Well we need to be able to identify organisms to do inventory and assessment and there isn’t a categorization based on only ecological function that exists. The ‘species’ concept mostly works when it’s used the way it used to be - discrete observable differences that correlate mostly with reproductive isolation. But strict genetic monophylly and separation of every possibly reproductively distinct entity into its own ‘species’ doesn’t work. At all. So we either need two different defintiions of species (in which case certainly the ecological one, not the taxonomic, would be most usable for iNat) or we need some way to find middle ground, which we aren’t now. I think some taxonomists want to use iNat to document and change proposed taxonomic entities when the intention of the site is to document individual organisms. I don’t think doing both at once works.
because many of the proposed entities can’t be classified in the field, or are changed constantly (change especially splits can make old data totally unusable so we just end up not changing it at all). The truth is species intergarde much more than most people believe and species are more of a cloud of similar intenties than something you can easily draw boundaries around. So we do the best we can, but we need something that works for community science and ecology.
Okay. work with me to lobby the iNat admis to let us as part of the iNat ID system then. Or else don’t take away the groups we have (or had). That’s my request to taxonomists, basically. We currently aren’t allowed to use ecological functional entities on iNat, nor have we even created such groups in most cases.
But it still is that, isn’t it? My professor (an evolutionary biologist) has told me that every species has a distinct morphology, even though that may only differ slightly from another species, but that are, nonetheless, present.
And so far it seemed to be that way even for hyper-split taxa (I have a key for all the different Dandelion species which is entirely based on morphological differences).
Is there currently any species (excluding bacteria, archaea, and many eukaryotic single-celled organisms) that is defined exclusively by genetic differences without any morphological ones?
Note that even if a species is morphologically distinguishable, it might not be distinguishable all the time, which causes problems if you’re identifying in the field. For example, North American jack-in-the-pulpits (Arisaema sp.) have recently been split up into several species mostly based on flower morphology (see this paper for an overview). For the record, I do think the changes are justified, but it now means that Arisaema plants are sometimes unable to be identified to species without seeing the flowers (you used to be able to identify Arisaema by seeing the leaves alone).
Very true. But that isn’t necessarily a new problem. Afaik some insects are only really morphologically distinguishable in their adult stage.
Or a related problem: some microscopic fungi actually have 2 different names. “Many true fungi, particularly in the Ascomycota and Basidiomycota, produce sexual and asexual states at different points in the life cycle. As a result, these fungi are associated with two names: one that refers to the anamorph at the time it was described and the other that refers to the teleomorph at the time that it was described.” (Gould, 2009)
Do you mean allowing more infrageneric groupings in general, or allowing non-monophyletic infrageneric groupings?
Laypeople have a particular understanding of “species” which is not exactly accurate (the way my old philosophy professor used “genus”, “species” was very confusing to me as an ecology nerd). Modern taxonomy is probably moving even further away from that broad understanding. But because of that layperson understanding combined with species being the “standard unit” of taxonomy, the iNat devs have understandably designed iNat to favour species identification.
I feel like the conflict here is a combination of tension between the common understanding of “species” and the modern taxonomic understanding, as well as a subtly separate tension between that modern taxonomic view and the species-biased design of iNat. To me it seems like the easiest link to change here is iNat. If modern taxonomy is moving away from species being a practical unit, and iNat is still forced to incorporate formal taxonomy (to which I don’t see any alternative), then the easiest resolution is to make iNat more flexible about using higher taxon units and “translating” them to laypeople.
Which brings me back to your quote at the top–are there structural issues preventing any particular species groups or complexes from being added currently, or is it just that curator attention hasn’t gone towards them yet?
There’s this feature request… (I feel like I saw a more recent thread about this but it didn’t come up in search, maybe this one) Do observation fields or projects as suggested there not work for your workflow? Like I don’t think mangroves or four-leaved clovers make sense as taxa on iNat but I still like being able to collect and filter them, so I made projects for them.
Okay this is an example where a non-monophyletic complex would be nice, but it’s not possible.
This genus split is a result of this paper (and others?) which argues pretty strongly that the genus Accipiter (goshawks and sparrowhawks) was paraphyletic with respect to the genus Circus (harriers). Even their lice aren’t each others’ closest relatives!
The problem is that Cooper’s Hawk (now Astur with goshawks) and Sharp-shinned Hawk (still Accipiter) are similar in size and shape and from a poor view are often difficult to separate. Before this any difficult observations would’ve been put to genus, but now they’ll need to go all the up to subfamily, at a level shared with 11 genera and like 73 different species. That’s a very imprecise identification when that there are only actually 2 species options being referred to by it.
However the genetics seems pretty clear, and all of the standard bird taxonomies are following the change that I know of. iNat would be striking out on its own to reject the change now.
