The modal preterit in Dutch (compared to German)

It has recently been argued in the linguistics literature that (some) tenses have a core meaning that can either be employed in the construction of temporal meaning or in the construction of modal meaning (Brisard 1989, Langacker 1991, De Mulder & Vetters 2002, a.o.). In line with these findings, I will try to account for the role of the preterit in Dutch, compared to the situation in German. This paper seeks to find out how the modal interpretation is compositionally computed, i.e. how the information encoded by the tense marker interacts with contextual factors to bring about the interpretation intended by the speaker.

In its basic use, the preterit serves to locate a SoA in the past. In other contexts, however, the temporal location is not the default reading[1]. I will argue that the temporal location comes about by means of context rather than verb-inherently. This paper will further specify the way in which the conceptual import of the Dutch preterit relates a state of affairs to the ground, starting from Langacker’s theory of grounding (1991), combined with the framework of mental spaces (Fauconnier 1994).

While the preterit in German clearly is a tense that can occur both in indicative or subjunctive mood (eg. kam vs. käme), in Dutch, it is not easy to decide whether we are dealing with either tense or ‘mood’ when coming across an instance like bleef. Even though in Dutch the only productive mood is the indicative, in some of its uses it has modal (subjunctive) meaning. Except for some old archaic formulas like ware or hatte the subjunctive mood has got lost. As a consequence, the contexts which are typically covered by the subjunctive II in German, are rendered by the indicative in Dutch. This can either be the preterit indicative of zullen (zouden) combined with the infinitive, or the ‘synthetic’ preterit indicative of the main verb. The fact that this is possible at all, must imply that there are at least certain similarities between the subjunctive on the one hand and the ‘modal’ preterit in the respective languages. The analysis I present here will also touch upon the conditions of interchangeability for the construction zou + infinitive and the preterit in Dutch.

(1a) Als hij goed studeerde, kreeg hij een beloning.

(1b) Als hij goed zou studeren, kreeg hij een beloning.

(1c) Als hij goed studeerde, zou hij een beloning krijgen.

(1d) Als hij goed zou studeren, zou hij een beloning krijgen.

Does the twofold verbal structure run parallel with differences in the speaker’s perspective?

Characteristics of the modal preterit

The morphological preterit forms in (1a) (studeerde and kreeg) can be interpreted both as both indicative and ‘modal’. In the former case we are dealing with a description or narration of a State of Affairs (SoA) in the past, whereas in the latter the SoA is hypothetical: the content of the subclause is not settled (yet) with respect to the mental space of the conceptualizer (potential inclusion is still treated as an open issue). Remarkably, the temporal meaning of both uses is opposite: viz. [+ anterior] vs. [- anterior]. To be described is which elements exactly contribute to either characterization of this bifurcation as far as Dutch is concerned.

The preterit in German

The German preterit behaves differently form its Dutch counterpart. Whereas – analogous to English – we can consign a distancing effect to the preterit (cp. Langacker 1991/ ANS), in German this tense cannot be analyzed independent of the mood it combines with. The preterit subjunctive for instance does not realize this property (otherwise it would be a double distancing, both by the subjunctive mood in se and the preterit tense in se), cp. Fabricius-Hansen (1999: 142):

Präteritale Morphologie markiert zwar auch eine Abweichung von der Interpretation, die der Indikativ Präsens im Default-Fall verlangt. […] Der präteritale Konjunktiv ist, dem Beitrag des Konjunktivs entsprechend, mit ‘Wirklichkeitsbezug’ aus der Sicht des aktuellen Sprechers unvereinbar, läßt aber den Zeitparameter ungeändert. Umgekehrt markiert der präteritale Indikativ im Default-Fall eine Verschiebung des Zeitparameters (Vergangenheitsbezug) bei ungeändertem Weltparameter (‘Wirklichkeitsbezug’) und ungeänderter Personenperspektive.

I will elaborate the claim that the tense-and-mood categories in German cannot be considered semantically compositional (i.e. the degree of distance from time, world or personal perspective comes about only in interaction with the category of mood). For Dutch, the relationship between past morphology and past semantics is not one-to-one either[2]. We then may arrive at the right temporal/modal interpretation by combining ‘the grammar of tense’ with available pragmatic information.

References:

- Brisard, Frank (1991) “The problem of temporality: tense and time in cognitive grammar” In: Antwerp papers in linguistics, 67.

- Fabricius-Hansen, Cathrine (1997): “Der Konjunktiv als Problem des Deutschen als Fremdsprache” In: Studien zu Deutsch als Fremdsprache III: Aspekte der Modalität im Deutschen – auch in kontrastiver Sicht. Hrsg. von Friedhelm Debus und Oddleif Leirbukt. Germanistische Linguistik 136. Hildesheim u. a. (Olms). S. 13-36.

- De Mulder, Walter & Carl Vetters (2002) “The French imparfait, determiners and grounding”. In: Frank Brisard (ed.) Grounding: The Epistemic Footing of Deixis and Reference. Berlin, New York: Mouton de Gruyter, p. 113-149.

- Fauconnier, Gilles (1994) Mental Spaces: Aspects of Meaning Construction in Natural Language. Cambridge: Cambridge University Press.

- Haeseryn, W., G. Geerts, K. Romijn, J. de Rooij & M.C. van den Toorn (1997), Algemene Nederlandse Spraakkunst. 2nd, revised edition, Groningen & Deurne: Martinus Nijhoff /Wolters Plantyn. [= ANS]

- Janssen, Theo A. J. M. (1989) "Die Hilfsverben werden (deutsch) und zullen (niederländisch): modal oder temporal. In Werner Abraham and T. Janssen (eds.) Tempus-Aspekt-Modus: Die lexikalischen und grammatischen Formen in den Germanischen Sprachen. Tübingen: Niemeyer, 65-84.

- Langacker, Ronald W. (1991) Foundations of Cognitive Grammar. Vol. II: Descriptive Application. Stanford: University Press.

- Mortelmans, Tanja (1999) Die Modalverben sollen und müssen im heutigen Deutsch unter besonderer Berücksichtigung ihres Status als subjektivierter 'grounding predications'. - Antwerpen : Universiteit Antwerpen.

- Nuyts, Jan (2001) Epistemic Modality, Language and Conceptualization: A Cognitive-Pragmatic Perspective. Philadelphia & Amsterdam: John Benjamins Publishing Company.

[1] Eg. Bleef je morgen maar thuis! (‘If only you’d stay at home tomorrow!’)

[2] If we presume the fitting of morphology and semantics to be a gradable function, it seems weaker for Dutch than for German.