The progress towards mathematization or, in a broader context, towards an increased “objectivity” is one of the main trends in the development of biological systematics in the past century. It is commonplace to start the history of numerical taxonomy with the works of R. R. Sokal and P. H. A. Sneath that in the 1960s laid the foundations of this school of taxonomy. In this article, I discuss the earliest research program in this field, developed in the 1920s by the (...) Russian entomologist and biometrician Eugen (Evgeniy Sergeevich) Smirnov. The theoretical and methodological grounds of this program are considered based on the published works of Smirnov as well as some archival sources. The influence of Smirnov’s evolutionary (mechano-Lamarckian) convictions on the development of this project of “exact systematics” is analyzed as well as the author’s attempts to establish a novel concept of “mathematical essentialism” in animal taxonomy. The probable causes of the failure of Smirnov’s project are viewed from both externalist and internalist perspectives, including the opposition to the use of quantitative methods in biology by some of the Lysenkoist ideologists in the USSR. A brief comparison of Smirnov’s research program with that developed 40 years later by Sokal and Sneath is provided. (shrink)