In this paper we account for the meaning and use of the three most frequently used connectives expressing forward causal relations in Dutch: daardoor (‘as a result’), daarom (‘that’s why’) and dus (‘so’). In an empirical, corpus-analytical study we have found that this part of the lexicon of connectives cannot be described in terms of distinctions well-known from accounts like `domains', levels or layers of meaning (see Kortman, 1997, for an overview), such as content, epistemic and speech act (Sweetser, 1990) or semantic - pragmatic (Sanders, Spooren & Noordman, 1992).

Alternatively, we have proposed to consider the difference between the connectives in terms of subjectivity, or, more precisely, in terms of the distance between the Speaker and the Subject of Consciousness (SOC) responsible for the causal relation. In such a subjectivity account (see also Degand & Pander Maat, 1999), the three connectives can be characterized as follows. In causal relations expressed by daardoor 'as a result', there is no SOC present because the causality is located outside of this SOC. In relations expressed by daarom 'that's why', there is a certain distance between Speaker and SOC, and in the cases of dus 'so', this distance is small or even absent. We have shown that this approach accounts for the actual distribution of the connectives in newspaper corpora (Pander Maat & Sanders, 2000).

In this paper we will develop this approach further. We have tested our subjectivity theory of the dus-daarom distribution by conducting two experiments in which language users were asked for their intuitions. Pursuing the idea that the difference between dus and daarom is determined by the continuity of the perspective and the amount of identification with the actor or concluder (cf. Kuno, 1987; Langacker, 1990), we have designed an experimental set-up in which this idea could be put to the test. We constructed discourse fragments with sentence pairs that could be connected by dus and daarom. Subjects were asked to state their preference for one of the connectives. We manipulated the ease of identification / distance in fragments like (1) versus (2) below.

1. D Het KNMI heeft voorspeld dat het 10 graden gaat vriezen vandaag.

Ik ga daarom /dus niet mee wandelen.

E The weather-forecaster predicted that there will be 10 degrees of frost.
I will daarom /dus not come for a walk.

2. D Ik heb gehoord dat het 10 graden gaat vriezen vandaag.
   Ik ga daarom /dus niet mee wandelen.

   E I heard that there will be 10 degrees of frost.
   I will daarom /dus not come for a walk

Results and theoretical implications will be discussed.

Degand, Liesbeth & Henk Pander Maat
1999 Scaling Causal Relations in Terms of Speaker Involvement. Paper delivered on the Levels of Interpretation Workshop, Edinburgh, July 6-9 1999.

Kortmann, Bernd
1997 Adverbial Subordination. A typology and history of adverbial subordinators based on European languages. Berlin/New York: Mouton de Gruyter.

Langacker, Ronald W.
1990 Subjectification. Cognitive Linguistics 1: 5-38.

Pander Maat, H. & T. Sanders.
2000 Domains of use and subjectivity. On the distribution of three Dutch causal connectives. In B. Kortmann & E. Couper-Kuhlen (red.) Cause, condition, concession and contrast: Cognitive and discourse perspectives. Amsterdam: Benjamins.

Sanders, Ted J.M., Spooren, Wilbert P.M., & Noordman, Leo G.M.
1992 Toward a taxonomy of coherence relations. Discourse Processes, 15, 1-35.

Sweetser, Eve E.
1990 From Etymology to Pragmatics. Cambridge etc.: Cambridge University Press.

Utrecht institute of Linguistics OTS, Universiteit Utrecht