The Nature and Practice

of Trust
Marc A. Cohen
First published in 2023

ISBN: 978-1-032-41513-0 (hbk)

ISBN: 978-1-032-41515-4 (pbk)
ISBN: 978-1-003-35846-6 (ebk)

Chapter 6

Trust as an intrinsic good, moral reason

to trust

(CC BY-NC-ND 4.0)

DOI: 10.4324/9781003358466-9
6 Trust as an intrinsic good,
moral reason to trust

Chapter summary
This chapter argues that trust is an intrinsic good (that is, a good to be
pursued apart from material or instrumental beneft), and so we have moral
reason to trust.

Political scientist Margaret Levi wrote, “trust is neither normatively good

nor bad; it is neither a virtue nor a vice [to trust]” (1998, p. 81), and Eric
Uslaner (2002, p. 49) cites this comment approvingly—agreeing in the case
of particularized trust. The present chapter argues that Levi and Uslaner are
wrong. Instead, trust is an intrinsic good (in addition to, but separate from
the material benefts), so we have moral reason to trust.
The trust literature documents the instrumental reasons to trust with ref-
erence to the material benefts. Much of this work focuses on the material
benefts in the economic space and within business organizations: econo-
mists suggest that trust can reduce transaction costs and improve economic
outcomes for businesses, while management academics have shown that
trust within organizations improves communication, fosters creativity,
encourages teamwork, increases productivity, and increases employee sat-
isfaction. More generally, Diego Gambetta argues that trust is necessary
for cooperation across all aspects of life; in a widely cited passage (and one
quoted in an earlier chapter) he describes trust as vital in situations “from
marriage to economic development, from buying a second-hand car to inter-
national affairs, from the minutiae of social life to the continuation of life on
earth” (1988, pp. ix–x). And Peter Nannestad summarizes work showing
that generalized trust is essential for “good social, political, and economic
outcomes in society” (2008, p. 422).
The philosophy and management literatures widely use the term “rea-
sons to trust” in a second and different sense, to refer to beliefs and expec-
tations about a trustee’s likely behavior, which provide reason to think a
particular trustee is trustworthy. The management literature refers to these
beliefs and expectations as antecedents of trust, and much of that literature
applies the framework developed in Roger C. Mayer, James H. Davis, and
DOI: 10.4324/9781003358466-9
 Trust as an intrinsic good, moral reason to trust 117
F. David Schoorman’s (1995) paper, which separates the antecedents into
three categories—the trustee’s ability, benevolence, and integrity.
Others use the term “reasons to trust” in a third sense, to mean that
the risks involved are warranted given the potential beneft. Philosophers
describe trust as “rational” in this sense. But what counts as “rational”
will be context dependent and subjective: negative experiences in the past
might make one person feel that a particular trusting act is irrational, when
another might fnd that trusting action rational; or, in an emergency a per-
son might trust others because the options are worse, making that trusting
act rational because of the context. Much work on the rationality of trust
is guided by an impulse to rationalize (in the sense of explain away) or even
deny the vulnerability involved in a trusting act. And this impulse obscures
questions about how trustors manage the vulnerability. Guido Möllering’s
(2001, 2006) account—discussed in Chapter 1—describes the trustor’s pro-
cess in terms of suspension: when A trusts B to do X, A “brackets” or
“suspends” doubt, acting as if there is no risk associated with B’s actions.
In some cases, Möllering suggests, this bracketing could be a matter of habit
and routine. And Chapter 1 argued that Möllering’s account is too narrow
because there are also cases in which A accepts vulnerability rather than
bracketing it in the sense just mentioned.
The point here—that trust is an intrinsic good, that we have moral rea-
son to trust—is absent from the literature on the material benefts and the
instrumental decision to trust. To be sure, there could be moral reason to
trust in particular contexts, if the benefts of trust themselves have some sort
of moral value: perhaps a person ought to trust his or her spouse to foster
a positive relationship, or one should trust to foster cooperation. But the
claim here is different: we have moral reason to trust across cases because
of the intrinsic good involved.
Levi can’t see this because she adopts an expectation-based conception
of trust; for her, A trusting B to do X means that A expects B to do X. The
(perhaps) most prominent example of such an account is in the management
literature, from Mayer, Davis, and Schoorman (1995), who—as a reminder
here—characterize trust as “the willingness of a party to be vulnerable to
the actions of another party based on the expectation that the other will
perform a particular action important to the trustor, irrespective of the
[trustor’s] ability to monitor or control that other party” (p. 712). And,
both accounts—Mayer, Davis, and Schoorman’s and Levi’s—are exten-
sions of Diego Gambetta’s thinking, which characterized trust in terms of
probabilities:

