relations’ with one another that have consequences for the reasonableness of trust among their members. Also, as noted above, one’s personal trust history bears on the question of whether and when trust is reasonable.  A history of betrayal can make it reasonable for someone to withhold trust even in relatively stable or long-term relationships.  Distrust of strangers, even in a trusting and trustworthy climate, can be a reasonable response to violent trauma.

Finally, keeping in mind that this must be understood against a background of one’s own social knowledge and location and history, we must consider the kinds of information that lead to the development of the attitude of trust in one person towards another person.  I have stressed that trust is grounded in a kind of emotional perception, but it does not arise ex nihilo.  Trudy Govier claims that there are five basic sources of trust, in descending order of status: direct personal knowledge (A encounters B), indirect personal knowledge (A knows C who knows B), book knowledge (A reads about B), ‘mediahand’ knowledge (television accounts, etc., of B) and knowledge based on social role (B is a priest, a doctor, a police officer, etc.) (Govier, 1993: 158-9).  Each kind of knowledge contributes to a sense of the trusted’s character.

Additional information comes in the form of social prejudices.  We interact daily with, and take some kind of trust attitude towards, others who are not well-known, either by oneself or by close associates.  In such cases, we often rely on what Margaret Urban Walker has called ‘socially salient identities’.  That is, we view the potential trusted as a representative of a social group (or multiple social groups) which has been defined in terms of commonly understood characteristics (Walker, 1998: 178).[4]  These representations, or what Walker calls ‘stereo-graphy’, are images that fuse representations of a group to one kind of bodily configuration.  Needless to say, such compelling, emotionally charged images can easily convey prejudice.  Although our own trust-histories can particularise our perceptions of the trustworthiness of others, the attitude we take towards unknown others is often not just a matter of individual psychology.  Rather, it reflects clear cultural patterns of representation, some of which are morally questionable or even morally malignant. 

While the question of the influence of prejudice on trust attitudes is an important one which cannot be explored here[5], it does raise directly the question of what, ethically, is at stake in trusting others.  Having given an account of trust that emphasises its intersubjective, affective, and responsive components, it is now possible to explore the moral significance of trust.  Annette Baier has pointed to one area of moral life where trust is vitally important: the cultivation of new members of the moral community.  Successful parenting implies trust between parents and infants or children.  Trust between parents and children enables the close relations which form the primary context for moral teaching and learning, that is, for producing moral agents at all.

However, there are several other ways in which trust is morally important. In my view, Baier’s account of the role of trust in moral life is somewhat limited by her conception of appropriate trust as ‘letting other persons … take care of something the truster cares about’, and as one of several virtues ‘democratically ruling in our souls’ (Baier, 1994: 10-11, 188). This definition encourages us to think of trust as something we do to protect things we value, and moves the discussion away from the relation between the truster and trusted (Baier, 1994: 105, 101).  As a result, Baier does not explore the links between trust and the development or maintenance of a distinctive moral identity.

The idea that moral philosophy should account for individuated moral identities by taking a longer view of moral lives has recently been taken up by several philosophers who share a dissatisfaction with the traditional emphasis on moral life as a series of discrete decisions.  These writers agree that a distinctive moral personality reflects not misguided partiality for the ‘dear self’ but a kind of self-definition and moral integrity which are valuable both in themselves and for the moral capacities they make possible. Having a moral identity involves valuing certain things over others, cultivating specific character traits, and having special moral responsibilities which arise in relationships with specific others.  In my view, this is the kind of approach required for seeing trust as more than the residue of reliably trustworthy behaviour.

One of the capacities required for having and refining a moral identity or moral persona is what Diana Meyers calls self-recognition.  Self-recognition is ‘the self-directed care that consolidates independent subjectivity’ (Meyers, 1994: 128).  The moral components of self-recognition include ‘endorsing sound values, developing one’s moral capacities, and living up to one’s ideals.’ (Meyers, 1994: 128).  As Meyers notes, the idea of a distinct moral identity, far from being exotic, shows up in everyday discourse, especially in morally reflective queries such as, ‘Do you want to be the sort of person who would do that’ or ‘Could I do that and still live with myself?’ (Meyers ,1994: 128-9).  Such questions reflect the fact that most people have a distinct moral ideal, a conception of the kind of person one hopes to be. Self-recognition involves developing and refining one’s moral identity, especially by enacting one’s moral ideal.

