Zur Abfolge im Mittelfeld des Deutschen. Eine methodische Etüde

This article reviews foundational research on the problem of unmarked word order in German. Focussing on the two most influential seminal papers by Lenerz (1977) and Höhle (1982), I argue that their attempts to determine unmarked, normal word order (either by definition or by explication) is flawed by the fact that both authors presuppose a certain empirical data base without offering a complete grammatical analysis of the data in question. I contend that a more comprehensive, linguistically satisfying analysis of these data will in turn presuppose a pre-given notion of unmarked, normal word order, thus making for a circular definition or explication.As is well-known, normal word order interacts with factors like point of view, thematic roles, animacy, and others. I will argue that influential suggestions for designing a precise theory of these interactions are unsuccessful on both methodological and empirical grounds. I suggest that the traditional modular analysis based on cumulation and treshold values is still the best model we have at present; however, a large and hitherto unresolved issue is the vast variety of contradicting acceptability judgments found in the literature. A careful analysis of these meta-data should enable us to determine paradigmatic core cases while, at the same time, leave room for deviations in various directions, and even for individual ad hoc preferences at the periphery.