There's a vaguely Kantian line of thought which goes something like this: in exercising practical reason -- or asking the question, 'What should I do?' -- one implicitly presupposes that this exercise of rational autonomy is itself important or worthwhile. So, for example, there seems something vaguely inconsistent about deciding to impair your future capacity for rational decision (at least if all else is equal). I'm not sure that's true -- this bootstrapping move is arguably fallacious -- but let's just suppose for now that there's a compelling argument somewhere in this vicinity. My question is: why think this yields any kind of deontological constraint ("never act so as to obstruct another's rational autonomy"), rather than a value to be promoted ("do that act that will maximize the general realization of rational autonomy")?
Consider the possibility of 'Rational Irrationality'. Suppose a madman will lobotomize me unless I take a pill that will just temporarily render me mentally incapacitated. Clearly, I should take the pill. I should temporarily forsake my rational autonomy, for the sake of protecting my capacity for autonomy into the (slightly) more distant future. This seems perfectly consistent with the bootstrapping line of thought. In asking myself, 'What shall I now do?', (let's grant that) I presuppose the value of my own rational autonomy. So I want to preserve my ability to continue to ask this question in future. And the best way to do that is to incapacitate myself (temporarily) in the present.
How would a Kantian disagree with this? How would they put the bootstrapping argument differently? It seems to me that the bootstrapping argument at best establishes the value of a rational will. I don't see how it has any bearing on the question how one ought to respond to this value (i.e. whether to 'respect' or 'promote' it). So I'm probably missing something...?