  文档下载格式:pdf 上传:2022-6-21 22:18 阅读:9次 页数:524页 大小:13.86 MB|预览The Germanic Languages - Wayne Harbert.pdf内容摘要: This page intentionally left blank
THE GERMANIC LANGUAGES
Germanic – one of the largest subgroups of the Indo-European language
family – comprises 37 languages with an estimated 470 million speakers
worldwide. This book presents a comparative linguistic survey of the full
range of Germanic languages, both ancient and modern, including major
world languages such as English and German (West Germanic), the Scandi-
navian (North Germanic) languages, and the extinct East Germanic lan-
guages. Unlike previous studies, it does not take a chronological or a
language-by-language approach, organized instead around linguistic con-
structions and subsystems. Considering dialects alongside standard varieties,
it provides a detailed account of topics such as case, word formation, sound
systems, vowel length, syllable structure, the noun phrase, the verb phrase,
the expression of tense and mood, and the syntax of the clause. Authoritative
and comprehensive, this much-needed survey will be welcomed by scholars
and students of the Germanic languages, as well as linguists across the many
branches of the field.
WAYNE HARBERT is Professor and Director of Undergraduate Studies in the
Department of Linguistics, Cornell University. He has published extensively
on syntactic topics, with a particular emphasis on historical syntax. His work
on Germanic languages covers a wide range of problems in historical Ger-
manic syntax and phonology, drawing on data from Gothic, Old English, Old
High German, Old Saxon, Old Norse, and Modern German.
CAMBRIDGE LANGUAGE SURVEYS
General editors
P. Austin (SOAS, London)
J. Bresnan (Stanford University)
B. Comrie (Max Planck Institute for Evolutionary Anthropology, Leipzig)
S. Crain (University of Maryland)
W. Dressler (University of Vienna)
C. Ewen (University of Leiden)
R. Lass (University of Cape Town)
D. Lightfoot (University of Maryland)
K. Rice (University of Toronto)
I. Roberts (University of Cambridge)
S. Romaine (University of Oxford)
N. V. Smith (University College, London)
This series offers general accounts of the major language families of the
world, with volumes organized either on a purely genetic basis or on a
geographical basis, whichever yields the most convenient and intelligible
grouping in each case. Each volume compares and contrasts the typological
features of the languages it deals with. It also treats the relevant genetic
relationships, historical development, and sociolinguistic issues arising from
their role and use in the world today. The books are intended for linguists
from undergraduate level upwards, but no special knowledge of the languages
under consideration is assumed. Volumes such as those on Australia and the
Amazon Basin are also of wider relevance, as the future of the languages and
their speakers raises important social and political issues.
Volumes already published include
Chinese Jerry Norman
The Languages of Japan Masayoshi Shibatani
Pidgins and Creoles (Volume I: Theory and Structure; Volume II: Reference
Survey) John A. Holm
The Indo-Aryan Languages Colin Masica
The Celtic Languages edited by Donald MacAulay
The Romance Languages Rebecca Posner
The Amazonian Languages edited by R. M. W. Dixon and
Alexandra Y. Aikhenvald
The Languages of Native North America Marianne Mithun
The Korean Language Ho-Him Sohn
Australian Languages R. M. W. Dixon
The Dravidian Languages Bhadriraju Krishnamurti
The Languages of the Andes Willem Adelaar with Pieter Muysken
The Slavic Languages Roland Sussex and Paul Cubberley
The Germanic Languages Wayne Harbert
THE GERMANIC
LANGUAGES
WAYNE HARBERT
CAMBRIDGE UNIVERSITY PRESS
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo
Cambridge University Press
The Edinburgh Building, Cambridge CB2 8RU, UK
Published in the United States of America by Cambridge University Press, New York
Information on this title: /9780521808255
© Wayne Harbert 2007
This publication is in copyright. Subject to statutory exception and to the provision of
relevant collective licensing agreements, no reproduction of any part may take place
without the written permission of Cambridge University Press.
