Daniel sent us this one — he wants to talk about biological taxonomy, the actual profession of it. Not just the greatest hits of Linnaeus and binomial names, but the working reality. Where do taxonomists actually sit? How does global coordination work across the different naming codes? What's a type specimen and why does it still matter in the age of DNA? And then the tension at the heart of it — we're losing species faster than we can name them, but universities keep cutting the very positions we need to do the naming. There's a lot to unpack here.
This is one of those fields where the public understanding is about two hundred years out of date. Most people think taxonomy is Victorian gentlemen with butterfly nets pinning specimens into mahogany drawers. And that still exists, but the field has been quietly revolutionized about four times since then.
Quietly revolutionized — that's the taxonomy brand, isn't it? Two centuries of methodological earthquakes, and the public image is still a watercolor of a finch.
You can't do conservation biology if you don't know what you're conserving. You can't track biodiversity loss without a baseline of what's there. You can't do drug discovery from natural products if you can't reliably identify the organism producing the compound. Taxonomy is the index of life on Earth, and the index is massively incomplete.
Let's start with the index itself. The prompt asks about the history, and I think we should cover that because it sets up why the modern system is the way it is. But I want to get to the working taxonomist's desk fairly quickly — the person who actually has to decide whether this beetle is a new species or just a slightly different shade of the same one.
So the quick arc. Aristotle gets credited with the first serious attempt at classification — he grouped animals by shared characteristics, things like "animals with blood" versus "animals without blood," which roughly maps to vertebrates and invertebrates. He had about five hundred species in his system. It was observational, it was logical for its time, and it held sway for nearly two thousand years.
Which says less about Aristotle's genius and more about how little anyone advanced the field for two millennia. The man classified dolphins as fish and nobody seriously challenged it until the Renaissance.
In fairness, a dolphin looks like a fish if you're working with the naked eye and no dissection protocols. But yes, the next real leap doesn't come until the sixteenth and seventeenth centuries, when exploration starts flooding European collections with specimens that don't fit the old categories. By the time Linnaeus arrives in the mid-seventeen hundreds, the classification systems are a mess — different authors using different names for the same thing, descriptions that run to paragraphs, no standardization at all.
Linnaeus's move was essentially the Dewey Decimal System for life.
That's actually a great way to put it. His big innovation wasn't the idea of grouping organisms — people were already doing that. It was the binomial nomenclature. Genus and species. Radically compress the identifier into something human-usable, and then organize those identifiers into a nested hierarchy — kingdom, class, order, genus, species. He published Systema Naturae in seventeen thirty-five, and by the tenth edition in seventeen fifty-eight, it had become the reference standard.
The tenth edition is the one that still matters legally, right? That's the starting point for zoological nomenclature?
The tenth edition of Systema Naturae, published in seventeen fifty-eight, is the official starting point for zoological nomenclature under the ICZN code. Anything named before that doesn't count. For plants, the starting point is Linnaeus's Species Plantarum from seventeen fifty-three. For fungi, it's also Linnaeus — but the modern fungal code has additional start dates for specific groups. It's a patchwork, which tells you something about how these systems evolve.
It tells you that taxonomy is law as much as science. Which I don't think most people appreciate. Naming a species isn't just describing something — it's a legal act with consequences for priority, for conservation status, for who gets credit.
That's where the type specimen comes in. Let's dig into that, because it's one of those concepts that sounds bureaucratic but is actually philosophically fascinating. A type specimen is a single physical specimen designated as the reference point for a species name. It's the anchor. If there's ever a dispute about what "Aus bus" actually refers to, you go back to the type. It sits in a museum collection somewhere, and it is the name.
The name isn't attached to a description, it's attached to a physical object.
The description can be revised, the species boundaries can be reinterpreted, but the type specimen is the fixed point. If someone later decides that what you called Aus bus is actually two different species, they designate one of them to carry the original name — that's called the lectotype — and the other gets a new name. The type specimen is the physical anchor that prevents names from floating off into ambiguity.
Which means that every species name ever given is tethered to a specific dead organism in a specific drawer somewhere. That's an enormous amount of physical infrastructure.
