Benchmarking in Taxonomy: A Commentary on the Role of the Holotype

Benchmarking in taxonomy is viewed both as establishing a specimen as standard of reference and as a process for optimizing that process. Here it is founded on vision theory that recognition of specimens, as for all objects, is personal to the observer and is based on stored exemplars (benchmark images) in their memory. A special feature of a holotype as a scientific benchmark is that it has been published with a Linnaean name permanently attached. This concept is generalized to an entity that includes all specimens published by subsequent taxonomists with that name attached (a labelled specimen database). As a record of usage it integrates all published images with a Linnaean name. It promotes an inquiry into processes for selection of such specimens. In the conventional model of practice taxonomists categorize specimens using their stored representations of already identified individuals; the process is immediate, acute and autonomous but is largely concealed; a specimen may be selected as a benchmark but its typicality is not revealed. As a remedy, a population model of practice is advocated wherein the basic autonomous visual process is supplemented by objective data about a specimen and the probability of its position within a potential source population.