1 Introduction

The aim of this special issue is to promote discussion on the value of truth and doxastic axiology. A key question of doxastic axiology is how to evaluate beliefs. There are three major approaches to it: the deontological one, the virtue theoretical one and the consequentialist one. The majority of the proponents of these approaches accepts the thesis that beliefs are primarily valuable because of their epistemic features. According to epistemic deontologism, certain epistemic norms determine whether it is permissible or not to have a certain belief (see Cohen 1984; Pollock 1987; and Feldman 2000). According to virtue epistemology, a valuable belief is an achievement of the epistemic agent manifesting an epistemic virtue in forming the belief in question (see Zagzebski 1996, 2003; Sosa and Bonjour 2003; Sosa 2007; Greco 2009; Riggs 2009; Greco and Turri 2012; Pritchard et al. 2018; Carter et al. 2015). Epistemic consequentialism says that, roughly, the epistemic benefits that follow from having a belief determine the value of the belief or the act of forming it (see Goldman 1979, 1986; Ahlstrom-Vij and Dunn 2014; Joyce 1998, 2009; Greaves and Wallace 2006; Leitgeb and Pettigrew 2010a, b, Pettigrew 2013).

How is the debate on doxastic axiology linked to the issue of the value of truth? According to a popular view, called veritism, the value of a belief is fundamentally provided by its truth or accuracy (see Goldman 1999; Pettigrew 2016; Pritchard 2016; Joyce 2018). Although the deontological approach, the virtue theoretical approach or the consequentialist approach to doxastic axiology do not entail veritism, the latter is often assumed more or less explicitly by the advocates of these three approaches. Nevertheless, within this debate, the role of truth remains partly controversial, since not everyone agrees that truth is the sole or the primary factor that provides value to beliefs, or that truth is even necessary to explicate doxastic value. Some contend that the value of beliefs is primarily provided by epistemic features but different from truth (e.g. their constituting knowledge), or that truth is just one among other equally fundamental epistemic values. Others argue that the value of a belief is provided only by non-epistemic features, or that the value of true belief is merely instrumental. If this is correct, even false beliefs might be valuable. Moreover, some contend that the value of truth can vary across different domains of discourse.

The contributions to this special issue on the value of truth cover four main topics. The first is the relation between truth and doxastic norms. A key question concerns the role of truth in determining our doxastic and epistemic norms (norms of belief, evidence, acceptance, testimony, etc.). The second topic is the primary units of epistemic evaluation, whether they are propositions or propositional attitudes as, for instance, most veritists usually assume, or something else, such as intellectual agents. The third topic is the nature of the value of truth, whether truth has a fundamental or an instrumental value depending on other, possibly non-epistemic, values. The fourth topic regards the plurality of the value of truth, whether the value of truth is domain-general and monistic, or domain-specific and pluralistic. In the following, we first elaborate on these four themes, then we summarize the main points of the contributions to this special issue.

The first topic is the relation between truth and epistemic norms, i.e. the role of truth in grounding doxastic evaluations. Let us start with the relation between truth and normativity. A seemingly trivial assumption could be that truth plays a central role in determining epistemic norms. However, some challenge the thesis that true beliefs are always more valuable than those that comply with norms about evidence but are false (see Cohen 1984; Feldman 2000; Wedgwood 2002). According to deontological epistemology, beliefs are evaluated on the basis of their conformity to certain epistemic obligations, regardless of the consequences and the sources of these beliefs (see Alston 1988; Vahid 1998; Nottelmann 2013; McHugh 2014; Sylvan and Ernest 2018). For instance deontic evidentialism is the view that it is epistemically rational for an agent to believe a proposition given her total evidence just in case her total evidence supports the belief (see Eder’s paper in this issue; also see Alston 1988; Vahid 1998; Conee and Fumerton 2008; Beddor 2015). In this case the (likely) truth of the belief in question does not necessarily determine whether an epistemically rational agent should or could believe it or not. By contrast, if epistemic rationality is understood as teleological and its aim is to produce true beliefs, then the norms of belief formation are considered appropriate when they serve that epistemic aim (see Littlejohn 2018; Wedgwood 2018; Wrenn 2016; for criticism see Berker 2013). For instance, if truth is the central epistemic aim, then truth serves as the basis of epistemic normativity: epistemic norms are such that following them results in (likely) true beliefs. However, if the end of inquiry does not determine epistemic norms, epistemic norms could be explained by something else than truth—for instance, the constitutive role that beliefs play in our cognitive life (see Nolfi 2015, and her contribution to this special issue).

The second topic is the value of truth and the nature of the units of epistemic evaluation. It is not obvious that truth must be a key value in epistemic evaluations. Many philosophers (for instance, the majority of veritists and consequentialists) agree that the basic units of epistemic evaluation are propositions or propositional attitudes, as they are the sort of things that can be true or false. However, virtue epistemologists take agents or communities to be the primary focus of epistemic evaluation, in which case the centrality of the value of truth becomes less clear (see Zagzebski 1996, 2003; Greco 2009, 2012; Sosa 2003, 2007; Riggs 2009). A suspect is, nevertheless, that the actual relations holding between veritism and apparently alternative views construing epistemic value in terms of intellectual virtues have not been adequately explored. One may defend a form of veritism just by taking intellectual agents as the primary subject of evaluation (see Pritchard, in this issue). Finally, even if propositions or propositional attitudes are taken to be the proper units of evaluation, a value different from truth—such as understanding—can be argued to be the basic epistemic value (see Kvanvig 2003).

