Abstract
The term "suppletion", introduced by Osthoff (1899. Vom Suppletivwesen der indogermanischen Sprachen.
Heidelberg: Universitatsbuchdruckerei Horning), was traditionally used to refer to an inflectional paradigm containing forms based on two or more etymologically different stems. In the last decades, however, it has been argued that etymology does not contribute to our understanding of the phenomenon, and it should be strictly defined on synchronic terms: simply as the peak point on the formal irregularity scale, regardless of the actual origin of the irregularity.
Under this approach, all forms reported by speakers as two potentially different lexical items are considered to be suppletive. To be able to determine what users of a living language consider to be a case of suppletion, it is possible to analyze data collected from speakers.
The situation is considerably more difficult for dead languages, which however have played an important role in the debate and provided many of the canonical examples. As a closest equivalent to eliciting the required information from a native speaker, the informed but from the present-day perspective naive expressions of linguistic introspection in the works of Late Latin Grammarians, namely their use of specific terms (defectivum, anomalum, inaequale) to refer to different degrees and lexical examples of irregularity, are highly valuable, as it also may reflect the difficulties confronted by non-native learners.