It's fragile infrastructure. Museums have type specimens that are centuries old. Some have been lost to wars, fires, floods. The Berlin herbarium lost type specimens during World War Two. The National Museum of Brazil fire in twenty eighteen — that was an absolute catastrophe. Entire holotypes gone. When the type specimen is destroyed, you can designate a neotype — a new specimen to serve as the reference — but it's a legal process, and it requires justification.
The holotype being the single specimen the author designated as the type.
Holotype is the one true reference specimen. There are also paratypes — additional specimens the author examined when describing the species. Syntypes when the author didn't designate a single holotype. Lectotypes chosen later from among the syntypes. Neotypes when everything's lost. It's a whole vocabulary of specimen curation, and it matters because these physical objects are the ultimate backstop for biological nomenclature.
Let's talk about the codes. The prompt mentions ICZN, ICN, ICTV, ICNP. That's four different rulebooks for naming different kinds of life.
It's actually a beautiful example of how decentralized global coordination can work. There's no world government of taxonomy, no UN body that sets the rules. Instead, each major group of organisms has its own community-governed code. The International Code of Zoological Nomenclature — ICZN — covers animals. The International Code of Nomenclature for algae, fungi, and plants — ICN, formerly the ICBN — covers those groups. The International Code of Nomenclature of Prokaryotes — ICNP — covers bacteria and archaea. And the International Committee on Taxonomy of Viruses — ICTV — handles viruses, though viruses are weird because there's debate about whether they're even alive.
Four codes, four communities, four sets of rules that don't always align. And these are maintained by international commissions that meet, debate, vote on amendments. It's governance without government.
It largely works. The codes agree on the big principles — priority, typification, publication requirements — but they differ in details. Under the zoological code, you can't name a species after yourself. Under the botanical code, you technically can, though it's frowned upon. The zoological code requires names to be published in a "permanent and widely accessible" medium — electronic publication is now allowed under certain conditions. The botanical code requires a Latin diagnosis for new taxa until twenty twelve, when they finally allowed English.
Until twenty twelve? So if you discovered a new orchid in twenty ten, you had to write the description in Latin?
You had to include a Latin diagnosis. You could write the full description in English, but you needed at least a short diagnostic section in Latin. This was a genuine barrier — you had taxonomists who were brilliant field biologists struggling to compose grammatically correct Latin descriptions. Dropping that requirement was a practical move that some traditionalists still grumble about.
The Latin requirement is one of those things that makes taxonomy seem like a guild from the seventeen hundreds. And in some ways it still is. The apprenticeship model, the emphasis on mentorship, the fact that you can't really learn to identify species from a book — you need to sit with someone who knows the group and learn to see what they see.
That's where we get to the working reality. Let's talk about where taxonomists actually sit today. The major centers are still the great natural history museums. The Smithsonian National Museum of Natural History in Washington has one of the largest collections — about a hundred and forty-eight million specimens. The Natural History Museum in London has about eighty million. The Muséum National d'Histoire Naturelle in Paris has around sixty-eight million. These aren't just display collections — they're active research infrastructure.
A hundred and forty-eight million specimens. That's a library where every book is unique and most of them haven't been read.
A significant fraction of those specimens are still undescribed. A researcher will collect something in the field, deposit it in a museum, and it might sit in a cabinet for decades before someone with the right expertise examines it and realizes it's new to science. The average lag time between collection of a specimen and its formal description as a new species is something like twenty-one years. That's the taxonomic impediment in one number.
Twenty-one years. So a specimen collected today, in the middle of what many are calling the sixth mass extinction, might not be formally named until the species is already gone.
That's not a hypothetical. There are documented cases of species being described from museum specimens where the original habitat has since been destroyed. The conservation status at the moment of description is already "possibly extinct." It's heartbreaking.
Let's walk through the process. Someone finds what they think is a new species. What actually happens?
Step one is comparison against existing type specimens and the published literature. You can't declare something new until you've eliminated the possibility that it's already been described. This means visiting museum collections, or borrowing specimens through inter-museum loans, or increasingly, examining high-resolution digital images. Many herbaria now have digitization programs — the New York Botanical Garden has digitized millions of herbarium sheets. But it's slow work, and many important collections remain offline.