The third topic discussed in this special issue is how and why truth is valuable. It involves many complications. Three sub-problems can be distinguished. First, whether or not truth is the fundamental epistemic value. For veritists, the answer is affirmative (see Goldman 1999; Pettigrew 2016). As said, some disagree and argue that other things such as certain propositional attitudes or states of the subject—prominently, understanding—bear the fundamental epistemic value (see Kvanvig 2003; Haddock et al. 2009). The second sub-problem concerns the nature of the value of truth: whether this is internal (i.e. it arises from the nature of truth itself) or is external (i.e. it is grounded in something else) (for discussion, see for example Goldman 1999; Sosa 2000, 2003, 2004; David 2001, 2005; Lynch 2004; Alston 2005; DePaul 2009; Baehr 2012). If the value of truth is internal, it is truth itself that makes believing, acquiring or promoting true propositions good or desirable. If the value of truth is external, then this value arises from other values—for example, the value of believing, acquiring or promoting propositions that foster intellectual or human flourishing (see Bader 2013 and Ferrari’s contribution to this special issue). Related to this question, one might ask in what sense, exactly, truth is valuable (e.g. practical, social, moral, etc.). This leads us to the third sub-problem: is the value of truth fundamentally epistemic, or is truth only instrumentally valuable as a means to other types of values? Some have argued that true beliefs lack any intrinsic epistemic value (see Gaultier 2017, Wrenn 2017). What looks pretty certain is that having or achieving true beliefs has some non-epistemic value (that is, social, practical, moral, etc.). Purist views about the value of truth, however, holds that at least in certain cases or domains, having true beliefs is also intrinsically valuable and remains so even if having those true beliefs lacks any social, practical value (see Goldman 1999; David 2001, 2005; Lynch 2004; Alston 2005; Baehr 2012). This view does not exclude that true beliefs typically have social and practical value, but it implies that (in some cases at least) truth is valuable for the sake of itself (such as, when the inquiry is motivated by mere curiosity). Some insist, however, that the value of truth is always instrumental (see McCormick’s contribution). For them, true beliefs are valuable only because they help us achieve or promote various practical, social, or moral goals.

The fourth topic concerns how alethic pluralism relates to the question of the value of truth. Alethic pluralism is the view that, roughly, there is more than one truth property. Propositions in different domains (everyday, scientific, moral, aesthetics, taste) may be true in different ways if they are true at all (see Wright 1992, 2001). Alethic monism, on the other hand, posits that all true propositions are true in the same sense. Pluralism is motivated by the intuition that certain theories of truth—such as correspondence theory—are the most plausible in domains of factual everyday claims as well as basic scientific claims, while in some other domains—e.g. those concerning matters of taste and aesthetics—certain forms of alethic anti-realism seem to be more plausible candidates. The alethic pluralist claims that truth is a substantive property in all areas of discourse (see Lynch 2009a, b), even in those where we seem to be less interested in knowing the truth (for instance, those concerning matters of taste). This suggests that the value of truth may vary across different domains (see Ferrari’s contribution).

This special issue contains ten articles that provide important and original contributions to these debates. Six papers have been authored by invited speakers to the conference “The Value of Truth”, which took place at the Institute of Philosophy of the Hungarian Academy of Sciences, Budapest, in November 2018. These authors are Anna-Maria Eder, Filippo Ferrari, Tihamér Margitay, Miriam McCormick, Kate Nolfi, and Duncan Pritchard. In addition, this special issue contains four call-for-papers contributions by Joseph Bjelde, Jim Hutchinson, Tommaso Piazza, and Daniel E. Weissglass.

The paper by Eder, Margitay, Nolfi, Piazza and Weissglass discuss relations between truth and doxastic norms of epistemic or non-epistemic rationality. Eder’s paper, “No commitment to the truth”, focuses on deontic evidentialism, according to which it is epistemically rational for us to believe propositions supported by our total evidence. Eder essentially contends that appealing to epistemic teleology to explain the normativity of epistemic rationality does not support the claim that we ought to believe what is rational to believe—it only supports the claim that we are permitted to do so. In arguing for this conclusion, Eder defends a special epistemic teleological position that involves no commitment to aiming at the truth.

In his paper, “Epistemology of testimony and values of science”, Margitay first argues that the intrinsic epistemic value of testimony in science—i.e. its serving as evidence for the truth of what is reported—is reducible to its moral and social value—i.e. its competent, conscientious, and honest performance. Margitay infers from this that competence, conscientiousness, and honesty also count as intrinsic epistemic values in science, and that the norms that follow from these values have both an epistemic and social function. In light of this thesis, in the second part of the paper, Margitay addresses the questions of why and under what conditions a hearer can rationally accept a testimony in science.

