The aim of the proposed paper is to investigate what (if any) was Peirce's theory of representation prior to the introduction of semiotic in Harvard Lectures of 1865. In 1861 Peirce defined metaphysics as the logical "analysis of conceptions" (W 1: 63).
The most fundamental of theses he designated bz the names for personal pronouns - I, It, and Thou (e.g. W 1: 45-49) which (besides other applications) serve in various manners as the metacategorical framework for generating systems (sic) of subordinated conceptions in terms of relations of which Peirce preferred to deal with diverse (predominantly) metaphysical problems.
I will concentrate on discriminating how by means of this apparatus Peirce thinks about representation. After indicating how Peirce supports his main thesis that representation is not conditioned by mind's capacity for conceptual synthesis, I proceed to explain his model of representation founded upon the relation of inherence of accidents in a substance, or in his terms, influxual dependency (W 1: 40) and analyze his transcendental argument that such conceived representation is a condition of possibility of arriving at true knowledge.
I claim that Peirce's semiotic can be seen as an actual fulfillment of his earlier metaphysical project of giving transcendental grounds of inference by accounting for how concepts can be inferentially generated out of other concepts, which is lacking in earlier theory.