Meyers does not address the question of trust, but it seems clear that self-recognition and self-trust are related phenomena, and that the latter, like the former, has a moral dimension. While any of us can have moments of insecurity and doubt, basic self-trust is necessary for the kind of moral reflection to which Meyers draws our attention.  A self-truster views herself as competent and benevolent;  she can count on herself.  Without self-trust, a kind of meta-introspection - a constant interrogation of one’s version of events, one’s motives, the value and shape of one’s moral ideal - would paralyse moral reflection.  Moreover, because self-recognition also involves acting on one’s moral ideal, the absence of optimism about one’s ability to follow through could undercut one’s drive to engage such reflective capacities. Also, because trust implies vulnerability, a deep and pervasive lack of self-trust can interfere with one’s ability to trust others.  I have defined trust as a certain attitude, a kind of optimism about how others will respond to one’s trust in them.  But such optimism is not a blind faith, a ‘cheery denial’, which is characteristic not of trust but of credulousness.[6]  If (as I shall discuss in detail below) trust requires recognition of the other, then it includes some awareness of the possibility of betrayal.  Trust entails a vulnerability which is minimised and/or reinterpreted by the trusting attitude, but which is never completely eradicated.  Thus, without self-trust, confidence in one’s ability to trust wisely or to survive betrayal by others can reduce one’s tendency to trust. The practice of self-recognition can also enhance self-trust.  This connection between self-trust and moral identity is reflected in the language of betrayal which people often use to describe acting in a way that conflicts with their moral ideals.  If honesty is a prominent feature of my moral identity, then my failure to be honest when it is called for can be as surprising, disappointing, and upsetting as being deceived by others.  Conversely, successfully enacting one’s moral ideal can increase one’s sense of optimism about facing moral challenges in the future.

While Meyers’s work is suggestive of the links between self-trust and moral ideals, connections between trust and such ideals are evoked by the work of Margaret Urban Walker.  According to Walker, individuated moral identities are necessitated by the fact that there are innumerable values located within acceptable moral limits.  Her investigation of such identities begins with her observation that people can and do make moral judgements which are ‘irreducibly particular’ (Walker, 1987: 176).  Particular moral judgements cannot be universalised because what makes them valid is not the connection between particular features of a situation and some moral rule, but their expression of an agent’s moral persona, the ‘existing or prospective uniformity over performances of this moral agent’ (Walker, 1987: 177).  The bindingness of such moral judgements arises from one’s sense of unique moral identity, not one’s sense of universal moral agency.  In short, particular judgements ‘constitute strong moral self-definition because by them one enacts and imposes on oneself particular constraints, and purports to stay or embark on a moral course not for all’ (Walker, 1987: 180).

Making particular moral judgements is not just something we happen to do, but represents the exercise of a special competence, ‘strong moral self-definition’ because they ‘consist in an agent’s assigning to particularised grounds a discretionary value (compatible with generally acceptable orderings of generic values) in the act of affirming a certain moral position on their basis’ (Walker, 1987: 177).  To illustrate the exercise of this special moral competence, Walker describes the situation of a man, Jeff, who is facing the decision of whether or not to institutionalise his increasingly dependent mother.  For Jeff, deciding what to do is not a matter of thinking up every possible course of action within minimal moral limits.  Rather, it depends heavily on which particulars he has granted relevance through progressive moral self-definition.  What makes his judgement ‘irreducibly particular’ is how the particulars are ranked and weighted relative to each other.  Is this a matter of betraying a deeply significant relationship in a time of need, compared with which the disapproval of a spouse is less significant?  A matter of what he owes his mother compared with what he owes his own children?  Or will moral reflection lead him to reinterpret his loyalty to his mother as a form of dependency which threatens his other adult relationships?

Clearly, once we take into account moral identities, ethical judgement looks very different from the ‘snap shot’ view, on which decisions are completed in an instant.[7] In her most recent work, Walker highlights this feature by referring not to moral personae but to ‘narratives of moral identity’ (Walker, 1998: 112). This term brings out the sense in which particular judgements reflect past choices and commitments, and bear on future ones.  It is also phenomenologically accurate, for it does justice to our characteristic tendency to represent moral problems as stories.  

Walker claims that moral identity ‘gives our deliberations greater focus and refinement’ (Walker, 1998: 112).  Assuming that both a tendency to trust and particular trustings are parts of a moral identity, then there is a very specific sense in which trust contributes to such refinement. Specifically, trust can aid moral judgement by keeping the moral landscape uncluttered. Trust can help us determine when moral judgement is called for by screening out morally insignificant events.  For example, if Romano trusts his wife, then neither her late nights at work, nor the occasional caller with the wrong number show up as signs of behaviour that deserve scrutiny.  Similarly, if Smith trusts her neighbour, then his inviting her inside his apartment, or forgetting to return her lawnmower do not set off any alarms.  Without trust in these situations, all of this behaviour would be open to the full range of possible interpretations.  Moral reflection would become so complex as to be impossible, and moral agency would be stunted.         