You're cross-referencing against potentially hundreds of described species, many of them known only from nineteenth-century descriptions and specimens that may have faded, been damaged, or been mislabeled.
The descriptions themselves can be maddeningly vague. A nineteenth-century entomologist might have described a beetle as "brownish, with antennal segments proportionately longer than in related forms." That's not diagnostic. So you're sometimes comparing a fresh specimen against a type that's missing its antennae and a description that's essentially useless.
The ghost of a Victorian entomologist is still causing problems.
Every taxonomist has stories about this. Once you've done the comparative work and you're confident you have something new, you prepare a formal description. This includes a diagnosis — the characters that distinguish your species from close relatives — a detailed description of the holotype, information about the collection locality, etymology of the name you're proposing, and usually a discussion of relationships. You submit it to a peer-reviewed journal. The reviewers check your work, examine your justification. If it's accepted, the name is published and becomes available under the relevant code.
The name itself has rules. It can't be offensive, it can't be unpronounceable, it has to be formed according to Latin grammar even if the words aren't Latin. You can name a species after a person, a place, a characteristic. You can name it after a celebrity — there are species named after David Bowie, after Darth Vader, after Beyoncé.
The Beyoncé horse fly — Scaptia beyonceae — was named because of its distinctive golden rear end. The taxonomist had a sense of humor. But naming after celebrities is a tiny fraction of new species. Most names are descriptive or geographic. And there's a genuine art to a good species epithet — something that's memorable, informative, and not already taken. Because priority matters, and if someone already used that name in your genus, you can't use it again.
We've got this process that's meticulous, slow, requires deep expertise, and is anchored in physical specimens. And we're doing it in a world where, by most estimates, we've only described a fraction of the species on Earth.
The numbers are staggering. Currently described species are somewhere around two million. Estimates for total eukaryotic species on Earth range from about five million to over a hundred million, with a recent best estimate around eight point seven million. That means we've described maybe twenty to twenty-five percent of the species on the planet. For some groups, it's much worse. Nematodes, fungi, deep-sea organisms, tropical insects — we've barely scratched the surface.
Those are just the eukaryotes. Bacteria and archaea are a whole different universe. Recent metagenomic surveys suggest there may be hundreds of thousands or even millions of prokaryotic species, and we've formally described about fifteen thousand.
Under the ICNP code, you can't validly name a prokaryote without a pure culture deposited in two culture collections in different countries. That requirement alone means the vast majority of bacterial diversity — everything we know only from environmental DNA sequences — can't be formally named under the current rules. There's an ongoing debate about how to handle "Candidatus" taxa — provisional names for organisms known only from sequence data. It's a code that was written for a pre-genomic era struggling to adapt.
Which brings us to DNA barcoding and the BOLD database. The prompt mentions these, and they're central to how taxonomy is trying to accelerate itself.
DNA barcoding is elegantly simple in concept. You pick a short, standardized gene region that varies between species but is conserved enough to amplify across a broad group. For animals, it's usually a segment of the mitochondrial COI gene — cytochrome c oxidase subunit one. For plants, it's more complicated — they use a combination of markers because COI doesn't work well for plants. For fungi, it's the ITS region. You sequence that marker for your specimen, and you compare it against the Barcode of Life Data System — BOLD — which now holds millions of barcode records.
Instead of comparing morphology against type specimens, you're comparing a genetic barcode against a reference library.
When it works, it's transformative. You can identify a fragment of tissue, a larva that looks nothing like the adult, a gut content sample. You can flag specimens that look identical but have deeply divergent barcodes — cryptic species that morphology alone would never reveal. Paul Hebert at the University of Guelph, who pioneered this, estimated that barcoding could accelerate species discovery by an order of magnitude.
It's not a replacement for traditional taxonomy, is it? It's a tool that depends on a reference library built by taxonomists.
That's the critical point. A DNA barcode is only as good as the reference database behind it. If the specimen that produced the reference barcode was misidentified — and misidentification rates in some collections are not trivial — then every identification downstream of that barcode is wrong. Barcoding doesn't eliminate the need for taxonomic expertise, it amplifies the consequences of getting it right.
The bottleneck remains the human expert who can verify that the specimen associated with that barcode is correctly identified. And those experts are retiring and not being replaced.