First published in print format 2006
ISBN-13 978-0-511-26885-4 eBook (EBL)
ISBN-10 0-511-26885-8 eBook (EBL)
ISBN-13 978-0-521-80825-5 hardback
ISBN-10 0-521-80825-1 hardback
ISBN-13 978-0-521-01511-0 paperback
ISBN-10 0-521-01511-1 paperback
Cambridge University Press has no responsibility for the persistence or accuracy of urls
for external or third-party internet websites referred to in this publication, and does not
guarantee that any content on such websites is, or will remain, accurate or appropriate.
to Myrlyn, Diane, and Jenny
CONTENTS
338806
List of tables x
List of language abbreviations xi
Acknowledgments xii
1 Introduction 1
1.1 Some remarks on the organization of this volume 1
1.2 Divergence and convergence in the Germanic languages 6
1.2.1 Germanic languages and Standard Average European 9
1.2.2 Typological classification 12
1.3 A survey of the Germanic languages 13
1.3.1 East Germanic 14
1.3.2 West Germanic 15
1.3.3 The North Sea Coast languages 17
1.3.4 North Germanic 19
2 The Germanic lexicon 21
2.1 Loanwords 22
2.2 Derivation 26
2.2.1 Compounding 29
2.3 Discourse particles 32
2.4 Phrasal verbs 36
3 The sound systems of Germanic: inventories, alternations and
structures 41
3.1 Segmental inventories and alternations 41
3.1.1 The obstruents: place and manner of articulation 41
3.1.2 The sonorants 53
3.1.3 The vowels 56
3.2 The suprasegmental phonology of the Germanic languages 65
3.2.1 Syllable structure and sonority 65
3.2.2 Length 74
3.2.3 Lexical stress and the foot in Germanic phonology 79
3.2.4 Word tone in Germanic 84
vii
viii Contents
4 The Germanic nominal system: paradigmatic and syntagmatic
variation 89
4.1 Nominal inflection 89
4.1.1 Historical prelude: Indo-European heritage and Germanic
innovation 89
4.1.2 Inflectional categories of the noun 91
4.2 The internal structure of the nominal phrase 122
4.2.1 Nominal phrases without nominal heads 122
4.2.2 Nominal phrases with pronoun heads 124
4.2.3 Nominal phrases headed by lexical nouns 126
4.3 Determiners 137
4.3.1 Weak quantifiers 137
4.3.2 Definite articles 141
4.4 Genitive phrases 148
4.4.1 Genitive phrases and determiners 150
4.4.2 Proper name possessors and pronominal possessors 155
4.4.3 The English ‘double genitive’ construction 158
4.4.4 The prenominal periphrastic possessive (Jan se boek)
construction 158
4.4.5 Special developments of the genitive in
possessive constructions 161
4.4.6 Meronymic constructions 164
4.5 Predeterminers 167
4.6 Discontinuous nominal phrases 168
4.7 Adjective phrases 171
4.7.1 Comparison 174
4.8 Pronouns in the Germanic languages 175
4.8.1 Personal pronouns 175
4.8.2 The referential properties of pronouns 196
4.9 The external syntax of noun phrases: subjects 214
4.9.1 Subject agreement 214
4.9.2 The typology and distribution of “expletive”
arguments 224
4.9.3 Subjects in imperative clauses 236
4.9.4 Derived subjects and the syntax of voice 237
4.9.5 Raising constructions 256
5 The verbal systems of Germanic: paradigmatic and
syntagmatic comparison 270
5.1 Historical prelude: the Indo-European heritage and
Germanic innovations 270
5.1.1 Types of verbal inflection 270
5.1.2 Categories of verbal inflection: the Germanic tense/mood
system and its Indo-European antecedents 272
5.2 Modal auxiliaries 285
5.3 Developments in the expression of tense in Germanic 292
5.3.1 Identifying periphrastic tense/aspect constructions 293
5.3.2 Future 297
Contents ix
5.3.3 Perfect tenses and past tenses 301
5.3.4 The meaning of the (present) perfect and the past 307
5.3.5 Progressive 315
5.4 Voice inflections and voice auxiliaries 316
5.4.1 Passive 317
5.4.2 Middle voice 322
5.4.3 The Scandinavian s-passive 327
5.5 Nonfinite verbal forms 329
5.5.1 The infinitive 329
5.5.2 The present participle and the English gerund 341
5.5.3 The past participle and the supine 345
5.6 Verbal valency 347
5.7 Head, complement and adjunct placement in the verb phrase 350
5.7.1 The relative order of the verb and its
complements within the VP 353
5.7.2 The order of objects 362
5.8 Phrasal verbs 366
6 The syntax of the clause 369
6.1 Sentence adverbs 370
6.2 The syntax of negation 376
6.2.1 Some definitions: negation, scope and polarity 376
6.2.2 Scope, polarity and syntactic position 377
6.2.3 Polarity items, negative concord and multiple
negation 379
6.2.4 Negative complementizers and pleonastic negation 382
6.2.5 Constituent negation with sentential scope 383
6.2.6 On formal differences between sentential negators and
constituent negators 392
6.2.7 On the typology of sentential negation in Germanic 394
6.3 The syntax of the left periphery: topics, verb-second and
subject/verb inversion 398
6.3.1 Main-clause/subordinate clause asymmetries in
verb position 400
6.3.2 The triggers of verb-second order and
the typology of V-2 404
6.4 Complementizers 415
6.4.1 That-clauses 415
6.4.2 Infinitive complements 417
6.5 Relative clauses, questions and other fronting constructions 420
6.5.1 A typology of Germanic relative constructions 420
6.5.2 Questions 473
6.5.3 Topic constructions and left dislocation constructions 478
References 482
Index 505
TABLES
4.1 Adjective endings in selected Germanic languages 99
4.2 Definite articles/demonstrative adjectives in selected Germanic
