On Trust and Trustworthiness
Sophie Vivian
A dissertation presented for the degree of Doctor of Philosophy
6 December 2019
School of Humanities and Languages The University of New South Wales, Sydney, NSW 2052, Australia Abstract
There has been a long-standing tendency in both the philosophical and non- philosophical literature in the English-speaking world to view trust and trustworthiness primarily as three-place relations. Trust is: A trusts B to do X, and trustworthiness is: B doing X for A. This three-place schematic reflects English language use regarding the word “trust” as well as the widely-held beliefs that: 1) cooperation and knowledge acquisition are normally underpinned by trust and trustworthiness, 2) trust is a ubiquitous cognitive expectation classifiable as a distinct kind of reliance, and 3) trustworthiness is the ubiquitous fulfilment of that expectation classifiable as a distinct kind of reliability (often moralised-reliability). In this dissertation, it is argued that trust and trustworthiness are not three-place relations, and that thinking of them in this manner distorts the role they play in our lives. Instead, this dissertation proposes that trust and trustworthiness are each one-half of a two-place relation, and emphasises the special, private, complex, and interactive nature of this relation, rather than the behaviours (cooperation/reliance etc.) and expectations (belief/obligation etc.) often associated with it. The classification of trust and trustworthiness as being non-cognitive and non-moralised is essential, since a definitive feature of trust and trustworthiness is their remarkable capacity for non-conformity when it comes to the rules governing areas of our lives such as: social-communication, normative-behaviour, moral reasoning and moral action, truth, knowledge, rationality and agency. Acknowledging this anomalous rule-breaking facet of our humanity, which is similar to love, facilitates a better understanding of human behaviour and it allows us to divorce important questions about trust and trustworthiness from other distinct questions in epistemology and ethics. For David, for showing me trust and making me worthy. Acknowledgements
Firstly, I would like to acknowledge my supervisor Michaelis Michael, for providing sound advice and considered guidance throughout my candidature. I would also like to thank the supportive staff in the Philosophy department at UNSW Sydney. In particular, my co-supervisor Melissa Merritt for introducing me to Iris Murdoch, and Markos Valaris for his role in sparking my initial interest in philosophy. I am grateful for the instruction provided by Professor Tony Attwood from Griffith University Queensland, who generously took the time to share his expertise in Theory of Mind with me. Both Paul Faulkner of The University of Sheffield and Richard Holton of Cambridge University kindly spared their time to answer my emails, and for this, I thank them. I am indebted to Dr. Kirsty McKenzie, my non-philosopher mentor; our conversations over the years were invaluable. Lastly, a (quiet) shout-out to the autistic community. In particular, the autistic academics: Dr Damian Milton, Dr Melanie Yergeau, and Dr Elena Chandler, as well as the advocates: Shona Davis, Kieran Rose, and Sara Harvey, and authors: Yenn Perkis, Laura James, and Katherine May. Along with countless others, these individuals gently but determinedly give their time to share knowledge and resources with anyone willing to listen and learn. On a personal note, the irreplaceable individuals who supported all the other aspects of my life as I undertook this dissertation – my parents Lainie and Tony, my children Margaret and Hume, my dear friend Hamish, and my David – provided me with the love and trust, and so the expertise, that it really took to write this dissertation. Thank you all so very, very much.
iv Thesis/Dissertation Sheet
Surname/Family Name : Vivian Given Name/s : Sophie Abbreviation for degree as give in the University calendar : PhD (Philosophy) Faculty : Arts & Social Sciences School : Humanities & Languages Thesis Title : On Trust and Trustworthiness
Abstract 350 words maximum: There has been a long-standing tendency in both the philosophical and nonphilosophical literature in the English-speaking world to view trust and trustworthiness primarily as three-place relations. Trust is: A trusts B to do X, and trustworthiness is: B doing X for A. This three-place schematic reflects English language use regarding the word “trust” as well as the widely-held beliefs that: 1) cooperation and knowledge acquisition are normally underpinned by trust and trustworthiness, 2) trust is a ubiquitous cognitive expectation classifiable as a distinct kind of reliance, and 3) trustworthiness is the ubiquitous fulfilment of that expectation classifiable as a distinct kind of reliability (often moralised-reliability). In this dissertation, it is argued that trust and trustworthiness are not three-place relations, and that thinking of them in this manner distorts the role they play in our lives. Instead, this dissertation proposes that trust and trustworthiness are each one-half of a two-place relation, and emphasises the special, private, complex, and interactive nature of this relation, rather than the behaviours (cooperation/reliance etc.) and expectations (belief/obligation etc.) often associated with it. The classification of trust and trustworthiness as being non-cognitive and non-moralised is essential, since a definitive feature of trust and trustworthiness is their remarkable capacity for non- conformity when it comes to the rules governing areas of our lives such as: social-communication, normative-behaviour, moral reasoning and moral action, truth, knowledge, rationality and agency. Acknowledging this anomalous rule-breaking facet of our humanity, which is similar to love, facilitates a better understanding of human behaviour and it allows us to divorce important questions about trust and trustworthiness from other distinct questions in epistemology and ethics.
Declaration relating to disposition of project thesis/dissertation
I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or in part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all property rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation. I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstracts International.
06/12/2019 ………………………….………………… ………………………….…………………. …………………………………………… Signature Witness Signature Date The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research.
FOR OFFICE USE ONLY Date of completion of requirements for Award:
v Inclusion of Publications Statement
UNSW is supportive of candidates publishing their research results during their candidature as detailed in the UNSW Thesis Examination Procedure. Publications can be used in their thesis in lieu of a Chapter if: • The student contributed greater than 50% of the content in the publication and is the “primary author”, ie. the student was responsible primarily for the planning, execution and preparation of the work for publication • The student has approval to include the publication in their thesis in lieu of a Chapter from their supervisor and Postgraduate Coordinator. • The publication is not subject to any obligations or contractual agreements with a third party that would constrain its inclusion in the thesis
Please indicate whether this thesis contains published material or not.
☒ This thesis contains no publications, either published or submitted for publication
Some of the work described in this thesis has been published and it has been documented in the relevant Chapters with ☐ acknowledgement
This thesis has publications (either published or submitted for publication) incorporated into it in lieu of a chapter ☐ and the details are presented below
CANDIDATE’S DECLARATION
I declare that I have complied with the Thesis Examination Procedure where I have used a publication in lieu of a Chapter, the listed publication(s) below meet(s) the requirements to be included in the thesis.
Name Signature Date (dd/mm/yy) Sophie Vivian 06/12/2019
Postgraduate Coordinator’s Declaration
I declare that the information below is accurate where listed publication(s) have been used in lieu of Chapter(s), their use complies with the Thesis Examination Procedure the minimum requirements for the format of the thesis have been met.
PGC’s Name PGC’s Signature Date (dd/mm/yy)
vi Originality Statement
‘I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.’
Signed ………………………………………………… Date ………………...……06/12/2019
vii Copyright Statement
‘I hereby grant the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the university libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all proprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation. I also authorise university Microfilms to use the 350 word abstract of my thesis in Dissertation Abstract International (this is applicable to doctoral theses only). I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted I have applied/will apply for a partial restriction of the digital copy of my thesis or dissertation.'
Signed …………………………………………………. Date ………………...……06/12/2019
Authenticity Statement
‘I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the con conversion to digital format.’
Signed …………………………………………………… Date ………………...…06/12/2019
viii Contents
On Trust and Trustworthiness i Abstract ii Acknowledgements iv Inclusion of Publications Statement vi Originality Statement vii Copyright Statement viii Authenticity Statement viii Contents ix
Introduction 1 Primer: On Trust and Trustworthiness 1 Outline: On Trust and Trustworthiness 7
10 On Trust 10 Overview of Part I, On Trust 10 I.1 Trust: in the Philosophical Literature 12 I.1.1 One-place trust 13 I.1.2 Three-place trust 20 I.1.3 Trust and rationality 25 I.1.4 Voluntariness of trust 33 I.1.5 Summary of Trust: in the Philosophical Literature 37 I.2 Dominant Views on Trust: Problems 39 I.2.1 Three-place trust: logic 39 I.2.2 Three-place trust: simplistic 41 I.2.3 Three-place trust: plus 49 I.2.4 Three-place trust explained 52 I.2.5 Summary of Dominant Views on Trust: Problems 58 I.3 The Object of Trust 60 I.3.1 The object of three-place trust: X 61 I.3.2 The object of three-place trust: B 65 I.3.3 The proper object of trust 71 I.4 Trust 84 I.4.1 Middlemarch 85 I.4.2 Trust and love 92 I.4.3 Trust in language 100 I.4.4 Trusting and monitoring 106 I.4.5 Self-trust 110 I.4.6 Deciding to trust 114 Summary of Part I, On Trust 118 ix Contents x
120 On Trustworthiness 120 Overview of Part II, On Trustworthiness 121 II.1 Trustworthiness: in the Philosophical Literature 123 II.1.1 Symmetricity of trust and trustworthiness 124 II.1.2 One-place trustworthiness 127 II.1.3 Three-place trustworthiness 136 II.1.4 Motives and trustworthiness 146 II.1.5 Summary of the literature on trustworthiness 155 II.2 Dominant Views on Trustworthiness: Problems 157 II.2.1 One-place trustworthiness: virtue 157 II.2.2 Three-place trustworthiness: scenarios 165 II.2.3 Three-place scenarios: discussion 172 II.2.4 Summary of dominant views on trustworthiness: problems 180 II.3 Rich Trustworthiness or Poor Trustworthiness? 181 II.3.1 Jones’ rich trustworthiness (rich reliability) 182 II.3.2 The ethical problem with rich trustworthiness 187 II.3.3 Summary of rich trustworthiness 200 II.4 Trustworthiness 201 II.4.1 The Invention of Lying 202 II.4.2 Mindreading and deception 208 II.4.3 A potential charge of hypocrisy 212 II.4.4 Løgstrup and trust/worthiness 215 II.4.5 The freedom to look again 222 Summary of Part II, On Trustworthiness 230
Conclusion 231 Summary: On Trust and Trustworthiness 231 Further Research: On Trust and Trustworthiness 234 Closing Remarks: On Trust and Trustworthiness 235 Bibliography 238
Love… faith… whatever word you want to use, we have to see that when we meet, we meet on a narrow ridge, and on one side is an abyss called individuality and if you fall into that abyss everyone you meet is only either useful to you, or, makes you feel better. And the other abyss on the other side is what I would call ‘collectivity’. It looks like community, but it’s really based on ‘we hate those people’ so it’s quite deceptive. And I think of the two, the collectivity is more lonely even than the individuality. But you know, on this narrow ridge we meet one another and that’s how love works. That’s how trust works… –Graham Long Wayside Chapel, Kings Cross Sydney
xi
Introduction
… fashion still demands that the real complexity shall go unexamined –Mary Midgely
This dissertation is a philosophical exploration of trust and trustworthiness as they exist in human relationships. It is in two symmetrical parts: Part I: On Trust and Part II: On Trustworthiness. The primary aim of this dissertation is to overthrow the current dominant views of trust and trustworthiness in philosophy: namely, that trust is a ubiquitous three-place epistemic relation that enhances our agency (similar to reliance or belief) and trustworthiness is its one/three-place ethical counterpart (similar to reliability or virtue). Instead, this dissertation suggests that trust and trustworthiness are best seen as two-place relations. A secondary aim of this research project is to characterise this two-place relation, by emphasising its inherent complexity and resistance to exacting definition. This introduction proceeds with a brief Primer on Trust and Trustworthiness to set out the historical and contemporary significance of The Problem of Trust, followed by an Outline on Trust and Trustworthiness, which sets out the structure of this dissertation proceeding.
Primer: On Trust and Trustworthiness
Discussion of trust in philosophy is not new. Plato, in his dialogue Charmides, asks us to consider how a layperson can determine the experts from the charlatans (380BC: 153a-76d). Famously, Plato was not optimistic, and the dialogue ends in aporia. Today, Plato’s concern with how non-experts could be fit to judge the assertions of experts has become widely regarded as ‘The Problem of Trust’ and discussion of “the problem” appears in many disparate areas of philosophical inquiry, in the context of a wider epistemological project of accounting for the transfer of knowledge. The conceptual link between trust and testimony as it exists in philosophy today, for example, is apparent in the Stanford Encyclopedia of Philosophy, which in 2006 included for the first time both an entry on Trust, and an entry on the Epistemological Problems of
1 Introduction Primer: On Trust and Trustworthiness 2
Testimony, each citing over 100 contemporary philosophers. Putting the bibliographies for these entries side-by-side, one is struck by the close commonality. The consequences of treating trust as a form of reliance on another person, or acceptance of their testimony, i.e. I trust you to show me the way, I trust you to tell me the truth etc., have been significant. The view is now common, for example, that it is not only knowledge gained from others (testimony) that requires trust, but that all knowledge (including self-knowledge) requires trust. ‘Trust’, in this sense, is as Niklas Luhmann describes:
confidence in one’s expectations... In many situations, of course, man can choose in certain respects whether or not to bestow trust. But a complete absence of trust would prevent him even from getting up in the morning. He would be prey to a vague sense of dread, to paralyzing fears. He would not even be capable of formulating distrust and making that a basis for precautionary measures, since this would presuppose trust in other directions. Anything and everything would be possible. Such abrupt confrontation with the complexity of the world at its most extreme is beyond human endurance. (1979: 4)
The view Luhmann is describing – trust as a phenomenon that is part of all our daily actions and interactions, so ubiquitous that its absence is “beyond human endurance” and would entirely incapacitate us – is present in the writings of many philosophers. According to Derrida:
elementary trust … is involved … in every address of the other. From the very first instant it is co-extensive with this other and thus conditions every “social bond”, every questioning, all knowledge. [my emphasis] (1998: 63)
As a ubiquitous pervasive cognitive expectation, trust functions as both a cause of, and potential solution to, what is problematic about all knowledge, and so agency, and so any attempt at collective agency. It is involved in every part of our epistemic and practical decision-making. Trust, in this sense, is what makes us agents in the first place and what enables us to extend our agency, and so in philosophy today it is as David Owens summarises:
Introduction Primer: On Trust and Trustworthiness 3
for Hume the psychological cement of human society was sympathy; for Hobbes it was fear. For many recent writers, both popular and philosophical, trust plays this foundational role. (2017: 214)
As a current example of trust playing this foundational role, Bernd Lahno (2017) claims that trust is what makes collective agency possible, because it is what allows us to overcome the inherent risk involved in being collective agents. To make this claim is to make trust an action that we take – an inherently risky action – that is rational only if one is in possession of certain knowledge:
Trusting behaviour is characterised by the risk that another person may act in undesired ways. A rational individual willingly incurs such risk only if she has reason to believe that others will respond cooperatively. (2017: 113)
For Lahno, trust is a cognitive rational expectation that the trustee will behave cooperatively, and so trustworthiness is behaving cooperatively in line with expectations (2017: 113-29). This conception leads trust to be regarded as a response to uncertainty about other people’s behaviour, and so rational only when we have reason to believe that others are worthy of our trust (i.e. will behave as expected). This important feature of collective agency is taken to be trust in many areas of academic inquiry that seek to understand and explain collective agency, i.e. social and political science, as Katherine Hawley describes:
trust and trustworthiness in collective contexts are widely discussed by social scientists. For example, in international relations there is debate about trust and distrust between states, other organizations or groups, and individual representatives of those collectives (e.g. Booth and Wheeler 2008). And ‘organizational trust’ is a recognized research topic in management studies, taking in not only corporations, but also non-governmental organizations and other complex group agents. Such research investigates both the influence of organizational contexts on our trust in individuals, and the ways in which trust is invested in organizations themselves (e.g. Saunders, Skinner, et al. 2010). (2017: 230)
Trusting governments and financial institutions, trusting experts and cyber technology; in some areas of academia, the concepts of trust and trustworthiness are being used to talk about issues of national, environmental, and technological security. In other areas –
Introduction Primer: On Trust and Trustworthiness 4 such as science – the concepts of trust and trustworthiness function differently, for example, when talking about the transfer of scientific knowledge and the process of scientific inquiry. In economics, trust and trustworthiness are about prediction and risk. These disparate senses of trust have the bare schematics in common: they take trust to involve three elements, 1) a truster, 2) a trustee, and 3) that which is entrusted (in many cases ‘truster’ and ‘trustee’ are to be given a broad interpretation so as to include non- human entities). Reflecting this, in philosophy today, the version of trust that has become dominant is three-place trust:
It has become normal, at least in Anglophone analytic philosophy, to think of it [trust] as fundamentally a three-place relation with an infinitival component: A trusts B to do C. (Domenicucci and Holton 2017: 149)
This strict simple three-place definition of trust, known here as: (A trusts B to do X), seems able to accommodate the ideas we have been exploring: namely, that: 1) trust is ubiquitous (since we depend on other human beings, systems, and entities to do things for us all the time), 2) trust is a kind of expectation that others will do things for us (behave in desired ways), and 3) trust is characterised by risk (since B might not do C for A). It is also apposite, according to many authors, because it is a distillation of the way we commonly use the word ‘trust’ in English (e.g. I trust you to return to me in one piece), and some, such as Philip Nickel, embrace the wide-scope ‘explanatory power’ of three-place trust, applicable as it is to a wide range of phenomena:
Trust should be able to explain cooperation, and its failure should help explain the emergence of cooperation-enabling institutions… What follows is the superiority of unrestricted views of trust, which take trust to be no more than the disposition to rely on others, over restrictive views, which require the trusting person to have some further attitude in addition to this disposition. (2017: 195)
This dissertation, in essence, categorically rejects this picture of trust and trustworthiness, arguing that the historical construal in English-speaking academia of trust as a feature of our cognitive/epistemic lives facilitating and implicating our knowledge, agency, risk, and rationality is mistaken. On the trust side, it is proposed that trust need not be able to explain cooperation in every instance, since not all
Introduction Primer: On Trust and Trustworthiness 5 instances of cooperation are instances of trust, and failures of trust do not always explain the emergence of cooperation-enabling institutions. Trust is more than a disposition to cooperate with others, rely on others, and expect them to do what we ask. It is more, even, than a disposition to react with resentment and betrayal when others fail to do what we expect of them. On the trustworthiness side, trustworthiness is more than being cooperative, honest, or compliant, and it is more than being these things with an attitude of goodwill or altruism. Trust and trustworthiness are not ubiquitous, but special. They take time to come into existence, doing so only in our close interpersonal relationships, and – even in those relationships – trust and trustworthiness do not involve a cognitive expectation (or risk) that others will do or be (or fail to do or be) anything in particular. Perhaps the most succinct way to capture what is to come in this dissertation is by quoting an exchange between a holographic Commander C. ‘Trip’ Tucker and Commander William T. Riker, characters beloved by fans of that fertile font of philosophical exchanges, the TV show Star Trek: Enterprise (Berman 2005):
Trip: I can count on one hand the number of people I trust. I don't mean trust like... 'I trust you aren't lying to me' or, 'I trust you won't steal my money'. I'm talking about the kind of trust where... you know someone's not gonna hurt you, no matter what; where you know they'll always be there for you, no matter how bad things get. You ever know anybody like that?
Riker: Yeah. One or two.
Unlike three-place accounts, the two-place account presented in this dissertation (A trusts B) and (A is trustworthy regarding B) is able to clearly differentiate trust and trustworthiness from the phenomena with which they are often confused: phenomena with which trust and trustworthiness no doubt interact in interesting ways but from which they are nonetheless distinct. This is essential, since – though we often use the terms interchangeably – that which often coincides with trust (i.e. reliance, belief, cooperation, expectation, hope etc.) and trustworthiness (i.e. fairness, honesty, cooperativeness, predictability, etc.) are not interchangeable with trust and trustworthiness, nor with one another. Given the prominent role trust has come to play in – as Domenicucci and Holton (2017) put it – “Anglophone analytic philosophy”,
Introduction Primer: On Trust and Trustworthiness 6 there is a pressing need to provide an account of trust and trustworthiness that instead captures their distinct and important role in our lives, as clear differentiating features of our anomalous rule-breaking special relationships; i.e. the ‘one or two’. As a final example to set the tone for what follows, I turn to an interview by Australian radio personality Richard Fidler with Graham Long (2017). Reverend Long is the Pastor and CEO of Wayside Chapel in Sydney’s King’s Cross– a refuge for domestic violence victims, the homeless, addicted, and otherwise desperate and abandoned. A sanctuary to receive second, and subsequent chances, Wayside is a place where trust and trustworthiness are at their most vital, most vulnerable. Though not religious myself, when I heard this interview on the radio around the middle of my candidature my ears pricked up. Like anyone who has lived in that part of Sydney for any length of time, I was familiar with Reverend Long and had seen him about from time to time as he met with Sydney's least lucky. Long was brought up by a fundamentalist preacher, and I knew already of his opinion that when it comes to human beings “any attempt to formularise is bullshit”. Having wrestled with his own trustworthiness in the public eye (the preacher who had an affair!) his humanism and not his religiosity had made him a local hero. Long sees his role as one of meeting people where they are, seeing them not saving them, and on this occasion, in answer to a question about the qualifications one needs to be effective (as he is) at the job that he does, Long remarked:
One of the sad things about the culture we live in, is we think things like trust, things like love… things like resilience, all kinds of so-called ‘personal qualities’ reside inside me. And it’s profoundly misunderstood because the truth is there’s no such thing as a single human being. There’s no such thing. The word ‘I’ could only ever describe half of something. The truth is I am not complete until there is at least two of us. My centre is not in here… it isn’t in here– it’s between you and me. And, while I live as an individual, I never know who I am. I can’t know who I am. It is only when I truly engage with you that I begin to find out who I really am. You might say, in a certain sense, you complete me. People who fall in love say that, but it’s true in every relationship. When we meet, you know, we are completed by the other. And you learn who you are through the other. We discover that my humanity is made by yours… and yours by mine.
Introduction Outline: On Trust and Trustworthiness 7
Trust and trustworthiness are each one-half of a two-place relation. They do not exist until there are two of us, and we make them, and break them, together. They are complex, veiled, full of contradiction, and they contribute to our sense of a shared humanity. At heart, this is what I spend the next hundred thousand words or so considering in this thesis On Trust and Trustworthiness.
Outline: On Trust and Trustworthiness
the life of one person is interwoven with the life of another. (Løgstrup 1956) translated (1997: 10)
Part I of this research project, On Trust, addresses the nature of trust. Section I.1 Trust in the Philosophical Literature begins by examining how trust has been understood in the contemporary philosophical literature. The picture that emerges is one in which three-place trust is – and has historically been – dominant. This reflects the questions philosophers have considered with regard to trust, i.e. various epistemic questions to do with agency, risk, knowledge, and rationality. Section I.1 thus makes a contribution in its original organisation of the seminal literature to highlight how trust has been framed as both a risk to, and solution to, an epistemic problem within the discipline of philosophy. Section I.2, Dominant Views on Trust: Problems, elaborates on the various versions of three-place trust, and highlights their limitations. I.2 makes its primary contribution in its exposition of these limitations. A secondary contribution of this section lies in the presentation and consideration of three explanations as to why philosophers have preferred not to abandon three-place trust despite its limitations. The aim of providing such explanations is to better understand the continued dominance of three-place trust in philosophy. Section I.3, The Object of Trust, contains the most significant contribution of Part I. In this section, I argue that it is essential to view three-place trust not as a kind of trust (let alone the dominant kind) but as a kind of reliance. The argument this dissertation makes for this shift in understanding relies on revealing an ambiguity regarding the object of trust that is inherent to the three-place trust logic. This section thus achieves
Introduction Outline: On Trust and Trustworthiness 8 the primary aim of Part I, which is to establish that three-place trust is not a kind of trust. This dissertation then offers some recommendations as to what sorts of things count as the proper objects of trust and trustworthiness. Accordingly, I.3 also contributes to the ongoing debate in the literature concerning what sorts of things can trust (and be trusted) i.e. persons, institutions, governments, systems, animals, etc. The final section of Part I, Trust I.4 presents an original discussion that emphasises the intricate private two-place nature of trust. Drawing on an example from George Eliot’s novel Middlemarch – the same example Victoria McGeer has used to make a related point – it is argued that trust is a complex experience that one person (the truster) has of another (the trustee). This section thus satisfies the secondary aim of Part I, which is not to provide anything like a definition or formula of trust, but to develop some novel ideas that make a small contribution to our understanding of the nature of trust. This section ends with a relaxed discussion that aims to anticipate and address some objections to the shift from a three-place understanding to a two-place understanding. Part II, On Trustworthiness, addresses the nature of trustworthiness. Section II.1 Trustworthiness: in the Philosophical Literature, begins as in I.1, by presenting how trustworthiness has been understood in the contemporary philosophical literature. Here, a distinct picture emerges. In the case of trustworthiness, various one-place conceptions are also widely adopted that view trustworthiness not as an epistemic notion, but as an ethical notion. Despite this underlying difference though, II.1 confirms that most philosophers still consider trustworthiness to be – at least in part – a three-place relation, notwithstanding any additional one-place moral requirements. Section II.1 demonstrates that this research project makes an original contribution in its proposal that trustworthiness is one-half of a two-place relation. Section II.2, Dominant Views on Trustworthiness: Problems, addresses the limitations of both one-place and three-place conceptions of trustworthiness. II.2 thus makes a substantial contribution to our current understanding of trustworthiness in its exposition of these limitations and satisfies the primary aim of Part II of this research project, which is to establish that trustworthiness is neither a one-place, nor a three- place relation.
Introduction Outline: On Trust and Trustworthiness 9
Section II.3, Rich Trustworthiness or Poor Trustworthiness, instead of rehearsing all the same epistemic arguments in relation to trustworthiness, shifts to focus instead on the views of one philosopher: Karen Jones. The primary aim of this section is to draw attention to the worrying ethical implications of her influential view (rich trustworthiness), implications that will extend to any view which is similar to hers in the relevant respects. This section thus contributes to a tertiary aim of Part II to demonstrate the potential ethical pitfalls of providing narrow behavioural definitions of concepts like trust and trustworthiness, which have historically been used to demarcate and establish our very personhood and place in a shared humanity. Finally, in symmetry with Section I.4 Trust, Section II.4 Trustworthiness makes use of another fictional example – the film The Invention of Lying – to develop our understanding of trustworthiness. The surprising link between trustworthiness and “mind-blindness” is expounded, improving our understanding of the nature of trustworthiness. In summary, Part I: On Trust and Part II: On Trustworthiness share parallel structures, primary and secondary aims, and make parallel primary and secondary contributions. The primary aim of both parts is to provide good reasons for thinking that we must shift our understanding from the dominant views. The secondary aim of both parts is to make a contribution to our understanding of trust and trustworthiness.
On Trust
The inner self who engages in these judgements – the essential person, the active self who really matters – was somehow being ignored and forgotten. –Iris Murdoch (2013: xi)
Currently, the dominant view in philosophy is that trust is a three-place relation (A trusts B to do X), or – less strongly – that there is a kind of trust which is three-place. It is the principal concern of Part I of this thesis to argue against the three-place conception of trust, on the grounds that it leads us to think of trust as an epistemically grounded, publicly observable phenomenon rooted in behaviour, which it is not. Instead, Part I suggests that there are more compelling reasons for thinking of trust as a private complex and variable experience that the truster has of their trustee, integrating a range of phenomena including our attitudes, beliefs, desires, intentions, memories, reactions, reasoning, emotions and so on, and manifesting – at times – in behaviours, though such behaviours are not to be identified with trust. So-called “three-placed trust” is better understood not as a kind of trust, but as a kind of reliance.
Overview of Part I, On Trust
Part I has four substantial sections. Section I.1 Trust in the Philosophical Literature reviews current philosophical literature on trust. As my aim in this research project is to reject wholly the view of trust that is dominant in the philosophical literature, this review is vitally important to setting up what follows. Section I.1 demonstrates that three-place trust is almost universally adopted, even among those who recognise other
10 I Overview of Part I, On Trust 11 elements to trust – such as an emotional or attitudinal element. I.1 also reveals that three-place trust has led to the “problem of trust” being largely understood to be an epistemic problem, concerning the transfer of knowledge. Section I.2, Dominant View on Trust: Problems provides further detail and examples of three-place trust and explores various problems with it, while observing some reasons why philosophers have chosen to adopt it in spite of its limitations. Section I.3 The Object of Trust moves on to critically examine the internal logic of three-place trust, concluding that three-place trust is actually a kind of reliance. Section I.4 Trust puts forward an alternative two- place account, making use of an analogy between trust and love, and an excerpt from George Eliot’s novel Middlemarch – which has previously been used by Victoria McGeer to highlight trust’s link to hope. Finally, Section I.4 also discusses some of the outcomes of the switch from three-place trust to two-place trust and anticipates and addresses some objections that may be raised.
I.1 Trust: in the Philosophical Literature 12
I.1 Trust: in the Philosophical Literature
Currently, there is no definite consensus either within or outside of philosophy regarding the nature of trust. While this is true of many similar phenomena, such as love (Helm 2017b), forgiveness (Hughes and Warmke 2017), loyalty (Kleinig 2017), and friendship (Helm 2017a), trust – unlike these other concepts – is used as a theoretical or technical term in a number of academic disciplines, most notably in biology, psychology, economics, and political and social science. As these disciplines often rely on theory and evidence external to their discipline (e.g. psychologists and political theorists use empirical data and theory from economics and vice versa) there is a worrying scope for confusion and error if what is meant by ‘trust’ in one discipline does not translate to another. In philosophy, as there is also no clear definition or consensus, organising the relevant literature on trust is inherently challenging and could potentially be done successfully in any number of different ways. This section identifies three broad themes that emerge – trust as an emotion, trust as an attitude, and trust as an epistemic state. Yet because one may coherently hold the view for example, that trust is an attitude which involves either an affective component, a cognitive component (such as beliefs), or both, neat lines simply cannot be drawn. This is further complicated by the fact that many philosophers writing on trust do not explicitly spell out their position regarding its nature. That said, it is often possible to glean a philosopher’s thinking from the specific problem they pay attention to. For example, if the philosopher is concerned with the rationality of trust, then it is likely they conceive of trust as an epistemic state. If a philosopher is concerned with our ability to decide to trust, then they may worry that trust is a non-voluntary emotion, or belief. Revealing a philosopher’s stance on the nature of trust often involves examining that philosopher’s question of interest vis-à-vis trust. Most philosophers discussing trust have been interested in the so-called risk of trust, and the question of when our trust is warranted or justified, with a smaller number paying attention to ethical questions that arise in relation to trust. What this shows is that most philosophers view trust as a cognitive/epistemic notion, similar to belief.
I.1 Trust: in the Philosophical Literature 13
Given these complicating factors, this dissertation organises the literature according to the bare schematics. Subsection I.1.1 One-place trust presents literature that conceives of trust as a one-place relation, and also literature which takes trust to be an attitude, connected to betrayal. Subsection I.1.2 presents literature on three-place trust, acknowledging that the many accounts which fit under this banner are diverse in their details. As paying attention to the question a philosopher is interested in can be so illuminating when it comes to saying what exactly they understand trust to be, I.1.3 Trust and Rationality and I.1.4 Voluntariness of Trust present two main questions of interest for philosophers, which serve to turn this messy history into a more coherent narrative.
I.1.1 One-place trust
When trust is conceived of as a virtue, emotion, character or personality trait, this locates the trust solely in the truster, making it a property or one-place relation. As a side note, it must be acknowledged that while some philosophical accounts of emotion view emotion as a two-place relation, for example Kenny (1963) and de Sousa (1987). Others have argued that many emotions – notably love – lack any propositional object (Kraut 1987; Rorty 1987). For simplicity here, emotions are treated as one-place relations, lacking any propositional object. If this turns out to be wrong, it will not force the conclusion that trust is an emotion, since various two-place emotions may be features of trust without trust being an emotion. Sometimes, accounts of trust as an attitude are one-place as well, but this is not true of every attitudinal account of trust. To be clear on the difference between one-, two- and three-place accounts, contrast:
1. A trusts (is a trusting person)
2. A trusts B
3. A trusts B to do X In simple examples: 4. Rasmus is a very trusting boy. 5. Rasmus trusts Iris. 6. Rasmus trusts Iris to look after his dinosaur while he is at school.
I.1 Trust: in the Philosophical Literature 14
1 and 4 are examples of one-place trust, 2 and 5 are examples of two-place trust, and 3 and 6 are examples of three-place trust. This subsection discusses literature in which trust is understood in the one-place sense captured in 1 and 4. Almost all philosophers who endorse the idea that we have one-place elements to our trust still embrace a three-place conception, where trust is a one-place response – a reaction – to what the trustee does. Purely one-place views, where trust is just an emotion, attitude, virtue, character or personality trait, are actually very rare. While a number of philosophers do mention a sense in which trust is a personality or character trait – for example Potter (2002: 8) – most qualify that this is not the sense of trust they are interested in. Indeed, while the view is more common in the literature on trustworthiness as we shall see in II.1, almost no philosophers write as though trust were a virtue for example, and it is usually only trustworthiness and not trust that is taken to be a moral notion. Notable exceptions to this are Nickel (2007), and Cohen and Dienhart (2013) who have suggested that the duty to be trustworthy can only be ascribed to a trustee by the very act of trusting, hence trust might be a moral notion, and McGeer, who has observed that being able to forgive is a “functional virtue” that trusters require to sustain trusting relationships (2008: 247). So, while the idea of trust as a virtue is rare, a number of philosophers do endorse the idea that trust is an emotion or that it contains an affective element. Though the idea that trust is an emotion is the least popular stance found in the literature, indications that trust does seem to have features similar to emotions appear in influential accounts of emotion, such as those of de Sousa (1990), Calhoun and Solomon (1984), Rorty (1980), and Lahno (2001). For example, Karen Jones has stated that trust – like other emotions – narrows our perception to certain “fields of evidence” i.e., those fields that support that emotion (Jones 1996: 11). For similar discussions, see also Baker (1987) and Webb (1992). Likewise, Carolyn McLeod (2015) has pointed out that if it is true that trust makes us blind to certain evidence – such as evidence of untrustworthiness – then the emotional account of trust holds some sway in its ability to account for this apparent “blindness-to-evidence” feature of trust. That said, Keren (2014) has provided an argument to show that a belief-based account of trust is able to accommodate this apparent feature, and so, again as Carolyn McLeod says, it is not clear why – on this
I.1 Trust: in the Philosophical Literature 15 basis – one should endorse a purely emotional account of trust. She says that while the idea holds some attraction, there does seem to be an undeniable distinction between trust and other more readily identifiable “typical” emotions, such as anger or joy (McLeod 2015). The reluctance to embrace a completely one-place account of trust reflects the current trend in philosophy to see trust as playing a role in an epistemic problem to do with testimony, as discussed already in Primer on Trust and Trustworthiness. The one- place account simply does not raise these epistemic questions, which philosophers are so interested in, such as: ‘How can trust in expert testimony be justified?’ and ‘How can it be rational to trust when trust requires us to overlook evidence?’ etc. Despite this, the pull of the emotional element to trust is strong, hence, it is not surprising to find the idea that trust is an emotion plus some other element explored in the literature. The term adopted in this dissertation for such views is hybrid, as they are schematically one/three-place in nature. As we shall see, hybrid views are popular as they allow philosophers to explore their epistemic questions while acknowledging other intuitive ‘non-cognitive’ elements to trust and trustworthiness. One example of hybrid thinking is evident in Stephen Darwall’s comment that even in epistemic three-place trust, where we “trust what others tell us”, we invariably “take it to heart”, putting our own personal feelings at stake (2017: 47). Another is in Bernd Lahno’s detailed argument for trust as an emotion, using an Aristotelian conception of emotion, which places the affective element of trust at the centre of his explanation, while also acknowledging that trust is an attitude “Trust does in fact possess emotional character in some important respects. It is an emotional attitude” (2001: 173). Similarly, Jones (1996) speaks of trust as an “affective attitude”. Like the emotional element, the attitudinal element of trust has likewise been acknowledged. Many philosophers speak of the “attitude of trust” as though to have an attitude of trust is to have one-place trust ( see, for example, Faulkner (2017: 20). While his is not a one-place view, Richard Holton (1994) no doubt provided the most influential ‘attitudinal account’ of trust. Particularly influential among those philosophers who show at least some concern with the metaphysical question regarding the nature of trust are Hieronymi (2008b), McGeer (2008), Walker (2006: 79), Faulkner
I.1 Trust: in the Philosophical Literature 16
(2007a, 2007b, 2015), Govier (1993, 1998), and Jones (2012). Indeed, the impact of Holton’s ‘reactive attitude trust’ on the metaphysics of trust is so significant that the Stanford Encyclopedia of Philosophy section on ‘The Nature of Trust and Trustworthiness’ now boldly asserts: “Trust is an attitude we have towards people we hope will be trustworthy, where trustworthiness is a property not an attitude.” (McLeod 2015). Yet among those who agree with Holton that trust “is an attitude” there remains disagreement about what sort of attitude trust is. Is it an emotional attitude as Lahno and Jones suggest, or a moral attitude (Mullin 2005), or even a belief-attitude (Hieronymi 2008b, 2008a; Wright 2010: 618)? Holton’s own view – the reactive attitude view – uses the concept of the participant stance first introduced by P.F. Strawson (1962) and involves a readiness on the part of the truster to feel betrayal. Trust occurs – according to Holton (1994) – whenever the truster stands ready to feel betrayal should the trustee fail to do what they have been entrusted to do. This view is complicated, as it allows that trust can have both an affective element, as well as involve beliefs (i.e. that the trustee is trustworthy) but maintains that it is nevertheless best explicated in terms of the reactions we have to the actions of the trustee. Holton (1994) provides what is ultimately a view of trust that is schematically three-place in nature; he acknowledges as much in Trust as a Two-Place Relation (Domenicucci and Holton 2017: Footnote 1). Significantly, the notion of betrayal is central to Holton’s 1994 discussion of trust and this proposal has provided a flashpoint of disagreement. Colin O’Neil, for one, agrees with Holton, but takes things further, arguing that a betrayal of trust cannot occur unless one is trusted. He says, “it is always a defence to a charge of having betrayed someone’s trust that there was no trust to betray” (2017: 71). O’Neil also claims that it is essential for a betrayal of trust to occur that the trustee be aware of the trust “the trusted person cannot betray trust that he or she is unaware of” (2017: 80). Against this, Thomas Simpson T. W. Simpson (2012: 553) has argued that in some cases the reactive attitude view does not do any distinctive work, because, as Philip Nickel similarly argues; merely saying so (that trust makes us stand ready to experience betrayal) cannot explain why the truster stands ready to feel betrayal, rather than just disappointment. Nickel says:
I.1 Trust: in the Philosophical Literature 17
Without an account of what attitudes comprise the moral attitude of trust in the first place, it is impossible to say what justifies a person in feeling disappointed or betrayed when trust fails. (2007: 318)
How people typically react, and whether they are justified in so reacting, are two separate questions not entirely made distinct in Nickel’s objection. However, if trust is conceived of as a three-place reactive attitude that necessarily causes betrayal when the trustee fails to do what they have been entrusted to do (otherwise it was never really trust in the first place), then Nickel and Simpson are right in pointing out that sometimes, this just isn’t so. As this dissertation suggests (II.1, II.2, II.3) the failures of trustees to do what they have been entrusted to do can cause any number of various actions and reactions, and philosophers do not get to say which ones are and which are not legitimate. Yet while Simpson and Nickel’s objection holds some sway, it does not do away entirely with the idea that trust might involve betrayal. That we do not always react to betrayals in the ways one might expect does not show that the notion of betrayal is not connected to trust and trustworthiness in some important way. As it stands, there is a notable hole in the literature when it comes to the notion of betrayal. Until this situation is rectified, betrayal cannot stand as a theoretical construct upon which trust is defined. Though some preliminary accounts can now be found on betrayal (Cogley 2012; Harding 2011; O’Neil 2017), the tendency has been for philosophers to deliver accounts that fit with a three-place conception of trust that they have already accepted. The accounts of betrayal on offer are invariably three-place as well– because trust is (A trusts B to do X) betrayal is (B failing to do X for A). Such descriptions of what it is to betray do not fit with the two-place account of trust offered in this dissertation. Moreover, a three-place account of betrayal does not fit with empirical research on betrayal (detailed in II.4), which shows that betrayal – of which infidelity is the prime example – consists not in our behaviour, but in something much more complex about the particular relationship two people share. One might say that the betrayal is in the promise, the history, the reason, the motive, the emotion… in the person, not in the act. Clarity on this is vital, since so many philosophers endorse the idea that what trusters risk (specifically) in trusting is betrayal: Baier (1994, Chapters 6-9), Holton (1994),
I.1 Trust: in the Philosophical Literature 18
Jones (2004), Walker (2006, Chapter 3), Hieronymi (2008), McGeer (2008), McMyler (2011, Chapter 4), and Hawley (2014). Recently, Paul Faulkner has explicitly endorsed the idea that the attitude of one- place trust is fundamental, with an argument based on language use:
If X trusts, there must be some Y that X trusts, but that X trusts Y does not in any way imply that X trusts more generally. So, of the two predicates, the one place one is arguably more fundamental. Thus, the heart of our notion of trust seems to be simply an attitude of trust, which may, but need not, take specific persons as its object, and which can support, but need not, the act of relying on persons. (2017: 120)
Personally, I do not share with Faulkner the intuition that (X trusts Y) does not imply that (X trusts) more generally. Surely, to trust Y, X must in some sense ‘have trust’. Analogously, to love anything is to have love, to hate anything is to have hate. While the idea of possessing a one-place ‘attitude of trust’ is not logically incoherent as Faulkner says, such a thing could only ever be – this dissertation argues in I.4 – one part of trust. The idea of trust as an emotion, virtue, character trait or attitude can only ever capture something about what it sometimes means to trust, since many elements of trust are only brought about in our interactions/relations with others. On this point, the notion of self-trust is sometimes characterised as one-place, since it involves only one individual. Philosophers who discuss self-trust include Foley (2005), Govier (1993), McLeod (2002), and Lehrer (1997). Though the thought that we can and do trust ourselves is well and truly alive in our folk-psychology, the question regarding the possibility (or perhaps even necessity) of self-trust in philosophy has not been settled. Indeed, it cannot be settled without first having settled the questions what it is to trust, what it is to have a self, and perhaps also what it is to possess knowledge. Yet answering such questions is not the primary focus of those who discuss self-trust. This is because – while it is often spoken of as though it were one-place – the notion of self-trust in play is actually a three-place epistemic notion – (A trusts A to do X). The questions that interest those discussing self-trust in the literature are invariably identical to the ones discussed by those who adopt three-place trust – questions to do with risk, agency and rationality.
I.1 Trust: in the Philosophical Literature 19
For example, Edward Hinchman (2003: 25-51) has argued that self-trust runs the same risks as trust in others, in the sense that reasonable trust presupposes trustworthiness in the trusted in both cases and so it can only be rational to trust ourselves if we have good reason to believe that we ourselves are trustworthy. Hinchman thus sees self-trust as a diachronic exercise of practical reason. Trudy Govier (1993: 101-20) argues that self-trust is a necessary condition for personal autonomy. She says:
to lack general confidence in one’s own ability to observe and interpret events, to remember and recount, to deliberate and act generally, is a handicap so serious as to threaten one’s status as an individual moral agent. (1993: 108).
Govier takes self-trust to involve “having a positive sense of one’s own motivation, competence, and integrity.” (1993: 110). Her view of trust is often difficult to distinguish from the process of having and developing confidence in oneself. She says “one may have more confidence in one’s own character and capacities than past experience and evidence would warrant, thus having too much self-trust in an epistemic sense” (1993: 115). Carolyn McLeod also develops a link between self-trust and agency, arguing that to be motivated to choose and act in accordance with our own values, we need to trust ourselves to do so:
Although some self-trust might be better than none at all as far as knowledge and autonomy go (for no self-trust would leave us incapacitated and with no opportunity to learn from our mistakes), justified self-trust is best overall. Without being justified in trusting ourselves to be good epistemic or autonomous agents, we cannot be either. (McLeod 2002: Section 3)
Similarly, Keith Lehrer (1997) argues that to be rational, we must first be worthy of self-trust, which he takes to involve a capacity to evaluate our own beliefs and desires. Lastly, Richard Foley (2005) contends that it is a necessary requirement of knowledge possession, that we be able to trust ourselves. These examples demonstrate that in the literature, self-trust is a matter of being able to rely on oneself to do something: to be honest with oneself, follow through on one’s own intentions, uphold one’s own
I.1 Trust: in the Philosophical Literature 20 promises to oneself, etc. Although it is often characterised as such, self-trust is not a one-place relation, but a three-place relation, involving – in a more accurate sense – two individuals (A at T1) and (A at T2) who have to rely on “one another” in their planning and deliberation. Self-trust is: (A at T1) trusts (A at T2) to do X. In summary, this subsection has shown that among those who endorse a logically one-place view, there are a number of characterisations: trust as an attitude, emotion, virtue, character or personality trait. Yet most philosophers that speak in terms of trust as one of these, on closer inspection actually endorse a hybrid one/three place view, which acknowledges some stable one-place element to trust while still speaking of trust as involving reliance, and so, as a three-place relation. This is also true of the notion of self-trust: the notion of self-trust in play in the literature, on closer inspection, is actually about our reliance on ourselves (practical or epistemic) and so, addresses trust as a three-place relation.
I.1.2 Three-place trust
To get a sense of the primacy of the three-place schematic in modern Anglophone analytic philosophy – the schematic I shall be arguing against – observe the following quotations, from leading philosophers on trust:
Trust is generally a three-part relation: A trusts B to do X. (Hardin 2002: 9)
Almost without exception, philosophical discussion of trust focusses on the three-place predicate: X trusts Y to φ. (Faulkner 2015: 424)
We usually trust people to do certain things—for example, to look after our children, to give us advice, or to be honest with us. (McLeod 2015: Section 1)
Take a stock example of trust. Jack asks his friend Jill if she will lend him her first edition volumes of Sterne’s Tristram Shandy, promising to return it to her within a week. (T. W. Simpson 2017: 178)
The clear majority of philosophers writing on trust take trust to be a three-place relation. Though their views differ considerably in detail, those adopting a logically three-place account of trust include Baier (1986, 1991a, 1991b, 1994, 2004), Baker (1987),
I.1 Trust: in the Philosophical Literature 21
Blackburn (1998), Cogley (2012), Coady (1992), Foley (2001), Daukas (2006), Dasgupta (2000), Faulkner (2007b, 2011), E. Fricker (1995), M. Fricker (1998), Fukuyama (1995), (Gambetta 1988), Goldman (2001), Govier (1993), Hardin (1996, 2002), Hardwig (1991), Hawley (2017), Hertzberg (1988), Hieronymi (2008b), Hinchman (2003, 2017), Holton (1994), Jones (1996, 1999, 2012, 2017), Keren (2014), Koenig and Harris (2007), Lahno (2001, 2017), Longworth (2017), McGeer (2008); McGeer and Pettit (2017), McLeod (2002), McMyler (2011, 2017), Mullin (2005), Nickel (2007, 2017), O’Neil (2012, 2017), Oshana (2014), Owens (2017), Pettit (1995), Potter (2002, 2013), Rorty (1980), Skyrms (2008), Silvers and Francis (2005), T. W. Simpson (2012, 2017), Smith M. N. Smith (2008), Eric M Uslaner (2000), Warren (1999), Webb (1992, 1993), and Zagzebski (2012). Most of the accounts provided by the above philosophers are very similar in their underlying structure, as well as in the illustrative examples provided in their discussions. For a more detailed sense of things, the following three examples should suffice: Stephen Wright’s necessary and sufficient conditions for three-place trust, Victoria McGeer and Philip Pettit’s necessary conditions for trust, and Philip Pettit’s illustrative descriptions of three-place trust.
These are what I think are the necessary and sufficient conditions for the statement ‘P trusts Q with something’ to be true: 1. P requires/requests something of Q. 2. P takes the participant stance described by Holton (1994: 67). 3. P believes that Q’s response will influence the situation. 4. P believes that Q is freely capable of choosing to do or not do as P would like. 5. P believes that Q believes they are being trusted or has access to evidence permitting them to reasonably form the belief. (Wright 2010: 618)
I will count as trusting you to do something X just insofar as three conditions are met. First, I manifestly rely on you to do X: I make clear to you my assumption that you will prove reliable in doing X. 2 Second, I assume that the manifest fact of my reliance will weigh with you as a reason for choosing voluntarily to X. And third, this assumption helps to explain or reinforce my relying on you. (McGeer and Pettit 2017: 15)
Think of the new resident who asks a neighbor to look after her home or pets or plants while she is away and gives them a key to her house. Think of the passenger who admits to not
I.1 Trust: in the Philosophical Literature 22
knowing the town and asks a taxi driver to get him to his destination by the quickest route available. Think of the person who asks a perfect stranger for directions on how to get somewhere and then follows these meticulously. Think of the customer who, finding that one store does not have something he wants, asks the salesperson for advice on where else to search. Or think of the visitor who asks a newsagent to recommend a good evening's read or asks a cinema attendant's opinion of the film showing. (Pettit 1995: 218)
As the above passages demonstrate, necessary and sufficient conditions, and illustrative examples using things like lending and borrowing, accepting testimony, or asking for assistance. Among those who endorse a three-place view, one or two-place conceptions of trust are occasionally briefly acknowledged. Jones, for example, displays recognition of two-placed trust when she says “I can find your trust burdensome, even when you have not entrusted anything in particular to me” (2017: 95). Stephen Wright explicitly recognises the distinction between two-place and three-place trust (2010: 616). Others who do this to some extent include: Hinchman (2003), Govier (1993), Foley (2001), McLeod (2002), Goering (2009), Jones (2012), and Potter (2013). Yet most find the notion of three-place trust to be the philosophically interesting one, with only a handful speaking in any real detail of either the one-place or two-place conceptions. Recently, others have made explicit arguments in favour of reversing the order of importance from three-place in favour of a one-place (Faulkner 2017), or two-place (Domenicucci and Holton 2017) conception. Faulkner’s position appears in stark contrast to his earlier work, in which he spoke of trust in the three-place manner, of “trusting speakers for the truth” (Faulkner 2007b, 2011). A similar transformation occurs with the views of Richard Holton. In his highly influential 1994 paper “Deciding to Trust, Coming to Believe”, Holton used a number of thought experiments when considering the nature of trust that were categorically three-place. In the main example, we are asked to imagine a situation in which we (the truster) have entrusted someone (the trustee) to catch us as we allow ourselves to lean backwards and fall. Now however, he endorses the view that the three-place notion is not the primary kind of trust at issue:
How should we think of trust? It has become normal, at least in Anglophone analytic philosophy, to think of it as fundamentally a three-place relation with an infinitival
I.1 Trust: in the Philosophical Literature 23
component; A trusts B to do C. Here we aim to question that idea… in giving an account of trust, this three-place relation provides us with the wrong place to start, and that we should start instead with the two-place relation, A trusts B, and work from there. (Domenicucci and Holton 2017: 149)
While both Faulkner’s and Domenicucci and Holton’s positions signal a critical reversal of the status quo, both take the three-placed notion of trust to be – at the very least – a legitimate kind of trust “we don’t deny the three-place relation exists, expressed in a perfectly natural English idiom, with a useful role to play” (Domenicucci and Holton 2017: 149). And more strongly, Paul Faulkner writes “Trust can be three- place, we trust one another to act in various ways–not to read our diary, to give a fair quote, and so on” (2017: 118). His newer account thus coheres with his earlier work in which he asserts “There is no question that three-place trust is central to our engagement with others.” (2015: 424). Hence, while these contributions signify an explicit recognition of the importance of other previously neglected forms of trust, neither of them explicitly rejects three-place trust, instead adopting it as a “kind of trust”. While such developments are in line with this dissertation, by leaving a place for three-place trust, and by not emphasising the complex private nature of trust, the ideas developed by both Faulkner and Domenicucci and Holton can certainly be differentiated from what is said in this dissertation. In contrast, I shall argue that trust properly understood cannot be regarded as a three-place relation at all since it leads to a quite different cluster of affective and cognitive phenomena not just in its relationality but also in its character. Aside from these two new exceptions, which both still endorse three-place trust as a ‘kind of trust’, the idea that trust is a three-place relation has been almost unanimously accepted by philosophers, even among those that identify strong moral or emotional elements to trust as we saw in the previous section. This is, as Thomas Simpson explains in his own take on the current state of affairs in the philosophy on trust, an effect of the seemingly intractable reality that trust involves cognitive elements, such as belief:
I.1 Trust: in the Philosophical Literature 24
It is unsurprising that those who observe the responsiveness of trust to moral and other non- epistemic reasons should construe trust as a non-cognitive mental state. What is surprising is a related but distinct family of views that are now current, on which trust is taken to be both cognitive and responsive to moral reasons. So distinguish non-cognitive non- evidentialism from cognitive non-evidentialism. What is the motivation for this latter view? Exponents are, I think, impressed by the need for belief that trust often seems to involve. It is not just that the truster acts as if she believes that the trusted will be trustworthy; rather, she actually believes that he is. It is that belief which constitutes the trust. They are also impressed by the moral dimensions of trust, such that interpersonal relationships ground the resulting belief, and this not merely through the epistemically uncontroversial process of giving privileged access to evidence. Cognitive non-evidentialism thus seeks to endorse both claims simultaneously. (2017: 182)
This passage demonstrates how views of trust have become considerably more refined and complicated in recent years, creating many hybrid accounts which acknowledge that we have both epistemic and non-epistemic reasons to trust. Simpson himself endorses a “Scope-restricted evidentialist constraint” that says that in some circumstances, trust is rational only if, on one’s total evidence, it is likely that B will φ (2017: 183). Yet despite this welcome acknowledgment that trust might involve both – as Simpson puts it – “cognitive and non-cognitive elements”, an intractable tension remains whenever trust is at all conceived of as fundamentally something that we do. This is because our so-called epistemic and non-epistemic ‘reasons for trust’ can conflict in epistemically devastating ways, rendering us irrational, one way or the other, whatever we end up doing. This longstanding ‘problem of trust’ will be discussed in the following section. In summary, this subsection demonstrated that trust is primarily conceived of as a three-place relation in the philosophical literature, even among those who recognise other dimensions to trust. And, even when it comes to the only two existing papers which argue explicitly that the three-place conception should not be considered central, both still maintain that (A trusts B to do X) is a legitimate and important kind of trust. Because of this three-place dominance, where trust is conceived of as something that we do (leave a diary on a desk) because of something that we believe/hope (our trustee will
I.1 Trust: in the Philosophical Literature 25 not read it), philosophers on trust have largely been concerned with trust as an epistemic problem.
I.1.3 Trust and rationality
This section addresses literature that has been concerned with an apparent tension between trust and rationality. This tension is often referred to as The Problem of Trust, or The Risks of Trust and it is frequently conflated with The Problem of Testimony. This subsection demonstrates that philosophers concerned with The Problem of Trust view trust as a three-place relation that is ubiquitous, and inherently risky. The dominance of three-place trust thus explains the widespread preoccupation with this very real epistemic problem as a problem of trust. Paul Faulkner has outlined what he sees as The Problem of Trust, which he takes to be a “sceptical problem”:
(i) we need to rely on another but recognise that doing so could have a worst outcome – we rely and the other proves unreliable, and – (ii) we know that this interaction is a one off (or one of a determinate number), and – (iii) we are entirely ignorant of the other’s individual motivations but recognise a general motivation to be unreliable. (2017: 111)
Faulkner says that given these facts, “reliance would seem to be irrational, and if it is, then a fortiori so too is trust” (2017: 111). Notice that Faulkner comes to this problem via his belief that trust is three-place – essentially involving reliance – and pervasive. He says:
thus, as with other sceptical problems, the issue is how to reconcile the philosophical result that reliance seems frequently to be irrational with the everyday fact that trust is pervasive. The difficult question is to account for how such reliance is rational under these conditions (2017: 112).
These ideas – that trust is pervasive or ubiquitous, three-place, and is inherently a matter of reliance and belief – are all apparent in Faulkner’s comments and they underlie much of the literature on trust, driving the concern that our trusting practices are irrational, and so also the many proposed solutions.
I.1 Trust: in the Philosophical Literature 26
Among those philosophers concerned with epistemic problems associated with testimony for example, the idea that trust is a ubiquitous three-place relation is so well accepted that for them, the ‘to do X’ in (A trusts B to do X) just means to tell the truth: (Faulkner 2017; E. Fricker 2006; Hinchman 2005; Keren 2007; McMyler 2011; Moran 2005). However, as Pamela Hieronymi put it “calculating the likelihood of her veracity, [is] precisely not trusting her. You are instead treating her like a good thermometer” (2008a: 222). What Hieronymi appears to be getting at here, is the worry inherent to three-place trust that if trust is to be differentiated from belief then trust cannot be just a cold calculation of our interlocutor’s veracity. Moreover, as Lars Hertzberg has pointed out, such an approach “disregards the fact that trust also enters into human relations in ways that have little to do with accepting the truth of statements” (1988: 309). Notwithstanding some scattered comments like Hertzberg’s, recognition that trust might be different from the three-place understanding is rare. Indeed, philosophers who think trust is a three-place relation are much more likely to not make sharp distinctions between the problems associated with trusting and the problems associated with our epistemic reliance on testimony, speaking at times as though the one just meant the other (Coady 1992; Daukas 2006; Faulkner 2007b, 2011; Foley 2001; E. Fricker 1995; Goldman 2001; Hardwig 1991; Jones 1999; Koenig and Harris 2007; McMyler 2011; Zagzebski 2012). The extent to which trust and testimony have become intertwined in epistemology is too great to explore in detail. However, Russel Hardin provides a typical example when he writes that trust is:
a cognitive notion, in the family of such notions as knowledge, belief, and the kind of judgment that might be called assessment. All of these are cognitive in that they are grounded in some sense of what is true…. The declarations ‘I believe you are trustworthy’ and ‘I trust you’ are equivalent. (2002: 7)
Asserting that trust and belief are effectively equivalent might sound like the sort of claim that is likely to be rare among professional philosophers. Let me provide just a few more examples that are not dissimilar from Hardin’s. Some have thought that trust is merely a ‘rational gamble’ that cooperation with others will ultimately payoff (Bates
I.1 Trust: in the Philosophical Literature 27
1988) others have thought: that trusting is a process of updating our expectations of other’s behaviour from the evidence we gain in our interactions with them (Gauthier 1986: 156); that trust is nothing but a rational belief about the character or the probable behaviour of another person (Dasgupta 1988; Gambetta 1988); or that trust is merely one aspect of self-interest driven and goal-oriented behaviour (Coleman 1990). All such outlooks put trust firmly in the realm of an instrumental rationality; as Philip Pettit put it “trusters identify reasons to trust others and trustees show that those reasons are good reasons” (1995: 202). Equating trust with other cognitive notions such as belief is one commonality among many of those who adopt three-place trust. Another is to view trust as inherently risky. Indeed, the riskiness of trust is generally held to be a central characteristic of trustful interaction (Lahno 2001: 171). Trust is risky – on three-place accounts – necessarily. The explanation given for this claim, is simply that if trust were not risky, then we would have no need to trust. This reflects the deeply held intuition that trust both goes ‘beyond the evidence’ and is ‘resistant to evidence’ captured in the following quotations:
Trust involves the risk that people we trust will not pull through for us; for, if there were some guarantee that they would pull through, then we would have no need to trust them. (McLeod 2015: Section 1)
One does not actually trust someone to do something if one only believes they will do it when one has evidence that they will. (Faulkner 2007b: 876)
[trust] go[es] beyond what the evidence supports. (McGeer 2008: 240)
What is the rational relation between trust and evidence? The dominant view among philosophers is that, in some sense, the importance of trust is that it goes beyond the evidence. If you are following the evidence, then you are not trusting. (T. W. Simpson 2017: 117)
These quotations expose a deep worry that trust is the sort of thing that necessarily disappears in the light of knowledge/evidence. This worry deepens even further when it is added to the idea that trust is pervasive.
