Derrida's early work is often quite plausibly thought to be about origins. Indeed, if we wanted to identify an origin for deconstruction itself, we might say that it comes to Derrida with the thought that the origin is irreducibly complex. ‘Originary synthesis’, as the work on Husserl was inclined to say, and soon enough ‘originary trace’. What Derrida calls ‘metaphysics’ tries to lead things back to an origin point that would be simple (call it ‘presence’): deconstruction involves the claim that in the beginning is a complexity that resists further analysis in the strict sense, and that simple origins are always only retro-jected after the fact in the more or less compelling stories or myths that metaphysics recounts. Derrida wants to account for the (undeniable) effects of presence by developing a ‘prior’ trace- or text-structure which allows for what looks like presence to emerge while never itself being describable in terms of presence.

But if, in the beginning of Derrida's thinking, it was most obviously about beginnings, at the end it was arguably more about ends. Metaphysics, finding itself always in the middle, in complexity (‘in a text already’, as the Grammatology says) tries not only to track back and then derive that complexity from a simple present origin point, but also to put that complexity in the (convergent) perspective of an end point or resolution. Complexity should come from something simple, says metaphysics, and should be headed towards something simple; and that final simplicity often enough involves a kind of recovery of the original simplicity.