Calling Time on Biological Nomenclature

Gathering Storm
Gathering Storm

I was writing a report on the role of nomenclators in PESI when I realized that (with a little tweaking and injection of dangerous opinions) one section would make a good blog post.

In order to facilitate the accurate exchange of taxonomic information, both within the taxonomic community and more widely in the biological and environmental sciences, the e-infrastructure needs to provide  two dictionary functions for scientific names of organisms i.e.

  1. A recognized list of the names used. To establish that any two studies are actually using the same names whilst accounting for spelling variants and homonyms as well as to facilitate consistency in spelling and presentation.
  2. A mapping between the names and descriptions of the taxa they are used for. To establish that any two studies are using the names in the same sense or compatible senses.

If the ICBN and ICZN codes required all names to be registered in a single or limited number of places then this would effectively fulfil the first function. Unfortunately neither the ICBN or ICZN codes require names to be registered. Neither do they require names to be published in a particular list of journals. They merely set out the conditions for effective publication. The  publications in which new names appear could be published anywhere and deposited in any library. There is no requirement for them to be peer reviewed.

Both codes require that the oldest name be used for a taxon yet provide no mechanism for ascertaining if the oldest name has been found. This means that any revisionary treatment of a group has the potential to destabilise the names in current use by uncovering older publications.

Although continuously updated the codes are still focused on a world of paper publication with a relatively small number of journals in each specialist field. The codes are run on the basis of consensus building and, because the need for registration is not yet universally recognised, no such registers have been established. ICZN is in the process of attempting to establish one in the form of ZooBank. A register for ICBN was proposed, then they fumbled and dropped the ball but may pick it up at some point in the future.

In contrast to these attempts the Bacteriological Code (ICNB) has an effective registration system that has been in place for nearly thirty years. All bacteria names have to either be in the Approved Lists of bacteriological names published in 1980 or published in the Journal of Systematic and Evolutionary Microbiology.

Unless some external body forces, and perhaps funds, the ICBN and ICZN to establish effective registers of names the community is forced to build and maintain separate databases of nomenclatural data – nomenclaturors such as IPNI, uBio, Index Fungorum, etc etc. Multiple databases with multiple versions of how to cite names.

Ultimately scientific names are only human readable strings for referring to biological taxa. As we move to an age where many if not all taxa (even plants!) will be defined in terms of their barcodes* and most biological data is handled by machines then the barcodes become the real identifiers for taxa and scientific names just so much eye-candy. Names are just the display labels for the links. The targets of the links are what is important and they will be defined in terms of barcodes embroidered with morphology.

I am afraid I can see the writing on the wall for old style nomenclature and with it a lot of traditional taxonomy – flame on TAXACOM!

* Before you have a knee jerk reaction to the thought of barcodes defining taxa rather than merely labelling them you might read Identifiers, Indentity and Me where I rather unskillfully advocate that an identifier erects a thing rather than things having identifiers. All we have to do is say “Anything that has this barcode is this thing”

11 Comments on "Calling Time on Biological Nomenclature"

  1. While it will be nice to have a particular place to register new names, what is essential before this process can start is to have a list of all the described species to date. There are no funded groups that have that as a goal and the few unfunded ones rely on data given to them by authors of that data [ie. fish base, spider catalog].

    Plants at least have a genus level catalog. Animals have no plans for this.

  2. The Southern California Association of Marine Invertebrate Taxonomists (SCAMIT) has been standardizing name usage and providing guidance and tools for proper identification of infaunal and epibenthic marine invertebrates in Southern California since 1982. We released the 5th Edition of our Taxonomic Listing of Macro and Mega-Invertebrates in July 2008. The indexed listing contains 2551 species across 18 phyla. It includes an updated systematic heirarchy with published synonmys and synonmys of incorrect usage regionally. We also include provisional names for species suspected of being undescribed which are fully documented by SCAMIT and its members. We are in the process of developing an online taxonomic database, with the species list as the backbone, which will link to a variety of resources (taxonomic information, images, occurance information, habitat information, references, links to other websites, etc) from dynamic species pages. An expanded west coast version of the species list beyond the Southern California region is a longer term goal. Our website, while accessable, is currently under revision and in need of updating. Edition 5 of the species list can be found there.

  3. Roger — this is all well-known stuff, and there are a series of semi-coordinated efforts moving forward on addressing these issues. The emergent technical foundation will most likely be GNA/GNI/GNUB, which is the first such technical foundatiuon that I have seen that (at least has the potential to) transcend all the partisan barriers that have prevented us from uinversally solving these problems decades ago. This does *NOT* represent “yet another grand nomenclator project”; rather it is an *architecture* (GNA) that can provide a common GUID infrastructure (GNUB) common to all of the existing initiatives, and assoicated tools to bridge text-string names to the GUID infrastructure (GNI). Plain-English documentation of this stuff is in the works. A meeting after e-Biosphere amongst representatives of the 3 major codes (ICBN, ICZN, Bacteriological Code) resulted in much more harmony and common vision than I would have predicted.

    By the way, the whole *point* of GUIDs for taxon names & concepts in my mind has been to represent the electronic link that exists behind the human-friendly text-string name-label.

