Friday, January 4, 2013

"Evolutionary" classification of Ranunculus, part 1

One of my hobbies, if that term is permitted, is to try and understand the stance of those who call themselves "evolutionary" systematists (ES) - the minority of taxonomists and systematists who desperately try to turn the wheel of progress back by a few decades and return to the time before formally accepted supraspecific taxa had to be circumscribed to be monophyletic. Especially at the end of 2010 and the beginning of 2011, I tried to find out if I had overlooked any decent arguments and if they were perhaps right in their assertion that the insistence on monophyly does not make sense.

After digging through dozens of publications and opinion pieces, some of them more intellectually coherent than others, I have concluded that most ES don't understand phylogenetic systematics (in particular the difference between tokogeny and phylogeny and that only species relationships matter), some of them don't even understand how parsimony analyses work, and all their arguments are built on fallacies or false premises except perhaps one, and that one is built on the assumption that we want to classify demonstrably ancestral species in a Linnean framework, which is virtually never the case. What is more, all their arguments have repeatedly been shown to be wrong or unfounded but they simply soldier on regardless; there is probably no point trying to convince many of them any more if it has not worked so far.

However, not for nothing did I start by saying that it is one of my interests. I remain curious if any of them comes up with something novel and intriguing, and in particular if they can make any new suggestions for what may be their fundamental methodological problem: the complete lack of a universal, objective and testable criterion for segregating paraphyletic residues from monophyletic groups nested within them.

Note that the search for such a criterion is doomed right from the start once we realize that evolution is gradual. The best criterion that ES can suggest is generally a long branch, i.e. a lot of morphological divergence, separating the paraphyletic residue from the nested clade. If the creationist caricature of the theory of evolution as of a frog giving birth to a dog were correct, then "evolutionary" systematics would be less of a problem, because in one fell swoop we would have a lot of divergence between (in this case) the paraphyletic frogs and the monophyletic canids. However, in reality there are no long branches, only the illusion thereof due to extinction, and thus any classification based on the length of the branch connecting the nested clade with the grade, as favored by ES, is broken by the discovery of a missing link, be it as a fossil or as a newly discovered living species.

But even if we restrict ourselves to living taxa only, the problem remains that it is unclear how long precisely a branch has to be for the paraphyletic taxon to be acceptable. ES love to mumble mysteriously about "information content", "adaptation", "ecological features", etc., but in the end the decision is still based on personal authority instead of an objective and universal criterion like monophyly.

With this in mind, I found it very interesting to recently discover a new paper by Hörandl & Emadzade (2012) presenting an "evolutionary" classification of the buttercup genus Ranunculus as a case study. Hörandl is one of the major botanical proponents of reversing the conceptual progress systematics has made in the 20th century. Let's see how she and her co-author classify, and contemplate their arguments and methodology!

Inspired by Deacon Duncan's critical discussions of apologetic literature, I will make a short series out of this, going through the publication from start to end, although not in a slavishly linear fashion. I will probably spend a disproportionally long time in the introduction and discussion sections, because it is there that the argument is built for accepting paraphyletic groups and even a first cursory reading shows that they provide a lot of material. It is also apparent that the paper will allow me to discuss several issues with implications well beyond the question of paraphyletic taxa in systematics.

In case there is a non-systematist actually willing to read this series, please be aware that it will be dense. I am writing quite verbosely as it is, and to keep my posts comparatively concise I will not explain most of the technical terms. It might therefore be advisable to keep Wikipedia open in another tab if you want to slog through this. Any systematist, phylogeneticist or taxonomist coming across these posts is sincerely invited to argue with me or correct me. However, I am not doing this because I expect to have a lot of readers anyway, but mostly for fun and to order and preserve my own thoughts.

Continue reading.


Hörandl E, Emadzade K, 2012. Evolutionary classification: A case study on the diverse plant genus Ranunculus L. (Ranunculaceae). Perspectives in Plant Ecology, Evolution and Systematics 14: 310-324.

No comments:

Post a Comment