Can anyone think of examples from science or any domain of inquiry where a conceptual distinction turns out to be unprincipled/spurious/epiphenomenal/etc. but is nonetheless still useful practically or methodologically? I.e., something for which we might say, “We know that the distinction between X and Y is spurious, but despite that, looking at things believed to be category X rather than category Y tends to be fruitful for learning about or investigating blah blah blah.”
The closest example I can think of is the idea that racial categories are biologically unprincipled but despite that are still important sociologically. But this is an imperfect example since it shifts the domain of inquiry between biology and sociology and I’d like a case where the baselessness and usefulness are internal to one domain.