languages 106
4.3 Second person pronouns in selected Germanic languages 193
5.1 Inflectional endings of active strong verbs in selected Germanic
languages 285
x
LANGUAGE ABBREVIATIONS
AFAfrikaansMEMiddle English
BNBokmål NorwegianMHG Middle High German
DADanish NFNorth Frisian
DUDutchNNNynorsk Norwegian
EFEast Frisian NONorwegian
EME Early Middle English OEOld English
ENEnglishOHG Old High German
FAFaroeseONOld Norse
FRFrisianOSOld Saxon
GEGerman PGPennsylvania German
GMC Germanic SAE Standard Average European
GOGothic SWSwedish
ICIcelandicWFWest Frisian
IEIndo-EuropeanYIYiddish
xi
ACKNOWLEDGMENTS
I would like to thank Abby Cohn, Sarah Fagan, Roger Lass and Paul Roberge for their
willingness to read and comment on portions of this work. It has benefitted substan-
tially from their attention and advice. Thanks are also due to Jessica Bauman for her
valuable assistance in the early stages, and to Steve Barganski for his thorough and
expert copy-editing. I owe a special debt of gratitude to Diane Harbert, who, in
addition to her constant support throughout this long project, devoted many hours to
proofreading the manuscript, catching a multitude of errors in the process. I gladly
accept responsibility for the probably numerous errors which remain.
xii
1
Introduction
1.1 Some remarks on the organization of this volume
No single volume can adequately address a topic area as broad as “The Germanic
Languages” in all of its aspects. It is necessary to single out a particular dimension on
which to focus. Languages can be looked at in their societal context, for example, with
attention to such questions as their use and significance in the communities of speakers
who employ them, their relationship with the associated cultures (including, for
example, literary uses), their demographics and their variation along geographical
and demographical dimensions. One can alternatively regard language from a histor-
ical perspective, as chronological sequences of divergences and convergences, states
and transitions. Each of these points of view has provided the organizational frame-
work for successful volumes on the subject. It is also possible, abstracting away from
their social, geographical, cultural and temporal contexts, to examine the languages of
the family as assemblages of grammatical units, rule systems and constructions. This is
the perspective which I will adopt here. The present volume is aimed primarily at those
who are interested in how the Germanic languages are put together – what they have in
common in terms of their linguistic organization and how they differ from each other
structurally. That choice in turn determines several other features of the organization
of the volume. In particular, I will not adopt the standard and often successful
approach of covering the territory by means of a series of self-contained descriptions
of individual languages. That encyclopedic approach is an ideal format for describing
languages in their socio-cultural setting, since the demographic, historical, cultural and
geopolitical situation of every language is unique. When the focus is on the grammat-
ical structures, patterns and inventories of the languages, though, such an organiza-
tional model becomes less ideal. For one thing, it necessarily leads to a large amount of
repetition. The Germanic languages are, after all, more alike than they are different,
and this becomes increasingly true the farther one descends the genetic tree. Once one
has read about the structure of the noun phrase in Swedish, for example, a description
of the noun phrase in Danish will present few surprises. Such a format is also not
1
2 1 Introduction
conducive to side-by-side comparison of the ways in which the languages accomplish
particular tasks, and so does not present a ready picture of structural commonalities
and differences across the family. I have therefore decided to organize the discussion
according to linguistic constructions and subsystems, rather than by languages. For
example, there is a section on vowel systems, a section on the expression of future
tense, and a section on relative clause formation, in each of which an individual
Germanic language may be mentioned or not, depending on whether it offers some-
thing of particular interest in connection with the grammatical phenomenon under
investigation. These decisions will no doubt make the volume less useful for readers
with certain purposes. In particular, since it does not include chapters on individual
Germanic languages, it does not provide a sense of how the grammars of individual
languages work as integrated systems. Fortunately, there are other volumes suited to
the interests of readers who want to inform themselves about the shape of individual
Germanic languages. König and van der Auwera 1994 is particularly to be recom-
mended. There are also volumes which approach these languages from a historical
perspective – most notably, the recent volume by Howell, Roberge and Salmons
(forthcoming). It is hoped that what is lost in the present treatment in terms of coherent
pictures of individual languages is compensated for by a clearer family portrait.