Currently it’s logistically impossible in the iNat taxonomy system for there to be a complex that includes both Cooper’s and Sharp-shinned because they can’t be daughter taxa of both their respective genera, and also daughter taxa of a complex taxon including both of them. I’m not sure if anyone would want to create a project or observation field for all these observations either since there must be thousands of them to go through… On the other hand I guess this filter sort of works.
tell that to the taxonomists currently upending plant taxonomy, and to the ‘POWO’ site they partially base it on.
Definitely the former, it’s a slam dunk. I don’t agree with making a hard line against the latter but it makes taxonomists REALLY upset.
I don’t agree because i think the taxonomic view of species has split from every other use of species including othe rscientific uses, so i feel like iNat is making a big mistake trying to follow it. Which i guessis the root of my whole taxonomy thing here, or a large part of it anyway.
I use them to some extent but it takes a LOT longer and so it just means i don’t reliably do it, or else I use iNat a lot less. I do use projects for some use and have started using tags to put in the scientific name i want to use when taxonomists try to change it to something else, but that doesn’t work for mapping or filtering with other specise, etc. I’ve also stopped doing IDs for other users except people i already know, because of the same.
I mean, it’s possible, because taxonomy is a human construct and humans choose what it is. It’s politically unfeasible on iNat, where one faction of thought currently controls how taxa are used.
it’s difficult to untangle, but with relational databases like the ones iNat presumably uses, it isn’t logisically impossible at all. It’s a choice. It doesn’t make sense to look at it other ways. You can agree with the choice, but you can’t present it like gravity or something. You fully trust genetics and think it should be the ironclad way to designate species. That’s a choice. Maybe in the end it will end up the right choice or maybe it won’t. Probably people will never fully agree, because humans usually don’t
Okay, we have that already so we’re mostly good there.
I mean that with the way the taxonomy system is programmed currently, it’s literally impossible for one species to be a daughter taxon of multiple parent taxa. It has to be a branching tree, there can’t be one leaf growing out of two different twigs.
That does cause issues in braided stream situations (like I think Eastern Tiger Salamander is part of both the tiger salamander complex and the unisexual salamander complex) but those situations are apparently rare enough to not be worth completely redesigning the whole system (which could take a huge amount of thought and effort) when there are many higher priority development goals. Kind of like sharing observations where that feature request won’t be implemented for a long time (if ever) because it apparently would require a rethinking of the fundamental philosophy of what an “observation” is.
we don’t though, people won’t make them for new species unless someone has included them in a paper, which is rare because the taxonomists writing these papers don’t prioritize maintaining data integrity of use of the system at all.
Yeah, that seems likely to be true.
Well i think that is part of the issue here. Genetics is showing it isn’t rare at all. But taxonomists insist on a braidless tree. In a braided rover system strict monophyly is sometimes just not possible.
Well yeah, that is true, as in my mind most improvement of the site pretty much stalled when it started growing too fast, they seem to be able to get funding/staff to improve the algorithm but that’s it. Otherwise the site is slowly depreciating (not blaming the iNat staff, they probably just can’t keep their head above the water keeping things running). Really obvious needed changes don’t happen, and the needed new app took many years and still doesn’t have most of the features it needs to b ea replacement yet. This is what happens when you prioritize growth over all else, but unfortunately that’s the way of the society iNat was formed in, and will have far greater consequences to our species than the issues on this site as I’m sure you well know. But it would have been nice if iNat had been able to exist without prioritizing exponential growth beyond what the admins could ever be asked to deal with.
And to bring it all back around, given that all, it’s frustrating to see dozens of curators working dilligently to upend taxonomy instead of helping with this important endeavor in other ways that would actually help us keep the community’s head above the water. Taxonomic revisionism is just one finger pushing the head down but when we are gasping to keep things going, that finger matters.
Do most large genera not already have an infrageneric taxonomy structure that new species will automatically be sorted into? Subgenera, sections, series etc.
some of them have coarse ones, but they aren’t consistently maintained and are prone to taxonomists deciding to upend those too. I think we need an automatic intragenetic entity for all splits, for instance, to maintain data integrity and allow a way to keep the older concept in the system. But the taxonomists get very angry about this because they might not be strictly monophylletic groups. Compromise isn’t an option at all, no matter how much difference it would make. it’s just got to be the genetic model uber alles and if you ask for something else you get harassed.
It seems to me that a big part of the upheaval is in where on the cladogram we draw that line. If we split a species into a clade of several related species, haven’t we merely moved the species boundary further out, that is, from before the branching to after? This is why I find the mushrooms so frustrating. One “field guide species” (for lack of a better term) is split into several species based on gene sequencing; but if those newly split species are all each other’s closest relatives, what is accomplished? The “field guide species” is equivalent to the monophyletic group.
