The idea that transcendental philosophy would have to be done from a special standpoint is implicit in Richard Rorty, Philosophy and the Mirror of Nature (Princeton University Press, 1979), p. 293, where Rorty writes of the "demand . . . for some transcendental standpoint outside our present set of representations from which we can inspect the relations between those representations and their object." Kant distinguishes "transcendental" and "transcendent" (see for example, A 296/B 352-53). In Rorty's phrase, 'transcendental' could be replaced by 'transcendent'. Not that that shows Rorty to be misusing 'transcendental'; he is suggesting that transcendental philosophy requires a transcendent standpoint. That is what I think we should dispute. When I wrote despairingly about Kant's "transcendental story" in Mind and World (pp. 41-43, 95-98), I was acquiescing, in a way I now regret, in a reading of parts of Kant's transcendental activity that fits Rorty's phrasing. (I would still disparage the philosophy such a reading finds in Kant.).
McDowell, Having the world in view, p. 446, n. 23.
One can agree with McDowell's claim that transcendental philosophy doesn’t necessarily require a transcendent standpoint and can still argue that in its Kantian version it does require a transcendent viewpoint. I think this is Habermas' position. Habermas terms the transcendental philosophy that doesn't require a transcendent viewpoint a weak transcendentalism. The crucial test here is whether the subject is conceived beyond change or amenable to change and hence within the ambit of history. If the subject is in principle within the ambit of history then transcendentalism doesn't imply a transcendent, if it is not, then it would imply a transcendent standpoint.
Habermas criticises the Kantian concept of “spontaneity of a subjectivity that is world-constituting yet itself without a world (weltlos).” The ranscendental subjectivity in the tradition of modern philosophy from Descartes to Kant and up to the phenomenological tradition places subject beyond the field of “constitution” in the sense that although subject “constitutes”, it is itself beyond any constitution [That what constitutes is pure spontaneity (beyond constitution) without being delimited by any limitations while it limits everything else (constitutes)]. Thus Habermas clearly recognises that Kantian transcendental idealism is “not compatible with the realist presupposition of a mind-independent world that places constraint not only on judgments of experience but also on the learning process of the judging subjects.” (TJ: 19, emphasis added). Despite various similarities in their position there is this crucial difference between Kant and Habermas: For Kant it is inconceivable that mind independent reality would constrain the learning process of “the judging subject.” The transcendental subject is outside the space and time where its activities can be constrained. Constraints only apply to an empirical subject. For Habermas on the other hand subjects capable of speech and action and of accomplishments and initiations are formed in interaction with mind independent reality and under its constraints. These constraints are for Habermas not mere hindrances but the conditions of the possibility of the formation of capable subjectivity.