A further practical consideration in favor of the present format is that it allows us to
sidestep the thorny question of how many Germanic languages there are, and which
varieties to devote chapters to. In volumes on language families in which the main aim
is the exhaustive description of particular languages it is usual to single out a particular
variety of each language as the object of that description. Most often, the written
standard variety is chosen (even though linguists recognize the privileged position of
standard languages to be largely a matter of historical accident), and nonstandard
dialects are given relatively short shrift. Such an a priori limitation would simply not
work in a study in which the main focus is the range of grammatical phenomena found
in the Germanic languages, since, as we will be seeing, the family abounds in highly
interesting and sometimes widespread linguistic developments which happen only to
be found in nonstandard varieties. The standard languages show a relatively high
degree of homogeneity, in part the result of their centuries of contact with each other
and other Western European standard languages as languages of high culture and
literature. The range of structural variation among these varieties is thus relatively
small in comparison with that found when nonstandard varieties are taken into
account. The division of the territory into, for example, a chapter on Dutch (repre-
sented by standard Dutch) and a chapter on German (represented by standard German)
is arbitrary not only because of the substantial variation that exists within the individ-
ual languages, but also because of the famous fuzziness of the boundaries between
languages in some cases. The Germanic languages include two notable dialect
1.1 Some remarks on the organization of this volume 3
continua – the West Germanic dialect continuum, encompassing Belgium, the
Netherlands, Germany, Switzerland and Austria, and the Scandinavian dialect con-
tinuum, encompassing Denmark, Sweden, Norway and parts of Finland – in which it is
impossible to draw non-arbitrary language boundaries between neighboring varieties
at any point (see Crystal 1987: 25).
There are also some more theoretical reasons for adopting the construction-by-
construction approach followed here. While I have attempted as much as possible to
keep linguistic theory in the background, this book is very much informed by the spirit
of recent “principles and parameters” approaches to linguistic variation – the idea that
languages are not free to differ from each other arbitrarily and without limit, but rather
that linguistic variation is constrained by general parameters of variation, and that
therefore structural differences across the languages of the family may be expected to
be patterned, rather than random. The construction-by-construction, side-by-side
format of the volume serves to highlight such patterns of variation as are found.
Once the decision was made to organize the presentation around patterns, paradigms
and constructions, rather than around languages, no principled reason remained for
including only the modern members of the family. If the book were an examination of
languages in context, partitioning the Germanic languages according to the salient
demographic property of having or lacking native speakers might make sense. Once
the focus is on structure, though, separating them on the basis of this criterion seems
plainly more arbitrary, since few structural differences correlate with this distinction.
Whether a particular variety belongs to early Germanic or modern Germanic is not
entirely unrelated to its structural characteristics; there are several features which unite
the postmedieval members of the family, and distinguish them from the early medieval
varieties. Some such differences arise by virtue of the fact that the later languages, but
not the earlier ones, were around to participate in pan-European diffusions of such
features as the distinction between formal and familiar forms of second person pronouns
and indefinite articles. Others arise because the later languages, but not the earlier ones,
participated in late parallel developments such as open syllable lengthening or the rise of
medial negators. More often, though, the linguistic features which turn out to have
general predictive value – whether a language is O(bject) V(erb) or VO, for example,
or whether it has lexically case-marked arguments or not – are ones which crosscut the
early Germanic/modern Germanic distinction. Therefore, it was decided for present
purposes to treat both the pre-modern members of the family and the living members of
the family on a par, to the extent possible – as different variants on a common theme. In
treating Gothic and Old High German, for example, side-by-side with Afrikaans and
Faroese, this volume differs from most other treatments of the Germanic languages.
