Tripartito = indimostrabile?

This article deals with a problem of theoretical stemmatics: the impossibility of positively demonstrating the existence of manuscript traditions with more than two branches. The assumption does not imply real criticality in the application of genealogic methods, but can lead to distortions in the analysis of manuscript traditions and in the selectio of variants. The question is addressed with particular reference to the Latin works of the Middle Ages, so far little examined in their complex by scholars. For these works, the incidence of traditions with three or more branches is apparently greater than for the works of classical literature; however, this may depend on a misuse of stemmatic representations, intended as a rough representation of axiological groups of manuscripts, without a strict genealogical analysis of their relationships.