Trust can also help morally praiseworthy behaviour show up for us.  If I trust a colleague, her offer to fill in for me at a meeting I cannot attend or teach my class if I am ill shows up as caring and concerned.  If I do not trust her, I may view her offers as attempts to usurp my position, or even to embarrass me.  If I trust my husband, I interpret his offer to make dinner on a night when it is my turn as helpful, not as an attempt to distract me from some illicit activity on his part.  In general, the effectiveness of moral sensitivity can be enhanced or diminished by the trust attitudes we have towards persons in their particularity.        

Assuming that a moral identity is public, or publicly shared, then we have a way to interpret the link between trustworthiness and trust which is more complex than the standard view of trust as a rational response to trustworthiness, which always has the effect of making trust a subsidiary phenomenon.  The key concept is integrity.  According to Walker, the kind of integrity which arises from one’s moral identity is relational.  It is a reliable accountability in terms of commitments and values, and a readiness to respond to the results of the accounting (Walker, 1998: 118).  This is in contrast to the standard definition of integrity as a form of internal cohesion of parts of the self which is strong enough to resist self-betrayal in tempting situations. Integrity as reliable accountability suggests a meaning of trustworthiness which has less to do with specific acts (lying, breaking promises, etc.) than with a kind of honesty in self-presentation.

This suggests that one way to look at trust is as a form of moral response to the self-presentation of others.  When we trust, we take an attitude towards someone which affirms his moral worth.  This is partly because trusting people involves seeing them as competent and benevolent, and gives rise to optimistic expectations of response to the trust placed in them.  But trust is also affirming because it accepts the trusted’s sense of self as it is revealed in his self-presentation.  Baier is right that we need others’ help in caring for our valued goods, but it is the attitude we take - towards this need, towards the vulnerability it connotes, and towards the one(s) who can respond - which makes trust morally significant.

Moreover, by trusting someone, I give that person certain moral opportunities, for cooperation, generosity, or beneficence, for example, which she might not otherwise have had.  Thus, when I trust someone, I help her exercise moral agency.  This feature of trust is overlooked in accounts which view trust solely as a response to prior actions of the trusted.  Trust is a forward-looking phenomenon, in that my perception of the trusted leads me to interpret her actions and intentions as benevolent.  This helps explain one oft-noted feature of trust: it tends to increase rather than decrease with use.  This is because trust, in its forward-looking aspect, always leaps ahead of what is rationally justifiable, gradually increasing in range or quality.

Trust as moral affirmation implies that trusting someone requires recognising him as a subject.  Anthropologist Judith Rollins’s ethnographic account of her fieldwork as a house cleaner in the suburbs of Boston provides an excellent example of what is missing in forms of ‘trust’ which do not include recognition of the trusted.  Rollins was regularly left alone in the homes she cleaned, which might seem to indicate trust.  However, Rollins’s descriptions of her invisibility to her employers caution us against imputing trust to them.  In one case, her employers left the house, locking her inside.  In another, they turned the heat off when they left for the day.  Rollins was so invisible to her employers that they frequently carried on deeply personal discussions while she was in the room.  On one such occasion she actually stopped cleaning, took out her field notes, and wrote for several minutes without them noticing that their ‘domestic’ was doing anything unusual (Rollins, 1997: 256).  I have described trust as an attitude we take towards another person.  Rollins’s employers were incapable of trusting her because her personhood was invisible to them.

That trust has a morally affirming feature is not surprising in light of the fact that people often speak of trustfulness as a character trait, and of character as an ethical notion.  Joel Kupperman defines character as one’s ‘normal pattern of thought and action, especially with respect to concerns and commitments in matters affecting the happiness of others or of [himself], and most especially in relation to moral choices’ (Kupperman, 1991: 17).  While a distrustful person might have a strong character, one which is resistant to temptations of various kinds, he cannot have a very good character.  The distrustful person limits the moral opportunities of those with whom he has contact, and his attitude towards others accuses them of ill will and/or incompetence.  Further, because distrust includes a pessimism about others, we tend to limit our dealings with those we distrust, thus denying them the opportunity to prove us wrong.  In this way, distrust becomes an intractable attitude without justification.  In a minimally decent society, it cannot be part of the life of a good person that he take up an attitude which is so morally discouraging to his fellow persons.

