From the Morphometrics at Stony Brook website
What Comes After Genus and Species?
More than two millennia after Aristotle proposed a hierarchical scale
of being for plants, animals and human beings, and two centuries after
Linnaeus proposed the binomial system that still provides the basic
framework for categorizing living things, scientists continue seeking
after the most neutralmethod for classifying plants and animals in
a biologically meaningful way. As researchers in many fields began
exploiting the capabilities of mainframe computers in the mid-20th
century, Distinguished Professor of Ecology and Evolution emeritus
Robert R. Sokal pioneered the introduction of statistical and computer
methods to make these classifications.
As he recalled almost thirty years later, “one day in 1953, during a casual bag lunch in the laboratory [of the Department of Entomology at the University of Kansas] when the conversation had turned to theory and practice of taxonomy, [I made] the brash claim that I could do a better job of classifying organisms by statistical means than by the traditional subjective approach. This view was challenged and before I knew it Earl Bell had bet me a six-pack of beer (then as now only 3.2% [alcohol] in Kansas) that it could not be done. Charles Michener agreed to furnish data on a group of bees and I started developing approaches initially based on early developments of cluster analysis in psychology. The result was the first modern paper on numerical taxonomy in North America.” (Michener C.D. and Soka1 R.R. “A quantitative approach to a problem in classification,” Evolution 11: 130-62, 1957)
Prof. Sokal soon met Peter Sneath, independently pursuing similar efforts at the Medical Research Council in London, and in 1963 the fruit of their first major joint effort, Principles of Numerical Taxonomy, described the new methodology. To preclude subjective judgments of similarity, they charted quantitative measures of an unprecedentedly large number of properties of biological specimens – characters – giving all of them equal weight, and applied clustering algorithms to the resulting similarity matrices. A widely read 1965 article Sokal published in Evolution with Joseph Camin, “A method for deducing branching sequences in phylogeny,” (Vol. 19, pp. 311-26) developed and tested the application of the new methodology to evolutionary studies by applying it to data based on a phylogeny, or historical tree of hypothetical organisms with a known evolutionary history that changed over time according to a simple evolutionary model.
In 1969, the year Sokal came to Stony Brook, he began convening annual meetings of what became known as the Numerical Taxonomy Group, which an early participant found to be infused with “the sense of a small band of pioneers exploring new territory.” The NT group, which consisted primarily of biologists, inspired the formation of the Classification Society of North America (CSNA), a more diverse organization drawing members from many fields, with a strong representation in psychometrics. Sokal and Sneath also helped found the International Federation of Classification Societies.
Although traditional taxonomists and evolutionary biologists initially resisted the statistical approach, and profound disagreement emerged between partisans of “phenetic” classification, based only on patterns of similarity and dissimilarity among the organisms being classified, versus those advocating “cladistic” classification, based on inductive estimations of the organisms’ evolutionary history – Sokal and Sneath had carefully included both in their own work – the methodological “revolution” is now complete. Because of their applicability to the analysis of any collection of items with similar properties where it is desirable to look at all the properties simultaneously to find relationships among the individual items, the methods of numerical taxonomy are not only accepted but valued in fields as divergent as archaeology, psychology, medicine, economics, and even the humanities, in addition to those that gave them birth.
Prof. Sokal was honored by his peers with honorary memberships in the Society for Systematic Zoology and the Linnaean Society and the presidencies of the Classification Society (for which he was invited to give the 25th anniversary address in 1993), the Society for the Study of Evolution and the American Society of Naturalists. He was appointed a SUNY Distinguished Professor in 1991.
Grateful acknowledgement is made for the information provided by Professor Sokal’s colleague, Distinguished Professor of Ecology and Evolution F. James Rohlf, himself an important founding contributor to this field. Current Contents No. 46, p. 14, Nov. 15, 1982, and “The Troubled Growth of Statistical Phylogenetics,” Joseph Felsenstein, Systems Biology, 50(4):465–467, 2001, were also consulted.