Such expressions as that famous one of Linnæus, and which we often meet with in a more or less concealed form, that the characters do not make the genus, but that the genus gives the characters, seem to imply that something more is included in our classification, than mere resemblance. I believe that something more is included; and that propinquity of descent,—the only known cause of the similarity of organic beings,—is the bond, hidden as it is by various degrees of modification, which is partially revealed to us by our classifications (Darwin, 1859, p. 413f).

Thursday, 19 March 2009

Myths that Evolutionary Taxonomists live by

Confused evolutionary taxonomists have once again made a stand in the pages of Taxon. The editorial by Brickell et al. (2008) represents a vote of no confidence in favour of paraphyletic groups - as if democracy in science has (or ever had) any valid scientific or empirical merit. This time the confusion stems from
    "Recent developments in taxonomic theory have resulted in the production of classifications of the Flowering Plants that are causing concern to all involved in horticulture — gardeners (both amateur and professional), nurserymen, landscape architects, foresters, designers, conservationists, and journalists, as well as to botanists engaged in many different, non-taxonomic disciplines and to other users of plant names generally" (Brickell et al., 2008:1047).
One wonders what those nasty molecular phylocodists and monophyly-peddling robbers of horticultural dignity are up to? Perhaps plotting horrid phenetic-cladogram-trees?

They certainly are a confused mob as the above mockery demonstrates. One way out of this wet paper bag is to see the world from a evolutionary point-of-view. Artificial classifications, which are useful in identifying plants, for example, are not necessarily evolutionary (in the sense of monophyletic). Some may turn out to be, but only empiricism will provide us with the necessary evidence. That is we need cladistic methods to test taxonomic claims of relatedness (i.e., monophyly). Evidence and empiricism, however, appear to be of no use to Brickell et al.(2008).
    "Cases such as these (and there are more that could be quoted) have arisen from a fundamentalist approach to cladistic methodology, which requires that a classification should not include paraphyletic taxa"(Brickell et al., 2008:1047).
I tire of saying this: Paraphyletic taxa are not of any use. They do not represent natural classifications. They are not a result of a common shared history. Paraphyletic groups are not anything other than names, just as 'leprechauns', 'unicorns' and 'griffins' are only names. Why then do horticulturalists and evolutionary taxonomists want them in their classifications?

This is because paraphyly is not a phylogenetic problem (phylogenies are essentially monophyletic - don't get confused with genealogies, which have nothing to do with classification). Paraphyly is taxonomic problem that evolutionary taxonomists refuse to face. If a group is paraphyletic, it means it has failed an empirical test for natural grouping. It needs to be revised. Revising groups is what taxonomists do best. Instead of embracing cladistics as a valuable tool, evolutionary taxonomists like Brickell et al. (2008) dismiss it because their favorite taxonomic groups under threat from revision. Acknowledging that one's group is paraphyletic and therefore requiring revision does not make you a bad taxonomist. Keeping non-existent groups however is. I don't want to say that Brickell et al. (2008) are 'bad taxonomists'. They a bunch of misguided evolutionary taxonomists who are confusing different things, namely artificial and natural classifications - an on-going problem since the 18th century. This confusion has led to several 'myths that evolutionary taxonomists live by'. I use Brickell et al. (2008) as an example.

Myth 1: If it ain't broke, don't fix it

One common misconception is that of historical 'significance' or 'pragmatism' in science. For example, 'Reptiles' is a wonderful term and describes all manner of organisms such as fire-breathing dragons, sea serpents and the Sea Devils from Dr. Who. (Remember them?) This does not mean that the Reptilia are immune to scrutiny or empiricism - in fact they're not. The same is true for taxa within the angiosperms
    "We, as horticulturists and horticultural taxonomists, wish to express our strong support for these pragmatic views, which will encourage the retention of familiar and widely used taxa [e.g., Dionysia, Dodecatheon, Soldanella, Omphalogramma, and Cortusa] which are distinctive and historically important" (Brickell et al., 2008:1047).
I empathize. Good names that are linked to poorly defined groups (which, incidentally is what makes them paraphyletic) sucks. But that's life ... sorry, that's systematics.

Myth 2: Taxonomy needs to be 'stable'

There is no such thing as a completely stable classification of living things. This is not because everything is fluid and 'moving' and 'unclassifiable'. As new evidence comes to light (e.g., molecular data), so do new discoveries. But Brickell et al. (2008) beg to differ
    "We are not against taxonomic change, which will continue to be a standard outcome of taxonomic research, but insist that horticulture needs a stable (though not static) classification and nomenclature that can be understood and applied effectively by horticulturists (and others) who exhibit a very wide range of levels of taxonomic sophistication.
Clearly they are against taxonomic change as that is what paraphyletic groups inevitably lead to - taxonomic change.

Myth 3: The needs of end-users are important

Let's face it, the end users of taxonomy are mostly other taxonomists. Regardless of the descriptions and keys out there, trilobite collectors and purveyors of fossils for instance, still insist on calling any large brimmed harpetid from the Devonian rocks of Morocco Scotoharpes. (The aforementioned genus does not occur in Morocco or in the Devonian). The concerns of end users is quite topical at the moment and will not be discussed in depth here (see Wheeler et al. 2004). The fact of the matter is that end users have to share the burden of changing taxonomies. This may make horticulture and conservation for example harder to do, but many are attempting to reduce this burden through employing new electronic media, which has created new emerging fields such as biodiversity informatics and cybertaxonomy.

