The Routledge Handbook of Trust and Philosophy

Judith Simon, Judith Simon

• Publisher: Routledge
• Year: 2020
• ISBN:
  9781134881673
• Edition: 1
• Topic: Law
• Subtopic: Estates & Trusts Law
• Index: Law

Trust is pervasive in our lives. Both our simplest actions – like buying a coffee, or crossing the street – as well as the functions of large collective institutions – like those of corporations and nation states – would not be possible without it. Yet only in the last several decades has trust started to receive focused attention from philosophers as a specific topic of investigation. The Routledge Handbook of Trust and Philosophy brings together 31 never-before published chapters, accessible for both students and researchers, created to cover the most salient topics in the various theories of trust. The Handbook is broken up into three sections:

I. What is Trust?

II. Whom to Trust?

III. Trust in Knowledge, Science, and Technology

The Handbook is preceded by a foreword by Maria Baghramian, an introduction by volume editor Judith Simon, and each chapter includes a bibliography and cross-references to other entries in the volume.

PART I

What is Trust?

1

Questioning Trust

Onora O’Neill

It is a cliché of our times that trust has declined, and widely asserted that this is a matter for regret and concern, and that we should seek to “restore trust.” Such claims need not and usually do not draw on research evidence. Insofar as they are evidence-based – quite often they are not – they are likely to reflect the findings of some of the innumerable polls and surveys of levels of trust that are commissioned by public authorities, political parties or corporations, or other organizations, are carried out by polling companies,¹ and whose findings are then published either by those who commissioned the polls, by the media or by interested parties.

Even when polls and surveys of public attitudes of trust and mistrust are technically adequate, the evidence they provide cannot show that there has been a decline in trust. That would also require robust comparisons with earlier evidence of public attitudes of trust and mistrust to the same issues – if available. And even when it is possible to make such comparisons, and they indicate that trust has declined, this may still not be a reason for seeking to “restore trust.” A low (or reduced) level of trust can provide a reason for seeking to “restore” trust only if there is also evidence that those who are mistrusted, or less trusted, are in fact trustworthy: and this is not generally easy to establish (see Medina, this volume).

In short, polls and surveys of attitudes or opinions do not generally provide evidence of the trustworthiness or the lack of trustworthiness of those about whom attitudes or opinions are expressed. Trust may be misplaced in liars and fraudsters, in those who are incompetent or misleading, and in those who are untrustworthy in countless other ways. Equally mistrust and suspicions may be misplaced in those who are trustworthy in the matters under consideration. Judgments of trustworthiness and of lack of trustworthiness matter greatly, but attitudinal evidence is not enough to establish them. Trustworthiness needs to be evidenced by establishing that agents and institutions are likely to address tasks and situations with reliable honesty and competence.

Evidence of attitudes is therefore not usually an adequate basis for claiming that others are or are not trustworthy in some matter. Yet such evidence is much sought after, and can be useful for various other purposes, including persuasion and reputation management. Here I shall first outline some of those other uses, and then suggest what further considerations are relevant for placing and refusing trust intelligently. Broadly speaking, my conclusion will be that doing so requires a combination of epistemic and practical judgment.

1.1 The Limits of Attitudinal Evidence

Investigations of trust that are based on opinion polls and surveys can provide evidence of respondents’ generic attitudes of trust or mistrust in the activity of types of institution (e.g. banks, companies, governments), or types of office-holder (e.g. teachers, scientists, journalists, politicians). Responses are taken as evidence of a trust level, which can be compared with trust levels accorded to other office-holders or institutions, and these comparisons can be tabulated to provide trust rankings for a range of types of institutions or office-holder at a given time. Repeated polling may also provide evidence of changes in trust levels or in trust rankings for types of institution or office-holder across time.