  4. Great post! Interetsing that the comments (and most of the “flames” in TAXACOM) seem to have missed your point that names are so last century, and that our obsession with them will seem odd to a generation brought up on DNA barcoding.

  5. DNA barcodes have many uses. Unfortunately they don’t act as barcodes as they currently cannot provide – and perhaps never will – unequivocal species-level identifications (e.g. land plants, cichlid fish, freshwater gastropods). Some of this is very simple: a fair percentage of currently recognised species are likely to turn out to be paraphyletic with respect to say COI. Gene trees aren’t species trees – right!
    So if we are interested in species then it would be unwise for DNA barcodes to function as or replace taxonomic names. I guess that’s where the comment about embroidering with morphology comes into play – in which case you’re well on the way to integrative taxonomy not barcoding!

    I think an interesting question is “What is the fundamental unit of biodiversity informatics?” My answer would be the scientific name (+ type specimens + publication), a notion very close to Zoobank’s registration unit. Far from scientific names disappearing in the future – I believe it is essential that they are restored to pre-eminence so that the upcoming millions of GUIDS and LSIDS can be nailed back to the one thing we think we know – name holding voucher specimen, including type specimens, in natural history collections. The anchors to all the links must lie in specimen vouchers (in the widest sense, e.g. specimens, DNA, videos etc). Without that the whole informatics enterprise will remain only tenuously linked to the targets. Indeed museum vouchers may be all we will know or can know for a rapidly increasing percentage of Earth’s pre-Anthropocene recent macrobiota.

  6. Once again late to the party. But then again, so is this blog post. Nothing really iconoclastic here (what is the point of a post if you are not going to break something?) and we have heard it before in various places. The distinction and relationship between species (which for argument’s sake we will assume to be real), variable human concepts of what these realish things might be, and the names we apply to these concepts (assuming them to represent the realish things) still have not been worked out (or if they have been worked out in the minds of the anointed few, they are are still far from understood in the minds of most taxonomists, let alone the general community).

    Destabilization by nomenclature does not worry me – it is just a bloody name, mostly of an organism no-one knows or cares about, and if it is going to create problems in the world of commerce, there are conservation provisions (at least in the botanical code to deal with this situation. What is a real problem is the disconnect between the species thing, the concept of what it might be and the name that might be applied to this concept – or more correctly stated, the assumption that there is a usable connection between a species, what a human thinks is a species and what a human chooses to call what a human thinks is a species. There is a chain of assumption here that, without documentation, really is quite unscientific.

    In another forum Rod Page opined of this opinion (paraphrasing from memory, sorry if I get it wrong Rod) that if taxonomy has to explicitly specify the concept to which a name applies, then it and we are rooted, to use an antipodean vernacular. A recurring nightmare is that Rod may well be right, and we are there right now.

    Registration is a bit of a side show. It will happen, in all codes. (btw, the ICBN did not fumble the ball – it took it firmly in hand and booted it out of the field because it was the wrong ball for the game!) The technology is there and it is not difficult. The sociology is a bit more problematic, but as long as the big egos resist the temptation to claim and impose authority and as long as it happens seamlessly as part of the process rather than as part of approval, I believe it will slide though and make it a bit easier to find and account for all names. But as far as accounting for all taxa, well, that is science…

    When it is eventually discovered, species may ultimately be defined by their ‘barcode’ (or dare I say it, their morphology) but if a human is going to communicate with another human about what they both think they might mean, the thing they will need to this is… hang on, it is on the tip of my toungue… a… a… got it! A name…

  7. @Jim Croft
    Ah well I didn’t actually say I was saying anything new did I! Most of this is recycling stuff that was published at the end of the last century in the Prometheus, Potential Taxon and other papers. What I have noticed in the intervening decade is that one has to keep going on and on till you are blue in the face explaining the same thing from fifteen different angles and people still don’t get it. Unfortunately for those who do get it and possibly had got it donkey’s years ago it all gets a bit boring.

    But I think I am reaching new levels of frustration with nomenclature. It really does seem pointless and stupid to have such a complex system that ensures the name you call something can never be guaranteed to be stable. If it can’t be done with human readable names how will anyone ever get it with GUIDs?

  8. @Roger Hyam
    Exactly. What I’m failing to say clearly enough is that the linking of solid nomenclatural data into taxonomic concepts would at least prevent the worst folksonomic horrors. Currently many names out there represent free-floating taxonomic concepts rooted in nothing at all but linked to misidentified snapshots or sequences or misplotted maps. Ideally the entire enterprise would have started with registered directories of ‘correct names’ (e.g. Zoobank registration units) and then later expanded to incorporate the multitude of discrepant taxonomic concepts. But that would have required natural history collection holders to have been on the ball…Still, its not too late to start.

  9. The PhyloCode, a code being developed to govern clade names, will required registration in a centralized, public database. (It also requires peer review.) See:

    Note, however, that the PhyloCode relies on other codes to provide specifiers.

    Barcodes don’t work for fossil taxa. (Well, not for most of them, anyway.)

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.