This is the third post in a series about the Taxon Name Linking Service project.
My last post was titled “Why Latin names just aren’t good enough anymore“. In that post I outlined the difficulty of combining data from multiple sources on the basis of the names written in those sources and why it would be good to use identifiers from a controlled vocabulary alongside the human readable names.
In this post I’m going to look in more depth at one way ambiguity is caused in the use of Latin names – when two names are written the same but possibly means something else. I’ll concentrate on plants in particular but many of the same things apply to animal names.
The glossary of the International Code of Nomenclature for Algae, Fungi and Plants defines a homonym as: “A name spelled exactly like another name published for a taxon at the same rank based on a different type (Art. 53.1)”. As mentioned previously names do not have fixed spellings. There are allowances for typographical or orthographical errors/variations and standardizations as well correction of terminations for the gender of the genus (see Art.50 and linked articles). Theoretically two names might have different letters in them but when one or both of them are corrected for gender and other spelling rules as applied today they might turn out to be homonyms – even though they did not have the exact same spelling when they were published or indeed for many years of use in the literature. We’ll put these matters to one side for now.
Spelled exactly like another
Despite much complexity the principle is simple. Homonyms are bad. A consistent classification should not contain taxa with names that have the same spelling. If two names exist with the same spelling then the later published one is considered an illegitimate name and can not be used as the correct name for a taxon. Unfortunately the nomenclatural code only requires a name to appear in a publication with a run of two copies, deposited in two libraries, to be considered validly published. It is therefore impossible for a botanist to know for sure if the name they are going to publish will turn out to be an illegitimate, later homonym at some point in the future. They can do due diligence and check IPNI and Google but this is no guarantee that there aren’t decades old publications sitting undiscovered in a library somewhere. Until there is mandated registration of names (which is highly unlikely) this will always be the case.
When we are combining data from the literature and multiple databases we can usually tell homonyms apart because they have different authors. Except that isn’t true! Name authors aren’t always cited. Furthermore until the 21st century there was no standard way of citing author names. Even if everyone had religiously followed the standard way of citing authors from 1997, when the first list was published, that would leave two hundred years worth of literature where the standard wasn’t followed. The result is finding two identically spelled names with different author strings might mean you are looking at a pair of homonyms or might not, depending on how you interpret the author strings. Working out if the authors are the same can often be easy for a human, but not always. It can be more error prone for a computer.
If you were writing a floristic or monographic account you might not come across homonyms very often and not think they were a big problem but the age of big data brings bigger problems. Looking at the current World Flora Online Plant List there are 200,569 species name entries that are spelled identically but have some difference in their author strings. Yes that is two hundred thousand entries (~100k pairs). Many of these need sorting out by a human. As an example:
They look like the same name and someone has just substituted an ampersand for the “ex”. Or is it the other way around? The literature would need checking. The correct one can’t be determined by a machine unless we know for sure that datasource 1 can be trusted over datasource 2.
These two look like entirely different names published on different types and indeed the A.Rich name was published in 1850 two years before the C.A.Mey. ex Turcz. name and so has priority. Name 1 is an illegitimate, later homonym.
Of course there is another reason why we might not be able to tell two homonyms apart on the basis of their author strings. Sometimes the same author publishes the later homonym themselves! Presumably they forget that they have already used that name or think there will not be confusion because the names are far enough apart in their classification. Here we can name and shame Albrecht Wilhelm Roth (1757-1834) as an example because he is far enough back in history not to mind.
Albrecht published the genus name Reichardia three times based on three different type species in Bot. Abh. Beobacht. 35 (1787) [wfo-4000032695], Catal. Bot. 2: 64 (1800) [wfo-4100002540] and Nov. Pl. Sp. : 210 (1821) [wfo-4000032694]. Homotypic genus names are particularly problematic as they cause confusion in any species placed in any of the versions of the genus.
We can tell Albrecht Roth’s names apart because they were published in different places but we need to go to those places to check that they weren’t based on the same type specimen. If he had been citing the same type then the names would not be homonyms they would be isonyms.