When we say we trust someone or that someone is trustworthy, we

implicitly mean that the probability that he will perform an action that
is benefcial or at least not detrimental to us is high enough for us to
consider engaging in some form of cooperation with him.
(1988, p. 217)
118 The practice of trust
There is a difference in the antecedents on the two views: for Levi, A expects
B to do X if doing X is in A’s own interest (this is Russell Hardin’s view
of trust in terms of “encapsulated interest,” Levi cites Hardin 1993; more
generally see Hardin 2006); where, for Mayer, Davis, and Schoorman, (as
mentioned earlier) A’s expectation depends on A’s assessment of B’s com-
petence, integrity, and benevolence. And other kinds of antecedents could
extend Mayer, Davis, and Schoorman’s account: for example, A’s expec-
tations about B’s doing (or not doing) X might involve emotional factors
rather than expectations or beliefs, in line with Karen Jones’s (1996) think-
ing. Or shared identity could foster positive expectations and therefore trust
(with Coleman 1990 and Kuwabara et al. 2007).
But, regardless of the antecedents, on the expectation-based conception
of trust, A will trust B and act if the risks involved are acceptable given the
potential benefts. Trust so-conceived is a matter of instrumental or, equiva-
lently, cost-beneft decision making. For example—and this is a standard,
schematic example—business A needs a supplier, B, to deliver a product by
a certain time and at a certain price (do X). A might trust B to do so, arrang-
ing the transaction without contractual safeguards or even without a con-
tract, and also without monitoring, because the two frms have a long-term
relationship (suggesting that the risk is low, the vulnerability is limited), in
order to reduce costs and speed up the process (the beneft). (This example
is presented from A’s perspective; note that business B also trusts A in agree-
ing to proceed without a contract, and business B also benefts from reduced
costs and a faster process.) There is no normative dimension and, with Levi,
virtue and vice are not relevant.
But if we see that trust relationships are given structure by commitments
and obligations that bind the parties, on the account developed in Chapters
1 and 2, then the situation is different.
Summarizing material presented in preceding chapters only briefy, on
the commitment conception of trust, when A trusts B to do X, A relies on
B’s commitment to do so. In particularized trust (as opposed to generalized
trust) the commitment is not derived from general or background moral
obligations. The commitment could be explicit (the prototypical case is one
in which B makes a promise and A relies on that promise) or it could be
implicit (the commitment could be embedded in roles, doctors for example
make a role-based commitment to exercise due care for patients). And there
could be signifcant ambiguity when those commitments are implicit; trust is
misplaced (in one sense) when there is confusion and/or disagreement about
the commitments. Note, though, this ambiguity and confusion is a common
problem in actual trust relationships; the commitment account refects this
problem, it doesn’t create the ambiguity and confusion.
Chapter 1 described the process of forming trust relationships (correctly
understood with reference to commitments) in terms of invitations: when
A wants to trust B to do X, A invites B to acknowledge and accept an
obligation to do so. When (or if) B accepts the invitation, B takes on that
 Trust as an intrinsic good, moral reason to trust 119
obligation, and then A can rely on B’s commitment with respect to that par-
ticular obligation. Trust-invitations can occur in the context of established
relationships, creating new obligations and, in that way, deepen and trans-
form the relationship. Or trust-invitations can establish obligations when
there is no pre-existing relationship. The invitation could also be initiated
by B: B might invite A to rely on B’s commitment to do X. Either way,
trust-invitations seek out a moral relationship (one given structure by com-
mitments and obligations), and trust-acceptance has a moral effect—after
trust-acceptance B owes the action X to A.
Understanding the process in these terms—as a description of what we
do when we trust—enables us to see and explain the moral reason to trust
with reference to the intrinsic good involved. Making this point is a key
motivation for the present project as a whole.
When A trusts B to do X, A relies on B’s commitment to do so. Apart from
