abstract Ryan Tonkens proposes that my Kantian approach to suicide intervention with respect to the mentally ill (2002) wrongly assumes that the suicidally mentally ill are rational and are therefore rational agents to whom Kantian moral constraints ought to apply. Here I indicate how the empirical evidence concerning the suicidally mentally ill does not support Tonkens' criticism that the suicidally mentally ill are irrational. In particular, that evidence does not support the conclusion that such individuals are systemically practically irrational so (...) as to undermine the attribution of at least minimal rational autonomy to them. A Kantian moral framework, albeit one developed in a non-ideal direction, remains applicable to such individuals. (shrink)