I.1 Trust: in the Philosophical Literature 28
For example, it has been proposed that we must have good or at least decent reasons for trusting others, most especially with regard to that which is important (E. Fricker 1995). Russel Hardin suggests that if my trust in you is rational, then “I make a rough estimate of the truth of [the] claim … that you will be trustworthy under certain conditions … and then I correct my estimate, or ‘update,’ as I obtain new evidence on you” (Hardin 2002: 112). Such reasons for our estimates or for our “trust updates” – according to Hardin – could come from inductive generalisations based on experience, or from our knowledge that there are social norms of trustworthiness (Hardin 2002: 130). Yet while Hardin takes it to be the case that our ‘reasons to trust’ must come in the form of direct evidence of trustworthiness, others have seen less obvious candidates. Philip Pettit for example, argues that people are naturally esteem-seeking, and so, they will do what we entrust them to do because they want to continue to be held in high esteem (1995). Yet this view is compatible with trustees’ acting out of self-interest, which some – for example: Wanderer and Townsend (2013: 9) – have argued is not compatible with trustworthiness. Also, Karen Jones says that it may be possible for our trust to be justified even when we our prediction of trustworthiness is not justified, since “our evidence for trusting need not be as great as the evidence required for a corresponding justified prediction” (1996: 15). While many such discussions are concerned with rationalising the trust involved in reliance, Paul Faulkner has argued that it is trust itself which rationalises reliance because “trusting and being trustworthy are actions that we intrinsically value; and in valuing behaviours that are so described we think that, other things being equal, we have a reason to act in these ways, and acting in these ways is the right thing to do” (2017: 125). The idea that trust can serve as its own justification for itself, or for reliance, is surprisingly quite a popular approach, with another form of this argument stemming from the recognition that there may be ‘therapeutic reasons’ to trust. The central idea behind therapeutic trust is that trust (reliance) might increase trustworthiness (reliability) since knowing that we are trusted (being relied upon) itself provides us with a reason to prove trustworthy (reliable). When trust is conceived of as a kind of reliance, then, at least in some cases, trust seems to increase the likelihood of trustworthiness, and so, this could provide the justification for trusting (McLeod 2015). Philip Pettit
I.1 Trust: in the Philosophical Literature 29 argues along these lines as well, claiming that we can rationally invest trust in others without any justification (or evidence) that the person we are trusting is trustworthy, because the act of trust itself can prove inherently motivating (1995: 220). Against this though, Victoria McGeer has argued that while this mechanism does account for some of the trustee’s trust-responsiveness, it cannot serve as an account of trust, as it is not a robust enough mechanism to ensure the rationality of trust (2008: 252). Like McGeer, Pamela Hieronymi is suspicious about the idea that therapeutic trust can make trust rational. She maintains that the ends for which we trust cannot provide reasons for us to trust in the first place, since trust always involves a trusting belief. Hieronymi claims that therapeutic trust is not pure or full-fledged trust, since people can rightfully protest that they are not really trusted in such cases (2008b: 230). For a similar argument, see Lahno (2001: 184-85). Yet Hieronymi’s argument against therapeutic trust rests on the claim that trust involves belief, and this claim too is contentious. Domenicucci and Holton (2017) have recently pointed to neuroscientific evidence suggesting that belief is not central to trust, since a higher level of Oxytocin, which they take to be correlated with a higher readiness to trust, appears not to increase the subject’s belief that the risks will be rewarded; they cite (Kosfeld et al. 2005: 154). This research, though, will only be relevant to philosophical arguments about trust if trust was in fact the thing that was being ‘tested for’ in this particular study. If this dissertation’s central point that trust is not a three-place relation is correct, then in many instances, what is being examined in such settings is in fact not trust, but cooperation. Thomas Simpson says that: “Not only is it compatible with trust that one is following the evidence. Further, and stronger, there are times when one’s trust is appropriate only if one is following the evidence.” (2017: 177). For Simpson, trust is a “rationally responsive mental attitude” (2017: 178), and as such, its “striking insensitivity to evidence” (2017: 181), is not an inattentiveness to evidence. Simpson’s attempt at rapprochement thus relies on a distinction between insensitivity to evidence, and inattentiveness to evidence, which he thinks marks the relevant difference between ‘rational trust’ and ‘irrational trust’. Lastly, in a wildly different move altogether (though with the same goal in mind), Victoria McGeer has argued that it is our capacity to hope that significantly underwrites our capacity to trust, and that since it is rational to
I.1 Trust: in the Philosophical Literature 30 hope for the best, then trust is rational as well (2008: 237-54). Elsewhere, McGeer has argued that hope is an essential feature of human agency, and so on her view, trust and human agency are linked (2004). What should be apparent by now, is that taking trust to be schematically three-place has cascading logical effects. First, if trust is (A trusts B to do X), then the boundaries between trust and other three-place relations becomes indistinct, and trust is more likely to be confused with cooperation, reliance or belief/acceptance of testimony. Second, since we rely on/believe/cooperate with other people all the time, then trust becomes ubiquitous: a thing upon which all our knowledge – including self-knowledge – hangs, since even when it comes to ourselves, we have to ‘trust’ (rely on/believe) our own sense perception. Trudy Govier has certainly come to this conclusion:
one may be called on to trust one’s perceptions and observations, interpretation of events and actions, feelings and responses, values and evaluation, memory, judgment, instinct, common sense, deliberation, choice, will, capacity to act, flexibility, competence, talent, and ability to cope with the unexpected. (1993: 108)
Like Govier, Thomas Simpson speaks of trust’s “near ubiquity in social interaction” (2017: 178), which makes trust indistinct from cooperation “Trust simply is the belief that another is likely to cooperate” (2017: 180). Others who speak of trust as though it were essentially a matter of cooperation include: Putnam et al. (1993: 171), Gambetta (1988), and Luhmann (1979). On three-place trust, so many concepts – trust, cooperation, reliance, belief – become indistinct. It is not surprising that philosophers embracing a three-place conception have become concerned with separating them out again. Indeed, the project of distinguishing trust from reliance and belief has become another great flashpoint in the current philosophical debate. Most suggestively, this only occurs when a three-place account of trust is accepted. On a two-place account, trust is much more straightforwardly distinct from any other three-place relations (such as reliance/belief) as discussed in detail in I.2. For those who do not adopt three-place trust, the epistemological debate and the project of distinguishing trust from reliance/belief have failed to hold much significance. Bernd Lahno’s account of trust as an emotional attitude, for example,
I.1 Trust: in the Philosophical Literature 31 views trust as a mechanism that helps us to cope with uncertainty, which is beyond the direct control of reason and so cannot be understood as a result of rational reflection (2001: 173). John Weckert similarly deemphasizes calculation, i.e. the rational assessment of beliefs and risks. He is critical of such epistemically-based accounts for overlooking for example, the trust that children have of their parents (2005). Mark Coeckelbergh also has a view of trust that emphasises the social aspect of trust, while de-emphasising the rational individualistic view (2012). Sadly, such rejections of the dominant view, and the subsequent concerns that trust and rationality present a pressing problem, are rare. Among those who adopt three-place trust, even if agreement were to be reached regarding which approach best reconciles trust and rationality, further substantive questions arise, such as: must the reasons that confer rationality be accessible to the truster in order to count? On this point, there is once again considerable disagreement, with some arguing that the reasons must be available for the trust to be rational (Hardin 2002; Hieronymi 2008b), and others arguing that they do not (Baier 1986; McLeod 2002; Webb 1993). Perhaps all we need is a list of ‘justifiers’ for trust, which the truster could consider in determining when to trust (Govier 1998; Jones 1996), or perhaps trust is rationally justified if and only if it is formed and sustained by ‘reliable processes’ (Goldman 1992: 113), whatever that turns out to be. As it makes trust a matter of practical action – of doing/believing something – under a three-place conception of trust, the tension between trust and rationality will always remain, and further questions about what “counts as a reason”, and then, what counts as a “reliable form” of that reason, will always arise. This is no bad thing in principle, but only if we have our picture of the thing under investigation right, and the questions we think are arising are actually arising. This dissertation, in its small way, is a call to refrain from delving too far into the thicket of epistemology with regard to trust, as the underlying assumptions driving this need to reconcile trust and rationality are problematic in a number of ways, one of which was recently captured nicely by Domenicucci and Holton:
It has often been contended that trust involves a form of vulnerability on the part of the person trusting; and this is often developed as a kind of ignorance. In some accounts this is
I.1 Trust: in the Philosophical Literature 32
understood as ignorance of whether the trust is well-placed: we trust insofar as we cannot be sure that the trusted will not let us down. That, however, is implausible: on such an account, as our knowledge of a person grows, so our trust in them must diminish. (2017: 151)
As Domenicucci and Holton say, clearly the very premise that trust and knowledge are somehow fundamentally and eternally conflicting is not reflected in human experience; we trust those we know better more, not less. Paul Faulkner has recently also gone against the conventional wisdom, asserting that trust is “unproblematic when one knows that the trusted is trustworthy” (2017: 110). Difference of opinion on this point is reflected in the Stanford Encyclopedia of Philosophy entry on trust, which states that “trust tends to give us blinkered vision, making us resistant to evidence that may contradict our optimism about the trustee”, while later acknowledging that it would be odd if it turned out that “what made an attitude justified destroyed that attitude” (Section 2). Part of the problem here, is that while the ‘riskiness of trusting’ is frequently discussed in the literature, it is not always clear precisely which notion of risk philosophers intend. The term ‘risk’ sometimes appears in the literature in a non- technical sense, as a rather vague situation in which it is possible (but not certain) that some undesirable event will occur. Discussion often slides between the notions of risk as the unwanted event, risk as the cause of that unwanted event, and risk as the probability of the unwanted event occurring. Since each conception of risk will have differing epistemic and ethical implications (Hansson 2018), there is a need for philosophers adopting three-place accounts to identify exactly which notion of risk is supposed to be relevant to trust. This is particularly relevant to those writing on the problems associated with ‘trusting testimony’, as in those discussions, the risk is taken to be just that we might acquire a false belief, since what we trust our interlocutor for is the truth (Faulkner 2007b). Yet since there are other candidates as well, such as a loss of self-respect (McLeod 2015), or unpleasant feelings of resentment or betrayal (Baier 1986: 235; Hieronymi 2008b; Jones 1996), mitigating against one ‘risk’ may do nothing to prevent the others. In summary, this subsection has presented an assortment of literature that takes The Problem of Trust to be an epistemic problem, which essentially involves a conflict
I.1 Trust: in the Philosophical Literature 33 between what trust requires of us with regard to risk and evidence, and what rationality requires of us with regard to risk and evidence. In doing so, this subsection has revealed that those who worry about this problem, take trust to be a three-place relation, and a ubiquitous feature of human life, present in virtually all interpersonal and informational transactions. The idea hopefully seeded is that by doing away with a three-place account of trust, we might avoid running into this problem altogether. Though this ‘justification problem’ will always remain as a problem of knowledge, by abandoning three-place trust the valuable concept of trust might rightly be spared from this particular philosophical snare.
I.1.4 Voluntariness of trust
This subsection addresses literature concerned with the voluntariness of trust. If trust is voluntary, then it makes sense to hold people both morally and epistemically accountable for their trust. Yet if trust is not voluntary, then a question hangs over how much – if at all – we can hold people to account for their trust. It is evident from the literature presented already that three-place trust is difficult to distinguish from either reliance or belief. Accounts which conflate trust with reliance are sometimes referred to as ‘non-cognitive accounts of trust’, while accounts that conflate trust with belief are sometimes referred to as ‘cognitive accounts of trust’, as described by (T. W. Simpson 2017). The debate over the voluntariness of trust results directly from this situation: conflation of three-place trust with reliance (non-cognitive accounts) make it seem obvious that we can decide to trust, while conflation of three-place trust with belief (cognitive accounts) make it seem equally obvious that we cannot. Hence, similar to the division over the rationality of trust, this particular division in the literature owes itself again to that underlying picture of trust as a three-place relation. To begin, trust is “obviously voluntary” when it is construed as non-cognitive (an action), as Benjamin McMyler writes:
In one general respect, it should be uncontroversial that we can decide to trust. We can decide to trust a mechanic to fix the car, decide to trust a neighbour to look after the kids, or decide to trust a friend to show up on time. We often find ourselves in situations in which we must deliberate about whether particular others are worthy of our trust concerning
I.1 Trust: in the Philosophical Literature 34
particular matters, and when the upshot of this deliberation is positive, it is natural to describe ourselves as having decided to trust. (2017: 161)
This general respect in which it should be uncontroversial that we can decide to trust, as identified by McMyler, only arises on a schematically three-place account, since it is only with a three-place account that trust can be viewed as an action, something that we can decide to do. On two or one-place accounts, trust is instead something that we are (one place trust) or something we embody/have/experience in relation to another person (two-place trust). Those philosophers who have been led to accept voluntarism about trust, have typically been so led by a combination of their acceptance of the three-place account, and their belief that trust is non-cognitive (Faulkner 2014b: 1979; Holton 1994: 63). Perhaps the most compelling and detailed argument in favour of the voluntary nature of trust was made by Richard Holton in his 1994 paper “Deciding to Trust, Coming to Believe”. This paper has been discussed in a number of contexts already (see I.1.1 and I.1.3), and in it, Holton rejects the idea that trust is cognitive (involves belief), with an analogy about riding a bicycle “to ride a bicycle you do not need to believe that you can do so. You need to act as you would act if you did believe” (1994: 64). In another example, Holton says that we can decide to trust a friend to be honest even if we do not believe that they will be honest. Indeed, if we already believed what the friend was saying, then – Holton says – we would have no need to trust them (1994: 74). Contained in these examples, which Holton takes to be paradigmatic examples of trust, is thus both the rejection of trust as cognitive, and the statement that trust is a non- cognitive action. For Holton – at least in 1994 – trust does not require belief, only reliance. This rejection of the idea that trust requires belief was shared by Paul Faulkner, when he said that “A can decide to trust S to φ just because trust need not involve the belief that S will φ” (2014b: 1979). As Stephan Darwall put it “Holton’s diagnosis of the difference between trust and reliance, then, is that trust is reliance from the participant stance” [his emphasis] (2017: 38); later, Darwall also points out that this is more or less Faulkner’s position, or at least it was in 2007.
I.1 Trust: in the Philosophical Literature 35
While Holton and Faulkner’s positions have changed somewhat by 2017, their earlier writings have been highly influential, leading many philosophers to ponder whether their denials of any necessary cognitive elements to trust are sound. Phillip Nickel disagrees with Holton that trust is a reactive attitude, instead taking it to be a moral attitude involving the ascription of obligations (2012: 309-31), yet his account also allows him to conclude that trust can be voluntary. Similarly, the approach taken by Victoria McGeer – trust as involving hope rather than belief – allows for the idea of voluntary trust. Because she views trust as “forward looking, anticipating the transformative effects of extending our trust” (2008: 242), she is able to take cases of therapeutic trust to be genuine cases of trust, and, if it is possible to trust therapeutically in the way McGeer describes, then trust must in some sense be voluntary. Common to views that see trust as voluntary is the idea of trust as something that we do in lieu of belief; a reliance act that is hopeful, moral, or ‘from the participant stance’. This denial of the belief element as necessary to trust is significant, since if trust involved belief then it becomes much harder to maintain that trust can be voluntary, given that most people are not voluntarists about belief (McLeod 2015). Against these non-cognitive accounts though, many have held the intuition that it is not possible to decide to trust for motivational, rather than epistemic reasons. This view is naturally common among those who do not make sharp distinctions between trust and belief, i.e. that the trustee is trustworthy (Daukas 2006; Hardin 2002, 2004; Hieronymi 2008b; Lahno 2001; Webb 1992). An early ‘cognitivist about trust’ was Annette Baier, who questioned whether people can trust “simply because of encouragement to trust” (1986: 244). She writes:
‘Trust me!’ is for most of us an invitation which we cannot accept at will—either we do already trust the one who says it, in which case it serves at best as reassurance, or it is properly responded to with, ‘Why should and how can I, until I have cause to?’ (1986: 244).
Some years later, Karen Jones wrote that it is an “obvious fact about trust” that it cannot be willed (1996: 15), since “we are generally not aware of our trusting and seldom bring it sufficiently clearly before our minds to endorse or reject it” (1996: 14). For Jones, trust “can no more be sincerely adopted in the face of a known and
I.1 Trust: in the Philosophical Literature 36 acknowledged absence of such grounds than a belief can be adopted in the face of a known and acknowledged lack of evidence” (1996: 16). Despite her strong intuition that the cognitive elements to trust cannot be dispensed with, Jones shows an appreciation of the sense in which trust can obviously be willed. She distinguishes trusting and entrusting, which is in actions, and actions – she agrees – are “paradigmatically, things that can be willed” (1996: 18). Thus, Jones deals with the problem not by rejecting the very idea that trust is a three-place kind of belief/reliance, but by distinguishing between reliance-trust (entrusting) and belief-trust (trusting). For Jones, there are two kinds of three-place trust, rather than – as this dissertation argues – one kind of two-place trust. Jones agrees that her “entrusting model” of trust does not sufficiently bring out the affective component of trust, because it “obscures the importance of optimism about the goodwill of another” (1996: 20). Jones says that trust and distrust are “partly constituted by patters of attention, lines of inquiry, and tendencies of interpretation” (1996: 20) hence, while it is impossible that trust be directly willed, it should be possible to cultivate them by other methods, such as controlling our patterns of attention, lines of inquiry, and tendencies of interpretation:
the rape victim whose trust in others has been shattered might set about cultivating trust because she sees herself as someone who is free-spirited and bold, and she does not wish to be the kind of person who is timid, protective of the self, and on the lookout for betrayal. (1996: 23)
This idea that trust must be either cognitive or non-cognitive, a belief or an action, historically has been accepted uncritically. Very recently though, Benjamin McMyler has presented a sophisticated view which aims to show why this is a false dichotomy. McMyler makes the case against the voluntariness of trust by separating the claim of non-cognitivism about trust from the claim of voluntarism of trust. By seeing that there are in fact these two distinct claims being made by philosophers discussing the voluntariness of trust, McMyler is able to argue that although trust is non-cognitive, it is nevertheless non-voluntary:
I.1 Trust: in the Philosophical Literature 37
non-cognitivism about trust and voluntarism about trust are often run together in philosophical discussions of deciding to trust such that evidence for one is taken to be evidence for the other. I think this is a mistake… even if one accepts non-cognitivism about trust, one should reject voluntarism. (2017: 163)
The way McMyler accomplishes this move, is essentially by rejecting an account of trust that is rooted in action, and assuming a ‘psychological attitude’ account instead. The result amounts to less than a total rejection of voluntarism, at least as some may understand it:
The fact that trust is non-voluntary does not mean that we are passive with respect to our trusting; that we can only manage our trust in the way that we can manage our pains and sensations. We can directly exercise agency over our trusting by settling the question (or set of questions) that the attitude of trust embodies, by actively taking the world to be a certain way. We can, in this respect, decide to trust others. Insofar as deciding to trust others is deciding to take the world to be a certain way, we cannot decide to trust others for reasons that we ourselves do not take to show the world to be this way. But this is not a limitation on our powers. It is simply a reflection of the fact that interpersonal trust is an attitude toward the world. It is a distinctive attitude to be sure, an attitude towards the soul, and this makes it liable to abuse and betrayal in a way that many other attitudes are not. (2017: 175)
McMyler says that trust, like belief but unlike action, cannot be adopted for reasons of inducements like threats or offers (2017: 168) and so trust is a “psychological attitude that is not voluntary” (2017: 172, footnote 15). His view thus has implications for the idea of therapeutic trust, since such occurrences would be understood as the actions we perform when we have good reasons to trust, rather than the trust itself – a result this dissertation is in agreement with. Indeed, McMyler’s view shares a number of similarities with the view put forward here, especially in its focus on the private, psychological nature of trust, and the result that trust can be voluntary, but only in the slower, cultivator sense.
I.1.5 Summary of Trust: in the Philosophical Literature
Section I.1 examined contemporary philosophical literature on trust. The aim of this section, rather than to appraise particular views, was to weave a coherent narrative
I.1 Trust: in the Philosophical Literature 38 within which the view presented in this dissertation can be situated. To that end, I.1 provided an overview of the current state of the philosophical literature on trust, identifying the dominant view, and the questions it raises. In Subsection I.1.1, we saw how among those that endorse a logically one-place view, there are a number of characterisations: trust as an attitude, emotion, virtue, character or personality trait, yet on closer inspection, most philosophers actually endorse a hybrid one/three place view. In Subsection I.1.2, we saw the extent to which trust is conceived of as a three-place relation in the philosophical literature – even when other conceptions are acknowledged, the three-place view is still held to be primary, or a legitimate and important kind of trust. In Subsection I.1.3, we saw how this dominance of three-place trust has led to The Problem of Trust being understood as an epistemic problem, which essentially involves a conflict between what trust requires of us with regard to risk and evidence, and what rationality requires of us with regard to risk and evidence. Lastly, in Subsection I.1.4, we saw how three-place trust leads trust to become confused with both reliance and belief, and how the debate over the voluntariness of trust results directly from this confusion.
I.2 Dominant Views on Trust: Problems 39
I.2 Dominant Views on Trust: Problems
In section I.1, the aim was to provide an overview of how the underlying three-place logic has led to the questions that philosophers currently pay attention to in relation to trust. In this section, the aim is to provide a fuller understanding of three-place trust itself, and to present its limitations. As the underlying logic is shared by all three-place views, I.2.1 begins with the bare schematics of a three-place relation. Then, I.2.2 presents three-place trust in its most simplistic form and addresses a recent (2017) argument for why we should adopt this form from Philip Nickel. As a number of philosophers have taken issue with simplistic three-place trust and provided various fixes to it, I.2.3 moves on to consider these ‘three-place trust plus’ versions, which essentially add to a simplistic account of various things, such as goodwill or a belief in the trustworthiness of the trustee. Finally, I.2.4 presents some reasons as to why three- place accounts remain dominant today, despite their limitations.
I.2.1 Three-place trust: logic
Three-place trust encompasses all views which take trust to be a three-place relation, logically, or schematically speaking. These views are also referred to as three-place notion trust or three-part trust. The first thing to do when probing the idea of trust as a three-place relation, is to get clear on what a three-place relation is. We can do this by looking at some examples of other three-place relations that are often confused with trust. Reliance is a three-place relation: A relies on B to do X. Testimony is another: A believes/accepts B’s word about X. Cooperation is when two (or more) parties, A and B, cooperate they do so to some specific end X. There are other examples, but these should serve to get the point across. Quite straightforwardly, to claim that trust is a three-place relation is the claim that trust involves three elements: a truster (A), a trustee (B), and that which is entrusted (X). Though the details can vary considerably, as was established in section I.1, most philosophers endorse and prefer to examine a three-place conception even when they acknowledge other conceptions, such as “general trust”, “one-place trust”, “fully- fledged trust”, “two-place trust” or “self-trust”. So, the bare schematic of three-place trust is:
I.2 Dominant Views on Trust: Problems 40
(A trusts B to do X)
Or, as Thomas Simpson writes: “trust is a tripartite relation, relating a trustor, to someone trusted, regarding some action that the trusted will perform. Schematically, A trusts B to φ” (2017: 178). To give this some substance, it would be good to introduce some examples from the literature. Paul Faulkner writes:
Here are some examples of trust. Leaving one’s closed diary on a desk where one knows one’s partner will see it. Not asking for a second quote when a mechanic says one’s car needs lots of work… Following a stranger’s directions. (2017: 109)
Similarly, Richard Holton uses a number of three-place examples, including: trusting an employee with the till, trusting someone to catch you, and trusting someone with your car (1994). Annette Baier speaks of trusting our enemies not to fire at us when we lay down our arms and put up a white flag (1986: 245), and Bernd Lahno uses the example of a mother trusting the babysitter to care for her daughter (2001). There are countless such examples to be found in the literature, all sharing the same underlying schematic structure. The most important thing for our purposes to notice about these examples, is that when trust is viewed this way, both trust (and subsequently trustworthiness) are public – they are actions, a matter of relying on a person, not a private matter of how we regard a person. If we take these examples at face value, then it seems that what determines whether trust is present in any case is the behaviour, publicly observable, of the truster. To see this, let us take a putative instance of trust that is three-place in logical structure, and similar to the examples above:
Hamish trusts Ana to hang out the washing.
In this example, the implication is that Ana is trustworthy if and only if she hangs out the washing. The problems with this implication are addressed in II.3. For now, let us focus on the truster, Hamish. What makes it the case that Hamish is trusting Ana? It seems like the only relevant facts here are to do with Hamish’s publicly observable behaviour. If Hamish is sitting inside with a cup of tea waiting for the washing to be
I.2 Dominant Views on Trust: Problems 41 hung, then, he is trusting. If Hamish is checking up on Ana, hanging the washing himself, or arranging for someone else to hang the washing instead, then he is no longer – or perhaps never was – trusting. Without some further specifications about what trust might involve, a three-place account leads us to believe that what goes on inside Hamish – his private thoughts, opinions, motives, beliefs, attitudes, emotions etc. – are not relevant. Without some further conditions added to the basic schematics of three-place trust, Hamish may behave as though he trusts Ana, all the while thinking her useless, careless, and incompetent. He can have any kind of private motive towards her or emotional reaction to her, so long as he keeps sitting there quietly drinking his tea and not behaving as though he doubts that Ana will hang the washing. Advocates of simplistic three-place trust are quite content with this result and are not particularly concerned with the fact that it collapses any meaningful distinction between trust and reliance. It is an especially prevalent view in areas which straddle the divide between science and the humanities: the political, social, and economic sciences. For disciplines that aim for scientific rigour, equating trust with reliant or cooperative behaviour makes sense, since doing so makes trust far easier to verify empirically. For those looking to explain complex social, economic or political behaviour, this simplistic three-place trust stripped bare of any private psychological elements, has been used to capture countless phenomena, from the behaviour of individual strangers playing ‘investment games’ (Berg et al. 1995), to the behaviour of states (Eric M Uslaner 2018), economies (Diekhöner 2017), exchange relations in a business environment (C. Lane and Bachmann 1998), and even to how human beings relate to cryptocurrency, sushi, and self-driving cars (Botsman 2017). Simplistic three-place trust is the ‘Theory of Everything’ to the social sciences.
I.2.2 Three-place trust: simplistic
This subsection presents three examples from the literature which have embraced simplistic three-place trust, from: economist Partha Dasgupta, political theorist Eric Uslaner, and philosopher Philip Nickel. I pay particular attention to Nickel’s view, as he argues extensively and explicitly for the simple (or as he calls is “unrestricted”) version
I.2 Dominant Views on Trust: Problems 42 of three-place trust, while Uslaner and Dasgupta adopt it less reflectively. This subsection aims to show that simplistic three-place trust fails to make trust distinct from other three-place relations, a situation that some do not appear to consider, and others consider unimportant or ever valuable, as in Nickel. We begin with Uslaner and Dasgupta. Eric Uslaner is interested in the relationship between democracy and trust and takes trust to be a kind of “rational gamble” that cooperation with others will ultimately pay off, as well as a commitment to “pro-social behaviour” (1999: 124). Trust, as he sees it, is individuals expecting other individuals to cooperate and do the right thing (distrust is expecting the opposite). This understanding leads Uslaner to prescribe a remedy for declining social capital. He suggests little league baseball as a sort of bootstrapping remedy “sports build social capital because they build self-confidence and teach respect for rules” (1999: 146). Uslaner’s reasoning is that if we do not trust one another, if we do not expect others to cooperate with us or behave in a pro-social manner, then we need to engage with one another in a rules-based activity, such as sport, in which we can come to possess the knowledge that ‘the others’ will cooperate with us and behave in a pro-social manner according to the rules. Hence we can begin to trust them again. On this conception, trust arises because of a lack of knowledge of how others will behave, and so the remedy for distrust is simple – provide them with the knowledge. Uslaner’s advice may indeed be effective for increasing cooperation, peace, and civic participation. However, it is not clear that trust and trustworthiness ever enter into it. If to be ‘trustworthy’ just means ‘to behave in a cooperative and pro-social manner’, what Uslaner is prescribing is that we increase cooperativeness and pro-social behaviour by increasing cooperativeness and pro-social behaviour. Granted, prescriptions of this nature can be extremely valuable. Yet it may be possible to achieve pro-social behaviour without increasing trust and trustworthiness at all. Suppose, for example, a historically uncooperative and antisocial society, marked by a history of intense infighting, were taken over by a strict authoritarian ruler who enforced cooperative and pro-social behaviour upon their citizens through fear of punishment or even death. With these incentives in place, we can expect cooperation and pro-social behaviour to increase. Yet we would have no reason to think that this
I.2 Dominant Views on Trust: Problems 43 change would cause any change in the underlying trust between individual citizens. Such a society may appear to be ‘high-trust’ from the outside – orderly, civil, and compliant – but this may just be an illusion of trust. Cooperation and civic participation are not the same thing as trust, since they can be compelled by other things, fear for example, mutual threats, or nationalistic sentiment. Not too dissimilarly from Uslaner, Partha Dasgupta interprets trust to be “correct expectations about the actions (his emphasis) of other people that have a bearing on one’s own choice of action when that action must be chosen before (his emphasis) one can monitor the actions of others.” (1988: 51). Hence, for Dasgupta, trust is A correctly expecting B to do X, when A lacks information about how B will behave. He says that trust is “not dissimilar to commodities such as knowledge and information” and emphasises that “the clause concerning the inability to monitor others’ actions in my definition of trust is crucial. If I can monitor what others have done before I chose my own action, the word ‘trust’ loses its potency” (1988: 51). The Problem of Trust for Dasgupta, like Uslaner, ultimately boils down to a lack of knowledge of how others will behave. This leads him to claim that “the problem of trust would of course not arise if we were all hopelessly moral” (1988: 53). By “hopelessly moral” Dasgupta clarifies that he means “always doing what we said we would do in the circumstances in which we said we would do it.” (1988: 53). Setting aside the detail that doing the moral thing may sometimes require of us that we act in a manner that is at odds with what we said we were going to do, this conception of trust – simple three- place trust – as we can see, is difficult to distinguish from other similar three-place relations, such as reliance, belief, or cooperation. For Uslaner and Dasgupta, the difficulty in distinguishing simplistic three-place trust and reliance seems not to have been deemed so problematic as to warrant a re- think of the underlying conception of trust. Their understanding of trust as a way of interacting with things in the world, is widely adopted in their fields. Yet in philosophy, the worry that simplistic three-place trust is too similar to reliance to be of any distinct use has had an impact, as we shall see in the following subsection, I.2.3. Yet one philosopher has gone to the point of arguing explicitly in favour of simplistic three- place trust.
I.2 Dominant Views on Trust: Problems 44
In a recent book contribution entitled “Being Pragmatic about Trust”, Philip Nickel argues that we must determine the concept of trust pragmatically, i.e. by our “scientific explanatory aims” (2017: 209). Put simply, Nickel’s argument is that we should adopt an unrestricted theory of trust that does not require trusters to have any psychological elements – emotions, attitudes, motives, beliefs and so on – since this restricts the number of instances of cooperation that can be “explained” by trust. Nickel says that trust “should be able to explain cooperation”, and that failures of trust (by which I take him to mean failures of trustworthiness), “should help explain the emergence of cooperation-enabling institutions” (2017: 195). For Nickel, unrestricted or simplistic views of trust – such as those adopted by Uslaner and Dasgupta – that take trust to be no more than the disposition to cooperate with or rely on others, should be favoured over restrictive views which require the trusting person to have some further attitude in addition to this disposition (2017: 196). Both trust and trustworthiness as presented in this research project would be classed as ‘restricted’ in Nickel’s terminology, since the psychology of the truster very much determines their presence. Nickel is very clear that he is arguing for the adoption of simplistic three-place trust on pragmatic grounds, and so before evaluating his argument, it may be helpful to appreciate his broader motivation. Nickel claims that most philosophical accounts of trust stand in opposition to pragmatism (2017: 200), and the approach taken in this dissertation may at first appear to do so, in its insistence on the distinction between the objectively observable so-called “practical effects” of trust and trust itself. One cannot, I argue, provide an adequate account of trust without reference to something other than its practical effects. This may at first appear to leave this dissertation open to the pragmatist’s complaint that it relies on “tender-minded a priori intuitions” about trust, while ignoring the “hard facts”. Yet those philosophers who have argued that there is something more to trust than the simplistic three-place picture reveals, have not come to do so just by thinking about it, revealing to themselves hidden “a priori truths”. Instead (though they may very well do so from an armchair) they have put forward considered views on trust by reflecting on their own and others’ lived experience as trusters. This is precisely to take in and examine the “hard empirical evidence”. Many accounts of trust which are restricted (in
I.2 Dominant Views on Trust: Problems 45
Nickel’s sense) are not at all “anti-pragmatic”. At the very outset, Nickel’s motivation for his approach is somewhat suspect, as it relies on the spurious claim that accounts of trust that go deeper than the “practical effects” of trust, are anti-pragmatic. For Nickel, it is legitimate to explain instances of cooperation as trusting or trustworthy not only when things like goodwill or optimism are absent, but even if things like fear, suspicion, greed, or selfishness are present. He says that we should leave open what might count as instances of trust so that “a range of possible motivations can potentially fit, instead of defining them away analytically a priori.” (2017: 199). As an example, he provides a description of one of the first encounters of European explorers of North America with Native American tribes. The account he offers us, which he takes to be an example of cooperation that we should be able to explain by appealing to the notion of trust, comes from the French explorer La Salle:
At nine o’ clock, doubling a point, [La Salle] saw about eighty Illinois wigwams, on both sides of the river. He instantly ordered the eight canoes to be ranged in line, abreast, across the stream ... The men laid down their paddles and seized their weapons; while, in this warlike guise, the current bore them swiftly into the midst of the surprised and astounded [Illinois people]. The camps were in a panic. Warriors whooped and howled; squaws and children screeched in chorus; some ran in terror, and, in the midst of the hubbub, La Salle leaped ashore, followed by his men. None knew better how to deal with Indians; and he made no sign of friendship, knowing that it might be construed as a token of fear. His little knot of Frenchmen stood, gun in hand, passive, yet prepared for battle. The Indians, on their part, rallying a little from their fright, made all haste to proffer peace. Two of their chiefs came forward, holding out the calumet, while another began a loud harangue, to check the young warriors who were aiming their arrows from the farther bank. La Salle, responding to these friendly overtures, displayed another calumet; while Hennepin caught several scared children and soothed them with winning blandishments. The uproar was quelled, and the strangers were presently seated in the midst of the camp ... Food was placed before them; and, as the Illinois code of courtesy enjoined, their entertainers conveyed the morsels with their own hands to the lips of [La Salle’s party]. (Parkman 1869: 158-59)
About this encounter, Nickel says “the unrestrictive view explains these outcomes in terms of trust, and the restrictive view cannot do so” (2017: 201), implying that this is
I.2 Dominant Views on Trust: Problems 46 obviously a severe failing of restricted accounts. But is it? Though there is clearly a ceasefire, and a suspension of outward hostilities, this interaction is far from friendly. Indeed, La Salle made “no sign of friendship” and upon approach his party “seized their weapons” and, having a “warlike guise”, came upon the Illinois people who are described as “surprised”, “astounded”, “in a panic”, and who “ran in terror”. Given these facts, it is not clear why we should be able to explain the ensuing ceasefire as an example of trust and trustworthiness. It is more accurate, and more useful to say that the Illinois Indians “made all haste to proffer peace” because of their desperation, fear or good sense, rather than their trust. This is especially so, seeing as we are told La Salle’s “little knot of Frenchmen stood, gun in hand, passive, yet prepared for battle”. We gain no practical benefits by describing tactical interactions involving guns, terror, power, and oppression as trusting. It is far more explanatory to say that the cooperation that occurred between these two fearful, suspicious and warlike groups was distrustful and strategic – a considered move on the part of the settlers to strike fear into the Indians by showing their strength, and an equally considered response by the chiefs to secure the safety of their people by compliance. Nickel, following Russell Hardin (2006), points out that such preliminary moves can result in trust developing (2017: 197), which may be true. Hostile violence can turn into strategic peace marked by cooperation, which can – in time – grow into trust. But the fact that such interactions can in the future produce trust is not a reason to describe them as already trusting. Nickel brings in trust as an explanatory device to account for cooperative behaviour that occurs before trust has properly developed, and in doing so, the unrestrictive view of trust actually restricts what we can say about such interactions. Nickel wants our “scientific aims to determine which account of trust we accept” (2017: 209) and so “philosophers should take the argument from explanatory potential on board as a way of grading accounts of trust” (2017: 212). Yet without getting into the ongoing debate in the philosophy of science about what counts as an explanation, there is general agreement even among those who favour pragmatism that everything a theory says about observable aspects of the world must be true. Explanations might omit some details, but they cannot say things that are outright false (Van Fraassen 1980) and it is
I.2 Dominant Views on Trust: Problems 47 just false to claim that interactions like this are trustful. How do we know this? Not by defining things away analytically a priori, but by examining the evidence: guns are used to make people cooperative when there is no trust. That is indeed one of their main functions. If there is trust between individuals, then they are not standing ready “prepared for battle”. The threat of violence here tells us all we need to know about the trust situation. Nickel claims that adopting simplistic three-place trust (unrestricted trust) would be of pragmatic benefit because “trust is messy” (2017: 209). He provides as evidence for this the claim that “our intuitive sense of what genuine trust is, is perhaps not shared universally” (2017: 209). This makes it unclear whether Nickel worries that trust is not universal, or that the concept of trust is not universal. He provides an explanation for the non-universality of trust/concept-trust, in a footnote:
In Dutch, ‘I rely on x’ and ‘I trust x’ can both be naturally translated as ik vertrouw [in, op] x, and ‘reliability’ and ‘trustworthiness’ can both be naturally translated as betrouwbaarheid. In French, the respective pairs are both translated as faire confiance en and fiable. Of course a distinction can be drawn, but it lies further from the surface than in English. (Nickel 2017: footnote 12)
Here, Nickel points out that the words ‘rely’ and ‘trust’ are close in both Dutch and French and takes this as evidence that trust/concept-trust is not universal. This is when, as he says, “a distinction can be drawn” between trust and its cognates even in these languages. This is puzzling. Supposing for a moment that the translation of ‘trust’ in language A is close to the equivalent of ‘hope’, and in language B is closer to the equivalent of ‘think’, then, this is not by itself evidence that trust/concept-trust is not universal. Translations of the German words Torschlusspanik (fear that time is running out to achieve lifegoals) or Fremdschaemen (feeling ashamed on someone else’s behalf) may differ considerably across languages, but this does not need to mean that these phenomena are not, in the relevant manner, shared. There may even be languages where the English word “trust” has no equivalent at all. We can come for the first time to possess words for things that we already have concepts for and are already doing with full awareness: ‘sexting’, ‘cybernating’,
I.2 Dominant Views on Trust: Problems 48
‘upspeaking’ and ‘barebacking’ for example. To demonstrate that people do not possess the concept of trust, we must actually demonstrate that they do not possess the concept of trust. To overthrow the common-sense idea that trust and trustworthiness are ‘universal’ i.e. shared human experiences like love, forgiveness, and betrayal, it is not enough to appeal to language differences, or even cultural differences in expression (this theme is explored in greater detail in II.3). Nickel’s argument is not just a denial that we might be able to produce a single coherent account of trust. He takes trust to fail to be a “natural kind” (2017: 210) in the strongest sense:
That medical scientists’ understanding of arthritis should determine the boundaries of the concept is plausible, but it is not similarly plausible that biologists, rather than psychologists, political scientists, economists, or philosophers, should determine the boundaries of the concept of trust. The second problem is that trust is a concept with heavy social and political significance in modern times (see, e.g. Baberowski 2014). Concepts like this, such as legitimacy and democracy, are public property, maybe even contested territory. The boundaries of such concepts are not dictated by experts alone. It appears, then, that there is no decisive reason to regard the concept of trust as being determined by a natural kind, nor by social facts that give experts authority over the concept. (2017: 211)
Somewhat misleadingly, Nickel may mean “determine” as “define” or he may instead mean something closer to “figure out”. I take no issue with Nickel’s claim that philosophers are not experts with freewheeling authority to define concepts. Yet this is true also of medical scientists. All of us are bound by what really is, and there is no reason in principal why a philosopher may not be able to use their specific skills to figure out the boundaries of the concept of trust. I have argued that if cooperation, reliance, and belief, can exist without trust, as they did between early European explorers of North America and Native American tribes, then they are not the same thing and it is better to adopt an account of trust that reflects this clearly. Here we can glimpse the first tangible benefit of the way trust is understood in this dissertation. Even if it were to be established that there are differences in trust and trustworthiness across cultures and throughout history, both in the way they are expressed in language and behaviour and so objectively observed, and in the subjective
I.2 Dominant Views on Trust: Problems 49 experience of trusters and trustees, a complex experiential account can accommodate such disparities. Diverse cultural and societal norms surrounding trust and trustworthiness – which acts count as violations, repentance and forgiveness for example – do not constitute trust and trustworthiness, but rather, express trust and trustworthiness in those cultures and societies. When we understand trust and trustworthiness as complex experiences, with varying manifestations, differences in the details are well tolerated. Rather than admonishing trust and trustworthiness for being “messy”, this dissertation has embraced their inherent complexity. In doing so, as Nickel rightly says, trust and trustworthiness cannot explain as many phenomena as a more “unrestricted” view of trust or trustworthiness might. This is held to be a positive attribute of two-place trust. A restricted theory of trust and trustworthiness, applied correctly, has genuine explanatory power. It is able to say that there was cooperation, but there was no trust, whereas an unrestricted theory of trust can only say of cooperation: there was trust. In summary, this subsection presented literature in favour of simplistic three-place trust. Philip Nickel’s argument for the adoption of simplistic three-place trust on pragmatic grounds was considered and rejected. Simplistic three-place trust is indistinguishable from reliance and cooperation. This is not a reason to favour it, as doing so does not allow us to explain more phenomena, we merely lose nuance where nuance is sorely needed.
I.2.3 Three-place trust: plus
Three-place trust plus includes all accounts which add any additional elements to simplistic three-place trust. Unlike simplistic accounts, which are entirely behaviouristic, ‘three-place trust: plus’ accounts are not. These accounts vary considerably, the motivation behind them is similar, with the aim being to distinguish trust from reliance, genuine trust from pretending to trust. As Philip Nickel observes, the motivation to make trust distinct from belief, reliance and cooperation is precisely what separates the treatment of trust in philosophy as compared to the social sciences (2017: 196).
I.2 Dominant Views on Trust: Problems 50
Yet precisely what needs to be added to trust to distinguish it from reliance, belief, and cooperation has become yet another flashpoint of disagreement in the literature, and it is spurred on by the insistence on maintaining an underlying three-place schematic. As Katherine Hawley put it:
What is the magic ingredient which distinguishes (dis)trust from mere (non-)reliance? Philosophers, inevitably, disagree: maybe a truster imputes appropriate motives to the trustee, including perhaps a concern with or responsiveness to the needs, desires, or indeed trust of the truster; or maybe a truster sees a trustee as morally obligated, committed, or accountable in appropriate ways. (2017: 231)
Interestingly, the growing number of ‘three-place trust plus’ accounts which Hawley is pointing to here, treat trust in a way that is strikingly reminiscent of the way in which knowledge – since the Gettier examples – is treated in Anglophone analytic philosophy. Here, we see knowledge being “fixed” by being analysed as justified true belief plus something else, such as an epistemic “anti-luck” condition (Zagzebski 1994: 72), a causal connection (Goldman 1967), or a reliability condition (Dretske 1989: 95), to make it non-accidental. These similarities between the way trust and knowledge have been treated by philosophers are increasingly being recognised (Domenicucci and Holton 2017: 149; Hinchman 2017: 56-59), and such parallels might be taken as further evidence that trust has been largely viewed as an epistemic notion. Three-place trust plus accounts are all the same in their underlying schematics, yet the details vary considerably. For example, Philip Pettit takes the philosophically interesting kind of trust to be a three-place “interactive trusting-reliance” in which A’s reliance on B to do X is manifest to B. Pettit explains that if we are trusting a bus driver, for example, to take us to the city centre, then that driver “knows that I am relying on him and knows that I am aware that he knows that.” (1995: 207). Hence, for Pettit, trust is like (or involves) a second or even third-order belief-like state. As we shall see in II.3, Karen Jones provides a parallel view – rich trustworthiness – which this dissertation argues is second-order reliability. More recently, Pettit has taken trust relations to involve a universal desire for the “good opinion of others” (2002: 354). For Bernd Lahno:
I.2 Dominant Views on Trust: Problems 51
A trusting person is disposed to react to a misuse of her trust in a particular and emotional way. This is due to the fact that the other is seen as a responsive person consciously engaged in interaction with the trustor. As the author of her acts she is held responsible and, thus, the expectations of the trusting person are normative in character. In contrast, perceiving the other from some distance like a mechanism governed by natural behavioural laws, an objective attitude in Strawson’s sense that would allow for pure factual expectations only, is incompatible with genuine interpersonal trust. (2017: 132)
Thus, for Lahno, the key elements of trust are the participant stance, and “connectedness” in the sense of shared norms and values. Meanwhile, Pamela Hieronymi discusses what she calls a “purist’s notion of trust”, according to which one person trusts another to do something to the extent that one trustingly believes that the other will do that thing (2008b). Colin O’Neil takes it that three-place trust requires an “expectation of performance” (2012: 308). Still others have proposed various additions, such as: hope (McGeer 2008), the participant stance (Holton 1994), an acknowledgement of the value of the trusting relationship (Hardin 2002), an expectation of goodwill (Jones 1996), a desire for the continuation of the trusting relationship (Baier 1986), a “feeling of trust” (M. Fricker 2007: 79), a belief in the trustworthiness of the trustee (Keren 2014: footnote 4), an emotion (Lahno 2001), a moral attitude (Mullin 2005), a commitment to a social norm (Coeckelbergh 2012), a disposition to experience Strawsonian reactive attitudes (Helm 2014), or even that trust is a gift (Mansbridge 1999). In summary, what we can see from these various proposed fixes is that philosophers have tended to choose one specific thing to add on to simplistic three-place trust. Trust is reliance plus belief in trustworthiness, or reliance plus the participant stance, or reliance plus a feeling of trust, or reliance plus a commitment to a social norm. However, despite considerable recognition that a straightforward three-place analysis of trust will not cut it, the inclination to abandon three-place trust altogether is curiously absent. This is odd, since what philosophers have added to make simplistic three-place trust reflect more accurately our actual experience as trusters, is invariably some private psychological/mental phenomenon. Nobody is arguing that, to trust, one must rely to ‘some sufficient degree’, or ‘with regard to certain things’. Trust is nowhere being
I.2 Dominant Views on Trust: Problems 52 distinguished from reliance by the claim that trust is ‘about’ serious or expensive things. Instead it is argued that one must be in a certain psychological state, or be prone to experiencing certain psychological effects, in order to be trusting. This demonstrates that there is already a good deal of recognition of the private experiential nature of trust, despite the oddly persisting confidence in the public nature of trust as an act.
I.2.4 Three-place trust explained
This subsection considers why philosophers have preferred to hold on to a three-place conception of trust, adopting ‘three-place trust plus’, or acknowledging three-place trust as a ‘kind of trust’, rather than abandoning the idea of three-place trust altogether. The primary explanation provided is to do with the way the word ‘trust’ is used in the English language. Two further explanations are then considered: 1) the persistent sense that what causes ‘trust-reactions’ are ‘trustworthy-actions’; and 2) the false idea that the only alternative to three-place trust is complete and total confidence in the trustee. Primarily, the inclination to continue to view three-place trust as an important kind of trust despite its known limitations, stems in part from the way the word ‘trust’ functions in language, particularly the English language. On this point, Stephen Darwall has said that the effort that goes into distinguishing trust from reliance in philosophy, is not aimed at settling a terminological dispute:
Used sufficiently broadly, ‘trust’ need not contrast with ‘rely’ or ‘reliance’. When we trust our cars or eyes to work properly, or perhaps even when we trust other motorists, no more need be going on than our relying on them. We simply proceed on the assumption that these will function or conduct themselves correctly or appropriately. Philosophers who write about trust, however, have mostly been concerned to distinguish trust from mere reliance of these kinds. (2017: 35)
And yet, the ways in which English speakers use the terms interchangeably can mean that we easily lose sight of the distinction, as Katherine Hawley notes:
We do not mark the trust–reliance distinction sharply in everyday language, and nor do we mark the trustworthiness–reliability distinction consistently. This everyday loose talk
I.2 Dominant Views on Trust: Problems 53
presents a problem to the theorist who hopes to explain the underlying differences between trustworthiness and reliability. (2017: 234)
Despite widespread recognition that the problem is not terminological, there is still an inexorable pull towards providing an account of trust that mirrors the way the word is used in everyday language, and ‘trust’ in English at least, is often used synonymously with ‘rely’. The conceptual confusion that has resulted from this pull is demonstrated by the kinds of example philosophers provide when they purport to be providing ‘examples of trust’ (see I.2.1). In English, ‘trust’ appears frequently in the three-place construction, and some philosophers writing in English have explicitly argued along such lines when defending a three-place trust account (Hardin 2002; McLeod 2015). Yet such arguments may only be convincing depending on which language one happens to speak. Richard Holton and Jacopo Domenicucci have pointed out that the three-place construction as it appears in English may be atypical, as it is not readily available in other languages; they note Latin, Italian, and French (2017: 150). Domenicucci and Holton take this to raise questions about the correctness of taking the notion of three-place trust at face-value and say that it points to an important area for further empirical investigation. That said, despite their scepticism about the concept being universal, they still do not suggest abandoning the idea that three-place trust is a kind of trust:
We don’t deny that this three-place relation exists, expressed in a perfectly natural English idiom, with a useful role to play. (2017: 149).
What Domenicucci and Holton do here in moving from the way the word ‘trust’ functions in language to allowing ‘different kinds of trust’ is actually quite a common ploy. We saw a similar move already in Nickel’s treatment of trust, though he provided contrary advice as to how we should manage the situation (I.2.2). Others who have tried to get at trust via language have been left with scepticism that there is only one kind of trust:
I.2 Dominant Views on Trust: Problems 54
What I doubt is that there is a distinctive attitude of trust that serves as a psychological underpinning of our sociality; rather different forms of trust are appropriate to different sorts of object. Owens (2017).
Comparable sentiments can be found in the work of Wright (2010); and Paul Faulkner (2017) has argued in a similar vein from language use that three-place trust – although again still a kind of trust – cannot be the fundamental kind. Instead of noting how the construction differs across different languages though, Faulkner makes his argument by noting that there is no three-place form of distrust in English “we do not say ‘X distrusts Y to φ’ – I don’t distrust my mechanic to deliver my mail!” (2017: 121). Essentially, Faulkner moves from the assumption – taken from Hawley (2014) – that there must be analytic connections between trust and distrust, to the claim that there is no three-place form of distrust, and hence, to the conclusion that the three-place form itself cannot be fundamental (2017: 120-21). Faulkner’s line of reasoning is problematic in the same way as Nickel’s and Domenicucci and Holton’s. The reality that there is no common three-place form of distrust in the English language does not mean that English speakers experience no three-place distrust equivalent to three-place trust. That the construction sounds odd to our ear, does not mean the idea or practice of distrusting a person to do a particular thing is so odd. Both Faulkner and Domenicucci and Holton (as well as many others, see: I.1), take the fact that there are two (or more) distinct ways we talk about trust – at least in English – as evidence that there are two (or more) distinct ways of trusting. Indeed, the limitations of three-place trust have meant that most philosophers acknowledge multiple ‘kinds of trust’, the arguments really being over which kind is fundamental, or interesting. This move is not – on the face of it – a bad way to go. When it comes to some concepts, language provides a reliable signpost. And after all, why should it not be the case that there are multiple kinds of trust? As Thomas Simpson writes:
That premise is: ‘trust’ is univocal. As such, ‘trust’ always has the same referent. I deny this. Listening to the way the word is used provides a compressed argument. Sometimes ‘trust’ is naturally understood as referring to a sort of affective attitude (‘I will trust my
I.2 Dominant Views on Trust: Problems 55
wife; I will not be jealous’); at other times to a conative one (‘Come what may, I will trust you to the end’); and at yet others to cognitive ones (‘I know you are honourable; I trust you’). Surely correctly, Annette Baier comments that ‘Trust, if it is any of these [affective, cognitive and conative], is all three’ (1994: 132). Indeed, sometimes it is not a mental state but action itself which is described as trust (‘The patrol followed the scout, trusting him to spot an ambush before it was too late’). It may be that there is an explanation of why it is felicitous to use ‘trust’ in each of these instances on which the term has an invariant meaning, perhaps expressible by necessary and sufficient conditions. It is all but certain, however, that any such conditions do not require the referent of ‘trust’ to be always the same kind of mental state. Further, I doubt that there are any such invariant conditions of use. The vulnerability of existing analyses of trust to counterexample is inductive evidence of this… The equivocity that ‘trust’ exhibits is not the same as that of ‘bank’. It is mere linguistic accident that ‘bank’ is ambiguous between a financial institution and the side of a river. It is no accident that ‘trust’ functions as an umbrella term that may refer to a variety of mental attitudes that, while non-identical, nonetheless share a range of similar features. (2017: 183-84)
I agree with Simpson that ‘trust’ does not always have the same kind of mental state referent. Indeed, in an important way, as we shall see in 1.5, the referent is complex, and can change depending on context. Here now though, is a different way to think about the things that Simpson has brought to our attention. Rather than capturing different kinds of trust (the conative, cognitive, or affective), each of the statements that Simpson points out merely capture just one part of trust. “I will trust my wife; I will not be jealous” captures the way in which trust makes us vulnerable and requires an ongoing effort. “Come what may, I will trust you to the end” points to the surprising strength of trust, and how it can shore up cooperation in the face of overwhelming adversity. “I know you are honourable; I trust you” points to the ability of trusting relationships to make us feel safe, secure, and certain. While “The patrol followed the scout, trusting him to spot an ambush before it was too late” probably just points to reliance, unless the scout is a trusted member of the patrol, and then it may point to the fact that trust can cause reliance behaviour. The point is that Simpson has not identified many varieties of trust, but many parts of trust, many upshots of trust, the many ways in which trust reveals itself in our feelings, thoughts, and actions.
I.2 Dominant Views on Trust: Problems 56
We could go the way of Faulkner, Simpson, Nickel, Domenicucci and Holton and others, and say that because the referent changes, across and within languages, cultures, and contexts, there are multiple kinds of trust. Alternatively, we could go the way this dissertation proposes and say that trust is complex, involving different things at different times, and producing multiple kinds of effects. The benefit of their way is that every time a philosopher provides a counter-example, pointing to a different way the word ‘trust’ is used, we do not need to go to great lengths to explain how this instance of trust fits (or does not fit) with our conception. We can just call it a different kind of trust. The downside of their approach is that it dilutes trust indefinitely and risks obscuring what all these ‘kinds’ share that make them ‘kinds of trust’ in the first place. The downside of my approach is that it may prove unwieldy, messy, and resistant to exacting philosophical analyses. But the upside of my approach is that we may just manage to acknowledge and capture the inconsistency, irregularity, and intricacy of something which is in actual fact all of these things. When it comes to trust, language use provides nothing but an exceptionally terrible starting point. This is true of other concepts, such as love. Trust, like love, is a part of human experience. We know this, because we recognise it. Across language and cultural barriers, we recognise trust, distrust, broken trust, betrayal, forgiveness, gratitude, closeness, friendship, loyalty, and love, in art, literature, poetry, history, song. We recognise them in our stories, many of which would make no sense – would be wholly unrecognisable – if their subjects (trust and trustworthiness) were not shared experiences. The way ‘trust’ functions in English provides one explanation for the continued acceptance of three-place trust despite the recognition of its limitations. Three-place trust remains dominant, even though there is plenty of precedent in the English language to capture a two-place conception. We do already speak coherently of trust that is not ‘about’ anything at all, without reference to anything entrusted: we trust her, she trusts him, you broke my trust, we trust one another, you have to learn to trust me again, we have lost our capacity to trust one another, the only person I’ve always trusted is you. We already speak about trust that has only two elements to it – a truster and a trustee. Note that this is not true of the other three-place constructions that are
I.2 Dominant Views on Trust: Problems 57 frequently confused with trust, such as reliance, dependence, cooperation and belief, which logically and grammatically require that they are ‘about’ something. Another explanation for the continued dominance of three-place trust stems from a deeper intuition that the three-place construction captures something importantly true about trust that any other construction would miss; it seems like trusting and being trustworthy really are about what we entrust one another to do. If someone says, “She broke my trust”, the natural response is “What did she do”? Indeed, any changes in our trust – whether for better or worse – appear to march in step with the actions of our trustees. Reflecting on those times when we have had our trust broken, when we have felt the sharp sting of a surprise betrayal, seems plainly to reveal that it was because of something that our trustee did: such as breaking a promise, lying, failing to return something, protect some treasured belonging, failing to follow through on a deal, or pull through with a favour, or the classic: infidelity. Given that trust and trustworthiness clearly manifest in acts of reliance and reliability, this concern is not easily dispensed with, and this dissertation cannot address it adequately until the following section. In this subsection, I will develop a two-place notion of trust, which, as we will see, nevertheless allows that there is a connection between trust and behaviour, without admitting that the behaviour constitutes the trust. As we shall see, once the confusion surrounding the object of trust that arises on three-place trust is addressed, this concern falls away. And so lastly, the final explanation considered here for the preference for three- placed trust stems from an apparent problem with two-place trust. The problem occurs when two-place trust – trust in a person, without reference to anything entrusted – is conceived of as a sort of irrational blind foolishness, a total confidence in someone else’s powers and abilities. This line of argument represents a historically entrenched way of thinking about trust and seems to be behind many philosophers’ distaste for anything other than a three-placed account. It is present in Howard John Neate Horsburgh’s distinction between three-place trust and trust in a person:
…sometimes when we say that A trusts B we mean that A has perfect confidence in B. Thus, we sometimes use the total absence of doubts and questionings as to the reliability of someone as our criterion for trust. (1960: 343)
I.2 Dominant Views on Trust: Problems 58
“Perfect confidence” in someone or the “total absence of doubts and questionings” as to their reliability is one thing, and probably very rare. But it is a mistake to think that this is two-place trust. At fault here is a false dichotomy – the difference between two-place trust and three-place is not the difference between trusting a person to do everything and trusting a person to do one specific thing. Trusting a person to do everything is complete reliance or total dependence: it is universal three-place trust. Trusting a person in the two-place sense, though, does not require any presumptions regarding the specific skills or interests of the trustee. Two-place trust does not presume that the trustee is competent or committed to do anything specific at all, let alone everything. Two-place trust involves only a truster and trustee, that is precisely the point of it. Trust as it is presented in this research project allows that we do indeed navigate our trusting relationships cognisant of each trustee’s abilities and limitations, including their limitations of character. Yet we need not assume that it is our trust that is the thing that is changing as our understanding of each trustee’s commitments and competencies change. Trudy Govier writes that trust and distrust are “often relativized to specific roles or contexts” (1993: 105). As this dissertation argues, our beliefs about people, predications about they will behave, and tendencies to depend on them are most certainly relativised to specific roles or contexts, but trust and trustworthiness are not. In summary, there are likely a number of explanations as to why philosophers have continued to prefer the three-place analysis of trust despite its known limitations. This subsection has considered three. First, in English, the three-place construction reflects language; second, there is a nagging connection between the growth and demise of trust and the successes and failures of our trustees that is not easily dispensed with; and third, there is an apparent limitation of the two-placed account that arises when two-placed trust is conceived of as ‘total dependence’.
I.2.5 Summary of Dominant Views on Trust: Problems
Section I.2 considered the notion of three-place trust in detail, and in various manifestations of its forms. Beginning in subsection I.2.1 with the bare schematics of three-place trust (which all three-place accounts share), we then saw various examples from the literature of three-place trust. Subsection I.2.2 presented the simplistic version
I.2 Dominant Views on Trust: Problems 59 of three-place trust and discussed three writers who have preferred to adopt it, focusing on Philip Nickel’s argument that there are pragmatic advantages to simplistic three- place trust. That subsection argued that simplistic three-place trust is indistinguishable from reliance and cooperation and rejected it as an adequate account of trust on those grounds. Subsection I.2.3 presented ‘three-place trust plus’ accounts, which aim to address the limitations of simplistic three-place trust with the addition of various private psychological elements, demonstrating that there is much disagreement as to what needs to be added to simplistic three-place trust to make trust distinct from reliance. Finally, subsection I.2.4 presented and considered three explanations as to why philosophers have clearly preferred not to abandon three-place trust despite its limitations, and despite the lack of agreement as to what additions to it need to be made. This subsection raised but did not address one of these explanations – the hunch that trust and trustworthiness really are a matter of what we do. The next section, The Object of Trust, presents a confusion that arises from the logic of three-place trust, demonstrating how – once this confusion is addressed – this hunch can be accommodated within a two-place framework.
I.3 The Object of Trust 60
I.3 The Object of Trust
I claim that trust is a two-place relation: (A trusts B). Though this is not all there is to it, schematically, it involves a truster and a trustee. The object of trust, then – on a two- place account – is quite straightforward: it is the trustee. What counts as the proper object of trust is a further matter affecting both two and three-place accounts equally. Yet on three-place accounts (A trusts B to do X), not only is it not immediately obvious what things might count as the proper object of trust, it is also unclear whether those who adopt three-place trust intend the object of trust to be taken as the trustee (B), or what is entrusted (X). In subsection I.3.1, I reveal this ambiguity regarding the object of trust that is inherent to the three-place trust logic. I then consider the proposal that the object of three-place trust is X, concluding that the idea of ‘trusting’ that a certain state of affairs will come to pass is better captured with notions such as: prediction, expectation, anticipation, confidence, calculation etc. Subsection I.3.2 considers the proposal that the object of three-place trust is B, revealing two possible ways in which ‘three-place trust in B’ might be interpreted: the conjunction of trust and an instance of reliance, or a causal relation between our trust and an instance of reliance. I propose that – rather than instances of trust – both of these possibilities are better expressed as simply: trust and reliance, or trust causing reliance. In subsections I.3.1 and I.3.2, this dissertation puts forward a significant and original recommendation of this research project; for conceptual clarity, it is essential in academic settings to adopt these more accurate ways of describing the phenomenon of so-called three-place trust. Then, I.3.3 addresses the question: What is the proper object of trust? demonstrating the wide variation in answers philosophers have given, from: anything, to only persons, to only things that are person-like, all the way to: most things except persons. Such wide variation, it is argued, underscores the importance of settling the metaphysical question before trying to tackle any ethical or epistemological issues that trust apparently throws up. The proper object of trust on the two-place account explored here, is anything that possesses the possibility of being worthy of it– anything that can both have, and induce in us, complex experiences of trust and trustworthiness. While trust and trustworthiness may not always be reciprocated in individual cases, the inherently reciprocal nature of
I.3 The Object of Trust 61 them – in the “proper” cases – is thus significant: while many things might induce genuine experiences of trust in us, only things which can also have full complex experiences of trustworthiness themselves count as the proper objects of trust.
