There could be situations in psychology where a good theory (not perfect, but as good as our theories about how to engineer bridges) would be described by (say) a 70-node causal graph, but that some of the more important variables in the graph anti-correlate with each other. Humans who don’t know how to discover the correct 70-node graph, still manage to pattern-match their way to a two-type typology that actually is better, as a first approximation, than pretending not to have a theory. No one matches any particular clinical-profile stereotype exactly, but the world makes more sense when you have language for theoretical abstractions like “comas” or “depression” or “bipolar disorder”—or “autogynephilia”.[9]
FYI I found these to paragraphs extremely helpful for understanding your basic point, which I don’t think that I had understood before (though, I haven’t read very much of your voluminous material, only little bits here and there).
FYI I found these to paragraphs extremely helpful for understanding your basic point, which I don’t think that I had understood before (though, I haven’t read very much of your voluminous material, only little bits here and there).