the instrumental benefts, A’s relying on B to fulfll a commitment manifests
respect for B as a moral agent, him- or her-self capable of being motivated
by commitments to others and so acting on those commitments. Put another
way, relying on another’s commitment and making oneself vulnerable mani-
fests respect because it treats the other as a person who can “recogniz[e] and
be responsive to our deliberately assumed vulnerability to their discretion-
ary powers” in making decisions about how to act (McGeer 2008, p. 248,
though she doesn’t talk about respect).1 There are other ways to talk about
respect here: person A deciding to trust B depends (typically) on A’s think-
ing that B is both capable of doing X and also trustworthy, where the sec-
ond is a matter of B’s integrity; so A’s decision to trust respects—meaning,
involves positive assessment of—both B’s abilities and B’s integrity. Some
talk of “honoring” others with this kind of positive assessment in the deci-
sion to trust (e.g., O’Neil 2012, p. 311), trust can be a signal of this positive
assessment. But the point that trusting others manifests respect and so is an
intrinsic good involves the specifc kind of respect just described: the trust-
ing party treats B as a moral agent. To trust is to respect in this sense and
is more than a (mere) signal of positive regard. And talk of respect in these
particular terms is not possible on expectation-based conceptions of trust:
on those conceptions, when A trusts B to do X, A only (merely) assesses
probabilities (B’s acting or not), vulnerabilities, and benefts—and then acts.
There is no relationship between A and B on such accounts.2
But the point is not that we have a moral duty to trust all others. Some
aren’t trustworthy, either because they lack the ability to perform the trusted
action (no one should trust the present author to fx their car because he
lacks the required knowledge), or because they’ve shown themselves to be
untrustworthy—unreliable, even dishonest (maybe a mechanic overcharged
me the last time my car needed new brakes). Here instrumental considera-
tions—questions about whether, and to what extent, a particular person
can be trusted—act as constraints on the decision to trust, constraints on
both the person(s) being trusted and the particular Xs involved. So, we have
120 The practice of trust
moral reason to trust others as a matter of respect (the intrinsic good), unless
other considerations show this to be too dangerous. And we could say, fur-
ther, that those who demonstrate untrustworthiness disqualify themselves
from the sort of respect involved here.
Separate from the moral reason to trust, we have moral reason to avoid
distrust for the wrong kinds of reasons.
In a recent paper, Jason D’Cruz (2019) also describes trust as “signaling”
respect, but his central concern is the disrespect involved in distrust and,
in particular, the pernicious effects of misplaced distrust. As mentioned, A
might distrust B because B lacks the ability to do X and/or because A thinks
B not trustworthy—and so A is not willing to rely on B’s commitment. And
these two components can come apart: A might think B has the requisite
ability but B’s commitments aren’t reliable; or A might have trusted B on
a number of occasions and might want to trust B, but doubt B’s ability for
some particular X. In both cases A will distrust B. But these points take A’s
assessment at face value, as empirical assessment informing deliberation.
D’Cruz is concerned with cases of distrust based on inaccurate and preju-
diced “construals” of the other person as incompetent, lacking integrity,
and/or threatening; he is concerned with cases of distrust as the product
of immediate assessment of other persons as different. We can miss these
kinds of cases because our accounts of trust and distrust treat the decision-
making involved in rational terms, as rational deliberation. But such cases
disrespect, and—with D’Cruz—we have an obligation to be skeptical of our
own attitudes; our trust/distrust practices, D’Cruz says, should “manifest[]
a spirit of skepticism, curiosity, and moral commitment,” and in particular
“skepticism about the warrant of one’s own felt attitudes of trust and dis-
trust” (p. 947).
This summary of D’Cruz’s thinking and the point about disrespect is per-
haps too dry and academic, and as such it might fail to capture the emotional
and social harm of systematic bias and distrust. One of D’Cruz’s examples
is drawn from Barak Obama’s memoir, Dreams from My Father (2007).
Obama describes his being shaped as a young black man by a “ledger of
slights,” including one in which

