36 While this way of balancing the legitimate claims of scientific naturalism and phenomenology may be attractive, is there any direct evidence that Nietzsche is sympathetic to it? Well, he evidently gives a central place, not only in his own practice, but also in his second-order reflections on it, to phenomenological psychology involving reference to ‘valuations’: ‘our opinions, valuations, and tables of the good certainly belong among the most powerful levers in the mechanism of our actions, but in each individual case the law of their mechanism is indemonstrable’ (GS 335). The most natural way of reading his point here is that in standard cases, personal-level phenomenological explanations are indispensable for satisfying our cognitive interests, although they are not sufficient explanations, since the personal level where we can properly speak of ‘actions’ is necessarily subtended by a subpersonal ‘mechanism’ the details of which are largely unknown. Nietzsche also suggests that, in certain non-standard cases, which according to him are more frequent than we are inclined to think, personal-level explanations are inappropriate and should ideally be replaced by what he calls ‘physiological’ explanations, presumably analogous to clinical practice in diagnosing and treating certain cases of depression. (See GM III 17; TI VI 6).