Valency and non‑finiteness

Abstract

Both terms seem to be contradictory to each other, since the classical verb‑centric valence theory, the follower of which is also Ulrich Engel, considers the verb, and, thus, also the formal feature of its finiteness as an inalienable criterion of a sentence proposition. Thus, non‑finite predications are simply considered as minimal utterances, whereas their “normal” status of clauses is mostly denied (cf. Zifonun et al. 1997). On the other hand, many languages ( Slavic languages, among others,but also old Germanic languages, Greek and partly Latin) have many full‑value clause structures being formally non‑finite. Similarly, modern German conveys numerous sentence‑like constructions with a full‑predicative propositional value. Many of them belong to the phenomenon of non‑finiteness (cf., among others, Riecke 2012, Kotin 2014), as in: Was vergangen, kehrt in der gleichen Gestalt nie wieder. If in the description of the valency structure of such sentences/ utterances the classical Engelian valence model is used, no significant differences between finite and non‑finite clauses can be registered, since within this model finite auxiliaries and copulas are considered as not valence‑dependent. However, the problem becomes more complicated, if we assume that copulas and auxiliaries possess at least structural valence and, thus, are regentia in respect to the hierarchically “lower” situated so called “full verbs” (cf. Eroms 2000). In this case, non‑finite sentences have to be considered as elliptical ones and treated accordingly to this status. This very complicated and complex theoretical problem of the dependence grammar is the subject of this paper containing an analysis of examples from various languages and different stages of their development, i.e. from synchronic, diachronic and typological perspectives.