[The version of this paper published by Oxford online in 2019 was not copy-edited and has some sense-obscuring typos. I have posted a corrected (but not the final published) version on this site. The version published in print in 2020 has these corrections.] Which is more fundamental, assertion or testimony? Should we understand assertion as basic, treating testimony as what you get when you add an interpersonal addressee? Or should we understand testimony as basic, treating mere assertion -- assertion without (...) testimony -- as what you get when you subtract that interpersonal relation? In this chapter, I’ll argue for the subtractive approach and for the more general thesis that its treatment of the interpersonal element in assertion makes understanding that interpersonal element the key to understanding how assertion expresses belief. My theory of belief-expression in assertion treats it as internalizing the transmission of belief in testimony. How we understand that internalizing move depends on how we conceptualize the interpersonal element in testimony. Since what I’ll call the Command Model does not give us the conceptual resources to make this move, we should adopt an alternative that I’ll call the Custodial Model, on which a testifier aims not to convince her addressee but to reason with him – to give him reasons to believe what she tells him grounded in her trustworthiness in thus attempting to influence him. The subtractive approach to assertion thus rests on a key distinction between the aims of reasoning and persuasion. (shrink)

An assurance theory of trust begins from the act of assurance – whether testimonial, advisorial or promissory – and explains trust as a cognate stance of resting assured. My version emphasizes the risks and rewards of trust. On trust’s rewards, I show how an assurance can give a reason to the addressee through a twofold exercise of ‘normative powers’: (i) the speaker thereby incurs an obligation to be sincere; (ii) if the speaker is trustworthy, she thereby gives her addressee the (...) reason. Each claim is controversial; my contribution lies in how I defend them together, linking the sincerity obligation to the provision of reasons. On trust’s risks, I defend a popular thesis about the nature of trust – that trust differs from mere reliance in how it risks betrayal – against the objection that the thesis ‘moralizes’ trust. Such betrayal need not be moral, I argue, because it may derive entirely from how the assurance purports to provide a reason. (shrink)

Can a reason to believe testimony derive from the addressee’s trust itself or only from reliability in the speaker that the trust perhaps causes? I aim to cast suspicion on the former view, defended by Faulkner, in favor of the latter – despite agreeing with Faulkner’s emphasis on the second-personal normativity of testimonial assurance. Beyond my narrow disagreement with Faulkner lie two broader issues. I argue that Faulkner misappropriates Bernard Williams’s genealogy of testimony when he makes use of Williams’s genealogical (...) argument in his own preferred assurance view of testimony. Though Williams doesn’t clearly articulate it, there is a deep reason why Williams’s genealogy cannot underwrite an argument for trust-based testimonial reasons. Can a genealogical argument underwrite any version of the assurance view? I sketch an assurance view of testimonial reasons that rejects Faulkner’s thesis that such reasons could be grounded in trust. Then I examine what it would take for that assurance view to receive genealogical vindication. (shrink)

Is autonomy – governing yourself – compatible with letting yourself be governed by trust? This paper argues that autonomy is not only compatible with appropriate trust but actually requires it. Autonomy requires appropriate trust because it is undermined by inappropriate mistrust. An autonomous agent treats herself as answerable for her action-guiding commitments, where answerability requires openness to the rational influence of external, critical perspectives on those commitments. This openness to correction makes one vulnerable to manipulation and can be exploited in (...) ways that undermine autonomy instead of supporting it. We argue that autonomy is undermined by inappropriate trust, and that the inappropriateness lies in the agent’s failure to manifest appropriate responsiveness to evidence of untrustworthiness in the one trusted. The problem isn’t that autonomy requires dispensing with trust but that trust must itself be governed by appropriate responsiveness to evidence of untrustworthiness. Our view confronts two challenges. One challenge lies in explaining why you shouldn’t treat your vulnerability to manipulation as a general reason to let your trusting disposition lapse into an untrusting one. We argue that there is no such general reason to let your responsiveness to evidence of untrustworthiness be supplanted by a suspicious disposition to seek and require positive evidence of trustworthiness. A second, and deeper, challenge arises from the fact that in some especially sophisticated cases of manipulation there is no evidence of untrustworthiness to which you could respond. We argue that you cannot rationally cope with even this unresolvable risk of exploited trust by trusting only yourself. (shrink)

