I’d promised some follow-up posts to two previous ones in which I describe other “issues” I perceived. The first was a continuation of what I felt was wrong about the argument of Scannella and Horner (here) in regards to their [attempted] subsuming of Torosaurus into Triceratops (leaving Triceratops as the only standing genus-rank taxon for the material at hand). The second was a continuation of the three things I found wrong with Greg Paul’s latest book, The Princeton Field Guide to Dinosaurs (here). This post will attempt to follow-up both of these posts. Happily, the issues I would raise on these posts are convergent.
It would take way too long to continue discussing something I’ve been discussing with various people to discuss it … again. Aside from that jumble of a sentence, this post should be much more succinct.
Let’s sum up a viewpoint:
Taxonomy is an art, not a science. It follows some rules, most of which are agreed upon by various organizations that have been followed, ostensibly to resolve order from chaos. One is tempted to scribe biological laws and rules to taxonomy that are independent of the convention of naming, but when you remove the name itself, all you have left is a series of biological organisms and nothing else. A name, therefore, is just a name.
If we think of taxonomy as just an art, then its practice in distinguishing fossil organisms must follow the sense of the aesthetic or political or personal rather than the scientific. If the science matters, then a name doesn’t; rather, the psychological tool of a language-neutral term for said organism merely puts aside common-language words for it. We may call it hund, dog, kalb, gau, cane, pies, erhar, qimmiq, jagua, aydi — but all of these are Canis lupus and each technical article, were it written in each of these languages, would be using that term. Science cares about reproduction of experiments, verification, etc. And while this does require the person making an observation sure he is perceiving the same type of animal another has, qualifying one’s perception is often just as tricky as making the biological identification.
In both previous posts, I called attention to what appears to be an error on the authors’ parts:
Scannella and Horner  infer that the material formerly ascribed as Torosaurus occupies a developmental trajectory that ends a progression began with what are commonly called Triceratops. On a biological level, this organized ontogeny is very sound, and makes a good argument toward plastic modification and remodeling of bone tissues. It also makes a good case for the perception of so-called agonistic or display features, namely the frill and horns. Scannella and Horner, as I mentioned before, claimed to reduce the taxa to a single form, Triceratops, while retaining two species in the latter, and failing to distinguish which of these species the now-invalid Torosaurus latus falls into. The numbers are off here, and this has a lot to do with the implications of the value of the genus-rank and that of the species-rank in Linnaean taxonomy. Without seeing the issue, perhaps, Scannella and Horner validate the quality of the genus over that of the species, and fail to identify synonymy as per the ICZN (remember those chaos-averting rules I mentioned above?).
But wait, there’s more…
This is not the first time authors have subsumed taxa for the sake of a perceived ontogenetic series, in many cases failing to justify variation or the lack thereof, or equating noted variation as being of ontogenetic form, for various other reasons. And it will not be the last time. Taxonomy, as I said, is an art: it has its practitioners, dilettantes, and its critics. Certainly many taxonomic lumps in the past have been upheld through time and recovery of better material; additionally, however, other taxa have been split because of new data. Stratigraphy plays a part (taxa in the same formation or bedding horizon that are very similar to one another may likely be the same taxon) while geography may also split taxa apart; these things are validated only through sampling and association, such that groups of fossils found together of similar animals may more than likely belong to the same type of animal. When there is variation, or a lack of overlapping data, this question becomes harder. Scannella and Horner argue through their samples that the material at hand renders a generic ontogeny that lumps Torosaurus and Triceratops; and this likely to be true, but I then question the self-same stratigraphy that appears to suggest that the two species of Triceratops, prorsus and horridus, are distinct as well as on morphological grounds. What, then, does this leave of the species resolution?
Greg Paul’s latest book makes a number of what some may consider extremely questionable or entirely unjustified assumptions of taxonomic lumping. Many taxa are joined together for the sole reason that the author appears to think that they are the same animal. It can be said Paul has previously justified this by pointing out that variation in the taxa Varanus and Canis are so broad that a great range of taxa can fit into them. Some authors have split these genera into subgenera, others into “species groups” with their own informal non-taxonomic labels, or have left the issue alone, best left untouched. Still others have argued that because a “genus” is an entirely undefined entity, without any agreed upon or morphological quality, that it is literally impossible to determine by appearance what is or isn’t a genus. Because of this, the practice of lumping taxa is questionable from the get go.
I would like to reiterate a small issue with splitting as well as lumping, and it presents the main reason I question the two works above as well as others in this series of posts on taxonomy.
Without ample justification on morphology or phylogeny, which informs the position of a fossil specimen or biological organism relative to its compatriots in comparison, the elective decision to re-brand that form as a part of an old taxon or as a new taxon is entirely unscientific. When it is purposed as scientific, or a paper or book is made with the intent of being scientific while practicing what is essentially art. I say “ample,” but sometimes, this is really just “any;” a great deal of works out there offer no justification but a perception on variation across another group of animals for which the author makes no attempt at quantifying. We must assume, for example, that Citipati (in Paul’s usage) has the same value or content as Canis in order to directly compare the variation or diversity among them. If, however, Canis represents a non-genus type of taxon, a clade, or a name for a lineage segment, then we are freer to relate taxa as we choose, split from it, lump into it, so long as we are clear that is why we are doing it. Using instead the myth of the genus as the justification makes it art.
If you split a taxon because it “looks different” or “might relate to something else,” but do not clearly justify this? Then you are practicing art. If you lump something because if fits the same general concept of “genus” but fail to justify what, in fact, a genus is? Then you are practicing art.
I am perfectly fine with naming new taxa from new material; I won’t ever do it myself (I think). I am also fine with finding synonymy so long as the grounds are solid and close to unshakable in the face of argument (and welcoming of argument). I do not care what to name, say, a new genus for an old species, or an old species for a new genus. What I care about is the process and the reasoning. I’ve become entirely critical of the process in general, and very cynical when presented with anyone attempting to raise genera from oblivion, new genera for species when the phylogeny doesn’t change or is unlikely to do so, or lumping because someone perceives ontogeny exists but fails to point out the relevant species variation in part of the ontogenetic sequence (and this isn’t even pointed at Scannella and Horner, although they did do this).
That is all.
 Scannella, J. B. & Horner, J. R. 2010. Torosaurus Marsh, 1891, is Triceratops Marsh, 1889 (Ceratopsidae: Chasmosaurinae): Synonymy and ontogeny. Journal of Vertebrate Paleontology 30(4):1157-1168.
 Paul, G. S. 2010. The Princeton Field Guide to Dinosaurs. (Princeton Univerity Press, Princeton.)