I.3.1 The object of three-place trust: X
The object of trust is the thing that trust targets– the thing that is trusted. On one-place accounts, there is no object of trust, trust is just something trusters have, it is just something that they are, similar to a mood or temperament. On two-place accounts, the object of trust is the trustee directly, not the trustee’s actions or words, since what is ‘entrusted’ is not part of trust on a two-place account. Yet if we examine the logical structure of three-place trust, it is not immediately obvious exactly what the object of trust is supposed to be. At least two possibilities present themselves–
1. The object of trust is X
2. The object of trust is B
The first interpretation says that the object of trust is X (an event, or a certain state of affairs), which happens to involve B. The second interpretation says that the object of trust is B, and yet, unlike on two-place accounts, the trust is not so much in them, as they are, but is in some way contingent on them performing the entrusted action. This distinction is nuanced, so an example may help. Taking a familiar example from the literature (the Trust Game discussed by Holton – see I.1.2) if an instance of trust is me falling backwards and ‘trusting’ that Damian will catch me, then, am I trusting that I will be caught (by Damian) or am I trusting Damian (to catch me)? We can appreciate the relevance of this distinction by thinking about the circumstances under which I might consider my trust to be broken. If I am trusting that I will be caught, and Damian fails to catch me but Sandra rushes in and catches me instead, then I seem to have no reason to consider my trust broken. On the first interpretation, Damian just happened to be who I thought would perform the trusted action, but it did not need to be him. Here, B serves as a placeholder that can be occupied by any individual or thing. But on the second interpretation, B is more than a placeholder, identifying always a single specific individual or thing. If the object of my trust is Damian and the same situation unfolds –
I.3 The Object of Trust 62 he fails at X, but Sandra succeeds – then my trust has been broken. We can naturally imagine being personally hurt by a friend who fails to do something for us, even if someone else comes to the rescue, and does it anyway. This ambiguity surrounding the object of trust inherent to all three-place accounts goes some way to explaining the various competing intuitions about trust evident in the current literature, as laid out in both I.1 and I.2. Yet both these possibilities – ‘trusting that X will be done (by B)’ and ‘trusting B (to do X)’ – are not what trust is. Such phenomena are better captured with different concepts entirely, or else, with two concepts, such as trust and reliance. To see this, let us consider each one. First:
The object of trust is X
In this interpretation, the trust is placed in X, and the trustee just happens to be the facilitator. If we say, “I trust the pilot to land the plane safely”, or “I trusted that the doctors would save her” then, these statements can mean that the actual person doing the landing, or the lifesaving does not matter. Sometimes, what matters is the outcome and the ‘trustee’ can be ‘swapped out’ for any other competent individual. When this is so, a ‘breach of trust’ occurs not when a specific trustee fails to do X, but just whenever X does not occur. Reflecting on our own experience, we see that such circumstances are commonplace; we regularly find ourselves in situations in which we are relying on a certain outcome X, where X depends in some way on someone or something B, though it matters very little which specific individual or thing serves as B to secure X. Indeed, this is perhaps the most common situation we find ourselves in with regard to our dependency on others. Typically, we do not know, nor does it matter, who delivers our mail, collects our garbage, grows our food, or checks the quality of our water. What matters is that whoever it is, they get it done and do it right. As it is the quintessential dependency relation that exists between individuals in a modern global world, why should we not think that the object of our trust could be the event itself, X? Surely, we can trust planes to land safely, economic downturns to right themselves, armies not to invade our borders, and election promises to be fulfilled. Indeed, this is how we already see the idea of trust being employed in a great deal of academic and journalistic discourse.
I.3 The Object of Trust 63
One could argue along pragmatic lines as to why we should understand trust as being in X: 1) we already use the notion of trust in X, 2) trust in X provides a simple way for us to describe complex situations, and 3) when people come across the concept of trust in X, for example in headlines like ‘Investors Trust in Economic Recovery Secured by Government Reforms’, they are not hopelessly confused but grasp the situation. Yet convenient as it is, such anthropomorphist metaphor has its perils. It is no accident that connotatively, trust is linked to things like: forgiveness, love, gratitude, and betrayal. Our concept of trust is linked to such phenomena, because trust, in our experience, is linked to such phenomena. Our position with regard to events though, is not. Strictly speaking, we are not betrayed by events, but by people. We do not forgive the past, but the people who wronged us in the past. We are not grateful to the plane, but to the pilots (or some of us perhaps to God). By accepting that the objects of trust are events, then, while we appear to gain an exceptionally diversifiable explanatory device, what we actually do is dilute an effective way of explaining one another. There is a difference between having something that we expected to happen fail to happen and having someone that we expected to do something for us fail to do it. If we want to hold on to this distinction, in academic settings, then we had better not use the same concept for both purposes. Recently, Stephen Darwall has drawn out this important distinction in discussing the difference between ‘expecting that’ and ‘expecting of’, taking only the latter kind of expectation to be associate with trust:
we can expect that something will occur or expect something from or of someone. Expectations take propositions or possible states of affairs as objects, whereas the object of an expectation of is a person (or group of collective persons). Expectation of both sorts impose standards. But where an expectation that says how things will be (or how we have reason to believe they will be), an expectation of says how someone should act. (2017: 36)
As Darwall describes, the kind of expectation involved in trust must be an expectation of, since “not living up to someone’s expectation is not to confound what she thinks will happen, it is to fail to meet her standards of how one should act or be” (2017: 37).
I.3 The Object of Trust 64
Hence, to capture trust’s distinctive connection to things like betrayal and personal disappointment (which is disappointment in someone) trust must involve an expectation of. Three-place trust in X involves B in the trust not as a distinctly individual person, but only as a logical placeholder, and, as Darwall says:
Interpersonal relations do not merely have persons as logical relata. They are constituted by people relating to [his emphasis] one another… the objects of reactive attitudes are therefore implicit addressees; the attitudes implicitly address their objects and invite or demand reciprocation in some way; they come with an implicit RSVP. (2017: 38)
Répondez s’il vous plaît is not something we can ask of the X’s in our lives (believe me, I have tried). Events, situations, states of affairs, and circumstances are influenced by our actions (and even perhaps by our attitudes), but they cannot respond to us. Many philosophers have recognised that there is a ‘bootstrapping’ mechanism inherent to trust, whereby trust that targets the untrustworthy can increase trustworthiness (Holton 1994; Horsburgh 1960: 346; Jones 1996: 6; Nickel 2007: 317). There is no similar bootstrapping mechanism to expectations, as Charles Dickens, through poor Pip, reminded us. If the object of three-place trust is X, then, what may broadly be termed our ‘trusting experiences and practices’, such as: 1. feeling betrayed 2. communicating resentment 3. holding one another to account 4. begging for forgiveness 5. giving the benefit of the doubt 6. pretending to trust to engender trustworthiness 7. assuming the best of someone 8. displaying gratitude 9. giving second chances
10. acting on a sense of loyalty. etc. … become senseless, inexplicable, and hollow, when these experiences and practices are in fact very meaningful and often successfully alter the outcome. This dissertation
I.3 The Object of Trust 65 finds no problem with the use of anthropocentric metaphor in everyday conversation (or even in sensationalist news headlines). However, it cautions against such use in serious academic settings, where claims are being made about the nature of reality. If broken trust can explain broken hearts, then it cannot also be used as a theoretical device in a purported explanation for the Chinese government’s political retaliation to the Australia government’s bungling of that latest trade deal. It is most unlikely that the same mechanism is the cause of both conditions. Happily, there is no need to use trust in academic settings to capture our experience of, and stance towards, events. When we ‘trust’ something to happen, indifferent to who specifically is involved, then this situation is better captured with concepts like: prediction, expectation that, anticipation, confidence, hunch, calculation etc. Using these alternative concepts instead of trust in academic settings to describe our relationship to events need not imply that we cannot have attitudes towards events, or emotional reactions to them. Nothing about the concept of prediction for example, precludes us from having some sort of private emotional reaction when our predictions turn out to be wrong. It is not as though we have cold emotionless expectations that on the one side, and hot sentimental expectations of on the other. Great expectations can produce great emotional suffering; again, we recall Dickens’ Pip. Nonetheless, there is a difference in the experience of having someone break our trust, and having a prediction fail to come to fruition. The complex nature of these experiences, though hard to characterise and no doubt sharing some overlapping qualities (we may grieve in both cases for example), are distinct. Three-place trust in X confounds this very real and very important distinction.
I.3.2 The object of three-place trust: B
This initially looks better, because here, the trust is directed at an entity which can presumably respond to it – the trustee. But what now are we to make of the X? It seems to just be hanging there, and we might well ask: what work is the concept of entrusting supposed to be doing in a three-place account, if we have already admitted that trust is in the trustee? Put another way, if who I trust is Damian, then how is my trust connected to my relying on him to catch me?
I.3 The Object of Trust 66
One idea is to understand three-placed trust as being comprised of two distinct things: 1) a trusting feeling, belief, or attitude towards a specific person, and 2) a specific instance of reliance in them, or acceptance of their testimony, connected together. This was Richard Holton’s move in 1994, when he argued that trust is “reliance from the participant stance”. Though it was not made explicit there, integral to Holton’s 1994 discussion was the idea that trust was made up of two things– an act of reliance, and an attitude of the participant stance. Yet if trust is a phenomenon that is made up of both a public act of reliance, and a private belief, feeling, or attitude, then the next question to ask is: what precisely is the nature of the connection between these two distinct phenomena, that together give rise to trust? I consider two possibilities: conjunctive and causal. Firstly, the connection between trust and reliance, between trust in a person and our reliance on them to do X, is conjunctive. The conjunctive move is a common one, as Domenicucci and Holton have recently noted :“many have tried to build trust conjunctively out of reliance: if one trusts, then one relies, and some other condition obtains” (2017: 153). A conjunctive account says:
(A is relying on B to do X) and (A trusts B)
To give an example of this approach in practice, it is good to examine Paul Faulkner’s thinking back in 2007:
A’s expectation is grounded by the conjunction of the belief that S recognises A’s dependence on S φ-ing and the presumption that this will move S to φ. This conjunctive ground then implies a further presumption. In presuming that S will be motivated to φ by A’s dependence on S φ-ing, and believing that S can in fact recognise A’s dependence, A comes to presume that S will indeed φ. This implied presumption, made for these reasons, is just the presumption that S will fulfil the expectation that A holds him to—that S will prove trustworthy. (2007b: 884)
For Faulkner here, trust involves both the truster’s private beliefs and presumptions, and a public act of reliance. One problem with the conjunctive approach in general, as Domenicucci and Holton recognise, is that it cannot explain a truster feeling betrayed
I.3 The Object of Trust 67 by a trustee’s failure if the truster’s trust in the trustee and their reliance on them merely happen to occur together (2017: 153). Moreover, there may be instances of trust even when there is no opportunity for reliance: “we do not automatically stop trusting someone when they lose the power to act in ways in which we relied on them” (2017: 154). On a conjunctive account, my trust and reliance can occur together without being in any way connected to one another. We rely on, cooperate with, and believe those we trust all the time, but we do not consider every instance of reliance, cooperation, or acceptance of testimony to be instances of trust. Sometimes my trusted spouse fails to remember to pick up nappies on his way home, even though I was relying on him to do so. Sometimes he tells me that it is not going to rain, and then it does. In many such instances of his unreliability (and there are many) my trust remains wholly unaffected. Equally, like a broken clock that is right twice every day, or an Immanuel Kant who takes a walk once every day, sometimes my darling proves most especially reliable without this being in any way connected to his trustworthiness. The fact that trust and reliance can occur together without it being a case of three-place trust shows that there must be something more to three-place trust than the conjunction of trust and reliance. Let us try again. Perhaps the two components of three-placed trust enjoy a causal connection. In that case we would have a genuine case of three-place trust only when:
((A is relying on B to do X) because (A trusts B))
In this interpretation, it is precisely the existence of a causal relationship between the two components itself that gives rise to three-place trust. Sometimes we rely on one another out of necessity, coercion, a sense of obligation, or the pressure exerted on us by particular social norms, and sometimes we rely on somebody for the simple reason that we believe them to be reliable. But on the causal account of three-placed trust, it is trust only when ((A’s trust in B) caused (A to rely on B to do X)). This looks promising, since it captures the intuition that it is more significant to trust somebody to do something than to merely rely on them to do it, even if we happen to share a background of trust with that person. The causal picture, unlike the conjunctive picture, is able to weed out those cases of unconnected coincidence between our trust
I.3 The Object of Trust 68 and our tendencies to rely on, cooperate with, and believe one another. As such, it effectively does away with the problem we saw with the conjunctive account. So far, so good. And yet, while there is no doubt a genuine phenomenon here – sometimes the way we feel about someone causes us to rely on them in ways that are distinctly significant – are we satisfied with saying that this – ((A is relying on B to do X) because (A trusts B)) – is what trust is? An analogy might help here. Love can cause us to act in ways that are ‘loving’ – devoted, caring, selfless, affectionate, passionate – but it need not do so. Sometimes love is unreciprocated and unwanted, and known to be so, and so remains hidden. Sometimes it is necessary to act as though we do not love when we do, and such pretence can even be considered an ‘act of love’ in itself. Here we can point to the behaviour of some Tutsi fathers and sons during the forced family rapes that occurred in the 1994 Rwandan Genocide, or, we can just remember that episode of Friends where Chandler pretends not to be in love with his best-friend Joey’s girlfriend. Love can cause us – require us even – to act as though we do not love. Trust is like this too. If it becomes essential to convince an enemy that we are not in cahoots, you and I might feign distrust in one another. Trust can cause and even require distrustful behaviour. Because of this, we should not accept an account of trust that says that trust is a phenomenon that occurs when how we experience someone causes us to behave in ‘trusting ways’ towards them, by relying on them, cooperating with them, believing what they say. Those advancing a three-place conception of trust say, “here is an example of trust”, and then talk of diaries being left on desks and people leaving their children with babysitters. This is like saying “here is an example of love” and then talking of meals being paid for and candles at dinner. Acts of trust and love come in many forms – reliance and romance are just the Hallmark versions. So, where does this leave us? The fact remains that we so often talk as though we trust people to do certain things, and this, at least at times, seems to be explanatory. Why do I choose to leave my children with my mother rather than a perfectly qualified stranger? Because I trust my mother! Sometimes trust is the thing that makes all the difference when it comes to our tendencies to rely on, cooperate with, and believe one another. Happily, the two-place account put forward in this dissertation finds no
I.3 The Object of Trust 69 problem with the idea that trust can play a role, sometimes a very significant direct causal role, in our acceptance of testimony and our decisions to rely on and cooperate with one another. A two-place account of trust can do this by categorising the phenomenon of so-called ‘three-place trust’ not as a kind of trust at all but as a kind of reliance; reliance that might have been caused by lots of things (necessity, coercion) but has, in this instance, been influenced to some notable degree by trust. Philosophers who have come some way to a similar understanding as I have, have chosen to hold on to the concept of three-place trust as a kind of trust (see I.1.2). While I gave possible reasons in subsection I.2.4., this is a mistake, and it leads to a good deal of unnecessary confusion. Here is another analogy: If we have elevated blood pressure (EBP) because we have a virus, then this is not a kind of virus (the EBP virus). Rather, it is a kind of elevated blood pressure (the kind caused by a viral infection). Similarly, if we are relying because we trust, then this is not a kind of trust, but a kind of reliance. Those who deal in explanations of human behaviour must be awake to the fact that reliance, cooperation, and acceptance of testimony can have many causes. Even when there is a background of trust, trust will not always be the culprit. Those who seek to explain human behaviour need to be on the lookout for coercion, compulsion, obligation, nativity, misunderstanding, intimidation, fear, faith, sympathy, or any one of many other things that might make one person rely on another, just as doctors need to be awake to the multiple causes of elevated blood pressure in their patients – even those patients who are known to have the EBP virus. By letting go completely of the idea that trust is a three-place relation, we lose nothing, and we accomplish much. We lose nothing, because there are plenty of other concepts at our disposal that are better placed to account for those cases three-place trust previously covered, as we have seen. As for what we accomplish: 1) we do away with the ambiguity surrounding the object of trust, thereby gaining a clear way of conceptually separating our experience of relying on events, from our experience of relying on trustees. And, 2) if trust is not a three-place relation, we do away with any difficulty in distinguishing between trust and all other three-place relations with which it has been historically confused. We also thereby, 3) leave epistemic problems to
I.3 The Object of Trust 70 epistemology, no longer getting in a knot about the ‘irrationality of trusting-beliefs’ or the ‘irrationality of trusting-behaviours’. Under a two-place framework, we can still ask about whether or not it is good, rational, or sound to let our trust influence our beliefs and behaviours in any particular instance, but we do not need to worry about any inherent absurdity regarding how ‘trusting beliefs’ that must be ‘resistant to evidence’ can possibly be rational. Disconnected from an epistemological concern about the rationality of trust, 4) it also no longer matters very much whether trust is voluntary. Free from such ambiguities, conceptual confusions, logical impossibilities, and epistemological concerns, 5) we are permitted to think about other aspects of trust which might be important to us, such as how we might cultivate deserved trust, and recover from broken trust. When we re-frame three-place trust as reliance because trust, all this is possible, 6) without having to let go of that stubborn intuition that there is a connection between trusting a person and relying on them, believing what they say. In summary, subsections I.3.1 and I.3.2 addressed the question – what is the object of three-place trust? Two possible interpretations were considered, both were rejected. The first interpretation – the object of trust is X – was unsuccessful because it conflated trusting someone with expecting something to happen. The second interpretation – the object of trust is B – led to two possible ways of characterising the relationship between (A trusts B) and (A relies on B to do X): conjunctive and causal. The conjunctive interpretation came up short because trust and reliance can co-occur without it being a case of trust. The causal account captured something important; however, I have argued that the causal relationship between our trust and our reliance is better understood as just that – a causal relationship – not as a kind of trust. Even if one accepts this picture and adopts the view put forward here that trust is a two-place relation involving A (a truster) and B (a trustee), a further question regarding the object of trust remains. B (the trustee) might not always be a person. Indeed, as we shall see, one philosopher David Owens argues that B is never a person. Now, we turn to considering the proper object of trust.
I.3 The Object of Trust 71
I.3.3 The proper object of trust
The previous subsection considered the question: what is the object of trust in three- place accounts? This subsection considers the proper object of trust. This issue has historically been seen as important because, in three-place trust, our trust has been thought of as rational, warranted, or well-grounded only when it targets that which is trustworthy (McLeod 2015). Such concerns are less pressing in a two-place account, since two-place trust that fails to correctly target its proper object poses no immediate problem to our practical or epistemic rationality. Yet even so, if trust is a kind of complex reciprocal two-place relation, then it is still worthwhile thinking about what an object must be capable of, to qualify as a party to such a relationship. Indeed, this question helps us to get a grip on the nature of the trust itself. This subsection first provides a brief overview of the current debate in the literature. As we will see, the discussion has concentrated on the issue of personhood. On this, as we will see, there are a number of possible stances philosophers have taken: 1) that some take it as given that non-person entities can trust, whereas others maintain that only persons can trust, 2) arguments are given that non-person entities are ‘person-like’ enough to trust, or, that they are not person-like enough to trust, 3) the question is side- stepped with a qualification that – whatever the reality – the philosopher is addressing trust only as it exists between persons/non-person entities. I then present three examples from the literature, each of which demonstrate how a philosopher’s stance on the proper object of trust reflects their understanding of the nature of trust. Finally, this subsection provides a partial answer to the question: What is the proper object of trust? which takes the form of a recommendation about how high we should set the bar if we are to keep trusting relations distinct from other kinds of relations. This answer is partial, because the proper object of trust is that which can be worthy of it, and so a full answer to this question will not become possible until Part II of this research project, when we come to consider what it is to be worthy of trust. To anticipate, reflecting the reciprocal nature of trusting relations, the answer is the same: the proper objects of both trust and trustworthiness are all those things which both can create and experience trust and trustworthiness.
I.3 The Object of Trust 72
Advocating for a conception of trust that is purportedly able to explain more phenomena, some philosophers maintain that certain non-person entities can be proper objects of trust. As we have seen (I.2.4), there is a significant motivation here to provide a conception that can be used as an explanatory device to predict and explain the behaviour of all sorts of non-person entities, from states to financial markets, in areas outside of philosophy such as economics, sociology, and international relations. Among those philosophers who write about a trust that is not interpersonal, there are various candidates for what is taken to be the object of trust. Some speak of trust in institutions (Govier 1997; Potter 2002; M. N. Smith 2008; Townley and Garfield 2013), others of trust in government (Hardin 2002; Eric M Uslaner 1999), trust in animals (Silvers and Francis 2005), and more recently trust in artificial intelligence (Coeckelbergh 2012; Taddeo and Floridi 2011). Sometimes, no explicit argument is given for why we should consider these non- person entities to be the proper objects of trust. Instead the focus is on questions that would only arise if the coherency of the idea had already been accepted, such as: How can we increase trust in governments? and What are the conditions that precede a breakdown of such trust? (see, for example, Warren (1999). At other times, it is argued that such entities are ‘person-like’ in the relevant respects, and so it may be proper to trust them (Domenicucci and Holton 2017). Reflecting the ongoing difficulties involved in settling the philosophical question of what qualifies as a person (Campbell and McMahan 2010; Olson 2007; Parfit 2012; Shoemaker 2011, 2012), one approach philosophers have taken is to side-step this question altogether and simply state that, whatever the reality here, they are speaking only of trust between human persons (Horsburgh 1960: 343; McLeod 2015; Mullin 2005: 316; Wright 2010: 616). Still others make no such explicit qualification, and simply discuss trust as though it were obviously a relation that existed only between human persons (Hertzberg 1988: 307-22; Lahno 2001: 171-89). In short, the issue of personhood is viewed as highly relevant to the proper objects of trust, whether it is argued for, argued against, or openly and expressly side-stepped. The present disagreement on this issue underscores the importance of having a firm answer to the metaphysical question: What is trust? prior to attempting any
I.3 The Object of Trust 73 epistemology or moral philosophy on trust, since whether or not you think non-person entities can be the proper object of trust will be determined by the underlying account of trust that you have adopted, and whether or not you believe non-person entities can be the proper objects of trust will influence your answers to any moral and epistemological questions. For example, Karen Jones says that trust is a form of optimism about the goodwill of another. She reasons that:
one can only trust things that have wills, since only things with wills can have good wills– although having a will is to be given a generous interpretation so as to include, for example, firms and government bodies (1996: 14).
For Jones, things that can have goodwill are legitimate targets for trust because she takes assumptions about the goodwill of the trusted to be part of what it is to trust. It is worth noting that in her later work, Jones takes assumptions of goodwill to be insufficient for trust (2012: 67), but this does not impact the current point that there is a link between the type of account a philosopher adopts (at any one point in time) and their position on the proper objects of trust. Like Jones, Domenicucci and Holton have asserted that some institutions might be the proper objects of trust. Yet this is not because they think institutions might possess goodwill, since goodwill is no part of their account of trust. Instead, they say institutions might be the proper objects of trust because of the reactive attitudes we invest in some institutions, such as Britain’s National Health Service, which – they assert – could only have “got going” in the first place as an object of the people’s trust, and thus it is right that we have invested it with such trust (2017: 156-59). Holton’s account of trust as a reactive attitude thus appears to have informed his and Domenicucci’s view that the proper object of trust might be anything that we can sensibly take a Strawsonian participant stance towards. Jones, and Domenicucci and Holton, talk predominantly of interpersonal trust while allowing that trust might sometimes involve non-persons if those non-persons are person-like enough in the ways relevant to their underlying ideas about trust. Almost all philosophers, even those who endorse trust in non-person entities explicitly, take interpersonal trust as given, and often, as primary. Yet one philosopher David Owens
I.3 The Object of Trust 74 has recently (2017) provided an exception. Working within a three-place logical framework, he has argued that persons are precluded from being the proper objects of trust. For Owens, the proper object of trust is any object which has a distinctive value which can be realised by trust. According to him, trust has no distinctive value of its own; rather, it is the thing that allows us to realise the value of its object:
I trust my car when I am prepared to drive it around (not merely sit in it), an attitude that enables me to realize the distinctive value of a car. I trust an apple when I am prepared to eat it (rather than use it as a football), an attitude that enables me to realize the distinctive value of an apple. A specific apple and a specific car are worthy of this trust when they actually possess the value distinctive of their kind. In the absence of trust, the value of a (trustworthy) apple and of a (trustworthy) car will be wasted. (2017: 216)
For Owens, trust is ubiquitous (2017: 214). So much so, that we cannot drive cars and eat apples without it (though we can sit in them and use them as footballs without it). Unlike apples and cars, though, objects that have no distinctive value cannot be the proper objects of trust:
There may be objects that are valued by us but which have no distinctive value; they are just valued in different ways by different people. Perhaps the moon is such an object, valued by some as a celestial adornment and by others for its gravitational pull. In that case there would be no such thing as trusting the moon. We may trust that tonight will bring a full moon but we won’t thereby be trusting the moon. (2017: footnote 5)
Persons, Owens thinks, are like the moon in this respect “it is doubtful whether people as such have a distinctive value, one that can be realised by trust” (2017: 217). Instead, Owens says that when we trust other persons, what we are doing is trusting some distinctive value they possess, such as their actions, technical skills, or particular character traits. So, I can trust you to be on time, since you have the distinctive value of promptness, but I cannot trust you (or the moon). This line of thinking leads Owens to conclude that “there is no general attitude of trust fitted to play a foundational role in our social theory. Trust is as various as the objects of trust” (2017: 229). It is not easy to place Owens’ thinking within our current framework. It is almost diametrically opposed to what I am arguing about trust here. The first thing to say is that
I.3 The Object of Trust 75 what Owens is describing here is a genuine phenomenon; there is a sense in which we relate to parts of the people in our lives, realising their distinctive and various values to us. Yet it is hard to understand why Owens thinks that relating to “parts of people”, is trust. For one thing, it is far too self-interested to be likely to elicit that which trust so often elicits in our trustees – loyalty, gratitude, care, and indeed, trustworthiness itself. Perhaps what Owens is describing is our confidence in, knowledge about, or previous experience with certain objects – including people – that allow us to use them to our advantage to realise their individual values. We use strangers in this way, and we use those we love and trust to further our own goals as well. But we also relate to those we love and trust in another way– in a way that is not about realising their distinctive value to us, rather it is about creating value in one another. It is good to keep these distinct ways in which we can relate to a single individual (demanding/trusting, loving/ receiving) separate in our minds. We can see from Jones, Domenicucci and Holton, and Owens how various answers to the question ‘What is the proper object of trust or what should we trust?’ are informed by a philosopher’s approach to the underlying metaphysical question about the nature of trust. If trust involves goodwill, then the trustee should have goodwill. If trust involves the participant stance, then the trustee should be the sort of thing we take the participant stance towards. If trust involves ‘realising an object’s distinctive value’, then the trustee had better possess a distinctive value. As I have said, the account presented here is no different in this regard. If trust is a complex private experience that induces a complex private experience (trustworthiness), then the trustee should be able to both induce, and have, a complex private experience. This situation is evidence of the symmetrical nature of trust and trustworthiness, a point made further in II.1.1. As Domenicucci and Holton astutely point out though, the question about the proper object of trust breaks down into the empirical: do people in fact have the same sorts of trusting attitudes towards non-person entities that they have towards people? – as well as the normative: is it appropriate or good for them to do so? (2017: 158). They are right to draw our attention to the importance of separating these questions, as in practice they are so intertwined as to be easily conflated. Even if we discover that people do in fact sometimes have experiences of trust that are caused by or directed towards non-person
I.3 The Object of Trust 76 entities, whether or not they should trust non-person entities will depend on whether or not those non-person entities are capable of responding (and responding appropriately) to trust. This creates a whole series of further empirical questions, and each individual case – animals, governments, systems, groups – would have to be addressed individually. Yet anyone aiming to settle these empirical questions with regard to any specific non-person entity will first have to specify what an appropriate response to being trusted or to breaking someone’s trust should be, and this is not an empirical question but a normative one. Although this is where debate in the literature on the proper object of trust has settled, I feel it is not my place to try to answer the empirical question regarding which entities qualify as proper objects of trust and which do not, since I claim that trust is a complex experience and I have no qualifications to say which entities have complex private experiences, and which do not. Instead, this dissertation makes a recommendation. There are a number of general motivations people can have for cooperating with, or relying on, one another. These include fear of sanctions, self- interest, a positive evaluation of cooperation, and a positive evaluation of friendly relations (Williams 1995: 118). Only the last of these has the potential to capture something about our trusting relationships. If we are to enhance our ability to explain human behaviour by keeping trusting relations distinct from our strategic and self- interested relations – membership, alliance, association, coalition, network etc. – then the requirements must be set quite high. While they may suffice to mend a fractured strategic alliance, in the case of broken trust, for example, the words of an apology, confession, or the act of undoing or compensating are not enough. Compensatory actions can be powerful and sufficient means of mending some relations, such as those between governments and their peoples or financial institutions and their customers, but it is not enough to ‘respond’ to breaking a friend’s trust this way. A bank may re-secure our business by compensating us, but a friend who steals from us would not be likely to secure our forgiveness just by giving the money back. A proper response to breaking someone’s trust requires genuine remorse, regret, distress, concern, guilt and other appropriate feelings. No matter how person-like it appears, an entity that cannot respond with real sentiment, reflection, and understanding, should not be trusted, since whatever
I.3 The Object of Trust 77 it is responding to – a need, a desire, a command, a line of code, a fiscal benefit, a law, a recommendation – it is not responding to us, to our trust. On this point, Domenicucci and Holton, speaking of trust in the banking institutions, comment that the banks seem to want the kind of discretion that we give to people when we genuinely trust them – to wit, banks want the freedom to do what they want without regulation (2017: 157). Taking this as potential evidence, they say that institutions can be person-like in many ways and we can, in some instances, interact with them as though they are persons without “absurd anthropomorphising” (2017: 158). However, what the banks want, which is to be allowed to do what they want without regulation, is actually very far from trust. In our trusting relationships, it is true that we want a certain amount of discretion. If we are being gratuitously micromanaged, then we may legitimately complain that we are not trusted. But trust is not incompatible with monitoring, though there is a commonly held belief that it is. To provide a recent example of this thinking, Paul Faulkner has agreed with Elster (2007) that, when we trust, we refrain from taking precautions against the trustee even when they could act in a way that might seem to justify precautions (2017: 109). Faulkner thinks that if we take such precautions – such as consolidating a handshake with a legally binding contract – then such actions preclude trust. Elsewhere he has said “Too thorough an assessment of the risk is inimical to trust” (2007b: 879). There is something importantly true about this idea – sometimes monitoring behaviour can reveal underlying distrust. Yet the opposite of micromanagement is not trust, but disengagement. Friends who do not register each other’s behaviour, far from trusting one another implicitly, fail to embody the kind of care and attention that trusting relationships require. Being a party to a trusting relationship involves ‘regulating’ one another, in the sense that we ‘monitor’ one another’s lapses of trust and trustworthiness, and – with humility if possible – lovingly bring these to one another’s attention. Our trust and trustworthiness are not fixed elements of our personalities, after all (I.1.1); they are something we create together in our relationship. This being so, in the course of a normal relationship occasional lapses are standard, and usually not fatal. As we have seen, monitoring has historically been viewed as wholly incompatible with
I.3 The Object of Trust 78 trust (see I.2.2), but in reality, there is no point at which monitoring necessarily turns into distrust. Other than this ongoing background monitoring, some situations call for a truster to monitor a trustee’s immediate behaviour intently. If the trustee is doing something difficult, dangerous, or important, a good truster may monitor their actions extremely closely. This – far from signalling distrust – may show nothing but the seriousness of the situation and a deep concern for all involved. For twenty-two years now, I have been a rock-climber, and that pastime involves a great deal of what is effectively intense micromanagement blossoming in relationships of profound reciprocal trust. Climbers who fail to monitor one another’s actions are not exhibiting their trust but failing to be worthy of it. We must allow context to determine what behaviour signals what. As for the banks, we have no good reason to consider their request for freedom without regulation as constituting a request for genuine trust. It is a request for freedom without regulation, nothing more. There is also something deeply concerning about trusting an entity that requests such things of us and yet cannot respond to us in the way that genuine trusters do, a fact that Domenicucci and Holton do note (2017: 158). Other than their claim that some institutions desire genuine trust, Domenicucci and Holton think that we may be able to trust some non-person entities because it makes sense in some situations to take the participant stance towards them, and this, they say, indicates genuine trust as opposed to mere reliance. Indeed, a number of philosophers have held that reactive attitudes signify trust in the strong sense that this is precisely how we should mark the distinction between trust and reliance (Baier 1994: Ch.6-9; Hieronymi 2008b; Holton 1994; Jones 2004; McGeer 2008; McMyler 2011: Ch.4; Walker 2006: Ch.3). If reactive attitudes like resentment and betrayal do signify trust, and if we can and should have them towards some institutions, then this would seem to compel us to accept that those institutions are the proper objects of trust. Here I think Katherine Hawley has got the right idea. Beginning from the observation that “even the most inflationary accounts of collective agency hold back from treating such entities as fully-fledged persons in every respect, on a par with individual human persons” (2017: 231), she asks whether the distinction between trustworthiness and reliance, which is so important when it comes to persons, remains
I.3 The Object of Trust 79 pertinent in the case of groups and organisations. Clearly observing the distinction between the relevant empirical and normative questions, Hawley says “certainly some of us seem ready to react to groups and organizations with attitudes like loyalty, gratitude, resentment, and a sense of betrayal. If groups are not genuinely trustworthy or untrustworthy with regard to their testimony, merely reliable or unreliable, then such reactions would seem to be mistaken” (2017: 243). Yet she does not give the standard reply that such reactions are always mistaken:
reactive attitudes connected to ‘trusting’ a group can sensibly be directed at individuals connected to the group. But is this really feasible where the audience does not know who these individuals are? Yes: you easily resent the person, whoever it was, who wrote graffiti on your front door, and your feelings are quite different about the wind which inconveniently blew litter into your garden. Likewise, you can resent the individuals who contributed to the publication of a misleading report, even if you do not know who those individuals are (2017: 244)
To provide a contrasting argument which may help us to appreciate the significance of Hawley’s insight in the current debate, Stephen Darwall says:
We take ourselves to be entitled to expect of other drivers that they will drive responsibly and to be justified therefore in objecting and having second-personal holding-accountable attitudes like resentment and moral blame when they do not. But when, however, one is cut up by an overly aggressive driver who is a stranger, it would seem odd to have participant responses that are distinctive of trust, to feel let down or disappointment in the driver, or personally hurt or betrayed. These would seem to imply some form of personal relation that would not exist in such a case. (2017: 46)
Hawley thinks we can take the participant stance towards strangers; Darwall disagrees. Yet here, both Darwall and Hawley are speaking truth. Darwall is right in his broader point that trust is the sort of thing that occurs only when we know someone personally, but Hawley’s observation, as I would put it, that elements of our trust – such as the response of resentment – can sensibly be directed at members of a group, even when they are unknown to us, is right too. How is this possible? Well, as Hawley perceives,
I.3 The Object of Trust 80 we do not need to move from the claim that resentment can sensibly be directed at a stranger, to the claim this is any indication of trust and trustworthiness:
there are individuals who may legitimately be resented for creating or perpetuating an agency whose main function is to mislead. Whether this involves a failure of trustworthiness, rather than some other sort of moral or political failure, may depend upon which detailed account of trust and trustworthiness we espouse, but reactive attitudes reach beyond the domain of trustworthiness and untrustworthiness in any case (2017: 245)
Reactive attitudes can “reach beyond trust and trustworthiness”, they can exist without it constituting a case of genuine trust/worthiness. Just as we can rely on a person or believe what they say without trust, sometimes we can resent a person without there being any underlying trust between us. We may not even know the person towards whom our resentment is directed. To Hawley’s observation, I add the related observation that we can trust a person and resent them without the two being in any way connected. Perhaps we resent our badminton partner, who is also our trusted dear old friend, just because she keeps beating us at badminton. On the two-place account presented here, any single part of trust and trustworthiness – an attitude of resentment, a feeling of betrayal, a belief in the trustworthiness of the trustee, a sense of security – can exist in any relationship without it necessarily constituting a case of trust and trustworthiness. Trust and trustworthiness are made up of a constellation of elements, which are not fixed. We can experience elements of trust towards entities and the unknown individuals within them, just as we can experience elements of love towards them. We might grieve the break-up of our favourite band or resent the ‘actions’ of our educational institution. We might feel secure with our current bank or believe our customers to be loyal. Yet such entities and the unknown individuals within them will still usually fall short of being the proper objects of trust, since the experiences they have of us and create in us are still importantly dissimilar and usually inferior to those we have of and create in one another in our close reciprocal interpersonal relationships. At the end of the day, having a friend is different from ‘being a friend’ of The National Parks Foundation. Hawley herself concludes that:
I.3 The Object of Trust 81
whatever our terminology, if the trustworthiness–reliability distinction lacks merit in the group case—or, in other terms, if there is a type of trustworthiness which individuals but not groups can exemplify—then this is of significance to wider debates. Even in the individual case, many public concerns about ‘trustworthiness’ are really about reliability, but they are pressing concerns nonetheless... many public concerns about trust and groups are in fact best construed as (genuine, pressing) concerns about trusting individuals in a group context. (2017: 247)
The issues discussed under the banner of ‘trust’ in international relations, sociology, economics, law and so on are extremely important. They are no less important for being about relations that are strategic, mutually beneficial, or self-interested, rather than trusting. If we are to address these concerns adequately though, then we must frame them adequately. If something is not an issue of trust but an issue of reliance or is not an issue of ‘trust in groups’ but an issue of trusting individuals within a group, then we need to say so – explicitly. Friendships and family relationships work in different ways from coalitions and alliances; they produce different effects and have different causes. We cannot use the same concept to explain both the micro and the macro of human interaction. This subsection has not as yet made any claims as to what the proper objects of trust are, and, as promised, it will not do so. Instead, I will leave this discussion with a story of trust which did fail to target a proper object of trust, in the hope that it leaves the reader with some sense of my position on this issue:
My daughter’s first love – at three years old – was a tiny plastic styracosaurus. When this miniscule dinosaur (like a triceratops but with only one frontal horn) went missing for a day once causing immeasurable distress, and then turned up inexplicably in plain sight on the bathroom floor, she experienced all the characteristic signs of broken trust. Initially concerned for his safety, when Styro was found she felt deeply betrayed by the nature of his cavalier reappearance. Her beliefs about him, about what sort of dinosaur he was, were shattered. This was no accident – Styro had gone missing on purpose, just to hurt her. She had stern words with him and sought revenge by refusing to give him his usual place (clutched tightly in her small left hand even when sleeping) for some days. After a while, when she was satisfied that he understood his transgression and was sufficiently repentant,
I.3 The Object of Trust 82
she began to forgive him, though she never quite forgot the sting of that occasion. For the rest of their time together, if she ever grabbed Styro wrong and he spiked her with his one- horn, or if he again went missing, she would bring up and rehash with him his previous wrongdoing, as though the background of that incident somehow informed his current motivations. Their history together provided a continuing struggle for her throughout their time together, and, in it, because of it, she grew reflective, morally aware, and found some emotional maturity.
I did then, and do now, find nothing wrong with taking this to be an example of trust. Misplaced trust, but trust, nonetheless. And I do not agree with the assessment that my daughter’s trust in her cherished styracosaurus was irrational – her trust was not epistemically problematic, rather, her belief that Styro could respond to it was. Moreover, this was a case of misplaced trust, not because Styro was not a person, nor of course because he was untrustworthy. He could also never do long-division, but that did not indicate that he had dyscalculia. It was misplaced trust because Styro was not capable of being worthy of trust, even under all the transformative power of her sincere and earnest trust (and it was, indeed, sincere and earnest). The proper object of trust is that which can be worthy of it, that which can respond to it, be transformed by it, experience it. What this requires in a two-place account is not ‘doing what is expected’, ‘providing accurate testimony’, or ‘proving reliable’. Trustworthiness is equally as complex an experience to the trustee as trust is to the truster. Dear Styro, though he manifestly created complex trust, just did not have what it takes to experience it. To summarise, this subsection first revealed that the issue of the proper object of trust has been conceived of as an issue of personhood, and then demonstrated how a philosopher’s stance on this issue will be informed by their underlying understanding of the nature of trust. Acknowledging that some non-trusting relations can produce some elements of trust, this subsection argued that this does not require us to concede that such relations are trusting. Even when they share some similarities, there is a difference between our strategic and self-interested relationships, and our trusting relationships. The proper object of trust and trustworthiness, this subsection proposed, is anything that can both have and create in others the experiences of trust and trustworthiness. I have remained silent on whether any non-persons can live up to this requirement as this is an
I.3 The Object of Trust 83 empirical question far beyond my ability to assess, though I suspect that most human beings (including non-speaking and pre-verbal human beings) and some border collies would make the grade, and all political establishments and plastic styracosauri would not.
I.4 Trust 84
I.4 Trust
Trust is a two-place relation. It involves a truster and a trustee and not what is entrusted. When we rely on someone because we trust them, thereby entrusting something to them, that is not an example of trust but an example of trust causing reliance. But that it is a ‘two-place relation involving a truster and trustee’, is not all that needs to be said about trust. This section aims to flesh out the bare two-place schematic, thereby providing further reasons to accept it; it does not provide any formulae, strict definitions, or necessary and sufficient conditions for trust. Instead I shall proceed in my efforts to capture trust by first, in I.4.1, making use of an exchange between three characters in George Eliot’s novel Middlemarch, which is rich in its exploration of trust specifically and human relationships in general, and then, in I.4.2, by highlighting the similarities between trust and love. What we are left with at the end of subsections I.4.1 and I.4.2 is hopefully an impression of trust not as a behaviour, nor a behaviour coupled with, or caused by, any single thing, whether attitude or emotion, reliance or belief. Rather, trust has a complex, manifold, and variable nature, that takes its shape in reflection of a specific other– the trustee. This section then moves on to consider some of the outcomes of shifting from a three-place view of trust that is rooted in the truster’s immediate behaviour with regard to a specific act of reliance, to a two-place view, which takes trust to be something complex that arises in the trusters experience of the trustee over time. The aim here is to clarify some consequences of this shift and anticipate and address some objections that may be raised. To that end, I have chosen to focus on the issues that were of particular concern for three-place trust, as laid out in I.1. Subsection I.4.3 revisits the issue of the way the word ‘trust’ functions in the English language. Subsection I.4.4 discusses in more detail how trust can be compatible with monitoring the trustee. Subsection I.4.5, discusses what happens to the notion of self-trust on an account of trust as a complex private experience, and subsection I.4.6 comments on what this shift means for the question of the voluntariness of trust.
I.4 Trust 85
I.4.1 Middlemarch
When looking to give an account of trust in human relationships, rather than just looking to how the word ‘trust’ is used by speakers of just a single language, it is quite right to look to literature that takes human relationships as its subject matter. The novel Middlemarch, A Study of Provincial life by George Eliot (1871-2) is considered one of the greatest works of fiction in English, though interestingly it is also regarded as a work of realism in its study of human relationships. Victoria McGeer first used the excerpt from Middlemarch that I shall be using in a discussion of trust in 2008. In that paper, ‘Trust, Hope and Empowerment’, McGeer identified the important connection between trust and hope. I use the same excerpt as she, as I believe there is much more in it that has been missed. McGeer’s significant insight is valuable, though I do not strictly agree with her conclusion that trust is made rational by being underwritten by hope, for reasons that should by now be apparent. Adopting an underlying ‘three-place trust plus’ logic, McGeer intentionally runs the public acts that trusting produces (reliance, acceptance of testimony), and the private experience of trusting a person together, clarifying that she takes the capacity to trust to mean both:
the capacity to cultivate an appropriate set of attitudes and a capacity to act in ways commensurate with those attitudes. Hence, when I speak of trust, I mean the capacity to operate, both mentally and materially, in hopeful or trustful ways (2008: 237)
Our works thus differ in that she takes trust to be three-place, and I take it to be two- place; this dissertation explicitly does not take the “capacity to operate materially” as trust, but rather this is the capacity to act on, or act in accordance with out trust (see I.3). Moreover, our underlying motivations are different, as McGeer aims to demonstrate how trust can be rational, and this dissertation has argued that trust requires no rational justification. Trust is not something that makes us rational by being underwritten by hope as McGeer says; instead rationality is a part of trust, something employed in the service of trusting well. McGeer brings us up to speed on the relevant relationships in the novel so well, that I shall paraphrase her synopsis in what follows, then quote the exact same passage. The
I.4 Trust 86 passage in Middlemarch involves a conversation that takes place between three characters: Dorothea Casaubon, Camden Farebrother, and Sir James Chettam, as they discuss a fourth character, a young doctor named Lydgate. Dorothea is an intelligent, idealistic and enthusiastic young woman, who has befriended Lydgate, a talented, but naïve young doctor who has fallen into debt and difficulty and is suspected in Middlemarch of a scandal (though the evidence against him is circumstantial). The scandal involves a dubious character, a man named Bulstrode. Farebrother is a poor but clever vicar, generous in his feeling, and – like Dorothea – a friend to Lydgate. Sir James is in love with Dorothea and has only her best interests at heart. The conversation between Dorothea, Farebrother and Sir James is as follows:
[Dorothea says:] ‘Mr Lydgate would understand that if his friends hear a calumny about him their first wish must be to justify. What do we live for, if it is not to make life less difficult to each other? I cannot be indifferent to the troubles of a man who advised me in my trouble, and attended me in my illness.’ . . . ‘But, Dorothea’, [Sir James] said, remonstrantly, ‘you can’t undertake to manage a man’s life for him in that way. Lydgate must know—at least he will soon come to know how he stands. If he can clear himself, he will. He must act for himself.’ ‘I think his friends must wait till they find an opportunity’, added Mr Farebrother. ‘It is possible—I have often felt so much weakness in myself that I can conceive even a man of honourable disposition, such as I have always believed Lydgate to be, succumbing to such a temptation as that of accepting money which was offered more or less indirectly as a bribe to insure his silence about scandalous facts long gone by. I say, I can conceive this, if he were under the pressure of hard circumstances—if he had been harassed as I feel sure Lydgate has been. I would not believe anything worse of him except under stringent proof. But there is the terrible Nemesis following on some errors, that it is always possible for those who like it to interpret them into a crime: there is no proof in favour of the man outside his own consciousness and assertion. ‘Oh, how cruel!’ said Dorothea, clasping her hands. ‘And would you not like to be the one person who believed in that man’s innocence, if the rest of the world belied him? Besides, there is a man’s character beforehand to speak for him. ‘But, my dear Mrs Casaubon,’ said Mr Farebrother, smiling gently at her ardour, ‘character is not cut in marble—it is not something solid and unalterable. It is something living and changing, and may become diseased as our bodies do.’ ‘Then it may be rescued and healed’, said Dorothea. ‘I should not be afraid of asking Mr Lydgate to tell me the truth, that I might help him. Why should I be afraid?... [I could]
I.4 Trust 87
ask for his confidence; and he would be able to tell me things that might make all the circumstances clear. Then we would all stand by him and bring him out of his trouble. People glorify all sorts of bravery except the bravery they might show on behalf of their nearest neighbours.’ . . . ‘It is true that a woman may venture on some efforts of sympathy which would hardly succeed if we men undertook them’, said Mr Farebrother, almost converted by Dorothea’s ardour. ‘Surely, a woman is bound to be cautious and listen to those who know the world better than she does’, said Sir James, with his little frown. ‘Whatever you do in the end, Dorothea, you should really keep back at present, and not volunteer any meddling with this Bulstrode business. We don’t know yet what may turn up. You must agree with me?’ he ended, looking at Mr Farebrother. ‘I do think it would be better to wait’, said the latter. (Eliot 1871-2: see 1994, Ch.72, 734–5)
If we can agree that Dorothea trusts Lydgate, while the other characters – Farebrother and Sir James – do not, then we can make some good headway. Though the other characters may not necessarily distrust Lydgate, it is fair to say that they have something less than full, proper or robust trust. We are interested in the nature of trust, so what exactly is it that Dorothea has that her interlocutors lack? Let us first start with the familiar three-place analysis (A trusts B to do X) and see how far it gets us. Filling the details from Middlemarch into a three-place construction, we get:
Dorothea trusts that Lydgate did not accept any bribe money from Mr. Bulstrode.
We also get:
Mr. Farebrother and Sir James do not trust that Lydgate did not accept any bribe money from Mr. Bulstrode.
The idea that Dorothea’s trust, and Mr Farebrother and Sir James’ distrust, are all epistemically grounded three-place cognitive assessments of Lydgate’s conduct with regard to Bulstrode’s bribe, does at first seem plausible. She does say “and would you not like to be the one person who believed in that man’s innocence, if the rest of the world belied him?” suggesting that her trust may indeed involve a cognitive belief that he has done as entrusted. Yet Dorothea’s trust does not seem to require the belief that he did not accept Bulstrode’s bribe. If this were so, then maintaining her trust would demand that she ignore the evidence against him, just as a number of epistemologists
I.4 Trust 88 have pointed out (see I.1.3). Indeed, what would we make of Dorothea’s trust if she had said “I trust Lydgate not to have accepted Bulstrode’s bribe, though I believe he did accept Bulstrode’s bribe”? The three-place analysis can make no sense of a person’s trust remaining robust in the face of the belief that the trustee may not have done as entrusted, and yet, this is exactly what happens. Dorothea does not ignore the evidence – on the contrary, her trust compels her to seek it out. She goes on to say: “I should not be afraid of asking Mr Lydgate to tell me the truth” so that this might “make all the circumstances clear”. Meanwhile the other men, lacking trust, urge her to leave matters alone. It is the distrusters in this scenario that recommend turning away from the evidence. Dorothea is not approaching the situation with a kind of irrational blind faith in Lydgate’s behaviour. Instead – ardent in her trust – she is clearly wide awake to the possibilities regarding what he may, or may not, have done. There is another way we could fit this situation into a three-place conception. Perhaps Dorothea’s trust is not in any specifics about what Lydgate has done, but is more general. Perhaps her trust is still three-place, but it is in his character:
Dorothea trusts Lydgate to have a trustworthy character.
She does say “there is a man’s character beforehand to speak for him”, suggesting that something like this might be going on. The problem is that when Farebrother points out to her that “character is something living and changing, and may become diseased as our bodies do”, Dorothea does not reject this possibility as she would need to do, again, if her trust was a three-place cognitive assessment of his character. Instead she maintains that even if it were true that his character has become corrupted, it may still be “rescued and healed”. It is not that she trusts Lydgate to have a trustworthy character, rather that her trust means that that she is not prompted by the possibility that his character has become “diseased” to think it is all hopeless and to give up on him. important measures of her trust at this point in time. But alone they are not enough. Farebrother also believes Lydgate is respectable, “a man of honourable disposition” he calls him. Yet Farebrother is able to maintain a pessimism about Lydgate that Dorothea cannot. He slides into despair, falling victim to the faulty reasoning – and it is faulty reasoning – that the corruption of a person is hopeless. Locked in such pessimism,
I.4 Trust 89
Farebrother is not able to see, Dorothea must remind him, that people have the capacity to change, to redeem themselves. Dorothea, unlike Farebrother and Sir James, experiences an attitude of optimism, of hope, towards Lydgate. And as McGeer rightly identifies, Dorothea’s capacity to hope is an important part of her trust. Indeed, hope is what sustains our trust when the evidence is irrefutably damning. Yet optimistic hope itself is built on other things. In Dorothea’s case, her hope requires her memory. She is able to recall a man she once knew, who “advised me in my trouble, and attended me in my illness” and this enables her to imagine that he may be brought back to this self, no matter how far he may have diverged from it. Imagination is thus also crucial: we cannot hope for a future that we cannot first imagine. So far, we have found that Dorothea’s trust consists of beliefs and attitudes, imagination and memory. Yet it is fair to say that her trust is constituted by desires as well. She desires firstly, to defend Lydgate. If she did not, then we might seriously doubt her trust. And she desires to seek the truth, and then, to set things straight, or to deal with his mistake however uncomfortable it may be. If he has become untrustworthy, then she desires to redeem him. Dorothea, like all good trusters, is strongly compelled by these desires: “what do we live for, if it is not to make life less difficult to each other?” she catechises. Yet although she desires to know the truth, this desire is not an emotionless cold cognitive desire, the kind a detective charged with investigating Lydgate might have. Dorothea’s desire to understand is driven by her emotions. She “cannot be indifferent” as Sir James is. Moreover, her words belie her outrage, indignation, sadness, and concern, and these can be added to our list of things that make up her trust. If she lacked such emotions, then she would not experience Lydgate in the way that she does, and she would not be compelled to act in his defence just as the other two men are not. Trusters are emotionally responsive to their trustees. But there is no list of emotions that constitute trust in general. What trusters experience emotionally will depend on the circumstances. We can imagine Dorothea’s trust manifesting in experiences of relief, joy, and a sense of gratification, were Lydgate to be exonerated.
I.4 Trust 90
What Dorothea does not experience is perhaps more telling. She is not uncaringly confident. It is not composed certainty or assuredness– it is not blind faith that she has. And she does not resent Lydgate nor feel betrayed by him. Though she wants to know the truth, she is neither suspicious nor doubtful, and she is not pessimistic. Moreover, although she may become entangled in his considerable difficulties, tarnishing her own reputation as Sir James rightly points out, she is notably unconcerned with her own fate. There is, in her trust, a certain selflessness, a generosity. In contrast, Sir James is markedly ungenerous, being concerned only with how the matters affect him, or affect other things that he cares about, such as Dorothea. Generosity, like so many of the complexities of trust, is also part of what it is to trust, a part that Eliot draws attention to when she says:
The presence of a noble nature, generous in its wishes, ardent in its charity, changes the lights for us: we begin to see things again in their larger, quieter masses, and to believe that we too can be seen and judged in the wholeness of our character.
Being able to experience other people in the way that Dorothea experiences Lydgate, with trust, is also not an accident of luck. Much of Middlemarch concentrates on how Dorothea refuses, though many opportunities to do so present themselves, to sacrifice her ideals and fall into cynicism and despondency regarding the people around her. When she is confronted with a seemingly endless stream of moral and social failings, cultivating her trust engages her reason. As we can see, she provides her interlocutors with reasoned arguments as to why trust should be the preferred approach, even in the face of so much disappointing humanity. But what enables her to take the stance that she does towards Lydgate is not reason alone. The idea that Dorothea’s trust in Lydgate is constituted by some simple three-place epistemically based trust that he did not do what he is accused of doing, is exposed as grossly inadequate in the light of the complexities of trust that are apparent in Eliot’s rich example. As is the idea that Dorothea trusts Lydgate “over certain domains” with the implied exclusion of others. And it is clear that she has not merely made a rational assessment of his behaviour, character, or of his particular “commitments and competencies” in the way a disengaged bystander might be able to do. Dorothea’s trust in Lydgate is not a three-place relation
I.4 Trust 91 at all. It is not the case that she trusts him to do or even to be anything. What’s more, it is not possible to capture Dorothea’s trust by tagging on to a three-place analysis any one single thing, such as an expectation of performance, goodwill, hope, or belief in his trustworthiness. Rather than interpreting the ongoing disagreements about trust as revealing that there are in fact ‘multiple kinds’ of trust, instead, I propose, it reveals that trust is multifarious in its inherent nature. Many phenomena that have been identified by philosophers before as being connected to trust are vibrantly present in Eliot’s example. But it is only by recognising the complexity of Dorothea’s experience that we might hope to account for her trust. Her ‘act of trust’ towards Lydgate is not rooted in – and so cannot be explained by – any single belief or attitude, nor any single emotion. Moreover, different phenomena will be in play and in different measures in different cases of trust: sometimes a steadfast belief in the trustworthiness of the trustee will be the most salient element; at other times, it may be the desire to defend them, or rely on them, or give them the benefit of the doubt. Occasionally our memories of the trustee will be what really matters, or our ability to reason that – however things look – they are still deserving of our trust. Domenicucci and Holton have come somewhat close to this assessment, in the context of their doubts that any of the things that have been identified as being parts of trust – a readiness to feel betrayal, belief that the trustee is trustworthy, tendency to give the benefit of the doubt, readiness to testify to others, readiness to rely, hopeful or optimistic view of the trustee etc. – are strictly necessary for trust (2017: 154). Where my account diverges from theirs can be gleaned from their discussion of the extent of trust (2017: 155-56). In that discussion, they ask how trust can be two-place when it is the case that our trust can be different in different spheres, i.e. we trust our partner except where alcohol is concerned or trust our plumber to fix the hot water system while not trusting them with our bank details, etc. The relevant passage is here:
the first thing to say is that the two cases we have given are rather different. The first involves a general attitude of two-place trust, with a qualification: the trust gives out when alcohol is involved. Such qualified trust is straightforwardly handled on the two-place approach. The two-place attitude is still fundamental. One arrives at the qualified attitude by
I.4 Trust 92
starting with it and knocking something off. Other cases involve a similar pattern, motivated by a welter of considerations about competence and motivation: ‘I don’t trust her when it comes to dogs’; ‘I don’t trust him once his relatives are involved’ and so on. The case of the plumber cannot be thought of in this way though. It is not as if one’s attitude to one’s plumber is like one’s attitude to one’s partner, except qualified in various ways... we don’t start with lots of trust and then reduce; quite the reverse; we start with rather a little, and then, perhaps if things go well, we add more. Nevertheless, we suggest that the two-place model is more revelatory, even in the case of the plumber. We still trust them, first and foremost, as a person; the variation comes from the fact that, given a typical relationship with a plumber, what is required to trust them as a person is radically less than is required to trust one’s partner. (2017: 155-56)
Taking seriously the idea that one might trust ones’ partner except where alcohol is concerned or trust one’s plumber only when it comes to fixing the hot water system, relies on a three-place logic underlying this line of reasoning. What Domenicucci and Holton call ‘two-place trust’, is really a universal three-place trust that is qualified: A trusts B to do everything, except X, Y, Z etc. In the case of people that we do not know very well such as plumbers and politicians, the three-place trust is just more qualified. On my two-place account however, we no longer have any need to justify how it can be the case that we typically trust people to do certain things while not trusting them to do others, since when we move to a genuinely two-place view, not some hybrid two/three- place version, then we reject the idea of ‘trusting people to do certain things’ altogether. In summary, this subsection presented an excerpt from George Eliot’s novel Middlemarch proposing that one character in that novel (Dorothea) trusted another (Lydgate). By reviewing this brief exchange, we found Dorothea’s trust to be constituted not by any single attitude, emotion, or belief, but by a number of things. Dorothea’s trust was in Lydgate, as he was, as he had been, and as he might become.
I.4.2 Trust and love
This subsection provides further colour to a two-place view of trust by drawing on the similarities between trust and love. Such similarities have been noticed before, and this subsection begins by acknowledging as much. Then, I.4.2 presents a series of interrelated observations about trust and love, in a relaxed format. This subsection does
I.4 Trust 93 not provide an account of trust or love as such. The aim here is to collect and so demonstrate a few ways in which trust and love are importantly alike. While it is outside the scope of this research project to delve deeply into the literature on love, I would like to make specific note of a number of philosophical accounts of love that are relevant to this discussion and may differ in subtle but important ways from my own (Abramson and Leite ; Ben-Ze'ev 2001; Brogaard 2015; De Sousa 2015; Grau 2004, 2010; Kolodny 2003; Pismenny and Prinz 2017; Smuts 2013, 2014b, 2014a). Both Becker (1996) and Lahno (2001) have previously mentioned the similarities between trust and love, and more recently Domenicucci and Holton have compared them as well, though they caution against drawing too close an analogy (2017: 152). Interestingly, Brian Skyrms has provided a detailed analysis of empirical evidence that is usually taken as providing reason to believe that trust has an evolutionary basis in cooperation, to argue instead that it has an evolutionary basis in love (2008). The most detailed philosophical comparison has come recently from Stephen Darwall, who argues that trust and love are both “second-personal attitudes of the heart”. He says:
Trust can be welcome in some ways and burdensome in others: a source of confidence and encouragement on the one hand, and the yoke of expectation on the other. But personal relationships are complicated in just these ways… When we trust someone, we implicitly invite him to trust himself also. We regard him as trustworthy and bid him to see himself this way too. Something analogous is true of love. To love someone is to see him as lovable and to invite him to see himself this way as well. Someone’s not loving himself is disturbing to those who love him not just because of its effects on his wellbeing, but because it can make it difficult, if not impossible, for him to accept their love. (2017: 48)
Darwall identifies that trust and love are welcome and burdensome invitations, which bid their objects to see themselves as they are seen; and that when those we love or trust cannot see themselves as we see them, this can impede the very relation itself. Such parallels, once we turn away from the idea that trust is an epistemically grounded three- place relation similar to reliance or belief, are easy to come by. In what follows, I make a number of further observations aimed at building a stronger sense of the likenesses
I.4 Trust 94 between the two. I focus on trust and love’s complex two-place experiential nature, and the way these features contribute to our sense of a shared humanity. Trust and love may be good or bad, challenging or straightforward, joyful or miserable, result in action that is moral or immoral, right or wrong, desired or detested, or both, at different times, and even at the same time. This makes them complex experiences. What ties instances of trust and instances of love together is the shared understanding we gain from becoming experienced trusters, lovers. Trust and love produce common lessons, common sympathies, hence they constitute important contributors to our sense of a shared humanity. No matter how different we are, no matter how differently our lives and relationships go, having experienced friendship with one true trustworthy friend who is genuinely loved affects us all in similar ways, provides us all similar opportunities for personal, social, and moral growth. We are subject to countless complex experiences throughout the course of our lives. Doing a PhD is a complex experience. Raising Irish wolf hounds is a complex experience. Complex experiences are unique and private, in the sense that they belong only to the person experiencing them. Yet complex experiences give rise to public knowledge via their commonalities, and salient features. When PhD students get together and speak of their PhD experiences, even if those experiences vary greatly (good ones and bad ones, engineering ones and philosophy ones), there is shared understanding, and this shared ground forms a particular kind of club. Many undertakings require endurance, and are frustrating, and lonely, but not in the same way as doing a PhD. People who have shared any common complex experience (living for a time in France, marrying into a difficult family) can relate to one another with understanding; and because most of us have loved and trusted, trust and love come to play their significant roles in our sense of a shared humanity. Unlike the PhD Club or the Irish Wolf Hound Club, the trust and love clubs have highly inclusive membership. Through language and cultural barriers, our understanding that the death of a loved one will be grieved, that betrayal explains revenge, that forgiving does not entail forgetting, that love and hate are contraries not contradictories, binds us all and allows us to share in one another’s histories, stories, purposes, and practices.
