In the past decade, grammaticalization has become a topic which has gained wide interest. Grammaticalization as viewed from a diachronic perspective is hypothesized to be prototypically a unidirectional phenomenon (see, among others, Heine, Claudi & Hünnemeyer 1991:4; Hopper & Traugott 1993:94 and Bybee, Perkins & Pagliuca 1994:13).

In this paper, I want to investigate the historical development of Dutch causal and temporal connectives, in order to find out:

(1) the way in which modern Dutch connectives have developed;

(2) whether this development is unidirectional;

(3) how this (unidirectional) development can be explained.

For the diachronic development of connectives unidirectionality has been claimed before and several clines have been proposed. Kortmann (1997:190-191), for example, shows that there are certain historical tendencies in the semantic change of connectives. His most concrete tendencies are the following:

(i) a. anterior time > cause

b. simultaneity / temporal overlap / co-occurence > contrast or concession

Traugott (1989:35; 1995:46) claims that a characteristic property of grammaticalization is that there is an increase in subjectification, which means that meanings tend to become increasingly based in the speaker’s subjective belief state / attitude toward the proposition. She shows that a process of subjectification took place in the diachronic change of English while. This subjectification theory can also be applied to the changes German weil underwent (see Keller 1995; Günthner 1996) and probably a parallel can be drawn with the development of Dutch terwijl.

Contrary to Traugott (1995:46-47), who has left her original hypothesis that changes go from propositional to textual to expressive meanings, Couper-Kuhlen & Kortmann (2000:6) suggest that the following hierarchy of semantic levels might play a role in the diachronic development of connectives (hierarchy taken from Crevels 2000:320):

(ii) content level > epistemic level > illocutionary level > textual level

A study by Van Loon (1997) suggests that the Dutch causal want has undergone a shift which is in line with this hierarchy.

Using new data about the historical changes of a selection of Dutch connectives, I will discuss whether these and other current theories are just descriptions or that they can really explain the unidirectionality in the diachronic development of connectives.