Most modal particles (MP) (e.g., G ja, F quand même, AmE like) have a non-modal counterpart (NMCP) as e.g., a sentence word, a conjunction or a preposition, often referred to as "homonym": ja 'yes', quand même 'even though', like 'similar to'. Recent years have seen an increasing number of studies investigating the historical development of these NMCPs into MPs (Abraham 1991, Autenrieth 1999, Diewald 1999, Meibauer 1994, Wegener 1998). The mainstream view seems to be that the NMCP developed a variant whose semantics progressively impoverished over the centuries, yielding the semantics of the MP as an abstract semantic core. Today's functional differences between the modal and the non-modal variants are then claimed to derive from the specification of this semantic core in different syntactic, textual, or pragmatic domains. I would like to challenge this view and make the following alternative claims:

The presumed common semantic core of MP and NMCP (e.g., "adversative relation" for G aber as a MP and 'but' as its NMCP, cf. Diewald 1999) is, in most cases, much too vague. The fine-grained semantic differences between MPs and their intricate distributional restrictions cannot be accounted for. Instead, I claim that the MP arises through metonymic semantic change (more precisely: delocutive semantic change) from its NMCP, yielding polysemy. The function of the MP is to evoke a speech situation that is typical for its use as a NMCP. Furthermore, NMCPs should be likely to be used in holophrastic (one-word) utterances, thereby being enabled to become "typical" for a specific type of speech situation.

For example, the German MP ja presupposes that the speaker assumes the content of the utterance to be obvious to the hearer, i.e. that the hearer agrees with her. As a reply to a question, the NMCP ja 'yes' (which is a holophrastic utterance) creates a speech situation where speaker and hearer agree on the propositional content of the question. Ja as a MP presupposes a speech situation which is created by ja as a NMCP. The two functions are in a metonymic relation, i.e., they are contiguous in a semantic frame. It is not the case that one function is a semantic impoverishment of the other one. Moreover, the relevant semantic frame is speech-related (the question-answer routine), i.e., one is dealing with delocutive semantic change.

An advantage of this model is that the MP need not be semantically more abstract than its NMCP; at the same time, a precise semantic link between MP and NMCP of a recurrent nature can still be maintained. Diachronic data from several languages will be provided as empirical evidence.

References

Abraham, W., 1991. Discourse particles in German: How does their illocutive force come about? In: W. Abraham, ed., Discourse particles. Descriptive and theoretical investigations on the logical, syntactic and pragmatic properties of discourse particles in German. Amsterdam.

Autenrieth, T. 1999. Heterosemie und Grammatikalisierung bei Modalpartikeln. Eine synchrone und diachrone Studie anhand von eben, halt, e(cher)t, einfach, schlicht und glatt. Ph.D. thesis, University of Tübingen.

Diewald, G., 1999. Die Entwicklung der Modalpartikel aber: ein typischer Grammatikalisierungsweg der Modalpartikeln. In: H.-O. Spillmann / I. Warnke, eds., Akten des 32. Linguistischen Kolloquiums in Kassel 1997. Frankfurt/M.

Meibauer, J., 1994. Modaler Kontrast und konzeptuelle Verschiebung. Studien zur Syntax und Semantik deutscher Modalpartikeln. Tübingen.

Wegener, H., 1998. Zur Grammatikalisierung der Modalpartikeln. In: I. Barz / G. Öhlschläger, eds., Zwischen Grammatik und Lexikon. Tübingen.