This chapter reviews a number of issues concerning Scandinavian object shift and object scrambling of the type found in the Germanic OV-languages. It differs from earlier reviews in that it adopts as its null hypothesis that the two phenomena should be given a unified treatment. An important reason for this is that object shift and scrambling are subject to similar effect-on-output conditions. This raises the question why object shift and scrambling behave differently with respect to, e.g., Holmberg’s Generalization. It will be argued that this is due to the fact that object movement is subject to various language-specific, violable constraints.

Introduction

This chapter provides a more general discussion of adpositions and their projections (adpositional phrases). Section 1.1 starts by providing a general characterization of the category of adpositions. Section 1.2 and 1.3 continue by discussing two possible ways of classifying the adpositions. The first classification is based on the relative position of the adpositions with respect to their complement (if any), which results in distinguishing the following four subclasses: prepositions, postpositions, circumpositions and intransitive adpositions/particles. The second classification, on the other hand, is based on the meaning of the adpositions, which results in distinguishing the following three subclasses: spatial, temporal and nonspatial/ temporal adpositions. Section 1.4 concludes with a discussion of a number of borderline cases, that is, elements that resemble adpositions in various respects, but for which it is nevertheless not entirely clear whether they should really be considered adpositions.

Characterization of the category adposition

This section provides a brief characterization of the category of adpositions. Section 1.1.1 discusses some characteristic properties of this category. Section 1.1.2 provides illustrations of the four basic types of adpositions: prepositions, postpositions, circumpositions and intransitive adpositions/particles. Section 1.1.3 concludes with a preliminary discussion of the syntactic uses of adpositional phrases. Since the purpose of the discussion below is to provide the necessary background for the remainder of our discussion on adpositional phrases, it is inevitably sketchy and far from complete; most issues mentioned here will be taken up again and discussed more exhaustively later in this study.

Properties of adpositions

There are several features that distinguish the class of adpositions from the other three main categories of words: verbs, nouns and adjectives. These are discussed in Subsection I. It is nevertheless difficult to design syntactic tests that single out the full set of adpositional phrases, although there are several tests that can be used in order to recognize certain syntactic or semantic subtypes. These will be discussed in Subsection II.

Differences from the other main categories of words

The subsections below discuss several respects in which adpositions differ from verbs, nouns and adjectives.

Verbs (V), adjectives (A), nouns (N) and adpositions (P) constitute the four major word classes. The present volume deals with adpositions and their projections (adpositional phrases or PPs). The general introduction in Chapter 1 provides a survey of the most distinctive syntactic, semantic and morphological characteristics of adpositions. Like the other categories, adpositions can project: in other words, they select complements and can be modified, which will be discussed in Chapter 2 and Chapter 3, respectively. After this, we will have a closer look at the syntactic uses of the adpositional phrases in Chapter 4. This study will be concluded in Chapter 5 with a discussion of the formation of pronominal PPs like er op ‘on it’, which consist of a so-called R-word like er and a preposition, and the phenomenon that these pronominal PPs can be split by extraction of the R-word. We will also address the complicated issue of co-occurrence and conflation of different types of R-words, although, strictly speaking, this goes somewhat beyond the main theme of this work.

General introduction

Dutch is an official language in the Netherlands, Belgium-Flanders, Surinam, Aruba and the Netherlands Antilles. With about 22 million native speakers it is one of the world's greater languages. It is taught and studied at more than 175 universities around the world (source: taalunieversum.org). Furthermore, Dutch is one of the most well-studied living languages; research on it has had a major, and still continuing, impact on the development of formal linguistic theory, and it plays an important role in various other types of linguistic research. It is therefore unfortunate that there is no recent comprehensive scientifically based description of the grammar of Dutch that is accessible to a wider international audience. As a result, much information remains hidden in scientific publications: some information is embedded in theoretical discussions that are mainly of interest for and accessible to certain groups of formal linguists or that are more or less outdated in the light of more recent findings and theoretical developments, some is buried in publications with only a limited distribution, and some is simply inaccessible to large groups of readers given that it is written in Dutch. The series Syntax of Dutch (SoD) aims at filling this gap for syntax.

Main objective

The main objective of SoD is to present a synthesis of currently available syntactic knowledge of Dutch. It gives a comprehensive overview of the relevant research on Dutch that not only presents the findings of earlier approaches to the language, but also includes the results of the formal linguistic research carried out over the last four or five decades that often cannot be found in the existing reference books. It must be emphasized, however, that SoD is primarily concerned with language description and not with linguistic theory; the reader will generally look in vain for critical assessments of theoretical proposals made to account for specific phenomena. Although SoD addresses many of the central issues of current linguistic theory, it does not provide an introduction to current linguistic theory. Readers interested in such an introduction are referred to one of the many existing introductory textbooks, or to handbooks like The Blackwell Companion to Syntax, edited by Martin Everaert & Henk van Riemsdijk, or The Cambridge Handbook of Generative Syntax, edited by Marcel den Dikken.

Introduction

This section is devoted to R-PRONOMINALIZATION and R-EXTRACTION. The first notion refers to pronominalization of the nominal complement of a preposition by means of a so-called R-word, which must precede the preposition. In (3) we give two examples with the R-word er, which will be glossed as there for reasons that will become clear when we discuss the examples in (6).

Pronominal R-words like er will normally be referred to as R-PRONOUNS, although we will also use the notation [+R] pronouns when they are discussed in contrast to [-R] pronouns like hij/hem ‘he/him’. The combination of an R-pronoun and its associate adposition will be referred to as a PRONOMINAL PP.

