The Position of Relative Clauses in German

This work focuses on two different aspects concerning the position of German relative clauses: adjacency and discontinuity with the antecedent noun phrase. Adjacency means that the relative clause is located straight after the antecedent, while discontinuity means that the relative clause is located/moved far away from the noun phrase it refers to. Discontinuity between antecedent and relative clause in German is very frequent with respect to other languages and, in a verb second language with OV-type like German, requires necessarily the lexicalization of the Nachfeld of the matrix clause. So, as well as post-verbal sentential complements or adjunction in German, the extraposition of restrictive relative clauses is another typical case of Nachfeld realization in German, which is not compulsory. While adjacency in German is allowed for every relative clause within all three fields of the matrix clause (Vorfeld, Mittelfeld and Nachfeld), discontinuity is subject to restrictions. The aim of this work is to outline these restrictions and to provide an explanation of the grammaticality or ungrammaticality of each case of discontinuity (Mittelfeld-Nachfeld, Vorfeld-Nachfeld and Vorfeld-Mittelfeld). Restrictions regarding this common word order phenomenon in German depend in the first place on the difference between restrictive and non-restrictive relative clauses; for each of the two typologies of relative clauses, every position of relative clause with respect to its antecedent will be taken into account. The first results on discontinuity with respect to these two kinds of subordinate sentences are based on the syntactic evidence that relative clauses in German are assigned two different syntactic derivations. This work will provide evidence of the fact that what has been called Head Raising Analysis (Kayne, 1994) and Matching Analysis (Sauerland, 1998) are actually suitable derivations for the German language, the former for restrictive relative clauses and the latter for non-restrictive relative clauses. Since the two types of relative clauses are assigned two different syntactic derivations, it is easy to assume that they also behave differently with respect to some of the usual syntactic operations as well as extraposition, which is the main focus of this work. One major difference arose by the following analysis is the fact that a restrictive relative clause can be extraposed when the antecedent is located in the first field of the main clause, hence at the beginning. Our hypothesis is that, in order to achieve a complete study of this particular case, we need to research another level of analysis and investigate the background structure of the main clause as well as its influence on an extraposed subordinate clause. Data will show that the focus and background structure of the antecedent is what allows extraposition at long distance. This kind of informational structure will also play a role for some marginal cases of extraposition with the antecedent located in the middle of the main sentence, in particular when there is ambiguity in the semantic reference of the relative clause. This work is mainly devoted to the syntactic aspects and restrictions of relative clauses and the related phenomenon of extraposition, which are topics of chapter one, two and three. Chapter four deals with cases of extraposition where the syntactic analysis is not enough and phonological and discourse-related aspects have to be taken into account. In the conclusion we will summarize the main findings of this work and it will try to present the set or rules for extraposing German relative clauses in a simple manner, which can eventually be implemented for a lesson on German relative clauses to advanced L2 students. More specifically chapter one provides an overview of previous literature on relative clauses in general and on the differences between restrictive and non-restrictive relative clauses. Moreover, it presents different theories of the word order phenomenon of extraposition regarding German relative clauses. Aim of this first chapter is to show that German relative clauses must be analysed independently from English relative clauses and that conclusions on extraposition that have been drawn from previous studies on the basis of English examples are not directly suitable for the German language as well. Chapter one will also bring evidence to the fact that, despite preliminary assumptions on the difference between types of relative clauses, non-restrictive relative clauses are often ignored as if assumption on the behaviour of restrictive relative clauses could tacitly work out also for them. One of our main goal of this work is to demonstrate that this si not the case. Restrictive and non-restrictive relative clauses behave differently because they have different syntactic status. Chapter two describes the different behaviour of relative clauses in German and, in particular, their position with respect to the own antecedent from a syntactic point of view. Data originally collected with a questionnaire shows clear restrictions on the position of relative clauses with respect to discontinuity. First of all we confirm with examples that the lexicalization of the field after the right verbal bracket of the main clause is required in order to allow grammaticality of discontinuity between relative clause and antecedent. Secondly we have a look at single cases of discontinuity. If the relative clause is located in the Nachfeld and the antecedent in the Mittelfeld or in the Vorfeld, restrictions depend on the nature of the relative clause (restrictive and non-restrictive relative clauses). Chapter two and chapter three will deal further with these syntactic differences between restrictive and non-restricitve relative clauses and we will try to provide a suitable derivation for both kinds of clauses which account for their differences with respect to some of the common syntactic operations, relationships and conclusions drawn on previous literature. Descriptive differences between restrictive and non-restrictive are very common but what hasn’t been discussed is the different status of these two syntactic structures. Since the first chapter gave an overview analysis of the major findings and hypothesis on relative clauses with minor comments on the suitability of the hypotheses found in the literature, chapter three will advance a theory on this matter. In particular we claim that restrictive and non-restrictive relative clauses are attached to the main clause in two different ways, as complement or adjunction. Chapter three shows that restrictive relative clauses are complement of the external lexical head D° while non-restrictive relative clauses are adjuncted on a much higher level of the sentence, that is the entire NP or DP of the antecedent. Chapter four investigates the interaction between syntactic and discourse-related restrictions on extraposition. It provides evidence that what has been claimed to be syntactic requirements for discontinuity is not enough to explain all the ungrammatical examples of extraposed relative clauses. Data shows that the discourse functions within the antecedent noun phrase, in the first place of the determiner, play an important role for this phenomenon. This chapter deals at the beginning with the concept of focus and accent in German and then with the focus/background structure of sentences and of potential antecedents NP/DP. This chapter tries to examine the interaction of these discourse related aspects with the data collected on the extraposition of relative clauses.