the older woman in my grandparents’ apartment building… became

agitated when I got on the elevator behind her and ran to tell the man-
ager that I was following her; [and her] refusal to apologize when she
was told that I lived in the building.
(Obama 2007, p. 80, cited in D’Cruz 2019, p. 942)

Obama describes a bewildering “obtuseness,” “as if whites didn’t know

they were being cruel in the frst place” (ibid.). Examples like this suggest
that, with D’Cruz’s point, that we have obligation to be skeptical of our
own trust/distrust practices—we have an obligation to avoid unwarranted
distrust as a way of avoiding disrespect.
 Trust as an intrinsic good, moral reason to trust 121
We might also think about the moral reason to trust and the intrinsic
good involved in another way.
The moral community can be defned as the set of persons willing to rely
on one another’s commitments. When A trusts B to do X, A relies on B’s
commitment, and A’s doing so locates (or puts) B in that moral community.
Or we could say that A’s trusting treats B as a member of that community.
(There may be other ways to think about the moral community, different
forms of moral community; this is one sort.) The process involved in form-
ing the moral community (as just defned) can be explained in terms of trust-
invitations: when A makes a trust-invitation to B, A invites B to join that
community, or if the two have a long-term relationship, then A reaffrms B’s
place in that community. Including others in this community is the social
form (or the social expression) of the respect just described, and so again we
have moral reason to trust as a manifestation of respect (the intrinsic good).
Moreover, so-including others would give them standing in the moral com-
munity (or acknowledge that standing) and, in that way, would provide
what John Rawls (2001) calls the social bases of self-respect. To distrust
another is to be unwilling to rely on that other’s commitments and so to
exclude that other from this moral community. Being so-excluded denies
social self-respect, and so—with D’Cruz—when this denial is grounded in
the wrong kind of reasons, exclusion will “marginalize and exclude indi-
viduals who have done nothing that would justify their marginalization or
exclusion” (2019, p. 933). Again, we have moral reason to avoid unwar-
ranted distrust.
There is widespread intuitive commitment to the idea that trust is impor-
tant beyond the instrumental value, this chapter explains that intuition with
reference to the moral community and to the manifestation of respect in trust.3