In fact, one must be at least moderately trustful in order to have a good character.  This becomes apparent in light of Kupperman’s arguments for the importance of character to moral life.  He claims that a person of strong character is more likely not only to display the kind of moral sensitivity which reveals the presence of moral problems and the contours of hard cases, but that the sense of commitment which a person of strong character must have can make virtuous conduct more likely to occur (Kupperman, 1991: 149). Like Walker, he forges connections between a person’s unique character and his moral choices, noting that some dilemmas, such as Sartre’s famous example of the young man who is torn between the needs of his mother and his country, are both unintelligible and unsolvable in terms of universal morality (Kupperman, 1991: 153-154).  A person who is at least moderately trustful has a characteristic tendency to take the self-presentation of others at face-value.  Without this, a person cuts himself off from the kinds of relationships which generate character-forming commitments.  According to Kupperman, commitments are not discrete obligations, but ways of being connected to other people, institutions, and causes that ‘amount to patterns of life’ (Kupperman, 1991: 153).

This connection between commitments and trust presents a way to connect trust to the good life, via Kupperman’s argument that having a strong (and moderately good) character is necessary for having a very good life.  Kupperman first makes the uncontroversial assumption that if a life is to be very good, it must be meaningful.  He then claims that only a person of strong character can have a meaningful life.  This is because a life without structural connections cannot be meaningful.  One way in which the parts of a life can be connected is through long term commitments and projects.  But such connections are much less likely to be sustained by persons whose loyalty to them is constantly threatened by changing circumstances, as in the case of people with weak character.  The other, more secure basis for meaning is strong character.  As Kupperman describes it, strong character is the ‘unifying thread’ running though the various episodes of a life.  Even when one’s character changes, continuities allow a life to hang together (Kupperman, 1991: 135-6).  It follows that there is a positive connection between having strong (and good) character and having a good life.  But if trust contributes to having a strong and good character, via the relationships and commitments it makes possible, then trust is an important part of a very good life.

In sum, trusting, both as self-trust and as trust in others, is a key moral value, whose importance can be specified independently of the value of trustworthiness. Of course, there can be too much trust or self-trust, and trust is not always morally preferable to distrust.  But we too often miss the moral value of trust when we focus - as does the strategic view - on the ways in which trust can go wrong.  As Baier argues, trust is morally important because it is required for producing moral agents.  But trust and self-trust are also essential for moral perception, deliberation and judgement.  Finally, trust is needed for having a good character, a prerequisite for having a very good life.[8]

References

Baier, A. (1994) Moral Prejudices. Cambridge, MA: Harvard University Press.

Brison, S. (1997) ‘Outliving Oneself: Trauma, Memory, and Personal Identity’, in Diana Tietjens Meyers (ed.) Feminists Rethink the Self.  Boulder, CO: Westview Press.

Dunn, J. (1988) ‘Trust and political agency’, in Diego Gambetta (ed.) Trust: Making and Breaking Co-operative Relations.  New York: Basil Blackwell.

 

Gambetta, D. (1988) ‘Can we trust trust?’ in Diego Gambetta (ed.) Trust: Making and Breaking Co-operative Relations.  New York: Basil Blackwell.

Govier, T. (1993) ‘Self-Ttust, autonomy, and self-esteem’, Hypatia. 8: 100-120.

Govier, T. (1997) Social Trust and Human Communities. Montreal and Kingston: McGill-Queen’s University Press.

Gratton, C. (1973) ‘Some aspects of the lived experience of interpersonal trust’. Humanitas.  9: 273-286.

Hardin, R. (1993) ‘The street-level epistemology of trust’.  Analyse und Kritik, 14 (1992): 152-76, reprinted in Politics and Society, 21: 505-29.   

Hardin, R. (1996) ‘Trustworthiness’, Ethics, 107: 26-42.

Jones, K. (1996) ‘Trust as an affective attitude’.  Ethics, 107: 4-25.

Kupperman, J. (1991) Character. Oxford: Oxford University Press.

Luhmann, N. (1980) Trust and Power (tr. H. Davies, J. Rafman and K. Rooney). London: John Wiley and Sons (original German edition, 1973).

Meyers, D. T. (1994) Subjection and Subjectivity.  New York: Routledge.

Oakley, J. (1993) Morality and the Emotions. New York: Routledge.

Rollins, J. (1997) ‘Invisibility, consciousness of the other, ressentiment’, in L. Lamphere, H. Ragone, and P. Zavella (eds) Situated Lives: Gender and Culture in Everyday Life. New York: Routledge.

Ruddick, S. (1983) ‘Maternal thinking’, in Joyce Treblicot (ed.) Mothering. Maryland: Rowman and Littlefield Publishers, Inc.

Walker, M. Urban (1987) ‘Moral particularity’, Metaphilosophy, 18: 171-185.

Walker, M. Urban (1998) Moral Understandings: A Feminist Study in Ethics. New York: Routledge.

Jessica Miller is an Assistant Professor of Philosophy at the University of Maine.  She specialises in ethical theory and applied ethics.  She is currently at work on a monograph on the ethics of interpersonal trust.

---