Epistemic norms, values and virtues

Listen to audio

1. We began by recalling the significance of the return of the virtues in morality and epistemology, which took place in reaction to both consequentialism and Kantian deontologism, and resulted in a return to the Aristotelian virtues in ethics and epistemology, with an emphasis on reliable cognitive faculties, intellectual virtues and responsibility. This gave rise to several programs: a weak one, aimed at integrating these elements into a classical framework, and a strong one, aimed at radically changing the model by advocating an "epistemology of virtues", as part of an agent-centered or agent-based approach. We have thus witnessed a restructuring of the field, a broadening of knowledge and truth to include other values such as understanding, or the flourishing of intellectual virtues associated with moral virtues. But does this solve the problems of defining knowledge and its value, and does it also respond to the skeptical challenge?

2. We then set out to analyze what is meant by "epistemic normativity " and, following P. Engel (2007). First of all, epistemic norms and standards can be understood as norms or rules of signification, norms specific to concepts, commitments or normative requisites of rationality, justifications or reasons one might have for believing such and such a proposition, general norms of knowledge and inquiry. Secondly, we need to consider the links between norms and values, their priority, their reduction to one another - and if so, which ones? Thirdly, we need to clarify the meaning of the concepts of norms, rationality and reasons. We talk about reasons to act and reasons to believe: how far can we push the parallel? Fourthly, what do we mean by the norm of rationality? We have seen that it meets two requirements: that of coherence and that of deductive closure:( a) we must not believe p and non-p; (b) we must believe q if we believe that p and that p implies q. We have evoked the problems raised by high-level norms: how do we analyze normative force and freedom (where does the limit of what we can or must violate pass?)? How do we understand the broad and narrow scope (detachment or non-detachment) of norms? What is the relationship between rational requirements and psychological reasoning? Fifthly, normative regulation and normative objectivity: how do norms guide or govern? Can norms be reduced to rules or conventions? Finally, we raised the question of the possible naturalization of epistemic norms: do norms "arise" from natural facts? What is the normativity of the mind? Is there a "hierarchy" of mental states? What links can be established between knowledge and metacognition? How can we solve the problem of integrating epistemic norms into a naturalistic conception of the world?

3. In a third section, we examined epistemic obligations and duties, and the now classic problem raised by the ethics of belief. We presented the terms of the Clifford/James debate and the lessons to be learned from it. For Clifford, "it is wrong, always, everywhere, and for anyone, to believe anything on the basis of insufficient evidence". For William James, on the contrary: it's not always wrong to believe on the basis of insufficient evidence. There are cases in which, following the path of preferring the injunction "we must know the truth" to the other "we must avoid error", it can be good (even rational) to believe things for which we have no good reason to believe, and even, in some cases, it can be good to believe against the evidence at hand. We need to distinguish, on the one hand, between epistemic justification and ethical justification, and on the other, between what may be valid (and even valued) at the level ofinquiry and what is valid at the level of epistemic justification. The lesson here is twofold. There is no strict parallel between the practical sphere and the epistemic sphere; but there is a relationship. Otherwise, there would be a disconnect between our cognitive and epistemic reasons or virtues, on the one hand, and our practical reasons or virtues, on the other.

4. We then wondered about the advantages of such an approach in terms of virtues. Why should virtues be preferred to epistemic duties in the first place? Secondly, what is the content of epistemic duties? Are they categorical or hypothetical duties? What problems does virtue pose in terms of "responsibility" (Montmarquet, Zagzebski)? We then looked at the difficulties of deontological conception, and the margin that could be envisaged between the (right) idea of an impossible will to believe and that of possible indirect control of belief formation, before tackling the question of intellectual virtue versus obligation, by analyzing the position of Ernst Sosa, and the triple condition (AAA) he proposes: accuracy, adroitness, aptness. To be precise, to be adroit, to be just or apt. As the archer can be(Virtue Epistemology, 2007, chap. 2, 22 sq.). For an archer's shot to be apt (well adjusted or just), it must not be precise and adroit, but precise(accurate) becauseadroit(adroit), in short, successful because competent. To this must be added a condition of sentience: S's belief that p is sentient if and only if, if p were not true, S would (probably) not believe that p - to be distinguished from the condition of security: a belief that p is secure as long as it would only have been held if (most probably) p.(ibid., 25). A belief can be secure without being sensible. Knowledge requires not full security, but relative security.

We can arrive at an analysis of subjective justification in terms of epistemic responsibility by following these principles:1) S is subjectively justified in believing p if and only if S's belief that p is epistemically responsible; 2) S's belief that p is subjectively justified if and only if it is the result of S's motivations to try to believe the truth (subjective condition), and if S's dispositions (abilities, powers, virtues) make S reliable ; 3) S knows that p if and only if (a) S is subjectively justified in believing that p, and (b) this results from S being objectively reliable in believing that p; 4) adding a clause like that proposed by John Greco(Achieving Knowledge, 2010), "Knowledge is a kind of success from ability". "Knowledge is a special case of more general normative phenomena - that of success from ability (or excellence, or virtue).

5. In conclusion, it has been asked whether the epistemology of virtue solves the problems of epistemology, as it claims. As P. Engel (2007, 146-153), it is first of all doubtful whether all epistemic virtues can be accorded the same status and need not be hierarchized, and also rather doubtful whether intellectual virtues and moral virtues can be so closely linked. What's more, one can have fulfilled all one's epistemic obligations and still hold unjustified beliefs. We also arrive at a questionable anti-realist conception of truth and knowledge. Finally, virtue is a concept that applies better to individuals than to a community. This does not mean that we are dealing with outdated ideals or epistemic norms that do not exist. But it does mean that we need to look for another model that is better able to respond to the twofold challenge to knowledge posed by relativism and skepticism.