> They worry that we are lumping together different groups of people based on behavioral evidence alone, when in fact their observable similarities mask important underlying differences, including differences between normal people who are troubled but basically healthy, and their pathological counterparts.

I think the pile of circumstantial evidence for this being the case is pretty large at this point. There is no single treatment that works for a majority of people with major depressive disorder. The types of schizophrenia are sufficiently different that using the same term for them is practically misleading and seems to be driven more by tradition than by any consistent system (cf. diabetes). There's at least one lab test that is believed to identify a type of depression; it seemingly fell out of favor because nobody figured out how to choose a treatment based on it and a negative result doesn't exclude an MDD diagnosis [1].

[1] https://en.wikipedia.org/wiki/Dexamethasone_suppression_test