From my personal experience, inat IDers for the most part are pretty good about not automatically sending observations to research grade and creating complexes where ambiguity arises. While the species concept can be a bit ambiguous at times, I think it is a useful tool as many others have stated above.
Also, I think as genetic tools become more powerful, cheaper, and efficient, we’ll begin to iron out at least some of that ambiguity. Historically, species were defined based on morphological differences, and only relatively recently has genetics begun to come into it in the last 20-30 years. You went from having maybe a few dozen characters to compare to, potentially, millions. As more and more groups are studied, we may find that “Hey, this species is actually a bunch of different species with very different genetics” or “These 2-3 species that looked different actually are almost genetically identical.”
Granted, its still not perfect (we still can’t seen to nail down how many species of giraffe there are, depending on the study you may get 1-6 different species), but its definitely improving things.
Is iNat too focused on species because of the focus on research grade and “numbers”? Yes
Is modern taxonomy and biology too focused on species to resolve cryptic and infinite species diversity? Yes
Are we as humans too focused on species labels as a means to define boundaries in a world where boundaries cannot easily and consistently be drawn, courtesy of an everchanging and constantly evolving world? Yes
Is this all a bad thing? Not necessarily.
I’ve said this before, but I think there’s often two issues that are conflated in this topic. The first is outright incorrect IDs and bad data due to people guessing or not doing research, or using the Computer Vision. The second is people who disagree on when a species name should be used, or not (and thereby kept at “genus” or “complex”. In the first case, this is a definite huge issue for a large database like iNaturalist. But the second case is subjective, and very case-by-case, and in the end I would argue it’s hardly a problem in the great scale of things.
The giraffe issue is the same as the Scrub-Jay issue; somewhat different animals living in different areas and not normally interacting. We can tell them apart, but is that really significant? Should they be species or subspecies or ?
There are no genetic criteria for species-level differences, or any other rank differences, though some people talk like there are. DNA is a tool, like morphology (which is a result of DNA) is a tool. DNA is really great for finding out what relationships are. It can’t tell us what to do with that information.
This is amazing. I love the wandering commentary about listening to frog calls with his future wife’s roommate and the flowchart is amazing. Also this:
We need to stop equating name changes with systematic “progress.”
under the old biological species concept, shouldn’t clonal animals all be considered reproductively isolated at an individual level? So many Aspidocelis to describe!
I’m mostly a herp guy, and it seems like the rat snake and king snake and copperhead splits have been very controversial, whereas the (equally major) milk snake split isn’t nearly as poorly received
I think another driver of “splitting” is that a lot of conservation priorities and money are determined by whether or not a group is distinct or not. There are concepts such as “evolutionarily significant units” and such as well (which don’t require naming a new species to confer protection). However, in my conversations with people I definitely think some people see “splitting” as a good because, if the new taxon is restricted, it can be determined to be threatened and protected more rapidly.
Not much I can add that hasn’t been said above, but I will say that there are “artificial” and “real” elements to the concept of species. As a taxonomists, we strive to make our species more “real” (i.e., testable and designatable empirically). Will we ever achieve this in an absolute sense? Probably not, because nature may be too complex to be defined fully. Can we find solutions that work in almost every case? Based on what I see in nature, I find that likely.
The argument on “what is a species” has been going on forever, but I often think of species as arbitrary labels from an evolutionary perspective, BUT in many cases they can be easily defined and diagnosed in an evolutionary snapshot (as to be “temporarily real” to be usable by scientists).
There are many instances of species being hard to define (lumping versus splitting), but as most of us know on here, there are a LOT of species that are quite distinct so as to be treated as real. I think that if two linages have achieved 100% reproductive isolation that they have become a true species that can never backcross again with a relative or ancestor. So a good example would be like a Mourning Cloak butterfly and a Compton Tortoiseshell butterfly. Same genus but different species and no confusion or interbreeding observed. These two entities can safely be treated as distinct entities in their own right. This is true with a lot of reef fishes where members of the same genus behave as unique biological entities and can be treated as such by scientists. Compare this to Heliconius butterflies or salamander ring species where there is a lot of blurring of where one ends and one begins.
That’s the part about splitting. Every herper accepts the Copperhead itself as a very distinct biological entity separate from any other N. American viper, so that notion of the “copperhead species” seems quite real as there is a large population of viper with copper coloration and no rattle that seems to be something unique and different from anything else. When you start trying to split this distinct clade, however, then you run into the evolutionary reality that the copperheads within this clade are still evolving and will not stop evolving. It’s crazy to think that if the lineage doesn’t go extinct, that many many more very different snake species could evolve with the “same ol’ Copperhead” as their common ancestor. Evolution is cool.
