Citation: Sandvik, H. (2005) The history of biological systematics – from convention to science and back again. – Talk given at the XXIVth Annual Meeting of the Willi Hennig Society in Fagernes.
Abstract: Biological systematics consists of a number of epistemologically very different disciplines, the most important of which are: (1) reconstruction of phylogenetic relationships (phylogenetics), (2) assignment of species to higher taxa (classification), and (3) assignment of Linnean categories to selected taxa. Not all systematic schools employed or employ all of those steps. Therefore, the final outcome of systematisation – a system of a given taxon – may or may not mirror the best available estimate of the taxon’s phylogeny. This is because, of the steps involved in erecting a system, only the reconstruction of phylogenetic relationships is a genuinely scientific activity. Accordingly, biological systematics has during the past fifty years witnessed a gradual reduction in importance of some non-scientific activities, viz. classification and Linnean categories. This might be viewed as a victory of science over arbitrariness, or of testability (objectivity) over both judgement (subjectivity) and convention (intersubjectivity). However, more or less parallel to this development, there has been a re-introduction of convention to systematics in certain systematic communities. Regardless of the label attached to this school of thought (transformed cladism, pattern cladism, parsimony-only phylogenetics, or the like), it has in a peculiar way reversed some of the most important achievements of what is sometimes called the Hennigian revolution. According to its proponents, the system has again become a goal in its own rights. This means that one cannot be sure whether or to which degree a specific system (i.e., a cladogram) is an estimate of phylogeny. The same discrepancy between system and phylogeny arose in some earlier, now largely abandoned, systematic schools. However, in those it was caused by the definition of – partially knowingly artificial (i.e., paraphyletic) – classes of species. Today, the practice of deliberately excluding some sources of data has essentially the same effect. To reject data that are regarded to have to do with evolutionary processes because they are "inference" as opposed to "observation", is a revival of the positivist assumption that theory-free observations exist.