This is the taxonomic impediment in its human dimension. University systematics positions have been cut for decades. Molecular biology departments absorbed the funding and the students — and molecular biology is important, I'm not dismissing it — but molecular biology without organismal expertise is like having a library of texts with nobody who can read the language. You can sequence everything, but you can't interpret what the sequences mean in an ecological or evolutionary context.
The prompt frames this as the irony of needing more taxonomists right as universities cut positions. And it is genuinely ironic. Biodiversity loss is accelerating. The public is more aware of extinction than ever. And the people who can actually tell you what's being lost are an endangered species themselves.
There was a paper a few years ago that referred to the "extinction of taxonomists" as a parallel crisis. In some invertebrate groups, there are literally only one or two people in the world who can reliably identify species. When they retire or die, that expertise vanishes. The specimens are still in museums, but the knowledge of what they represent is gone.
You can't train a replacement from scratch in a few years. Taxonomic expertise is cumulative. You spend a career learning a group — the morphology, the variation, the literature, the type specimens. It's an apprenticeship that takes decades.
There are efforts to address this. The Global Taxonomy Initiative under the Convention on Biological Diversity has been trying to build capacity, especially in biodiversity-rich countries that lack taxonomic infrastructure. GBIF — the Global Biodiversity Information Facility — has been aggregating occurrence data from museums and surveys worldwide, making it freely accessible. That's a massive achievement — over two billion occurrence records now. But data aggregation doesn't create taxonomists. It makes the work of existing taxonomists more efficient, but it doesn't replace the human expertise.
Integrative taxonomy — the prompt mentions that too. This is the attempt to combine multiple lines of evidence, right?
Integrative taxonomy is the current best practice. You don't rely on morphology alone, or barcodes alone. You combine morphology, multiple genetic markers, ecology, behavior, biogeography. The idea is that each line of evidence has weaknesses, but converging lines of evidence give you confidence. A barcode difference plus a morphological difference plus a geographic separation — that's a strong case for a new species. A barcode difference alone, with no other corroboration, is more ambiguous.
Which is a higher standard than most of the two million described species were ever held to. The vast majority of species were described on morphology alone, by a single author, with no genetic data at all.
That's an uncomfortable truth. If we applied the standards of integrative taxonomy retroactively, a significant fraction of currently accepted species would probably fail to meet them. But we don't have the resources to re-examine everything. So we live with a taxonomic literature that's a palimpsest — layers of different standards from different eras, all treated as valid until someone has the time and expertise to revise them.
That's exactly what it is. And the people doing the revising are working on shoestring budgets in institutions that are perpetually under threat.
Let me give you a concrete example of the funding situation. The National Science Foundation in the United States has a program called Systematics and Biodiversity Science — that's one of the primary funding sources for taxonomic research in the US. Its budget has been essentially flat for years, while the cost of doing the work — sequencing, fieldwork, publication — has gone up. Grant success rates are low. Early-career taxonomists are competing against established labs for a shrinking pool of money.
This is in the United States, which has relatively strong institutional support. In many biodiversity-rich countries, the situation is far worse. You've got countries with enormous undescribed diversity and almost no taxonomic capacity.
This is what's sometimes called the "taxonomic imperialism" problem — specimens collected in the Global South, deposited in Northern Hemisphere museums, described by Northern Hemisphere taxonomists, and the country of origin never builds the capacity to do the work itself. There have been efforts to change this — the Nagoya Protocol on access and benefit-sharing is one attempt to ensure that countries of origin benefit from the use of their genetic resources. But it's complicated, and it hasn't fundamentally shifted where taxonomic expertise is concentrated.
We've got a field that is philosophically rich, practically essential, and structurally precarious. What does the optimistic case look like?
The optimistic case is that technology lowers the barriers. High-resolution imaging makes type specimens accessible from anywhere. DNA barcoding and metabarcoding accelerate discovery. Machine learning starts to assist with morphological identification — there are already AI systems that can identify plants from photos with reasonable accuracy. Citizen science platforms like iNaturalist generate occurrence data at scales that professional surveys could never match. And some of that data feeds back into GBIF and becomes part of the permanent record.
