Nederlandse Taalkunde - Volume 15, Issue 3, 2010

Dutch language users can use connective patterns to express backward causal relations. Sub-omdat patterns (omdat, ‘because1’, followed by a subordinated clause) and co-want patterns (want, ‘because2/since/for’, followed by a coordinated clause) can be used in both spoken and written Dutch. However, only in spoken Dutch, a third pattern might be used: the co-omdat pattern. A possible parallel phenomenon occurs in German. We will describe the co-omdat pattern against the background of a comparison of this so called “epistemic weil” pattern and present a first study on the co-omdat phenomenon. We analyse the co-omdat pattern from two linguistic perspectives: cognitive text analysis and spoken discourse analysis. We argue that the congruency seen in the resulting two analyses of co-omdat can be taken as strong support for the acknowledgement of a co-omdat construction in spoken Dutch.

Most current views on grammaticalisation consider reanalysis to be the basic mechanism underlying grammaticalisation. A comparison of subject doubling data from West Germanic and Romance varieties shows that this view faces problems. The grammaticalisation of pronouns into clitics and agreement markers is less likely to proceed into the final stages in West Germanic than in Romance dialects. This difference can be related to the fact that some of the obvious analogical extensions for the use of clitics in Romance are not or to a lesser extent available in West Germanic. Hence, the grammaticalisation pathway of a certain pattern is crucially determined by the pattern’s potential for paradigmatic extensions rather than by its liability to reanalysis.