Character‐based phylogenetic Linnaean classification: taxa should be both ranked and monophyletic

Abstract

Three primary philosophical positions have emerged among systematists concerning the relationship between phylogenetic analysis and biological classification. The first of these, expressed in a letter to Taxon in February 2005, coordinated by I. Nordal and B. Stedje and signed by 148 additional botanists (Nordal & Stedje, 2005), maintains that paraphyletic should be accepted because of the alleged logical incompatibility between the strict requirement of monophyly and the use of Linnaean ranks, and the claim that with monophyletic taxa results in classifications that are solely on descent at the expense of modification. According to this view, phylogenetic analysis and taxonomy are complete ly separate endeavors that may inform one another but that should not be strictly linked in any necessary way. The second view, held by proponents of phylogenetic nomenclature (e.g., De Queiroz & Gauthier, 1992) as exemplified by the PhyloCode (http://www.ohiou.edu/ phylo code/), is that should be monophyletic by def inition and that the only way to make classification sys tems adequately and satisfactorily phylogenetic is to abandon Linnaean ranks completely. In this view, phylo genetic analysis is used not to test the monophyly of but to reveal their identities and membership, which therefore could only be known with certainty if we could reconstruct with certainty the true tree of all life. The third, intermediate position, the one to which we sub scribe and which we believe is the prevailing view among systematists, states that monophyly and Linnaean ranks are not only compatible but complementary and that should be both ranked according to the so-called Linnaean system and circumscribed so as to reflect puta tive monophyletic groups based on current evidence. In this view, taxonomy and phylogenetic analysis are dis tinct but tightly linked endeavors: each taxon is a human construct whose membership is determined by taxono mists based on character state distributions, and phyloge netic analyses simultaneously provide information about which groups are and are not monophyletic and which character states are and are not homologous. In a twist that we find both ironic and unfortunate, the first two positions described above have come to be considered complementary because they are united in claiming that the criterion of monophyly and the use of Linnaean ranks are incompatible, and at least some mem bers of each of these two camps have suggested that we should have two parallel systems of classification, one that is Linnaean but not strictly phylogenetic and a sec ond that is strictly phylogenetic but not Linnaean. We find these suggestions deeply disturbing and believe that to adopt either or both of these systems would be coun terproductive and detrimental to systematics as a science. It is our view that the so-called obsession with monophyly is part of a historical and intellectual pro gression, in which, as methods to understand and classi fy knowledge of biological diversity?fundamental human endeavors?have improved, those methods have been incorporated into the science of systematics. We are therefore obsessed with monophyly for the same rea sons we are obsessed with explaining diversity by Darwin's theory of evolution by natural selection or with formulating and testing hypotheses according to the sci entific method. To abandon any of these would represent a step backwards in scientific progress. On the other hand, the same can be said about the so-called Linnaean hierarchy and the rules of nomenclature put forward in the ICBN (Greuter & al., 2000), which also represent progress in the science of taxonomy. Moreover, as has been pointed out repeatedly, abandoning the system that has served us well for over 200 years would cause mas sive and completely unnecessary confusion. We further contend that the arguments put forward by proponents of both of the first two positions described above are based on misconceptions, misinterpretations, and misguided logic. The claim that paraphyletic groups are inevitable in the Linnaean system is based on the inappropriate (and impossible) desire to construct classifications from phy logenetic trees rather than cladograms, and the corre sponding assumption that we must classify to higher