Band 37, Heft 3

The following contribution is dedicated to questions about the interface between interactionally based studies on the grammar of everyday language use and “usage-based” approaches from Construction Grammar. Studies of grammatical structures in interactional use make it clear that concepts of grammar which assume homogeneous, static categories are only partly suitable for the description of spoken language, since grammatical structures in communicative practice are more heterogeneous, open and context dependent than postulated. At the same time, these analyses show that prefabricated patterns form an important means of solving communicative tasks precisely in oral communication, which takes place under time pressure. These factors make usage-based approaches from Construction Grammar attractive as a framework for a practically oriented examination of grammar, though many divergences exist all the same. On the basis of empirical analyses of pseudo-clefts and wenn constructions in spoken German, the article discusses the possibility of extending usage-based approaches from Construction Grammar, which aims to integrate the circumstances of language production and reception in communicative practice into the analysis, and therefore the fact that we use language in order to carry out social activities in the process of interaction.

The concept of ‘constructions’ covers the range from fully unanalyzed or “frozen” units to abstract and productive schemas that can be used to form new utterances, even with verbs that usually have a different valency (coercion). The dynamic nature of the construction as well as its functional grounding makes the construction particularly suitable for describing and explaining the course of language acquisition. Distributional analyses of the linguistic structures used across development as well as experimental tests on the productivity of these structures are the methodological means to assess the degree of freedom of constructions used at each point in development. In the earliest stages of development children rely heavily on bottom up processes and generalize only slowly. But with growing productivity they are able to generate new utterances top down.

The notion of constructions—as the basic semiotic units of language varying in complexity, size and degree of abstraction—has proved to be useful in diachronic research in general and in grammaticalization studies in particular. Grammaticalization scenarios have been modelled as a unidirectional succession of specific types of constructions, thus providing a means for depicting the interdependence of pragmatic and contextual feature on the one hand and language internal features on the other. However, the endpoint of grammaticalization processes, i.e. the integration of formerly autonomous signs into tightly organized paradigmatic structures (grammatical paradigms), so far has not been in the scope of constructional approaches. It is argued that existing models of grammaticalization scenarios should be expanded by a fourth stage, the stage of paradigmatic integration. It is suggested that a grammatical paradigm represents a particular type of construction. Thus, the stage of paradigmatic integration represents the formation of a fourth type of construction.

This paper is concerned with the concession-relation on epistemic-level, herein after referred to as ‘concession’. It is shown that there is a spectrum of means to encode concessions in German: the forms vary from rather “idiomatic” constructions to less specified constructions, from lexico-grammatical constructions up to hardly encoded forms, which have to be inferred from discourse context. Furthermore, a method is demonstrated to reconstruct hardly encoded forms of concessions.

If signs of type X regularily cooccur with signs of type Y, this can be explained in two ways: (i) X is required by the valency of Y, or (ii) there is a complex construction Z that requires both X and Y. I argue that, when deciding between (i) and (ii), grammarians should consider, among other things, whether the cooccurence of X and Y resembles prototypical instances of (i). This recommendation is based on the fact that the properties of the prototype of (i) show that valency is a morphosyntactic feature of word forms (like plural). Therefore, if the cooccurence of X and Y does not resemble the prototype of (i), treating it as an instance of (i) would establish a wrong parallel to morphosyntactic phenomena. And, on the other hand, if it does resemble this prototype, treating it as an instance of (ii) would miss this parallel. – In the perspective of this proposal, I discuss some phenomena whose status between (i) and (ii) is controversial.

Shifts in a word's valency (such as increases or reductions, e.g. of a verb) are alterations to a core valency. One precondition of valency shift is that the verb provides semantic and conceptual opportunities for this shift. Another precondition is that, in a given language, constructions have already evolved in which these shifts can be actualized. Valency shift is construction inheritance: a given verb inherits a construction for which it had not previously been licensed from another verb that is licensed for this construction. Valency inheritance is the inheritance of a verb's valency by a noun construction. Valency theory and construction grammar are therefore complementary. Our concept of shift and inheritance of valencies and constructions is prototypical and analogical. In this context, the paper discusses identity, polysemy and homonymy of constructions, using the nominative-accusative construction as an example.

Using the much debated German particle verb construction as an example, some basic questions of Construction Grammar are investigated: constructional meanings, compositionality of meaning, the role of lexical verb bases, morphological and/or syntactal analysis. It is argued that the traditional mode of presentation in historical-comparative grammar (as practiced by Neogrammarians like Hermann Paul) is “constructional” in spirit and mirrors empirical concern for constructional change and constructional meaning. Prototypical verb particles are analyzed within a pattern of grammaticalization and lexicalization of adverbial relators stemming from deictic “imperatives” comparable with English elements like Down!, Up!, Away!, Out!. These elements are the sources for the evolution of prepositions and copredicative adverbials used to build patterns of “phrasal verbs” and particle constructions.

Theories of grammar typically have some way of accommodating fixed or semi-fixed idiomatic expressions in addition to expressions derived compositionally by general abstract rules. Construction Grammar differs from most other theories in that it takes such idiomatic expressions as a model for the grammar of natural languages as a whole. Linguistic structures at all levels of complexity and schematicity are uniformly modeled as “constructions”, i.e. form-meaning pairs with unpredictable formal or semantic properties. This view of grammar is plausible only to the extent that abstract and complex linguistic structures can be shown to constitute such form-meaning pairs. In this paper, I present an analysis of the so-called “modal infinitives” in German (structures consisting of haben ‘have’ or sein ‘be’ and an infinitive with zu ‘to’). I argue that German modal infinitives can be described adequatlyadequately only if we assume that they are constructions in the sense of Construction Grammar. While their general function of encoding possibility and obligation can be predicted from the meaning of haben / sein and the infinitival verb phrase, at least one of their more specific conventionalized meanings, “obligation without alternatives”, is more specific than anything that could be derived compositionally from these component parts. Since modal infinitives are part of the “core grammar” of German, this analysis strengthens the case for an analysis of natural language grammars in terms of constructions.

This article analyzes, for the first time, the lexicographic construction of gender in more recent editions of German dictionaries (from 1980 onwards). The article starts by presenting a survey of studies on this topic in other languages (French, Swedish, English). Then the dictionary entries of Frau ‘woman’ and Mann ‘man’ are scrutinized with respect to their definition(s), their structure, and their use in sentences of exemplification. The study shows that a surprisingly high degree of stereotypes still exists, including grammatical differences such as men occurring more frequently in subject positions and women dominating in object positions. While some dictionaries represent clear progress with respect to the representation of gender over time, others are dominated by androcentric attitudes.