I.4 Trust 95
Our sense of a shared humanity does not just come from trust and love of course. There are many private experiences that we all ‘share’– pain for example, or sunshine. Yet unlike pain and sunshine (or puberty and existential angst), trust and love are two- place experiences one person has of another. Trust and love take specific and irreplaceable objects. By taking a unique object and so having in each instance a unique form, trust and love are doubly unique. When we love, we experience the particular beloved in a particular way, in a way that is different from how we usually experience things of that sort, and different from how we experience anything else that we trust/love. The objects taken by love and trust thus become special. What I mean by special is this: we cannot undo the loss of a deceased child by having another. We cannot replace the trust lost in a lost friend with the trust gained in another. At most, new objects of trust/love can fill the void left by the ones who have gone; but new loved ones, new trustees, will fill it in their own unique ways. Because people are all different, because our history together will be different, each object will be loved or trusted differently – not more or less necessarily, but differently. My daughter inspires more devotion; my son, more joy; my daughter more considered attention; my son, more playful attention. My trust of my mother is complex in ways my trust of my partner never will be, yet my partner can betray in ways she cannot, and, even so, my forgiveness of him comes more quickly. The whole complex character of my trust or love for each of them responds to them and is shaped over years into its own particular form. That is why we grieve the loss of love and trust so much: because their objects are uniquely special, they are truly irreplaceable. The uniqueness and specialness of each instance of love or trust means that we can only ever produce broad themes and patterns in our representations of them, rather than strict formulae, or conditions. Such broad themes have been so well-represented by humanity, that they seem to become worthless platitudes– love dies slowly, love is dangerous, love knows no bounds, love hurts, love changes everything, trust is precious, trust starts and ends in truth, trusting is hard – knowing who to trust even harder, trust takes years to build seconds to break and forever to repair. Yet these well-worn platitudes, which tickle us when we first encounter them (that is why ten-year-olds devote themselves to pop songs) contain powerful truths that strike us only when we
I.4 Trust 96 come to actually live them ourselves. Like all complex experiences, we come to know trust and love only via first-hand experience. No matter how many romantic or revenge novels we read, such impractical knowledge is practically useless. Hence, each generation makes the exact same mistakes in love and trust as the previous generation. One understanding we gain about them as we experience love and trust is their complexity. As a child I imagined love would be a new kind of emotion – a sort of euphoric exhilaration, like the falling-out-of-myself excitement of going on holiday. It was. But this was only one tiny part of it. With experience, I understood that love is sometimes manifest in this feeling – especially when it is romantic – but love is not just a feeling. Like trust, love is not just a desire for the beloved, or a belief in the loveliness of the beloved, or a hope that the beloved will reciprocate the love, or an optimism that everything will work out all right in the light of love, or a feeling of butterflies when the beloved is near, or a disposition to become distressed if what is loved is in peril and grieve if what is loved is lost. Love is all of these, and more, and sometimes it is none of these. Sometimes love is all darkness and desperation (and not just in the bedroom). Trust and love cannot be captured with general necessary and sufficient conditions because only trusters and trustees, lovers and beloveds, get to say what is necessary and sufficient for their trust or love. Each person’s trust or love comes with its own unique set of conditions, and the conditions change depending on who is asking for it. There are commonalities, but there is no set list of human betrayals. What counts as a betrayal depends on you and me and our shared history. Some relationships survive infidelity, others do not. Some love fortifies during periods of long illness, other love distorts or fades. Some trustworthiness is rooted in honesty, some in benevolent lies. There are patterns, hence there are platitudes, but no general conditions can be given which tell us how love and trust will behave, what effects they will have, what will mend them or break them. All this is entirely dependent on the individuals involved and the nature of their unique relationship. Embracing trust and love as complex two-place relations need not entail that trust and love have no explanatory power. Far from it. The understanding we gain in our experience with them is precisely what allows us to grasp the complexities and inconsistencies of human behaviour, human relationships. We can watch foreign films
I.4 Trust 97 and read translated novels, and, even if we do not fully understand the cultural practices surrounding them, we have no trouble usually recognising who is in love, whose trust has been betrayed, and so, why they may be elated, grieving, seeking revenge, etc. Trust helps us to understand one another, on the level that we inhabit in our everyday lives, in our everyday interactions with those we live with, care for, and need. Trust and love cannot explain cooperation (in a general sense), economic systems, or sweeping social or political movements, but they can help us to understand, for example, why John moved out of Sarah’s apartment even though he had only just moved in; why Lucy left us all (her dearest friends) to be with Vyacheslav in Siberia; or why Aunty Jude yells at everybody over nothing every year on the anniversary of Cousin Martin’s death. These are not things we understand well as small children, when we are often left frightened and confused by the behaviour of adults as it relates to love and trust. Trust and love are also alike in that they both begin when we first encounter the person who will go on to become someone we trust/love. We cherish our children’s births and old friends and lovers reminisce over their first chance meetings because this is where the story starts, no matter how long it takes for love and trust to blossom. We can sometimes pinpoint a ‘moment’ when it became apparent to us that we loved or trusted somebody, but this is to have a realisation of something that is already happening. It is not really ‘the beginning’ as it were. Such realisations happen because much is already happening in the background. George Eliot calls this sort of thing being:
humiliated by the sudden discovery of a fact which has existed very comfortably and perhaps been staring at us in private while we have been making up our world entirely without it. (1871-2: Ch.35)
Because people grow and change, trust and love, which tie people together, never stay long in their current forms. If they do, they can fracture. Iris Murdoch’s character, the priest Brendan Craddock, says “people often start by falling in love, and they go on for years and years without realising that the love must change into some other love which is so unlike it that it can hardly be recognised as love at all” (Murdoch 2010).
I.4 Trust 98
The love we have at the beginning of a relationship is more like blind desire, the trust, more like blind faith. As they develop, love and trust involve not a turning away, but a turning toward the beloved or trustee, seeing them as they are, and holding our ground in the face of their imperfections, limitations, and failures. As such, it could be said that trust comes in degrees, with it tending to grow and deepen with the passage of time, or break and wither as the result of certain events. The strength of trust and love to hold their ground lies in their very complexity. Emotions and beliefs, attitudes and virtues, these can be capricious things. But because love and trust involve so many complex elements, when one element is challenged or waning, we can use another to shore them up. If our belief that our trustee is not a liar is overthrown, we can remember a time when they were bravely honest. If our optimism that our beloved is worthy of our trust is challenged, we can reason that they deserve a second chance. If we find we care less than we should, we can desire to care more, and spark love anew with this new kind of desire. A belief or a feeling is not enough: it takes many tools to craft the disappointments we throw at one another over the course of any relationship into something that will last the decades. Because of their complex natures, love and trust break in similar manners. It is said that love dies slowly but trust breaks suddenly. Yet broken trust works more like a fatal stab wound. The act of betrayal is sudden and fatal, but even so, the full death of trust is slow. Some parts die abruptly, like the beliefs we had about our trustees that can change in an instant, but other parts take longer to diminish, like the hope, the memories. And, even beyond all hope, still, we may look – as Tim Hardin sang – to find a reason to believe, our rational mind searching for a way back to one another even long after love and trust are dead. Because of the similar natures of trust and love, trusters and lovers also share a form of vulnerability. Trusters and lovers are not vulnerable because they might have made a mistake, being ‘wrong’ in their love and trust, but because trust and love can sour into guilt, grief, bitterness, resentment, regret, paranoia, insecurity, self-doubt, and betrayal. These experiences are not parts of love and trust’s contraries (distrust and indifference). They are what trust and love can become if they are not properly regarded or not properly tended to.
I.4 Trust 99
What causes trust and love to sour into their curdled forms might be the behaviour of the object of trust or love; an ‘untrustworthy trustee’. But it does not have to be. Trust and love can sour following an ‘act of betrayal’ of course, but souring can also occur because of something that has gone wrong with the truster/lover. The suspicion, doubt, paranoia, bitterness, resentment, hatred, coldness, confusion or indifference that constitute distrust or a lack of love can be induced by illness and disease, or a lack of maintenance on the part of the truster or lover. It is not always the lover or trustee’s fault that they are unloved, distrusted. When trust and love are understood as experiences, rather than as acts, we can explain such unfair and unwarranted changes without attributing any undeserved blame to those who should still be trusted, loved. Trust and love are both two-place relations of a similar complexity and play a similar role in our lives. They are both two-place relations because they are experiences of something– the trustee, the beloved, the object of trust or love – and they are complex because they involve a whole host of things occurring over some time. Love and trust are best understood, not as kinds of feelings, beliefs, attitudes, actions, etc., but as complex experiences of their particular irreplaceable objects. And since – though their precise forms are always unique and special – trust and love are common experiences, they play a valuable role in our sense of a shared humanity, helping us to understand and relate to one another even across cultural or language barriers. Indeed, language provides the final comparison – ‘trust’ functions in English quite a lot like ‘love’. Developing an account of the nature of love by starting from the way the word ‘love’ is used – I’d love it if you would go away and leave me alone to work in peace! – would likely result in a very confused account. Philosophers on love tend not to suggest that we must ‘expand the concept of love’ so that it has greater explanatory power, covering requests for cups of tea – I’d love another cuppa… or false- compliments – I love the way that dress hides your ankles… We are confident when it comes to love that these uses of ‘love’ are wholly irrelevant to the analysis. This dissertation has highlighted that the three-place analysis of trust reflects a common construction in the English language (I.2.4). It is interesting that in English, love has its own three-place equivalent: ‘I’d love you to get me another cup of tea’. Yet nobody is proposing any three-place account of this ‘kind’ of love (A loves B to get T). As the
I.4 Trust 100
Stanford Encyclopaedia of Philosophy entry on love makes it clear right at the outset, many common uses of the term ‘love’ form no part of the philosophical analysis of love (Helm, 2017). When addressing the question what is love, philosophers have not provided short reductive definitions, necessary and sufficient conditions, or logical formulas, but lengthy and colourful essays, such as Harry Frankfurt’s The Reasons of Love. ‘Trust’ and ‘love’ are common, convenient and co-opted words that do not pick out any single phenomenon– trust you to do something like that… I trust you have enjoyed your flight with Air New Zealand – these are Karen Jones’ examples (1996: 5) . That doesn’t mean, however, that trust and love are not singular phenomena. We must, when investigating trust, begin with the shared understanding we gain in our experience as trusters, and not with that ‘beguiling devil’, language. In summary, this subsection brought forth a number of interlinked parallels between trust and love as contribution to this dissertation’s case that we should treat them in similar manners in academic settings. In philosophy, we already treat love as a complex inter-relational phenomenon that plays an important role in contributing to our sense of a shared humanity. Given their intuitive and manifest similarities, we must treat trust in a comparable way. The following four subsections of 1.5 raise and briefly discuss some objections and upshots of this proposed pivot in thinking about trust.
I.4.3 Trust in language
Even if we accept that trust can be a two-place relation, the fact is we more commonly use the word ‘trust’ in the three-place construction, and we do so in contexts that really do seem to be about trust. Are you proposing we do away with the three-place construction?
This dissertation has argued that language is not a good starting point to argue for any particular conception of trust. I add here, that it is also not a good starting point to argue against any particular conception of trust either. Domenicucci and Holton have proposed that because in English there is no three-place syntactic construction of distrust, we should be cautious of the notion of three-place trust:
Strikingly… even in English there is no three-place syntactic construction of distrust. We do not say that we distrust someone to do something. We simply distrust, or mistrust, a person.
I.4 Trust 101
But if distrust is in some important way a contrary of trust, and so inherits the basic form of trust, that suggests that trust itself is primarily an attitude to a person. (2017: 150)
This dissertation agrees with Domenicucci and Holton, and Katherine Hawley (2014) that distrust is the contrary of trust, and so, as this dissertation has it, distrust is a two-place relation: a complex way one person might come to regard or experience another. Yet any argument that moves from a lack of any particular construction in any particular language to the claim that the concept or the phenomenon itself does not exist is suspect, as I have discussed already (I.2.2). Though it is unnatural to say so in English, the idea of three-place distrusting someone ‘to do X’ – in the sense that we are suspicious of their competence or motives with relation to some specific area – is not so bizarre. The idea of a three-place form of distrust that is not absolute but rather causes a failure to rely in some instances, is, I think, perfectly within our grasp– even if it is, in the end, a kind of confusion. That the two-place construction of distrust is unnatural in any given language is not an argument for two-place trust. To their credit, Domenicucci and Holton do say that this observation is only suggestive, and, acknowledging that linguistic considerations cannot suffice to make their point (2017: 151) they go on to provide firmer arguments. This dissertation, as it relates to the word ‘trust’, should not be read as a proposal that we change language but as an argument that language is not a good starting point to analyse trust (I.2.4). What has been argued, is that all those apparent three-place ‘examples of trust’ (leaving your diary open on your desk, lending your car to a friend) obscure the important reality that trust and behaviour are distinct: we might have trusted and not acted on it, or acted as though we trusted when really, we did not. When we ‘see’ an ‘act of trust’ (or an ‘act of love’) such as cooperative activity (or an affectionate embrace), we are always making an inductive inference. It is a consequence of the shift to a two-place understanding as I have presented it in the previous subsection (I.4.2), that the only person whose trust (or love) we might have certainty of is our own, since we are the only person undergoing the complex private experience. This understanding clarifies how trust is risky: trust is risky not because we cannot be certain what the trustee will do, but because we cannot be certain that our trustee is experiencing us the way they would need to in order to be worthy of our trust.
I.4 Trust 102
Karen Jones says:
acting as if you trusted and genuinely trusting could have the same result only on the assumption that there is no perceptible difference between the behaviour that would be produced from trust and the behaviour that would be produced from acting as though you trusted [which she takes to be an implausible situation] (1996: 22)
Jones is likely correct that we can perceive a difference between trust and trust- pretence, since we do get a sense sometimes that we are not trusted, not really. However, taking ourselves to be trusted when really, we are not, is hardly an implausible situation. People do discover (and are shocked to discover) that they are not really trusted when they have been going along for some time perceiving (wrongly) that they are. Such discoveries do not always occur via a sudden perception of the lack of trust; a frank admission or a third-party disclosure is often what brings things to light. Now and then there is no perceptible difference between trusting and behaving as though we trust; we can be deceived not only about who is trustworthy, but also about who is trusting. It is better to embrace a two-place account that locates trust in the private complex ongoing experience the truster has of their trustee, and not in their public behaviour, otherwise how do we make sense of those times in which we only ‘find out’ through admissions, that we are not trusted, or worse, not loved. Making the shift to a two-place understanding of trust need not entail doing away with the three-place construction in language or with the idea of three-place trust altogether. What is prescribed is that we 1) understand three-place trust differently, and 2) take more care with language in certain settings. Regarding 1) – the alternative picture of three-place trust that was presented in I.3.2 enables us to hold on to the idea of trust qua act by shifting our understanding of three-place trust as a kind of trust, to three-place trust as a kind of reliance. Examples of so-called three-place trust are examples of reliance that has been affected, caused or influenced by trust and so constitutes a special kind of ‘trusting-reliance’. Nothing really is lost, but more precision and clarity has been gained. If someone says – for example – “the patient trusted that the doctor prescribed the correct medication” we do not need to think this is nonsense, or nothing to do with trust. We can simply interpret this as meaning that the
I.4 Trust 103 patient believed the doctor to have prescribed the correct medication, because he trusted her. Similarly, the statement: “I trust him as a father and not a doctor” doesn’t make much sense on my two-place conception. What we are saying in such cases, is that we have a trusting relationship with a person (our father), and yet, we would not rely on him in some particular capacity (medical care). On two-place trust, trust can feature in explanations of behaviour, and we can maintain the all-important distinction between trust and behaviour. Regarding 2: this dissertation is not proposing that we do away with any talk of “trusting to do”. It rather argues that because the three-place trust construction can capture two distinct things (trust and reliance), or a relationship between two distinct things (trust causing reliance) or only one thing (just reliance) we must take care in using it. Moreover, often the three-place construction is used in contexts that are not even about trust or reliance. Take the following three-place sentence:
I don’t trust Chloe when it comes to dogs.
This may have several intended meanings:
1. Don’t let Chloe look after your dog! She is so irresponsible with animals. 2. Don’t let Chloe go to the dog show – she’ll come home with another puppy! 3. Chloe is terrified of dogs and behaves very irrationally around them. 4. Chloe is ignorant and misinformed about dogs. Sentences containing the three-place trust construction have various functions: to warn, threaten or manipulate, to justify, induce guilt or compliance, to reassure, comfort, encourage or pressure. In saying “I don’t trust Chloe when it comes to dogs”, the speaker may not share any background of trust with Chloe at all. What is more, even if the speaker does trust Chloe, this may be no part of what is being communicated. When we say to our teenager (whom we do in fact trust) “I am trusting you with the car”, use of the word ‘trust’ may just be serving as a warning to drive safely, and when we say to our partner (whom we do in fact trust) “I was trusting you to clean up the house and you’re just sitting here playing Solitaire!” we may just mean to induce guilt. We can use ‘trust’ in a three-place construction when there is no trust to speak of. In such cases, ‘trust’ is just a synonym for ‘rely’. And, we can use ‘trust’ in a three-place construction
I.4 Trust 104 in the context of a background of trust with the person we are referring to, and still not actually mention trust. We can also use ‘trust’ in a three-place construction and not refer to trust or reliance, but instead just mean to warn, induce guilt etc. When seen this way we realise, though a relatively common word, just how little we talk about trust. The suggestion put forward here, is that: 1) given that trust is not a three-place relation and 2) given that the three-place construction may not mention trust, or 3) may mention trust in an ambiguous manner, those discussing trust in some settings treat three-way constructions with caution. In some instances, especially when there is a pressing need to emphasise the role trust has played in the outcome (X), rephrasing restrictive clause (that) three-place trust statements as non-restrictive clause (which) statements can be a useful tool for English-speakers to keep these separate notions apart, so that the reader understands the role of trust and its significance. For example, if we think trust has played a significant role in the outcome, then instead of:
Parents trust that vaccines administered by their family general practitioners are safe, but they are less trusting when it comes to vaccines administered by doctors at medical centres.
Researchers could write:
Parents trust their family general practitioners, which we suggest played a role in the higher rate of compliance in these settings than in medical centre settings, where doctors and patients have often never met before.
The first three-place restrictive-clause way of framing the situation (A trusts that…) is less informative and more ambiguous. This may be what is intended. Trust may not be the key factor, or we may not know whether or not it is. Perhaps the researchers who found this discrepancy in compliance mean ‘trust’ synonymously with ‘believe’ and go on to explain that parents have been led by a false-news reports to believe that medical centres are stocked with inferior vaccines. Yet if it is suspected that trust was the key factor, then the second non-restrictive clause two-place way of framing the situation (A trusts B, which…) unambiguously points to trust, to the relationship, as the factor that made all the difference.
I.4 Trust 105
Trust and belief, trust and compliance, are separate issues, even though they can be powerfully connected. If it is important, then being reminded with non-restrictive clause language that links trust and reliance, rather than implying trust is reliance, may prompt those reading (policymakers for example) to better interpretations. In this case, the non- restrictive clause version would likely prompt the reader to reason that relationships are important in some medical contexts. For instance, if we want to increase parental compliance with the childhood vaccination schedule in medical centres, then creating opportunities for better continuity of care in medical centres may prove more effective than, say, our current policy of making sure they are heavily stocked with vaccine information pamphlets. With the shift to two-place trust, those meaning to report on the ways trust is shaping our behaviour, have a clear way to conceptualise and communicate their point. As a final example, imagine that you meet an elderly couple on a train. They appear utterly harmless and friendly, and they invite you to stay at their home that evening. You graciously accept, and the evening goes well. However, in the morning you awake to find that your wallet has been stolen, and the elderly couple are nowhere to be found. Immediately you jump to the (correct) conclusion that it is them who have stolen your wallet: you have been tricked, hoodwinked, ripped off, robbed. What is your reaction? Of course, individuals will differ, but I suggest that a likely reaction would involve a prompt trip to the local police station to report the crime. Internally, you may experience anger, annoyance at the inconvenience of it all, and perhaps some sense of betrayal. You may feel less safe than you did before and be less likely to rely on other strangers in the future. You may say out loud “I trusted those two!”, so, isn’t this a straightforward case of broken trust? No doubt, it has some of the hallmarks. Yet the distinction I am making can be seen if we think about what happens next. In this example, your main concern isn’t to find out why your robbers stole from you, nor to reform them. There is no complex reasoning to engage in to understand the events, nor complex emotions to overcome. You have been robbed; it happens. Your only concern is to report these two charlatans, so they may be caught. Your mind is on getting your possession back, and perhaps preventing others from suffering the same fate. In contrast, suppose that you are staying with your own dear brother for the night.
I.4 Trust 106
Again, you awake to find your wallet has been stolen, and your brother is nowhere to be found. What is your reaction then? In contrast to the elderly strangers, here, you are unable to jump quickly to any conclusion. The first thing you feel is probably confusion– could my own brother really have stolen from me? You may entertain the thought– if we were robbed by a third party, did something bad happen to him? If it becomes apparent somehow that it really was your own brother who did the deed, then likely, you do not call the police at all. Instead of anger at this point, you feel concern– why has he done it? Is he on drugs? Has he lost his job and felt too ashamed to tell me? Indeed, your mind is not on getting your wallet back, but on locating your brother and ensuring he is safe. Because a relationship of trust underpins this example, and does not underpin the first example, your experiences of having your wallet stolen in these two cases are wholly distinct. Whatever our ways of speaking, one way we can get clear on whether or not a case of reliance was underpinned by trust, is to see what happens when that reliance is let down. Do we jump quickly to conclusions of guilt, or do we entertain other possibilities? Do we immediately seek justice, or do we first and foremost seek understanding? Do we allow our anger to guide our actions, or are our immediate angry reactions quickly overcome by a deeper concern? Again, I am not drawing sharp distinction with my examples. Individuals may indeed trust complete strangers they meet on trains wholly and immediately, experiencing – if those strangers let them down – all that they would experience if a close old friend or family member let them down in the exact same way. Yet I think this is probably rare. At least in adults who have cultivated mature and healthy trusting relationships, there will be a distinction, even if we do not mark it adequately in the English language.
I.4.4 Trusting and monitoring
Traditionally, philosophers have worried that monitoring destroys trust. If I am monitoring or “checking up” on you, then I do not really trust you. How can you say that trust is compatible with such monitoring? Surely monitoring is the very stuff of distrust!
Good. This objection gets us to the heart of things. To refresh, thinking of trust as incompatible with monitoring is what has led philosophers to think of The Problem of
I.4 Trust 107
Trust as a problem of epistemic and practical rationality, as discussed already in I.1.3. In demonstration of the prevalence of this line of thinking, observe the following remarks from prominent philosophers on trust:
substantial trust may be characterized by two related features: (1) it involves making or maintaining judgements about others, or about what our behaviour should be towards them, that go beyond what the evidence supports; and (2) it renounces the very process of weighing whatever evidence there is in a cool, disengaged, and purportedly objective way. The problem, then, is to explain how trust of this sort can be rational. (McGeer 2008: 240)
since trust inherently involves risk, any attempt to eliminate the risk through rational reflection could eliminate the trust at the same time… trust tends to give us blinkered vision: it makes us resistant to evidence that may contradict our optimism about the trustee. (McLeod 2015, Section 2)
[trust] shields from view a whole range of interpretations about the motives of another and restricts the inferences we will make about the likely actions of another. Trusting, thus opens one up to harm because it gives rise to selective interpretation, which means that one can be fooled, that the truth may lie, as it were, outside one’s gaze. (Jones 1996: 12)
I believe [that my friend] is innocent. I do not, however, come to believe she is innocent, despite the evidence, by weighing and balancing present evidence against her past record…. What others regard as evidence against her isn’t considered by me as evidence at all. (Baker 1987: 3)
As this dissertation has shown (subsection I.2), philosophers on trust have been led by their adoption of three-place trust to the conviction that good trusters overlook evidence, selectively interpret evidence, go beyond the evidence, renounce the very process of weighing the evidence objectively, and are resistant to evidence. According to the dominant view, good trusters have blinkered vision, do not engage in rational reflection and put themselves at risk both practically and epistemically. Yet we saw in subsection I.4.1 with Dorothea that this is not an accurate characterisation of trust. Good trusters do not need to have blinkered vision; indeed, it was the distrusters in this example, Sir James and Farebrother, who turned away from the evidence. These philosophers are describing a feature of wishful thinking, blind
I.4 Trust 108 faith. Faith goes “beyond what is ordinarily reasonable, in the sense that it involves accepting what cannot be established as true through the proper exercise of our naturally endowed human cognitive faculties” (Bishop 2016: Section 4). Trust though, requires no such thing. Though much work has gone in to distinguishing trust and reliance, this common way of understanding the relationship between trust and evidence shows that philosophers need to start paying more attention to the distinction between trust and faith. When we make the shift to two-place trust, and so see clearly that trust is not a kind of reliance but is instead a complex experience one person has of another, it becomes easier to keep sight of the fact that it is not the monitoring per se that affects our status as trusters, but the complex experience behind our monitoring. For instance, if the truster is checking up on the trustee just to see how they are getting along, because they have thought of some relevant information that might help the trustee succeed, or because some new information has come to light that the task is far more difficult than first thought, then they are not compromising their trust. So long as they continue to experience the trustee as worthy, trusters can monitor their trustees, check up on them, devise contingency plans, seek reassurance, or organise assistance for them. This kind of ‘benevolent monitoring’ is important for maintaining our practical and epistemic rationality – not to mention not dying when rock climbing – and it can be done without affecting our trust. If trust is a complex way in which we experience another person, then trust can only be undermined by a significant change in that complex experience. I say significant, because if we think about the people whom we trust the most, we may notice that these relationships historically involve many little doubts, moments of resentment, moods of insecurity, worries about unreliability, flashes of duplicity. Even the best trusters do not maintain a perfectly serene optimism all the time, because even the best trustees are not perfect. Struggling with such provocations when they arise is not at odds with trusting, it is part of the experience of trusting. This is another way that trust is like love. False love is like false trust in that it is blind to its object’s failures and limitations. We come to know trust and love’s genuine forms by the way in which they, unlike their false forms, somehow miraculously thrive in the face of grim truth. A significant change in
I.4 Trust 109 trust, like falling out of love, is a complex manifold experience; a deep hopelessness and pessimism, persistent negative thoughts about the trustee, entrenched beliefs that the trustee is compelled by ill-will or malice, strong feelings of betrayal and resentment, gripping desires to end the relationship, and so on. As trust’s opposite, distrust is a complex experience as well – it is a far more serious condition than a touch of cynicism. Though it has been argued that trust is compatible with monitoring, this dissertation is not proposing that monitoring never indicates distrust. Like the withdrawal of affection that often accompanies fading love, it is quite a reliable sign. But just as reliance does not constitute trust, monitoring does not constitute distrust. One way we know this is because we can argue back and forth about it, all the while agreeing on the facts. Adolescents complain “You never trust me, you’re always checking up on where I am and what I’m doing!”. To this, the parent replies “I do trust you, it’s other people I don’t trust, and I need to make sure you’re safe”. It may be that the adolescent thinks the monitoring is occurring because they are not trusted, the parent knows it is because others are not to be trusted. If monitoring constituted distrust – if behaviour was all there was to it – then such disagreements would be puerile and easily settled by the evidence. What we find though, is that such disagreements are not puerile, and they are not easily settled by the evidence. In time honoured fashion, with the same evidence planted firmly between them, adolescents and their parents argue back and forth for years like this without getting anywhere, generation after generation. What settles such arguments is not more evidence, but experience. Until we ourselves have been charged with the responsibility of caring for someone who is not yet fully capable of caring for themselves, we cannot appreciate this feature of trust; it is possible to trust someone and need to know where they are and who they are with at all times. One way to understand the confusion about trust and monitoring, trust and rationality, comes from acknowledgment that trust involves a kind of optimism, coupled with the misconception that optimism itself puts us into conflict with the facts. As Michaelis Michael and Peter Caldwell (2004) argued convincingly in “The Consolations of Optimism”, optimism is an attitude, not a set of beliefs. Being optimistic is not a matter of how one takes reality to be; optimists and pessimists do not disagree on the facts:
I.4 Trust 110
The wishful thinker really does think that things are other than they are. Wishful thinking is by its very nature not a rational cognitive strategy, whereas optimism involves no such irrationality. Wishful thinking is a belief-like attitude, and it can lead at least temporarily to a kind of cheerfulness. But it can also be said that it is irrational since it involves founding beliefs on inadequate evidence and ignoring contrary evidence. Only the optimist is happy while seeing things as they really are. (Michael and Caldwell 2004)
In the context of the current discussion, we could say that distrust involves an attitude of pessimism, trust involves an attitude of optimism, and truster and distruster do not necessarily disagree on the facts about their trustees. Trusters can monitor their trustees, review evidence objectively; the optimism component of trust does not give us ‘blinkered vision’. In my experience at least, the pessimism that has led to my own failures at giving people the benefit of the doubt has more often put me in conflict with truth than my optimism has. I suspect this is not because more people are good than bad and so optimism is the best ‘stab in the dark’ strategy. Instead, I agree with those who have pointed out that people are usually both good and bad, but that treating them as though they are good gives them more reason, more opportunities to be so. I discuss this more in II.4.4, but for now I will say that trust and trustworthiness are features not of moments, but of relationships. How others respond to our momentary failures of trustworthiness can determine what happens to it next. That is why Bishop Myriel of Digne is the real hero of Lés Misérables.
I.4.5 Self-trust
If trust is a complex private experience, then what of the notion of self-trust? Can we have a complex private experience of ourselves?
Self-trust, as it is currently conceived of in philosophy, is a cognitive notion that has been put forward as playing a significant role in our practical and epistemic rationality. For example, Keith Lehrer (1997) talks about how assessing the truth of our beliefs necessitates that we be able to ‘trust ourselves’ to do this work; Richard Foley (2005) argues that knowledge requires self-trust; and Trudy Govier (1993) and Carolyn McLeod (2002) argue that agency requires self-trust, since we cannot act autonomously unless we trust ourselves in this regard.
I.4 Trust 111
In this dissertation I have not made any argument against the claims that: 1) to plan effectively, it helps if one believes one’s own beliefs; and, 2) to make sound decisions, one must rely on one’s own reasoning. Indeed, we might start to question a person’s agency when they fail in these regards. What I have done, is argue that the situation described by these philosophers is not one in which trust plays any vital – let alone necessary – role. When Richard Foley talks about the necessity of trust as a “leap of intellectual faith” (2005: 18) and “the need for which cannot be eliminated by further inquiry” (2005: 20) for example, I understand him to mean that to have knowledge we need to be able to rely on our own belief-forming mechanisms. This is in line with his comment that “living a normal life requires trusting that one’s opinions are generally correct, and the faculties and practices that give rise to them generally reliable” (2005: 3). What is ubiquitous in our intellectual and practical lives is that we might be wrong, might have made a mistake. This situation calls for us to press on in the face of uncertainty, but it does not necessitate trust. As Part I has shown, many other things can play this role. We often press on in the face of uncertainty in the absence of trust, instead spurred on by hope, faith, acceptance, coercion, or any number of other reasons and compulsions. Trust does not ubiquitously manifest in every instance of decision, belief, and action, but manifests only in our close interpersonal relationships. Trust is not there every time we sit on a chair and expect it not to collapse, believe our colleague’s scientific findings, act according to our prior reasoning, follow a stranger’s directions, or hand our teenager the car keys. On two-place trust, we reframe the important epistemic concerns that Lehrer, Govier, Foley and others are getting at as follows: when I believe that my belief is true, I have a second-order belief, not a trust in my belief, and, when I use my prior reasoning to make a decision, I rely on my prior reasoning, I do not trust my prior reasoning. If the two-place picture is correct, and trust is a complex experience we have of the trustee, then self-trust is the ability to experience oneself as trustworthy: to experience oneself in the way one would experience another person that one trusted, with all the history and complexity that this involves. It is beyond the scope of this dissertation to address this proposal in any detail, but we can talk a little about what two-place self-
I.4 Trust 112 trust may require. First, we would need to have a self, and then we would need to be able to experience that self. This feels a little incoherent: do we have a second self that experiences a first self? Along with Iris Murdoch and Galen Strawson, I struggle to find even one self, let alone two. Perhaps just as other people gain a certain realism in our experience of them, so we too gain a certain realism in our experience of ourselves. Thought about this way, maybe self-trust just is what it is to ‘have’ a self. If two-place self-trust can be made to be coherent, then a person with self-trust would come to experience themselves the way that Dorothea experiences Lydgate: with gratitude, imagination, hope, optimism, beliefs about trustworthiness, an absence of doubt and suspicion, and a tendency to feel self-betrayal, self-resentment, self- forgiveness, etc. I leave it up to the reader to examine their own experience and see if they find any truth to these proposals. If there is a problem in principle with two-place self-trust, then it is that there must be a certain opacity to trust. This is a significant theme of Part II so I shall not go into too much detail here. For now, I will say that although some knowledge is not incompatible with trust, trust requires us to experience the trustee in a way that might not reflect reality. Hope, imagination, or giving the benefit of the doubt, are parts of trust that require us not to have full knowledge – extending both backward and forward in time – of the object of our trust. Omniscient beings cannot trust. This does introduce a potential incoherency – in order for two-place self-trust to be possible, it must be possible that we sometimes do not know what we think, what we want, why we are doing what we are doing, who we may become in the future. If I examine my own experience, this, at least, is a proposal I personally can get on board with, and, it becomes even more salient with regard to two-place self-distrust, which is where I shall finish this discussion. On two-place trust, self-distrust does not manifest when our reasoning or memory fails us. Self-distrust is not a failure to put thoughts together properly or forget what we had previously decided (though these conditions may confound self-trust, even so, they are problems of self-reliance, self-belief). Self-distrust is a far more complex experience: it is when we come to doubt that we have our own best interests at heart, suspect we might be out to harm ourselves, cannot maintain optimism about ourselves,
I.4 Trust 113 cannot relax in our own company, imagine no good future for ourselves, give up on ourselves, and – in the most troubled case – desire to end our relationship with ourselves. Such a complicated way of experiencing one’s self would be highly distressing, highly destabilising, and does seem to describe very well the difficult-to- manage experience of some individuals who we today describe as mentally ill or addicted. If self-distrust eludes us in mental illness or addiction, then, regaining it would be part of any robust and enduring recovery. When looking at the factors which precipitate and sustain recovery from severe mental illness, this is precisely what Davidson and Strauss found. In their report, Sense of Self in Recovery from Severe Mental Illness, Davidson and Strauss (1992) identify a “sense of self” as essential to recovery, which they take to involve a number of things including: imagination, belief in the face of counterevidence, the ability to experience one’s self in a way that might not reflect reality (all the while keeping track of any evidence), an accurate “personal inventory” of current capacities, an awareness of potential, and critically, hope. It is notable that their account of a sense of self that gets shattered in mental illness and repaired in recovery is strikingly close to my account of two-place self-trust. Moreover, what they discovered when they looked at people who had been hospitalised for severe mental illness and recovered, was that the process of recovery – of regaining a sense of self (or self-trust) – is far from solitary. At critical points it is either undermined or bolstered by others:
It is important to recognise the influence of others’ appraisals – whether spoken or implicitly conveyed through actions and attitudes – on the [mentally ill] person’s own process of taking stock of the self. A highly critical and over-involved family milieu can be seen as interfering with the development of an active sense of self by fostering negative appraisals and undermining a [mentally ill] person’s efforts... On the other hand, a supportive social network can be seen as facilitating the development of a positive sense of self by providing one with a sense of self-worth through belonging and nurturing self- initiated action. (Davidson and Strauss 1992: 142-43)
I.4 Trust 114
In other words, the self-trust that is essential to recovery can be given us, and taken from us, by other people. If we find ourselves deprived of such a trusting social network, then we are doubly at risk. In such cases, the mechanisms we employ to bootstrap trust in others may be useful at bootstrapping self-trust. Indeed, cognitive behavioural therapy, which is essentially a matter of getting an individual to challenge and change their negative beliefs so that a more complete attitudinal and effective change might follow, is likely one such ‘bootstrapping mechanism’ already in widespread clinical use (Westbrook et al. 2011). In light of the current discussion, some other things we might try are if we find ourselves in such a predicament are: giving ourselves the benefit of the doubt, foregoing judgement on ourselves, interpreting evidence against ourselves in the most favourable light, looking harder within ourselves for reasons to believe we are worthy of our own trust, and giving ourselves many opportunities to prove to ourselves that we are. That said, the prospects of successful self-bootstrapping of self-trust may be grim, so long as one remains in the throes of the distressing disturbance itself. As such, to underscore the importance of the interpersonal element even when it comes to self-trust, I shall end by suggesting that to recover, in place of the agonising, time-consuming, and expensive work of self-examination, it may be more advantageous to try to establish a new, more favourable social network. This advice is in line with the significant outcome of a large cross sectional survey into global patterns of mental health discrimination, which found that “the main source of reported discrimination [of those diagnosed with a mental illness] is from family members, which is also the source of the most reported support” (Lasalvia et al. 2013: 59). In short, if you’ve got a good family, good; if not, jump ship. The difference could be between recovery – i.e. recovering a sense of self/self-trust – and continued psychological distress. As a recently recovered friend of mine with a highly stigmatised diagnosis and a highly critical family said to me “from now on, friends and lovers… forget the others!”
I.4.6 Deciding to trust
If trust is a complex experience, involving all manner of things, then how does two- place trust answer the question about the voluntariness of trust?
I.4 Trust 115
The argument over the voluntariness of trust reveals best the underlying ontological disagreements about trust. When trust is conceived of as an emotion or a belief in the trustworthiness of the trustee, then we cannot decide to do it. When trust is conceived of as an act of reliance or cooperation, then, we can. When trust is understood as a two- place complex experience involving all sorts of phenomena, the answer – unsurprisingly – is more complex. Complex experiences are the sorts of things which, by their nature, contain both elements that are within our control and elements that are not. I can decide to enrol as a PhD student, but I cannot decide whether or not I will get to graduate and so wear the floppy hat. I can decide to holiday with Frank and Marsha, but I cannot decide whether I will find their company exhilarating or nauseating. When we make the shift from trust as an epistemic stance or leap one takes with regard to some specific situation, to trust as an ongoing interpersonal tête-à-tête, then an imprecise answer to the question of the voluntariness of trust is much better tolerated. Again, it is beyond the scope of this dissertation to go into detail as to the various imprecise ways in which trust is and is not voluntary in a two-place account. However, a few details are intriguing. A truster who feels they experience too little trust might resolve to rely on their trustee more or to take their word as true despite any reservations they may have, thus providing their trustee with a chance to prove themselves. This ‘fake it till you make it’ technique can be highly effective, and it does involve a decision, as a number of philosophers have noticed (Holton 1994; Horsburgh 1960: 346; Jones 1996: 6; Nickel 2007: 317). Yet we must be clear about what is happening here. When a person undertakes such a bootstrapping of their trust, they are not ‘trusting though they do not trust in order to trust’ (for that hardly makes sense). What they are doing is acting as though they trust in order to produce genuine trust. Therapeutic trust, as it is called in the literature, is not trust. Although to believe that the trustee is honest or reliable is not what it is to trust, such a belief can certainly help get things going. Pretending to experience a person with trust can sometimes produce trust, precisely because trust manifests in complexity. Once we believe the trustee is reliable and honest, we may feel gratitude towards them, develop an optimistic attitude towards them, then over time develop a desire to defend them and to give them the benefit of the doubt, and eventually develop the potential to feel
I.4 Trust 116 betrayed by them, etc. Pretending is not all bad. It can have a cascading effect; pretending is how all relationships get going. None of us trust at the outset. Instead we follow social norms: we are credulous and polite, we rely, we cooperate, and we see what comes of it. In opposition to this, Victoria McGeer considers the example of the parents who allow their teenager to borrow the family car to be an example of trust, even when it involves:
no comfortable feelings of optimism or confidence but is instead characterised by explicitly imposing on them [the teenager] certain normative expectations (McGeer 2008: 241)
She rejects the proposal that therapeutic trust is not really trust but merely acting as if we trusted, arguing that it is trust because:
it is an attitude that both empowers us in our trust – making it possible for us to think and act in trustful ways – and empowers them through our trust, by stimulating their agential capacities to think and act in trustful ways (McGeer 2008: 242)
McGeer is right insofar as there is a sense in which part of experiencing trust is experiencing the hopeful beginnings of trust, just as part of experiencing a PhD is hoping you will be accepted into the programme. If you end up with trust, or with a PhD, then the hopeful beginnings will end up being part of the story. But if things do not work out and you cannot get that initial burst of hope to spur on all the other complicated set of things that constitute trust, or a PhD, then you never trusted, or completed a PhD. Dashed hopes are one thing, but broken trust is quite another. The parents who have “no comfortable feelings of confidence or optimism” but are instead merely “imposing normative constraints” on their child, do not trust their child. Normative constraints can be imposed on all sorts of people– those we dislike, hate, fear, and most commonly, on those we have only just met and have no specific feelings about at all. Enemies and strangers are people we can believe, and choose to rely upon, but we must not equate this ‘trusting behaviour’ with trust. Sometimes we accept a person’s testimony because we believe it is true, and sometimes because that is the polite thing to do, normatively speaking. At other times we accept a person’s testimony
I.4 Trust 117 because we trust them. The source and force of many outwardly indistinguishable “trusting behaviours” vary in all sorts of philosophically intriguing ways. We can certainly decide to try for trust, but we cannot guarantee we will experience any. This is partly because of the complex nature of trust, but it is also partly because of the reciprocal nature of trust. Some trustees make it easy for us to lean in and experience greater trust, and others make it difficult. What is more, each one of our trusting relationships affects the others: a bad sod can rot the whole lot. This goes the other way as well; one good strong trusting relationship can act like the giant old tree that creates a canopy of shade to protect small seedings as they mature. A healthy deep- rooted relationship of trust with even just one person generates in us the ability to produce more. At risk of stretching the analogy too far: ancient giant trees, when they fall, flatten and destroy everything below. We need to talk more in philosophy about deciding to trust, but discussion should not be around which experts should we trust, how do we decide which network to trust in the current fake-news climate. These are important discussions, but they are not about trust. We need to talk about the importance of making good decisions about who we let into our personal private lives in the first place, before it even comes to trust. We need to talk about the decision between trying to forgive and letting go. We need to talk about how our actions impact on our trusting relationships. We need to talk about those times when we need to decide between doing the right thing and doing what is best for our trustee, doing the rational thing and doing what’s best for our trustee. These are not moral questions or epistemological questions. They are questions about how best to get along with one another.
I Summary of Part I, On Trust 118
Summary of Part I, On Trust
Part I began in Section I.1 by reviewing the current literature on trust, outlining the major issues philosophers have seen in relation to trust, and establishing that they arise because the dominant view in philosophy is that trust is a three-place relation. Sections I.2 and I.3 then demonstrated that trust is not a three-place relation, which satisfied the primary aim of this research project. In Section I.3, I put forward an alternative picture of three-place trust: three-place trust is reliance-because-trust. When we rely on a person because we trust them or take their word as part of the trust that we have in them, these are kinds of reliance, not kinds of trust. This research project is able to account for the way we commonly use the word ‘trust’ in the English language, and the intuition that such usage does in fact on occasion mention trust. Section I.4 presented an original discussion about trust that emphasised the private nature of trust as a complex experience one person (the truster) has of another (the trustee)– the object of trust. Only when we identify the object of trust as the trustee can we account for what is inimitable and distinctive about trust, and so overcome the problems with the three-placed account identified in Section I.2. This two-place shift in thinking about trust was able to incorporate much of what philosophers have identified before as distinctive features of trust, and so, unifies many important, and previously incompatible, insights about trust under one account. Section I.4 addressed some outcomes of two-place trust as they relate to the major points of division in the literature revealed in Section I.1. When we think of trust as a two-place relation, an experience one person has of another, it is much easier to maintain the conceptual separation between trust and any other three-place relations. This has been a long-standing issue in the philosophy on trust, with a great deal of effort spent on conceptual distinctions between trust and reliance, trust and belief, and trust and cooperation. Acknowledging trust’s status as a two-place relation also has benefits for our understanding of the rationality of trust, which has been of considerable concern to philosophers as outlined in subsection I.1.3. In any situation where there does happen to be a causal connection between trusting a person and relying on that person, a two-place account allows us to keep sight of the fact that there are actually two distinct and independent things going on that we should
I Summary of Part I, On Trust 119 be awake to: our practical or epistemic dependence on a person, and our trusting relationship with that person. If I trust you, and am currently relying on you to do something, then I should be awake to the practical and epistemic risks and benefits of my reliance that pertain directly to your success or failure, as well as any evidence that suggests how things might go in this regard, and I absolutely should update my beliefs and actions in light of any new information I receive regarding your potential success or potential failure. To not do so – to look the other way when relevant evidence is presented – is irrational, as many have pointed out, and could be dangerous to both of us. When trust is conceived of as a complex ongoing experience, trusters can monitor their trustees in the moment without destroying trust. The tension between trust and rationality thus dissolves.
On Trustworthiness
There is truth and then again there is truth. For all that the world is full of people who go around believing they've got you or your neighbor figured out, there really is no bottom to what is not known. The truth about us is endless. As are the lies. –Philip Roth (2000)
Given the dominance of three-place trust, it should not be surprising that, currently, trustworthiness is most often conceived of as a three-place relation as well: as a matter of the trustee doing what is entrusted (keeping a secret, looking after a possession, returning a borrowed item etc.) for the truster. On the strong version of this view, a person becomes trustworthy just when they have successfully done what they have been entrusted to do. On a weaker version, a person is trustworthy prior to their successfully doing what is entrusted, and even in some cases if they fail, so long as they remain both competent and committed to doing what they have been entrusted to do. Yet while three-place trustworthiness is again dominant, it is more common to find one-place trustworthiness in the literature: trustworthiness as a stable virtue or character trait, that takes no specific object. However, on close examination, most apparently one- place accounts are in fact hybrid one/three-place, in that they take the personality of the trustee and their actions into account when deciding a person’s worthiness of trust. A small number of philosophers have recognised the major problem with three- place trustworthiness. Acting in a ‘trustworthy manner’ is not sufficient, since one could be merely pretending to be trustworthy, doing what one has been entrusted to do this
120 II Overview of Part II, On Trustworthiness 121 time in order to gain trust for what are ultimately underhanded reasons– the ‘Mr Fox and Jemima Puddle-Duck’ effect. Just as in the literature on trust, the addition of various motivating factors such as goodwill, selflessness, conscientiousness, or benevolence, creating a corresponding ‘three-place trustworthiness plus’ is quite common. The primary aim of Part II of this research project is to argue against the various one and three-place views of trustworthiness in favour once again of a two-place view, in which trustworthiness, like trust, is a complex way of experiencing a particular individual– the trustee. Trustworthiness is not an immutable feature of our personality but a relational quality. It is not an action, and it is not an action guided by any single motivation such as conscientiousness or loyalty. Trustworthiness does not exist until there are at least two of us: we find it alongside trust (and love) in what two individuals create in their complex, historical, reciprocal experiences of one another. The reason this dissertation gives equal weight to establishing the two-place nature of trustworthiness is because trust and trustworthiness are two elements of a single phenomenon; not only do they share important similarities regarding their logical structure and ontological commitments, they are responses to one another. An account of the nature of trust is half-finished without an accompanying account of the nature of trustworthiness.
Overview of Part II, On Trustworthiness
Part II has four substantial sections, three of which mirror those of Part I. The difference is that while Part I focused more on the epistemological implications of three-place trust, Part II focusses more on the ethical implications of one/three-place trustworthiness. Section II.1 Trustworthiness: in the Philosophical Literature reviews the current philosophical literature on trustworthiness. The picture that emerges is again one in which three-place accounts are dominant. Although in the case of trustworthiness, various one-place conceptions are also commonly adopted. Section II.1 Dominant Views on Trustworthiness: Problems explores the limitations of one-place trustworthiness and three-place trustworthiness. Section II.2 Rich Trustworthiness or Poor Trustworthiness? addresses the considerable ethical implications of the influential
II Overview of Part II, On Trustworthiness 122 view of Karen Jones, who has written more on the nature of trustworthiness than any other modern philosopher. Then, just as in I.3 Trust, Section II.3 Trustworthiness draws again on an example from fiction to highlight some thought-provoking truths about trustworthiness. This time, I use a lighter example: the Ricky Gervais film The Invention of Lying to explore the philosophical consequences of a world in which people cannot deceive one another.
II.1 Trustworthiness: in the Philosophical Literature 123
II.1 Trustworthiness: in the Philosophical Literature
Just as in the literature on trust, there is currently no clear consensus among philosophers about the nature of trustworthiness. That said, also as in the literature on trust, most philosophers take trustworthiness to be a three-place relation; a matter of the trustee doing what the truster has entrusted them to do:
Three-place trustworthiness: (B does what A entrusted B to do)
Yet a significant number of philosophers instead write as though trustworthiness were one-place; an immutable personality trait, or a cultivatable character trait, that may or may not be virtuous:
One-place trustworthiness: (B is trustworthy)
This dissertation understands trustworthiness to be a complex experience the truster has of the trustee:
Two-place trustworthiness: (B is trustworthy to A)
While there is some preliminary movement towards two-place trust in the literature as acknowledged in Section I.1, no equivalent move towards two-place trustworthiness is evident in the current philosophical literature. In contrast to the literature on trust, there is far less literature available on trustworthiness, with very few philosophers examining the notion in any detail. Indeed, a recently published collection of essays by philosophers working in this area, edited by Paul Faulkner and Thomas Simpson, contains ‘trust’ in the title of twelve of fifteen essays and yet only one that mentions trustworthiness in the title, i.e. “Trustworthy Groups and Organisations” by Kathrine Hawley (2017). Of the hundreds of works cited in this research project, only a handful contain any detailed discussion of the nature of trustworthiness. Reflecting this, the Stanford Encyclopaedia contains an entry on trust dating back to 2006; yet more than a decade later, it still does not contain an entry on trustworthiness. Furthermore, each year a number of conferences and workshops on the philosophy of trust take place at universities around the world, such as: the Trust &
II.1 Trustworthiness: in the Philosophical Literature 124
Belief Workshop, Cambridge 2016; the conference ‘Trust’, at the University of Manchester in 2014; the ACU Workshop on Trust, ACU 2017; and the Workshop and Conference on Social Trust, BGSU 2018. Yet I was unable to find any during my candidature (2014‒2019) devoted to the philosophy of trustworthiness. Philosophers clearly find the notion of trust more stimulating. As demonstrated in Part I, this is likely due to the tendency to view trust as posing problems for our epistemic and practical rationality. Given the lack of obvious literature, this section will begin in subsection II.1.1 by demonstrating how the symmetrical nature of trust and trustworthiness means one can ‘find’ ideas about trustworthiness embedded in the trust literature, thus establishing the very possibility of the current enterprise. Subsection II.1.2 presents literature that appears, or claims to be, one-place. Subsection II.1.3 presents literature on three-place trustworthiness. Subsection II.1.4 presents literature that aims to fix three-place trustworthiness with the addition of various motivating factors, the ‘three-place trustworthiness plus’ accounts. Subsection II.1.5 provides a summary of the philosophical literature on trustworthiness.
II.1.1 Symmetricity of trust and trustworthiness
The project of reviewing the literature on trustworthiness when there is apparently so little of it may appear futile. Yet Karen Jones suggests that we may succeed in understanding any particular philosopher’s position on trustworthiness by “glimpsing backward” from their stance on trust, thus making it possible to discover literature on trustworthiness already embedded in the trust literature:
Philosophers have written a lot about trust, but we have been surprisingly silent about trustworthiness. There are scattered remarks about it in discussions of trust, so the silence is not complete, but the problem of understanding trust and trustworthiness has been pursued largely from the trust side. Despite this comparative silence, we can discern what philosophers must have been thinking about trustworthiness from what they have said about trust, since accounts of trust typically contain reflections of an implicit account of trustworthiness, glimpsed backward like writing viewed in a mirror. (2012: 61)
II.1 Trustworthiness: in the Philosophical Literature 125
Elsewhere, Jones advises that reversing the focus of the approach may in fact prove more fruitful:
Part of the point of switching to approach the problem of understanding trust and trustworthiness from the trustworthiness end is that the normative role of these concepts comes more clearly into view when approached from this direction. (2012: 61)
Jones would be right about the possibility of glimpsing trustworthiness from trust only if it were true that most philosophers write as if they share important symmetries, a prospect which Russell Hardin disagrees with. Hardin says that philosophers often mistake trustworthiness for trust (1996: 28), and that trust and trustworthiness are not reflective of one another, but rather, are deeply dissimilar:
… trust and trustworthiness are not analogous or symmetrical…, because one can be disposed to trustworthiness without any risk… A relationship cannot make you worse off if you are merely trustworthy in it. It can, however, make you substantially worse off if you are trusting in it… (2002: 37)
On a two-place view of trust and trustworthiness as put forward here, Hardin’s assessment of differing risk profiles is not correct (see I.3.5). There have always been risks associated with trustworthiness, and these risks are no less significant than the risks associated with trusting. Indeed, the risks are of a similar type. Namely, pessimism, bitterness, resentment, insecurity, self-loathing, the loss of relationships, etc., that can occur when, for example, trustworthy trustees are not afforded the trust that they deserve. Moreover, even if there were differing risk profiles to trust and trustworthiness, this alone cannot suffice to show that they are not symmetrical in other important respects. Either way, for the purposes of reviewing the literature, it does not matter whether trust and trustworthiness are in fact symmetrical. Since most philosophers do adopt mirroring accounts of trust and trustworthiness, it is possible to ‘glimpse’ many philosophers’ stance on either trust or trustworthiness from their corresponding account, as Jones says. In my own exploration of the literature, I found numerous instances of the symmetry Jones refers to. For example, Nickel says that it is not possible for us to trust
II.1 Trustworthiness: in the Philosophical Literature 126 a person to do something, if we think it is optional or supererogatory, and so, trusting involves an obligation ascription to the trustee. Once we understand his position on trust, it becomes clear that being trustworthy for Nickel is performing a task that we have an obligation to perform, ascribed to us by the ‘act’ of trust (2007). Also, because Baier provides the following formula for trust– (A trusts B with valued thing C), she takes trustworthiness to be a matter of (B looking after valued thing C for A) (1986: 236). Similarly, Trudy Govier takes self-trust to be a necessary condition for both autonomy and self-respect, and so self-trustworthiness on her account is necessary for autonomy and self-respect as well. This becomes clear when she says, “to trust a friend is to believe her motivations (toward oneself) emerge from affection, care, and concern and not from dislike, ambition, or egoism” (1993: 104), and so the “motives of trustworthiness” for Govier are – unsurprisingly – affection, care, and concern. Explaining the symmetry in his view, Stephen Wright says:
My account of what it is to be trustworthy follows from what it is to trust someone because when I accounted for trust, I was careful to distinguish it from examples of expectation or reliance, arguing that trust had a higher value than either of these. My account of trustworthiness also maintains the distinction between cases of trust and reliance, distinguishing between being reliable or corresponding with expectations and being trustworthy. The key link between my account of trust and my account of trustworthiness is the trustee’s ability to freely choose whether or not to disregard the trust that has been invested in them. (2010: 622)
Jones herself, though she has more recently expressed doubts about the symmetry of trust and trustworthiness (2017: 103), takes trust to be a three-place relation and adopts a corresponding three-place account of trustworthiness (see II.3). And, the two-place version of trustworthiness presented in Part II mirrors that of trust presented in Part I. Such symmetry, it is suggested here, reflects a widespread appreciation that trust, and trustworthiness are partial concepts in need of their other halves. Indeed, Hardin’s own Hobbesian accounts of trust and trustworthiness mirror one another as well. Hardin says, “I trust you because I think it is in your interests to attend to my interests in the relevant manner” (2002: 4), and so, predictably, trustworthiness for him is a matter of attending to the trustee’s interests because it is in our interests to
II.1 Trustworthiness: in the Philosophical Literature 127 do so. We know this because he argues that trustworthiness can be augmented or diminished by things like institutional backing and formal contracts– things that help to eliminate the risk of cooperation. Such assurances, he says, make us more rather than less trustworthy (1996: 31-32). As Stephen Wright explains:
essentially, according to Hardin’s theory, as the opportunity for the trustee to act against the trustor’s wishes decreases, the trustworthiness of the trustee increases. Hardin’s theory is underpinned by a belief that if people are trustworthy then they will do what we want them to. (2010: 621)
The symmetry that exists between the various accounts of trust and trustworthiness is the product of the constraints of logic interacting with the ontological commitments – as well as the question of interest – of each philosopher. If a philosopher is concerned with an epistemological ‘problem of trust in testimony’ for example, then trustworthiness will be a matter of speaking truthfully. If a philosopher is concerned with the ethical problems introduced by trust, then trustworthiness will be seen as a moral imperative. If, like Hardin, the philosopher sees the world as a nasty and brutish place where self- interested actors play games with one another as they play out their own mercifully short lives, then trust and trustworthiness will be calculated plays aimed at maximising self-interest. This symmetry creates a convenient situation for anyone trying to understand a philosopher’s stance on either trust or trustworthiness, but it also hints at something more tangible; unless we are all totally wide of the mark, it is likely that this prevalent symmetry in accounts of trust and trustworthiness does reflect some genuine symmetry in the world.