The descriptions offered here are not theory-neutral; I doubt that it is possible to do
linguistic description in a truly theory-neutral way. My particular training and
4 1 Introduction
inclination as a syntactician (working within the Government-Binding/Principles and
Parameters tradition) necessarily informs my descriptions of particular phenomena, the
kinds of explanations I offer for instances of variation (in which the assumption of
parameters will play a prominent role, for example), and even, to an extent, the kinds
of structures and phenomena singled out as worthy of description and explanation, as
well as those left out of consideration. A scholar with a different theoretical orientation
would imaginably have produced a somewhat different work. Nonetheless, an effort
has been made to keep theoretical assumptions in the background in order to make the
descriptions accessible to all readers with a background in linguistics, and to deploy
theory-specific terminology only when it substantially contributes to the efficiency of
the description.
The goals of the work are fundamentally synchronic: to identify and describe
structural similarities and differences across the Germanic family. Nonetheless, it will
be seen that discussion of linguistic history intrudes with some frequency. There are
various reasons for this. For one thing, many of the accounts offered for the distribu-
tion of features across these languages are typological in nature. Many claims are made
of the following sort: languages in a subgroup of the Germanic languages share a
feature Y because they share a linked property Z from which the presence of
Y follows. The validity of such typological linkages is supported by showing that they
vary together over time – that when Y arises by historical change, Z appears too.
Second, there are some shared features of the family or subgroups within it whose
appearance and distribution can only be explained in historical terms: features which
exist only because of historical facts of inheritance or borrowing. Some of the more
interesting cases involve differences in the uses to which inherited “junk” (Lass 1988)
are put. See, for example, the discussion of the weak/strong adjective contrast in
Section 4.2.3.3 and the discussion of the development of the reflexive/nonreflexive
possessive distinction in German (Section 4.8.2.1.1).
Discussions of phonology and the lexicon are accorded less space than the discussion
of morphology and syntax. The particular choices made with respect to how much
attention to give to each of these topics reflect, besides space limitations and the particular
interests of the author, the fact that morphosyntactic aspects of the grammar are more
amenable to the systematic contrastive treatment adopted here; their side-by-side
investigation holds out the most promise of helping us to answer one of the central
questions of the volume: In what systematic ways are the Germanic languages alike
and in what ways different? Aside from prosodic phonology, it is difficult to make
typological statements about the sounds or the vocabulary of the Germanic languages
of comparable generality to those possible for the morphosyntax of these languages.
This book does not separate the treatment of morphology (or “accidence”) from the
treatment of syntax in the way that is familiar from most handbooks of Germanic
1.1 Some remarks on the organization of this volume 5
languages. This fact, too, is related to its central goal of systematic, side-by-side
comparison of the Germanic languages; there are numerous cases in which some
languages in the family use inflectional morphology to encode particular structural
relations among elements, while others avail themselves of syntactic means for this
purpose. So, for example, Gothic and some of the Scandinavian languages have
passive affixes – a matter of morphology – but these are functionally equivalent to
the periphrastic passive markers of other languages in the family, which are syntactic
in nature since they are free morphemes. Similarly, some of the languages of the
family exhibit case inflections – a matter of morphology – but the same grammatical
relationships which are encoded by means of these are encoded by means of free
morphemes (in the form of prepositions) in other languages. Treating the two separ-
ately because one is a morphological phenomenon and the other a syntactic phenom-
enon would, of course, obscure the fundamental point of their functional equivalence.
Instead, I have chosen a different organizational scheme, based on lexical classes.
