Luis Rosa

This paper challenges Sosa’s account of the epistemic propriety of suspension of judgment. We take the reader on a test drive through some common problem cases in epistemology, to see how the account of suspension deals with them. Section 1 outlines Sosa’s account of accurate and apt suspension. Sections 2 and 3 focus on the skeptical scenario to argue that Sosa makes it too easy for suspension to be accurate and apt. Section 4 applies Sosa’s account of accurate and apt suspension to BonJour’s case of Norman the clairvoyant. The upshot is that Sosa makes accurate and apt suspension not only too easy but also too hard.

The paper investigates whether and how AI systems can realize states of uncertainty. By adopting a functionalist and behavioral perspective, it examines how symbolic, connectionist and hybrid architectures make room for uncertainty. The paper distinguishes between epistemic uncertainty, or uncertainty inherent in the data or information, and subjective uncertainty, or the system's own attitude of being uncertain. It further distinguishes between distributed and discrete realizations of subjective uncertainty. A key contribution is the idea that some states of uncertainty are interrogative attitudes whose content is a question rather than a proposition.

Why should we avoid incoherence? An influential view tells us that incoherent combinations of attitudes are such that it is impossible for all of those attitudes to be simultaneously vindicated by the evidence. But it is not clear whether this view explains what is wrong with certain akratic doxastic states. In this paper, I flesh out an alternative response to that question, one according to which the problem with incoherent combinations of attitudes is that it is impossible for all of those attitudes to be simultaneously knowledgeable. This alternative response explains what is wrong with akratic combinations of attitudes using commonly accepted epistemological theses. The paper still shows how this proposal is able to explain the badness of incoherent combinations involving the absence of attitudes, suspended judgment, and credence. Finally, it is suggested that this picture can be generalized to the realm of practical rationality as well.

ABSTRACT In this paper I present reasons for us to accept the hypothesis that suspended judgment has correctness conditions, just like beliefs do. Roughly put, the idea is that suspended judgment about p is correct when both p and ¬p might be true in view of certain facts that characterize the subject's situation. The reasons to accept that hypothesis are broadly theoretical ones: it adds unifying power to our epistemological theories, it delivers good and conservative consequences, and it allows us to assess processes of reasoning involving attitudes of suspended judgment.

The paper proposes to expand modal doxastic logic with ascriptions of attitudes toward questions, or attitudes whose contents are questions, besides ascriptions of belief toward propositions. The relevant attitudes are the attitude of being in doubt and the attitude of rejecting a question, respectively. In order to accommodate such ascriptions, modifications on the canonical framework of doxastic logic are called for. Interrogative complements are added to its grammar, and questions are included among the possible semantic values of expressions. New valid schemas and consequence relations are grounded on the suitably modified framework—schemas and consequence relations that correspond to principles of coherence or structural rationality. The relevant principles are concerned with the relationship between belief, doubtfulness, and rejection, as well as how doubtfulness and rejection spread from one question to another in an ideally rational mind.

This article argues that the traditional taxonomy of doxastic attitudes leaves out an important alternative. That is the attitude of rejecting the question. Sundry examples of the phenomenon are presented and their unifying features are identified. One rejects a question when one treats it as if none of its answers were true. One can reject not only non-polar questions, but also polar ones (questions of the form whether p). The stance of rejecting a question is further distinguished from interrogative attitudes in general, and from the third stance of the traditional taxonomy in particular.

Intuitively, an inquirer should avoid inquiring into pseudo-questions. It turns out to be difficult to make more precise sense of that norm, however. According to one view, pseudo-questions are not legitimate questions at all. But that entails that it is not even in principle possible for an inquirer to inquire into a pseudo-question—for inquiry is always inquiry into some question or other. So if the target norm isn’t supposed to be one that every inquirer trivially abides by all the time, we better have another way of understanding what a pseudo-question is. In this paper, I explore different ways of drawing the distinction between legitimate and pseudo-questions, each of which leads to a different precisification of the initially formulated norm. It turns out that, whereas some of those precisifications are indeed norms of inquiry, others are not.

