Taxonomy is caught between the search for the "perfect" theory and an elusive biological variability. The lack of major advances in issues related to how "species" and other taxonomic categories are defined suggests that perhaps we should avoid excessively rigid formalism in this regard. The risk is a separation between elegant but useless theories and confusing applications of the taxonomic tools. Communication is one of the main functions of taxonomy, and stability one of the main parameters that taxonomy users should be sensitive to. An excess of stability may generate anachronistic consequences while continuous revisions may make the tool of taxonomy scarcely practical. The current tendency pushes toward more and more fragmentation of biologically valid taxa. While taxonomy specialists enjoy such challenges, many taxonomy users feel a bit nervous and discouraged when trying to use a tool that is constantly changing. Debates over taxonomy would seem particularly unrewarding for fields with limited samples and scarce biological diversity, such as palaeoanthropology. In this context, where the information available is rarely sufficient to supply consistent taxonomical evidence, there are frequently excessive efforts to create debate on species separations. The risk is that we maintain the debate on a purely theoretical level, or else we distrust a reliable use of taxonomy. A compromise (and recommended) choice between these two extremes would be to rely on shared and reasonable interpretations of homogeneous evolutionary units, without diving into fine-grained issues that will remain, however, unresolved. Taxonomy should be a tool, not the goal, of the evolutionary biologist. Our mind needs discrete and recognizable objects to structure our perception of reality. There is no reason to expect that nature works the same way.