[jl6]

A definition that is debatable at the edges doesn’t have to be debatable at the center, so I don’t think it follows that we need to completely throw out the concept. Nature doesn’t provide neat categories, but speciation is nonetheless a real natural phenomenon that isn’t perturbed by having fuzzy boundaries.

[Myrmornis]

Yes I completely agree with every word of this. I hope I didn't give the opposite impression. Clearly the sexual organisms living in one place belong to discrete groups, defined by what breeds with what, no objection to using the word species for those groups in sympatry, and the process by which those species come into existence is extremely important and worth studying (I used to do that). As you say, the problem is that the definition doesn't apply in all situations (e.g. populations that don't live in the same place).

That shouldn't be controversial, but unfortunately the idea that you can and should try to "determine" whether such populations are conspecific is completely entrenched in biology; has been since Linnaeus basically.