In his book The Significance of Consciousness, Charles Siewert argues that some of our phenomenal features are intentional features, because we are assessable for accuracy in virtue of having these phenomenal features. In this paper, I will, first, show that this argument stands in need of disambiguation, and will emerge as problematic on both available readings. Second, I will use Thomas Szanto’s recent ideas to develop a deeper understanding of the difficulties with Siewert’s argument. Szanto emphatically contrasts the Husserlian, constitutive (...) conception of intentionality with the mainstream, representational conception. If we interpret Siewert’s ideas in representational terms, it will be possible to add to my critical objections. However, I will suggest that it is also possible to interpret, or perhaps to modify, Siewert’s views in Husserlian constitutive terms, thereby addressing the objections raised in the present paper. (shrink)