Zachar & Kendler do recognise that there are differences between the controversies. As they point out, the Pluto decision involved a new definition of 'planet'. They suggest in the case of homosexuality that the "compromise was also a definition, namely that a psychiatric disorder involves distress, social occupational dysfunction, or both". I'm not convinced, though, that the definition of mental disorder was changed as such. The argument about the definition of mental disorder as involving distress, social occupational dysfunction, or both was used, but I don't think there was a fundamental redefinition of 'mental disorder'. It was more about whether homosexuality should be seen as a mental disorder. Homosexuality, in fact, should never have been regarded as a morbid condition. As Zachar and Kendler note,
What homosexuality is like psychologically in an environment where it is considered a disorder will meaningfully differ from what it is like in an environment where it is considered a normal variation in sexual orientation.Zachar & Kendler do say that a "compelling argument can be made that Pluto never should have been classified as a planet", but I can't see that it could have been foreseen that the definition of 'planet' would change.
There is an argument that the definition of mental disorder was wrongly changed in DSM-IV, in that the functional/organic distinction was abolished (see previous post). Robert Spitzer (2001), who chaired the task force for DSM-III, that removed homosexuality from the manual, always maintained that DSM-III took a neutral approach to causation and that it is not covertly committed to a biological approach to explaining mental illness. There is truth in this observation. DSM-III's primary motivation was to introduce operational criteria to replace what it considered to be vague and imprecise definitions of mental disorder. True, it was associated with reaffirmation of an implicit “medical model” with a focus on brain mechanisms and a positivistic approach to science, but the link between DSM-III and biological aetiology is merely associative, not necessarily logically causal. A classificatory system in itself is not necessarily biomedical. This is illustrated by the fact that DSM-I was influenced by the reaction types proposed by Adolf Meyer, despite Meyer's concern about the general over-emphasis on psychiatric diagnosis (see my book chapter). In fact, Meyer had "no use for the essentially ‘one person, one disease’ view” and regarded the “…statistics published…[as] a dead loss…and an annual ceremony misdirecting the interests of staff”. DSM-5 did set out to change the definition of mental disorder but failed (see eg. previous post).
Of course, as Zachar & Kendler say, competition for “scientific authority” among competing groups is a normal part of scientific progress. Psychiatry has been challenged as unscientific for putting the decision about homosexuality out to a vote. Zachar & Kendler make clear that they are not reifying scientific authority but, by making the comparison with the Pluto case, do seem to be wanting to defend the scientific nature of psychiatry.
The real problem, though, is the reification of psychiatric disorder. Apart from organic disorders, mental disorders are not concrete entities, like Pluto. There will inevitably be limitations in the application of psychiatric diagnosis, whatever way symptoms and signs are grouped and conceptualised. Psychiatric practice needs to acknowledge this state of uncertainty. The concepts of psychiatric disorder do not need to be abandoned for this reason. They merely need to be recognised for what they are, as attempts to describe prototypical or ideal types of psychological states. They are merely idealised descriptions of those aspects of reality that interest us. As unobservable hypothetical constructs they are not capable of description in a natural scientific sense (see eg. my article).
Psychiatric diagnoses are simply categories justified by clinical utility. They are working concepts for clinicians. The value-laden nature of psychiatric diagnosis is not a sign of scientific deficiency but of its meaningful nature. We must stop too easily assuming a functional mental disorder is an entity of some kind and thereby using such an association as a justification for treatment. To quote from Adolf Meyer (see my paper):
Very often the supposed disease back of it all is a myth and merely a self-protective term for an insufficient knowledge of the conditions of reaction