Myth 4: Molecular systematists and cladists are all phylocodists

This is a myth that has been exacerbated by Brummitt (2006, 2008). Not all molecular systematists and cladists agree with the phylocode. In fact some of the most ardent critics of the Phylocode are cladists who use molecular data (e.g., see Nixon et al. 2003). Moreover, supporting monophyletic taxa does not automatically make you a Phylocodist or anti-Linnean. Here is an example from (Brickell et al., 2008:1047)
    "(cf. Brummitt in a note to a colleague: ‘By any logical consideration either one has a monophyletic system with an infinite number of nodes but no ranks, for which the PhyloCode is designed, or you have the Linnaean system with ranks at very few levels, and paraphyletic taxa’".
Classifications are not divided into 'the Phylocode' versus 'Linnaean taxonomy'. This dichotomy is false. The Linnaean system of taxonomy remains silent about paraphyly or monophyly. Biological classification consist of artificial and natural systems, the modern Linnaean System belonging to the later. As taxonomists, we aim to find natural groups (a.k.a 'monophyletic groups') in our Linnaean System. But paraphyletic groups, like Linnaeus sexual system, are artificial. They may be useful in identifying organisms, but they do not reflect natural evolutionary groups and should be exempt from our classifications. Brickell et al. (2008) are misguided and confused if they are to believe that paraphyletic groups are 'natural' or even evolutionary in anyway.

Myth 5: What does the Molecular data mean?

There are many ways to tell someone to p*ss off and this is a beaut:
    "In saying this we do not wish to imply that phylogenetic studies are unimportant or uninteresting; only that the purpose for which they are produced is not applicable to horticultural needs and practices. (Brickell et al., 2008:1047-1048)"
I agree. Molecular trees (which is what Brickell et al., 2008 are referring to above) do not have any characters listed at their nodes. If horticulturalists are to follow our lead and adopt new groups based on molecular data, then show us the characters that support it as a monophyletic group. If the group is paraphyletic then do the revisionary taxonomy. There is however a catch. Molecular systematists do not necessarily do all the work. Saying that something is paraphyletic and in need of revision without any morphological evidence is hard for any taxonomist or horticulturalist to swallow. I think that Brickell et al.(2008) are on to something here and it is well worth pursuing. Consider this myth busted.

Unfortunately Brickell et al. (2008) do not qualify for this year's Pewter Leprechaun although their attempts at misusing paraphyly have reached a particular zenith.

Brickell, C.D., Crawley, M., Cullen, J., Frodin, D.G., Gardner, M., Grey-Wilson, C., Hillier, J., Knees, S., Lancaster, R., Mathew, B.F., Matthews, V.A., Miller, T., Noltie, H.F., Norton, S., Oakeley, H.J., Richards, J., Woodhead, J. (2008). Do the views of users of taxonomic output count for anything? Taxon 57:1047–1048. Nixon, K. C., J. M. Carpenter, and D. W. Stevenson. 2003. The PhyloCode is fatally flawed, and the "Linnaean" system can easily be fixed. Bot. Rev. 69: 111–120. Wheeler, Q. D., Raven, P. H., Wilson, E. O. 2004. Taxonomy: Impediment or expedient? Science 305: 285.

1 comment:

David Marjanović said...

For example, 'Reptiles' is a wonderful term and describes all manner of organisms such as fire-breathing dragons, sea serpents and the Sea Devils from Dr. Who.I'm so stealing this...

No, I won't steal it, I'll properly cite it in a paper of mine, assuming I'll ever find the time to get back to that particular manuscript.

Classifications are not divided into 'the Phylocode' versus 'Linnaean taxonomy'.Not least because phylogenetic nomenclature doesn't consist of hacking the tree into pieces and stuffing those into boxes of sorta kinda predetermined size. It leaves the tree intact and ties labels to defined places on it (not necessarily nodes, BTW); it is, in short, not classification.

Molecular trees [...] do not have any characters listed at their nodes.That's not true, even though those characters are boring to read: "substitution at position X of C by G", "deletion at position Y"...

Molecular data are just as useful as morphological data. They come with their own advantages (easy sampling of humongous amounts of data -- cutting the time requirements for a serious study of a large group by orders of magnitude; no worrying about how to delimit states; knowable probabilities for each possible state change) and their own disadvantages (alignment; how to treat gaps; greater susceptibility to long-branch attraction; possibility for wrong models of evolution to screw everything up; inapplicability to most fossils, and therefore greater susceptibility to long-branch attraction again). They need to be studied as carefully as morphological data and are just as trustworthy (provided that the analyses are comparable in scope, which usually isn't the case -- there are very few decently-sized morphological analyses out there).

Now, for an almost perfect character, see SINE and LINE insertion sites... except those, too, are not applicable to anything without copious amounts of preserved nuclear DNA.