Eva Schmidt

Explainable artificial intelligence (XAI) aims to overcome the opacity of black box systems, i.e., to make them understandable to suitable stakeholders. In this chapter, I investigate how understanding depends on how much is at stake in a context. I support the intuition that understanding is sensitive to the stakes with a pair of cases. I further use this pair of cases to spell out how exactly the stakes affect understanding, particularly, outright understanding why. To do so, I connect discussions of the concept of understanding with debates on pragmatic encroachment and on inductive risk. I zoom in on one necessary condition on understanding specifically, viz. the epistemic justification condition, according to which the beliefs involved in understanding have to be supported by sufficient evidence and have to cohere sufficiently well. I argue that, where the stakes are high, there are more stringent standards for sufficient evidence and coherence, and that this explains the stakes-sensitivity of understanding.

In this paper, I explicate pragmatic encroachment by appealing to pragmatic considerations attenuating, or weakening, epistemic reasons to believe. I call this the ‘Attenuators View’. I will show that this proposal is better than spelling out pragmatic encroachment in terms of reasons against believing – what I call the ‘Reasons View’. While both views do equally well when it comes to providing a plausible mechanism of how pragmatic encroachment works, the Attenuators View does a better job distinguishing practical and epistemic reasons to believe. First, this view does not appeal to the costs of believing falsely as reasons against believing; second, because of this, it does not run the risk of tearing down the wall between practical and epistemic reasons bearing on belief. I underpin the Attenuators View with a virtue‐theoretic account of how pragmatic encroachment attenuates epistemic reasons and close my discussion by considering some objections against such a view.

This book contains twelve original contributions from leading epistemologists, which connect the debate over epistemic dilemmas with fundamental issues in contemporary epistemology. In an epistemic dilemma, any doxastic response of an agent will violate an epistemic requirement. Recently, there has been a surge of interest in epistemic dilemmas. However, few attempts have been made to systematically connect epistemic dilemmas to a wider range of issues in epistemology and normative theorizing. The present volume begins to fill this lacuna. Some of its contributions explore new responses to epistemic dilemmas and others present novel epistemic dilemmas, which connect in interesting ways to other areas of epistemology. Several contributions discuss links between epistemic dilemmas and the nature of epistemic and practical reasons, investigate the nature of evidence or of doxastic attitudes, or explore dilemmas arising in collective or oppressive contexts.

One influential response to apparent higher-order dilemmas implies that agents can rationally both believe p on the basis of their evidence and simultaneously believe that their evidence does not support believing p. This possibility of rational epistemic akrasia seems to call into question the Enkratic Requirement, which prohibits believing a proposition p according to one’s lower-level evidence, while believing that one’s lower-level evidence does not support believing p. In this chapter, we explore two ways to defend the Enkratic Requirement. First, evidentialists about epistemic justification are committed to holding that an ex post (or doxastically) justified belief, as part and parcel of being ex post justified, is accompanied by a corresponding higher-order belief that the first-order belief is supported by the evidence. In a dilemma case, this higher-order belief comes into direct rational conflict with the subject’s other higher-order belief, that her evidence does not support the belief – so the subject has contradictory beliefs. In this way, the so-called ‘level-splitting’ strategy for the claim that being akratic is not per se irrational is called into question. Our second argument for the Enkratic Requirement presupposes normativism about belief and appeals to the conditions of having epistemically permissible beliefs. We sketch an argument starting from an influential view about the normativity of belief: that irrational beliefs are impermissible beliefs. From this claim, together with some widely held as-sumptions about the notion of rationality, we show that agents who display akratic combi-nations of attitudes hold impermissible beliefs.

Artificial intelligent (AI) systems used in medicine are often very reliable and accurate, but at the price of their being increasingly opaque. This raises the question whether a system’s opacity undermines the ability of medical doctors to acquire knowledge on the basis of its outputs. We investigate this question by focusing on a case in which a patient’s risk of recurring breast cancer is predicted by an opaque AI system. We argue that, given the system’s opacity, as well as the possibility of malfunctioning AI systems, practitioners’ inability to check the correctness of their outputs, and the high stakes of such cases, the knowledge of medical practitioners is indeed undermined. They are lucky to form true beliefs based on the AI systems’ outputs, and knowledge is incompatible with luck. We supplement this claim with a specific version of the safety condition on knowledge, Safety*. We argue that, relative to the perspective of the medical doctor in our example case, his relevant beliefs could easily be false, and this despite his evidence that the AI system functions reliably. Assuming that Safety* is necessary for knowledge, the practitioner therefore doesn’t know. We address three objections to our proposal before turning to practical suggestions for improving the epistemic situation of medical doctors.

