Abstract

This article examines two case studies of cognate expressions in English and in French, which have developed partly in the same and partly in different directions. One case is the pair actually: actuellement, the other is the set in fact: en fait/de fait/au fait. Monolingual research on their present-day meanings and the study of their translation paradigms bring to light semantic and pragmatic overlap as well as differences between the members of each set. The study also looks at their historical development and compares the stages the expressions have gone through in the two languages concerned. The diachronic data indicate partially parallel paths of development, with salient divergences in some cases.

The empirical crosslinguistic diachronic study of these two sets has a mainly theoretical aim, i.e., to contribute to a further understanding of the processes of grammaticalization. Building on existing debates on the issue of the rise of discourse markers and the extent to which they instantiate cases of grammaticalization, the article considers the following questions anew in the light of the results of the empirical data: (i) How can we explain that words with the same origin develop pragmatic functions in one language but not in the other, or that they do so at a much later stage? (ii) Does the case for a unidirectional development towards (inter)subjectification stand if we consider the two case studies? (iii) Which criteria for classifying the two cases as examples of grammaticalization are fulfilled?

While the case for crosslinguistic synchronic research on discourse markers has been argued in previous studies (see e.g., Aijmer and Simon-Vandenbergen, Linguistics 41: 1123–1161, 2003, Pragmatic markers in contrast (Studies in Pragmatics 2), Elsevier, 2006; Aijmer et al., Pragmatic markers in translation: A methodological proposal, Elsevier, 2006), and the usefulness of a panchronic crosslinguistic approach of discourse markers has also been recently shown (see especially Hansen and Strudsholm, Linguistics 46: 471–505, 2008), the present article lends further supports to this thesis by showing that it is through comparison of partially parallel processes that the complex issues of actuation of a change and motivation for change become more transparent. Further, the formal differences of the sources of each set at hand as well as the present-day syntactic and pragmatic behavior of the different members favor an approach to lexical and grammatical categories as nondiscrete and point to the need for a revision of the traditional conceptualization of grammar and grammatical classes.