Nolfi’s paper, “Epistemic norms, all things considered”, defends a form of action-oriented epistemology, according to which our capacity for belief is embedded within, as a constitutive component of, our capacity for action. Nolfi argues that her action-oriented epistemology provides us with a good explanation of the fact that our evaluative practice seems to be premised on the assumption that what we epistemically ought to believe helps to determine what we ought all-things-considered to believe. The paper criticises traditional attempts in epistemology to make sense of this phenomenon—prominently, the view that truth is the aim of belief-forming and, thus, the ground of normative reasons.

Piazza’s contribution, “The value of truth and the normativity of evidence”, focuses on McHugh’s recent suggestion to vindicate the thesis that evidence is normative (i.e. that it determines which doxastic attitude we ought to take in each case) by adducing the assumption that truth is epistemically valuable. Piazza argues that some of the difficulties that afflict McHugh’s strategy can be overcome. Yet he contends that this strategy ultimately fails because it commits us to acknowledging non-evidential reasons for believing. Piazza examines the possibility that we could have non-evidential reasons for believing because we can believe things on the basis of non-epistemic considerations, but he rejects the principal strategies to show that we can believe on non-epistemic considerations.

In his paper, “Is belief evaluation truth sensitive? A reply to Turri”, Weissglass analyses the recent debate on whether the truthfulness of a proposition is or should be a relevant factor in our evaluation of beliefs. The most popular view on this matter—i.e. non-factivism—holds that we should judge the reasonableness of a belief independently of its truth. By contrast, factivism holds that our ordinary evaluations of beliefs are truth-sensitive. Recent empirical studies by Turri seems to support factivism. This paper argues, however, that Turri’s factivist conclusion is actually unsupported.

The problem of the value of truth and the units of epistemic evaluation is explored in Pritchard’s and Hutchinson’s contributions. Pritchard’s paper, “Intellectual Virtues and the Epistemic Value of Truth”, offers a novel defence of veritism relying on an account of intellectual virtues. Pritchard reviews that main objections recently raised against veritism and argues that they don’t actually refute the thesis that truth is the fundamental epistemic value. He suggests that what is problematic is, not veritism in itself, but the erroneous way in which it has been unpacked. Pritchard proposes an original way to account for veritism that explains it in terms of what an intellectually virtuous subject would value, and so in terms of an intellectually virtuous inquiry.

Hutchinson’s paper, “Why can’t what is true be valuable?” focuses on an important feature of the recent debate on the value of truth. Hutchinson notes that the participants in this debate typically assume that what is alethically valuable is not the things that are true—i.e. propositions—but only certain attitudes, states or activities associated with them—e.g. believing, knowing, or investigating. Hutchinson scrutinizes the five major arguments for this assumption, and finds all of them flawed or seriously problematic in different respects. Although he ultimately puts forward no argument for the thesis that it is propositions that are alethically valuable, the paper offers some promising starting points for such a case.

The next two contributions—McCormick’s and Bjelde’s—concentrate on the question of how and why truth is valuable. McCormick’s article, “Value beyond truth-value: a practical response to skepticism”, interweaves the investigation of the value of truth with the problem of responding to the sceptic. McCormick argues that the value of truth is not intrinsic but only instrumental: true beliefs are ultimately valuable because they contribute to both individual and collective flourishing. She contends that if contributing to flourishing is what provides truth with its value, this must count as a doxastic value, and that scepticism can be answered by insisting that some beliefs can be justified by appealing to this doxastic value. The paper answers various potential objections and contrasts this response to the sceptic with similar replies, such as Crispin Wright’s entitlement theory and Susanna Rinard’s “pragmatic skepticism”.

Bjelde’s paper, “Anything but the truth”, argues that truth is not a fundamental epistemic value. The argument is two-fold. Bjelde first contends that if truth were a fundamental epistemic value, it should have a central role in the best explanation of certain epistemic evaluations. But this is not the case. For example, evaluations of beliefs as epistemically justified should receive their best explanation by adducing truth as an epistemic value. Yet these epistemic evaluations are explained at least as well by positing evidential support, rather than truth, as the central epistemic value. Bjelde also argues that although there is a set of evaluations which truth seems to best explain, according to veritists, these evaluations are not actually epistemic. For example, truth has been asserted to best explain why we generally have pro tanto reasons to desire true beliefs, but the evaluation explained is not an epistemic one. The paper finally analyses the assumptions of this two-fold argument in order to delimitate the boundaries of its effectiveness.

Finally, Ferrari’s paper “Alethic pluralism and the value of truth” investigates whether and to what extent truth is valuable. It does so by isolating the value question from other normative issues while importing into the debate about the nature of truth some key distinctions from value theory. After analysing paradigmatic cases of disagreement in different fields, Ferrari suggests that there is significant variability in the value of truth in different areas of discourse. He then investigates how pluralistic accounts of the nature of truth deal with this variability, and finds out that it is a problem for certain popular versions of truth pluralism exemplified by Michael Lynch’s alethic functionalism. Ferrari proposes two ways out for advocates of these theories that require structural changes in them.