In his classic paper “Two Concepts of Consciousness” David Rosenthal says “…if consciousness is essential to mentality no informative, nontrivial explanation of consciousness is possible” (2005 p 22). The claim is, roughly, that if all mental states are conscious and if there are no mental states that can occur unconsciously, then we cannot explain what makes a mental state conscious. But this assumes a certain way of characterizing the claim that all mental states are consciousness.
In a really excellent paper that just came out in Philosophers’ Imprint by Allison Simmons, Cartesian Consciousness Reconsidered, she argues that Descartes’ can understand consciousness as a kind of self-representation and can account for ‘unconscious’ mental states by appealing to gradations of consciousness and what she calls ‘phenomenal confusion’. I think she makes an excellent case that a Cartesian picture of the mind can go a long way towards providing an explanation of conscious and (seeming) unconscious states.
Yet even so, I think we can see that the initial claim made by Rosenthal is still true in a sense. The Cartesian moves made by Simmons all rely on a prior account of mentality. That is, the Cartesian explanation given relies on a prior understanding of sensations (as having objects as their ‘objective reality’, which we would translate as ‘having representational content’) and explains consciousness in terms of a thought having as it objective reality a first-order state (i.e. consciousness is the representation of our first-order states). In fact Simmons seems to make much the same point in her paper. If we interpret Descartes as claiming that thoughts and consciousness are identical then we will not be able to explain it. But, as she shows, this is not something that a Cartesian is committed to. So even if one is attracted to the view that all mental states are conscious one doesn’t need to give up on an explanation of what consciousness is.