Isonyms
The nomenclatural code states under point 6.3 “When the same name, based on the same type, has been published independently at different times, perhaps by different authors, then only the earliest of these “isonyms” has nomenclatural status. The name is always to be cited from its original place of valid publication, and later isonyms may be disregarded.” In the glossary isonym is defined as: “The same name based on the same type, published independently at different times perhaps by different authors. Note: only the earliest isonym has nomenclatural status.”
An example of an isonym pair is Artemisia serotina Bunge [wfo-0000126669] which was published in Beitr. Fl. Russl. 165 (1852) and also published in Mém. Acad. Imp. Sci. St.-Pétersbourg Divers Savans 7: 340 (1854) [wfo-1000076712]. Really these aren’t separate names at all. They actually are “the same” (homo in Latin) names and may not warrant separate name entries in a database. But they can always be indistinguish from real homonyms without research.
“ex” in the authors
The nomenclatural code allows the author or authors who validly publish a name to acknowledge a previous author or authors who published the name but not validly by including their names followed by ‘ex’ and then their own names in the author citation. This occurs in two of the examples above. In many ways names with “ex” in the authors string are analogous to isonyms. Both are only concerned with a single name (linked to a single type and spelled in the same way) published in two places but in the case of “ex” names the original publication was not valid and the subsequent publication was valid. With isonyms it is the other way around. The first publication is the valid one and the subsequent publication is superfluous. There is no term in the code to describe these names with multiple places of publications that have “ex” in the author citation.
It’s all Latin and Greek to me!
To distinguish between homonyms, isonyms and “ex” names the places of publication and type specimens need to be known but they are not known until the names have been fully researched. The code does not provide a term to refer to names spelled exactly alike for which this information is unknown – which is usually the case if you come across to identical names in the literature or a database. In fact the terminology used by the code is just plain confusing. It uses two prefixes from different roots both with the same meaning and meaning “the same”. The Latin prefix (homo– in homonym) is used to indicate that words are spelled the same but in fact refer to different names. Homonym therefore does not mean “same name” it means different names written the same. The Greek prefix (iso– in isonym) is used to indicate the words refer to the same name object (signified by the type). A clearer approach would have been to use homograph (homographic names) for names written (graphed) the same and homonym for the same name (what we use isonym for now). Homonyms would then be a subset of homographs where we know that identical words refer to different types. Unfortunately it is far, far too late to make any changes to terminology.
Does any of this really matter?
All this is very important from the perspective of a taxonomist working to produce a coherent classification. To the eight billion other people on planet Earth, even to other biologists, it appears to be of little import. People use scientific names to reference taxa (organisms). To do this they need a real or implied classification that governs the taxonomic placement of that name. Its nomenclatural status is not important. They assume that has all been sorted out by a taxonomist – and it should have been. In the case of both isonyms and “ex” names there really is only one name and so using either version of the name (if there is a detectable difference) should make no difference to which organism is being referred to. These things only matter to taxonomists. There is only the possibility of taxonomic confusion with true homonyms, where the same name string might refer to different taxa. This is where the work of the taxonomists is important but hidden from the rest of the world. Only by curating their work in global checklists like WFO can taxonomists firstly hope to know whether they are dealing with homonyms at all and secondly share that knowledge with the wider community. This will reduce the creation of new homonyms and eventually account for practically all the existing name clashes. This is especially so for homonyms that may occur in taxonomically disparate groups governed by the same nomenclatural code, like orchids, mosses, fungi, which are rarely worked on by the same people. Ironically codes of nomenclature for macro organisms prevent this problem from being solvable because they do not mandate registration of names and are unlikely to do so in the foreseeable future. There can be a de facto solution in the form of large, consensus classifications, but there is unlikely to ever be an “official” solution to the problem of homonyms in biology.
From the point of view of TNLS and matching names homonyms, isonyms and “ex” names are a big problem if they haven’t been resolved by the taxonomic community. If a search returns several names but no opinion on which is the correct name to use then the results are useless for “regular people”. But for a taxonomist with an interest in nomenclature they might appear as a fascinating problem to solve!
For my next post we will flip from Latin to the Greek prefix “syn” and talk about what we really mean when we say a name is a synonym.
This project has received funding from the European Union’s Horizon Europe Research and Innovation programme within the framework of the TETTRIs Project funded under Grant Agreement Nr 101081903.
Leave a Reply