Notes
1 Victoria McGeer’s account of therapeutic trust—discussed in the next chapter—
focuses on the way A trusting B can empower B, fostering B’s ability to recog-
nize and respond to vulnerability. She suggests, we have reason to trust others
as a way of “stimulating their [the trusted parties’] agential capacities to think
and act in trust-responsive ways” (p. 242), without weighing evidence about
the trusted parties’ trustworthiness. (The usual example is a parent trusting a
teenager with the family car. McGeer thinks hope is involved in motivating such
trust.) So McGeer would say that to trust is to foster the potential for such
capacities in others, the point here is more general: A trusting B recognizes those
capacities in others and so is to respect.
2 This point—that there is no trust relationship on expectation-based accounts—
might be surprising. But if, with Mayer, Davis, and Schoorman, when A trusts
B, A is willing to be vulnerable to the actions of B based on A’s expectation that
B will perform a particular action important to A, irrespective of A’s ability to
monitor or control B, there’s nothing for B to do; A assesses B’s ability and char-
acter, forms an expectation, and acts. We might wonder if B even knows he or
she is involved. In the philosophy literature Philip Nickel discusses an example
in which A “hopes that [B] will realize he is being trusted” (2007, p. 317), also
122 The practice of trust
without seeing this one-sidedness as a problem. And on Baier’s (1986) widely
discussed account, A trusts B when A relies on B’s goodwill; but still, B’s goodwill
is something for A to assess, there is no process for A coming to trust B, there’s
no actual relationship between A and B, and it could come as a surprise to B that
A relies on B’s goodwill. The same gap is present if we think of trust as a decision
or an action (as opposed to an attitude), if A trusts B to do X when A decides to
act (or when A acts) on expectations that B will do X. Again, the process here
involves A’s assessment of B’s trustworthiness along with the potential benefts
and risks. B plays no role.
This point—that B isn’t involved, the process of establishing a trust relation-
ship is entirely one-sided—is another reason to reject expectation-based accounts
and adopt the commitment conception. Put the other way, if expectation-based
conceptions of trust allow for one-sided cases, they are incomplete in an impor-
tant way. But if trust involves commitments, then there must be a process in place
for A coming to trust B to do X, and that process must involve B in some way. As
noted, the process could be implicit, but the prototypical case is one in which B
makes a promise to A and then A acts, relying on that promise; the implicit cases
are ones in which B makes a commitment by adopting a role, joining a group,
etc. Examples were presented in Chapters 1 and 2. To be sure, a trust relation-
ship could be transactional: A might trust B to do X once, with no plans for
future interaction of any sort; think of a temporary, mutually benefcial business
partnership. But we can still talk of “relationships” here in the sense that B has
obligations to A, and B had to be involved in putting the obligation(s) in place.
3 The paragraph at the beginning of this chapter mentioned Eric Uslaner’s (2002)
book because he cites Levi’s comment approvingly. Uslaner distinguishes between
particularized trust and generalized trust, and at points uses the terms “moralis-
tic trust” and “generalized trust” as synonyms. When he agrees with Levi about
trust not being a virtue or a vice, it is only with respect to particularized trust—
because Uslaner thinks that particularized trust is necessarily strategic, meaning
instrumental.
Three further points are important here. First, even though Uslaner uses the
term “moralistic trust,” he is not arguing that we have moral reason to trust.
That is the point most relevant in connection with the present chapter. To be
sure, in the frst (conceptual) part of his book (the introduction and frst two
chapters) Uslaner makes and repeats a number of claims with normative dimen-
sions—trust in others (strangers) is based on an “ethical assumption” that others
share fundamental values; “people who trust others will seek to better the lives
of those who have less”; we trust others “when we perceive a shared fate”; and
so on (all p. 2). Some of this is supported by the empirical work that makes
up the majority of his book, especially the frst point about optimism. So, with
Uslaner: “Moralistic trust [trust in everyone including strangers] is predicated
upon a view that the world is a benevolent place with good people, that things
are going to get better, and that you are the master of your own fate” (p. 23). But
none of this amounts to saying that we have moral reason to trust. And, others
have shown—empirically—that generalized trust in strangers is related to one’s
experience with strangers, so evidence does play a role in generalized trust, con-
tra Uslaner’s narrower understanding (see the references to Markus Freitag and
Richard Traunmüller’s work in Chapter 2).
Second, the conceptual part of Uslaner’s book is frustrating, at points madden-
ing. One example (among many): Uslaner writes, “Moralistic trust is the belief
that others share your fundamental moral values” (p. 18), and on the same page
he also explains that “Placing trust in others [moralistic trust] does not require
agreement on specifc issues or even philosophies. Instead, it is a statement of
toleration of different ideas” (p. 18).
 Trust as an intrinsic good, moral reason to trust 123
Uslaner’s conceptual account of generalized trust as “moralistic trust” is par-
ticularly problematic. He sees particularized trust in in-group members and gen-
eralized/moralistic trust in all persons as fundamentally incompatible: for him,
particularized trust is (as just mentioned) necessarily strategic while generalized/
moralistic trust can’t be strategic, because one can’t trust strangers for instru-
mental reasons on the basis of evidence about their trustworthiness—because
they are strangers. So, for Uslaner a person is either a particularized trustor or
a generalized/moralistic trustor. We should reject this claim for a number of
reasons. Most directly, there is no conceptual reason a person couldn’t trust
in-group members for strategic reasons while also trusting strangers without
evidence, not for strategic reasons. Also, as just mentioned, empirical research
shows that a person’s willingness to trust strangers depends in part on that per-
son’s past experience with trusting in-group members and strangers, so evidence
is relevant in those cases. Separately, Uslaner wants to draw out a distinction
between evidence-based trust in in-group members and trust in all (including
strangers) grounded on an optimistic world view, and he can do this, but the lat-
ter is different from what others refer to under the heading of “generalized trust.”
So at best it’s not clear how to map Uslaner’s thinking onto the usual categories.
That said, third, Uslaner’s empirical analysis of the factors supporting gen-
eralized trust is very interesting, especially when he presents data at odds with
Robert Putnam’s work.

