Marga Reis

Linguistic evidence : empirical, theoretical, and computational perspectives

The renaissance of corpus linguistics and promising developments in experimental linguistic techniques in recent years have led to a remarkable revival of interest in issues of the empirical base of linguistic theory in general, and the status of different kinds of linguistic evidence in particular. By contrasting evidence from different sources with respect to particular research questions we may gain a deeper understanding of the status and quality of the individual types of linguistic evidence on the one hand, and of their mutual relationship and respective weight on the other. The present volume is a collection of papers which address these issues in relation to specific linguistic research problems, thereby helping to establish a better understanding of the nature of linguistic evidence in particularly insightful ways.

What do Wh-imperatives tell us about Wh-movement?

This paper is concerned with two variants of Wh-movement, +Wh-question-movement and topicalization. Drawing on German material, we argue that they crucially differ as to the features of the landing sites: Wh-question movement is movement of +Wh-phrases into A-bar positions marked by +Wh, which assigns clausal scope to the +Wh-phrases; topicalization is movement of XP-phrases into A-bar positions unmarked by +Wh.Since +Wh-phrases are also XP-phrases, our account predicts that +Wh-phrases may undergo not only +Wh-question movement, but also topicalization, i.e., that they can be Wh-moved without the scope effects typical of +Wh-interrogatives. This prediction is borne out by the existence of Wh-imperatives in German, i.e., long distance extractions of ±Wh-phrases into imperative clauses, which we discuss in detail. It is shown that +Wh-imperatives presuppose complements with an initial +Wh-phrase, which is topicalized into the matrix clause, thus showing conclusively that scope assignment is independent of Wh-movement.

Zur Grammatik (vor allem) konditionaler V1-Gefüge im Deutschen

Abstract Standardly, verb-first (V1) conditionals are considered to be mere variants of wenn-conditionals; accordingly, left-peripheral V1-clauses are analyzed as embedded into the prefields of declarative apodosis clauses, just like their V-end counterparts. We challenge this view, proposing instead that dependent V1-clauses are syntactically unembedded/unintegrated, and, consequently, that postposed declarative apodosis clauses are either V2-declaratives with prefield ellipsis or V1-declaratives. We argue our case by presenting evidence that (i) wenn-clauses differ considerably from V1-clauses in semantic distribution, (ii) unlike wenn-clauses, V1-clauses do not meet the criteria for syntactic embedding, (iii) the alternatives entailed by (ii) for the structural analysis of a postposed apodosis both have empirical support. As for a syntactic analysis of V1-structures suited to these findings, we argue that a CP adjunction analysis is currently the best option available. We also point out connections between the semantic restrictions on V1-conditionals and the semantics of V1-interrogatives that are suggestive of a common semantic core, which raises hope that our background vision (ultimately, all dependent V-to-C clauses are semantically licensed substitutes for genuine subordinates) may also be true for V1-clauses.

On the form and interpretation of German Wh-infinitives

This paper is about a minor and so far neglected sentence type in German: wh -root infinitivals such as Wohin gehen? ‘Where (should one/we) go?’. I concentrate on describing their formal and interpretive properties in detail, and relate these properties in a nonarbitrary manner by suggesting ways in which the salient aspects of their interpretation, that is, (the specifics of) their modality, their subject interpretation, and their illocutionary use, could be derived from the interpretively relevant properties of their form. Earlier versions of this paper were presented at the Zentrum fur Allgemeine Sprachwissenschaft, Typologie und Universalienforschung-Berlin (3/02, 5/02), the University of Pennsylvania (4/02), and the University of Frankfurt (6/02). Thanks to all the audiences for valuable comments and suggestions. Special thanks for constructive discussion and help with specific points of the paper are owed to many: Rajesh Bhatt, Veronika Ehrich, Ewald Lang, Manfred Krifka, Ingo Reich, Inger Rosengren, Joachim Sabel, Hubert Truckenbrodt, Ede Zimmermann, Ilse Zimmermann as well as the editor, Mark Louden, and two anonymous referees. A preprint version of this paper (here considerably revised) appeared as Reis 2002.

Is German V-to-C movement really semantically motivated? Some empirical problems

Abstract 1. The central issue for a linguistic theory of sentence types is to what extent their meanings (‘sentence moods’) can be compositionally derived from their form. T[ruckenbrodt]s paper is a valiant, and altogether fascinating attempt to identify a specific contribution of the central formal feature of German sentence types – verb position viz. finite verb movement to C (V-to-C ) – to these meanings. His central claim is that V-to-C is triggered by a context index in C containing crucial sentence mood elements that must be checked in order to yield the appropriate illocutionary interpretations of V-in-C root clauses. In short, according to T, German V-toC has a purely semantic viz. illocutionary motivation. It is this central claim that I will take issue with in the following.