The comment starts with a brief exposition of the Eroteric View put forth by Artūrs Logins. I then provide one friendly comment on the exact form of the normative question which is central to the view, and suggest that in addition to the question, ‘Why ought S to φ?’, Logins should take the question, ‘Why is S permitted to φ?’ as definitive of normative reasons. In a more critical comment, I reflect on how normative explanatory reasons and normative reasoning reasons relate, calling into question Logins’s claim that the two kinds of reasons are fundamentally distinct. First, I argue that being a normative explanatory reason near enough guarantees being a normative reasoning reason; second, I argue that normative explanatory reasons as well as normative reasoning reasons can be weighed, and that they can be weighed against each other. But this causes trouble for the Eroteric View.

Current debates about reasons and reasoning often draw comparisons between epistemic and practical reasons and reasoning and presuppose substantial unity between the practical and epistemic domains. This stance seems to conflict with a stark Humean contrast between the two domains: With respect to practical reasons and reasoning, Hume highlights the role of impressions, especially the passions, in motivating and rationalizing action, while apparently downplaying the potential relevance of beliefs, reason, or reasons. With respect to epistemic reasons and theoretical reasoning, he urges us to proportion our belief to the evidence (EHU 10.4), which suggests a significant epistemic role for evidential reasons. My contribution ar-gues that there is nonetheless a convincing and interesting unified Humean account of reasons and reasoning across both domains. To present my case, I first narrow down the unity claim that I am interested in. I then closely engage with Hume’s claims in epistemology, the philos-ophy of action, and metaethics to develop the unified Humean account.

In these replies, I react to comments on my paper “Facts about Incoherence as Non-Evidential Epistemic Reasons”, provided by Aleks Knoks, Sebastian Schmidt, Keshav Singh, and Conor McHugh. I discuss potential counterexamples to my claim that the fact that the subject’s doxastic attitudes are incoherent is an epistemic reason for her to suspend; whether such incoherence-based reasons bear on individual attitudes or only on combinations of attitudes; the prospects of restricting evidentialism about epistemic reasons to reasons to believe; whether incoherence-based reasons are truly epistemic; the alleged normative and motivational expendability of incoherence-based reasons; the possibility of incoherence-based reasons to suspend without actual belief in the incoherent propositions; the relationship between suspension, inquiry, and incoherence; and the nature of suspension of judgment.

Can the evidence provided by software systems meet the standard of proof for civil or criminal cases, and is it individualized evidence? Or, to the contrary, do software systems exclusively provide bare statistical evidence? In this paper, we argue that there are cases in which evidence in the form of probabilities computed by software systems is not bare statistical evidence, and is thus able to meet the standard of proof. First, based on the case of State v. Loomis, we investigate recidivism predictions provided by software systems used in the courtroom. Here, we raise problems for software systems that provide predictions that are based on bare statistical evidence. Second, by examining the case of People v. Chubbs, we argue that the statistical evidence provided by software systems in cold hit DNA cases may in some cases suffice for individualized evidence, on a view on which individualized evidence is evidence that normically supports the relevant proposition (Smith, in Mind 127:1193–1218, 2018).

This comment raises two worries for Crane’s view of religious beliefs and their contents. First, I argue that his appeal to inferentialism about the contents of dispositional beliefs cannot fully avoid the problem of inconsistent beliefs. For the same problem can be raised for occurrent thought, and the inferentialist solution is not available there. Second, I argue that religious beliefs differ from ordinary beliefs with respect to their justification in cases of peer disagreements. This suggests that noncognitivism about religious beliefs, which Crane opposes, is the correct view after all.

In this paper, I explicate pragmatic encroachment by appealing to pragmatic considerations attenuating, or weakening, epistemic reasons to believe. I call this the ‘Attenuators View’. I will show that this proposal is better than spelling out pragmatic encroachment in terms of reasons against believing – what I call the ‘Reasons View’. While both views do equally well when it comes to providing a plausible mechanism of how pragmatic encroachment works, the Attenuators View does a better job distinguishing practical and epistemic reasons to believe. First, this view does not appeal to the costs of believing falsely as reasons against believing; second, because of this, it does not run the risk of tearing down the wall between practical and epistemic reasons bearing on belief. I underpin the Attenuators View with a virtue-theoretic account of how pragmatic encroachment attenuates epistemic reasons and close my discussion by considering some objections against such a view.

