A brief history of the species concept in virology and an opinion on the proposal to introduce Linnaean binomial virus species names

The species concept used in virology is based on the logic of the Linnaean hierarchy, which views a species class as the lowest abstract category that is included in all the higher categories in the classification, such as genera and families. As a result of this class inclusion, the members of a species class are always less numerous than the members of higher classes, which become more numerous as one moves up in the hierarchy. Because species classes always have fewer members than any of the higher classes, logic requires that they need more qualifications for establishing membership than any of the higher classes. This invalidates the claim that a species could be defined by a single property present in all its members. Species were only accepted in virus classification in 1991, because virologists assumed that it would lead to the use of Latin species names, which they rejected. Anglicized binomial species names have been used by virologists for more the 40 years and are popular because they consist of a virus name followed by a genus name that most virologists are familiar with. The ICTV has proposed to introduce a new Latinized virus species binomial nomenclature using the genus name followed by a hard-to-remember Latinized species epithet that bears little resemblance to the name of the virus itself. However, the proposal did not clarify what the advantage is of having to learn hundreds of new unfamiliar virus species names. In 2013, the ICTV changed the definition of a virus species as an abstract class and defined it as a group of physical objects, which induced virologists to believe that a virus species could be defined by a few characteristics of the viral genome. In recent years, thousands of viral sequences have been discovered in metagenomic databases, and the ICTV has suggested that it should be possible to incorporate these sequences in the current ICTV virus classification. Unfortunately, the relational properties of these hypothetical viruses that result from their biological interactions with hosts and vectors remain in the vast majority of cases totally unknown. The absence of this information makes it in fact impossible to incorporate these metagenomic sequences in the current classification of virus species.