Sustainability of annotated resources in linguistics: A web-platform for exploring, querying, and distributing linguistic corpora and other resources

We report on finished work in a project that is concerned with providing methods, tools, best practice guidelines, and solutions for sustainable linguistic resources. The article discusses several general aspects of sustainability and introduces an approach to normalizing corpus data and metadata records. Moreover, the architecture of the sustainability platform implemented by the authors is described.

Gegen die Kompositionstheorie der Affigierung

In einer Auseinandersetzung mit dem Aufsatz von Tilman N. Höhle (Über Komposition und Derivation: Zur Konstituentenstruktur von Wortbildungsprodukten im Deutschen. In: ZS l (1982), 76-112) vertritt und verteidigt Marga Reis gegen Höhle die These, daß sich die Kompositionstheorie der Affigierung nicht halten läßt, soweit man sie als Wortbildungstheorie versteht. Auch als Wortstrukturtheorie aufgefaßt, führt sie nach Marga Reis zu Schwierigkeiten.

ZUM GRAMMATISCHEN STATUS DER HILFSVERBEN

Hauptverb eines zugrundeliegenden S sein muß.« Die Basiskategorie >Verb< umfaßt somit neben echten Verben auch Adjektive, Kopula und Hilfsverben; [ ± Aux] ist lediglich Subkategorisierungsmerkmal. Im weiteren möchte ich am Material der deutschen PassivVarianten auf einige syntaktische Schwierigkeiten hinweisen, die die weithin akzeptierte >Auxiliaries äs Main Verbs<-These (im folgenden AMV-These) synchron wie diachron mit sich bringt. Das Referat gliedert sich wie folgt: In § l diskutiere ich kurz die für die weitere Argumentation relevanten Regeln EQUI-NP-TILGUNG (imfolgendenEQUI)undSUBJEKT-ZU-SUBJEKT-ANHEBUNG (im folgenden SS A), in § 2 vor allem den problematischen Fall des bekommen-Pa&sivs; in § 3 werden die in § 2 gegen die AMV-These entwickelten Argumente verallgemeinert. Abschließend wird gezeigt, daß auch die herkömmliche und derzeit einzig mögliche alternative Auffassung der >Hilfsverben als Hilfsverben< den Fakten nicht gerecht wird; mögliche Ursachen dieser Aporie werden angedeutet.

On the Form and Interpretation of Echo Wh-Questions

We argue that echo wh-questions are autonomous grammatical structures, and that their distinctive formal properties determine their semantics and pragmatics. Echo wh-sentences contain a wh-phrase in which the wh-element is narrowly focused, and a phrasal Q operator. Wh-phrase and Q operator determine a question semantics. Focus on wh determines the discourse appropriateness conditions for echo wh-questions, namely that an answer to the question be given in the context. A compositional analysis is provided that derives this interpretation.

Kaum-Gefüge im Deutschen - Grammatik und Pragmatik

German kaum constructions have remarkable grammatical and pragmatic properties, which have so far received little attention. This paper aims to provide an in-depth study of these properties, the principal claim being that the core pragmatic properties of all variants of kaum constructions can be derived in a ‘compositional’ rather than ‘constructional’ way. Major descriptive findings are, on the grammatical side, (i) that the clause introduced by kaum is always unintegrated, no matter whether it is clearly a main clause (when followed by an als or und clause) or not (when followed by an asyndetic V2 or “V1” clause), and, on the pragmatic side, (ii) that there is indeed a mismatch between syntactic and communicative ±subordination, (iii) that kaum constructions basically share the ‘narrative’ value of V1 declaratives distinguishing both from basically ‘descriptive’ V2 declaratives, (iv) that the additional ‘dramatizing’ effect these constructions assign to the sequence of events expressed is variable, its quality and strength being almost totally derivable from regular interpretive effects of the lexical and syntactic means typically (but variably) employed in kaum constructions. In conclusion, these results are evaluated with respect to the principal claim cited above, and two promising areas for testing it more widely are pointed out. 0. Worum geht’s? 1. Kaum-Gefüge – Grammatik 1.1 Der Erstsatz 1.2 Der Zweitsatz 1.2.1 Obligatorik 1.2.2 Strukturvariabilität 1.2.3 Stellung 1.3 Gefügestatus und -struktur von kaum-Konstruktionen 1.3.1 kaum-als-Gefüge 1.3.2 kaum-(und)V2-Gefüge 1.3.3 kaum-„V1“-Gefüge 2. Kaum-Gefüge – Pragmatik 2.1 Diskrepanz zwischen syntaktischer und kommunikativer Über-/Unterordnung 2.2 Veranschaulichungs-/Verlebendigungseffekt der kaum-Gefüge 2.2.1 Informationsstruktur von kaum-Gefügen: Affinität zu thetischen Strukturen