Introduction

In this chapter, I will explore the idea of a “hinge epistemology,” i.e., a theory about justification (and knowledge) which gives center stage to Wittgenstein's notion of hinges. I will first put forward some basic methodological considerations regarding the relationship between merely exegetical work on Wittgenstein's texts, with special reference to On Certainty (OC), and more theoretically committed work, which aims at developing suggestions which can be found in that text, even though not clearly attributable as such to his author (Section 2). I will then summarize in broad outline the main problems raised in OC and the most influential interpretative lines it has given rise to so far (Sections 3–6). In light of the initial methodological considerations, I will review some contemporary attempts at developing Wittgenstein's ideas in an anti-skeptical direction, such as Crispin Wright's and Michael Williams’ (Section 7). I will argue that, their intrinsic merits notwithstanding, they fail to take proper measure of Wittgenstein's own position. I will then close by sketching my own version of hinge epistemology and by highlighting points of contact and disagreement with Wittgenstein's own views (Section 8).

Some Basic Methodological Considerations

Let me start with some platitudes. When engaged in doing history of philosophy—analytic or otherwise—as opposed to doing philosophy as such, we are all first and foremost involved in a process of interpretation of more or less remote texts. Such interpretations can't be assessed merely on the basis of their own specific philosophical interest and merits, if they are meant as exegeses. Rather, they will have to be evaluated with respect to the text of which they are purported interpretations, as well as to other historical evidence pertaining to what we may generically call the “context” in which the work was created. I believe there is a strong tendency, especially in analytic philosophy, to forget such a methodological requirement and thus to end up presenting more how a text struck one, as it were, than what that text itself tried to say, in light of its author’s intentions and context. However, to put it baldly, rational reconstructions without textual and contextual evidence, or else supported by very little textual and contextual evidence, seem to me to be “empty” —to use a piece of Kantian terminology. That is to say, they are devoid of historical content.