Eugene Gaffney, turtle expert from the American Museum of Natural History & David Krause, mammal expert from the Department of Anatomical Sciences at Stony Brook University, both in New York State, US, have recently described a new fossil pelomedusoid turtle from the Maastrictian Maeverano Formation of Madagascar. This isn’t ordinarily news, and certainly not for Gaffney, who trots the globe describing new turtles, but something in the paper, not quite based on the skull, caught my eye:
What, exactly, is gained from loading the nested taxa shown here with rank names? The text surely doesn’t need them, and in fact terms “magnafamily,” “superfamily,” and “family” are used in the text only next to the names they are applied to, thus it is always “magnafamily Podocnemidera” or “superfamily Podocnemidoidea,” never “magnafamily” or “Podocnemidera”. The only term used in isolation is “family,” and this is usually tengential to referring to one of the named “families” in the text, such as Euraxemydidae. [n1]
The use of the terms are themselves odd simply because they are redundant in the bulk of the text, which is subtly although not noticeably lengthened by their presence; excluding them wouldn’t have shortened the text, and simply referring to the clade names themselves would have been sufficient references for the reader. Including them, conversely, adds to the weight of the reader having to keep the terms in mind, thus enforcing their existence on the reader, rather than making them secondary or corollary to the discussion. This is one of the effective problems of ranks in the Linnaean System.
Removal of rank names does not result in any dramatic shift, loss of effective communication, nor does it create confusion, when usage of those rank names is redundant by their own authors’ usage with the names that they are applied to.
I made a special point in my discussions of the horrendous problems with ranks and the Linnaean System by also noting that authors will place taxa into the context that their ranks afford or reflect special attention, and even when not explicit, their nomenclature follows this. This is present in discussions of taxa that we talk about, such as Caretta, Tyrannosaurus, or Homo, and not what we really mean by these terms, being the species themselves. They are proxies for these names, capitalized and everything, placed in premier importance. In rankless systematics, one can do away with this over-riding structure and abandon use and even sense of the term “genus,” which causes “Caretta” to be just a name including caretta, much how Hominidae includes Homo. When you have various nesting names, when they only contain a name that only contains another all the way down to the species (e.g., Opisthocomiformes, Opisthocomidae, Opisthocomus, hoatzin) you are merely dealing with nested and redundant taxa, and when they are clades, they have explicit definitions which render the value any has over another moot. I think this is the greatest fear that keeps Linnaean Systematics running, that abandonment of it will cause people to think “How will I know how my clades are arranged!?” Your phylogeny, my dear Dr. Gaffney. [Dr. Gaffney has not actually said this: I am paraphrasing Doyle. I am also repeating something I have been told in person and in correspondence from those who still advocate use of the system, often with the appellate "What other system do we have?" to which I can only answer "Any system without ranks is better than one with them." Let the taxa speak for the themselves.]
[n1] Note that while I italicize the names of clades in this text, I do not do so when quoting or citing, as this preserves the original intent and citation of the authors.
Gaffney, E. A. & Krause, D. W. 2011. Sokatra, a new side-necked turtle (Late Cretaceous, Madagascar) and the diversification of the main groups of Pelomedusoides. American Museum Novitates 3728:1-28.