II.1.2 One-place trustworthiness
Though a small number of philosophers endorse the idea that trust is an emotion, one- place trust is not a common view to be found in the trust literature, as we saw in subsection I.1.1. A few philosophers do acknowledge that there is a sense in which trust can be one-place – being an inherently trusting person – though most tend to move quickly past this sense to a three-place sense, taking the latter to be the philosophically interesting notion: for example, Faulkner (2017). Discussion of one-place trustworthiness, however, is much livelier. Still, identifying one-place views is not
II.1 Trustworthiness: in the Philosophical Literature 128 straightforward. In the literature, one-place trust and trustworthiness are also sometimes called ‘thick’ trust and trustworthiness, in contrast to ‘thin’ or three-place trust and trustworthiness. Jones and McLeod have adopted these terms, but reference to ‘thick’ and ‘thin’ trust/worthiness is most prevalent in other disciplines, such as political and social science. To make matters more confusing, there are other terms in play as well. Eric Uslaner, for example, refers to this kind of one-place trust as ‘moralistic (generalized) trust’ (2003). Following Holton and Faulkner, this dissertation uses the terms one-place trust and one-place trustworthiness for two reasons. First, these terms capture best the essential and most basic difference between the various views, and second, unlike ‘thick’ and ‘thin’, they allow for a natural intermediate concept– two- place trustworthiness, which is the one endorsed here. As should be familiar from I.1.1, on one-place trust and trustworthiness there is no object of trust and trustworthiness: no particular person or action that our trustworthiness attaches to. Rather, trustworthiness is a property of an individual:
(B is trustworthy)
This bare construction leaves many questions unanswered: is trustworthiness an emotion, character trait or virtue, good or bad, desirable or undesirable, stable or momentary, etc. Philosophers adopting one-place accounts usually take trustworthiness to be a virtue and so morally good, and also relatively stable. For example, Collin O’Neil draws a distinction between moralised-trust and non-moralised trust, where moralised-trust puts the trustee under a moral obligation to do what they have been entrusted to do (2017: 70-88). He says that this kind of trust can make a moral difference because it can “aggravate a prior wronging by making its consequences worse for the truster, thus making the wronging more serious” (2017: 80). Both Jones (2004) and Walker (2006) have also provided accounts of how trust can make us more vulnerable to moral harm through failures of trustworthiness, though Jones has since argued that the “norms of trust and trustworthiness” have the potential to sit uneasily with moral norms. Jones now takes it to be the case that there is a distinct set of norms for trust and trustworthiness, and these answer to the “pressing interest we have in their being people whose agency we can directly recruit to enhance the effectiveness of our
II.1 Trustworthiness: in the Philosophical Literature 129 own” (2017: 97). Few philosophers have argued explicitly, as Jones does, that trustworthiness is not a moral notion, another recent exception coming from Edward Hinchman, who says that the obligation at the heart of trustworthiness is rational, and not moral (2017). Instead most philosophers either explicitly endorse or make comments to suggest that they support what can be considered a ‘moralised version’ of trustworthiness, including: Baier (1986)(1986), Cogley (2012), Frost-Arnold (2008), Horsburgh (1960), Lagerspetz (1998), Løgstrup (1956), McLeod (2002), Pettit (1995), Potter (2002), Nickel (2007), O’Neil (2012), Hardin (2002), and Rose and Mishler (2011). Yet even among philosophers who agree that trustworthiness is a moral notion, accounts still differ considerably from one another in detail, making assessment of the merits of one-place trustworthiness overall challenging. This is further complicated by the fact that many philosophers take virtues in general to involve both character and action. In terminology familiar to this dissertation, such accounts make trustworthiness a sort of hybrid one/three place notion. For example, Nancy Potter thinks of trustworthiness as an Aristotelian conception of virtue, defining a trustworthy person both in terms of their actions and their inherent qualities, as:
one who can be counted on, as a matter of the sort of person he or she is, to take care of those things that others entrust to one and (following the Doctrine of the Mean) whose ways of caring are neither excessive nor deficient. (2002: 16)
“As a matter of the sort of person he or she is” suggests that Potter takes trustworthiness to be one-place, yet “to take care of those things that others entrust to one” suggests a three-place interpretation. Like Potter, many philosophers speak at times as though trustworthiness were a matter of doing what we are entrusted to do, and at other times, as though it were something more inherent and immutable. Some, like Philip Pettit, are helpfully clear about their intention to adopt a hybrid account. Pettit explicitly articulates that trustworthiness is “trust-reliableness”, which is being three-place trustworthy, plus possession of the relevant one-place qualities, which he takes to be: loyalty, virtue and prudence (1995: 208-12). More recently, Pettit, along with Victoria McGeer, has provided yet another hybrid account, where trustworthiness is both a
II.1 Trustworthiness: in the Philosophical Literature 130
“standing sensitivity” and a “situational sensitivity” (2017: 31). They describe trustworthiness as a “standing sensitivity to reasons of trust that you (the trustee) bring to the encounter” (2017: 26), though this is to be coupled with a situational sensitivity to the presence of the truster that “enhances trustworthiness” and increases the likelihood that we will do as the truster bids. Basically, their idea is that we may have some underlying trustworthiness in us, but by manifestly relying on us the truster increases both the dependability of our reliance as well as its durability: we are both more likely to comply, and more likely to comply durably in the face of any disruptors:
How trust-responsive you are in dealing with me may be a function of many factors: my nature, our relationship, and the company that we share with others, such as others who serve as witnesses to my relying on you in a certain way. In short, it may be a function of the context in which you operate as well as a function of your inherent character. It may be a context dependent or ecological capacity, not a capacity that is fixed only by how you are in yourself. (2017: 28)
Like McGeer and Pettit, Paul Faulkner also explicitly states that trust, (A trusts B to do X), is a “metaphysically hybrid notion, in that it describes an action, that is done with a certain attitude, which is best described as trustful” (2017: 119). Though Faulkner is speaking here of trust, it is likely that he would endorse a corresponding view of trustworthiness (see II.1.1). Regrettably for our purposes, most others adopting a hybrid account do not do so explicitly, and, given that we are already at a disadvantage in trying to glean a philosopher’s stance on trustworthiness from their thoughts on trust, when it comes to reviewing the merits of apparently one-place accounts of trustworthiness, it is not always possible to say whether it is the trustworthiness itself that is supposed to be a matter of a person’s actions, or the underlying qualities of a person that serve to motivate those actions, or both. Potter’s idea that it is both fits with a common understanding of virtues whereby possession of a virtue is not just some latent quality ‘in us’ but requires us to expect, value, desire, feel, act, and react in certain characteristic ways (Hursthouse and Pettigrove 2016). Yet even if it is right that it is not enough to feel and think like a trustworthy person unless we also act like one, a problem
II.1 Trustworthiness: in the Philosophical Literature 131 arises when we try to say what happens to our trustworthiness if we are ‘trusted to do something untrustworthy’. This possibility does not arise on a two-place account. However, on a hybrid account it leads to the following impasse: the three-place element seems to require that we do what we are trusted to do, but the virtue element seems to require that we do the right thing, which may not be what we have been entrusted to do. This is a problem that Karen Jones has considered. She has argued that, if we think of trustworthiness as a virtue, then we must think of untrustworthiness as a vice, but since it is conceivable that conflicts would occur where being trustworthy to one person requires being untrustworthy to another, the virtue account seems to demand that we exhibit a vice, and so cannot be correct (2012: 84). Stephen Wright makes the same point (2010: 626). And relatedly, Carolyn McLeod, worries that virtue accounts are ill- equipped to deal with the problem of unwanted trust, as since we can be trusted to commit a crime or conceal a murderer, then surely it would not be virtuous to take care of those things that others have entrusted to one (2015). Essentially, Jones, Wright and McLeod are aligned in their contention that ‘trustworthy action’ sometimes requires us to act immorally in a way that destroys any chance of maintaining our virtue, hence, trustworthiness cannot be virtuous. Notice, though, that this problem appears only when we take it to be the case that actions are a part of our trustworthiness – when we assume that trustworthiness is at least partly three-place. This problem is not a problem for one-place trustworthiness but a problem for hybrid accounts that remain ambiguous as to whether a person’s virtue is in them – in some immutable sense – or in their actions. Not all those who embrace a virtue account would agree that actions are a part of the assessment of virtue: for example, (Kawall 2009; Watson 1990). As Hursthouse and Pettigrove note:
Whereas consequentialists will define virtues as traits that yield good consequences and deontologists will define them as traits possessed by those who reliably fulfil their duties, virtue ethicists will resist the attempt to define virtues in terms of some other concept that is taken to be more fundamental. (2016)
If this assessment of virtue is correct then those who think doing what is entrusted is a part of our trustworthiness – Nancy Potter, Phillip Pettit, and Paul Faulkner – may be
II.1 Trustworthiness: in the Philosophical Literature 132 making consequentialist claims rather than virtue-theoretic claims. McLeod’s, Jones’s and Wright’s criticisms will fail to undermine trustworthiness-as-virtue. This is because we cannot be ‘entrusted’ to commit morally dubious acts as McLeod suggests, and we cannot get into conflicts over whether our trustworthiness requires that we do A or B as Jones and Wright suggest, since we cannot be entrusted to do anything at all. Trustworthy-as-virtue, on such a conception of virtue, is what we are. What this would mean – among other things – is that trusters lose any ability to determine our worthiness of trust. We could go our whole lives failing to do what others have ‘wrongly entrusted’ us to do and still be trustworthy, in the same way that we could go our whole lives failing to do what others wrongly think would be the courageous thing to do and still be courageous. Yet one might be critical of this result. As H.J.N. Horsburgh nicely put it:
once one has decided to distrust someone in one respect that distrust tends to spread to other aspects of one’s dealings with him and one’s attitude tends to settle into a rigid mould of distrust. (1960: 351)
The point here being that trusters appear to have some ability to give and to strip a person of their trustworthiness, in the sense that, if a trustee lets a truster down in a significant way, perhaps they are not worthy of that particular truster’s trust, even if they were acting virtuously. The best of friends isn’t always on the side of right; rather, they are always on our side. If effective arguments are to be levelled against virtue accounts of trustworthiness, then they must meet them on their terms. To do this, we must first sort out precisely what is meant by the claim that trustworthiness is a virtue. This is a difficult thing to do, given that there are several differing interpretations of what virtues are, for example: eudaemonist, agent-based, exemplarist, target-centred, and Platonic. Second, reviewing the literature on trustworthiness fails to give any clear sense of whether trustworthiness is supposed to be a simple virtue in and of itself, or whether it is supposed to be made up of several underlying virtues, and if so, then which ones. For example, Silvers and Francis (2005) – in contrast to Pettit’s loyalty, virtue and prudence – identify the relevant characteristics as consistency, reliability and honesty, while Paul Faulkner at one point identified sincerity and accuracy (of speakers) as
II.1 Trustworthiness: in the Philosophical Literature 133 requirements for trustworthiness (2007a: 310); and trustworthy individuals, according to Hardwig, are those who are truthful, competent, conscientious, and have adequate epistemic self-assessment (1991: 700), while Becker identifies three key areas of trustworthiness as: benevolence, conscientiousness, and reciprocity, or “meaning well, playing by the rules, and playing fair”. Becker dismisses loyalty, honesty, courage and temperance on the basis that they are not part of the core traits that are connected with our ability to feel secure, and hence to trust (1996: 53). This disagreement is actually quite telling; like the disagreements in the literature on trust, what it suggests is that trustworthiness is complex, involving potentially many things, at different times. Again, because we are usually working from an account of trust and not trustworthiness, it can be hard to say in many instances whether the philosopher is proposing that trustworthiness is a virtue in and of itself or whether it is some sort of ‘composite virtue’. This challenge aside, mention of things like honesty, conscientiousness, loyalty, and prudence makes trustworthiness seem like a moral concept, and thus the purview of ethicists. Yet some have argued that trustworthiness is an epistemological or intellectual virtue. Intellectual virtues are characteristics that promote intellectual flourishing, or which make for an excellent cognisor (Turri et al. 2017). When trustworthiness is understood as ‘saying what we mean, meaning what we say, and doing what we say we will do’, as it is in the philosophical literature on testimony (Adler 2013), it is easy to see how it might come to be considered necessary for intellectual flourishing both in an individual and within an intellectual community, as a number have claimed, see for example: Morton (2010), Stocker (2013), Kashdan and Silvia (2009). Understood as an intellectual virtue, trustworthiness facilitates the transfer of knowledge, while as a moral virtue, trustworthiness facilitates moral action. Yet these are two very different things, and they can put us into conflict with ourselves, and with each other as we go about our practical lives. While they no doubt interact with one another, it is doubtful whether any single concept – trustworthiness – could adequately capture both the phenomenon of being a morally good person, and the phenomenon of being a good knower. Especially seeing as sometimes doing the right thing requires us to forgo knowledge, for example, if others are gossiping about a third party it may be
II.1 Trustworthiness: in the Philosophical Literature 134 that we ought to walk away and not listen, even if by doing so we miss out on acquiring information. In reality, navigating the competing requirements of our intellectual and moral lives is an ongoing and deeply intertwined process. In philosophy, the argument over who gets custody of trust and trustworthiness in the philosophical divorce between epistemology and ethics has led to some intriguing attempts at rapprochement. Paul Faulkner has called trustworthiness an “ethical virtue” with “epistemic consequences” (2014a: 204). This conception has led him to speak of the trustworthiness not of a person, but of a “piece of testimony” (2014a: 191). Understood as a property of language and not of persons, language itself becomes ethically virtuous: a rather odd result. In another attempt at unifying the apparent epistemological and ethical elements to trustworthiness, Stephen Wright argues that trustworthiness is a “virtue of rationality” that requires us to consider the value of the trust that has been placed in us in deciding how to act:
here are what I take to be the necessary and sufficient conditions for ‘Q is trustworthy’ to be true: 1. Q acknowledges the value of the trust that is invested in them. 2. Q uses this to help rationally decide how to act. (2010: 622)
Wright shows concern that a trustee might do what they were entrusted to do but only for selfish, hence potentially unethical, reasons– a situation he finds intolerable:
Because I have set up a characterisation that for Q to be trustworthy all that Q has to do is appropriately value his truster’s trust in him and decide how to act accordingly, one possible objection might be that if Q was such a selfish person that they recognised that P was trusting them to do something that might not be in Q’s own interests but after weighing everything up decided that they did not want to do as P asked them to then they could prove themselves to be (under my account) trustworthy, but we would generally want to say that in such a case Q has not demonstrated trustworthiness at all. (2010: 622)
The obvious solution is to simply stipulate that any action fails to be trustworthy if it is motivated by selfishness. Yet Wright notices a problem with the obvious solution:
II.1 Trustworthiness: in the Philosophical Literature 135 acting rationally sometimes requires acting selfishly. But if acting in a trustworthy manner requires us not to be motivated by selfishness, then, at least in some situations, it seems we cannot hope to be both trustworthy and rational. Unhappy with this result, Wright proposes an alternative solution. He goes ahead and precludes any selfish act from being trustworthy yet maintains that all trustworthy acts are rational, by arguing that when we prioritise self-interest, we fail to acknowledge the value of the trusting relationship:
I think that by prioritising his own selfish interest, Q has not fulfilled the requirement of acknowledging the value of the trust placed in him, part of clause 1 of my account. The reason for this is that I think that for him to be considering the value of the trusting relationship appropriately (and therefore being trustworthy under my account) Q must consider the trusting relationship appropriately in relation to other motivations. By arriving at a conclusion that it seems that no rational person would end up at- that lazy self interest would be more valuable than or should take priority over a trusting relationship, I think that Q has set himself up for the charge of not considering the value of the trust properly. (2010: 623)
For Wright, the trustworthy thing and the rational thing are always one and the same because trust is so valuable that it naturally trumps self-interest. Wright thus takes himself to have established trustworthiness as a virtue by aligning it with rational action (2010: 625). The problem this dissertation finds with Wright’s solution is that selfishness is not an on/off thing. Many of the things we altruistically do for others have benefits to self. Requiring trustworthy trustees to be wholly unselfish will likely preclude us all; I provide argument for this claim in II.1.4. To summarise, one-place trustworthiness is a view that says that our worthiness of trust takes no object. It is not about actions, nor the consequences of those actions, nor the people involved in those actions and what they may want from us. It is instead an inherent characteristic of a person, often moralised. This subsection demonstrated that most accounts that may at first appear to be one-place, such as those that take trustworthiness to be an ethical or epistemological virtue, are not one-place at all, but are in fact hybrid one/three-place views. Hybrid views take actions – doing what we were entrusted to do – to be an essential factor in determining our trustworthiness. For
II.1 Trustworthiness: in the Philosophical Literature 136 better or worse, hybrid views thus make the opinions, wishes, desires and expectations of the truster, as well as the trustee’s ability to satisfy those desires and live up to those expectations part of their trustworthiness. As they contain the three-place element, hybrid views of trustworthiness face the same limitations as standard three-place trustworthiness.
II.1.3 Three-place trustworthiness
When taken to include the ‘hybrid’ one/three-place views identified in the previous subsection, three-place trustworthiness is the dominant view to be found in the literature, in the sense that most philosophers speak, at least at times, as though trustworthiness is a matter of the trustee doing what the truster has entrusted them to do. At its most basic, whether or not character or motives are taken to be a part of trustworthiness, three-place trustworthiness, like three-place trust, is the view that trustworthiness involves these same three elements: a truster, a trustee, and that which is entrusted. On the strong version of three-place trustworthiness, the trustee must successfully do what the truster has entrusted them to do. On the weaker version, the trustee is only required to be competent and committed to doing what the truster has entrusted them to do. The weaker version thus allows us to remain trustworthy in circumstances where we are prevented from doing what we were entrusted to do through no fault of our own. Importantly, both the strong and the weak versions of three-place trustworthiness makes a specific act, entrusted to us by the truster, a part of our trustworthiness. The strong version requires that we successfully execute that act, and the weak version requires just that we have the relevant ability to execute it (competence), and intention to do so (commitment). Which version – strong or weak – is preferred, tends to vary from discipline to discipline. In philosophy, the strong version is less popular than the weak version, but some still do adopt it, for example T. W. Simpson (2017) and Hinchman (2017). Outside of philosophy, the strong version is dominant. For example, trust and trustworthiness are recognised as central concepts in science (Whyte and Crease 2010); but more precisely, it is strong three-place trustworthiness that is adopted in relation to
II.1 Trustworthiness: in the Philosophical Literature 137 science. When scientists and philosophers of science speak of the need for scientists to be trustworthy, they tend to mean that it is essential that scientists successfully do what they are entrusted to do, not just be competent and committed to doing it. What this means in practice is contentious, and it is not necessary to go into details here, but in short it involves adhering to certain standards of method and accuracy (for an overview see Andersen and Hepburn (2016). As an example of how the concept of trustworthiness takes on various roles in the philosophy of science literature, take the following passage, which comes from a feminist perspective on epistemology, an anthology entitled Engendering Rationalities:
The dependency of scientists on other scientists – of peers on peers within shared institutional settings – while less obviously irrational needs to be called into question in the light of what are widely acknowledged to be the problematic ways in which power and privilege shape the workings of the practices meant to ensure trustworthiness. But there is an additional and deeper problem than those involving the trustworthiness of scientists and scientific practices. All along those who have been the authorised knowers have been, in subtle and complex ways dependent on those whom they would not have acknowledged, except, perhaps, in the most purely theoretical of terms as their peers; and those forms of dependency have gone unacknowledged and unaccounted for in assessing the trustworthiness of knowledge claims (Scheman 2001: 41).
As this passage demonstrates, problematically, there is an ambiguity in the way the concept of trustworthiness is applied in the philosophy of science literature. Sometimes trustworthiness is understood as a property of scientists – reliable testifiers, as in (Lang and Hallman 2005), and at other times as a property of scientific knowledge – reliable testimony, as in (Millstone and van Zwanenberg 2000). Whether trustworthiness is taken to be a property of scientific knowledge or of scientific knowers, discourse to the effect that science/scientists must ‘be trustworthy’ has become the norm since 1985, when John Hardwig articulated the philosophical dilemma posed by large teams of researchers. As the body of knowledge in science grows too large for any one person to possess, the members of scientific community are not in a position to evaluate the results of other members' work. This creates a situation in which the members must take
II.1 Trustworthiness: in the Philosophical Literature 138 one another’s results ‘on trust’. The framing of this issue as an issue of trust and trustworthiness is evident in Hardwig’s wording:
Modern knowers cannot be independent and self-reliant, not even in their own fields of specialization. In most disciplines, those who do not trust cannot know; those who do not trust cannot have the best evidence for their beliefs. In an important sense then, trust is often epistemologically even more basic than empirical data or logical arguments: the data and the arguments are available only through trust. (1991: 693)
“Those who cannot know; those who cannot trust.” For Hardwig, trust and trustworthiness are clearly cognitive epistemological notions, and he places them at the centre of knowledge. If ‘trustworthiness’ as he uses it is understood as reliability, then there is no problem. However, because unlike trust, trustworthiness is often understood to be a moral or social notion (see II.1.2 and II.1.4), Hardwig’s claim risks being confused with a requirement that scientists be friendly with one another, or good moral citizens. Yet Hardwig is saying something far less contentious: science cannot progress unless scientists follow the scientific method, collaborate, and accept one another’s results. Given Part I of this dissertation, we can see that the ‘trustworthiness of scientists’, in Hardwig’s sense, is just the reliability of scientists at doing science. Hardwig’s point is about evidence and how the sheer amount of it now available outstrips any one person’s ability to absorb and assess it. It is not about scientists becoming friends. It is not about scientists ‘doing what the good man would do’. It is about scientists being reliable and cooperative and accepting one another’s testimony. Hardwig’s use of the terms ‘trust’ and ‘trustworthiness’ to capture this situation has had significant downstream effects. Today, his Epistemology of Mass Collaboration problem is not the only one discussed as if it were a problem of trust and trustworthiness. Other related problems, such as those to do with reproducibility and replicability, testimony, peer disagreement, epistemic relativism, evidence, and judgement aggregation, are currently framed as problems of trust and trustworthiness as well even though they too are epistemological problems to do with knowledge. What’s more, these important epistemological problems are discussed not only by philosophers of science but also by philosophers of law such as Talbott and Goldman (1998) and
II.1 Trustworthiness: in the Philosophical Literature 139
Laudan (2006); as well as by social epistemologists such as List and Pettit (2011), Coady (1992), Gilbert (1989), and Goldman (2001, 2011). Examples of questions posed in this arena include:
1) Can a layperson justifiably identify which (professed) expert to trust? 2) Is a speaker S trustworthy for hearer H only if H has positive evidence or justification for the reliability of this particular speaker S? As should be familiar from Part I, we can reframe these epistemic questions without reference to trust and trustworthiness without any conceptual loss, thereby reserving the concepts trust and trustworthiness for their vital roles in accounting for our experience of one another in our close interpersonal relationships:
3) Can a layperson justifiably identify which expert to believe?
4) Is a speaker S reliable for hearer H only if H has positive evidence or justification for the reliability of this particular speaker S? These questions come into view as epistemological and not ethical or interpersonal, when we consider that those working on solutions to them have tended to propose epistemically-targeted solutions. Michael Blias for example, argued that trustworthiness can be modelled as a strategy– the correct strategy for all members of a scientific community (1987: 370). And Philip Kitcher has proposed a prescriptive decision theoretic models as solutions to ‘trust problems’ (Kitcher 1993). Today, such decision theoretic solutions are commonly applied to numerous areas of academic inquiry. Many of those concerned about ‘trustworthiness’ as it relates to governments, financial institutions, and trans-national corporations adopt a risk-assessment position on trustworthiness, and so naturally adopt decision theoretic solutions, where self-interest is presumed to be the norm (Coeckelbergh 2012; Govier 1997; Hardin 2002; Potter 2002; Taddeo and Floridi 2011; Townley and Garfield 2013). Such epistemically-targeted solutions will not solve interpersonal issues of trust and trustworthiness, and, they are not always adequate even in these domains. Torsten Wilholt has argued that scientific inquiry – his argument would naturally extend to legal, social, political and medical settings – engages ethical norms as well as epistemic norms and a ‘deeper attitude of trust’ based on shared values is required for things to go well (2013). Wilholt’s proposal echoes Nancy Daukas’ distinction between epistemic
II.1 Trustworthiness: in the Philosophical Literature 140 trustworthiness, which is “a character trait that supervenes on a relation between first- and second-order beliefs, including beliefs about others as epistemic agents” (2006: 109), and moral trustworthiness, which is one who is “benevolent, and they sincerely believe what they say, say what they believe, and behave consistently with those beliefs” (2006: 110). Though Wilholt addresses his concern to science and Daukas to politics and society, the key point made by both is that strong three-place trustworthiness is unsatisfactory, even as it is applied to these domains. Their point is highly pertinent: when it comes to certain human endeavours, epistemically targeted solutions may not suffice to ‘fix’ things, since good outcomes can require more than good information. We can have moral deficits, and interpersonal deficits, and in some arenas – government, academia, international relations, medicine and science – these deficits may cause just as many problems as truth deficits. To secure good outcomes in medical contexts for example, doctors no doubt need to be knowledgeable. But they may also need a decent moral compass and good interpersonal skills. If we separate out these three requirements: 1) knowledge and expertise, 2) an ethical conscience, and 3) trustworthiness (understood as one half of an interpersonal relationship shared with their patient), rather than using ‘trustworthiness’ to refer ambiguously to all three, then it will be easier to identify precise deficits and propose targeted solutions. Looking briefly even further afield, many studies in relationship marketing have been based upon the Commitment-Trust Theory developed by Morgan and Hunt (1994). Within this theoretical framework trust is described as “confidence in the exchange partner’s reliability and integrity” and trustworthiness described as “the desire or intention to maintain a valued relationship into the future” (1994: 23). The Commitment Trust Theory is aimed at supporting mutual advantage in exchange relationships. Similarly, the now famous Trust Game, designed by Berg et al. (1995) is an experiment of choice in behavioural economics designed to measure so-called trust in economic decisions. Yet what the Trust Game actually reveals is our tendency to disregard crude self-interest and instead pay attention to an equally crude ideal of fairness under a very limited range of controlled conditions. The Trust Game is not a game of trust, but a game of prediction and punishment.
II.1 Trustworthiness: in the Philosophical Literature 141
So far, we have been discussing the application of strong three-place trustworthiness to communities (political, social, scientific) where epistemic agents need to rely on other epistemic agents for information. Yet, as philosophers have long noticed, it is not only other agents that agents must rely on in this way, as discussed already in relation to self-trust (I.4.5). Edward Hinchman and Trudy Govier are interested in how the problem of epistemic reliability in others applies to our selves, and both utilise strong three-place trustworthiness in their discussions. Hinchman says that if you are not trustworthy, or if you have good evidence that you are not, then you should not, by the norms of rationality, trust yourself, since “trust presupposes trustworthiness in the trusted” (2003: 26). Similarly, Govier says that “A is trustworthy if and only if A is disposed to behave [when contextually appropriate] as though her epistemic status is S if and only if her epistemic status is S” (1993: 111). For Govier, the problem is that:
We have to make judgements about what is going on, make decisions and implement them, and do this ourselves. If we are insecure about our own values, motives, and capacities, we cannot think and act effectively. (1993: 106)
To solve this problem, she uses the notions of trust and trustworthiness in an extensive sense, so that:
One may be called upon to trust one’s perception and observations, interpretation of events and actions, feelings and responses, values and evaluation, memory, judgement, instinct, common sense, deliberation, choice, will, capacity to act, flexibility, competence, talent, and ability to cope with the unexpected. (1993: 108)
For Govier, trust in another, and so knowledge gained from others, can only be maintained so long as a person is “self-trustworthy” (1993: 117). In opposition to this, and resonating with Descartes, Mark Owen Webb implies that self-trust is not required for self-knowledge when he says “trust is necessary if one wants to have knowledge of anything other than [my emphasis] one’s own immediate experience” (1993: 260). Whatever the merits of Webb’s contention, what is important to recognise for our purposes here, is that the worry about being “self-trustworthy” in Hinchman and Govier’s sense is a worry about how, for our reliance on ourselves to be rational, we
II.1 Trustworthiness: in the Philosophical Literature 142 ourselves must have reliable perceptive, observational, and deliberative methods and systems. This should be uncontroversial. Just as it is used by Hardwig and others, the strong three-place sense of ‘trustworthy’ employed in discussions of self- trustworthiness, is so close to the idea of being a reliable epistemic agent that it is indistinguishable. The terms are interchangeable without conceptual loss. The central contention in such ‘problems of trustworthiness’ is that successful knowledge acquisition and transfer requires reliable knowledge acquisition and transferring systems. This dissertation poses no contrary argument to this; the argument here is that this problem is not a problem of trust or trustworthiness because the assumption that it involves trust and trustworthiness is mistaken. Being strongly three-place trustworthy is the same as being reliable. Such conceptual overlap is ill-advised, because the concept of reliability does not capture all there is to care about when it comes to dependency relations involving human beings. The socially and morally determined way in which we interact and hold one another (and ourselves) to account explains why some have instead adopted the weak version of trustworthiness, where it is enough that the trustee be both competent and committed to do what they have been entrusted to do (Dooley and Fryxell 1999; McLeod 2015; M. J. Smith and desJardins 2005, 2008). According to this view, what matters is not success as such, but a combination of ability and intention. This would not fit the conception of trustworthiness as it is commonly used in science, as being competent and committed to conveying reliable scientific information is not commonly taken to be enough to produce trustworthy science. In that context, success is what matters, not intention (Ioannidis 2005; Nyhan et al. 2014). Yet while strong three-place trustworthiness may capture perfectly well the importance of accuracy to the scientific enterprise, it is limiting as to what it can capture about the inherently fallible practitioners of science – scientists – as it makes no distinctions between 1) a scientist who engages in premeditated scientific fraud for personal gain, 2) a scientist who innocently publishes misleading information, and 3) a scientist who published false information to save their colleague and friend from public humiliation. If we think such distinctions matter to how we treat and understand the people who are scientists, then we are better served by
II.1 Trustworthiness: in the Philosophical Literature 143 reserving a more nuanced and ‘restricted’ version of trustworthiness so that we can respond appropriately. Weak three-place trustworthiness finds trustworthiness where there is competence and commitment, even if accident, adverse circumstance, mistake, or confusion sometimes results in a kind of blameless failure of accuracy. Being weakly trustworthy is distinct from being reliable; this is a good thing, as we now have two distinct concepts that serve to explain two distinct phenomena. If we adopt weak three-place trustworthiness, then we can have a trustworthy scientist (competent and committed) who, due to the unforeseeable, produced an unreliable result. In contrast, if we adopt strong three-place trustworthiness, all we can get out of the same scenario is an untrustworthy scientist. Yet weak three-place trustworthiness is not without its critics. One problem for the competency condition in particular, noted by Karen Jones, is that it does not seem enough that we are competent at doing something if we fail to let others know that we are. To fix this problem, Jones distinguishes between three-place trustworthiness and “rich trustworthiness”, which she takes to require:
not only competence in a domain, but also competence in assessing my own competence, so that I neither signal competences I do not have, nor hide my light under a bushel. I need to engage in ongoing reflective self-monitoring of my own competences so that I know them and their limits. (2012: 76)
This dissertation addresses Jones’ rich trustworthiness in detail in II.3. For now, I will say that once again, what Jones is picking up on here is that possession of the relevant knowledge or skill may not be enough if we cannot know that we possess it, and do not have the skills to competently signal this to ourselves and to others. It is again the same observation that Hardwig, Govier, Hinchman, and countless others have made, only on a meta level. Possessing and transferring knowledge successfully requires reliable possession and transfer structures and systems all the way up and down for things to go well. What so many are framing as a ‘problem of trust’ is just the pervasive reality that epistemic problems will arise when certain parts of a system rely on other parts to transmit and receive information, and errors and faults (both intentional and accidental)
II.1 Trustworthiness: in the Philosophical Literature 144 are possible at multiple levels throughout the system. This describes the situation we are in regarding knowledge of ourselves (since our senses can deceive us) and knowledge gained from others (since others can deceive us), hence it is the status quo in self, scientific, social, political knowledge… indeed all knowledge. The risks that both strong and weak three-place trustworthiness emphasise are taken to stem from the fact that we have relationships with others whose behaviour we can neither predict nor control (Gambetta 1988; Luhmann 1979) and not from the fact that we have relationships with others whose complex selves are, at least to some extent, veiled from us (and probably to some extent from themselves as well). The commitment condition in weak three-place trustworthiness may at first seem to fix this problem, by taking the trustworthiness away from the agent’s success and putting it back in the agent, in their intentions. However, as some have noticed, a trustee can be committed to doing what they have been entrusted to do for many different reasons. The commitment condition alone does not specify that a trustee be committed for the sorts of reasons we would ordinary think of as trustworthy. For example, some philosophers identify the force of norms or social constraints as compelling the commitment of a trustworthy trustee (Dasgupta 1988; Hardin 2002; O’neill 2002). As Carolyn McLeod recognises, on this view, what she calls the social contract view, the commitment stems from the social norm itself and not from, say, benevolence or care. The problem with social contract views, McLeod says, is that while social constraints “can shore up trustworthiness, they cannot account for trustworthiness altogether” (2015: Section 1). She points to Nancy Potter’s argument that, on the social contracts view, a sexist employer – the kind who treats female employees well only because he thinks he would face legal sanctions if he did not – would be trustworthy (2002: 5), which seems counterintuitive. McLeod tells us that:
An alternative to the social contract view is a view [the encapsulated interests view] according to which those who are trustworthy are motivated by their own interest to maintain the relationship they have with the trustor, which in turn encourages them to encapsulate the interests of that trustor in their own interests. (McLeod 2015: Section 1)
II.1 Trustworthiness: in the Philosophical Literature 145
This is the view adopted by Russell Hardin, discussed already in II.1.1. Yet, as McLeod points out, it too is problematic:
Consider how it applies to the sexist employer. He is not motivated by an interest to sustain his relationships with female employees: if he could easily fire them, or even avoid hiring them altogether, then he would do that. He is therefore not trustworthy. (2015: Section 1)
Both the social contract view and the encapsulated interests view are instances of weak three-place trustworthiness, since both contain the commitment condition. And yet, they are both instances of “risk-assessment views” of trust or trustworthiness, as Jones calls them (1999: 68), which do not require the commitment to stem from what we would ordinarily think of as trustworthy motives. This means that they allow for no distinction between being trustworthy and merely acting in a trustworthy manner or pretending to be trustworthy. Responding to this intuitive limitation, several philosophers require of the commitment condition that the commitment be of a certain type, a certain kind of motive. These views will be discussed in the following section. Lastly though, while almost no accounts challenge directly the assumption that trustworthiness is three-place, one notable exception can be gleaned in the work of Lars Hertzberg. In a 1988 paper “On the Attitude of Trust”, Hertzberg observed that reliance has a more or less specific content: one relies on a person for particular purposes. Reliance is like factual belief in the sense that “my relying on someone is conceptually independent of whatever attitude I take him to have in other respects” (1988: 312), whereas “when I trust someone, it is him I trust; I do not trust certain things about him” (1988: 315). Given these comments, Hertzberg can perhaps be viewed as a proponent of two-place trustworthiness, recognising as he does that trust is placed, and so presumably trustworthiness must reside, in persons, not in their actions. It is telling that he is rarely cited in the three-place trust/worthiness literature. To summarise, this subsection addressed literature on three-place trustworthiness. Three-place trustworthiness is a view that says trustworthiness involves the trustee doing for the truster, that which is entrusted. It splits into two versions: strong and weak. The strong version is adopted in the science, and social and political science literature, and it is indistinguishable from reliability. The weak version is more
II.1 Trustworthiness: in the Philosophical Literature 146 prevalent in the philosophical literature on trust, and it allows that the trustee might fail to do what they have been entrusted to do through no fault of their own. Yet by itself, the weak version says nothing about the underlying character or motivations of the trustee, and so, without further specification, counterintuitively counts trustees with untrustworthy motives as worthy of trust.
II.1.4 Motives and trustworthiness
Motives are thought to be relevant to trustworthiness in the following way: a trustee might appear to be trustworthy, in the sense that they are helpful, cooperative, share accurate information with us, and generally do what we entrust them to do, yet such a trustee may be committed to behaving in this manner only for reasons that are incompatible with trustworthiness: Mr Fox, in the Beatrix Potter story “Jemima Puddle- Duck”. At least according to our common-sense platitudes, the intentions of the trustee matter to their trustworthiness. Indeed, this is how we make that important conceptual differentiation between false and true friends, who may appear to us – at least at first – as indistinguishable in their behaviour. Yet both the strong and weak versions of three- place trustworthiness say nothing about the sorts of motives a trustworthy person must have. The trustee does not have to like the truster, care about them, want the best for them (or not be planning on eating them). The strong version requires just that the trustee successfully does what they were entrusted to do, and the weak version requires just that they are competent and committed in doing it. Both versions are compatible with the trustee having selfish or even malicious motives for doing what we ‘entrust’ them to do. This subsection discusses accounts of trustworthiness which are not entirely behaviouristic. Among those who recognise and try to address this limitation with three-place trustworthiness, there is little agreement in the literature as to what kind of additional motive a person needs to possess (or lack) for doing what is entrusted. Aligning trustworthiness with morality, Carolyn McLeod says that a sense of duty is a sufficient motivator to ensure trustworthiness. This is because – she argues – all we need to have to be able to trust strangers to be decent is the presumption that they are committed to common decency (2002: 21-27). Elsewhere she summarises her own view “ultimately,
II.1 Trustworthiness: in the Philosophical Literature 147 what I am presuming about the stranger is moral integrity, which has been thought of as the relevant motive for trust relations” (McLeod 2015: Section 1). She then goes on to identify others, i.e. (Cohen and Dienhart 2013; Nickel 2007), who similarly classify the relevant motive as duty or moral obligation. Some talk not of a sense of moral obligation, but of a recognition of dependency. For example, Paul Faulkner says:
A trusted party S is trustworthy, in a circumstance defined by A’s (affectively) trusting S to φ, if and only if S sees A’s depending on his φ-ing as a reason to φ and φs for this reason. (2011: 148).
And similarly to Karen Jones who I shall discuss in detail in II.4, Amy Mullin argues that those who are genuinely trustworthy can be motivated by a different sort of non- moral commitment– a commitment to a particular social norm (2005: 316). Being motivated by a commitment to a social norm, sense of duty, or recognition of dependency though, strikes some as inadequate, since one could be committed in these ways and still feel ill-will towards the truster (which seems to preclude trustworthiness). Will-based accounts say that trustworthy trustees are always motivated to do as they are entrusted out of goodwill. Discussion of the relevance of goodwill to moral philosophy finds its origins in Plato (Protagoras and Republic, Book 4) and Aristotle (De Anima, see esp. III.10), see also Price (2011), but application of the concept of goodwill to trust and trustworthiness originates in the work of Annette Baier. Baier’s own motivation for applying the concept of goodwill is very clearly to use it to distinguish trust/worthiness from reliance/reliability:
What is the difference between trusting others and merely relying on them? It seems to be reliance on their good will toward one, as distinct from their dependable habits, or only on their dependably exhibited fear, anger, or other motives compatible with ill will toward one, or on motives not directed on one at all. (1986: 234)
Today, broader issues to do with goodwill are taken up under the heading Moral Motivation, and debate is lively with wide-ranging implications (Rosati 2016). It is no
II.1 Trustworthiness: in the Philosophical Literature 148 surprise to see the familiar arguments of moral motivation playing out in the literature on trust/worthiness. Much of the contention when it comes to the concept of goodwill involves agreeing on precisely what ‘having a good will’ involves: is it a cognitive notion or a non-cognitive notion? For example, on the cognitive side, some have claimed that a good will is the will of one who is committed only to make decisions that she holds to be morally worthy and who takes moral considerations in themselves to be conclusive reasons for guiding her behaviour (Johnson and Cureton 2017). One example of such cognitive goodwill applied to trustworthiness comes from Colin O’Neil, who takes goodwill to not include any benevolent feelings, just respect for our “rights”:
By “good will,” I do not mean that the trustee’s will is benevolent, but merely that it is responsive to the trustee’s obligations. We can trust those who do not care how things go for us—we can even trust enemies who hate us—so long as we expect them to be responsive to our rights. (2012: 309)
Cognitive views align trustworthiness with moral action rooted in reason rather than emotions or desires. Because of the emphasis on obligation and respect for rights, cognitive views can render trustworthiness indistinguishable from moral action. They tend to be adopted by those who view the alternative as too narrow, precluding us from being parties to trusting relationships with, as O’Neil says, those “enemies who hate us”. If we apply the reasoning of Part I of this dissertation to O’Neil’s contention, then ‘trusting enemies who hate us to be responsive to our rights’ is instead understood as relying on enemies who hate us to behave in morally appropriate ways. Like O’Neil, according to Jones the goodwill involved in trustworthiness is cognitive and involves no “friendly feeling”:
We are not to take 'goodwill' as synonymous with 'personal friendly feeling', or will-based accounts would be obviously too narrow. Sometimes those we trust have the relevant goodwill just in virtue of being morally decent, or honest, or caring about fulfilling their duty. (1999: 68)
II.1 Trustworthiness: in the Philosophical Literature 149
If we are talking about people doing the right thing, morally speaking, without any interpersonal factors at play, then it may make sense to adopt cognitive goodwill, since in this case we may not care whether those people care (affectively) about us, as long as they compliantly respect our rights. However, as Stephen Wright has pointed out with regard to Jones’ cognitive trustworthiness, it is similar to Hardin’s risk-assessment view, since both make trustworthiness essentially a matter of compliance:
In order to be trustworthy I must act because you are counting on me and I believe the fact that you are counting on me to be a compelling reason to act according to Jones. The reason that I find this unsatisfactory is that this response can lead Jones to a position similar to the one that Hardin is in, in that she demands compliance in order for someone to be trustworthy, in Jones’ case being motivated by the fact that someone is trusting them. If someone being trustworthy means that they take the fact that someone is trusting them to be a reason that compels them to do whatever they are being trusted to do then trustworthiness is associated for Jones with compliance. If Jones associates trustworthiness with acting as a trustor would like then her view of trustworthiness becomes similar to Hardin’s view that being trustworthy is doing what someone has trusted you to do. The reason that I believe that Jones’ view becomes closer to Hardin’s is because their end products become the same (compliance) and it is just the method of getting there that differs. (2010: 623-24)
Compliance (doing what we are entrusted to do) is potentially problematic, as it can put us into conflict with morality, and Jones herself worries about this implication, as we saw in II.1.2. Those applying a cognitive version of goodwill to trustworthiness may not avoid the intuitive problem laid out at the beginning of this subsection. This is because we do not want those whom we trust just to be concerned with fulfilling their duties or respecting our rights. We want them to be ready to abandon those norms/duties/rights for us if the need arises. This seems to require that they care about us, for, what else could enslave our reason (including our moral reasoning) in the way it is commonly enslaved in our interpersonal relationships but genuine affection? David Hume at least thought that goodwill and benevolence – in the affective sense – come together. In An Enquiry into the Principles of Morals, Hume says:
II.1 Trustworthiness: in the Philosophical Literature 150
It may be esteemed, perhaps, a superfluous task to prove, that the benevolent or softer affections are ESTIMABLE; and wherever they appear, engage the approbation, and good- will of mankind. (1751: Section II, Part I)
If we apply Hume’s non-cognitive (or not wholly cognitive) goodwill to trustworthiness, requiring as it does an element of affective-benevolence, then this gets us closer to our goal of distinguishing those who are committed to doing what we have entrusted them to do from a place of trustworthiness, and those who are committed from a place of untrustworthiness. Sadly though, Hume’s non-cognitive goodwill is not the conception most philosophers applying the notion of goodwill to trustworthiness intend. This is puzzling, since the primary point of employing the notion of goodwill in an account of trustworthiness is to distinguish a trustworthy person from a reliable person who may be acting out of maliciousness/selfishness. At least, this was Baier’s original intention, as we have seen. Hume’s conception of goodwill gets us closer, but goodwill as it was understood by Aristotle may not be able to serve to make the distinction we need between false friends (who may be reliable) and true friends (who are trustworthy). In Nicomachean Ethics, Book VIII, Aristotle reflects on three-types of friendship: those based on utility, pleasure and the good. All three require goodwill, though the first two kinds of friendship are self-serving, since under these circumstances, friends are motivated by their own utility and pleasure, not by anything essential to the nature of the friend. Hence, Russell Hardin’s position, for example, where “motive” just refers to “…the interest we have in maintaining particular relationships that makes it in our interest to be trustworthy” (1996: 52) is compatible with the trustee having that motive only for self- interested reasons, just as in Aristotle’s first two examples of friendship. Goodwill is – or has become at least – a moral concept, and aligning trustworthiness with moral action will always bring us back to the same problem Stephen Wright saw with Jones’ view, i.e. what happens when a truster trusts us to do something immoral? In that case it seems, we are ‘damned if we do and damned if we don’t’. Wright avoids using the notion altogether, arguing that trustworthiness need not involve any goodwill since the kind of vulnerability goodwill is a response to is not a part of trust: “I see no reason for an account to include goodwill as Baier suggests, or even the absence of ill
II.1 Trustworthiness: in the Philosophical Literature 151 will” (2010: 619). Instead he talks of the difference between “I value you”, versus “I value my relationship with you”, with only the former “I value you” being understood by him as trustworthy (2010: 622). Valuing a person in and of themselves is a step in the right direction. However, in Part I (I.4) this dissertation has argued that even more is needed: trusters and trustees must become special to one another. It is not enough to value a truster or trustee as a person, for their personhood, because they have ‘inalienable rights’. We can value a stranger in this way. In trust and trustworthiness, we must value the object specifically, in a way that makes it irreplaceable not in a general sense, but irreplaceable to us. In relation to Wright’s view, we could be in the position of valuing a person more than we value our relationship with them, but still be valuing them for selfish or underhanded reasons (a result Wright himself aims to avoid – see II.1.2). This is plausibly true of employer/employee associations in larger companies, where there may not be any relationship between the parties to speak of (they may not even know one another) and yet, the employee is still valued by the employer for the specific role they fulfil. This point may be purely terminological though, as Wright could potentially counter that what the employer values in such cases is his relationship with his employee, in the sense that he values being an employer and having an employee (in that role), and not the person that is the employee. Like Wright, Philip Pettit addresses the relationship between trust/worthiness, and the motive of selfishness in his 1995 paper “The Cunning of Trust”. Pettit is working from a three-place epistemic conception of trust and trustworthiness, whereby both are ubiquitous (1995: 218) responses to reasons. He says:
Trust materializes reliably among people to the extent that they have beliefs about one another that make trust a sensible attitude to adopt. And trust reliably survives among people to the extent that those beliefs prove to be correct. Trustors identify reasons to trust others and trustees show that those reasons are good reasons. (1995: 202)
Showing that “those reasons are good reasons” for Pettit means showing ourselves to be reliable. When it comes to trustworthiness, or “trust-responsiveness” as he calls it, Pettit
II.1 Trustworthiness: in the Philosophical Literature 152 identifies that one of the reasons we have to be trustworthy (to do what we are entrusted to do) is intrinsically selfish:
the trust-responsiveness that I have in mind is not a trait that many will be proud to acknowledge in themselves. It is the desire for the good opinion of others and it counts by most peoples' lights, not as a desirable feature for which they need to strive, but rather as a disposition – a neutral or even shameful disposition – that it is hard to shed. The fact that it [trust] can be supported by such a disposition shows a certain cunning on the part of trust. Trustors do not have to depend on the more or less admirable trustworthiness of others; they can also hope to exploit the relatively base desire to be well considered. (1995: 203)
Pettit is drawing our attention here to those situations in which we ‘do what we are being trusted to do’ out of a desire for the good opinion of others. Given that this ‘reason to be trustworthy’ is frequently present, we seem often to have a selfish (and so potentially untrustworthy) reason to be trustworthy. If we think this reason is generally motivating, then, trusters seem to have a corresponding (potentially untrusting) reason to trust. The ‘reasons to trust and be trustworthy’ that Pettit identifies here are not denied in this dissertation’s two-place trust/worthiness. Instead, they are reframed as a reason to behave reliably and a reason to believe that others will behave reliably. Both Wright and Pettit address trustworthiness as it relates to selfish motives from a three-place perspective. Wright displays concern that selfish motives may underlie doing as we are entrusted to do, and Pettit views selfishness as an additional motivating reason to do as we are entrusted to do. Also addressing trustworthiness and selfishness from a three-place perspective, Jones argues that selfish motives for doing as we are entrusted to do are not incompatible with trustworthiness. She says that a concern for maximising the wellbeing of the truster over one’s own wellbeing cannot underpin and indeed can sometimes invalidate trustworthiness, since the truster may not value their own wellbeing, or value other things – such as the trustee – more highly (1996: 10). She says “one is not trustworthy unless one is willing to give significant weight to the fact that the other is counting on one, and so will not let that consideration be overruled by just any other [my emphasis] concern one has” (1996: 8), suggesting that selfish motives can play a part, but they must be measured. For Jones, a trustworthy trustee must consider the expectations of the truster and use this as their primary motivation
II.1 Trustworthiness: in the Philosophical Literature 153 when deciding how to act (1996: 6). Jane Mansbridge agrees with Jones and recognises “altruistic trust” and “self-interest trust” as both legitimate kinds of trust and suggests that altruistic trust is actually more fragile since people are more likely to balk if they have no self-interest at all in the trust/trustworthiness (1999: 290-309). Though this dissertation does not recognise “altruistic trust” and “self-interest trust” as kinds of trust, with regard to the general point this dissertation is in alignment with Mansbridge. This is because, for one thing, sustaining relationships of trust is inherently good for us: the stress that can result from living with betrayal and distrust is bad for our health (DeLongis et al. 1988; McEwen 2008; Rasmussen et al. 2009). For another thing, the presence of self-interested ‘you scratch my back, I’ll scratch yours’ agreements in trusting relationships may actually be essential to maintaining them, particularly when responsibilities are shared– such as in parenting. Indeed, successful close interpersonal relationships require a kind of long-term reciprocity that is rooted in each party successfully communicating and getting their own selfish needs met (Buunk and Schaufeli 2011; Knudson-Martin 2013). And lastly, as H.J.N. Horsburgh has pointed out there are moral benefits to trustworthiness as well:
one of the main ways in which one individual can affect the moral growth of another is by trusting or distrusting him for each instance of distrust erects barriers to intercourse and cooperation, thereby limiting the scope – and with it the moral opportunities – of the distrusted person’s life. (1960: 350)
Though trustworthiness may at first intuitively seem incompatible with selfishness, being trustworthy is rationally self-interested. It is good for our health, the health of our relationships (which confer benefits to us), and it is good for our moral development. But these reasons cannot serve to explain why we are trustworthy. Trustworthiness is also, at heart, stunningly selfless. Yet the fact that trustworthiness can be both selfless and selfish should not lead us to think that there is both “altruistic trustworthiness” and “selfish trustworthiness”. There is just us, people who are trustworthy, and we are both altruistic and selfish in our interpersonal relationships. My daughter, who sometimes makes my heart sing with her courageous protectiveness and generosity towards her younger brother, also manipulates him into giving her his lollies and slams her bedroom
II.1 Trustworthiness: in the Philosophical Literature 154 door in his face. The selfish/selfless nature of the sibling relationship does not undermine its status as trustworthy. Often our selfish siblings are the most trustworthy individuals any of us are lucky enough to encounter in our lifetimes. To undermine two-place trustworthiness, just as to undermine two-place trust, what is needed is a significant and prolonged change in the trustee’s experience of the truster (see I.4.6). The presence or absence of a single motive underlying a single act, will not do it. On two-place trustworthiness, because doing what we are entrusted to do is not part of our trustworthiness, trustees can have multiple and even incongruous motives for their reliability without their trustworthiness necessarily being affected. This fits with the contemporary idea that a complex of motivational elements underlies even relatively simple behaviour. As neurobiologist Eleanor Simpson notes:
Many different factors influence motivation, including the organism’s internal physiological states, the current environmental conditions, as well as the organism’s past history and experiences. (2016: 4)
In assessing a person’s trustworthiness, this dissertation is not suggesting that motives are unimportant. Indeed, in II.2, we will see various ways in which they can affect trustworthiness. But given the sophistication and changeability of our motivational landscape, motives over time are what matter. For example, deep underlying intractable motives, not surface level ones like the passing desire to be well-regarded by those watching us as we do what we are being ‘entrusted’ to do, right here, right now. And, being only one part of trust and trustworthiness, motives are only ever one part of what matters. In summary, this subsection demonstrated that a number of philosophers agree that doing what we have been entrusted to do, or being competent and committed to doing it, may not be enough for trustworthiness. In particular, selfish motives are thought to be of concern. Yet rather than abandoning three-place trustworthiness, various additional motivational elements have been added to three-place trustworthiness: possessing a sense of duty, goodwill, benevolence, lack of selfishness, consideration of the value of the relationship, responsiveness to dependency, willingness to give significant weight to being counted on, etc. Much in the same way that trust is often viewed as reliance plus
II.1 Trustworthiness: in the Philosophical Literature 155 something else (hope or belief) as identified in I.2.1., motive accounts view trustworthiness as reliability plus something else, though there is little agreement – once again – on what must be added to three-place trustworthiness (reliability) to rescue it from any intuitively untrustworthy underpinnings. If we adopt three-place trustworthiness, discussions of how motives matter to trustworthiness risk being taken in the immediate sense, as relating to a specific act of reliability, instead of in that deeper ongoing sense that actually matters to our worthiness of trust.
II.1.5 Summary of the literature on trustworthiness
Subsection II.1.1 began with a demonstration that trustworthiness is symmetrical with trust, in the sense that a philosopher’s stance on trustworthiness tends to mirror their position on trust, thus confirming that the project of reviewing the literature on trustworthiness was possible, despite an apparent lack of it. Subsection II.1.2 considered literature that takes trustworthiness to be a one-place relation – a virtue or character trait – revealing that truly one-place accounts are rare. Most apparently one-place accounts are in fact hybrid one/three-place, and as such will be liable to all the same problems that standard three-place views face. Subsection II.1.3 considered literature on three- place trustworthiness, which separates into strong and weak versions. Both versions, however, failed to make trustworthiness incompatible with intuitively untrustworthy motives. The final subsection, II.1.4, considered literature that aims to fix this problem with the addition (or preclusion) of various motivational elements as necessary conditions. Overall, Section II.1 confirmed that most philosophers consider trustworthiness to be a three-place relation, or a hybrid one/three-place relation: it involves a truster doing what the trustee entrusts them to do, sometimes because of the sort of person that the trustee is, sometimes because of the sort of motivation that they have. Section II.1 has demonstrated that this research project makes an original contribution in its conception of trustworthiness as a complex two-place relation. Again, as in Part I, I suggest that the many disagreements about trustworthiness that this literature review highlighted, rather than being evidence that ‘definitions and intuitions differ’ and so there are multiple ‘kinds’ of trustworthiness, is instead evidence of the very complexity of the thing under
II.1 Trustworthiness: in the Philosophical Literature 156 investigation, with most philosophers correctly identifying a part of this inherently complex and contradictory facet of our lives.
II.2 Dominant Views on Trustworthiness: Problems 157
II.2 Dominant Views on Trustworthiness: Problems
This section further explores the limitations of framing trustworthiness as a one, or as a three-place relation. Though some limitations were already discussed in the previous Section, as my primary aim in Part II of this research project is to change the dominant view from one in which trustworthiness is conceived of something that we do, or something we are, to something that we have in relation to a specific other (the truster), additional arguments are provided here. To that end, subsection II.2.1 explores problems with one-place trustworthiness. Since one-place trustworthiness is most often understood as a virtue (see: II.1.2), I address my discussion to the idea that trustworthiness is a moral notion, though the primary argument, that trustworthiness cannot be one-place because it takes an object, applies to the proposal that trustworthiness is a non-moral character trait with equal force. Together, subsections II.2.2 and II.2.3 explore the limitations of three-place trustworthiness, namely, all versions of three-place trustworthiness cannot handle the complex realities of trust and trustworthiness as they actually present themselves in human relationships.
II.2.1 One-place trustworthiness: virtue
On a personal note, when I began this research project, I was sympathetic to a virtue account of trustworthiness, since virtues, many have recognised, are complex, in the sense that they encompass many aspects of ourselves:
A virtue is an excellent trait of character. It is a disposition, well entrenched in its possessor—something that, as we say, goes all the way down, unlike a habit such as being a tea-drinker—to notice, expect, value, feel, desire, choose, act, and react in certain characteristic ways. To possess a virtue is to be a certain sort of person with a certain complex mindset. A significant aspect of this mindset is the wholehearted acceptance of a distinctive range of considerations as reasons for action. (Hursthouse and Pettigrove 2016)
This description of virtue shares some striking similarities with what I have said already in Part I about trust: it ‘goes all the way down’, involves a ‘complex mindset’, engages our reason and emotion, and is ‘well entrenched’. Moreover, as was touched upon in II.1.4, and will be argued for further in Sections II.3 and II.4, trustworthiness requires us to have a certain motivational landscape, rather than a single precise motivation, and
II.2 Dominant Views on Trustworthiness: Problems 158 this is something which is often held to be true of virtue (Slote 2001: 99-100; Zagzebski 2004: 154-60). Still, although there is much to a virtue account that fits with trustworthiness as it is presented in this dissertation, two significant irreconcilabilities remained: virtues do not take an object, and when taken to involve phronesis or practical wisdom, virtues result in action that is, morally speaking, the right thing to do. When I say that virtues do not take an object, or that virtues do not discriminate, what I mean is that – at least in their ideal forms – virtues are one-place. If we are honest then this is a description of the sort of person that we are (or aim to be), to all (or most) people and in all (or most) circumstances. Though we may fall short of the ideal, virtues are not ‘pegged’ to specific others in the same way that trust and trustworthiness are. Because of this, unlike virtues, trustworthiness can compel immoral action. We often use ‘trustworthy’ as a synonym for ‘virtuous’ in everyday conversation, and, trustworthiness so often calls upon our virtue, so often requires that we embody or act in ways that are morally praiseworthy, or which have ethical consequences, that it is easy to see why one might be convinced of the idea that trustworthiness itself must be morally good. This reflects the situation in the trust literature, where ‘trust’ is often used as a synonym for ‘rely’, and this, coupled with the fact that trust often causes reliance, leads us to confuse trust with reliance (see: I.3.3). Indeed, a significant number of philosophers writing on trust speak as though it were uncontentious that trustworthiness is virtuous, as we saw in II.1.1. Many who do so, take trustworthiness-qua-virtue to involve doing what we have been entrusted to do, interpreted as a form of moral action. As Paul Faulkner has recently explained:
trustworthiness is more than reliability in that it is evaluative as well as descriptive. The trustworthy act, like the opportunistic one, is done in response to someone’s depending on one in certain ways, but unlike the opportunistic act, it is the appropriate thing to do – where “appropriate” here amounts to a quasi-moral evaluation. (2017: 110)
Yet the idea that trustworthiness really does have some kind of intrinsic moral value has been the subject of some criticism, as Faulkner himself has noted (2017: 117). The most obvious criticism, which Philip Nickel pointed out over a decade ago, is that
II.2 Dominant Views on Trustworthiness: Problems 159 trustworthiness flourishes among the non-virtuous just as fiercely as it does among the virtuous. As Nickel put it, this is similar to how “the bad do love selflessly their own kind” (2007: 309). Recently, Karen Jones has elaborated on this theme, arguing that the “norms of trust and trustworthiness” have the potential to be in conflict with morality (2017: 102). She says:
You can be perfectly trustworthy with respect to some people without being a good person; indeed, being trustworthy with respect to some people requires [her emphasis] that you not be a good person… What can be demanded of friends is not subject to a moral filter and morally bad people can be good friends… Being trustworthy might require you to do the morally wrong thing and there may be no broader perspective from which to adjudicate the conflicting normative perspectives to arrive at a judgement as to what, all things considered, is the thing to do. (2017: 102-07)
So, while Jones and Nickel observe that trustworthiness cannot be virtuous because trustworthy people can do immoral things (and do so in the name of trustworthiness), Stephan Darwall provides us with a different reason to reject a moralised view of trustworthiness. For Darwall, trustworthiness (again like love) cannot be a moral attitude, as even in a friendship it is “nothing we can claim from someone or hold her accountable for” (2017: 47). In demonstration of this, Darwall points us to the difference between trust/worthiness, and promising. He says:
While it is possible to accept someone’s promise without trusting that the promise will be kept, it is not possible to accept an invitation to trust without trusting. Nothing else could count as accepting such an invitation. Unlike the relation between promiser and promise, that between trustor and trustee is not a(n impersonal) relation of right. It is a kind of personal relationship in which the trustor and trustee make themselves vulnerable to one another personally [his emphasis] rather than juridically. […] the kind of mutual recognition that is most distinctive of personal relationships like friendship and love is not intrinsically deontic or juridical. It is a mutual openness to forms of personal relation, including trust, that we have no authority to demand of one another, but that we nonetheless hope for and are understandably personally hurt or let down when it is not reciprocated. (2017: 46)
II.2 Dominant Views on Trustworthiness: Problems 160
On trust and trustworthiness as they are presented in this dissertation, Darwall’s observation that it is possible to accept someone’s promise without “trusting that the promise will be kept”, is taken to mean that it is possible to accept someone’s promise without believing that the promise will be kept, or without relying on the promise. This reframe though, is not vital to his point – unlike the relation between promiser and promisee – that the relation between truster and trustee is one we have no right to demand of one another. This is partly because, as complex two-place relations involving beliefs and emotions, trust and trustworthiness are not wholly voluntary (see: I.6.4), but it is also because they are historically determined, and intrinsically special. If we accept them, then together the above observations contributed by Nickel, Jones, and Darwall should lead us to strongly doubt that trustworthiness is a virtue. To these, this dissertation adds a related observation: unlike the virtues (and unlike non- moral character traits), trustworthiness takes an object, trustworthiness discriminates. As the argument here contains what should be by now familiar ideas and relatively uncontroversial claims, I shall make it brief. This dissertation has already stated its case (see: I.3) that trust discriminates, or ‘takes an object’ (the trustee). Trust is, at the very least, a two-place relation, and – given the symmetrical nature of trust and trustworthiness (see: II.1.1) – we can expect trustworthiness to be two-place as well, taking as its object the truster. Yet it is worthwhile exploring this proposal further here, with the added insight that the current contrast with virtue can bring. From section I.3, remember, in one-place accounts there is no object of trust/worthiness. Trustworthiness is just something trustees have, it is just something that they are, similar to a mood or temperament (B is trustworthy). If trustworthiness were one-place then it might well be a virtue, since virtues are one-place; virtues do not take specific things (individuals) as their objects. If we are honest or kind then this is the sort of person we are, expressed in the way we notice, expect, value, feel, desire, choose, act, and react in characteristic ways (Hursthouse and Pettigrove 2016). If I only display kindness towards my friends, or towards people of my own cultural background, then I do not possess the virtue of kindness. We cannot attribute a virtue to an agent based on a single observed action or even a series of similar actions (Sreenivasan 2002).