This book concerns the nature and the norms of inquiry. It tackles not only philosophical issues regarding what inquiry is, but also issues regarding how it should and should not be executed. Roughly put, inquiry is the activity of searching for the true answers to questions of our interest. But what is the difference between empirical and armchair inquiry? And what are the right and the wrong ways to inquire? Under what conditions should one start inquiring? Which questions are such that one should not inquire into them? The book offers answers to these questions. It argues that competent armchair inquiry only makes explicit what was already implicit--the inquirer already had the answer to her question before inquiring into it, though this was not explicit to her. It also argues that we should avoid inquiring into questions whose answers are unknowable to us, in the instrumental sense of 'should', and that different modes of inquiry are called for, depending on which type of information is available to the subject. These answers are rigorously argued for, and they stem from a unified framework for modeling the activity of inquiry.

The paper is concerned with the rational requirements for suspended judgment, or what suspending judgment about a question rationally commits one to. It shows that two purported rational requirements for suspended judgment cannot both be true at the same time, at least when the entailment relation between questions is understood a certain way. The first one says that one is rationally required to suspend judgment about those questions that are entailed by the questions that one already suspends judgment about. The second one says that one is rationally required to believe the presuppositions of the questions one suspends judgment about. Two plausible solutions to the problem are discussed. One of them explicates the relation of entailment between questions in an alternative manner, and the other one rejects the presupposition requirement. Either way, a closure requirement for suspended judgment can be maintained which is analogous to the closure requirement for belief. Whereas belief is ideally closed under entailment between propositions, suspended judgment is ideally closed under entailment between questions.

Epistemologists often describe subjects as being capable of adopting a third kind of categorical doxastic stance regarding whether something is the case, besides belief and disbelief. They deploy a variety of idioms in order to ascribe that stance. In this paper, I flesh out the properties that the third kind of categorical stance is supposed to have and start searching for the best ways to ascribe it. The idioms ‘suspends judgment about whether’ and ‘is agnostic about whether’, among others, are found to be unfit to play the desired role. In the end, I suggest that ‘is in doubt as to whether’ is our best choice among the alternatives surveyed here.

There are many ways in which a speaker can confuse their audience. In this paper, I will focus on one such way, namely, a way of talking that seems to manifest a cross-level kind of cognitive dissonance on the part of the speaker. The goal of the paper is to explain why such ways of talking sound so twisted. The explanation is two-pronged, since their twisted nature may come either from the very mental states that the speaker thereby makes manifest, or from how the speaker chooses to express themselves (even if there is nothing wrong with their mental states). So-called ‘Moore-paradoxical’ utterances are but one example of the phenomenon, and the explanation of what is wrong about them is subsumed under a more general explanation here—one that captures also the twisted-ness of utterances whereby questions are raised or intentions expressed.

In this paper I address the question of what makes deduction reliable (when it is reliable). A successful answer to that question would help us understand how deduction can expand and generate knowledge. I explore two answers to the main question. The first one makes use of the notion of formal-logical entailment, whereas the second one makes use of the notion of metaphysical entailment. The latter is found to be superior to the former, in that it allows us to explain the reliability of a wider class of reliable types of deduction.

Tyler Burge and Crispin Wright both distinguish two forms of warrant: entitlement and justification. But they use these terms in very different ways. Entitlement for Wright is a non-evidential, a priori rational right to claim knowledge against the skeptic. Wright’s project engages the skeptic. Entitlement for Burge is a truth-conducive good route to knowledge that does not involve reasons. Justification is the route that involves reasons. Burge’s project falls within moderate foundationalist, competence-based approaches to knowledge. Burge’s project examines the structure of knowledge. The chapters of the volume are introduced. The chapters in Part I engage Burge’s project. Part II engages and extends competence and moderate foundationalist approaches. Part II engages Wright’s project.

In the first part of this paper, we argue that the claim that a subject S believes that ϕ on the basis of good reasons cannot be the only type of explanation why S rationally believes that ϕ. Explaining attributions of rationality only by means of the notion of a belief being based on good reasons generates one version of the problem of regress of reasons. In the second part we flesh out a hypothesis according to which some beliefs are rationally held by a subject S in virtue of the fact that S knows how to produce good reasons for holding them. In the third part we offer reasons for accepting the relevant hypothesis, and we argue that it is more successful than foundationalist, coherentist and infinitist theories in explaining the truth of some attributions of epistemic rationality. In the last part we address the problem of the regress of reasons and sketch a solution for it based on our hypothesis.