However, attitudinal evidence about trust levels and rankings has limitations. Most obviously, a decline or rise in reported levels of trust in specific types of institution or office-holder across time can be detected only by repeated polling, using the same or comparable methods. However, for the most, part polling was less assiduous and frequent in the past than it is today, so reliable comparisons between past and present trust levels and rankings are often not available. And where repeated and comparable polls have been conducted, and reliable comparisons are feasible, the evidence is not always that trust levels have declined, or indeed that trust rankings for specific types of institution and office-holder have changed. For example, in the UK journalists and politicians have (for the most part) received low trust rankings in attitudinal polls in the past and generally still do, while judges and nurses have received high trust rankings in polls in the past and generally still do.

Even where polling suggests that levels of trust have declined, the evidence they offer must be treated with caution. Polls and surveys collate evidence about informants’ generic attitudes to types of institution or to types of office-holder, but cannot show whether these attitudes are well-directed. But while the evidence that polls and surveys of trust levels provide cannot show who is or who is not trustworthy, trust rankings can be useful for some other quite specific purposes. I offer two examples.

One case in which trust rankings are useful is in summarizing consumer rankings of the quality and performance of standardized products or services, when these rankings are based on combining or tabulating the views of a wide range of consumers. For example, trust rankings of hotels or restaurants, of consumer durables or retail outlets, can be usefully informative because they are not mere expressions of generic attitudes. Such rankings reflect the experience of those who have bought and used (or tried to use!) a specific standardized product or service, so can provide relevant evidence for others who are considering buying or using the same product or service.

However, it is one thing to aggregate ratings of standardized products and services provided by those who have had some experience of them, and quite another to crowd-source views of matters that are not standardized, or to rely on ratings of standardized products or services that are provided by skewed (let alone gerrymandered) samples of respondents. Reputational metrics will not be a reliable guide to trustworthiness if the respondents whose attitudes are summarized are selected to favor or exclude certain responses (see Origgi, this volume).

A second context in which the attitudinal evidence established by polling can be useful is for persuading or influencing those who hold certain attitudes. Here attitudinal evidence is used not as a guide for those who are planning like purchases or choices, but for quite different (and sometimes highly manipulative) purposes.

Some uses of polling evidence are entirely reputable. For example, marketing departments may use attitudinal evidence to avoid wasting money or time cultivating those unlikely to purchase their products. Some are more dubious. For example, political parties may use evidence about political attitudes, including evidence about attitudes of trust in mistrust in specific policies and politicians, to identify which “demographics” they can most usefully cultivate, how they might do so, and where their efforts would be wasted. In some cases this is used to manipulate certain groups by targeting them with specific messages, including misinformation and disinformation.² Information about generic attitudes, and in particular about generic attitudes of trust and mistrust, can be useful to marketing departments, political parties and other campaigning organizations whether or not those attitudes are evidence-based because the evidence is not used to provide information for others with like interests or plans, but as a basis for influence or leverage. I shall not here discuss whether or how the use of polling results to target or orchestrate campaigning of various sorts may damage trust – including trust in democracy – although I see this as a matter of urgent importance in a digital world.³

1.2 Judging Trustworthiness

Once we distinguish the different purposes to which attitudinal evidence about trust levels can be put, we have reason to reject any unqualified claim that where trust is low (or has declined) it should be increased (or “restored”). In some situations increasing or restoring trust may be an improvement, and in others it will not. Nothing is gained by raising levels of trust or by seeking to restore (supposed) former levels of trust unless the relevant institutions or office-holders are actually trustworthy. Placing trust well matters because trustworthiness is a more fundamental concern.⁴

Seeking to gain (more) trust for institutions or individuals that are not trustworthy is more likely to compound harm. The point is well illustrated by the aptly named Mr. Madoff, who made off with many people’s money by running a highly successful Ponzi scheme that collapsed during the 2008 banking crisis. It would have been worse, not better, if Madoff had been more trusted, or trusted for longer – and better if he had been less trusted, by fewer people, for a shorter time, since fewer people would then have lost their money to his scam. Aiming to “restore” or increase trust will be pointless or damaging without evidence that doing so will match trust to levels of trustworthiness. By the same token, mistrust is not always damaging or irrational: it is entirely reasonable to mistrust the untrustworthy (see D’Cruz, this volume).