This paper treats two questions about the relation between trust and the will. One question, about trust, is whether you can trust ‘at will.’ Can you trust despite acknowledging that you lack evidence of the trustee’s worthiness of your trust? Another question, about the will, is whether you can exercise your will at all without trust – at least, in yourself. I treat the second question as a guide to the first, arguing that the role of trust in the will (...) reveals how you can trust at will. The key lies in distinguishing two ways of being responsive to evidence. On the one hand, you cannot trust someone – whether another person or your own earlier self – whom you judge unworthy of your trust. On the other hand, trust does not require a positive assessment of trustworthiness. I thus take issue with Pamela Hieronymi’s analysis of trust as a “commitment-constituted attitude.” Your trust is indeed constrained by your responsiveness to evidence of untrustworthiness, but you need undertake no commitment akin to or involving a judgment that the trustee is trustworthy. We can see why trust is not a commitment-constituted attitude by seeing how trust itself plays a role in forming a commitment. (shrink)

How far, if at all, do our intrapersonal and our interpersonal epistemic obligations run in parallel? This paper treats the question as addressing the stability of doxastic commitment in the two dimensions. In the background lies an analogy between doxastic and practical commitment. We’ll pursue the question of doxastic stability by coining a doxastic analogue of Gregory Kavka’s much-discussed toxin case. In this new case, you foresee that you will rationally abandon a doxastic commitment by undergoing a shift in the (...) context in which you doxastically deliberate. The case reveals an important stability condition on doxastic commitment that plays no role in cases modeled on Ulysses and the Sirens, the scenario that has shaped previous attempts to draw an analogy between doxastic and practical commitment. The stability condition thereby revealed is not diachronic -- it does not involve a shift in belief -- but we can construct specifically diachronic cases that reveal an important intrapersonal epistemic obligation. This obligation arises when you expect the shift in belief by ‘projecting’ yourself into a future doxastic context -- thus vindicating a restricted, but only this restricted, application of Bas van Fraassen’s Reflection Principle. Despite some important differences, we can use these diachronic cases as a model for interpersonal cases that reveal an important role for doxastic context in the epistemology of disagreement. But the roles played by projection and context in disagreement ensure that there is no general obligation to be responsive to the opinion of an epistemic peer who disagrees with you. (shrink)

Practical commitment is Janus-faced, looking outward toward the expectations it creates and inward toward their basis in the agent’s will. This paper criticizes Kantian attempts to link these facets and proposes an alternative. Contra David Velleman, the availability of a conspiratorial perspective (not yours, not your interlocutor’s) is what allows you to understand yourself as making a lying promise – as committing yourself ‘outwardly’ with the deceptive reasoning that Velleman argues cannot provide a basis for self-understanding. Moreover, the intrapersonal availability (...) of such a third perspective is what enables you to commit yourself ‘inwardly.’ Here I offer an alternative to Christine Korsgaard’s account of practical commitment, on which committing yourself requires identifying yourself with a principle. You needn’t identify yourself with a principle, I argue, because the unity at which you aim when you commit yourself is a unity not with your acting self but with a later perspective, where the relation is one of self-intelligibility, not self-justification, and therefore needn’t be mediated by principles. This ‘twice-future’ perspective – neither your present intending nor your (once-)future acting but a third perspective that looks back on that relation – plays the intrapersonal role played in interpersonal commitment by potential co-conspirators. Kantians are therefore right to link your ability to commit yourself with your ability credibly to express that commitment to others. But the linkage generates a strikingly unKantian result. The nature of agency cannot provide an apriori basis for honesty because what enables you to commit yourself is what also enables you to lie. (shrink)

How must you think about time when you form an intention? Obviously, you must think about the time of action. Must you frame the action in any broader prospect or retrospect? In this essay I argue that you must: you thereby commit yourself to a specific prospect of a future retrospect – a retrospect, indeed, on that very prospect. In forming an intention you project a future from which you will not ask regretfully, referring back to your follow-through on that (...) intention, “What on earth was I thinking?” I argue that this broader attitude expresses the self-accountability necessary for practical commitment. (shrink)

