It’s a debated issue, not a great embarrassment.
This article is a good guide to the topic: https://plato.stanford.edu/entries/species/
This article is a good guide to the topic: https://plato.stanford.edu/entries/species/
Words are just labels we attach to concepts that we find ourselves wanting to refer to. My claim is that there is no concept in allopatry that's worth naming.
Have a look at Hey's underappreciated book (he's the first author cited in the article you linked).
It is an embarrassment: because there's a simple answer (the one I just gave), and because the confusion over "species concepts" has persisted for decades and confused generations of biologists, and hundreds, if not thousands of flawed papers have been published claiming "sufficient divergence" of allopatric populations.
I think it’s more of a deeper problem with language and reality more generally. To me it seems like species is a noun, like other nouns, but reality is ultimately a process. What is needed is a “verb-based” language for describing reality.
A lot of “process philosophers” like Whitehead, Deleuze, etc. have argued similar things.
So yeah I suppose I agree it’s ultimately a taxonomy problem, but in order to fix that I think you essentially need to reinvent language entirely. So it’s not a quick fix for biologists, it’s a systemic issue that I’m not sure really has a solution from within science itself.
I know this sort of doesn’t engage directly with your comment, but process philosophy is a personal interest and I wanted to link to the SEP, heh.
I’ll check out those books though, thanks.
Do you know of comparable "messes" like the one in biology that you suspect were caused by a lack of more process-based language?
Are there some human languages that are more process-based than English?
It makes me wonder whether what the LLMs have captured in their training data might be effective for looking over different languages to see if there are any interesting different ways of handling "species".
I can't seem to find much on the idea that certain languages are more suited to this than others. Theoretically a verb-oriented language would be better.
To me it seems like a giant philosophical project: to create a language that is process-first. It might simply be beyond the capability of humans to actually use such a language in the world, though: which is where AI might be interesting, as it doesn't have the kinds of cognitive limitations on delineations like humans do.