Aligning trust with trustworthiness, and mistrust with untrustworthiness is not simple. It requires intelligent judgment of others’ trustworthiness or untrustworthiness, and the available evidence may not reveal clearly which agents and which institutions are trustworthy in which matters. Typically we look for evidence of reliable competence and honesty in the relevant activities, and typically the evidence we find underdetermines judgments of trustworthiness or lack of trustworthiness.

There are, I think, two quite distinct reasons why this is the case. The first is that available evidence about trustworthiness may be epistemically complex and inconclusive. Judging any particular case will usually require two types of epistemic judgment. Judgment will be needed both to determine whether a given agent or institution is trustworthy or untrustworthy in some matter, and to interpret evidence that could be taken in a range of ways. Both determining (alternatively determinant, or subsumptive) and reflective judgment are indispensable in judging whether some institution or office-holder is trustworthy in some matter.⁵

But determinant and reflective judgment are only part of what is needed in judging trustworthiness and lack of trustworthiness. Judgments of trustworthiness also usually require practical judgment. Practical judgment is needed not in order to classify particular agents and institutions as trustworthy or untrustworthy, but to guide the placing or refusal of trust where evidence is incomplete, thereby shaping the world in some small part. Practical judgment is not judgment of a particular case, but judgment that guides action and helps to shape a new or emerging feature of some case.

Decisions to place or refuse trust often have to go beyond determining or interpreting existing evidence. For example, where some agent or agency has meager powers it may make sense to place more trust in them than is supported by available evidence or any interpretation of the available evidence, for example because the downside of getting it wrong would be trivial or because the experience of being trusted may influence the other party for the better. In other cases there may be reason to place less trust than the available evidence or any interpretation of that evidence suggests is warranted, for example because the costs of misplacing trust would be acutely damaging.⁶

Making practical judgments can be risky. If trust and mistrust are badly placed, the trustworthy may be mistrusted, and the untrustworthy trusted. Both mismatches matter. When we refuse to trust others who are in fact trustworthy we may worry and lose opportunities, not to mention friends and colleagues, by expressing suspicions and by intrusive monitoring of trustworthy people. Those who find their trustworthiness wrongly doubted or challenged may feel undermined or insulted, and become less sure whether the effort of being trustworthy is worthwhile. And when we mistakenly trust those who are in fact untrustworthy we may find our trust betrayed, and be harmed in various, sometimes serious, ways. So in placing and refusing trust intelligently we have to consider not only where the evidence points and where it is lacking and where interpretation is needed but also the costs and risks of placing and misplacing trust and mistrust. Judging trustworthiness is both epistemically and practically demanding.

1.3 Aligning Trust with Trustworthiness in Daily Life

The epistemic challenges of placing and refusing trust well despite incompleteness of evidence are unavoidable in daily life. A comparison with the equally daily task of forming reliable beliefs is suggestive. Both in scientific inquiry and in daily life we constantly have to reach beliefs on the basis of incomplete rather than conclusive evidence, and do so by seeking relevant evidence for specific claims and by accepting that further check and challenge to those beliefs may require us to change them. Both in institutional and in daily life we constantly have to place or refuse trust on the basis of incomplete rather than conclusive evidence of others’ trustworthiness.

However trust and mistrust can be placed intelligently by relying on relevant evidence about specific matters, by allowing for the possibility that further evidence may emerge and require reassessment, and also by addressing practical questions about the implications of trusting or refusing to trust in particular situations. In judging others’ trustworthiness we often need to consider a fairly limited range of specific evidence. If I want to work out whether a school can be trusted to teach mathematics well, whether a garage can be trusted to service my car, or whether a colleague can be trusted to respect confidentiality, I need to judge the trustworthiness of that particular school, garage or colleague in the relevant matter – and any generic attitudes that others hold about average or typical schools, garages or colleagues will be (at most) marginally re...