Table 1, which is based on the classification of the pronouns developed in Section N5.2, shows that all third person, non-anaphoric [-R] pronouns that can be used independently as arguments have a [+R] counterpart. The fact that these [+R] pronouns all contain an /r/ is responsible for their name.

Table 1 shows that the [-R] pronouns can be divided into four main groups by means of the features [±NEUTER] and [±HUMAN]. For the referential, demonstrative and relative pronouns (group A), the gender distinction is the most prominent one, although it must be noted that most nouns that refer to [+HUMAN] entities are [-NEUTER] as well; nouns with the feature constellation [+NEUTER] and [+HUMAN] are typically diminutives (exceptions are the noun meisje ‘girl’, which has the formal characteristics of a diminutive, the noun kind ‘child’, and a number of nouns with a negative connotation like wijf ‘bitch of a woman’ and mens ‘person’). For the interrogative and quantificational pronouns (group B), only the distinction between [+HUMAN] and [-HUMAN] is relevant. As illustrated in (2), and to be discussed more extensively in Section 5.1, [+R] pronouns are typically, but not exclusively, used as alternative realizations of [-HUMAN] pronouns; see Table 2 in Section 5.1.7 for an overview.

Introduction

This section discusses the complementation of adpositional phrases. Sections 2.1-2.3 will show that, in the core case, the complement of an adposition is a noun phrase: adpositional phrases are not common as complements of adpositions, although there is a small set of prepositions that may occur with them; adjectival phrases acting as the complement of an adposition are extremely rare, if possible at all. Section 2.4 will show that complementation by finite and infinitival clauses is readily possible, but normally involves an anticipatory pronominal PP (that is, a PP with an °R-pronoun as its complement); this section will pay special attention to formations like voordat ‘before’, which is analyzed in traditional grammar as a complex subordinator, but which may involve the preposition voor ‘before’ followed by a finite clause introduced by dat ‘that’. Although adpositional and adjectival phrases are not common as complements of adpositional phrases, they do occur as the predicative part of the so-called absolute met-construction, which will be discussed in Section 2.5.

Nominal complements

Complements of adpositions are normally noun phrases. A distinction must be made between noun phrases with a determiner and (singular) bare noun phrases, that is, noun phrases without a determiner. As is to be expected, the first are normally referential in nature; the noun phrase het kantoor ‘the office’ in (1a) just refers to a building, and it is claimed that Jan is working there. The bare noun phrase kantoor in (1a′), on the other hand, does not refer to a specific building, and the PP does not refer to a specific location; instead, it is claimed that Jan has an occupation that in some way is related to the noun: he may be an office or administrative worker. Similarly, (1b) expresses that Jan is located at the office, while (1b′) simply expresses that Jan is at work.

Introduction

Generally speaking, adpositional phrases resist modification with the exception of the spatial and temporal ones. So, whereas the non-spatial/temporal adpositional phrase in (1a) cannot be modified, the spatial and temporal adpositional phrases in (1b&c) can.

Note, however, that this does not hold for focus particles like alleen ‘only’ and juist ‘especially’ in (2a&b), nor for pas ‘only’ in (2c), which can also be used to modify non-temporal/spatial PPs. We will not discuss this in the present chapter because these particles can be construed with all lexical categories (see Barbiers 1995:ch.3), and thus deserve a discussion in their own right.

Section 3.1 discusses modification of spatial adpositional phrases and will show that the hypothesis that spatial adpositions denote a set of vectors is very helpful in describing the modification possibilities. Section 3.2 will show that the description of the modification possibilities of the spatial adpositional phrases can be straightforwardly extended to the temporal ones. Section 3.3 concludes by discussing the rare cases of non-spatial/temporal adpositional phrases that may undergo modification.

Spatial adpositional phrases

This section is concerned with modification of spatial adpositional phrases. It will be shown that the hypothesis that spatial adpositions denote sets of vectors is very helpful in describing the modification possibilities. For an extensive discussion of the vector hypothesis, we refer the reader to Section 1.3.1.1.

Introduction

This section is concerned with the syntactic uses of adpositional phrases shown in Table 1. For all these syntactic uses we will investigate at least the following four syntactic properties: the possibility of °topicalization, °scrambling, °PP-over-V, and °R-extraction.

Adpositional phrases used as arguments

The use of adpositional phrases in positions that are normally occupied by a nominative or an accusative noun phrase is restricted to constructions in which the adpositional phrase acts as the °logical SUBJECT of a complementive: the two examples in (1) illustrate, respectively, a copular and a vinden-construction, in which the locational prepositional phrase onder het bed ‘under the bed’ functions as the SUBJECT of the nominal predicate een leuke plek ‘a nice spot’.

Example (2) provides similar examples in which the directional postpositional

phrase de berg op functions as the SUBJECT of the adjectival predicate gemakkelijker

‘easier’.

It has been argued that in examples like Het is warm in de kamer ‘It is warm in the room’, the PP in de kamer ‘in the room’ also functions as a SUBJECT of the adjectival predicate warm ‘warm’, albeit that the pronoun het ‘it’ is used as an anticipatory pronoun introducing the SUBJECT-PP. Constructions like these are not be discussed here but in Section A6.6.3.

Examples in which an adpositional phrase acts as the SUBJECT of a main verb cannot readily be constructed and generally sound quite forced. Adpositional phrases can, however, readily be used as complements of lexical heads; they frequently occur as complements of verbs, adjectives and nouns, as illustrated in (3).