II.2 Dominant Views on Trustworthiness: Problems 161
This does not mean that to possess a virtue we must exhibit it infallibly. Possessing a virtue is a matter of degree, and most of us fall far short of the ideal (Athanassoulis 2000). But if we show kindness predictably in relation to some people and not others then this is not kindness but favouritism. If we are gentle with the cat but kick the dog, then we are not gentle, but instead cruel (and fond of cats). If the King seeks fairness and justice only for his companions, shirking fairness and justice otherwise, then he cannot be said to be a fair and just ruler. Virtues can be weak, and akrasia, or a lack of control, may preclude them from being exercised appropriately on occasion, but they do not – or at the very least aim not – to discriminate in these ways. At least in their ideal forms, virtues do not take particular objects but apply indiscriminately, whoever happens to be involved. The virtuous king is honest, benevolent, fair etc. to all his subjects, the virtuous nurse to all his patients, the virtuous teacher to all his students. Yet, unlike the disturbing idea that my virtue might apply discriminately, there is nothing inherently upsetting about the idea that my trust applies discriminately. I can trust only my doctor and not this doctor, my sister but not my brother, the dog but never the cat. The discerning nature of my trust does not seem to undermine it, in fact, it seems to be one of its most distinguishing features. The reason we trust some people and not others, is – as we all know – because only some people are worthy of our trust. Trust discriminates, because trustworthiness discriminates. Like the virtues, trustworthiness is a ‘well-entrenched complex mindset’, involving reason and emotion, which goes ‘all the way down’. But unlike the virtues, both trust and trustworthiness are ‘pegged’ to a specific individual. This leads to both epistemic and practical consequences. It means that you and I can disagree about who is worthy of our trust and both be right. Your dear old school friend Simon is worthy of your trust: you could ring him up in your hour of need and he would heed the call. But your dear old school friend Simon is not worthy of my trust. Though I am sure, as you say, he is a good person, I have never met him. If I were to ring this stranger up and disclose my deepest darkest secrets or beg him to advocate for my innocence, he would probably hang up on me. Where discrimination leads us to doubt a person’s virtue, a lack of discrimination leads us to doubt a person’s trust and trustworthiness. Consider the person who treats
II.2 Dominant Views on Trustworthiness: Problems 162 everyone as their best friend: confiding in strangers, becoming emotionally attached to work-colleagues, expecting loyalty from outsiders, reacting with resentment when their lofty relationship expectations are not matched. We know this person as one who lacks mastery of trust and trustworthiness. We identify them instead as naïve, gullible, troublesome. Insensitive to the time, history, attention, care – specialness – that genuinely close interpersonal relationships require, such a person is potentially dangerous, or else, they are dangerously vulnerable. That trust and trustworthiness are discriminating, discerning, selective, does not entail that we regard those we do not trust, those who are not worthy of our trust, with suspicion or hostility. Positive everyday interactions with most people do not require trust and trustworthiness but are very well facilitated by nothing more than a general optimism and openness, and an adherence to the relevant social and cultural norms. It is vital, as we shall see (II.3), that mastery of local social/cultural norms are not confused with trust and trustworthiness. Just as not all those who can love and be worthy of it are skilled in the relevant social/cultural norms associated with romance/courting, so not all those who can trust and be worthy of it are skilled in the relevant social/cultural norms associated with reliance/reliability. In II.3 it is argued that we must not deny trust/worthiness (love/worthiness) to those who struggle with the local behavioural norms surrounding them. Though trustworthiness is not a virtue, the interplay between trustworthiness and morality is intimate, and interesting in its own right. Trustworthiness interacts with the moral realm in several ways, for example: 1) by requiring virtue or moral action, 2) by introducing genuine moral conflict into our lives, 3) by introducing struggles between morality and trustworthiness, and 4) by changing moral dilemmas into trustworthiness predicaments, with ethical consequences. It is beyond the scope of this research project to discuss all the ways in which our trustworthiness intersects with our morality in detail, though, before moving on, I shall elaborate on 1 to 4 very briefly. 1) Our trustworthiness can require morality, as the ability to sustain close interpersonal relationships often demands that we exhibit virtue, and engage in moral reasoning, moral action. Lacking a good moral compass can damage trusting relationships just as surely as anything can. Yet trustworthiness can also require that we
II.2 Dominant Views on Trustworthiness: Problems 163 exhibit a vice, as Jones says (2012: 84), and as I shall explore further in the remainder of II.2. If it is a requirement of trustworthiness that we possess a moral sensibility, then it is also a requirement of trustworthiness that we know when to ignore that moral sensibility. 2) Our trustworthiness can introduce moral struggles between virtue and vice, courage and fear, generosity and self-interest, into our lives. When this happens, our trustworthiness itself is not involved directly in the struggle. In this role, trustworthiness is an initiator, a catalyst. Analogously, there is a pressing moral choice that requires a captain’s decision. The captain, and only the captain, has the authority to make this moral choice. The role of trustworthiness, like the role of captain, can introduce moral dilemmas, moral choices, and moral consequences into our lives that we cannot ‘fob off’ onto anybody else. 3) Our trustworthiness itself can conflict with our morality. When this happens, we find ourselves in a struggle between the pressures of trustworthiness – the pressures of interpersonal relationships – and the pressures of morality. Again, analogously, the captain may be put in a position of having to decide between doing the right thing, ethically speaking, and following the admiral’s orders. The right thing to do is obvious and uncontroversial. The decision she needs to make is not between right and wrong, but between morality and the captain’s burden. Like this, our trustworthiness can place us in situations where we know what we ought to do, but this does not settle for us what to do. Trustworthiness is a force at least as strong as morality, and their winds can blow in opposing directions. 4) Our trustworthiness has the power to help us decide on moral dilemmas, by changing them into trustworthiness predicaments with ethical consequences. The captain must decide between saving two crew members, ensign Reicher and ensign Kim (she cannot save both). These kinds of situation are identified as symmetrical moral dilemmas (Sinnott-Armstrong 1988: Ch.2), the quintessential example being Sophie’s Choice (Styron 1979). The bald ethical question underlying our captain’s choice is– given that I can save only A or B, ought I save A or B? Obviously, this question is hopeless, and provokes no moral intuitions– a true dilemma. Our captain’s two-week
II.2 Dominant Views on Trustworthiness: Problems 164 intensive ethical training module undertaken back at the academy cannot help her here. She might as well go ahead and resort to eeny, meeny, miny, moe. Mercifully though, in real life, things are not usually so devoid of any relational content. Hopeless symmetrical moral dilemmas are rare. In real life, it is not A and B, it is our sister and our brother, our friend Susan and our friend Sarah. The choice is between two different individuals with whom we share two different relationships. Because no two relationships are exactly alike, no two instances of trustworthiness are exactly alike: there is no such thing as a ‘symmetrical trustworthiness dilemma’ as there are symmetrical moral dilemmas. Sometimes, when we are in a symmetrical moral dilemma, we can use the trustworthiness difference to make the choice. Our choice can be the right choice, maybe not ethically speaking, but ‘trustworthily’ speaking. Whatever we choose, we will walk away with dirty hands, but that was always going to happen anyway. Moral dirt just gets everywhere. This subsection began on a personal note, and so I shall end it on one. My personal hope is that instead of viewing trustworthiness as a virtue, as a moral imperative, the curious manner in which our close interpersonal ties to one another can shift, shape and break our morality will be explored more by philosophers in the future. Iris Murdoch said “anything which alters consciousness in the direction of unselfishness, objectivity, and realism is to be connected with virtue” (1997: 104). To be trustworthy is not to have our consciousness altered in the direction of selfishness, subjectivity, or illusion, but it is to have it altered in the direction of those few special others who have (subjective) significance for us. Yet though they face their imperfect objects and not some ideal morality, it is in our acquaintance with trust and trustworthiness, with loving and being loved, that we learn the first and the hardest moral lessons. Trust and trustworthiness are where our moral selves begin. With them we acquire care, forgiveness, gratitude, and hope. Trustworthiness is not virtue, not the mountainous heights of our morality. It is the sun and rain on the plains of our moral selves: flooding and scorching, but ultimately, giving life to our moral landscape To summarise: philosophers have made important contributions regarding the ways in which trustworthiness conflicts with morality: trustworthiness can require that we take immoral action, and it is not something we have any right to claim from anybody
II.2 Dominant Views on Trustworthiness: Problems 165 else. To these, this dissertation has added a related observation, that unlike the virtues, which are (or aim to be) one-place, trustworthiness discriminates, it takes an object. That said, trustworthiness and trust interact with morality, with our moral selves, in several interesting ways. Four of these ways, were briefly outlined.
II.2.2 Three-place trustworthiness: scenarios
To recapitulate, strong three-place trustworthiness is the view that a person is trustworthy just when they do what they have been entrusted to do, and a person is untrustworthy just when they fail to do what they have been entrusted to do. Weak three-place trustworthiness is the view that a person is trustworthy just when they are competent and committed to doing what they have been entrusted to do and are untrustworthy when they are either incompetent or are not committed to doing what they have been entrusted to do, or both (see II.1.3). Various versions of three-place trustworthiness also require the trustee have, or lack, certain motives (see II.1.4). The aim of this section is to provide some good reasons for thinking that all versions of three-place trustworthiness cannot handle the complex realities of trust and trustworthiness as they actually present themselves in human relationships. To accomplish this aim, this subsection introduces two ordinary characters: Rasmus, playing the part of the trustee, and Iris, playing the part of the truster. These two characters are placed into ten different scenarios. In each scenario, certain facts remain the same: 1) Iris and Rasmus are having a telephone conversation. 2) Iris asks Rasmus where he is: “Where are you?” she says. 3) Rasmus is in a pub named The Prancing Pony. 4) Iris ‘trusts’ Rasmus to tell her the truth, namely, that he is in The Prancing Pony. According to strong three-place trustworthiness, Rasmus counts as trustworthy only in those scenarios in which he tells Iris that he is in The Prancing Pony. According to weak three-place trustworthiness, Rasmus counts as trustworthy only in those scenarios in which he is both competent and committed to telling Iris that he is in The Prancing Pony. According to motive accounts, Rasmus is trustworthy when he, in addition to doing or being competent and committed, possesses the motive relevant to the account: goodwill, a lack of selfishness, etc. While these four basic facts remain the same, in each scenario additional contextual information will be introduced. As this
II.2 Dominant Views on Trustworthiness: Problems 166 happens, all versions of three-place trustworthiness begin to fail to consistently secure the intuitively correct result regarding Rasmus’ trustworthiness.
Scenario 1 Selfish Lie Rasmus is in The Prancing Pony. Iris calls to ask him where he is, as she is hoping to meet up with him for a drink. Rasmus, cowardly and not as keen on Iris as she is on him, lies, saying that he is in another pub, The Dancing Dolphin. Iris goes to The Dancing Dolphin expecting to meet Rasmus, only to find that she has been stood up. In Scenario 1, Rasmus did not do as he was entrusted to do, and though we assume he could competently have done so, he was not committed to doing it. He also acted selfishly, spinelessly avoiding the awkward encounter. Rasmus comes out as untrustworthy in both the strong and the weak versions, and any version of three-place trustworthiness that requires him to have unselfish/benevolent motives. So far, so good.
Scenario 2 Honest Mistake Rasmus is in The Prancing Pony. However, having been given false information by a sober-enough and friendly-seeming patron, he mistakenly believes himself to be in The Dancing Dolphin. Rasmus is desperately trying to meet up with his dear old friend Iris, but alas, he unwittingly conveys his false belief to her – an honest mistake. When they eventually figure out the error and find one another, these two old friends share a good laugh, and there ae no bitter accusations of betrayal or earnest appeals for forgiveness. Counterintuitively, strong three-place trustworthiness renders poor Rasmus untrustworthy in Scenario 2, since according to the strong version, remember, Rasmus is worthy of Iris’ trust only when he actually succeeds in doing what she entrusts him to do. Between Scenarios 1 and 2, we see how strong three-place trustworthiness is unable to distinguish between someone who is cowardly and deceitful, and someone who is just excusably confused. On the weak version, Rasmus is certainly committed to telling Iris where he is, so he satisfies that condition, and we assume that he can normally tell people accurate information about his location even though he failed to do so this time. So here in Scenario 2, we see the weak version successfully doing what it is meant to do, rendering Rasmus trustworthy despite his honest mistake.
II.2 Dominant Views on Trustworthiness: Problems 167
Scenario 3 Recurrent Inattentiveness As in Scenario 2, except that Rasmus has a habit when he is drinking of getting preoccupied chatting and often fails to bother to get his facts straight. Rasmus is both competent and committed to doing what Iris entrusted, and he had no selfish or ill motives towards Iris. However, this is not the first time his inattentiveness has meant that Iris has ended up spending half her night going from pub to pub trying to find him because he provided her with false information. Understandably, she is beginning to feel resentful and wonders if he really does care about her at all. Rasmus fails in this instance to be strongly three-place trustworthy, and some may hold the intuition that this is the right outcome. However, he may end up being weakly three-place trustworthy (a result that may unnerve us), since he has both commitment, and competence. One benefit of strong three-place trustworthiness is that – because it condemns every failure as untrustworthy – it never accidently lets off any guilty party. Weak three-place trustworthiness, though it saves some innocent versions of Rasmus from the unfair charge of untrustworthiness as we saw in Scenario 2, can fail to condemn some guilty versions. Moreover, requiring Rasmus to have benevolent motives underlying his behaviour, in this instance did not work to get us to the right outcome. This is because a trustee’s motives are not all that trusters care about. Our relationships have history, and that history partly determines the weight of any current transgression. Rasmus has been letting Iris down in small ways for years. Without reference to the history between two people, Rasmus’s motives in a single-instance snapshot cannot tell us all we need to know about his worthiness of Iris’ trust.
Scenario 4 Accidental Trustworthiness In Scenario 4, Rasmus tells Iris the truth… but only because he believes himself to be talking to someone else. Iris is calling from their mutual friend Fia’s phone. Rasmus has historically been a diligent, caring, and reliable friend to Fia, and his commitment to telling her that he is in The Prancing Pony is right now wholly benevolent. However, The Prancing Pony is a busy and loud establishment and Rasmus does not realise that he is actually speaking with Iris – a girl he loathes and would have lied to if he had known he was speaking to her.
II.2 Dominant Views on Trustworthiness: Problems 168
Counterintuitively, in Scenario 4, Rasmus is trustworthy on all accounts. As discussed already in relation to three-place trust (I.2.3), three-place trustworthiness is, like the Gettier cases, subject to a form of ‘accidental success’, where we want to say that a person fails to be trusting/trustworthy to their trustee (Iris), but our account leaves us with no way to do it.
Scenario 5 Benevolent Lie As in Scenario 1, Rasmus has no commitment to telling Iris the truth at all. Again, he outright lies, this time saying that he is still at work when he is in fact drinking merrily at The Prancing Pony with a group of their mutual friends. Yet, unlike in Scenario 1, in Scenario 5 Rasmus has a good reason for lying. Iris’ friends have all met up in secret this evening with the express purpose of organising a surprise birthday party for her, planned for the following weekend. Rasmus has failed to tell the truth and was never committed to it, so in Scenario 5 he is unjustly rendered untrustworthy on both strong and weak three-place trustworthiness. Sadly, the exonerating reasons behind his lack of commitment seem to make no difference. Specific motivations are requirements added to three-place trustworthiness; being unselfish or benevolent does not serve to make a person worthy of trust according to three-place accounts if they were never even remotely committed to doing what the truster entrusts. This is highly counterintuitive, since Rasmus’s failure to do what Iris is entrusting him to do, far from making him unworthy of her trust, seems to be an integral part of what it is to be a decent trustworthy friend to her.
Scenario 6 Misidentified Untrustworthiness In Scenario 6, as in Scenario 5, Rasmus lies to Iris when she calls and asks him where he is. Only this time, he went along to the surprise-party-planning-gathering to get all the details so that he could later ruin the happy surprise. Rasmus, rubbing his hands together in wicked glee, intends to wait for the most devastating moment to let slip to Iris about the surprise party in order to get back at her for some imagined slight. Here, both strong and weak three-place trustworthiness positively identify Rasmus as untrustworthy. This is some small victory. The problem is, neither version can
II.2 Dominant Views on Trustworthiness: Problems 169 identify the reason why Rasmus is not worthy of Iris’ trust. The strong version tells us that Rasmus is untrustworthy because he failed to do, and the weak version because he was not committed to doing, what Iris entrusted him to do. Yet Rasmus is untrustworthy not because he lied to Iris (her true friends did that), nor because he was uncommitted to truth (again, neither were her true friends), but because he is out to get her! If Rasmus is lying with malicious intent, then the strong and weak versions might succeed in rendering him untrustworthy (which is what we want) but for the wrong reason.
Scenario 7 Manipulative Honesty In Scenario 7, Rasmus finds himself at The Prancing Pony this evening on a disastrous blind date his mother has set up for him. Rasmus excuses himself from the table and shares this mortifying truth with Iris when she unexpectedly calls for a chat. Iris thinks her dear sweet friend Rasmus is confiding in her. Yet she could not be more wrong. Rasmus despises Iris and wants her fired from their mutual workplace (preferably with considerable humiliation). He has reckoned that the best way to make this happen is by befriending her, so that he can learn some incriminating facts about her that he can use against her in the future. His openness, honesty and friendliness towards Iris are false… a ticking time bomb… The strong version of three-place trustworthiness can find no fault with Rasmus here, yet neither can the weak version, since Rasmus is both competent and committed to doing what Iris entrusts him to do… for now. If we require Rasmus’ actions to be underpinned by benevolent motivations, then we may potentially secure the right answer in Scenario 7, but as we have already seen, these additional motivational requirements will not fix things in every case (Scenarios 3,4,5), and, there are others…
Scenario 8 Untrustworthy Truster In Scenario 8, Rasmus is Iris’ employee, and, having worked hours of unpaid overtime every night this week, he has left the office a few minutes early and headed to The Prancing Pony to enjoy a well-deserved pint. When Iris calls to ask him where he is, Rasmus desperately wants to lie, yet he is utterly terrified of Iris and has not the guile to do it. Iris is not a trustworthy truster. She does not engender genuinely loyal, honest,
II.2 Dominant Views on Trustworthiness: Problems 170 caring employees. Her aim is unquestioning obedience, and she secures ‘trustworthiness’ with fear-tactics. Unable to discriminate between an unwilling subordinate and a willing trustee, both strong and weak three-place trustworthiness render Rasmus trustworthy in Scenario 8. And, since his motivations for telling Iris the truth were not malicious, here again, as in Scenarios 3, 4, and 5, the addition of this motivational requirement makes no difference. In Scenario 8, all versions of three-place trustworthiness result in the wrong outcome. The outcome is wrong, because Rasmus is not trustworthy here, but compliant. Cruel masters cannot trust their slaves, which is why they need their iron fists. Scenario 8 highlights two important things. 1) When evaluating trustworthiness, we consider the counterfactual: would Rasmus have lied, if he could have brought himself to it? Such counterfactuals, though, are not considered relevant in any accounts of trustworthiness I have seen; 2) Trust and trustworthiness are qualities of relationships; the truster’s intentions, personality, and habits, i.e. their trustworthiness can disturb the trustee’s worthiness of trust too.
Scenario 9 Moral Trustee In Scenario 9, Rasmus is largely indifferent to Iris. He does not know her very well and means her neither good nor ill will. Moreover, he is a conscientious person, and has thought long and hard about right and wrong. He has made diagrams, done moral calculations, read all the great philosophers, and has come up with a rule – always tell the truth so long as the benefits of doing so are not outweighed by the costs. This time, circumstances were such that his moral calculations happened to be in favour of doing what Iris entrusted him to do, so, that is what he did. Rasmus, fully competent and genuinely committed, with no ill or selfish motives, did what Iris entrusted him to do. As such, according to all versions, he is trustworthy. However, had his moral calculations come out differently, Rasmus would have lied, and (this is the important bit) he would have done so no matter what the consequences of his lie would have been for Iris. Although Rasmus can be ‘trusted to do the right thing’, can Iris trust him? I think not. A devout consequentialist like Scenario 9 Rasmus will do the
II.2 Dominant Views on Trustworthiness: Problems 171 right thing by their trustee very often, but the astute truster knows that friendships with devout consequentialists (devout anything) are potentially dangerous.
Scenario 10 Social Norm-follower Finally, in Scenario 10, Rasmus does not much like Iris – he finds her boring – but again he tells her the truth, harbouring a general sense of human goodwill towards her. But this time it is not because he thinks it is the right thing to do, but simply because it is what people do. Rasmus is not a morally virtuous person. He is a common or garden variety social chameleon, and he has found himself living among people who are for the most part friendly, polite, helpful and prone to telling the truth unless there is a good reason not to. When Iris called and asked him where he was, he reflexively told her… because… that’s what people do. Rasmus did, and was competent and committed to doing, what he was entrusted to do, and he lacked any ill motive. All versions render Rasmus trustworthy in Scenario 10, and some might agree with this assessment. Still, my sense at least is that Scenario 10 Rasmus, though he is not made untrustworthy by these actions, has not done enough to deserve trustworthiness, since his doing what Iris entrusted him to do had nothing to do with her. He was just behaving normally, according to the behavioural norms of his society. Would those norms have been different, if his society underwent a social shift, he would have behaved differently. In summary, it is surprisingly easy to produce scenarios that draw out the seemingly endless limitations of three-place trustworthiness. In addition to the ten presented, there are more complex Gettier-like scenarios in which Rasmus tells the truth but only by accident; mistaken identity scenarios in which Rasmus has no ill-will but only because he has wrongly believed the person before him is his friend; moral dilemma scenarios where Rasmus lies but only because telling the truth would betray someone else; mental illness/confusion scenarios where Rasmus harbours ill-will that is wholly irrational; truster-fault scenarios where Iris is to blame for Rasmus’ failure; compulsion scenarios where Rasmus does what Iris entrusts but did not seem to have any choice in it; unavoidably self-interested scenarios where doing what Iris entrusts benefits Rasmus. As this subsection demonstrated, when we actually apply three-place trustworthiness to
II.2 Dominant Views on Trustworthiness: Problems 172 the complex linguistic, ethical, social, cultural, technological, emotional, interpersonal, and historical conditions under which human beings operate in and weigh their dealings with one another, all versions of three-place trustworthiness routinely produce counterintuitive and conflicting results.
II.2.3 Three-place scenarios: discussion
Subsection II.2.3 discusses how the failure of three-place trustworthiness to handle the everyday contextual complexities presented in the previous subsection, provides us with reason to doubt that it can do the job of accounting for trustworthiness as it actually exists in human relationships. Regarding the proposition that the concept of trustworthiness should instead be framed in such a way as to provide a wide-scope explanatory device suitable for use in multiple areas of academic inquiry, this dissertation once again suggests that there are other concepts better suited to this task. What we need is an account of trustworthiness that can differentiate between what we do, and who we are in relation to special others. Such an account would have no necessary and sufficient conditions, no particular motivational requirements, no set formulae. It would amount to a discussion about trustworthiness, not a definition. This Subsection presents a handful of philosophers whose intuitions diverge significantly from my own as I have offered them in the previous subsection, with the aim of shedding some light on the distinct nature of the phenomena each of us is speaking to. Before proceeding with this though, it is with regret that I feel I must defend my approach. Recently, in the philosophy on trust literature, it has become popular to shy away from thought-experiments such as the one provided in II.2.2, precisely since ‘intuitions differ’ about them. Those who have made arguments along these lines include Hardin (2006), Faulkner (2017), Jones (2017), Nickel (2017), and T. W. Simpson (2012). Instead of inquiring as to what trust and trustworthiness are, the current fashion is to ask what role we want the concepts of trust and trustworthiness to play in our conceptual scheme. For example, Paul Faulkner provides an account, not of trust and trustworthiness as such, but of “our understanding of these notions” (2017: 117), while Karen Jones provides a “conceptual role argument” that “begins by looking at the normative point of
II.2 Dominant Views on Trustworthiness: Problems 173 having such concepts”. She asks “what conceptual role do we want [trust and trustworthiness] to do?” (2017: 99). Russell Hardin says that the only valuable account of trust and trustworthiness is the one which can yield “explanations of behaviour and social institutions” (2006: 16), and Philip Nickel advises that we should grade accounts of trust based on their “explanatory potential”, where those accounts that “explain” more phenomena being more highly rated (2017: 195-213). Lastly, Thomas Simpson contends that because of the multiplicity of definitions of ‘trust’, we cannot say that trust/worthiness is this, since counterexamples can always be found. He says that all we can do with a notion like trust is to give a genealogical account of what is presumably a “family of concepts” sharing a “single ancestor” which he identifies as cooperation (2012: 551). Philosophy, as it is understood in this dissertation, is the business of asking what there is (and how do we know that we know it), and not asking what we want there to be (and how do we define it). Whether or not this dissertation succeeds, it has set out to say something of value about trust and trustworthiness and takes this project to demand the examination of human relationships. Though Rasmus and Iris are imaginary, they have not been forced into any incredible ‘purely philosophical’ (meaning purely imaginary) thought experiments. There are no zombies, brains in vats, homunculi, or wretched little girls locked away in black-and-white rooms. They are ordinary people doing ordinary things such as accidently providing false information, telling little lies to conceal surprises, and engaging in moral reasoning and social norm-following. What our intuitions tell about how these contextual influences affect their worthiness of trust, we can take to tell us something about what sort of thing trustworthiness is, and, where our intuitions diverge can tell us where our assumptions diverge, not where trustworthiness branches off into another kind. For example, Domenicucci and Holton (2017: 152) discuss a scenario similar to the ones I have presented, taken initially from Hieronymi (2008b). This scenario involves Jim and Jules, who have agreed to meet at a restaurant. Jules thinks – quite unreasonably – that Jim will not show up. Domenicucci and Holton hold the intuition that if Jim does show up, then he could rightly complain that Jules’ lack of confidence betrays a lack of trust. Yet if Jim does not show up, then, according to them, he forfeits
II.2 Dominant Views on Trustworthiness: Problems 174 his right to the complaint (2017: 154). Here, my own person intuitions diverge from theirs. By my lights, a trustee’s entitlement to the complaint that they are not really trusted depends always on why they failed to show up, not whether or not they actually did. If Jim’s failure to show up was not his fault, then Jules’s complaint that the failure signals a lack of trustworthiness is undeserved. Once we get a handle on the different intuitions that such thought experiments draw out, our assumptions become clearer. In this case, I hold the intuition that I do because I assume that our trustworthiness is not negated by events well beyond our control. Once this assumption is clear, reasons for thinking that it is sound can be provided, such as: in the actual world, in our actual interpersonal relationships, events beyond our control tend not to strip us of our worthiness of trust. Jim loses the entitlement to complain about Jules’s lack of trust only if his failure was the result of a lack of trustworthiness, not when it was the result of, say, sudden illness or accident. The fact that there is a difference between Domenicucci and Holton’s intuitions and my own, reflects our different assumptions about trustworthiness, and, as this dissertation has argued, the complex reality of trustworthiness, and not the reality that there are “many kinds of trustworthiness”. It is not evidence that we should give up the game and just decide what we want trustworthiness to be. It is an indication that one of us is right about trustworthiness and one of us is wrong (or that neither of us knows what we are talking about). The point is, such disagreement is valuable. If your own intuitions about my ten scenarios differ from those presented here, then good: tell me why, and we shall be doing some philosophy. Moving on, given the inability of strong three-place trustworthiness to secure (what I take to be) the correct result in so many of our ten scenarios, those whose intuitions agree with mine might wonder why it exists as an account of trustworthiness at all. As the Danish philosopher Knud Ejler Løgstrup put it, trustworthiness cannot be “merely a matter of fulfilling the other person’s expectations and granting his or her wishes” (1997: 7) as the strong version has it. In answer to this, this dissertation has already provided a historical explanation for the existence of strong three-place trustworthiness stemming from the way the word ‘trustworthiness’ has been used synonymously with ‘reliability’ in academic discourse in the philosophy of science (see: II.1.3).
II.2 Dominant Views on Trustworthiness: Problems 175
Yet even in the most recent (2017) philosophy on trust literature, strong three-place trustworthiness has a number of active proponents. Edward Hinchman has said that in all cases other than the special case of promissory trust, trust cannot be what he calls “disappointed yet unbetrayed” (2017: 59-63). In our language, Hinchman takes it that a trustee cannot fail to do what they have been entrusted to do and still be worthy of trust, since failure quite literally constitutes betrayal. Giving an example of a trustor (Andrew) entrusting the safekeeping of a leaf to a trustee (Bernice) he says:
Could trust be disappointed yet unbetrayed? No. If Bernice disappoints his trust by failing to produce the result – keep the leaf safe – that Andrew is relying on her to produce, then she betrays his trust. (2017: 59)
For Hinchman, success is a necessary condition for trustworthiness. He qualifies this with a further necessary condition that a trustee must “manifest concern” for the needs of the trustor. Yet, he clarifies that this concern “need not be felt”, only make us “disposed to act” (like automaton) as if we are concerned (2017: 59). He says:
Two conditions are individually necessary and jointly sufficient for unbetrayed trust: (i) the trusted must actually produce the result that the trusting is relying on her to produce, and (ii) the trusted must do so in a way that manifests appropriate responsiveness to a subset of the trusting’s needs: specifically, those needs recognition and acknowledgement of which inform the agreement at the core of the trust relation. (2017: 60)
If we apply Hinchman’s framework to Scenarios 6 & 7 Rasmus – the Rasmus who ruined the surprise party and the Rasmus who is trying to get Iris fired – then, both Rasmuses are trustworthy. While Scenario 2 Rasmus – the Rasmus who is understandably lost – is not trustworthy. In fact, Hinchman’s account finds trustworthiness where there is none, and untrustworthiness where there is trustworthiness, in every single one of our ten scenarios. Disagreement with regard to outliers or ambiguous cases is one thing, and this subsection began by explicitly welcoming the clarity it can bring. But Hinchman’s intuitions diverge from my own to such an extent that we simply cannot be speaking of the same phenomenon. The question is, which one of us has identified trust/worthiness, and which one of us is speaking to something else (assuming one of us has got things more or less right). By
II.2 Dominant Views on Trustworthiness: Problems 176 my own reasoning, if Rasmus’ failure was innocent, honest, or for Iris’ own benefit, or if his success was malicious, underhanded, or part of some greater scheme to harm her, then such details must make a difference to his trustworthiness according to whatever account we adopt, again, since such details would make a difference to whether or not we trusted Rasmus if we were Iris. We may still rely on him, take his word as true, or cooperate with him, but such acts can be accomplished in the absence of trust. Colin O’Neil, though, clearly sides with Hinchman:
A betrayal of trust violates an obligation to do specifically as trusted. If you trust me to pick you up, it is possible for me to betray your trust by failing to pick you up, even if I take steps to ensure that your interests are protected by `arranging another ride for you, or by giving you a timely warning. For this to be the case, the relevant obligation (in being trustworthy) must be to do specifically as trusted. (2017: 71)
He also says:
A betrayal of trust is only a failure to do as trusted. If you trust me to pick you up, but don’t trust me not to steal from you, then only my failure to pick you up can betray your trust. The obligation (again, in being trustworthy) must be exclusively an obligation to do as trusted. (2017: 71)
According to O’Neil, then, Rasmus does betray Iris if, for example, he successfully arranges for their mutual friend to call instead and tell her that they are in The Prancing Pony because his own phone is out of battery, but he does not betray her trust if he tells her (truthfully) that they are in The Prancing Pony but plans to murder her shortly after she arrives. In O’Neil’s view, our dear thoughtful friends who go out of their way to ensure someone else supplies us with correct information when they cannot provide it themselves are untrustworthy, but our murderers, if we are trusting them to murder us and they manage it, are worthy of our trust. Surely Hinchman and O’Neil, and others who adopt strong three-place trustworthiness, are speaking not of trust and trustworthiness, but of credibility, reliability, predictability, expectedness. In “Being Pragmatic About Trust”, remember (I.2.2), Philip Nickel recommends the adoption of an analysis of trustworthiness that
II.2 Dominant Views on Trustworthiness: Problems 177 makes no mention of the trustee or trustor’s thoughts, emotions, or intentions, so that the concepts may be more broadly applicable (2017: 195-213). Specifically, by removing private mental content that is difficult to determine empirically from the account, Nickel hopes the concept of trustworthiness will become useful to science. A compelling (though stale) idea, this is just what the radical behaviourist B.F. Skinner (1974) and his contemporaries tried to do mid-last century – though not with trust – with the whole of psychology. The notion that trust can be described and explained without making reference to inner mental events or internal psychological processes is straight out of the behaviourist textbook. As Wilfred Sellers said, all a person needs to be described as a behaviourist is that they insist on confirming “hypotheses about psychological events in terms of behavioural criteria” (1963: 22); this is just what proponents of strong three-place trust and trustworthiness are insisting we do. Thanks to the cognitive revolution, radical behaviourism is no longer how we do psychology. It should no longer be how we do philosophy. The result of stripping trust/worthiness of all their complex veiled humanness, is the lumping together of innocent and culpable failures, both moral and interpersonal, and identifying all of them as betrayals of trust. The upside of this, as we saw in Scenario 3, is that by condemning every failure as untrustworthy, we will not accidentally exonerate any guilty party. But the unacceptable downside is that we condemn many innocent parties and give away the ability to make an important conceptual distinction that we in fact do make in our dealings with one another in the actual world. Moreover, without considering the private inner world of the trustee, close counterfactual scenarios, and the history the trustor and trustee share, we severely limit what we can say about the nature of their innocence or culpability. In many instances, a trustee’s innocence or culpability will have nothing to do with their failure per se, and everything to do with why they failed or succeeded, or what has happened between them and the trustor in the past. Determining trustworthiness is no straightforward affair: it is more art than science, being not an investigative activity, but a creative one. Science has many other concepts at its disposal that are far more appropriate to the task than trustworthiness. Concepts that do not suggest any relationship: precise, accurate, predictable, certain, consistent, and of course reliable, to name a few. With so many
II.2 Dominant Views on Trustworthiness: Problems 178 concepts better suited, we need not make trustworthiness ‘valuable to science’. Instead, along with love, trust, and forgiveness, trustworthiness must remain one of the illustrative and clarifying devices of the humanities. These concepts, left vague and imprecise, can capture what is imprecise, complex, inconsistent, and contradictory. They can explain us to one another as they always have done. Not all those who favour a three-place account of course, favour the strong version. But as we have seen, ‘fixes’ to what is still a three-place account – requirements of benevolence, unselfishness, competence, or commitment – still often fail in common everyday examples. Unlike Hinchman and O’Neil’s accounts, Stephen Wright’s account, for example, allows that a trustee be able to disregard the trust that has been placed in them, when it is appropriate to do so, without losing their trustworthiness (2010: 622). Accordingly, in many of our scenarios in which strong three-place trustworthiness fails, Wright’s account does manage to secure the intuitively correct outcome for Rasmus. Yet because Wright’s account locates trustworthiness not in the consideration of the value of the trustee, but in the value of their trust (see II.1.4) it makes Scenario 7 Rasmus – the one who is out to get Iris – trustworthy, since he no doubt values her trust (albeit for underhanded reasons). Somewhat absurdly, what solves the problem in accounts which have specific motivational requirements in some cases, becomes the problem in others. For example, Wright’s account precludes a trustee from having any selfish motives, but again, though this condition helps sometimes to secure the correct result, in others, it leads us to preclude the trustee from being counted as worthy of trust on rather spurious grounds. Wright says we must include a no-selfishness clause in an account of trustworthiness since no rational person would end up at the position that “lazy self-interest would be more valuable than, or should take priority over, a trusting relationship” (2010: 623). While this dissertation is in agreement with Wright’s assessment that trusting relationships are always more valuable than lazy self-interest, the reality is that we regularly end up acting in a contrary manner, and often, it is our reason – our ability to rationalise – that got us there. Failing to do what we have been “entrusted to do” just does not equate to a failure to weigh the value of a relationship appropriately on most occasions. The reason why we do not weigh our relationships when we make decisions
II.2 Dominant Views on Trustworthiness: Problems 179 about what to do in most everyday circumstances is that we know our relationships are not in jeopardy in most everyday circumstances. Countless small failures of selflessness do not manage to strip us of our worthiness of trust (or for that matter our worthiness of love). What long-term relationship has not survived a letdown, has not survived the charge ‘but I was counting on you’? Trust and trustworthiness, ongoing complex experiences, in one way are the ability to endure such lapses and not to be made or broken by them. When a three-place account is adopted, even when it is tempered with motivational requirements as Wright’s account is, we always end up back at the same place with single-instance failures. Failures of benevolence or care, failures of selflessness or honesty, failures of competence or commitment, failures of morality or sociality, failures to do what we are entrusted to do – all are capable of destroying our trustworthiness. This dissertation is not suggesting that this never happens. Some acts are so egregious that they do, in an instant, destroy any chance we might have of remaining worthy of our trustee’s trust. Even when this happens though, it is not the act itself, but what the act reveals – an unforgivable lack of care. Our trustees are not transformed from worthy to unworthy if they put themselves first now and then, or if they do not always manage to texture our interactions with palpable goodwill (‘no, for the last time I do not know where your bloody socks are… you can bloody well go to school without them… I don’t care if you get bloody blisters… now get in the car!’). Here, I think, is a good point to say something about the practice, so connected to trust/worthiness, of demanding explanations. Demanding explanations of those we trust is a crucial intermediary step between our epistemic-update (‘you didn’t do what I entrusted you to do’) and our ‘trust-update’ (‘you betrayed me in an unforgivable way’, cue resentment, betrayal, grief, and withdrawal from the relationship). When we share history, share a relationship with someone, and they blunder, we pause, and we usually ask them why. When it comes to our trust-updates as opposed to our epistemic-updates, there is so much more to consider. One way we can frame the problem with three-place trustworthiness is that it is too set in the public world. It is interested only in what actually happened in the publicly-observable phenomena: did the trustee do what they were entrusted to do? Three-place trustworthiness does not permit the consideration of
II.2 Dominant Views on Trustworthiness: Problems 180 what might have come to pass. This is an unacceptable failing, as in determining trustworthiness, we are not just interested in what actually happened, as we are when we make epistemic updates. Trust does not pass judgement on so little information, but always asks the question “what was it you were trying to do”?
II.2.4 Summary of dominant views on trustworthiness: problems
Section II.2 highlighted the limitations of one-place trustworthiness and three-place trustworthiness. Subsection II.2.1 observed that unlike the virtues, which are (or aim to be) one-place, trustworthiness takes an object, and briefly outlined some ways in which trustworthiness and trust interact with morality. Subsection II.3.2 demonstrated that although various versions of three-place trustworthiness do manage to pick-out trustworthiness in some scenarios, they cannot do so consistently, or without error. In II.2.3, I defended the utility of such ‘thought experiments’ and debated what it means that intuitions vary with regard to them. This dissertation proposes that those whose intuitions vary considerably from those presented here, are speaking to some other phenomenon, such as cooperation, expectation or reliability.
II.3 Rich Trustworthiness or Poor Trustworthiness? 181
II.3 Rich Trustworthiness or Poor Trustworthiness?
Since Karen Jones has written more on the nature of trustworthiness than any other modern philosopher, her influential view – rich trustworthiness – warrants special consideration. While Jones has contributed a number of important insights to the growing philosophical literature on trust and trustworthiness, in this section I want to draw attention to an unexpected and worrying result of her view: specifically, that it precludes certain groups of people from being considered as candidates for trustworthiness based on their communication differences or difficulties alone. This problem is part of a wider one; the consequence of providing narrow definitions of concepts like trustworthiness, concepts that have historically been used to demarcate and establish our very personhood and place in a shared humanity, is that we risk excluding those whose experiences and expressions do not fit our definitions, potentially contributing to their dehumanisation. This section contains two separate arguments for rejecting Jones’ view. In subsection II.3.1, I provide an overview of rich trustworthiness, underscoring that rich trustworthiness is a three-place account of trustworthiness. In the case of rich trustworthiness, what is expected of the trustee over and above being competent and committed to doing what they have been entrusted to do, is that they successfully signal their competence and willingness to be counted on. It is proposed that this is more accurately described as rich reliability, rather than rich trustworthiness. In II.3.2, I provide an ethical argument against accepting rich trustworthiness, on the basis that accepting it as an account of what it is to be worthy of trust incorrectly and potentially harmfully precludes many individuals from the possibility of being counted as trustworthy, and, no one should be considered untrustworthy for lacking the kind of social skills that figure in Jones’ account of rich trustworthiness. To be clear, my ethical argument here does not render trustworthiness itself a moral notion. Like trust, trustworthiness is an interpersonal notion. The argument that follows regards the ethical consequences of treating trustworthiness as a matter of our reliability.
II.3 Rich Trustworthiness or Poor Trustworthiness? 182
II.3.1 Jones’ rich trustworthiness (rich reliability)
Let us first examine Jones’ motivations for rich trustworthiness. We saw in II.2.1, that Jones is one philosopher who is awake to the problems associated with moralised trustworthiness. She provides an account of trustworthiness that distances itself from “moralised norms” by describing what she takes to be the distinctly non-moralised “norms of trust and trustworthiness” captured by the complaint “but I was counting on you”:
‘But I was counting on you!’… is the signature complaint of those who think that their trust has been betrayed. Thus, by coming to understand when and why the complaint has normative force, we can explore the shape, source, and force of central norms governing trust and trustworthiness. Many moralize those norms [of trust and trustworthiness] and analyse trust as ascribing, or even grounding, an obligation to be trustworthy. Against such moralizing moves, this chapter argues that the norms of trust and trustworthiness are not themselves moral, have the potential to sit uneasily with moral norms, and require external moral regulation. Nevertheless, these norms [of trust and trustworthiness] have genuine normative force, a force that derives from the pressing interest that we have as finite, reflective, and social creatures in being able directly to recruit the agency of another. (2017: 90)
As this passage demonstrates, Jones’ general overarching motivation is to separate trustworthiness from morality and instead highlight its apparent role in human agency. This aim makes sense, given that she sees the primary purpose of trust and trustworthiness as devices that extend our agency via the non-coercive recruitment of the agency of others (2017: 101-03). As Carolyn McLeod nicely summarises:
Jones argues that the concepts [trust and trustworthiness] exist essentially because of the need we have to be able to count on one another. Hence, in her view, trustworthiness involves a willingness to factor into our practical deliberation the likelihood that others are counting on us; and it can even require that we signal this willingness to others. (2015: Section 1)
This provides us with the background motivation we need. Now for the background assumptions. As we know, what all three-place accounts have in common is that they
II.3 Rich Trustworthiness or Poor Trustworthiness? 183 share the same underlying logical construction, taking trust and trustworthiness to involve minimally 1) a truster, 2) a trustee, and 3) that which is entrusted. Jones endorses her own version of weak three-place trustworthiness, whereby the trustee must be both competent and committed to doing what they have been entrusted to do, by the truster (2012: 70-1; 2017: 99). Yet because she also takes trustworthiness to be a device the very purpose of which is to extend our agency, Jones astutely recognises a problem with her own version of three-place trustworthiness. The trustee’s competence and commitment alone are not enough to ensure that dependency relations run smoothly in the way that they commonly do. To extend our agency via dependence on the trustee, to take advantage of our trustee’s competence and commitment, we need to know about it. The trustee must also be competent at signalling their particular competencies and willingness to be relied upon to the truster, otherwise such dependency relations may never be realised, as she explains:
we want the competent who can be counted on in the ways we need to identify themselves, and we want those who are not up for a particular form of dependency, whether because they lack the competence or the inclination, to identify themselves before we count on them in ways that are apt to be disappointed. We want those we can trust regarding a particular domain to signal their trustworthiness to us, so we can work out where—and where not—to turn. (2012: 74)
Jones rightly identifies that “counting on” one another in the way she describes is a non- moralised way in which we extend our agency via recruiting the agency of others, and, as she says, it is impeded if we do not signal (communicate) effectively to one another. In short, Jones’ observation can be summed up as: ‘what use is it that I could have counted on you, if you didn’t tell me that I could?’ Hence, she distinguishes trustworthiness (weak three-place trustworthiness) from rich trustworthiness, and defines the latter in the following way:
B is richly trustworthy with respect to A just in case: (i) B is willing and able reliably to signal to A those domains in which B is competent and will take the fact that A is counting on her, were A to do so, to be a compelling reason for acting as counted on and (ii) There is
II.3 Rich Trustworthiness or Poor Trustworthiness? 184
a non-trivial number of relatively central domains in which B will be responsive to the fact of A’s dependency in the manner specified in (i). (2012: 74; 2017: 94)
About this definition, notice that being both competent and committed in a particular domain and able to reliably signal that competence and commitment, precisely because it still pertains to a particular domain, is still just to be three-place trustworthy (reliable); it is simply to be better at it. Like all three-place accounts, Jones’ rich trustworthiness makes trustworthiness a matter of the trustee’s success or failure. What rich trustworthiness adds, over and above our success in simply being competent and committed, is our success in signalling as much. In contrast, the two-place trustworthiness endorsed here involves only the truster and a trustee (A and B), and how they experience one another over time. It is not a requirement of two-place trust that the truster signal their intention to count on the trustee, and it is not a requirement of two-place trustworthiness that the trustee signal that they can be counted on. The fundamental difference between the two-place account argued for here and any three-place account, is that according to a two-place account ‘what is entrusted’ or ‘what we are being counted on to do’ does not form any part of what it is to trust or be worthy of it. What distinguishes trust from reliance then on a two-place account, is that reliance involves the third-place X (‘counting on’, ‘depending on’ or ‘entrusting’ to do something specific), while trust, does not. For Jones though, counting on and being counted on, with regard to a specific domain, are what differentiate trust and trustworthiness from reliance and reliability. This distinction is similar to Stephen Darwall’s distinction between ‘expectations of’ (B to do X) and ‘expectations that’ (B will do X) (see I.3.1):
To count on something or someone is to embed in your plans and goals an expectation that, if false, means you risk being left worse off than you otherwise would have been. Counting on is related to the notion of predictive expectation, or expectation that (Hollis 1998), but having a predictive expectation is neither necessary nor sufficient for counting on something or someone. The toppling of predictive expectations can be met with mere surprise, even pleasant surprise. Not so when the things that we count on coming to pass fail to occur. That failure comes, by definition, at a cost and so is met with disappointment,
II.3 Rich Trustworthiness or Poor Trustworthiness? 185
frustration, let down, or—in cases where what we are counting on is the agency of another and our relationship is one of trust—with feelings of betrayal. (Jones 2012: 91)
According to Jones – and others (see I.1.1) – if we trust someone, and they fail to do what we are counting on them to do, then we will by definition experience “feelings of betrayal”. What distinguishes trust and reliance for Jones isn’t that one is two-place and the other is three-place. Instead, it is betrayal, which is supposedly a reaction we necessarily have when we were trusting, and the trustee fails to do what we were counting on them to do. Yet, as we have seen many times now, we can trust someone and be counting on them, and they can fail to do what we are counting on them to do without this leading to any feelings of betrayal, and without it in any way impacting their worthiness of our trust. In certain circumstances good trustees will fail to do what we are counting on them to do because of their worthiness of our trust, and bad trustees will succeed in doing what we are counting on them to do because of their unworthiness. This is because we are not always best-placed to know where to place our normative expectations: trusting, moral, predictive, or otherwise (see II.2). Jones’ rich trustworthiness does not avoid any of the pitfalls we saw in II.2, since all the complicated contextual realities, which we observed in the scenarios apply equally to our failures at signalling. In many normal everyday circumstances, our success or failure at signalling correctly what others can count on us for will simply have nothing whatsoever to do with our trustworthiness. Mistakes, misunderstandings, and misinterpretations are commonplace, and though they no doubt impact our ability to extend our agency, they usually do not make anyone “feel betrayed” or strip anyone of their worthiness of trust. Moreover, in many circumstances we can fail to signal correctly what we can be counted on for because we are worthy of trust, and we can succeed in signalling correctly because we are unworthy. Such situations are captured when we say things like:
I was always coming to get you, but I didn’t tell you that you could count on me to pick you up because I wanted it to be a surprise!
II.3 Rich Trustworthiness or Poor Trustworthiness? 186
I told her she could count on us to pick her up, now she’ll be waiting there for us like a lamb to the slaughter!
There are ‘trustworthy bad signallers’ and ‘untrustworthy good signallers’ just as there are ‘trustworthy immoral people’ and ‘untrustworthy moral people’. When he was in his last months with us, my father-in-law was unable to signal reliably. He would tell us that we could count on him to pick up something for dinner, and then he would forget. In the end we could not count on him for anything much at all. He did not even know who my children (his grandchildren) and I were. He would say to us “Who are you? It doesn’t matter… I know that you are someone that I love”. Somehow, he held on to the strong sense that we were special to him, and as this sense continued to guide his conduct towards us, he remained worthy of our trust to his last notwithstanding his declining social communication abilities. My mother-in-law, in contrast, with all her faculties and adept at social communication: “You lot can expect to count on me for nothing!” (and she was right) has never been worthy of my trust, and never will be. With rich trustworthiness, Jones has identified social-communication norms, norms related to our shared interest in extending our agency, and she has distinguished them from other moralised norms. But she has not identified trustworthiness and trust. Being au fait with the norms of social communication does not guarantee anything in terms of our worthiness of trust, just as being au fait with the moral norms does not. Jones makes a failure to signal correctly what we can and cannot be counted on for a failure of rich trustworthiness. But a failure to signal correctly is a failure of social communication, not a failure of trustworthiness. As ever, complex enigmatic trust and trustworthiness continue to defy philosophical attempts to define them and neatly carve them up. When we start from the wrong place to begin with, with a three-place conception, then tagging a condition or two on such as requiring the trustee possess ‘goodwill’ or in the case of rich trustworthiness, ‘good signalling’ will not do. Though being able to recruit the agency of others is often a valuable upshot of trusting relationships as this dissertation has acknowledged (I.3.2), this is not what trust and trustworthiness are. After all, we can non-coercively recruit the agency of those we do not know, those we do not like, those we do not trust– that is precisely what currency and contracts are for.
II.3 Rich Trustworthiness or Poor Trustworthiness? 187
With rich trustworthiness, Jones has drawn our attention to the significance of successful social communication in dependency relations. She has reminded us that if we are to effectively extend our agency via others, then we must avoid any confusion about what we can, and cannot, be counted on for. She has also demonstrated that we hold one another to account for failures of social communication in a non-moralised way. What rich trustworthiness can do is provide a framework to conceptually separate moral norms and social-communication norms, and appreciate the conditions required for those social-communication norms to extend our agency successfully. But it cannot sort out those who are worthy from those who are unworthy of our trust. What Jones has identified in rich trustworthiness is a kind of ‘second-order reliability’, whereby the chances of extending our agency via dependence on others are improved by successful communication. For accuracy, just as we must understand trustworthiness as reliability, we must understand rich trustworthiness as rich reliability. In this subsection, I have referred to arguments made in previous sections to support my claim that we must reject Jones’ rich trustworthiness as an account of trustworthiness. To accept rich trustworthiness as trustworthiness is to make a category error. I instead suggested that Jones’ important contribution is better understood as rich reliability. I turn now to providing an ethical argument against rich trustworthiness.
II.3.2 The ethical problem with rich trustworthiness
Only months before her death aged 99, in a debate with Simon Blackburn and the neuroscientist Colin Blakemore, Mary Midgley commented on Blakemore’s attempt to gain traction on what it means to be human by excluding some people such as those with advanced dementia:
I’m shocked and appalled by the move, of making definitions of what it is to be human such that you can cut some people out. Being human is a very complicated thing, and as it is a kind of compliment, if you start taking it away, you exclude a person from the group of which we are all a part. This is no trivial matter. I don’t know what makes anybody want to do it. (Midgley et al. 2016)
Those who see it as their place to do so, have tended to define humanness via a process of exclusion, rather than inclusion. Those who are excluded, and so dehumanised, are
II.3 Rich Trustworthiness or Poor Trustworthiness? 188 done so on the basis of what they supposedly lack: intelligence, rationality, self- determination, autonomy, individuality, integrity, subjectivity, sentiment, violability, morality, pain-sensitivity, and worthiness of trust (Haslam 2006). As Caroline Howarth wrote in “Race as Stigma: Positioning the Stigmatized as Agents, not Objects”:
The stigmatising representation of black youth as untrustworthy [my emphasis] and so criminal pervades these boys’ everyday encounters – as they see that they – or more precisely their skin is seen and treated in a particular way. This… leads to elaborate systems of social exclusion and marginalisation. (Howarth 2006: 447)
As many political and social scientists have recognised, calling our enemies ‘untrustworthy’ has historically been used as a way to dehumanise them, and so justify their murder:
Over time, prejudice paves the way for demonization and dehumanization. Name calling begins. Enemies describe each other as animals, aliens, or witches. Yesterday's neighbours become untrustworthy dogs... uncivilised savages without morals or ideology. Gradually, enemies cease to be human; they become a scourge that needs to be expelled or "cleaned" away. (Carter et al. 2015: 306)
And:
The foreigner, for instance, is seen at once as "wicked, untrustworthy, dirty," and "uncanny, powerful and cunning." Similarly, according to the cannons of race prejudice, contradictory qualities of exceptional prowess and extraordinary defect - ascribed to Orientals, Negroes, Jews or any other group - together make them a menace toward whom customary restraints on behavior do not obtain. The main conscious emotional concomitants of partial dehumanization, as with prejudice, are hostility and fear. (Schwebel 2003: 66)
Rhetoric to the effect that an out-group ‘lacks trustworthiness’ helps them be seen as subhuman, and, this can escalate a conflict by promoting prejudice and violence (Cassidy et al. 2017; Kassin et al. 2014). In contrast, the charge of unreliability has not been used to dehumanise in the same way. This is not surprising. While both can be pejorative, ‘untrustworthy’ is chosen over ‘unreliable’ to describe those we wish to justify violence towards because the former capitalises on our intuitive sense that one
II.3 Rich Trustworthiness or Poor Trustworthiness? 189 who is unworthy of my trust is one who is unworthy of me, below me, less than me, whereas one who is unreliable might still be worthy of me (for example, children). So, while ‘unreliable’ is still a negative charge, it does not carry the same weight, and is unlikely to have the significant adverse implications that follow a charge of ‘untrustworthy’. In this dissertation, I have been arguing that confusing trust with reliance, rich- trustworthiness with rich-reliability, is likely to have deleterious consequences for our theoretical understanding of a number of diverse phenomena. But this is only a consequence for our clear-thinking. There is a far more insidious danger of adopting strict three-place accounts of trustworthiness that find trustworthiness not in us but in our behaviour, and one that is becoming all too prevalent in philosophy at that. By narrowly defining the parameters of how fundamentally humanising qualities like trustworthiness are expressed, we can unethically and unjustifiably contribute to the dehumanisation of entire groups of individuals. We have a pressing obligation to reject any three-place accounts that, via their definitions, formulae, and necessary conditions, exclude in these ways. To see how this all-too-familiar wrong has played out in relation to rich trustworthiness, observe how – to ensure the signalling process between truster and trustee runs smoothly – Jones says that trusters and trustees must adhere to the ‘norms of trust and trustworthiness’ (social-communication norms). She identifies these, on the trust side, as:
I. Pay attention to the signalling of others regarding the kinds of dependencies that they may be up for from people like you.
II. Be aware of the different understandings of the background network of social assumptions that form part of the standing fabric in which signalling takes place so that you don’t misinterpret signals.
III. Take responsibility for communicating those dependencies, responsiveness to which cannot be assumed given shared understandings of who can be counted on by whom for what. While on the trustworthiness side:
II.3 Rich Trustworthiness or Poor Trustworthiness? 190
I. Be aware of the trustor’s location in standing social systems of signalling and, as required, counter-signal.
II. Be aware of different understandings of the background network of social assumptions that form part of the standing fabric in which signalling takes place and of what will be counted as counter-signalling.
