The Problem of the Epistemic Status of Rules: Wilfrid Sellars on the Material Rules of Inference

Abstract

The epistemic status of our access to the world has always given rise to vivacious debates in Western philosophy. The crucial point of these discussions was the connection between conceptual and non-conceptual elements of our relation to the world. The core of this problem can be also formulated as the question of connection between semantic and syntactical rules of our epistemic relation to the world. Could we know anything about some state of affairs by using only syntactical rules, which relate logical terms to one another, or we also need semantic rules, which connect the formal syntax of our epistemic relation to the world with extra-conceptual characteristics of each state of affairs?

Wilfrid Sellars’ approach to this problem is a critical reaction to the traditional theories of rules. The main point of his criticism of the traditional theories of rules, whose origins Sellars attributed to Kant’s conception of rules (which in my argument is an erroneous attribution), could be summed up as follows: the difference between syntactical and semantic rules of our epistemic relation to world is founded on the Myth of the Given. In his Inference and Meaning (1953) Sellars elaborated a theory of material rules of inference which tries to neutralize the traditional approach to the problem (above all, Carnap’s theory of logical syntax of language). I am of the opinion – and it is my major claim in this paper – that Sellars’ argument in IM suggests a way of avoiding the Myth (even if this article was published earlier as EPM). I argue that Sellars’ argument consists in three basic steps: (1) irreducibility as well as (2) operative normativity of the material rules of inference and (3) linguisticality of descriptive terms.