References
Baier, A. 1986. “Trust and antitrust.” Ethics 96, no. 2: 231–260.
Coleman, J.S. 1990. Foundations of Social Theory. Cambridge, MA: Belknap Press.
D’Cruz, J. 2019. “Humble trust.” Philosophical Studies 176, no. 4: 933–953.
Gambetta, D. 1988. “Can we trust trust?” In Trust: Making and Breaking
Cooperative Relations, edited by D. Gambetta, 213–237. New York: Basil
Blackwell.
Hardin, R. 1993. “The street level epistemology of trust.” Politics and Society 21,
no. 4: 505–529.
Hardin, R. 2006. Trust. Malden, MA: Polity Press.
Jones, K. 1996. “Trust as an affective attitude.” Ethics 107, no. 1: 4–25.
Kuwabara, K., et al. 2007. “Culture, identity, and structure in social exchange: A
web-based trust experiment in the United States and Japan.” Social Psychology
Quarterly 70, no. 4: 461–479.
Levi, M. 1998. “A state of trust.” In Trust and Governance, edited by M. Levi and
V. Braithwaite, 77–101. New York: Russell Sage Foundation.
Mayer, R.C., J.H. Davis, and F.D. Schoorman. 1995. “An integrative model of
organizational trust.” Academy of Management Review 20, no. 3: 709–734.
McGeer, V. 2008. “Trust, hope, and empowerment.” Australasian Journal of
Philosophy 86, no. 2: 237–254.
Möllering, G. 2001. “The nature of trust: From Georg Simmel to a theory of
expectation, interpretation and suspension.” Sociology 35, no. 2: 403–420.
Möllering, G. 2006. Trust: Reason, Routine, Refexivity. Oxford: Elsevier.
Nannestad, P. 2008. “What have we learned about generalized trust, if anything?”
Annual Review of Political Science 11: 413–436.
Nickel, P.J. 2007. “Trust and obligation-ascription.” Ethical Theory and Moral
Practice 10, no. 3: 309–319.
124 The practice of trust
Obama, B. 2007. Dreams from My father: A Story of Race and Inheritance. New
York: Crown.
O’Neil, C. 2012. “Lying, trust, and gratitude.” Philosophy and Public Affairs 40,
no. 4: 301–333.
Rawls, J. 2001. Justice as Fairness: A Restatement. Cambridge, MA: Harvard
University Press.
Uslaner, E.M. 2002. The Moral Foundations of Trust. Cambridge: Cambridge
University Press.