For the plants in my area, iNat doesn’t even use the common names that are in use. I don’t mean regionally-specific names for widespread plants (the plants are only found here), the “common names” listed are ones that nobody actually uses in the real world.
We already see that, like when ecologists send one bee specimen to be identified and then use that name for everything that shows up on a flower.
So, basically isotopes or allotropes. And yes, if you’re dealing with things like radioactivity or tensile strength, if you don’t take those into account, you are doing bad science, with potentially very bad repercussions for industrial labs and chemical manufacturing. Just like if you’re looking at bee visitation on specific flowers, you need to actually identify which species of bees are visiting which plants or else your data will be inaccurate.
I find it very funny that you picked this analogy. Particularly mentioning carbon and iron where isotopes and crystal phases are especially relevant!
well it’s kind of intentional. For sure there are tons of different phases and grey areas amidst elements, but most people still need to use ‘iron’ as a concept.
A split at the genus level is still a split. When an observation of Genus Anas gets bumped back to Anatidae because Spatula got split off, the effect is the same. Couldn’t tell if it was a female green-winged teal or blue-winged teal? Well, now it’s just a generic “Ducks, Geese, and Swans.”
I find it harder to set much store by monophyly now that the orange includes the grapefruit. Not only are they morphologically distinct, but they had independent origins – they evolved at different times in different places. If we can have a known polyphyletic species (orange/grapefruit), none of this cladistic stuff makes much sense anymore. So if I meet someone who likes orange juice but dislikes grapefruit juice, I get to think of them as silly because it’s all the same thing now.
Meanwhile, virtually indistinguishable entities are being defined as different, as if “the six species of lemons can’t be told apart by appearance or taste; it requires microscopic dissection” – and they’re all bumped back to Section Limon. Or maybe Genus Citrus. Family Rutaceae?
It does? That taxonomy isn’t something I follow closely, but I thought Citrus was one of those hot apomictic messes like Crataegus and Amelanchier invalidates key assumptions of the biological species complex. The approach I’ve seen taken in these genera that makes the most sense to me is to name the basal diploid species (where known) and accept that most material from the genus can’t be delineated into clean species but can to some extent be described in terms of which basal species contributed to it, and perhaps some degree of proportion.
This has been the approach in Boechera, except that many of the triploid and polyploid apomicts of hybrid origin form distinctive self-reproducing lineages that also carry species names. They just don’t interbreed among themselves due to their apomictic reproductive mode, which is one example of why the traditional Biological Species Concept needs supplementation to represent the reality that nature presents.
Honestly I’d never have thought that taxonomy could bring out such passion and such polarity in people. I find it both fascinating and, in a way, also disturbing, because one of the very few really clear points emerging is that…
Now this is something I live with all the time. I’m married to a phytosociologist and some of my best friends are plant taxonomists. And never the twain shall agree. Well, never is a big word, but while both are absolutely respectable scientific approaches, it’s impossible to ignore the critical differences in the mindsets of the two specialisations. A now-retired professor in phytosociology once summed it up as: a phytosociologist looks for similarities and features that unite, a taxonomist looks for differences and features that divide. Now clearly that’s a tremendous over-simplification, but without making any judgements, it does contain more than a grain of truth.
To get back to the original question:
As I’ve said before, iNat is conceived precisely to identify observations and it’s pretty hard to see how it could do anything but at least try and focus on species. But as things stand, I don’t think that “getting a nice green Research Grade stamp” is necessarily the only motivation for trying to ID an observation to an agreed species level. It’s clearly gratifying to the observer and also the IDer, but it also has the very practical utility of removing that observation from the “Needs ID” pile. Of course, as @sedgequeen points out, both genus and species complex can be confirmed to Research Grade, but I believe very few of us non-specialists have the confidence, or the presumption, to click that “No, it’s as good as it can be” box, even if, given the photographic evidence presented, we’re 98% sure that even getting the observation to genus is going to be a struggle. On a personal level, I can, of course, just tick the “Reviewed” box and for me that observation disappears. But every IDer who passes through still has to wade through an ever-growing pile of dubiously IDable observations to find those he/she finds worth dedicating time to.
This thread is perhaps not the right place to explore this further, but I share @charlie’s fears that “if iNat is forced down that path it will collapse in the next 5-10 years if not sooner”. Which would really be a great pity for all of us.
I thought phytosociologists studied plant communities, in other words, which plant species tend to grow together. So don’t they rely on the taxonomists to tell them how to identify their constituent plants?
Reminds me of a recent comment by some botanist at a conference - along the lines of “a species should be identifiable in the field”. (~If it takes a million-dollar machine, yuck…)