The article discusses two claims from Seiberth's book Intentionality in Sellars: A Transcendental Account of Finite Knowledge, both of which bear on the question of what it takes to be in touch with the world. Seiberth claims, first, that the philosophical method known as transcendental analysis, which Sellars adopts from Kant, is more basic than Sellars's other methodological commitments, including the method of providing a conceptual analysis of the manifest and the scientific image of man-in-the-world. I ask whether the results (...) of transcendental analysis should be applied to the manifest image. Does Sellars think that the manifest image fulfills the transcendental conditions for any conceptual system that is a) about the world of which it is a part and that b) allows those using it to gain knowledge of that world? On Seiberth's reading of Sellars, the answer seems to be negative. Second, Seiberth claims that there is only one kind of vertical relation between the conceptual and the real, the non-semantic relation Sellars calls picturing. I contest that claim, arguing that Seiberth gives too strong an interpretation of Sellars's denial that meaning statements are relational statements connecting a word with an independently existing object. This precludes Seiberth from seeing that Sellars, just like Kant, is an empirical realist in a robust sense. I also argue that picturing provides a meagre substitute for what in my reading of Sellars we might call vertical 'semantical' relations between words and things after all. (shrink)