III. Do not explicitly invite trust and then fail to follow through. IV. Develop reflective awareness of capacity so as to avoid overly optimistic signalling. (2017: 106). Though this list is not meant to be comprehensive, the significance of our adherence to these norms cannot be overstated. They supposedly govern our very entry into trust relationships (2017: 106). Most concerningly, Jones claims that:
Because we [individuals who have no impairments in social communication] have a theory of mind, we can make decisions that take into account the mental life of others. There is thus available to us [but not to those who are impaired] a distinctive way of responding to the fact of other agents’ dependency through recognising that very dependency. (2017: 99)
For Jones, the norms of trust and trustworthiness are possible only for those who possess a ‘Theory of Mind’ (ToM). Jones goes so far as to call this ability the “heart of trust” (2017: 100). She also claims that the connected ability to pay attention to instilling and maintaining sensitivity to the fact that others are counting on us is one of our “core people-making practices” (2017: 97), suggesting that deficits in this regard risk ‘unmaking’ us as persons. At another point, she proposes that being able to count on people meeting us in our dependency “makes possible the distinctive vulnerability of intimacy” (2017: 100), again suggesting that one who is impaired in their ToM would find intimacy impossible. So, why is this a problem? To answer this, we need to understand something about the theory Jones makes central to her account of trustworthiness. For those who are not familiar with it, the ToM theory, adopted in philosophy by the drolly-dubbed ‘theory-theorists’ and accepted widely in social and cognitive psychology, asserts that attributions of mental states to other people are explained by each of us possessing a naïve theory of mentalising. For example, if I know that you are counting on me to buy you a copy of
II.3 Rich Trustworthiness or Poor Trustworthiness? 191
The Guardian and leave it on your desk for you, then I can infer that you will not buy your own copy of The Guardian, and also that you will be surprised and annoyed if I were to instead buy you a copy of my own preferred reading material, The London Review of Books. An impacted ToM supposedly explains why some individuals have difficulty making the correct inferences in situations like this, and a functioning ToM supposedly explains how we can ‘read one another’s minds’ to, among other things, count on one another to do things for us, and so extend our agency. Given that Jones sees trustworthiness as a device the purpose of which is to extend our agency (2017: 100-03), it is not surprising that she has linked this apparent ToM ability to our ability to be worthy of trust. The potential harm from using a ToM model to delineate the trustworthy from the untrustworthy comes in the detail that there are any number of groups who struggle with social communication and so may not – typically through no fault of their own – be able to reliably adhere to Jones’s ‘norms of trust and trustworthiness’. Indeed, the seemingly benign list of seven norms of trust and trustworthiness Jones provides turns out to be highly exclusionary. For example, it is widely taken to be the case that ToM deficits explain the social communication difficulties found in: autism (Baron-Cohen et al. 1985); schizophrenia (Sprong et al. 2007); attention deficit hyperactivity disorder (Korkmaz 2011); major depressive disorder (Lee et al. 2005); various specific language impairments (Nilsson and de Lopez 2016); certain traumatic brain injuries (Bibby and McDonald 2005); alcoholism (Uekermann and Daum 2008); dementia and Alzheimer’s patients (Gregory et al. 2002); blind children (Bedny et al. 2009); minimally verbal children with cerebral palsy (Sandberg and Hjelmquist 2003); children with Down’s Syndrome (Zelazo et al. 1996); Parkinson’s patients (Saltzman et al. 2000); frontal lobe patients (Rowe et al. 2001); and children with specific language impairment (Miller 2001). These are pathologised populations, and some may agree with the consequence of Jones’ account that human beings belonging to these groups are excluded from the set of those who might count as worthy of our trust. However, ToM delays also occur in otherwise perfectly ‘normal’ deaf children who are late signers because their parents do not sign (Woolfe et al. 2002), and, oddly enough, in children born in Iran and China,
II.3 Rich Trustworthiness or Poor Trustworthiness? 192 though they appear to acquire other mentalising abilities at an earlier stage, which has been taken as suggesting that different cultures may prioritise different mentalising abilities (Shahaeian et al. 2011), though this explanation is not endorsed here. As for those of us who thought we were off the hook, ToM deficits are found in those who are sleep-deprived (Erwin 2016), those experiencing significant or chronic emotional or physical pain (Zunhammer et al. 2015); and lastly, a decline in the social communication abilities attributed to ToM is also part of the normal ageing process (Duval et al. 2011). As Jones would have it, untrustworthiness comes to us all in the end. If we add these groups together, then a significant proportion of the population is at any one time excluded from the possibility of being counted as trustworthy based on a set of largely irrelevant measures ranging from age to culture to exhaustion. This dissertation is not arguing against the proposal that we should be cautious of relying on or counting on those who are exhausted, grieving, experiencing cognitive problems, or clinically depressed etc. to do things for us. However, there are many things wrong with the idea that an impacted ToM makes a person unworthy of trust. First, for decades now the very idea that normal human social cognition (if there is such a thing) involves the utilisation of a Theory of Mind has been drawn into serious question by philosophers and cognitive scientists (Ferrari et al. 2003; Gallese and Goldman 1998; Iacoboni and Dapretto 2006; Oberman and Ramachandran 2007; Rizzolatti and Arbib 1998; Rizzolatti et al. 2001; Rizzolatti and Craighero 2004; Umilta et al. 2001). As John Duffy and Rebecca Dorner say in their paper ‘The Pathos of Mindblindness’:
At present, there is no empirical evidence to support or disprove ToM hypotheses. A grand search in the neuroanatomical literature for the locus of the autistic ToM deficiency has been inconclusive, which is perhaps inevitable given that no one really knows where the seat of the “mind” exists in the brain, if it does exist after all. The absence of empirical data means that the degree to which ToM is accepted by researchers and the general public ultimately comes back to the rhetorical power of a story that asks us to accede to its rational, emotional, and moral views of the world. (2011: 205)
II.3 Rich Trustworthiness or Poor Trustworthiness? 193
Binding a theory of trustworthiness to a suspect theory of mentalising is itself a suspect move. Setting this awkward truth aside though, the fact remains that even if such a capacity exists, and even if some groups of people lack it, this must not automatically disqualify them from the possibility of being worthy of trust. Firstly, a lack of trustworthiness has been used to dehumanise certain groups in the past as already discussed. This is itself enough of a reason for us all to steer well clear of providing exclusionary definitions of it. Secondly, the evidence. Those individuals who do have difficulties with social communication – difficulties with Jones’ norms of trust and trustworthiness – are simply not thereby untrustworthy. The sheer number of groups implicated by Jones’ account means that this dissertation cannot review the evidence regarding the potential trustworthiness of every one of them. Given this, a pertinent way to proceed is to take on the strongest case – the group with the most severely impacted ToM – and see what the evidence for untrustworthiness is among this population. Autistic persons, by diagnostic stipulation, are impaired specifically and significantly when it comes to Jones’ so-called ‘norms of trust and trustworthiness’. For this group, things like ‘being aware of the different understandings of the background network of social assumptions’ and ‘being aware of your location in standing social systems of signalling’ are impacted to a greater or lesser extent (Baron-Cohen et al. 1985; Tomasello et al. 2005; Wing and Gould 1979). As this is the group most associated with ToM deficits (Happé 1994), autistics are most implicated as unworthy of trust according to Jones’ account and all other accounts which view behaviour, especially communicative or socio-normative behaviour, as the defining feature. This implication is potentially serious, since the charge of untrustworthiness adds weight to an already large body of work dehumanising autistics, as written about extensively by Melanie Yergeau (2013, 2017), and as captured in these two quotes by well-known autism researchers:
It’s as if they [persons with autism] do not understand or are missing a core aspect of what it is to be human. (Falcon and Shoop 2002)
[Autistics lack] one of the quintessential abilities that makes us human. (Baron-Cohen 1997)
II.3 Rich Trustworthiness or Poor Trustworthiness? 194
In philosophy, the dehumanisation of autistics typically occurs in the process of attempts to define humanising concepts, such as personhood, the self, and – as in Jones – agency. Philosophical discourse that uses autistics to attempt to explicate various concepts is present in, for example, Goldman (2006), Currie and Ravenscroft (2002), Blair (1996), Kennett (2002), and Farkas (2017). The extension of the dehumanising implication to philosophy is to be expected, given that denying the autistic person personhood via rhetoric to the effect that they lack certain humanising attributes has been routine since the very conception of the condition. The ‘founding father’ of the current Gold Standard behaviourist treatment for autism (Applied Behavioural Analysis) is the Norwegian-American psychologist and professor at the University of California, Los Angeles (UCLA), Ivar Lovaas. In an interview with Psychology Today in 1974, Lovaas described his approach to a distressed autistic child named Beth who was under his care. When Beth resorted to self-harm in an effort to self-regulate, Lovaas recounts:
I just reached over and cracked her one right on the ear. She was a big fat girl, so I had an easy target… I let her know that there was no question in my mind that if she hit herself again, I was going to kill her. (Chance 1974: 78)
Such violence, which was routine in his practice, was no doubt enacted with a clear conscience by Lovaas, given his beliefs about autistic people generally. In the same interview, this founding father of their ‘treatment’ said:
You see, you start pretty much from scratch when you work with an autistic child. You have a person in the physical sense – they have hair, a nose and a mouth – but they are not people in the psychological sense. One way to look at the job of helping autistic kids is to see it as a matter of constructing a person. You have the raw materials, but you have to build the person. (1974: 76)
Such attitudes sadly, have not died with the passage of time. At the Judge Rotenberg Centre in the United States, autistic people are still today made to wear back-packs that deliver to them a painful electric shock of 41 milliamps – 10 times the amount used in most stun guns – when they ‘misbehave’ (Pilkington 2018). Yet the evidence for the widespread mistreatment of autistics is nowhere more apparent than in research that
II.3 Rich Trustworthiness or Poor Trustworthiness? 195 looks into the autistic life expectancy. While autism as a condition itself does not shorten life expectancy, the average life expectancy of a person with autism in the United States is 36, with suicide, infanticide, and medical neglect being some of the leading causes of death (Guan and Li 2017). Recent research from the Autism CRC of UNSW Sydney revealed that the mortality rates of those with ASD are more than twice those of the general population in New South Wales, Australia (Hwang et al. 2019). Given this ongoing dehumanisation, violence, and injustice, influential thinkers – especially ethicists! – must proceed with extreme caution when it comes to implicating this vulnerable group of individuals by denying them anything connected with personhood, agency, humanity. Jones is most likely wholly unaware of the potential harm to these individuals of making trustworthiness a matter of social communication. But others, I think, are not so innocent. In one example, the likes of which is rarely seen in modern times outside of the extreme dehumanising ‘animalisation’ propaganda of genocidal conflict (Haslam 2006: 252-53), Steven Pinker, in his book The Blank Slate, alleged:
together with robots and chimpanzees, people with autism remind us that cultural learning is possible only because neurologically normal people have innate equipment to accomplish it. (Pinker 2002: 62)
Pinker’s book was highly influential, being endorsed by the likes of Richard Dawkins, Francis Fukuyama, and Daniel Dennett. By “cultural learning” Pinker means “equipped to discern other people’s beliefs and intentions” (Chapter 4) and so “cultural learning” in Pinker’s sense is comparable to Jones’ social signalling '“norms of trust and trustworthiness”. Autistics supposedly lack these abilities, and so, according to Pinker, they are no better than robots or chimpanzees. At least according to Jones, they are only unworthy of our trust. Such rhetoric, such insinuation, leaves these human beings vulnerable to those who may wish to justify the inhumane treatment they are receiving. So far, I have been building an appeal case to cease and desist producing exclusionary definitions of trustworthiness, purely on ethical grounds. But supposing that the ethical argument is no good, that it is somehow justifiable to contribute to the dehumanisation of some human beings if indeed they do not have the equipment for
II.3 Rich Trustworthiness or Poor Trustworthiness? 196 trustworthiness, then, what does the evidence say about the worthiness of trust regarding the population most associated with ToM deficits? Taking even just a cursory glance at it, we see that despite their apparent social communication differences, despite their difficulties in adhering to the supposed “norms of trustworthiness”, the evidence that autistic people are not proper candidates for trustworthiness (let alone humanity itself) is not forthcoming. Indeed, despite a persistent trope to the contrary, autistic people have typical or even excessive emotional empathy (Brewer and Murphy 2016; Markram et al. 2007), and do not have an increased risk for violence (Woodbury-Smith et al. 2006). In fact, more recent research has revealed that autism reduces the risk that an otherwise high-risk individual will be convicted of a violent crime (Heeramun et al. 2017). Moreover, autistics are frequently noted as being characteristically fair, loyal, and honest (Attwood 1997: 32), making them regularly the victims – not the perpetrators – of deliberate and malicious deception (Bargiela et al. 2016; Brown-Lavoie et al. 2014). Lastly, like the rest of us, autistics often deeply desire friendship and a sense of belonging, yet unlike the rest of us they regularly endure both mental and physical exhaustion in their tireless efforts to camouflage their autism in order to fit in and connect with others who do not experience the world in the same way as they do, and so do not share their social communication style (Hull et al. 2017; Livingston and Happé 2017). If we presume – and I think we can safely presume – that trustworthiness has more to do with friendship, belonging, care, honesty, loyalty, fairness, non-violence, empathy, and a tendency to be open and non-deceptive than it does to do with following social norms or extending our agency, then autistics are no doubt worthy candidates for being parties to healthy trusting relationships. Making trustworthiness a three-place matter of our ability to do X, follow certain norms, or signal reliably to one another, is not only unjustly exclusionary though. It also runs the risk of being dangerously inclusionary. On tests designed to measure a person’s Theory of Mind, those with diagnosed antisocial personality disorder (previously known as psychopaths) – individuals who show a pathological lack of guilt and remorse at harming or deceiving others – show no ToM deficits, performing significantly better than those whom Jones’ account excludes (Blair 1996; Dolan and Fullam 2004). This is
II.3 Rich Trustworthiness or Poor Trustworthiness? 197 a group with no trouble signalling, no trouble with Jones’ norms of trust and trustworthiness. Indeed, their proficiency with Jones’ norms makes them especially good at lying and manipulation, which require good perspective-taking abilities. Again though, it is not just these pathologised populations that are bizarrely implicated. Chillingly, the richly trustworthy are all around us. The deceitful politician, successful con-artist, corrupt businessman, conniving co-worker, and calculating ‘frenemy’ all adhere closely to Jones’ norms for trust and trustworthiness, as to excel in these areas one must: ‘recognise the zones of their competence’, ‘monitor new expectations for compatibility with current ones’, and ‘recognise the expectations different people are liable to form’ and ‘what they will count as a signal’. Jones may counter that such individuals would not qualify on her account, as, by harming us, they could not possibly be doing what we counted on them to do. However, this is to overlook the insidious danger of master manipulators. As we saw in II.2.2, their very success lies in their ability to convince us that things which are bad for us are good, things which will harm us we need, things which are dangerous are benign. Such individuals count on us counting on them to do just what we are counting on them to do! We saw an example of this too in the scenarios presented in II.2.2. Though we should not have needed it, the evidence tells us that rich trustworthiness condemns innocent groups and renders those with highly questionable qualifications the standard bearers. Yet this is still not the most disturbing upshot of Jones’ account. Not only does rich trustworthiness condemn those with social communication problems as untrustworthy, it also precludes us from being trustworthy towards them. This is because communication is a two-way street, as Jones herself recognises (2017: 104), and problems typically go both ways. This situation adheres for all our implicated groups (i.e. social communication is likely to be strained between two people when one is grieving and the other is not, one is depressed and the other is not, etc.), but it is again nowhere more evident than in relations between autistics and non-autistics. Though for a long time autistics were blamed exclusively for the social communication difficulties that exist between autistics and non-autistics, emerging research is showing that non- autistics experience just as much difficulty in taking into account the mental life of autistics and signalling to them appropriately, and empathising with them appropriately,
II.3 Rich Trustworthiness or Poor Trustworthiness? 198 as the other way around (Milton 2012, 2017; Sasson and Morrison 2017; Sasson et al. 2017; Sheppard et al. 2016). Non-autistics – with no supposed ToM impairments – are impacted in equal measure when it comes to ‘reading the minds’ of autistics (Heasman and Gillespie 2018). Indeed, the ‘signalling problem’ between these two groups is two- sided (Brewer et al. 2016), leading one to muse that perhaps to autistics, Stephen Pinker looks and sounds much like a chimpanzee, or a robot; taking in data from the surrounding environment and spitting it back out unreflectively. Two final pieces of evidence are relevant. In their interactions with one another, autistics often overcome their apparent difficulties with the “norms of trust and trustworthiness” by finding novel methods for social communication, thereby achieving innovative intersubjectivity that is not readily identifiable within the standard normative framework (Heasman and Gillespie 2018). It has been noted that this is similar to how Deaf Culture provides a non-standard normative framework for social communication among the deaf (H. Lane 1989). Lastly, the apparently automaton-like behaviours of autistics in childhood, many autistic adults now tell us, were actually meaningful attempts at communication and connection. What were taken to be empty ‘robotic’ or ‘animal’ bodily sounds and movements and so ignored – or worse, violently punished – were actually displays of emotion or affection, expressions of thoughts and desires, attempts at humour, expressions of empathy, love, trust, and understanding (Jaswal and Akhtar 2018). Perhaps, given their vast historical experience of being misjudged, misunderstood, and underestimated, autistics may have developed the skills required to afford Stephen Pinker the benefit of the doubt he never gave them. When it comes to the ToM that Jones places at the centre of our ability to be worthy of trust, the reason one group gets held responsible for the signalling upsets and not the other, one group gets dehumanised and not the other, is, as vice president of the Autism Association of New England Philip Schwarz explained in an interview:
non-autistic people too, are rather lousy at understanding the inner states of minds too different from their own – but the non-autistic majority gets a free pass because they assume that the other person’s mind works like their own, and they have a much better chance of being right. (Szalavitz 2009)
II.3 Rich Trustworthiness or Poor Trustworthiness? 199
We must recognise that we are all rather poor at intuitively understanding – and so signalling appropriately – to minds that are different from our own. When social communication fails, this does not mean the person we are trying to communicate with, trying to reach intersubjectivity with, trying to extend our own agency with, does not have a mind, does not have agency, is not human, is not worthy of our trust. All the groups implicated by rich trustworthiness are experiencing the world in a different way. Successful signalling between individuals who are different is going to be fraught; we have known this since before we ever encountered other cultures, since diversity has always been part of humanity, even at home. Diversity in experience and expression can of course hinder social communication, and as such can encumber our attempts to extend our agency, count on one another. But it cannot make one unhuman, unworthy of trust. It is both unethical and, in many instances, false to adopt an account of trustworthiness that has this implication. On closer inspection, Jones’ ‘heart of trust’, which supposedly emerges from a shared Theory of Mind, is more likely not much more than a fluke. Our “distinctive way of responding to the fact of other agents’ dependency through recognising that very dependency” seems to hinge on those other agents being like us. What we think of as the ability to ‘read someone’s mind’, then, is perhaps the rather less impressive coincidence of happening to possess a similar mind, and what we think of as intimacy and a shared-humanity is nothing more than mirror-gazing. The fruitful frontiers of trust and trustworthiness with their characteristically hopeful leaps beyond what is known and familiar, belong to those who expand beyond these measures, not to those who exist safely within them. Reaching understanding and intersubjectivity, forgiveness and optimism, hope and care, with someone who is very different to you and communicates very differently to you is the highest expression of trust and trustworthiness. Jones wraps up her account with the maxim that “successful trust relations are those that match up would-be dependents with those willing and able to be counted on in the way required” (2017: 103), but one might think that the classifieds section of the local newspaper does this well enough. There is no need to evoke trust and trustworthiness to adequately describe this phenomenon.
II.3 Rich Trustworthiness or Poor Trustworthiness? 200
II.3.3 Summary of rich trustworthiness
This dissertation agrees unreservedly with Jones that similar minds may communicate, and so cooperate, more effectively with one another than those which are dissimilar. This section has argued that this certainly does not delineate the trustworthy from the untrustworthy, the human from the unhuman. Given the ethical implications that follow the charge of untrustworthiness, that have historically been connected to the charge of untrustworthiness, we must re-frame rich trustworthiness as rich reliability. Then we can maintain Jones’ important insight without excluding any potentially worthy candidates. To capture trust and trustworthiness, we need lengthy accounts, not short definitions. We value simplicity in philosophy, and that is no bad thing. But what we lose when we define trustworthiness in a line is not just understanding. We leave behind half of humanity.
II.4 Trustworthiness 201
II.4 Trustworthiness
Like trust, trustworthiness is a two-place relation, involving a truster and a trustee. Of course, that is not all there is to say about it. Yet given that the concept of trustworthiness has been used to dehumanise certain groups in the past as we saw in II.3.2, this Section will not be providing any necessary or sufficient conditions, nor formulas/definitions of trustworthiness. This dissertation aims not to preclude any group from the possibility of being counted as worthy of trust based on their inability to meet my chosen criteria. Because trust/worthiness is a two-place relation, the only person with any authority to decide if any particular individual is worthy of their trust is the truster, not any philosopher. In the counterpart to this section, I.4 (Trust), this dissertation demonstrated how trust can – not as a matter of necessity or sufficiency but just as a matter of circumstance – encompass many different things: emotions and attitudes, beliefs and virtues, thoughts and plans, history and memory. This enterprise was useful in Part I, as it helped us to appreciate just how complex and involved trust can be. But here, though they do apply, it is unnecessary to rehash these same arguments in relation to trustworthiness. Sections II.2 and II.3 adequately demonstrated that, just like trust, our trustworthiness can involve a vast array of antithetical phenomena depending on context: truth and lies, assiduousness and reckless abandon, morality and immorality, norm-following and norm-breaking. The aim of this Section instead is to develop the surprising link between trustworthiness and ‘mind-blindness’. Most accounts of trustworthiness, as we saw in II.1, in one way or another place trustworthiness in opposition to deception/opacity. The trustworthy do what we: ‘trust them to do’ ‘expect of them’, are ‘counting on them to do’, provide ‘true testimony’, and ‘signal accurately’. All these are ways of saying that trustworthiness is about transparency, about knowing what sort of person the trustee is (one-place), or what they will do (three-place). This interpretation of trustworthiness as we saw in II.3, carries the implication that people who are different to us and so less transparent to us are unworthy of our trust. In contrast, this dissertation takes the view that the lack of transparency that exists between us all provides us all with remarkable opportunities. What remains hidden between us, all that we might be wrong about with
II.4 Trustworthiness 202 regard to one another, affords us the chance not only to be other than what people think we are, but also for other people to hope that we might be other than as we appear. To accomplish this aim, subsection II.4.1 presents some dialogue from the film The Invention of Lying. The imagined version of humanity portrayed in that film, honest to a fault, wholly transparent and incapable of deception, is palpably not brimming with trust and trustworthiness. II.4.2 clarifies the human capacity for ‘mindreading’ or mentalising; II.4.3 addresses a possible charge of hypocrisy; II.4.3 further characterises two-place trustworthiness set against the view of the Danish philosopher Knud Ejler Løgstrup. Finally, in II.4.6, I make some closing comments on the relevance of P. F Strawson’s insight in “Freedom and Resentment” (1962) to the current discussion. By the end of this section, my hope is that we have a greater appreciation of the extent to which the veil that exists between us all is responsible for, and sustaining of, trust and trustworthiness, and how trust and trustworthiness have to do not with lifting this veil, but with living with it.
II.4.1 The Invention of Lying
Evidently drawing on his philosophical training gained at University College London (he intended to study biology but changed to philosophy after two weeks, and received upper second-class honours), in The Invention of Lying (2009), the British comedian Ricky Gervais has created a philosophical thought-experiment of some relevance: A world where not only is lying not yet ‘invented’, but concealment of thoughts, beliefs, desires, reactions, motives, emotions, attitudes – pretence of any kind – is unheard of. This is a world where the contents of other minds are not private, not veiled, but public. Whatever another person is doing, thinking, or feeling, whatever their aim or desire, whatever they believe to be true is painfully apparent since the private goings-on of people in this world are reported in all their brutal, revolting, and occasionally poignant totality. This subsection contrasts dialogue from the beginning of the film, when the protagonist is not yet able to lie, with dialogue taken from later in the film, when he can. These dialogues take place between the protagonist Mark (played by Ricky Gervais) and his love interest Jennifer (Jennifer Garner), and between Mark and his friend Frank
II.4 Trustworthiness 203
(Jonah Hill). Mark is, the narrator tells us, a “loser”. He is unattractive and has made little progress in either his personal or professional life. Jennifer is beautiful, good- natured, and “way out of his league”. Frank is unemployed, lonely, and suicidal.
Mark & Jennifer Without Deception
[Mark knocks on Jennifer’s door for a blind-date.]
Jennifer: Hi. You're early. I was just masturbating.
Mark: That makes me think of your vagina. I'm Mark, how are you?
Jennifer: A little frustrated at the moment. Also, equally depressed and pessimistic about our date tonight. I'm Jennifer. Come on in. Um... Just wait there. I need to finish getting ready. While doing that, I might realise I'm still horny and try to finish masturbating without you hearing.
[Jennifer leaves Mark standing by the door and goes into her adjacent bedroom. The conversation continues, at a louder volume.]
Mark: I feel awkward now about being early.
Jennifer: Yeah, I'm disappointed that you're early and not really looking forward to tonight in general, but the thought of being alone the rest of my life scares both my mother and me equally.
Mark: Sure. I'm thinking you've started masturbating, cause it's like… it's too quiet. And you said you were going to try and do it without me hearing. I'm worried the restaurant I've picked isn't expensive enough for you. It's all I can afford in my situation. I know I'm in my 40s, but I haven't got any financial assets to speak of. Also, my boss said he's probably going to fire me this week.
[Jennifer walks back into the room, looking refreshed and dressed, ready for dinner.]
Jennifer: I just masturbated. Where are we eating tonight?
Mark: A little place called la Bonisera in West Hollywood.
Jennifer: You obviously don’t have very much money, but that’s not necessarily a deal breaker.
Mark: I have very little money.
II.4 Trustworthiness 204
Jennifer: I also don’t really care about a guy who knows all the latest hippest restaurants.
Mark: I don’t know any of them.
Jennifer: In fact, there are very few things in life that I care about all that much. The only things I have to offer myself or anyone else are my good looks and my affected sense of quirkiness, which artistically inclined men interpret as intellect. In fact, I think my best trait is that I’ve made very few mistakes: socially, academically, financially or romantically. I take very few risks and therefore lead a relatively happy and light- hearted existence. Mostly though, I’m a kind sweet person with the potential of becoming a vital and interesting human being the day I take the energy I expend on hyper self-reflectivity and apply it to actual action in the reality of my life.
Mark: I found that boring and started thinking about this place’s fish tacos.
Mark & Jennifer With Deception (i.e. after Mark has ‘evolved’ the ability to lie)
Mark: I’m happy, you see, like that happy chap over there.
Jennifer: What, sleeping, ugly fatty?
Mark: Well...
Jennifer: He's not happy.
Mark: Well, how do you know?
Jennifer: What do you mean? Look at him. He's a loser.
Mark: You can't tell that from just looking at the guy. He could be the world's greatest poet. Well, he probably is a loser. Bad example. But I'm saying, it doesn't have to be, just by... Okay. What do you see when you see this fellow here?
Jennifer: Short, sweaty, bald guy.
Mark: Right. But he's carrying a briefcase. He's in a hurry. He's probably off to a really important meeting. He's probably a high-powered businessman.
Jennifer: You see more than I do.
Mark: Well, because, if you look... What do you see when you see those guys?
Jennifer: Mmm...Two nerdy losers in hats?
II.4 Trustworthiness 205
Mark: Yeah. Good. Good observational skills. But what I mean is, look beyond just their appearance and look at them. They're holding hands. They're in love.
Mark & Frank Without Deception
[Mark and Frank meet in the elevator.]
Frank: Hi Mark. How are you?
Mark: Not so good. Last night I went on a date with a girl who I’ve had a crush on for years who will most likely never call me again and I’m pretty sure I’m going to get fired today. You?
Frank: I spent the whole night throwing up pain killers because I’m too scared to take enough to kill myself.
[The elevator door opens.]
Mark: See you tomorrow
Frank: Bye.
Mark & Frank With Deception
Frank: Mark, how's it going?
Mark: Good. How's it going with you?
Frank: Pretty awful. I was up last night researching on the Internet, like, suffocation suicide. I think that's what I'm gonna do tonight. Well… See you later.
[Mark nods, and Frank begins to walk off. But then…]
Mark: Frank.
Frank: Yeah?
Mark: Don't do that.
Frank: Why? I mean... You know, I'm miserable. I don't think anyone would care.
[Mark hesitates…]
Mark: I'll care. [an obvious lie]
Frank: But you're a loser, which doesn't really count.
[Mark nods, turns away again, but then…]
II.4 Trustworthiness 206
Mark: Things… things are gonna… be okay.
Frank: They are?
Mark: Yeah. You're gonna meet someone. You're gonna be happy.
Frank: I shouldn't kill myself?
Mark: Definitely not.
Frank: Okay, so I don't need to kill myself. Wow! I thought... I thought that like... That suffocation idea, I thought that was like a really good idea, you know?
Mark: It wasn't.
Frank: Wow! My night has opened up!
The Invention of Lying comes to a close with Mark and Frank beginning to form a real friendship and Mark and Jennifer falling in love. Mark and Frank’s relationship moves from an association based on transparency and truth, to a friendship based on compassion and care. Truth (like morality and rationality) becomes a consideration, not a command, and it is to be weighed against other considerations, most notably, what is best for the truster. With Jennifer and Mark, their relationship deepened following an act of honesty rather than deception, but honesty in the context of his ability to deceive. Jennifer finds out about Mark’s ‘super-power’ and is so taken with his care in not deceiving her into being with him (when he easily could have), that she finds him irresistible. As the two of them struggle to come to grips with the moral/interpersonal novelties thrown up by his ability to deceive and her knowledge of it, they agree: “Everything is hard. Nothing is easy anymore.” The narrator closes by telling us that the ability to deceive eventually spreads through the population, creating “A world without honesty, a world with pretence, a world with fiction, a world with flattery, and most importantly, a world with true love – a world very much like our own.” The Invention of Lying is not a refined work of humanistic genius on par with what was presented in this section’s counterpart (I.4). George Eliot’s Middlemarch and The Invention of Lying are not of the same force. As Manohla Dargis of The New York Times put it:
II.4 Trustworthiness 207
For the most part, Mr. Gervais prefers to shock us with our own brutality...[with] unvarnished truths [that] begin to feel heavy, cruel. (2009)
Yet as she also points out:
Lying becomes a means to transcendence… [The Invention of Lying] is a melancholic defence of deceit. (2009)
A defence of deceit. This is a world where none of the quintessentially human forms of deception are possible. There is no: two-faced pretence, false-modesty, insincerity, mockery, sarcasm, lying, negging, gaslighting, bullshitting, misdirection, misleading etc., and yet, it is no social utopia. This film upends any simplistic treatment of what it means to be worthy of trust by showing us a version of ourselves that is honest, truthful, transparent, but also ugly, heartless, and bare. In the beginning, all the ingredients for trustworthiness seem to be there, or at least, none of the obvious ingredients for untrustworthiness are present, and yet these transparent individuals cannot engage in anything like the special relationships of trust that we are so familiar with. Here, the distinction between the public self and the private self is virtually non- existent. Such radical psychological transparency, the standard story goes, would make people profoundly more trusting and trustworthy. Yet contrary to what much of the literature on trustworthiness suggests (see: II.1, II.2, II.3) this transparent version of humanity, as Gervais imagines it, is neither loving nor trusting. Instead their relationships are hollow, self-interested, contractual, rational, and immoral. The people in The Invention of Lying follow Jones’ norms of trust and trustworthiness in an exemplary manner, signalling to one another their intentions without hesitation or obfuscation; they know just who they can count on, and for what. Because of this, they can ‘trust to do’ in that three-place sense with very little epistemic risk, having unfettered access to one another’s emotions, thoughts, and aims. Gervais wrote, directed, and starred in the film, and seems to have intended foremostly to provide a mocking commentary on the role of lying with regard to religion. But this crass Hollywood rom-com-cum-philosophical-thought-experiment somehow manages to do something more. The Invention of Lying, at heart, is the story of one man’s sudden and unexpected psychological emancipation. Yet the less
II.4 Trustworthiness 208 significant aspect of the protagonist’s journey is that he gains the freedom to lie to others. More importantly, the veil between himself and the rest of humanity provides him with the psychological freedom to imagine, hope, forgive, give the benefit of the doubt, and challenge his own motivations towards, perceptions of, and beliefs about others. The relevant message is that where truth and transparency can be prisons, lies and deception can be liberators. The extension of our agency may be about the ability to get others to signal reliably and do what we expect, however, being worthy of one another comes when our own freedom to choose who to be with regard to others, allows us the freedom to choose how to regard others. I take this theme up again in II.4.5. In summary, this subsection contrasted dialogue from the beginning of the film The Invention of Lying, when the protagonist is not yet able to lie, with dialogue taken from later in the film, when he can. The protagonist’s only friendship and romantic relationship is greatly improved by the veil that miraculously descends between himself and the rest of humanity: a veil that the rest of humanity still lacks. Once he lacks transparency, Gervais’ protagonist begins to show recognisable signs of trust and trustworthiness: optimistic interpretations of others, giving the benefit of the doubt, a sensitivity to the truster that overrides any sensitivity to rationality or truth, noble lying and courageous selfless honesty. The Invention of Lying will continue to be evaluated in the context of the discussions presented in the following subsections.
II.4.2 Mindreading and deception
You have linked trust and trustworthiness to a “veil” that supposedly exists between people in our world. A veil, you say, that the people in The Invention of Lying lack. Yet is there really some mysterious veil that exists between us? Surely, we do have mentalising abilities and these facilitate us rationally trusting one another.
In our world, the private contents of other people’s minds are sometimes apparent to us. People often tell us what they are doing, thinking, feeling, desiring, imagining, etc., just as they do in The Invention of Lying. And even when they do not openly narrate for us their own inner goings-on, now and then, we can look at another person and come to know such veiled happenings anyway. This dissertation does not take the view that we are incapable of possessing any knowledge of other minds.
II.4 Trustworthiness 209
The view presented here, is that despite our successes, despite the knowledge we have of other’s thoughts, beliefs and emotions, we often fail at mentalising (‘reading other people’s minds’). There are at least two ways we might go about confirming this. One way is to carry out systematic empirical investigations. As I have touched upon already (II.3.2), mentalising researchers have postulated a ToM brain-module that is supposedly responsible for the human capacity for mentalising. The way in which such research into mentalising has been conducted should be of interest to philosophers, especially those referring to ToM when making claims about agency, personhood, trustworthiness, etc. The most highly regarded investigation into human mentalising by perhaps the most prominent researchers in the field is described in a paper entitled ‘Do Triangles Play Tricks? Attribution of mental states to animated shapes in normal and abnormal development’ by Abell, Happe and Frith (2000). Their method, as described in their abstract, is as follows:
Computer-presented animations were used to elicit attributions of actions, interactions and mental states. Two triangles moved around the screen according to one of three conditions. Descriptions of the animations were rated according to accuracy and type of description. Adults predominantly used action descriptions for Random animations (e.g. bouncing), interaction descriptions for Goal-directed (G-D) sequences (fighting), and mentalising descriptions for Theory of Mind (ToM) sequences (tricking). (2000: 1)
What this study revealed, was that children with autism used mentalising descriptions less often than typically developing 8-year-olds, and “frequently referred to mental states that were inappropriate to the animation” (2000: 1). This discrepancy between autistic and non-autistic descriptions of the triangles, has been widely taken as evidence that children develop, in the normal case, a Theory of Mind, the ability to ‘mindread’. To appreciate what is happening here though, note that the triangles lack any anthropomorphised features (face/colour/voice) and are simply black triangles moving about on a white screen, occasionally bumping into various other objects, and speeding up or slowing down. When asked the key question What are the triangles are doing? Those with postulated ToM deficits had “inappropriate” responses, for example, where
II.4 Trustworthiness 210 typically developing children said that the triangles were “coaxing” or “mocking” one another, atypically developing children would say something else, either providing precise physical descriptions of the triangle’s movement patters, or else, providing “false” mentalising interpretations. To be clear, what is purportedly evidencing Theory of Mind here is the existence of apparently false accounts of the mentality of triangles. Even more concerningly, the refreshingly literal and creatively divergent responses of those with supposed ‘ToM deficits’ are being taken as evidence not only that they lack the ability to understand others’ mental states, but even that they lack inner mental states of their own. An hypothesis widely (and wildly) postulated long before such “empirical proof” (Baron-Cohen 1997: 6; Kanner 1943: 242). Though the “evidence” is both feeble and methodologically frightening, as we have already seen it has proven to be an irresistibly exciting prospect to philosophers– real-life zombies! Philosophers could be valuable allies to those with language, communication, and social differences, if, rather than latching on to popular mentalising research in their discussions about personhood, free-will, agency, and trustworthiness, they instead examined its gross methodological, ontological, epistemological, and ethical flaws. Thankfully, serious concerns are now being raised regarding the validity and soundness of such investigations from within the discipline itself (Dalton et al. 2005; Gernsbacher and Frymiare 2005; Kapp et al. 2013). Duffy and Dorner summarise this push-back in “The Pathos of "Mindblindness": Autism, Science, and Sadness in "Theory of Mind" Narratives”:
the condition of “mindblindness” is itself an act of imagination, a story told by cognitive scientists about the evolution of the mind and individuals who have followed that evolutionary path. In the absence of definitive physiological or neurological markers, ToM remains a construct, a theory of a theory, its “reality” an outcome of laboratory tests and the rhetorical efforts to persuade others of the validity of those tests. (Duffy and Dorner 2011: 209)
Given this, we should all be cautious of any supposedly scientific claims regarding our “ToM brain module” or our “mind-reading abilities”. This is especially so, since further investigations into mentalising have managed to establish something rather
II.4 Trustworthiness 211 embarrassing– those with ‘typical’ results on standard mentalising tests consistently overrate their own abilities. As it turns out, it is normal to be not nearly as good at mentalising as you commonly think you are (Ames and Kammrath 2004; Epley and Caruso 2008; Hall et al. 2009; Realo et al. 2003). As the philosopher of mind Shannon Spaulding, who has spent much of her career examining this phenomenon, put it:
We consistently and substantially overrate our ability to accurately judge others’ mental states and interpret social interactions… whatever the cause, the consensus from the empirical literature is that mind misreading is very common. (2016: 423)
Philosophers, she goes on, have tended to focus exclusively on the so-called ‘telepathic’ abilities of typical adults while neglecting to focus in any systematic way on the much more common mind misreading that occurs in all of us. Somewhat ironically – Spaulding notes – though philosophers do examine mind-misreading as it exists in chimpanzees, children, and (unsurprisingly once again) those with autism, we disregard it in ourselves, falling victim to the very same Self-Serving Attributional Bias that contributes significantly to our mind-misreading in the first place (2016, (2016: 423). The internal logic of ToM may itself be to blame for this oversight. As David Smukler writes in his essay ‘Unauthorized Minds: How ‘Theory of Mind’ Theory Misrepresents Autism’, ToM models of mentalising are inadequate in large part because they fail to consider what those who supposedly lack this module themselves think and feel (2005: 11-24). In effect, ToM theory is itself “mindblind”. Had we looked to our own experience instead of the one-sided exclusionary rhetoric present in current popular mentalising research, we might have come to appreciate all this better, as the author Philip Roth did:
You fight your superficiality, your shallowness, so as to try to come at people without unreal expectations, without an overload of bias or hope or arrogance, as untanklike as you can be, sans cannon and machine guns and steel plating half a foot thick; you come at them unmenacingly on your own ten toes instead of tearing up the turf with your caterpillar treads, take them on with an open mind, as equals, man to man, as we used to say, and yet you never fail to get them wrong. You might as well have the "brain" of a tank. You get them wrong before you meet them, while you're anticipating meeting them; you get them
II.4 Trustworthiness 212
wrong while you're with them; and then you go home to tell somebody else about the meeting and you get them all wrong again. Since the same generally goes for them with you, the whole thing is really a dazzling illusion empty of all perception, an astonishing farce of misperception. And yet what are we to do about this terribly significant business of "other people," which gets bled of the significance we think it has and takes on instead a significance that is ludicrous, so ill-equipped are we all to envision one another's interior workings and invisible aims? Is everyone to go off and lock the door and sit secluded like the lonely writers do, in a soundproof cell, summoning people out of words and then proposing that these word people are closer to the real thing than the real people that we mangle with our ignorance every day? The fact remains that getting people right is not what living is all about anyway. It's getting them wrong that is living, getting them wrong and wrong and wrong and then, on careful reconsideration, getting them wrong again. (1998: 35)
We get one another wrong, frequently. There is a veil between us. The significance of our so-called ‘telepathic’ or ‘mindreading’ abilities is overstated. In the previous section we saw how Jones, adopting those influential but spurious theories of mentalising, identified our so-called ‘mind-reading abilities’ as being necessary for trust and trustworthiness. This rendered many sections of the population unable to be counted as candidates for trust and trustworthiness. As for everyone else, it now turns out our frequent failures at mentalising would preclude us too. In The Invention of Lying, Gervais has created creatures with no veil between themselves and the rest of the world. This makes them different from us because they can effectively mindread, while we – all of us – cannot. We can make sound inferences about one another, and this ability facilitates us rationally relying on one another. But it is the veil between us that facilitates trust and being worthy of it.
II.4.3 A potential charge of hypocrisy
Aren’t you doing exactly what you have accused others of doing? It seems that in your account, autistics and others with supposed ToM deficits would be precluded as candida tes for trust and trustworthiness based this time on their apparent inability to deceive.
II.4 Trustworthiness 213
II.3 presented evidence that those with supposed ToM deficits, such as those with autism, far from being untrustworthy as was the consequence of Jones’ rich trustworthiness, are instead characteristically honest, typically learn to lie later than their non-autistic peers (Peterson et al. 2005), and prefer to engage with others with candour throughout their lives (Atherton and Cross 2018; Atherton et al. 2018). Historically, this preference for truthfulness has laid such individuals open to the charge that they themselves must lack imagination and so the ability for deception (Baron- Cohen 1992; Oswald and Ollendick 1989). If one adopts this line of thinking, then one would likely come to view such individuals as similar to the characters in The Invention of Lying – open-books with empty pages, “mind blind” versions of the rest of us. As we saw, a number of thinkers have come to this conclusion, imagining those with ToM deficits as effectively real-life philosophical zombies. If this picture were true, then I am indeed doing what I have accused others of doing, only in my account, autistic people (and others with supposed ToM deficits) would be precluded from being candidates for trust and trustworthiness based on their inability to deceive. Hypocrisy is a serious charge, but in this case, it does not stick. First, many of those who were written off by their performance on ToM tests can in fact both use and detect deception in natural settings as proficiently as the rest of us (Li et al. 2011; Lombardo et al. 2016; Martinez-Murcia et al. 2017; Rutherford and Ray 2009), which once again seriously calls into question the relevance of ToM as a theory of human mentalising (Atherton et al. 2018; Gallagher 2004). Indeed, autistics – women and girls in particular – often develop a considerate form of empathetic deception that aims to make non- autistic people feel more comfortable in their presence (Lai et al. 2017). This ‘autistic camouflaging’ fools even the experts quite remarkably, as suggested by Livingston and Happé (2017). Second, there is mounting evidence demonstrating how grossly the private world of many of those with apparent ToM deficits has been mischaracterised. The world of the autistic person, far from being empty and thin, is now known to be significantly more intense and intricate than typical human experience (Mottron et al. 2009; Remington and Fairnie 2017). While autistic behaviour and facial expressions can belie these
II.4 Trustworthiness 214 complex depths, a higher perceptual capacity is a core feature of autism, present from childhood (Swettenham et al. 2014). Furthermore, many of those with apparent ToM deficits appear to have superior emotional processing when it comes to inferring those particularly complex counterfactual emotions – regret and relief – that are no doubt essential for successful moral and interpersonal relations (Black et al.). Many of those who supposedly cannot mentalise are in fact exceptional ‘deceivers’, in the sense that it is we who have failed to understand them, more than the other way around. Yet even if this dissertation had not been able to draw on the substantial empirical evidence here, it would not have mattered. Trust and trustworthiness as this dissertation describe them, as two-place relations, are applicable to any individual that has a private experience of the world that is hidden from the rest of us. This, I take to extend to all human beings. The only entities that the account presented here excludes are those which have no private experience (governments, systems, plastic styracosauri) and the gods. Styro was precluded from the possibility of trustworthiness because he had no private experience of his truster (II.2.1). The gods, if they existed, would be precluded because of their omniscience and omnipresence; the gods know all there is to know about us, which is presumably why they get to judge us, while we are told to refrain from judging one another:
It is the Lord who judges me. Therefore, do not pronounce judgment before the time, before the Lord comes, who will bring to light the things now hidden in darkness and will disclose the purposes of the heart. (Corinthians 4:1-21)
Bringing light to things now hidden in darkness. Disclosing the purposes of our hearts. In a salute of sorts to its own limitations, humanity has long imagined such a being. One who could lift the veil and see us in our completeness, both in terms of what we did and what we would have done, what we are and what we may have become. One who therefore is qualified as ultimate judge, jury, and executioner. Whatever one’s view of religion, the idea is an important one. It leads us to question – if faith and judgement are for the gods, then what is left for us mind-blind “mere human beings” (Romans 2: 1-3)? I say trust and trustworthiness are for us – all of us – and, I say, they are even more spectacular.
II.4 Trustworthiness 215
II.4.4 Løgstrup and trust/worthiness
Apart from saying that they require the veil and are complex two-place relations, can you do any more to further characterise the place of trust and trustworthiness in our lives?
It may be of use here to draw on the similarities between trust/worthiness as they have been presented in this dissertation and the work of the Danish philosopher Knud Ejler Løgstrup. In ‘Trust is Basic’ (2017), Robert Stern made the connection between the philosophical literature on Trust and the central role trust plays in Løgstrup’s major work The Ethical Demand (1956). Note that I refer to the English translations of 1997 and 2007. Stern points out that in contrast to the standard story in modern western philosophy, Løgstrup has provided us with a thought-provoking examination of the place of trust and trustworthiness in our lives that, as Stern notes, has been largely overlooked. Although Løgstrup’s view differs from that presented here in many respects – most notably in the background religious/moral nature of his discussion and his ‘ubiquitousness’ proposal that trust is “essential to every conversation” (1997: 14) which I have argued against in numerous places – the similarities are worth mentioning. Broadly, Løgstrup’s view and my own are alike in that neither of us sees trust/worthiness as epistemic notions, the purpose of which is to extend our agency. Instead, we both highlight trust’s unique relation to love, humanity, and interdependence, and not – as is usually taken to the be the case in contemporary Western debates – the limits of our rationality, epistemological status, moral status, or normative practices (see: I.1 and II.1). Løgstrup says in the 1997 English translation:
Not to let the other person emerge through words, deeds, and conduct, but to hinder this instead by our suspicion and by the picture we have formed of him or her as a result of our antipathy is a denial of life. (1997: 14)
Løgstrup’s remark that trust is so vital to human beings and their relationships that distrust is a “denial of life” is not to be taken in the literal sense (as death). Instead, a “denial of life” is a denial of what a person is or may become. For Løgstrup, trust and distrust, trustworthiness and untrustworthiness, are not three-place relations. Distrust is a “picture we have formed” of one another, while trust is a process of letting the trustee
II.4 Trustworthiness 216 emerge over time. In this way, trust is a form of open-minded and open-hearted monitoring; an ongoing, partly self-reflective process, which involves overcoming our own antipathy, our own impulse to unreflexively accept the suspicious pictures we form of one another. This is very like the position advanced in this dissertation. I have discussed how trust and monitoring are compatible (I.4.4), and how trusting relationships are those in which we create one another (I.3.3), and I have presented an example from The Invention of Lying (II.4.1), where the protagonist’s own ability to be more than as he appears (to step behind his own veil) grants him the perspective that others may also be more than as they appear. On a Løgstrupian interpretation of The Invention of Lying, it could be said that– instead of denying life by accepting the limiting antipathetic ideas he has formed of those around him, the protagonist Mark learns to let others emerge: “you can’t tell just from looking at the guy… he could be the world’s greatest poet” he retorts to Jennifer’s bald observation of “sleeping ugly fatty”. To this I add, that this trustful ‘watch and wait and see’ approach of letting others emerge can be passive or active. Sometimes trust requires of us only that we get out of our own way, overcome our own suspicions and antipathies so that we can see people as they are already showing themselves to be. This is a passive way of letting others emerge as it requires us to engage only with ourselves. Yet, at other times, trust is active, in the sense that our trustful engagement with others comes to play a substantial role in them becoming who they become. In The Invention of Lying, Mark is active in the creation of Jennifer and Frank. At first, Jennifer uncritically adopts the normative framework of the world she lives in, whereby people marry not on the basis of love or friendship, but strictly appearances. Her ideas about her own preferences on this matter are stuck fast, as she communicates to Mark:
I do like you and I do enjoy your company. And if we were to get together and procreate, I would like the offspring that are carrying half my genetic code to be well taken care of, financially stable. I also think you'd make a good father and a good husband, which I like. Unfortunately, none of that changes the fact that you'd still be contributing half the genetic code to our children and I don't want little fat kids with snub noses.
II.4 Trustworthiness 217
Before he is able to lie – to step behind his own veil – Mark can only resignedly accept Jennifer’s attitude as the unshakeable truth. But armed with his own newfound capacity to ‘look again’, Mark is able to charitably afford this capacity to Jennifer too, even in the absence of any evidence that she is capable of it. It is his hopeful rejection of her antipathy towards him that allows her to eventually come to reject this ‘unshakeable’ truth about herself:
You're chubby and you have a snub nose... but you're smart. You're kind. You're the sweetest man I've ever met. You are definitely the most interesting person I know. And you are fun to be with. And you see the world in a way that nobody else sees the world, and I like the way you see the world. And you're my best friend. You make me happier than anyone I've ever known. I know what I want. I want little fat kids with snub noses.
Accentuating the two-place interpersonal back-and-forth nature of the trusting relationship, we see that this capacity to ‘look again’ is one that Jennifer never would have realised without being first credited it by Mark. Her nascent trustworthiness is the result of his nascent trust. We see the same active dynamic playing out with Mark and Frank, with Frank’s own will to live having gone missing in his experience of himself. Mark is able to find Frank’s will to live for him in the form of a lie, which in this case has another name– hope (see II.4.1). If we relate these dynamics back to what Løgstrup says, in The Invention of Lying, Jennifer and Frank have denied themselves. They have failed to embody self-trust, and self-trustworthiness (see further discussion in I.4.5) and so it is their own antipathies and suspicions about themselves that are preventing their emergence. They cannot be worthy of others’ trust, because they do not have self-trust. But the remedy comes not from within the self but from outside. It is just as Reverend Long said: “None of us is complete until there is at least two of us. My centre isn’t in here– it’s between you and me. It is only when I truly engage with you that I begin to find out who I really am. We discover that my humanity is made by yours… and yours by mine” (see: Introduction). When we attribute humanity, morality, or other characteristics or capacities to one who cannot find them in themselves though, we do not give them the relevant capacities. To say this is to attribute a mystical, almost god-like, property to ourselves.
II.4 Trustworthiness 218
By repudiating the suspicions and antipathies others have of themselves we do no more than sweep away the cobwebs, providing them with the freedom to be what they already are or become what they already always could have been. To fail to do this, to fail to trust, is not necessarily to be false, socially unacceptable, or even in many cases immoral. Yet as Løgstrup says, it is always a “denial of life”, as by limiting how we see others we limit how they see themselves and so limit what is possible for them. When we limit others, we also limit ourselves, since so much of what makes a person the irreplaceable individual that they are comes from what happens between them and others in their close interpersonal relationships. Stern nicely summarises this element of Løgstrup’s view:
In these [trusting] relations we take people at face value as they present themselves to us and connect to them directly, rather than forming a certain image or picture of their character, a theory about what makes them tick, and using that to define them for us. (2017: 289)
“A theory about what makes them tick”… another term for this is Theory of Mind (II.3). Theorising about what makes people (or animated triangles) tick may very well be crucial for everyday human social interactions – for analysing, judging, and inferring others' behaviours. Such theorising this dissertation has acknowledged may be useful for extending our knowledge and agency (see II.3) and vital to cooperative endeavours (see I.2.2). Yet because “my humanity is made by yours and yours by mine” theorising about someone, about who they are or what they will do, always involves an infringement on their freedom. Not just because we might be wrong, but because even if we are right, the theory itself denies and limits the person being theorised about. Trusting and being worthy of it, like loving and being worthy of it, are an important alternate way of relating to one another without a theory. As Stern says:
(1) As living creatures, we have the capacity for renewal, which must be realized if our lives are to go well. (2) This capacity cannot be realized if we are confined by the picture or theory imposed on us by others. (3) Trust involves relating to another without a picture or theory. (4) So being trusted by others enables us to function in the right way. (5) To distrust
II.4 Trustworthiness 219
is to impose a picture or theory on someone. (6) So to distrust someone is to risk blocking their capacity for renewal, and thus to prevent their life going well. (2017: 287)
Our trusting hero from Part I (Dorothea) did not have a theory about her trustee (Lydgate). Her distrustful interlocutors (Farebrother and Sir James) did. She was willing to ‘watch and wait and see’, and then, if Lydgate proved to have done wrong, Dorothea was willing to reform him and renew him with her trust, which includes as part of it a hopeful view of him (see: I.4.1). When we refrain from the “impulse to investigate the other person’s character” (Løgstrup 1997: 13), we give people the freedom both to be what they are, and to become something other than what they are. This trustful way of relating to one another is very close to love (see I.4.2), and according to Løgstrup, close to all the “spontaneous expressions of life” including sympathy and mercy:
In love and sympathy there is no impulse to investigate the other person’s character. We do not construct an image of who he or she is … If, on the other hand, we are not in sympathy with the other person … then we begin to form a picture of the other’s character … However, when we are in the direct association with that person this picture usually breaks down; the personal presence erases it … Only where the proof of her unreliability has in the most positive sense become an ingrown distrust, or where the irritation and antipathy have shut me off completely, does the picture continue to stand. (1997: 13)
Reverend Long regards his role serving the needs of Sydney’s homeless as one of meeting people – “seeing them not saving them” (see Introduction) – and this is very like Løgstrup’s idea of being in “direct association” with one another: the personal presence erasing the picture we have formed, the theory we have constructed. In this quotation as well, we find in Løgstrup yet another idea this dissertation has addressed; that it is not always the trustee’s fault that they are not trusted (I.4.2). A distrustful picture of the trustee might be sustained either by evidence: “proof of unreliability” or else by something gone wrong in the truster: “the irritation and antipathy” which can shut us off completely from one another. Stern comments on why he thinks Løgstrup finds it essential to emphasise this point:
As someone who lived through the Nazi experience both in Germany and in Denmark, and the consequent erosion in relations of trust that this entailed, this must be seen as the
II.4 Trustworthiness 220
fundamental lesson Løgstrup learned: not that society could not function in such conditions, because in some sense it did, but that it is still a pathological form of human life, in which important goods were lost as different and ‘deficient’ kinds of inter-relations took hold that required people to be committed to a limited way of understanding one another as living beings. (2017: 289)
This “pathological form of human life” fed by its “limited way of understanding one another” is manifest in the world that Gervais imagines too. Society functions in The Invention of Lying, and on the surface it looks very much like our own: people have houses, jobs, cars, they get married and procreate. Yet they are palpably devoid of Løgstrup’s “spontaneous expressions of life”, sympathy, mercy, love and trust. They have theories of themselves and of one another, granted, based on very good evidence, but this is all they have to navigate their world. This deficient way of relating to people, of failing to understand them as living beings, is evident in the historical treatment of those with autism (see II.3.2) and the historical treatment of minority groups in general. It is how La Salle’s men related to the Illinois Indians (see I.2.2). Conventional social norms, including those of honesty, non-violence, and of doing what we are being counted on to do and signalling so effectively, Løgstrup and I agree (see II.3.1) are not to be identified as trust and trustworthiness. Stern again summarises this element of Løgstrup’s trust:
We do not create or bring about trust as a practice or norm, in the way that we bring about practices or norms like driving on the left, marriage, or even property, which govern our various social institutions in ways that we hope are for the best. These practices or norms are brought into being by us in a contractual or quasicontractual manner and are thus goods that we introduce into the world and over which we have control. But there are other structures which are fundamental to life itself, of which we are part, that we could not bring about in this way as without them we could not come to be at all, and trust (along with the other sovereign expressions of life) are structures of this sort. (2017: 291)
Conventional norms are goods which may induce peace, cooperation, information transfer, and social stability, but we can be without them. We have been without them. Many sectors of the world’s population continue to be without them. Without the
II.4 Trustworthiness 221 intersubjectivity gained by trust and trustworthiness and the other sovereign expressions of life however, we are no longer ourselves. Instead, we are grotesque versions of ourselves living only in the ghoulish imaginations of warped British comedians. Though they possess far superior knowledge about one another than we do, the characters in The Invention of Lying are all dreadfully unhappy, damaged, defeated, and resigned. In Løgstrup’s view and in my own (see I.2, I.3, I.4), trust does not deny knowledge, rather, what we know about a person already is quite irrelevant. Trust requires no justification, and, for Løgstrup, this includes the potential justification that trust is the rational approach as it can help us to retain a positive view of life. According to Løgstrup, if you trust the other for the good such trust brings you, then this is just another way to cut yourself off from the person concerned by focusing in on yourself instead, a point he develops in relation to mercy:
Mercy consists in an impulse to free another person from suffering. If it serves another purpose, such as stabilizing society, it is replaced by indifference towards the other person’s suffering. The ulterior motive transforms mercy into its own opposite. This is why the spontaneous expressions of life [including trust] defy all justification. The very moment we seek to give a reason for them, we make them contingent upon that which we present as our reason, and they become corrupted right then and there. We have made them a means to obtain a goal other than their own: a means for the goal that is present in the justification. (2007: 128)
This dissertation has argued that trust and trustworthiness are essentially personal private experiences (I.4.2), that do not preclude selfishness (II.1.4). This may seem to be a point of difference between Løgstrup and myself. Yet I have also continuously maintained that trust and trustworthiness are not inwardly focused. The focus of them both is always the object of trust/worthiness – the truster/trustee. Being two-place relations, trust and trustworthiness involve the experience two individuals have of one another and as such, they are wholly collaborative and interactive (see I.3, I.4.2, and II.2.1), aside from the pathological case where things go terribly wrong (see I.4.2). In the normal case, people who trust and are worthy of it emerge together, in response to one another, and, as Løgstrup says in relation to mercy, their motives can transform
II.4 Trustworthiness 222 them: trusters into distrusters, those who are worthy into those who are unworthy (as I have argued in II.2.2, and II.2.3). To summarise, trust and trustworthiness arise, not because of the need we have to count on one another, but because of the situation we are in: one in which we can never fully see, never fully comprehend, another person. One in which we can always be wrong about someone, even if we have got them right today. A situation in which we have hidden depths, private experiences, and so, the freedom to let one another emerge, the freedom to help one another emerge. Refraining from letting those other limiting ways of regarding one another rule, with theories and rationalisations, is trust. Love gives us this extraordinary capacity too. On these points, the Danish philosopher Knud Ejler Løgstrup and I appear to be in agreement.
II.4.5 The freedom to look again
But surely it is what a person did that determines their trustworthiness, otherwise, how do you explain the fact that our trust is broken when people do or fail to do certain things?
To answer this, it the inherently reciprocal nature between trust and the participant stance. In “Freedom and Resentment” (1962), addressing directly the free-will/determinism debate, P.F. Strawson showed us that to hold one another accountable in the ways that we do, we must – in some important sense – be free. My purpose in relating my own work to Strawson’s is not to make claims about what Strawson would have said of my trust cases (we cannot know), but to make a new claim– that whatever Strawson’s view, navigating our trusting relationships does indeed involve the kind of shifting between the objective and participant stance that I characterise. In the trust cases I am interested in, which are different from Strawson’s own cases, we shift back and forth between the objective and the participant stance: indeed, this shifting is part of the trust-process. The applicability of Strawson’s discussion to cases such as this is bolstered by interpretations taken from secondary literature, for example, as discussed by Snowdon and Gomes:
Strawson’s purpose [in “Freedom and Resentment”] is to dissolve the so-called problem of determinism and responsibility. He does this by drawing a contrast between two different
II.4 Trustworthiness 223
perspectives we can take on the world: the ‘participant’ and ‘objective’ standpoints. These perspectives involve different explanations of other people’s actions. From the objective point of view, we see people as elements of the natural world, causally manipulated and manipulable in various ways. From the participant point of view, we see others as appropriate objects of ‘reactive attitudes’, attitudes such as gratitude, anger, sympathy and resentment, which presuppose the responsibility of other people. These two perspectives are opposed to one another, but both are legitimate. In particular, Strawson argues that our reactive attitudes towards others and ourselves are natural and irrevocable. They are a central part of what it is to be human. The truth of determinism cannot, then, force us to give up the participant standpoint, because the reactive attitudes are too deeply embedded in our humanity. Between determinism and responsibility there can be no conflict. (2019 Section 8.5)
Setting aside the issue of determinism, one influential view of trust (as discussed in 1.2.1) is that trust is an action that we take (reliance) from the participant stance. To refresh, on this view, when we trust, we rely, and open ourselves up to experiencing reactive attitudes – i.e. resentment and betrayal – should our trustee fail to do what we are entrusting them to do. Richard Holton is celebrated for having first made the connection between Strawson’s insight and trust in “Deciding to Trust, Coming to Believe” (1994), but more recently, Stephan Darwall has argued that actually Baier and Jones share with Holton a similar view of trust as a Strawsonian reactive attitude (2017). A number of others have since adopted this view. This has all been discussed already in I.1.1. The new idea explored in this subsection is that, while Strawson’s insight can shed some light on an interesting feature of our trusting relationships, the widespread adoption of three-place trust/worthiness has led the connection between the participant stance and trust/worthiness to be mischaracterised. Trust and trustworthiness are not ‘actions we take from the participant stance’; rather, trust and trust worthiness involve experiencing one another, which itself involves navigating the complex problem of when and when not to take the participant stance towards one another. This proposal goes against common wisdom on the nature of the connection, precisely because those who have made the connection between trust and Strawson’s
II.4 Trustworthiness 224 argument in “Freedom and Resentment” take trust to be, in Darwall’s words, a “species of reliance” (2017: 35) that is seen ultimately as a kind of reaction toward the behaviour of others. Indeed, the freedom relevant to trust has been understood as the freedom to entrust and the freedom relevant to trustworthiness as the freedom relevant to doing what is entrusted. According to Holton:
You let yourself fall because the others will catch you. Or at least that is what they told you they would do. You do not know that they will. You let yourself fall because you trust them to catch you… Resentment and gratitude are examples of the particular attitudes that we feel towards people when they act [my emphasis] in certain ways… if the people in these dealing were to hurt us, we would typically feel resentment; and if they were to help, we would typically feel gratitude. We are ready to take particular reactive attitudes should they act [my emphasis] in certain ways. (1994: 63-67)
Here, Holton views trust in that classic three-place cognitive sense; something that we do when we lack knowledge of how others will behave, hence, trustworthiness is doing what others expect us to do. On any three-place understanding, Strawson’s participant attitudes are relevant to trust if we understand trust to be an action that we take ‘from the participant stance’, and so a trustee is worthy of trust when they take action, doing what we entrust them to do. However, to view trust this way is to ignore, I think, the important nuance in Strawson’s point. As he says, the participant attitudes reflect:
how much we actually mind, how much it matters to us, whether the actions of other people—and particularly some other people—reflect attitudes towards us of good will, affection, or esteem on the one hand or contempt, indifference, or malevolence on the other. (1962: 5)
What we actually mind is not the actions but what the actions reflect: how others experience us – with goodwill, affection, esteem, or contempt, indifference, malevolence. Does our trustee experience us as Dorothea experiences Lydgate, or as Sir James experiences Lydgate? Is Rasmus telling the truth because he is Iris’ friend or because he is pretending to be Iris’ friend? Though the language is couched in terms of action, we do not take the participant stance towards people’s actions, but towards
II.4 Trustworthiness 225 people. We react to what they think and feel about us, or rather, given the veil, our best guess of what they think and feel about us. This facet, the veil, again, is all-important. It matters that we might be wrong about one another, indeed, it is precisely this that allows us the freedom to choose when to take the objective stance, and when to take the participant stance. We can decide to adopt objective thinking: ‘he couldn’t help it– it was an accident’ or participant thinking: ‘you betrayed me– you did it on purpose’. Trust is not a matter of adopting participant thinking already, before anything happens, and standing ready to experience betrayal should our trustee fail (Holton 1994: 66); trust can just as much require us to remain steadfast in objective thinking, even when the trustee fails, even when it seems without question that we have been betrayed. We can see this in The Invention of Lying. In Gervais’ big-screen thought- experiment, we are confronted with characters who do have freedom of action, as well as the capacities for reflection, deliberation, decision-making, and self-determination on a level not too dissimilar from our own. They can “decide to trust”, in Holton’s sense that they can decide to fall backwards, counting on someone to catch them, and they can “decide to be trustworthy” in the sense that they can decide to catch the person who is counting on them. It is true that they differ from us in possessing superior knowledge of their trustee’s intentions (at least at the time of asking), but they still lack certain knowledge of the outcome. Intentions can change in an instant, and, in the trust game, all sorts of extraneous things can intervene in the moments as we are letting ourselves fall, meaning that they cannot predict the future any better than we can. As such, the Trust Game works (involves risk) in their world too, requiring them, like us, to take a leap of uncertainty. Yet despite all these similarities in agency and epistemic status, what we see in The Invention of Lying are characters who do not take the participant stance towards one another. With no veil, there is defeated resignation instead of moral outrage; hurt acceptance in place of resentment. They treat one another as elements of the natural world, causally manipulated and manipulable in various ways. Of course, unlike us they lack the freedom to speak lies, but lacking this freedom alone cannot fully explain why they only take the objective stance towards one another. Imagine that our friend (or
II.4 Trustworthiness 226 someone we thought to be our friend) is drugged somehow to speak only truth. In this single way they would be very much like the characters in The Invention of Lying. Yet if our friend were to say something hurtful to us that they could not help saying while drugged, then we may still be upset at them; not for saying it – they couldn’t help that – but even just for thinking it. In our world, it is not only important how others treat us, but how they privately regard us, how they experience us. We hold people responsible not only for their actions, but for their opinions, beliefs, and attitudes (A. M. Smith 2005). This makes sense, given the freedom we have in this regard. Take for example, that reactive attitude so often linked in the literature to trust (see I.1)– betrayal. When it comes to betrayal, infidelity must be considered the prime example. In The Invention of Lying, an unfaithful lover is just an unreliable monogamous partner. Without the accompanying lying or sneaking around, the harm ends with the act. This is comically captured in a marriage ceremony scene late in the film, where the couple’s ‘vows’ consist of the rather feeble ‘promise’ to stay with one another for “as long as they want to”. In our world though, the accompanying deception is by far the most damaging aspect of infidelity. As infidelity researcher Lisa Firestone explains, the real villain behind infidelity is not the affair [the sexual act] itself, but the many secrets and deceptions built around it:
Deception and lies shatter the reality of others, eroding their belief in the veracity of their perceptions and subjective experience. The betrayal of trust brought about by a partner's secret involvement with another person leads to a shocking and painful realization on the part of the deceived party that the person he or she has been involved with has a secret life and that there is an aspect of his or her partner that he or she had no knowledge of. (Firestone et al., 2006)
We can shatter one another’s reality because our reality with regard to one another, unlike theirs, is made of glass-guesses. When explaining our reactive attitudes – why we feel betrayed – we tend to speak in terms of what he did… what she said… how they behaved. ‘I’m resentful because she lied’, we say. But the act is just the catalyst. More precisely, when we take the participant stance towards someone, what we are really reacting to is our own ideas about them, our own interpretation of them, of who they are
II.4 Trustworthiness 227 in the greater scheme of their reasons, motives, and emotions and it is most important, most vital, that it is a guess, that we might be wrong. In Gervais’ world, people fail to take anything other than the objective stance towards one another not because they have no freedom with regard to action, but because they cannot choose to either: imagine that things are other than as they appear, give the benefit of the doubt, use their reason to let that which they do not understand pass unjudged, try to move into a more optimistic frame of mind about one another, reflect on their own motivations towards one another, try to overcome their own emotional responses towards one another, or look again and see one another in a better light. They cannot be held accountable for their failures of trust and trustworthiness, not because they cannot be held accountable for their actions – for the most part they can – but because they have no freedom with regard to their own experience of one another. Without hidden private subjectivity, they manage only the ‘purely objective’ view of one another as posing problems simply of ‘intellectual understanding, management, treatment, and control’ (Strawson 1962: 13). Just like the people in The Invention of Lying, we too can respond to evidence about a trustee’s trustworthiness. This is to have our mind changed for us about them. But unlike them, who only have this passive avenue of change open to them, we have the freedom (within personal limits) to change our own mind about a person, without any outside help at all. This is not the same as blind faith, for it is not about closing our eyes to truth. Rather, it is a matter of holding appearances firmly in one eye’s gaze, while using the other eye to see something – even just some remote possibility – that was obscured in bare-blinding reality. This “looking again” is discussed by Iris Murdoch in her development of an ethics of vision, as opposed to an ethics of choice (1957). This ethics of vision for Murdoch involves the human capacity to “picture ourselves” and then come to resemble the picture (Murdoch 2001: 234) and so in this way, is very like Løgstrup’s ethics. Strawson’s argument in “Freedom and Resentment” can shed some light on trust and trustworthiness, as it helps us to see that navigating our trusting relationships requires us to treat one another in various ways depending on the context, sometimes holding one another to account for our failures of reliability, sometimes not. But trust is
II.4 Trustworthiness 228 no more ‘reliance from the participant stance’ than it is ‘reliance from the objective stance’. Moreover, there is one worrying part of Strawson’s discussion that does not sit well with the predominant message of Part II of this dissertation. The participant stance is something we take towards some groups of people and not others:
We can have direct dealings with human beings without any degree of personal involvement, treating them simply as creatures to be handled in our own interest, or our side’s, or society’s—or even theirs. In the extreme case of the mentally deranged, it is easy to see the connection between the possibility of a wholly objective attitude and the impossibility [my emphasis] of what we understand by ordinary interpersonal relationships. Given this latter impossibility, no other civilized attitude is available than that of viewing the deranged person simply as something to be understood and controlled in the most desirable fashion. To view him as outside the reach of personal relationships is already, for the civilized, to view him in this way. For reasons of policy or self-protection we may have occasion, perhaps temporary, to adopt a fundamentally similar attitude to a ‘normal’ human being; to concentrate, that is, on understanding ‘how he works’, with a view to determining our policy accordingly, or to finding in that very understanding a relief from the strains of involvement. (1962: 11)
Strawson says that the participant stance is impossible to take towards some groups of individuals, i.e. the “mentally deranged”. This makes the alternative objective stance, in Long’s words, to treat a person as a problem to be solved, rather than as a person to be met (Long 2017). Any account that makes trust into ‘reliance from the participant stance’ will end up being exclusionary, just as Jones’ account is (II.3). It may well be “civilized”, but it is not human to treat anybody as entirely outside the reach of personal relationships. Taking relief from the strains of personal involvement is something we do with everyone on occasion (particularly our in-laws). However, if taking the objective stance becomes something that we do consistently with any individual or any particular group of individuals, then we are refusing those individuals participation in human relationships, and so are dehumanising them. Moreover, there is a treacherous (but ultimately just) twist to dehumanisation: the dehumaniser is thereby dehumanised. Coming full circle back to that rich source of philosophical exchanges Star Trek: Next Generation (Berman 1990), I end with some reflections between the non-human
II.4 Trustworthiness 229 android Lt. Commander Data (capable only of the objective stance), and his very much human commander, Commander William T. Riker:
Riker: In all trust, there is the possibility of betrayal. I'm not sure you were... prepared for that.
