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, 3 April 2008

The Problem of Similarity

Systematics and Biogeography has a problem: similarity. Ever since Goethe, naturalists and biologists have been rejecting similarity. It is the foundation of artificial classifications, non-evolutionary groupings and the basis for many arguments against evolution (i.e., homology).

Similarity implies that organisms are similar and not the same, that is it remains silent about sameness. The difference between being similar and the same is astronomical. Any two things in the universe can be similar. It is not a discovery. It demands no explanation. It is a means unto itself. However, when two things are discovered to be the same, they require explanation. This is when the study of evolution begins. By denying sameness -- or ignoring it -- we remain in the realm of artificial classification.

So why, then, is similarity so popular? All methods in molecular systematics use "similarity methods", herein phenetics, in order to measure nothing more than similarity. No evolution is (or can be) discovered, nor even touched upon. Molecular trees are simply meaningless in the context of evolution. They tell us nothing about sameness and therefore demand no explanation whatsoever. Regardless of this fact, all molecular systematists seem to explain similarity as if it means sameness. A molecular tree is generated and not discovered. They are means unto themselves. The gargantuan task of sequencing, aligning and building trees to find similarity ends with nothing at all. We are by no means poo-pooing similarity methods (phenetics).

Similarity methods are vital for understanding in non-evolutionary fields, such as geology. The chemical composition of rock is important for classification and identification. The same is true for biological keys and other artificial classifications. They helps us identify organisms based on their characteristics. The key will still work if the characters are homologous, not homologous or a mixture of both. Similarity will never be able to show which is which. Phenetics is useful outside of systematics and biogeography and evolutionary biology as its popularity shows. But popularity alone will not validate phenetics, or any similarity, in evolutionary biology.

What do we do with all the data, the matrices and the trees, produced by phenetics? We hope that their owners have fluked it - actually found a meaningful evolutionary, that is a monophyletic group. The sad news is that they'll never know.

Of course, molecular data have meaning and we should not be understood as attempting to trash molecular systematics. We simply feel they have been sold short. Maximum likelihood, parsimony optimization, and so on, are all kinds of phenetics: they are similarity methods. They are useless in the pursuit of evolutionary patterns, namely homology and monophyly. Molecular systematists need to stand up and shake loose the shackles of similarity, realize that their data and their methods are two separate issues and question those that wrongly promote similarity methods as "evolutionary".