Classification and diagnosis and Part II: Pathogenesis and aetiology) begin from the fact that "few, if any, breakthroughs in basic scientific research have led to substantive improvements in psychiatric clinical practice". They are right to conclude from this situation that there is a need for refocusing of research agendas in psychiatry.
However, they then start from the wrong premise. They still want to "endow psychiatry with a mechanistic, neuroscientifically informed basis”. They, therefore, produce a list of 17 problems for psychiatry, created by asking an international group of scientists and clinicians to state what they perceive as “the single most important problem or hypothesis" that needs to be addressed to meet this objective.
How long will it take for psychiatry to realise that its very nature is that it has "major conceptual and practical challenges"? It's no good expecting research to bypass this situation.
Psychiatric diagnoses are simply categories justified by clinical utility. Their value-laden nature are not a sign of scientific deficiency but of their meaningful nature. Mental disorders are not natural kinds and there are inevitably fuzzy boundaries between different syndromes. Psychiatry needs to avoid the reification of diagnostic concepts.
As far as aetiology is concerned, genes set the boundaries of the possible but environments define the actual nature of mental disorder. The human brain is socially constructed in the literal sense that brain cytoarchitecture itself is fashioned by input from the social environment. Minds are enabled but not reducible to brains.
As I keep saying, please do not misunderstand me. Of course mental states map onto the brain. But the localisation of function to structure in the brain is not a new problem. Psychiatry can still be practiced even though there is a mind-brain problem. There's no need to create 17 pseudo-problems to solve by research before progress can be made.