Hoeven diachroon

Abstract

The diachrony of Dutch hoeven ‘need’ This paper presents a diachronic investigation into the semantic and grammatical evolution of the Dutch negative polarity modal auxiliary Hoeven (‘need’). The study is corpus-based, working with representative samples of occurrences of the verb and of its predecessor, the (originally) full verb behoeven (‘need’), from different stages of Dutch, from Old Dutch until today. It shows how, after the initiation of a tendency to auxiliarize in behoeven in Early New Dutch (which largely ends again after that period), the short form Hoeven splits off from the latter in that same time period and takes the lead as the auxiliary variant from then onwards. It also reveals, however, that in the course of New Dutch Hoeven starts to develop new autonomous uses, in a way comparable to what has happened in the other central modal auxiliaries in Dutch (a case of collective degrammaticalization). Moreover, it shows how, from its emergence onwards, this modal auxiliary develops a wide range of modal and related meanings in a very short time, no doubt due to a strive for (semantic) analogy with the other (and much older) central modals.

Directionals and re-autonomization in Dutch modals

“…This use is transitive in all four verbs. This is attested for kunnen and hoeven: there are numerous 'old main V' instances in our present data, or in the data in Nuyts (2013) and Nuyts et al (2018), which also preserve the original meanings 'to know' in kunnen (cf. (6a)) and 'to need' in hoeven.…”

“…Even if it were plausible, this would not change the fact that the process involves degrammaticalization. As argued before (Nuyts, 2013;Byloo and Nuyts, 2014;Nuyts and Byloo, 2015;Nuyts et al, 2018), analogy does play a central role in the evolution of the modals, not only structurally but also semantically, even if in other ways than in the directionals hypothesis. Yet analogy and (de)grammaticalization are not mutually exclusive.…”

“…There is, to some extent, a division of labor between these verbs in terms of negation. As indicated in Section 1, hoeven is a negative polarity item (it occurs in negative contexts in 80% of the instances in END and in 100% in PDD; see also Nuyts et al, 2018). Moeten, on the other hand, occurs less often in negative contexts than kunnen and mogen (no doubt because hoeven occupies the 'negative space').…”

“…Corpus studies into the diachrony of these forms from Old Dutch till the present (Nuyts, 2013;Nuyts et al, 2018Nuyts et al, , 2019 have shown that until Early New Dutch their development is compatible with a regular grammaticalization scenario (e.g. Hopper and Traugott, 2003).…”

“…The ergative variant occurs occasionally in END and ND, but largely disappears in PDD (there is one instance in our PDDW data). The cause for this alternation is to be found in the fact that behoeven 'need', the source of hoeven (see Section 1), was originally a verb with quirky case asignment, with the needer marked dative, and the thing needed marked genitive (there are remains of this pattern in the Early Middle Dutch data in Nuyts et al, 2018). When Dutch lost its case system, the two variants emerged as alternative adjustments of the valency pattern of (be)hoeven to a situation with unmarked, 'regular', subjects and objects (the same happened in many other quirky case marking verbs).…”