This paper engages two debates about trust, deriving from two distinct questions about the nature of trust. The first asks how to define trust. Does trusting B to φ involve anything more than relying on B to φ? The second asks about the normative structure of trust. Does trust most fundamentally embody a two-place or a three-place relation? I’ll defend a new position in the second debate that yields an equally new position in the first. The standard three-place model highlights (...) ‘A trusts B to φ.’ The two-place model prefers ‘A trusts B.’ On my new three-place model, both are less fundamental than ‘A φs through trust in B.’ We do, of course, trust people to do things. But the most fundamental form of trust lies in doing something through trust in a person. I call this the Assurance View of trust because I’ll argue that trusting involves accepting another’s invitation to trust – an assurance, in effect, that the other is relevantly trustworthy. Trust thereby differs from mere reliance by how it makes available a reason to rely on the trusted. In mere reliance, your reason to rely on B is exogenous to the reliance relation, emerging, for example, from B’s track record of relevant reliability. When you trust B, by contrast, you take yourself to have a reason to rely on B that is endogenous to your trust, a reason grounded not merely in B’s reliability, exogenously considered, but in B’s ongoing responsiveness to your relevant needs. The Assurance View’s key insight emerges, I’ll argue, from cases wherein trust is disappointed – the trusted does not do what you trusted her to do – yet not thereby betrayed. (shrink)

It’s a popular idea that memory resembles testimony insofar as each can ‘preserve’ epistemic warrant. But how does such ‘preservation’ do its epistemic work? I have elsewhere developed an assurance theory of testimonial warrant. Here, I develop an assurance theory of preservative memory. How could the ‘preservation’ of warrant through memory work through an assurance? What would even count as an intrapersonal assurance? I explain each form of preservation by contrasting the relation that preserves warrant with a pathological alternative. My (...) assurance theory of testimony theorizes the illocutionary norms governing the speech act of telling by explaining what goes wrong when there’s what Austin calls an ‘abuse.’ But what norms govern an intrapersonal memory relation? There are two sides in a memory relation: the earlier self judges that p, then memory keeps the later self in epistemic touch with that judgment by ‘preserving’ its warrant. But this is not an illocutionary relation. (When you ‘tell yourself that p,’ you are trying to ‘convince yourself’ that p and as such typically do not yet even judge that p.) I argue that the earlier self nonetheless offers an assurance by extending an invitation to trust. Your earlier self judges that p in part by projecting a future in which you will continue to trust yourself on the question whether p when you lose access to the evidence that informed the judgment. That projection of trust in yourself posits a worthiness of trust that ought to lead others to accept your testimony that p. Assurance in memory and assurance in testimony thus ought to work together. This aspect of the parallel emerges vividly when we see how it breaks down in the pathology of gaslighting – more than merely illocutionary abuse! – which targets specifically how testimony and memory work together in inviting trust relations. (shrink)

[This draft is dated November 2007. I wrote it while I was a fellow at the Center for 21st Century Studies at UW-Milwaukee, in 2005-06, and published it only on the Center's website as a working paper. Many of the core ideas in this paper wound up in "Receptivity and the Will," Nous 2009, "Assertion, Sincerity, and Knowledge," Nous 2013, and "Assurance and Warrant," Philosophers' Imprint 2014 -- though formulated rather differently. What follows is the original abstract.] This working paper (...) extends a project that I’ve been developing for several years on the role of trust, including self-trust, in rationality and personal autonomy. I’m interested equally in epistemic questions (what to believe) and in practical questions (how to act), though the present paper addresses only the former. In an earlier paper on the epistemic side of my project, “Telling as Inviting to Trust” (Philosophy and Phenomenological Research, May 2005), I offered an explanation of how you can get an entitlement to believe something simply by trusting someone when she tells you it’s so, the entitlement deriving not merely from the assurance but from the trustworthy character of the one who gives it. The present paper adapts my explanation of believing-on-another’s-word to belief as such, via the observation that you can’t form a belief without trusting someone’s judgment – if only your own. I view intrapersonal reasoning about what to believe as an internalization of the interpersonal cases I analyzed in that earlier paper. In each instance, belief-formation has a second-personal dynamic: an invitation to trust on one side, trusting acceptance of that invitation on the other. If I’m right, the dynamic at the heart of all belief manifests not autonomy but a trusting heteronomy, whereby you render yourself appropriately receptive to judgment – your own or another’s. (The idea that judging is inviting self-trust first occurred to me in 2002, but I didn’t get a chance to develop it properly till 2005-06, my fellowship year at the Center for 21st Century Studies. I appreciate the support.). (shrink)