This paper presents a counterexample to the principle that all epistemic reasons for doxastic attitudes towards p are provided by evidence concerning p. I begin by motivating and clarifying the principle and the associated picture of epistemic reasons, including the notion of evidence concerning a proposition, which comprises both first- and second-order evidence. I then introduce the counterexample from incoherent doxastic attitudes by presenting three example cases. In each case, the fact that the subject’s doxastic attitudes are incoherent is an epistemic reason to suspend, which is not provided by evidence. I argue that this incoherence fact is a reason for the subject to take a step back and reassess her evidence for her conflicting attitudes, and thus a reason to suspend all of them. Suspending judgment enables the subject to revise attitudes where appropriate and thus (typically) to arrive at a set of coherent and well-supported attitudes. I then address a dilemma for my proposal and, in conclusion, briefly suggest a picture of epistemic reasons on which they are to be understood against the background of the subject’s virtuous intellectual conduct.

This volume celebrates the work of Hans-Johann Glock, a philosopher renowned for both his exegesis of Wittgenstein and his many contributions to debates in contemporary philosophy. It brings together 16 new essays by up-and-coming and distinguished philosophers engaging with Glock’s work, and it concludes with a "Reflections and Replies" chapter in which Glock responds to his interlocutors. Glock’s distinctive philosophical voice features a rare combination of a Wittgenstein-inspired approach with a willingness to break away from Wittgenstein to tackle problems in an open-minded way. The broad selection of essays included in this volume reflects Glock’s wide-ranging philosophical interests and demonstrates the potential of applying Wittgensteinian insights to advance current systematic debates in philosophy. The chapters discuss Wittgenstein’s philosophy, metaphilosophy, truth and language, animal minds and agency, and reasons and normativity.

This article investigates whether religious experience can be conceived in such a way that the perceiver's religious expertise (via cognitive penetration or perceptual learning) contributes to the justificatory power of the experience. It also considers what kind of content religious experience would have to have to be able to justify standard types of religious beliefs. It argues that, against first impressions, religious expertise cannot supplement perceptual justification. At the same time, to the extent that religious experience has singular contents or Gestalt contents, it may well be able to justify religious belief in virtue of such contents.

We argue that explainable artificial intelligence (XAI), specifically reason-giving XAI, often constitutes the most suitable way of ensuring that someone can properly be held responsible for decisions that are based on the outputs of artificial intelligent (AI) systems. We first show that, to close moral responsibility gaps (Matthias 2004), often a human in the loop is needed who is directly responsible for particular AI-supported decisions. Second, we appeal to the epistemic condition on moral responsibility to argue that, in order to be responsible for her decision, the human in the loop has to have an explanation available of the system’s recommendation. Reason explanations are especially well-suited to this end and we examine whether – and how – it might be possible to make such explanations fit with AI systems. We support our claims by focusing on a case of disagreement between human in the loop and AI system.

Previous research in Explainable Artificial Intelligence (XAI) suggests that a main aim of explainability approaches is to satisfy specific interests, goals, expectations, needs, and demands regarding artificial systems (we call these “stakeholders' desiderata”) in a variety of contexts. However, the literature on XAI is vast, spreads out across multiple largely disconnected disciplines, and it often remains unclear how explainability approaches are supposed to achieve the goal of satisfying stakeholders' desiderata. This paper discusses the main classes of stakeholders calling for explainability of artificial systems and reviews their desiderata. We provide a model that explicitly spells out the main concepts and relations necessary to consider and investigate when evaluating, adjusting, choosing, and developing explainability approaches that aim to satisfy stakeholders' desiderata. This model can serve researchers from the variety of different disciplines involved in XAI as a common ground. It emphasizes where there is interdisciplinary potential in the evaluation and the development of explainability approaches.

This paper maintains that objectivism about practical reasons should be combined with pluralism both about the nature of practical reasons and about action explanations. We argue for an ‘expanding circle of practical reasons’, starting out from an open-minded monist objectivism. On this view, practical reasons are not limited to actual facts, but consist in states of affairs, possible facts that may or may not obtain. Going beyond such ‘that-ish’ reasons, we argue that goals are also bona fide practical reasons. This makes for a genuine pluralism about practical reasons. Furthermore, the facts or states of affairs that function as practical reasons are not exclusively natural or descriptive, but include normative facts. That normative facts can be reasons justifies a pluralism about reason explanations, one which allows for what we call enkratic explanations in addition to teleological ones.

In this comment on Katherine Dormandy's paper «True Faith», I point out that the clash she describes between epistemic norms and faith-based norms of belief needs to be supplemented with a clear understanding of the pertinent norms of belief. I argue that conceiving of them as evaluative fails to explain the clash, and that understanding them as prescriptive is no better. I suggest an understanding of these norms along the lines of Ross’s (1930) prima facie duties, and show how this picture can make sense of the clash.