Our main topic is epistemic agency, which can be either free or unfree. This aligns with a distinction between two sorts of knowledge, the reflective and the animal. We first take up the nature and significance of these two sorts of knowledge, starting with the refl ective. In a second section we then consider the nature of suspension and how that relates suspension to higher orders of meta-belief. Finally, we consider a distinction in epistemology between animal competence and refl ective justification. All of these topics and distinctions are important for virtue epistemology, in ways to be considered.

The gist of modal epistemology is expressed in the idea that you fail to know if you do believe truly but it is seriously possible for you to believe falsely. According to subjunctivism, this idea is captured by certain subjunctive conditionals. One formulation invokes a safety condition—“If S had believed P, then P would have been the case,” while the other invokes a sensitivity condition—“If P had been false, S would not have believed that P.” According to simple subjunctivism, such conditionals do not contrapose and Sosa derives important epistemological consequences which favor safety from this difference. However, simple subjunctivism is inadequate. I return to Goodman and his analysis of factuals and propose modal stability, which is restricted sensitivity or enhanced safety as a proper epistemic condition for the non-accidental connection between the basis for the belief and the relevant facts of the matter. The idea of modal stability combines robustness (benefi ts of safety) with responsiveness to facts (benefi ts of sensitivity) and recovers the original motivation for the relevant alternatives theory—when testing for claims of knowledge that p we ask what might be the case if not-p, but we ignore irrelevant possibilities. Epistemic modal conditions should be expressed in terms of conditionals of connection which contrapose within the limits of relevance.

Ernest Sosa has made and continues to make major contributions to a wide variety of topics in epistemology. In this paper I discuss some of his core ideas about the nature of knowledge and scepticism. I start with a discussion of and objections against the safety account of knowledge—a view he has championed and further developed over the years. I continue with some questions concerning the role of the concept of an epistemic virtue for our understanding of knowledge. Safety and virtue hang very closely together for Sosa. All this easily leads to some thoughts on epistemic scepticism and on Sosa’s stance on this.

The paper discusses Sosa’s view of intuitional knowledge and raises the question of the nature of reflective justification of intuitional beliefs. It is assumed, in agreement with Sosa, that pieces of belief of good researchers are typically reflectively justified, in addition to being immediately, first-level justified. Sosa has convincingly argued that reflective justification typically mobilizes and indeed should mobilize capacities distinct from the original capacity that has produced the belief-candidate for being justified, in order to assess the reliability of the original capacity. It has to go beyond justifiers that are of the same-kind (“homogeneous”) as first-level immediate ones, in order to enlarge the circle of justification (and thus avoid viciousness), and is, therefore, holistic and coherentist. But if this holds, it seems that reflective justifi cation of armchair beliefs, presumably produced by intuition and some reasoning, should revert to empirical considerations testifying to the reliability of intuition and reasoning. Therefore, it typically combines, in an articulated way, a posteriori elements contributing to the thinker’s reflective trust in her armchair capacities. In short, the paper argues that Sosa’s own view of second-order justification goes better with a more aposteriorist view, if it does not even force such a view.

Frank Jackson’s famous Knowledge Argument moves from the premise that complete physical knowledge about experiences is not complete knowledge about experiences to the falsity of physicalism. Some physicalists (e.g., John Perry) have countered by arguing that what Jackson’s Mary, the perfect scientist who acquires all physical knowledge about experiencing red while being locked in a monochromatic room, lacks before experiencing red is merely a piece of recognitional knowledge of an identity, and that since lacking a piece of recognitional knowledge of an identity does not entail lacking any pieces of knowledge of worldly facts, physicalism is safe. I will argue that what Mary lacks in her room is not merely a piece of recognitional knowledge of an identity and that some physicalists have failed to see this because of a failure to appreciate that Mary’s epistemic progress when she fi rst experiences red has two different stages. While the second epistemic stage can perhaps be plausibly considered as acquiring merely a piece of recognitional knowledge of an identity, there is good reason to think that the fi rst epistemic stage cannot be thus considered.

G. E. L. Owen, in his infl uential paper “Inherence,” talks of “vink,” a name he has created for a particular shade of the color pink, and this “vink” serves as an individual in the Aristotelian category of quality. Owen was one of the first to aim to discredit the belief that J. L. Ackrill and his camp espoused, the belief that Aristotle thought that “general attributes are not in individuals, particular attributes are not in more than one individual.” I postulate that there is nothing here that does not preclude the existence of transferable nonsubstantial particulars, and base this view on passages from Aristotle’s Categories and certain examples found in Ammonius’s commentary and On Colors. Given this, a nonsubstantial particular of “vink” would not have to rely on having inhered in just one particular body to have existence, however, it would have to inhere in at least one particular body.