Data [an android]: Were you prepared, sir?
Riker: I don't think anybody ever is.
Data: Hmm... Then it is better not to trust?
Riker: Without trust, there's no friendship, no closeness. None of the emotional bonds that make us who we are.
Data: And yet you put yourself at risk.
Riker: [smiles] Every single time.
Data: Perhaps I am fortunate, sir, to be spared the emotional consequences.
Riker: Perhaps.
To summarise, if Strawson’s discussion in “Freedom and Resentment” illuminates our trusting relationships, then it is not as a dichotomous divide – the trustworthy on the participant-side and the untrustworthy on the objective-side – but as a reminder of how our trusting relationships involve the continual, alert-but-blind navigation of this divide. Taking the objective stance spares us the emotional consequences, but it dehumanises others, and in doing so ourselves. This is most unfortunate.
II Summary of Part II, On Trustworthiness 230
Summary of Part II, On Trustworthiness
Section I.1 reviewed the philosophical literature on Trustworthiness. Section II.2 demonstrated the limitations of the current dominant understandings of trustworthiness in the philosophical literature. Section II.3 highlighted the potential ethical problems with current dominant understandings of trustworthiness via one influential philosopher’s view. Section II.4 made use of the film The Invention of Lying to further characterise the role that trustworthiness plays in our lives. Two-place trustworthiness, as it was presented in this dissertation, is one way we might experience and so relate to a special individual in our lives; one who trusts us. Trusters and trustees are all of us who live a veiled existence between the zombies and the gods. Trust acknowledges: ‘I can’t see you; you could be anything you like, but here I am choosing to make myself vulnerable to you’. Trustworthiness answers: ‘You can’t see me, I could be anything I like, but here I am choosing to make myself worthy of you’.
Conclusion
To trust … is to lay oneself open (Løgstrup 1997)
This dissertation has two symmetrical parts. Part I, Trust, argued against the dominant epistemic view of trust in philosophy – three-place trust – and suggested a paradigm shift. So-called ‘three-placed trust’ is better understood not as a kind of trust, but as a kind of reliance. Trust is instead presented as a two-place relation, and this dissertation further characterised this bare schematic. Part II, Trustworthiness, has also argued against the dominant view of trustworthiness in philosophy, paying particular attention to the potential ethical implications of the dominant view, rather than the epistemological implications focused on in Part I. Trustworthiness in Part II has been presented as the other half of a trusting two-place relation, and again this dissertation further characterised this bare schematic. This conclusion proceeds with a more detailed synopsis of this dissertation in the section that follows, Summary; then provides recommendations for further work in this area in Future Research; and finishes with some final Closing Remarks.
Summary: On Trust and Trustworthiness
Both Part I and Part II began with an extensive review of the relevant philosophical literature on trust and trustworthiness respectively. In I.1 Trust: in the Philosophical Literature, this dissertation has demonstrated that the dominant view in philosophy is that trust is a three-place relation. Examination of the literature has revealed that the dominance of three-place trust led to ‘The Problem of Trust’ being understood as an epistemic problem, involving a conflict between what trust requires of us with regard to risk and evidence, and what rationality requires of us with regard to risk and evidence, leading trust to become confused with both reliance (implicating our practical reasoning) and belief (implicating our theoretical reasoning). In II.1 Trustworthiness: in the Philosophical Literature, this dissertation has demonstrated a symmetrical three- place dominance, though in the case of trustworthiness one-place views that understand
231 Conclusion Summary: On Trust and Trustworthiness 232 trustworthiness as a moral notion – a virtue – have been shown to have been given more prominence. Sections I.1 and II.1 thus have effectively set the stage for the arguments that follow, renouncing the dominant views. Following on from their respective literature reviews, both Part I and Part II have presented problems with, and limitations of, the dominant views as laid out. Dominant Views on Trust: Problems (I.2) considered the notion of three-place trust in detail, examining the reasons for three-place dominance and the various manifestations of its forms. The more simplistic forms were shown to be indistinguishable from reliance and cooperation and were rejected on these grounds. This dissertation has demonstrated that the more intricate ‘three-place trust plus’ accounts found in the literature aim to fix the limitations of simplistic three-place trust and acknowledged that such accounts have much to add to our understanding of trust. Ultimately, though, this dissertation has demonstrated that by retaining the three-place schematic, ‘three-place trust plus’ accounts do not manage to avoid the problems faced by their more simplistic counterparts. This dissertation was able to incorporate what was found to be of merit in these ‘three-place plus trust plus’ accounts into its own two-place account presented in Trust (I.4). In Dominant Views on Trustworthiness: Problems (II.2) this dissertation has highlighted and addressed the limitations of one-place trustworthiness and three-place trustworthiness. With regard to one-place trustworthiness, this dissertation has revealed that unlike virtues or character traits, which are (or aim to be) one-place, trustworthiness takes an object, and so cannot be a virtue or character trait. With regard to three-place trustworthiness, this dissertation has made use of an original thought experiment involving two individuals (a truster and a trustee) to demonstrate that although various versions of three-place trustworthiness do manage to pick-out trustworthiness in some circumstances, they cannot do so consistently, or without error. Three-place trustworthiness was rejected on those grounds. In I.3 The Object of Trust and II.3 Rich Trustworthiness or Poor Trustworthiness? the symmetry between Parts I and Part II diverged, with I.3 having undertaken an exposition of the underlying logic of three-place trust, and II.3 having undertaken an examination of the ethical implications of three-place trustworthiness. This reflected the parallel divergence in how the concepts of trust and trustworthiness have been treated
Conclusion Summary: On Trust and Trustworthiness 233 by philosophy: trust is more often understood as an epistemic notion with epistemic consequences, and trustworthiness as an ethical notion with ethical consequences. To that end, section I.3 demonstrated an underlying ambiguity in the logic of three-place trust regarding the object of trust. This dissertation has offered a remedy to this situation by providing an alternative way of understanding so-called three-place trust and trustworthiness as kinds of reliance/reliability that may have involved trust/worthiness, rather than as kinds of trust/worthiness that have definitionally involved reliance and reliability. In section II.3 this dissertation has used an influential example from the philosophical literature on trustworthiness to draw attention to the potential ethical consequences of adopting a three-place view of trustworthiness that is rooted in behaviour. Specifically, by framing trustworthiness as a matter of a trustee’s success or failure at adhering to certain social norms of behaviour, many groups of people have been precluded from being considered as candidates for trustworthiness based on criteria that are 1) orthogonal to their worthiness of trust, and 2) potentially harmful. This dissertation established that these grounds are sufficient for rejecting behavioural- based three-place accounts of our worthiness of trust. The symmetry in this dissertation re-emerged in sections I.4 Trust and II.4 Trustworthiness, with both these sections having further characterised the bare two- place schematic argued for in the previous sections. To do this, I.4 made use of an excerpt from George Eliot’s novel Middlemarch, and II.4 made use of the Ricky Gervais film The Invention of Lying. Both these humanistic art forms, in their very different ways, were shown to be expositions of trust and trustworthiness as they actually exist in human relationships, and they both shed light on the severe limitations of trust and trustworthiness understood as moralised or cognitively informed actions. In addition to making use of these humanistic resources to further characterise two-place trust and trustworthiness, I.4 and II.4 both finished by addressing some potential criticisms and implications of the shift to a two-place view. The shift from three/one-place trust/worthiness to two-place trust/worthiness, as presented in this dissertation, has allowed us to divorce epistemological and ethical situations, implications, and questions that arise in our interactions with one another from the interpersonal relationship itself, while acknowledging that such situations,
Conclusion Further Research: On Trust and Trustworthiness 234 implications, questions, and dilemmas do arise in and because of our interpersonal relationships, and do affect them. Truth, morality, reason, duty, love, and trust are intermingling and yet separate forces pushing and pulling us about in complex ways, and so this shift should be of use to ethicists and epistemologists who tackle these complex and interrelated issues.
Further Research: On Trust and Trustworthiness
The conceptual shift from three-place trust/worthiness to two-place trust/worthiness presented in this dissertation has left much room for further characterisation. Future work could investigate, for example, the various ways in which being parties to trusting relationships affects our moral and rational decision-making. When trust/worthiness is non-moralised and non-cognitive, the interplay between the trusting, the moral, and the rational elements becomes more intricate and interesting, as discussed briefly in subsection II.2.1. Two-place trust and trustworthiness as they have been presented in this dissertation could also play an important role in psychology, not only in accounting for the success and break-down of relationships as they do now, but potentially also in helping to frame some forms of mental illness and treatment, especially in relation to the notion of complex self-trust, as touched upon in subsection I.4.5. As some aspects of trust and trustworthiness are voluntary according to a two-place account and some are not, as discussed in subsection I.1.4, philosophers may be interested in which aspects are which, and how these voluntary and non-voluntary components interact to bolster or undermine our trusting relationships. This dissertation has highlighted the similarities between trust and love, but it did not examine the relationship between them. This would be another important area of investigation; for example, philosophers may ask: which is primary (trust or love), can either one survive the loss of the other, and is love a two-place relation as well? Indeed, more work needs to be done than was possible in this dissertation to characterise the symmetrical nature of trust and trustworthiness as two-halves of a two-place relation,
Conclusion Closing Remarks: On Trust and Trustworthiness 235 and whether love has two similar halves (love and love-worthiness) is an interesting proposal. This dissertation has provided numerous reasons for rejecting the increasingly held view that trust should be given a wide-scope interpretation so as to be able to explain all instances of cooperative behaviour. This may provide the beginnings of a framework for researchers to separate out broad categories of causes underlying cooperative endeavours, with trust being only one possible cause among many: selfishness, mutual- benefit, boredom, coercion, mercy, love, sympathy, etc. This work should be of use to historians as well as to political and social scientists, or any researcher faced with the difficult task of elucidating human behaviour both on an individual as well as a collective scale. Lastly, given the ethical implications addressed in section II.3 in relation to one influential account, it would be pertinent for philosophers to investigate whether any other theories of either trust or trustworthiness that have come out of philosophy have similar ethical implications. As suggested in II.3, it is my hunch that they will. On this point, it would be in line with the message of inclusivity brought forth in this dissertation for philosophers to reflect on how mind-blindness, rather than mind- reading, gives rise to important aspects of our humanity, since unlike so-called “mind- reading”, mind-blindness – the veil – is a feature of humanity that is not exclusionary, but is one we all share.
Closing Remarks: On Trust and Trustworthiness
This dissertation opened with some thoughts from Pastor Graham Long of the Wayside Chapel in Sydney’s Kings Cross. As he is, in a way, this dissertation’s ambassador for two-place trust and trustworthiness, returning to Long is a fitting way to close. The “philosophy of Wayside” – which is a way of life for Long – does not come from any “tricky seven-step answer that we worked out of some, you know, helping book or The Bible or anywhere else; it just isn’t there” (Long 2017). This way of life is not religious or moral, but is instead a humanistic way of treating those who comes to them for assistance as a person to be met rather than as a problem to be solved:
Conclusion Closing Remarks: On Trust and Trustworthiness 236
If they come and they’re driven by: what can you give me? you know, we haven’t got a lot to give. If you arrive thinking this is an intensive care unit so that we can make you feel better in your lostness it’ll be very disappointing because we’re not here to meet all your needs. We’re here to meet you. Call you out into a healthier place which is a place of community where you take responsibility for yourself. In other words, you are responsive to others. That’s a much more difficult journey than… just give me this… just give me that… everything I know about me has come from others and... and… it’s always two-way.
This dissertation, at heart, has been a plea to shift how we understand trust and trustworthiness in philosophy from solutions to the problems other people create in our lives – problems of an ethical, agential, epistemological, or risk/reward nature – to a way of meeting people, creating one another in our experience of one another. The bare schematics of this shift went from one/three-place trust/worthiness to two-place trust/worthiness. Fleshed out, this means that rather than being attributes that we have (one-place) or behaviours that we do (three-place), trust and trustworthiness are two- place relations that emerge when we lay ourselves open and let others emerge. Trust and trustworthiness meet a person where they are at, with open eyes, heart, and mind, blinded not by faith, nor by the facts as they are, nor by the judgments we might have carried. Meeting a person in this manner requires us to acknowledge the veil between us: the veil that means there is always a chance we have got one another wrong; always a chance that if we look harder, we might see something different, something new, something that we missed before. The rewards of this way of experiencing another human being can never serve as the reason why we do it, but they are manifest, and so can be explicated. Describing the rewards of the experience of being trusted, Long recounts to Richard Fidler his attendance at the death of a newborn baby whose parents had found out days before he was born that he would not survive due to a rare genetic condition:
The mum sang him sweet songs and eventually his life ebbed away. And, the incredible honour it was to be in that room. I felt like I received something even though we lost this little boy. The trust these people put in me to just be there and be a… a tower of weakness for them… that’s what I do. I don’t have magic tricks. I don’t know anything. I’m just there. And I was there with these people in this awful moment and in some ways, I think they
Conclusion Closing Remarks: On Trust and Trustworthiness 237
invited me in because they knew I wasn’t going to tell them this all happened for a good reason. I was just with them. And the trust they put in me was, man… one of the treasures of my life. I have no right to be there. That they would ask me to be there is just… an honour beyond words.
Two-place trust and trustworthiness, as presented in this dissertation, do not require us to know anything or have any magic tricks. The relationship of trust is special, but we do not need to display any special skill at social norm-following, communication, moral development, intelligence, or rationality to be let in. These things can enhance, and they can undermine our trust and trustworthiness, which require our presence not our reasons, our openness not our action, our vulnerability as much as our fortitude. The complicated way in which two individuals’ lives can entangle means we can experience almost anything in our trusting relationships, our loving relationships. These relations will break any conditions we place on them in our attempts to define them. Their ability to do this, to expand our experience in wholly unexpected and contradictory ways, is precisely why we value them so much.
* * *
Bibliography
Abell, F., Happé, F., and Frith, U. (2000), 'Do triangles play tricks? Attribution of mental states to animated shapes in normal and abnormal development', Cognitive Development, 15 (1), 1- 16.
Abramson, Kate and Leite, Adam 'Love, Value, and Reasons', The Oxford Handbook of Philosophy of Love.
Adler, Jonathan (2013), 'Epistemological problems of testimony', Stanford Encyclopedia of Philosophy
Ames, Daniel R and Kammrath, Lara K (2004), 'Mind-reading and metacognition: Narcissism, not actual competence, predicts self-estimated ability', Journal of Nonverbal Behavior, 28 (3), 187-209.
Andersen, Hanne and Hepburn, Brian (2018), 'Scientific Method', Stanford Encyclopedia of Philosophy
Athanassoulis, Nafsika (2000), 'A Response to Harman: Virtue Ethics and Character Traits', Proceedings of the Aristotelian Society, 100, 215-21.
Atherton, Gray and Cross, Liam (2018), 'Seeing More Than Human: Autism and Anthropomorphic Theory of Mind', Frontiers in psychology, 9, 528-28.
Atherton, Gray, et al. (2018), 'What am I thinking? Perspective-taking from the perspective of adolescents with autism', Autism, 1362361318793409.
Attwood, Tony (1997), Asperger's syndrome: A guide for parents and professionals (Jessica Kingsley Publishers).
Baier, Annette (1986), 'Trust and Antitrust', Ethics, 96 (2), 231-60.
--- (1991a), 'Sustaining Trust', Tanner Lectures on Human Values (13; Salt Lake City: University of Utah Press).
--- (1991b), 'Trust and Its Vulnerabilities', Tanner Lectures on Human Values (13; Salt Lake City: University of Utah Press).
--- (1994), Moral prejudices: Essays on ethics (Cambridge, MA: Harvard University Press).
--- (2004), 'Demoralization, trust, and the virtues', Setting the moral compass: Essays by women philosophers, 176-90.
Baker, Judith (1987), 'Trust and rationality', Pacific philosophical quarterly, 68 (1), 1-13.
238 Bibliography 239
Bargiela, Sarah, Steward, Robyn, and Mandy, William (2016), 'The Experiences of Late- diagnosed Women with Autism Spectrum Conditions: An Investigation of the Female Autism Phenotype', Journal of Autism and Developmental Disorders, 46 (10), 3281-94.
Baron-Cohen, Simon (1992), 'Out of Sight or Out of Mind? Another Look at Deception in Autism', Journal of Child Psychology and Psychiatry, 33 (7), 1141-55.
--- (1997), Mindblindness an Essay on Autism and "Theory of Mind" (The MIT Press).
Baron-Cohen, Simon, Leslie, Alan M, and Frith, Uta (1985), 'Does the autistic child have a “theory of mind”?', Cognition, 21 (1), 37-46.
Bates, Robert H. (1988), 'Contra contractarianism: some reflections on the new institutionalism', Politics Society, 16 (2-3), 387-401.
Becker, Lawrence C (1996), 'Trust as noncognitive security about motives', Ethics, 107 (1), 43- 61.
Bedny, Marina, Pascual-Leone, Alvaro, and Saxe, Rebecca R. (2009), 'Growing up blind does not change the neural bases of Theory of Mind', 106 (27), 11312-17.
Ben-Ze'ev, Aaron (2001), The subtlety of emotions (MIT press).
Berg, Joyce, Dickhaut, John, and McCabe, Kevin (1995), 'Trust, Reciprocity, and Social History', Games and Economic Behavior, 10 (1), 122-42.
Berman, Rick (1990), 'Star Trek: The Next Generation', in Robert Scheerer (ed.), Legacy (4: Paramount).
--- (2005), 'Star Trek: Enterprise', in Allan Kroeker (ed.), These Are the Voyages... (4: Paramount).
Bibby, Helen and McDonald, Skye (2005), 'Theory of mind after traumatic brain injury', Neuropsychologia, 43 (1), 99-114.
Bishop, John (2017), 'Faith', Stanford Encyclopedia of Philosophy
Black, Jo, et al. 'Intact counterfactual emotion processing in autism spectrum disorder: Evidence from eye-tracking', Autism Research, 0 (0).
Blackburn, Simon (1998), Ruling passions (Clarendon Press Oxford).
Blair, R. James R. (1996), 'Brief report: Morality in the autistic child', Journal of Autism Developmental Disorders, 26 (5), 571-79.
Blais, Michel J. (1987), 'Epistemic Tit for Tat', The Journal of Philosophy, 84 (7), 363-75.
Botsman, Rachel (2017), Who can you trust?: How technology brought us together and why it might drive us apart (Hachette UK).
Bibliography 240
Brewer, Rebecca and Murphy, Jennifer (2016), 'People with autism can read emotions, feel empathy', Scientific American, 12.
Brewer, Rebecca, et al. (2016), 'Can neurotypical individuals read autistic facial expressions? Atypical production of emotional facial expressions in autism spectrum disorders', Autism Research, 9 (2), 262-71.
Brogaard, Berit (2015), On romantic love: Simple truths about a complex emotion (Philosophy in Action).
Brown-Lavoie, S. M., Viecili, M. A., and Weiss, J. A. (2014), 'Sexual Knowledge and Victimization in Adults with Autism Spectrum Disorders', Journal of Autism Developmental Disorders, 44 (9), 2185-96.
Buunk, B. P. and Schaufeli, Wilmar (2011), Reciprocity in Interpersonal Relationships: An Evolutionary Perspective on Its Importance for Health and Well-being (10).
Calhoun, Cheshire and Solomon, Robert C. (1984), What is an Emotion?: Classic Readings in Philosophical Psychology (Oxford University Press).
Campbell, Tim and McMahan, Jeff (2010), 'Animalism and the varieties of conjoined twinning', Theoretical Medicine and Bioethics, 31 (4), 285-301.
Carter, Judy, Irani, George, and Volkan, Vamik D (2015), Regional and Ethnic Conflicts: Perspectives from the Front Lines (Routledge).
Cassidy, Brittany S., et al. (2017), 'Configural face processing impacts race disparities in humanization and trust', Journal of experimental social psychology, 73, 111-24.
Chance, Paul (1974), 'A conversation with Ivar Lovaas about self-mutilating children and how their parents make it worse', Psychology Today, (January 1974), 76-84.
Coady, Cecil A. J. (1992), Testimony: A Philosophical Study (Oxford: Clarendon Press).
Coeckelbergh, Mark (2012), Growing Moral Relations: Critique of Moral Status Ascription (Palgrave-Macmillan).
Cogley, Zac (2012), 'Trust and the trickers problem', 53 (1), 30-47.
Cohen, Marc A. and Dienhart, John (2013), 'Moral and Amoral Conceptions of Trust, with an Application in Organizational Ethics', Journal of Business Ethics, 112 (1), 1-13.
Coleman, James S. (1990), Foundations of social theory (Harvard University Press).
Currie, Gregory and Ravenscroft, Ian (2002), Recreative minds: Imagination in philosophy and psychology (Oxford University Press).
Dalton, Kim M, et al. (2005), 'Gaze fixation and the neural circuitry of face processing in autism', Nature neuroscience, 8 (4), 519.
Bibliography 241
Dargis, Manohla 'Ricky Gervais on Why the Awful Truth Is Awful',
Darwall, Stephen (2017), 'Attitude (of the Heart)', in Paul Faulkner and Tom Simpson (ed.), The Philosophy of Trust (Oxford University Press), 35.
Dasgupta, Partha (1988), 'Trust as a commodity', in D. Gambetta (ed.), Trust: Making and Breaking cooperative relations (47; New York: Basil Blackwell), 72.
--- (2000), 'Trust as a commodity', Trust: Making and breaking cooperative relations, 4, 49-72.
Daukas, Nancy (2006), 'Epistemic Trust and Social Location', Episteme, 3 (1-2), 109-24.
Davidson, Larry and Strauss, John S. (1992), 'Sense of self in recovery from severe mental illness', British Journal of Medical Psychology, 65 (2), 131-45. de Sousa, Ronald (1987), The Rationality of Emotion (Cambridge, MA: The Massachusetts Institute of Technology Press).
--- (1990), The rationality of emotion (The MIT Press).
--- (2015), Love: A very short introduction (415: Oxford University Press, USA).
DeLongis, Anita, Folkman, Susan, and Lazarus, Richard S (1988), 'The impact of daily stress on health and mood: psychological and social resources as mediators', Journal of personality and social psychology, 54 (3), 486.
Derrida, Jacques (1998), Acts of religion (Routledge).
Diekhöner, Philipp Kristian (2017), The Trust Economy: Building strong networks and realising exponential value in the digital age (Marshall Cavendish International Asia Pte Ltd).
Dolan, M. and Fullam, R. (2004), 'Theory of mind and mentalizing ability in antisocial personality disorders with and without psychopathy', Psychological Medicine, 34 (6), 1093- 102.
Domenicucci, Jacobo and Holton, Richard (2017), 'Trust as a Two-place Relation', in Paul Faulker and Tom Simpson (ed.), The Philosophy of Trust (Oxford University Press).
Dooley, Robert S. and Fryxell, Gerald E. (1999), 'Attaining Decision Quality and Commitment From Dissent: The Moderating Effects of Loyalty and Competence in Strategic Decision- Making Teams', Academy of Management Journal, 42 (4), 389-402.
Dretske, Fred (1989), 'The need to know', in Marjorie Clay and Keith Lehrer (eds.), Knowledge and Skepticism (Westview Press).
Duffy, John and Dorner, Rebecca (2011), 'The pathos of" mindblindness": Autism, science, and sadness in" theory of mind" narratives', Journal of Literary & Cultural Disability Studies, 5 (2), 201-15.
Bibliography 242
Duval, Céline, et al. (2011), 'Age effects on different components of theory of mind', Consciousness and Cognition, 20 (3), 627-42.
Eliot, George (1871-2), Middlemarch: A Study of Provincial Life (by Mary Ann Evans), 2 vols. (1-2: Harper & brothers).
Elster, Jon (2007), Explaining Social Behavior: More Nuts and Bolts for the Social Sciences (Cambridge: Cambridge University Press).
Epley, Nicholas and Caruso, Eugene M (2008), 'Perspective Taking: Misstepping Into Others’ Shoes', in Keith D. Markman, William M. P. Klein, and Julie A. Suhr (eds.), Handbook of imagination and mental simulation (New York: Psychnology Press), 295.
Erwin, Julie (2016), 'The Effects of Sleep Quality on Theory of Mind'.
Falcon, M. and Shoop, S. A. 'Stars 'CAN-do' about defeating autism', USA Today (updated 10 April 2002)
Farkas, Katalin (2017), 'Philosophy Bites', in David Edmonds (ed.), Katalin Farkas on Knowing a Person (London).
Faulkner, Paul (2007a), 'A Genealogy of Trust', Episteme, 4 (03), 305-21.
--- (2007b), 'On Telling and Trusting', Mind, 116 (464), 875-902.
--- (2011), Knowledge on Trust (Oxford University Press).
--- (2014a), 'A Virtue Theory of Testimony', Proceedings of the Aristotelian Society, 114 (2pt2), 189-211.
--- (2014b), 'The practical rationality of trust', Synthese, 191 (9), 1975-89.
--- (2015), 'The attitude of trust is basic', Analysis.
--- (2017), 'The Problem of Trust', in Paul Faulkner and Tom Simpson (ed.), The Philosophy of Trust (Oxford University Press).
Ferrari, Pier Francesco, et al. (2003), 'Mirror neurons responding to the observation of ingestive and communicative mouth actions in the monkey ventral premotor cortex', 17 (8), 1703-14.
Foley, Richard (2001), Intellectual trust in oneself and others (Cambridge University Press).
--- (2005), 'Universal intellectual trust', Episteme: A Journal of Social Epistemology, 2 (1), 5-11.
Fricker, Elizabeth (1995), 'Telling and Trusting: Reductionism and Anti-Reductionism in the Epistemology of Testimony', Mind, 104 (414), 393-411.
--- (2006), 'Second-Hand Knowledge', Philosophy and Phenomenological Research, 73 (3), 592-618.
Bibliography 243
Fricker, Miranda (1998), 'Rational authority and social power: Towards a truly social epistemology', Proceedings of the Aristotelian Society (JSTOR), 159-77.
--- (2007), Epistemic injustice: Power and the ethics of knowing (Oxford University Press).
Frost-Arnold, Karen Louise (2008), 'The epistemological importance of trust in science', (University of Pittsburgh).
Fukuyama, Francis (1995), Trust: The social virtues and the creation of prosperity (Free Press New York).
Gallagher, Shaun (2004), 'Understanding Interpersonal Problems in Autism: Interaction Theory as an Alternative to Theory of Mind', Philosophy, Psychiatry, & Psychology, 11 (3), 199- 217.
Gallese, Vittorio and Goldman, Alvin I. (1998), 'Mirror neurons and the simulation theory of mind-reading', Trends in cognitive sciences, 2 (12), 493-501.
Gambetta, Diego (1988), 'Can we trust trust?', Trust: Making and breaking cooperative relations (New York: Blackwell), 213-37.
Gauthier, David (1986), Morals by agreement (Oxford University Press on Demand).
Gernsbacher, Morton Ann and Frymiare, Jennifer L. (2005), 'Does the Autistic Brain Lack Core Modules?', The journal of developmental and learning disorders, 9, 3-16.
The Invention of Lying (Warner Bros., 2009), Gervais, Ricky and Robinson, Matthew (dir.).
Goering, Sara (2009), 'Postnatal Reproductive Autonomy: Promoting Relational Autonomy and Self Trust in New Parents', Bioethics, 23 (1), 9-19.
Goldman, Alvin I. (1967), 'A causal theory of knowing', The Journal of Philosophy, 64 (12), 357-72.
--- (1992), Liaisons: Philosophy meets the cognitive and social sciences (the MIT Press).
--- (2001), 'Experts: Which Ones Should You Trust?', Philosophy and Phenomenological Research, 63 (1), 85-110.
--- (2006), Simulating Minds: The Philosophy, Psychology, and Neuroscience of Mindreading (Oxford University Press USA).
--- (2011), 'A guide to social epistemology', Social epistemology: Essential readings, 11-37.
Govier, Trudy (1993), 'Self‐Trust, Autonomy, and Self‐Esteem', Hypatia, 8 (1), 99-120.
--- (1997), Social trust and human communities (McGill-Queen's Press-MQUP).
--- (1998), Dilemmas of trust (Montreal and Kingston: Mcgill-Queen's University Press).
Grau, Christopher (2004), 'Irreplaceability and unique value', Philosophical Topics, 32 (1/2), 111-29.
Bibliography 244
--- (2010), 'Love and history', The Southern Journal of Philosophy, 48 (3), 246-71.
Gregory, Carol, et al. (2002), 'Theory of mind in patients with frontal variant frontotemporal dementia and Alzheimer’s disease: theoretical and practical implications', Brain, 125 (4), 752-64.
Guan, Joseph and Li, Guohua (2017), 'Injury Mortality in Individuals With Autism', 107 (5), 791-93.
Hall, Judith A., Andrzejewski, Susan A., and Yopchick, Jennelle E. (2009), 'Psychosocial Correlates of Interpersonal Sensitivity: A Meta-Analysis', Journal of Nonverbal Behavior, 33 (3), 149-80.
Hansson, Sven Ove (2018), 'Risk', Stanford Encyclopedia of Philosophy
Happé, Francesca (1994), 'An advanced test of theory of mind: Understanding of story characters' thoughts and feelings by able autistic, mentally handicapped, and normal children and adults', Journal of autism Developmental disorders, 24 (2), 129-54.
Hardin, Russell (1996), 'Trustworthiness', Ethics, 107 (1), 26-42.
--- (2002), Trust and trustworthiness (4: Russell Sage Foundation).
--- (2004), Distrust: Manifestations and management (New York: Russell Sage Foundation).
--- (2006), Trust (Cambridge: Polity Press).
Harding, Matthew (2011), 'Responding to Trust', 24 (1), 75-87.
Hardwig, John (1991), 'The Role of Trust in Knowledge', The Journal of Philosophy, 88 (12), 693-708.
Haslam, Nick (2006), 'Dehumanization: An Integrative Review', 10 (3), 252-64.
Hawley, Katherine (2014), 'Trust, Distrust and Commitment', Noûs, 48 (1), 1-20.
--- (2017), 'Trustworthy Groups and Organizations', in Paul Faulkner and Tom Simpson (ed.), The Philosophy of Trust (Oxford University Press).
Heasman, Brett and Gillespie, Alex (2018), 'Perspective-taking is two-sided: Misunderstandings between people with Asperger’s syndrome and their family members', Autism Research, 22 (6), 740-50.
Heeramun, Ragini, et al. (2017), 'Autism and Convictions for Violent Crimes: Population-Based Cohort Study in Sweden', Journal of the American Academy of Child & Adolescent Psychiatry, 56 (6), 491-97.e2.
Helm, Bennett (2014), 'Trust as a reactive attitude', in David Shoemaker and Neal A. Tognazzini (eds.), Oxford Studies in Agency and Responsibility (2: Oxford University Press), 187-215.
Bibliography 245
--- (2018), 'Friendship', Stanford Encyclopedia of Philosophy
--- (2017), 'Love', Stanford Encyclopedia of Philosophy
Hertzberg, Lars (1988), 'On the attitude of trust', Inquiry, 31 (3), 307-22.
Hieronymi, Pamela (2008a), 'Responsibility for believing', Synthese, 161 (3), 357-73.
--- (2008b), 'The reasons of trust', Australasian Journal of Philosophy, 86 (2), 213-36.
Hinchman, Edward S. (2003), 'Trust and Diachronic Agency', Noûs, 37 (1), 25-51.
--- (2005), 'Telling as Inviting to Trust', 70 (3), 562-87.
--- (2017), 'An Assurance Theory of Trust', in Paul Faulkner and Tom Simpson (ed.), The Philosophy of Trust (Oxford University Press), 51.
Holton, Richard (1994), 'Deciding to trust, coming to believe', Australasian Journal of Philosophy, 72 (1), 63-76.
Horsburgh, H. J. N. (1960), 'The Ethics of Trust', The Philosophical Quarterly (1950-), 10 (41), 343-54.
Howarth, Caroline (2006), 'Race as stigma: positioning the stigmatized as agents, not objects', 16 (6), 442-51.
Hughes, Paul M. and Warmke, Brandon (2019), 'Forgiveness', Stanford Encyclopedia of Philosophy
Hull, Laura, et al. (2017), '“Putting on My Best Normal”: Social Camouflaging in Adults with Autism Spectrum Conditions', Journal of Autism and Developmental Disorders, 47 (8), 2519-34.
Hume, D. (1751), An enquiry concerning the principles of morals (London: A. Millar: Catherine Street in the Strand).
Hursthouse, R and Pettigrove, G (2017), 'Virtue Ethics', Stanford Encyclopedia of Philosophy
Hwang, Ye In, et al. (2019), 'Mortality and cause of death of Australians on the autism spectrum', Autism Research, 0 (0).
Iacoboni, Marco and Dapretto, Mirella (2006), 'The mirror neuron system and the consequences of its dysfunction', Nature Reviews Neuroscience, 7 (12), 942.
Ioannidis, J. P. A. (2005), 'Contradicted and Initially Stronger Effects in Highly Cited Clinical Research', ACC Current Journal Review, 14 (10), 6.
Bibliography 246
Jaswal, Vikram K. and Akhtar, Nameera (2018), 'Being vs. Appearing Socially Uninterested: Challenging Assumptions about Social Motivation in Autism', Behavioral and Brain Sciences, 1-84.
Johnson, Robert and Cureton, Adam (2018), 'Kant's Moral Philosophy', Stanford Encyclopedia of Philosophy
Jones, Karen (1996), 'Trust as an affective attitude', Ethics, 4-25.
--- (1999), 'Second-Hand Moral Knowledge', The Journal of Philosophy, 96 (2), 55-78.
--- (2004), 'Trust and Terror', in Peggy DesAutels and Margaret Urban Walker (eds.), Moral Psychology: Feminist Ethics and Social Theory (Lanham, Rowman and Littlefield), 3-18.
--- (2012), 'Trustworthiness', Ethics, 123 (1), 61-85.
--- (2017), '‘But I Was Counting On You!’', The Philosophy of Trust (Oxford: Oxford University Press).
Kanner, Leo (1943), 'Autistic disturbances of affective contact', Nervous child, 2 (3), 217-50.
Kapp, Steven K, et al. (2013), 'Deficit, difference, or both? Autism and neurodiversity', Developmental psychology, 49 (1), 59.
Kashdan, Todd B and Silvia, Paul J (2009), 'Curiosity and interest: The benefits of thriving on novelty and challenge', Oxford handbook of positive psychology, 2, 367-74.
Kassin, S., Fein, S, and Markus, H. R. (2014), 'Conflict escalation and reduction', Social Psychology (9th edn.: Wadsworth), 330.
Kawall, Jason (2009), 'In Defense of the Primacy of the Virtues', Journal of Ethics and Social Philosophy, 3 (2), 1-21.
Kennett, Jeanette (2002), 'Autism, empathy and moral agency', The Philosophical Quarterly, 52 (208), 340-57.
Kenny, Anthony (1963), Action, Emotion and Will (London; New York: Routledge and Kegan Paul; Humanities Press.).
Keren, Arnon (2007), 'Epistemic Authority, Testimony and the Transmission of Knowledge', Episteme, 4 (3), 368-81.
--- (2014), 'Trust and belief: a preemptive reasons account', Synthese, 191 (12), 2593-615.
Kitcher, Philip (1993), The Advancement of Science (New York: Oxford University Press).
Kleinig, John (2018), 'Loyalty', Stanford Encyclopedia of Philosophy
Knudson-Martin, Carmen (2013), 'Why Power Matters: Creating a Foundation of Mutual Support in Couple Relationships', Family Process, 52 (1), 5-18.
Bibliography 247
Koenig, Melissa A and Harris, Paul L (2007), 'The basis of epistemic trust: Reliable testimony or reliable sources?', Episteme, 4 (03), 264-84.
Kolodny, Niko (2003), 'Love as valuing a relationship', The Philosophical Review, 112 (2), 135- 89.
Korkmaz, Baris (2011), 'Theory of mind and neurodevelopmental disorders of childhood', Pediatric Research, 69 (5), 101R-8R.
Kosfeld, Michael, et al. (2005), Oxytocin Increases Trust in Humans (435) 673-6.
Kraut, Robert (1987), 'Love De Re', Midwest Studies In Philosophy, 10 (1), 413-30.
Lagerspetz, Olli (1998), Trust: The tacit demand (1: Springer Science & Business Media).
Lahno, Bernd (2001), 'On the Emotional Character of Trust', Ethical Theory and Moral Practice, 4 (2), 171-89.
--- (2017), 'Trust and Collective Agency', The Philosophy of Trust (Oxford: Oxford University Press).
Lai, Meng-Chuan, et al. (2017), 'Quantifying and exploring camouflaging in men and women with autism', Autism, 21 (6), 690-702.
Lane, Christel and Bachmann, Reinhard (1998), Trust within and between organizations: Conceptual issues and empirical applications (Oxford University Press).
Lane, Harlan (1989), When the mind hears: A history of the deaf (Vintage).
Lang, John T. and Hallman, William K. (2005), 'Who Does the Public Trust? The Case of Genetically Modified Food in the United States', Risk Analysis, 25 (5), 1241-52.
Lasalvia, Antonio, et al. (2013), 'Global pattern of experienced and anticipated discrimination reported by people with major depressive disorder: a cross-sectional survey', The Lancet, 381 (9860), 55-62.
Laudan, Larry (2006), Truth, Error, and Criminal Law: An Essay in Legal Epistemology (Cambridge Studies in Philosophy and Law; Cambridge: Cambridge University Press).
Lee, Lisa, et al. (2005), 'Mental state decoding abilities in clinical depression', Journal of Affective Disorders, 86 (2), 247-58.
Lehrer, Keith (1997), Self-trust: A study of reason, knowledge, and autonomy (Wiley Online Library).
Li, Annie S., et al. (2011), 'Exploring the Ability to Deceive in Children with Autism Spectrum Disorders', Journal of Autism and Developmental Disorders, 41 (2), 185-95.
List, Christian and Pettit, Philip (2011), Group agency: The possibility, design, and status of corporate agents (Oxford University Press).
Bibliography 248
Livingston, Lucy Anne and Happé, Francesca (2017), 'Conceptualising compensation in neurodevelopmental disorders: reflections from autism spectrum disorder', Neuroscience Biobehavioral Reviews, 80, 729-42.
Løgstrup, Knud E. (1956), Den etiske fordring [The ethical demand] (Copenhagen, Denmark: Gyldendalske Boghandel Nordisk Forlag).
--- (1997), The ethical demand (translated), eds Hans Fink and Alasdair MacIntyre (Notre Dame, Indiana: University of Notre Dame University Press).
--- (2007), Beyond the ethical demand (University of Notre Dame Press Notre Dame, IN).
Lombardo, Michael V., et al. (2016), 'Unsupervised data-driven stratification of mentalizing heterogeneity in autism', Scientific Reports, 6, 35333.
Long, Graham 'Graham Long and the 'quantum of compassion'', Conversations [Radio], (updated 7 Dec 2017)
Longworth, Guy (2017), 'Faith in Kant', The Philosophy of Trust (Oxford: Oxford University Press).
Luhmann, Niklas (1979), Trust and power (Toronto: Wiley).
Mansbridge, Jane (1999), 'Altruistic trust', in Mark E. Warren (ed.), Democracy and trust (Cambridge University Press), 290.
Markram, Henry, Rinaldi, Tania, and Markram, Kamila (2007), 'The intense world syndrome-an alternative hypothesis for autism', Frontiers in neuroscience, 1, 6.
Martinez-Murcia, F. J., et al. (2017), 'On the brain structure heterogeneity of autism: Parsing out acquisition site effects with significance-weighted principal component analysis', Hum Brain Mapp, 38 (3), 1208-23.
McEwen, Bruce S. (2008), 'Central effects of stress hormones in health and disease: Understanding the protective and damaging effects of stress and stress mediators', European Journal of Pharmacology, 583 (2), 174-85.
McGeer, Victoria (2004), 'The Art of Good Hope', The ANNALS of the American Academy of Political and Social Science, 592 (1), 100-27.
--- (2008), 'Trust, hope and empowerment', Australasian Journal of Philosophy, 86 (2), 237-54.
McGeer, Victoria and Pettit, Philip (2017), 'The Empowering Theory of Trust', The Philosophy of Trust (Oxford: Oxford University Press).
McLeod, Carolyn (2002), Self-trust and reproductive autonomy (5; Cambridge MA: MIT Press).
--- (2018), 'Trust', Stanford Encyclopedia of Philosophy
Bibliography 249
McMyler, Benjamin (2011), Testimony, Trust, and Authority (Oxford University Press).
--- (2017), 'Deciding to Trust', The Philosophy of Trust (Oxford: Oxford University Press).
Michael, Michaelis and Caldwell, Peter (2004), 'The Consolations of Optimism', in David Benatar (ed.), Life, Death and Meaning: Key Philosophical Readings on the Big Questions (Rowman & Littlefield).
Midgley, Mary, Blakemore, Colin, and Blackburn, Simon (2016), 'In Search of the Self', Philosophy for our times (The Institute of Art and Ideas).
Miller, C. A. (2001), 'False belief understanding in children with specific language impairment', J Commun Disord, 34 (1-2), 73-86.
Millstone, Erik and van Zwanenberg, Patrick (2000), 'A crisis of trust: for science, scientists or for institutions?', Nature Medicine, 6, 1307.
Milton, Damian (2012), 'On the ontological status of autism: the ‘double empathy problem’', Disability & Society, 27 (6), 883-87.
--- (2017), 'An Autistic Critical Reflection on the Dual Action Process Model of Intensive Interaction', Intensive Interaction Newsletter, (60), 2-2.
Moran, Richard (2005), 'Getting told and being believed', Philosophers' Imprint, 5, 1-29.
Morgan, Robert M. and Hunt, Shelby D. (1994), 'The Commitment-Trust Theory of Relationship Marketing', Journal of Marketing, 58 (3), 20-38.
Morton, Adam (2010), 'Epistemic Emotions', in Peter Goldie (ed.), The Oxford Handbook of Philosophy of Emotion (Oxford University Press), 385--99.
Mottron, Laurent, Dawson, Michelle, and Soulières, Isabelle (2009), 'Enhanced perception in savant syndrome: patterns, structure and creativity', Philosophical Transactions of the Royal Society B: Biological Sciences, 364 (1522), 1385-91.
Mullin, Amy (2005), 'Trust, Social Norms, and Motherhood', Journal of Social Philosophy, 36 (3), 316-30.
Murdoch, Iris (1957), 'Metaphysics and ethics', Existentialists and Mystics: Writings on Philosophy and Literature, 59-75.
--- (1997), 'The Sovereignty of Good over Other Concepts', in Roger Crisp and Michael Slote (eds.), Virtue ethics (Oxford University Press), 99-117.
--- (2001), 'On ‘God’and ‘Good’', in Marjorie Grene (ed.), The Anatomy of Knowledge, 233-35.
--- (2010), Henry and Cato (Random House).
--- (2013), The sovereignty of good (Routledge).
Nickel, Philip J. (2007), 'Trust and obligation-ascription', Ethical theory and moral practice, 10 (3), 309-19.
Bibliography 250
--- (2012), 'Trust and Testimony', Pacific Philosophical Quarterly, 93 (3), 301-16.
--- (2017), 'Being Pragmatic about Trust', The Philosophy of Trust (Oxford: Oxford University Press).
Nilsson, Kristine Kahr and de Lopez, Kristine Jensen (2016), 'Theory of Mind in children with Specific Language Impairment: A systematic review and meta‐analysis', Child development, 87 (1), 143-53.
Nyhan, Brendan, et al. (2014), 'Effective messages in vaccine promotion: a randomized trial', Pediatrics, 133 (4), e835-e42.
O’Neil, Collin (2012), 'Lying, Trust, and Gratitude', Philosophy & Public Affairs, 40 (4), 301- 33.
--- (2017), 'Betraying Trust', The Philosophy of Trust (Oxford: Oxford University Press).
O’neill, Onora (2002), Autonomy and trust in bioethics (Cambridge University Press).
Oberman, Lindsay M and Ramachandran, Vilayanur S (2007), 'The simulating social mind: the role of the mirror neuron system and simulation in the social and communicative deficits of autism spectrum disorders', Psychological bulletin, 133 (2), 310.
Olson, Eric T (2007), What are we?: A study in personal ontology (Oxford University Press).
Oshana, Marina (2014), 'Trust and Autonomous Agency', 91 (3), 431-47.
Oswald, Donald P. and Ollendick, Thomas H. (1989), 'Role taking and social competence in autism and mental retardation', Journal of Autism and Developmental Disorders, 19 (1), 119- 27.
Owens, David (2017), 'Trusting a Promise and Other Things', The Philosophy of Trust (Oxford: Oxford University Press).
Parfit, Derek (2012), 'We Are Not Human Beings', Philosophy, 87 (1), 5-28.
Parkman, Francis (1869), The Discovery of the Great West: An historical narrative (London: Boston: Little, Brown).
Peterson, Candida C., Wellman, Henry M., and Liu, David (2005), 'Steps in Theory-of-Mind Development for Children With Deafness or Autism', Child Development, 76 (2), 502-17.
Pettit, Philip (1995), 'The Cunning of Trust', Philosophy & Public Affairs, 24 (3), 202-25.
--- (2002), Rules, Reasons, and Norms (Oxford: Oxford University Press) 424.
Pilkington, Ed ''It's torture': critics step up bid to stop US school using electric shocks on children', The Guardian
Pinker, Steven (2002), The blank slate: The modern denial of human nature (Penguin).
Bibliography 251
Pismenny, Arina and Prinz, Jesse (2017), 'Is Love an Emotion?', The Oxford.
Potter, Nancy Nyquist (2002), How can I be trusted? A Virtue Theory of Trustworthiness (Lanham, Maryland: Rowman & Littlefield).
--- (2013), 'Narrative Selves, Relations of Trust, and Bipolar Disorder', Philosophy, Psychiatry, and Psychology, 20 (1), 57-65.
Price, A. W. (2011), Virtue and Reason in Plato and Aristotle (Oxford University Press).
Putnam, Robert D, Leonardi, Robert, and Nanetti, Raffaella Y (1993), Making democracy work: Civic traditions in modern Italy (Princeton university press).
Rasmussen, Heather N., Greenhouse, Joel B., and Scheier, Michael F. (2009), 'Optimism and Physical Health: A Meta-analytic Review', Annals of Behavioral Medicine, 37 (3), 239-56.
Realo, Anu, et al. (2003), 'Mind-reading ability: Beliefs and performance', Journal of Research in Personality, 37 (5), 420-45.
Remington, Anna and Fairnie, Jake (2017), 'A sound advantage: Increased auditory capacity in autism', Cognition, 166, 459-65.
Rizzolatti, Giacomo and Arbib, Michael A (1998), 'Language within our grasp', Trends in neurosciences, 21 (5), 188-94.
Rizzolatti, Giacomo and Craighero, Laila (2004), 'The mirror-neuron system', Annual Review Neuroscience, 27, 169-92.
Rizzolatti, Giacomo, Fogassi, Leonardo, and Gallese, Vittorio (2001), 'Neurophysiological mechanisms underlying the understanding and imitation of action', Nature reviews neuroscience, 2 (9), 661.
Rorty, Amelie Oksenberg (1980), Explaining Emotions, ed. Amélie Rorty (232; Berkeley: University of California Press).
--- (1987), 'The historicity of psychological attitudes: love is not love which alters not when it alteration finds', Midwest Studies in Philosophy, 10 (1), 399-412.
Rosati, Connie S. (2017), 'Moral Motivation', Stanford Encyclopedia of Philosophy
Rose, Richard and Mishler, William (2011), 'Political trust and distrust in post-authoritarian contexts', in Sonja Zmerli and Marc Hooghe (eds.), Political trust: Why context matters (ECPR Press), 117.
Roth, Philip (1998), American Pastoral (Vintage Books).
--- (2000), The Human Stain (Houghton Mifflin).
Rowe, A. D., et al. (2001), '"Theory of mind" impairments and their relationship to executive functioning following frontal lobe excisions', Brain, 124 (Pt 3), 600-16.
Bibliography 252
Rutherford, M. D. and Ray, Darrien (2009), 'Cheater detection is preserved in autism spectrum disorders', Journal of Social, Evolutionary, and Cultural Psychology, 3 (2), 105-17.
Saltzman, Jennifer, et al. (2000), 'Theory of mind and executive functions in normal human aging and Parkinson's disease', Journal of the International Neuropsychological Society, 6 (7), 781-88.
Sandberg, Annika Dahlgren and Hjelmquist, Erland (2003), 'The non-specificity of theory of mind deficits: Evidence from children with communicative disabilities AU - Dahlgren, Svenolof', European Journal of Cognitive Psychology, 15 (1), 129-55.
Sasson, Noah J and Morrison, Kerrianne E (2017), 'First impressions of adults with autism improve with diagnostic disclosure and increased autism knowledge of peers', Autism Research, 1362361317729526.
Sasson, Noah J, et al. (2017), 'Neurotypical peers are less willing to interact with those with autism based on thin slice judgments', Scientific reports, 7, 40700.
Scheman, Naomi (2001), 'Epistemology Resuscitated: Objectivity as Trustworthiness', in Nancy Tuana and Sandra Morgan (eds.), Engendering Rationalities (Albany, NY: State University of New York Press), 23-52.
Schwebel, Milton (2003), Behavioral science and human survival (iUniverse).
Sellars, Wilfrid S. (1963), 'Philosophy and the scientific image of man', in Robert Colodny (ed.), Science, Perception, and Reality (Ridgeview: Humanities Press), 20-63.
Shahaeian, Ameneh, et al. (2011), 'Culture and the sequence of steps in theory of mind development', Developmental psychology, 47 (5), 1239.
Sheppard, Elizabeth, et al. (2016), 'How easy is it to read the minds of people with autism spectrum disorder?', Journal of autism developmental disorders, 46 (4), 1247-54.
Shoemaker, Sydney (2011), 'On What We Are', in Shaun Gallagher (ed.), The Oxford Handbook of the Self (Oxford University Press).
--- (2012), 'Against simplicity', in Georg Gasser and Matthias Stefan (eds.), Personal identity: Complex or simple (Cambridge University Press), 123-36.
Silvers, Anita and Francis, Leslie Pickering (2005), 'Justice through Trust: Disability and the Outlier Problem in Social Contract Theory', Ethics, 116 (1), 40-76.
Simpson, Eleanor H. and Balsam, Peter D. (2016), 'The Behavioral Neuroscience of Motivation: An Overview of Concepts, Measures, and Translational Applications', in Eleanor H. Simpson and Peter D. Balsam (eds.), Behavioral Neuroscience of Motivation (Cham: Springer International Publishing), 1-12.
Simpson, Thomas W. (2012), 'What Is Trust?', Pacific Philosophical Quarterly, 93 (4), 550-69.
--- (2017), 'Trust and Evidence', The Philosophy of Trust (Oxford: Oxford University Press).
Sinnott-Armstrong, Walter (1988), Moral Dilemmas (B. Blackwell).
Bibliography 253
Skinner, B. F. (1974), About Behaviourism (London: Jonathan Cape).
Skyrms, Brian (2008), 'Trust, Risk, and the Social Contract', Synthese, 160 (1), 21-25.
Slote, M. (2001), Morals from Motives (Oxford: Oxford University Press).
Smith, Angela M. (2005), 'Responsibility for Attitudes: Activity and Passivity in Mental Life', 115 (2), 236-71.
Smith, Matthew Noah (2008), 'Terrorism, shared rules and trust', Journal of Political Philosophy, 16 (2), 201–19.
Smith, Michael J. and desJardins, Marie (2005), 'A Model for Competence and Integrity in Variable Payoff Games'.
--- (2008), 'Learning to trust in the competence and commitment of agents', Autonomous Agents and Multi-Agent Systems, 18 (1), 36.
Smukler, David (2005), 'Unauthorized Minds: How “Theory of Mind” Theory Misrepresents Autism', Mental Retardation, 43 (1), 11-24.
Smuts, Aaron (2013), 'In Defense of the No-Reasons View of Love'.
--- (2014a), 'Normative reasons for love, part II', Philosophy Compass, 9 (8), 518-26.
--- (2014b), 'Normative Reasons for Love, Part I', Philosophy Compass, 9 (8), 507-17.
Snowdon, Paul and Gomes, Anil (2019), 'Peter Frederick Strawson', Stanford Encyclopedia of Philosophy
Spaulding, Shannon (2016), 'Mind Misreading', Philosophical Issues, 26 (1), 422-40.
Sprong, Mirjam, et al. (2007), 'Theory of mind in schizophrenia: meta-analysis', The British Journal of Psychiatry, 191 (1), 5-13.
Sreenivasan, Gopal (2002), 'Errors about Errors: Virtue Theory and Trait Attribution', Mind, 111 (441), 47-68.
Stern, Robert (2017), '‘Trust is Basic’: Løgstrup on the Priority of Trust', The Philosophy of Trust (Oxford: Oxford University Press).
Stocker, Michael (2013), 'Intellectual and other nonstandard emotions', On Emotions: Philosophical Essays, 9, 99.
Strawson, Peter Frederick (1962), 'Freedom and Resentment', Proceedings of the British Academy, 48, 1-25.
Styron, William (1979), Sophie's Choice (Random House).
Swettenham, John, et al. (2014), 'Seeing the unseen: Autism involves reduced susceptibility to inattentional blindness', Neuropsychology, 28 (4), 563-70.
Bibliography 254
Szalavitz, Maia 'Asperger's theory does about-face', The Star
Taddeo, Mariarosaria and Floridi, Luciano (2011), 'The case for e-trust', Ethics and Information Technology, 13 (1), 1-3.
Talbott, William J. and Goldman, Alvin I. (1998), 'Games Lawyers Play: Legal Discovery and Social Epistemology', Legal Theory, 4 (2), 93-163.
Tomasello, Michael, et al. (2005), 'In search of the uniquely human', Behavioral brain sciences, 28 (5), 721-27.
Townley, Cynthia and Garfield, Jay L (2013), 'Public trust', Trust: Analytic and Applied Perspectives, 95-108.
Turri, John, Alfano, Mark, and Greco, John (2008), 'Virtue Epistemology', Stanford Encyclopedia of Philosophy
Uekermann, Jennifer and Daum, Irene (2008), 'Social cognition in alcoholism: a link to prefrontal cortex dysfunction?', Addiction, 103 (5), 726-35.
Umilta, Maria Alessandra, et al. (2001), 'I know what you are doing: A neurophysiological study', Neuron, 31 (1), 155-65.
Uslaner, Eric M (1999), 'Democracy and social capital', in Mark E. Warren (ed.), Democracy & trust (Oxford University Press), 121-50.
--- (2000), 'Producing and consuming trust', Political Science Quarterly, 115 (4), 569-90.
--- (2018), 'The study of trust', in Eric M Uslaner (ed.), The Oxford handbook of social and political trust (Oxford University Press), 3-14.
Uslaner, Eric M. (2003), 'Varieties of trust', European Political Science, 2 (3), 43-49.
Van Fraassen, Bas C (1980), The scientific image (Oxford University Press).
Walker, Margaret Urban (2006), Moral repair: Reconstructing moral relations after wrongdoing (Cambridge University Press).
Wanderer, Jeremy and Townsend, Leo (2013), 'Is it Rational to Trust?', 8 (1), 1-14.
Warren, Mark E (1999), Democracy and trust (Cambridge University Press).
Watson, Gary (1990), 'On the primacy of character', Identity, character, and morality: Essays in moral psychology, 449-83.
Webb, Mark Owen (1992), 'The epistemology of trust and the politics of suspicion'.
Bibliography 255
--- (1993), 'Why I know about as much as you: A reply to Hardwig', The Journal of Philosophy, 90 (5), 260-70.
Weckert, John (2005), 'Trust in cyberspace', in Robert J. Cavalier (ed.), The impact of the internet on our moral lives (Albany: State University of New York Press), 95-117.
Westbrook, David, Kennerley, Helen, and Kirk, Joan (2011), An introduction to cognitive behaviour therapy: Skills and applications (Sage).
Whyte, Kyle Powys and Crease, Robert P. (2010), 'Trust, expertise, and the philosophy of science', Synthese, 177 (3), 411-25.
Wilholt, Torsten (2013), 'Epistemic Trust in Science', The British Journal for the Philosophy of Science, 64 (2), 233-53.
Williams, Bernard (1995), 'Formal Structures and Social Reality', in Bernard Williams (ed.), Making sense of humanity: and other philosophical papers 1982-1993 (Cambridge University Press), 111-22.
Wing, Lorna and Gould, Judith (1979), 'Severe impairments of social interaction and associated abnormalities in children: Epidemiology and classification', Journal of autism developmental disorders, 9 (1), 11-29.
Woodbury-Smith, M. R, et al. (2006), 'High functioning autistic spectrum disorders, offending and other law-breaking: findings from a community sample', The Journal of Forensic Psychiatry & Psychology, 17 (1), 108-20.
Woolfe, Tyron, Want, Stephen C, and Siegal, Michael (2002), 'Signposts to development: Theory of mind in deaf children', Child development, 73 (3), 768-78.
Wright, Stephen (2010), 'Trust and Trustworthiness', Philosophia, 38 (3), 615-27.
Yergeau, Melanie (2013), '“Clinically Significant Disturbance: Theorists Who Theorize Theory of Mind,”', The Disability Studies Quarterly, 33 (4).
--- (2017), Authoring autism: On rhetoric and neurological queerness (Duke University Press).
Zagzebski, Linda Trinkaus (1994), 'The Inescapability of Gettier Problems', The Philosophical Quarterly, 44 (174), 65-73.
--- (2004), Divine motivation theory (Cambridge University Press).
--- (2012), Epistemic Authority: A Theory of Trust, Authority, and Autonomy in Belief (Oup Usa).
Zelazo, P. D., et al. (1996), 'Theory of mind and rule use in individuals with Down's syndrome: a test of the uniqueness and specificity claims', J Child Psychol Psychiatry, 37 (4), 479-84.
Zunhammer, Matthias, et al. (2015), 'Theory of mind and emotional awareness in chronic somatoform pain patients', PloS one, 10 (10), e0140016.