Epistemologists take themselves to disagree about whether there are situations where it is rational for one to believe that p and rational for one to believe that one’s evidence does not support p (rational akrasia). The embedded sentence ‘one’s evidence does not support p’ can be interpreted in two ways, however, depending on what the semantic contribution of ‘one’s evidence’ is taken to be. ‘One’s evidence’ might be seen as a sheer indexical or as a descriptive singular term. The first interpretation makes the relevant kind of rational akrasia impossible, whereas the second one makes it possible. But the proposition that is taken to be expressed by ‘one’s evidence does not support p’ by each of these interpretations is not the same. We thus have a rational reconstruction of views that are labeled as being for and against the possibility of rational akrasia according to which they do not really contradict each other.

Why should we avoid incoherence? An influential view tells us that incoherent combinations of attitudes are such that it is impossible for all of those attitudes to be simultaneously vindicated by the evidence. But it is not clear whether this view successfully explains what is wrong with certain akratic doxastic states. In this paper I flesh out an alternative response to that question, one according to which the problem with incoherent combinations of attitudes is that it is impossible for all of those attitudes to be simultaneously knowledgeable. This alternative response explains what is wrong with akratic combinations of attitudes using commonly accepted epistemological theses. The paper still shows how this proposal is able to explain the badness of incoherent combinations involving the absence of attitudes, suspended judgment and credence. These explanations deploy the notions of knowledge and being in a position to know, instead of the notion of responding properly to the evidence. Finally, I suggest that this picture can be generalized to the realm of practical rationality as well.

In this paper I present reasons for us to accept the hypothesis that suspended judgment has correctness conditions, just like beliefs do. Roughly put, the idea is that suspended judgment about p is correct when both p and ¬p might be true in view of certain facts that characterize the subject’s situation. The reasons to accept that hypothesis are broadly theoretical ones: it adds unifying power to our epistemological theories, it delivers good and conservative consequences, and it allows us to assess processes of reasoning involving attitudes of suspended judgment.

In this paper, I flesh out an account of the inferential basing relation using a theory about how humans reason: the mental models theory. I critically assess some of the notions that are used by that theory to account for inferential phenomena. To the extent that the mental models theory is well confirmed, that account of basing would be motivated on empirical grounds. This work illustrates how epistemologists could offer explications of the basing relation which are more detailed and less empirically risky.

How does rationality bind the agnostic, that is, the one who suspends judgment about whether a given proposition is true? In this paper I explore two alternative ways of establishing what the rational requirements of agnosticism are: the Lockean–Bayesian framework and the doxastic logic framework. Each of these proposals faces strong objections. Fortunately, however, there is a rich kernel of requirements of agnosticism that are vindicated by both of them. One can then endorse the requirements that belong to that kernel without thereby committing oneself to the problematic implications that stem from either of the aforementioned proposals.

Logic arguably plays a role in the normativity of reasoning. In particular, there are plausible norms of belief/disbelief whose antecedents are constituted by claims about what follows from what. But is logic also relevant to the normativity of agnostic attitudes? The question here is whether logical entailment also puts constraints on what kinds of things one can suspend judgment about. In this paper I address that question and I give a positive answer to it. In particular, I advance two logical norms of agnosticism, where the first one allows us to assess situations in which the subject is agnostic about the conclusion of a valid argument and the second one allows us to assess situations in which the subject is agnostic about one of the premises of a valid argument.

A proposição expressa por “Deus existe”, se é verdadeira ou falsa, ela é ou necessariamente verdadeira/falsa ou não necessariamente verdadeira/falsa. Em outras palavras, se G é capaz de ter um valor de verdade v, então ela é ou necessariamente v ou contingentemente v. por “Deus” eu quero significar um ser sobrenatural, com uma mente poderosa e imaterial que supostamente criou o universo. Certamente existem outros significados que estão vinculados a esse termo em certos contextos, mas os argumentos que eu irei avaliar aqui só fazem sentido sob essa interpretação.

In this paper I deal with epistemological issues that stem from the hypothesis that reasoning is not only a means of transmitting knowledge from premise-beliefs to conclusion-beliefs, but also a primary source of knowledge in its own right. The idea is that one can gain new knowledge on the basis of suppositional reasoning. After making some preliminary distinctions, I argue that there are no good reasons to think that purported examples of knowledge grounded on pure reasoning are just examples of premise-based inferences in disguise. Next, I establish what kinds of true propositions can to a first approximation be known on the basis of pure reasoning. Finally, I argue that beliefs that are competently formed on the basis of suppositional reasoning satisfy both externalist and internalist criteria of justification.