INaturalist is an interesting case. It's essentially a social network for people who take pictures of organisms, with an AI that suggests identifications and a community that confirms or corrects them. And it's generated data that has led to the discovery of new species.
There are species that were first noticed because someone posted a photo on iNaturalist and a taxonomist saw it and said, wait, that doesn't look like anything I recognize. That's new — a pipeline from casual observation to formal description that didn't exist fifteen years ago.
It also raises the question of what counts as a taxonomic act. If an AI suggests an identification, and a community of amateurs confirms it, where does the professional taxonomist fit in?
The professional taxonomist does the work that can't be crowdsourced. Revising a genus — that means examining every type specimen, resolving synonymies, deciding which names are valid and which are not, producing a classification that reflects evolutionary relationships. That's not something an AI or a community of enthusiasts can do. It requires deep knowledge of the group, access to collections, and the authority to make nomenclatural acts that are valid under the relevant code.
Authority is the right word. Taxonomy is one of the few sciences where individual judgment still carries legal weight. When a taxonomist designates a lectotype or synonymizes two species, that act is registered and has permanent consequences. It's a kind of scientific jurisprudence.
That's why the career squeeze matters so much. You can't automate jurisprudence. You can't outsource it to an algorithm. It requires human judgment, and human judgment requires humans who have spent years learning to exercise it.
We're left with this paradox. The tools for doing taxonomy have never been better. DNA sequencing is cheaper than ever. Digital collections are growing. AI-assisted identification is improving. GBIF makes global biodiversity data freely available. And yet the number of people qualified to use these tools to actually describe and classify life is shrinking.
I think part of the solution has to be a cultural shift in how biology departments value organismal expertise. For decades, the prestige has been in molecular biology, in genomics, in bioinformatics. Those are important fields. But a biology department with nobody who can identify the organisms being sequenced is incomplete. It's like a literature department where nobody reads the original languages.
The musical equivalent of a philharmonic that only studies waveforms and never listens to the music.
And the students know this. There's genuine student demand for natural history courses, for field biology, for taxonomy. But universities often don't offer them because the faculty lines have been converted to other specialties. It's a supply problem, not a demand problem.
The question the prompt is really asking — underneath the history and the mechanics — is whether we're going to invest in the human infrastructure to finish cataloguing life on Earth before large portions of it disappear.
That's a question that goes beyond science. It's a question about what we value as a civilization. Do we want to know what shares the planet with us? Do we want to understand the full diversity of the biosphere we're part of? Or are we content to let most of it go unnamed and unstudied?
The Linnaean project isn't finished. We've been at it for nearly three centuries, and we're maybe a quarter of the way through. And the remaining three-quarters is disproportionately the small, the obscure, the hard-to-reach, the understudied. The charismatic megafauna are mostly done. What's left is the beetles, the nematodes, the fungi, the deep-sea things that nobody photographs for nature documentaries.
Terry Erwin's famous fogging studies in the Amazon rainforest canopy in the nineteen eighties gave us the high-end estimates of tropical arthropod diversity — he extrapolated from the beetles collected from a single tree species to suggest there might be thirty million arthropod species globally. That number is debated, but the point stands. Most of animal diversity is arthropods. Most arthropod diversity is beetles. And most beetles have never been seen by a taxonomist.
The beetles inherited the earth, and we haven't even finished taking attendance.
Now, Hilbert's daily fun fact.
Hilbert: During the Cold War, the US Army's Camp Century base in Greenland — a nuclear-powered facility buried in the ice sheet — accidentally created a new lava tube ecosystem when meltwater from the reactor's cooling system carved channels through the ice and exposed mineral-rich basaltic rock. The resulting geothermal microhabitat supported chemosynthetic bacteria that had never been documented in Arctic environments, an unintended consequence of trying to hide missiles under a glacier.
The US Army accidentally invented a hot tub for extremophile bacteria under the Greenland ice sheet.
Nuclear-powered, no less. The Cold War produced some truly strange bycatch.
This has been My Weird Prompts. Our producer is Hilbert Flumingtop. You can find every episode at myweirdprompts dot com. If you enjoyed this, leave us a review wherever you listen — it helps more people find the show. We'll be back with a new episode soon.