Yeah this is the question, if strict monophyly is the rule. Theoretically I default to preferring lumps, but then when I look at specific examples then lumping doesn’t look appealing at all either…
For example Elegant Tern and Sandwich Tern are clearly different species. Having Elegant as a western-orange-billed subspecies of Sandwich would be super weird.
Harriers are clearly not goshawks or sparrowhawks and are their own group that should have their own genus.
I think my position with dandelions is maybe analogous to my thoughts on the whole cladistics thing in general… Theoretically I like phenotypic categorization and lumping (relative to the current state of things), but in practice I feel like the current philosophy has to work itself out before the pendulum can swing back to something more practical. “The only way out is through.”
Like with dandelions the current North American model from Flora of North America is inadequate because it fails to account for the real diversity of the genus outside of the continent, some of which have shown up in North America and there aren’t good names to apply to them. On the other hand the European model is nice and thorough but excessively tedious (to put it lightly) for practical use.
What we really need is a balanced model somewhere in between, but who in the world is going to create that model? Only someone who’s had to confront both other options and is frustrated with both of them (and also has a crazy amount of time and funding to work on it). I also don’t feel like iNaturalist is a suitable place to experiment with alternative taxonomy models like that.
this sounds like a strong argument for holding bins and complexes, so we don’t ruin a bunch of data for a transient philosophical debate.
it seems like literally the best place possible because it’s not beholden to any academic institution (for now), it’s more open source and ‘democratic’ than other places, and it can be more flexible than government or academic databases. Like you’ll literally never find a better place to do this important work. It’s here or nowhere at all. But if you don’t want to spend time on it, and prefer to do other things instead, that’s totally valid. I think the holding bins and complexes are the best start if we are to find a compromise, though i favor just rejecting the splitting and revisionism entirely.
I don’t think experimental taxonomizing should be done anywhere where it affects actual specimen records, whether museum specimens or iNat obs
You could do this kind of experimentation in a blog or something, or in an academic setting, publish your plans and get feedback on them, etc. But it should be somewhat settled before applying it to a database where it will affect real records. Until then the changes should only be hypothetical.
can you explain to me how one could do work on revising taxonomic classification systems with a blog? I’m sorry but that just doesn’t seem like a good faith suggestion. And i already explained why an academic setting isn’t always accessible. A community effort to do so with a site like iNaturalist is the only real option. And there are ways it could be done without disrupting what others are trying to do. Let the revisionists ‘use fields and tags’ like they always suggest for others. Or maybe they could make up fan fiction style blogs for their microspecies they found in their yard? Cos when i hear ‘blog’ in 2024 i don’t think a good faith suggestion at all.
Can we have this standard for taxonomic changes on the site too? like a hard requirement that a paper is peer reviewed and maybe has a year or something to account for any further issues before we ‘affect real records’? Because that would be a huge step forward from what we have now which is… nowhere close to that.
in theory, experimental taxonomizing could be done with tagging and it wouldn’t affect any actual specimen records.
for example, in japan they put ficus erecta pollen into a few different varieties of ficus carica. there was a big disparity in compatibility… the seedlings of one carica variety all died shortly after germination, while the seedlings of another carica variety grew relatively vigorously. this shouldn’t be surprising since carica has been widely cultivated for many centuries from china to africa. naturally we would expect to see quite a bit of diversity within such a wide ranging species.
taxonomists typically aren’t horticulturalists or breeders so we can’t reasonably expect them to take cross compatibility into account when drawing subjective lines between species. but in my estimation big differences in cross-compatible, which are inherently genetically based, are not a minor detail.
there’s no reason that carica cultivars can’t be split into new species based on their cross-compatibility with other ficus species. these new species wouldn’t have to be official or formal, they could simply be tags that anyone could apply to any carica cultivar.
flickr allows you to add multiple tags to other people’s photos. can’t imagine such a system being subject to much abuse.
Oh no I’m serious. I agree there are a lot of issues with academic settings; as a result a lot of the cool research these days is happening on anonymous blogs. This guy’s blog is mostly about them. This includes personal iNat journals too; a lot of novel research and exploration is being done there. Like we can write out taxonomy hypotheses and discuss them on iNat; I just mean we shouldn’t be implementing them on the whole website.
The way I’m imagining you’d do it in a blog is basically writing out academic proposals but in a more informal way and doing distributed peer review through other random interested people commenting on the blog, rather than the normal formal academic process. The same way we discuss and give feedback about how iNat functions here on the forum.
Maybe we need a modification of the last sentence there? E.g. plants are supposed to follow POWO, but POWO doesn’t cover infrageneric taxonomy so curators end up following primary literature anyway to curate those aspects.
