The Nature of Representational Content: Original but Not Intrinsic

Abstract

What makes a representational content representational? A widespread assumption, often unstated, is that content-hood is an intrinsic feature: a determinate fact, written into a state, of what it represents. This paper argues that the assumption cannot be sustained, and that its retirement dissolves a family of famous puzzles. Building on a companion account of the representational asymmetry, on which the generative act of representing is never among the contents it produces, we run the asymmetry in the reverse direction: what makes a content representational is never itself represented. Searching a content for its content-hood yields only further contents, exactly as searching for the act does. Representational status is therefore role-deep: constituted in the act’s sorting, fully real as role-filler, with nothing representational beneath the role; original, but not intrinsic. Act and content are accordingly two roles of a single occurrence, not two items in sequence. Dennett saw the deflationary half of this and mislocated its ground, relativizing content to an interpreter’s stance and inviting a regress of ascribers; constitution in the producing closes the regress. The account is then tested where the two-items picture visibly breaks: the temporal anomalies of consciousness and the Orwellian/Stalinesque dispute, whose dissolution is recovered without Dennett’s verificationism and with phenomenal realism intact.