Chapter 4, for example, treats the morphosyntax of nouns and the other lexical
categories (adjectives, determiners and pronouns) with which they are associated,
and the syntax of the phrases in which these categories participate. Chapter 5 is
devoted to verbs and their phrases. Within each of these discussions, there is a
secondary division into a discussion of the paradigmatic properties of these lexical
items followed by a discussion of their syntagmatic properties. Paradigmatic relation-
ships are the relationships obtaining between an expression and other expressions
which are substituted for it in different contexts. Case paradigms, and their prepos-
itional phrase equivalents in languages without case (e.g., the man, to the man, of the
man. . .) are instances of paradigmatic relationships. The syntagmatic relationships of a
linguistic expression are the relationships which hold between it and non-equivalent
expressions with which it is concatenated in forming larger linguistic expressions. The
relationship between a subject and a verb, for example, is of this type. To a certain
extent, this distinction overlaps with that between morphology and syntax, since, for
example, case paradigms are a matter of morphology, and putting together a noun
phrase and a verb phrase to form a sentence is a matter of syntax, but the two are not
entirely isomorphic.
By its nature, a survey of this sort consists largely of reports of previous scholarship.
This work owes a great debt to the centuries-long tradition of description of Germanic
languages, and, in particular, to a recent spate of reference grammars and grammatical
sketches of high quality for individual languages. The reader may find the following to
be of particular interest: Allan et al. 1995; Bandle 2002, 2005; Booij 1995, 2002b;
Braunmüller 1991; Collins and Mees 1996; Donaldson 1981, 1993; Engel 1988;
Haugen 1982; Holmes and Hinchcliffe 1994; Jacobs 2005; Katz 1987; König
and van der Auwera 1994; Kristoffersen 2000; Lass 1994; Lindow et al. 1998;
61 Introduction
Lockwood 1995; Mitchell 1985; Tiersma 1999; Zifonun et al. 1997. However, the
exercise of creating a construction-by-construction comparison of all of the Germanic
languages (and attempting to fill in the considerable gaps in the available descriptions
of the older languages in particular, as required by that exercise) has turned up
occasional patterns and generalizations which had not been observed before.
1.2 Divergence and convergence in the Germanic languages
Germanic (hereafter, GMC) is, in the first order, a genetic concept. The GMC
languages share many properties and constructions by virtue of common ancestry.
Common inheritance is the reason, for example, that they signal inflectional contrasts
by a mixture of suffixation and alternations in root vowel. It is also the reason that they
have only a single inflectional past tense, and do not distinguish between preterite and
imperfect, for example. The GMC languages share the first of these characteristics
with other languages with which they are more remotely related, including the
neighboring Celtic and Romance languages. The second, however, is a GMC innov-
ation, which sets GMC apart from these other branches of Indo-European (IE), the
larger family to which it belongs, including Celtic and Romance. Among the other
distinguishing characteristics of the GMC languages which set them apart from their
IE ancestor are:
the fixing of the accent on the root or first syllable of the word (Section
3.2.3) and the possibly related tendency to reduce final syllables
the incorporation of verbal nouns and verbal adjectives into the verbal
paradigms as infinitives and participles
the reduction of the system of inflectional tenses to a simple contrast
between non-past and past and the possibly related tendency to
introduce new periphrastic constructions for the expression of tense
and aspect
the introduction of a class of weak verbs, with “dental preterites”
the systematization and restructuring of vowel alternations (ablaut) in the
signaling of tense contrasts in the inherited strong verbs (Section 5.1)
the reduction of the IE inventory of moods by conflation of the subjunctive
and optative (5.1.2.2.1)
the reduction of the inherited system of inflections for verbal voice and
the consequent introduction of periphrastic passives (Section 5.4)
the introduction of a strong/weak inflectional distinction in adjectives
(4.2.3.3)
the introduction of definite articles
1.2 Divergence and convergence in the Germanic languages7
the reduction of the IE case system to four core cases (nominative,
accusative, dative, genitive, with occasional survivals of other cases)
(4.1.2.3.1)
the development of a productive class of so-called weak nouns, based on
IE n-stems
the introduction of relative pronouns (6.5.1.1) based on demonstrative
and (secondarily) interrogative pronoun paradigms
the introduction of verb-second word order (Section 6.3).
In some cases, the common inheritance of the GMC languages has taken the form of
an inherited dilemma, to which the individual languages have responded with individ-
ual and original solutions. This is illustrated in the interesting example of the varied
treatment of weak and strong adjective endings, for example, as discussed in 4.2.3.3.
Some of the GMC languages are more closely related than others. The precise
nature of these genetic relationships has historically been a matter of dispute (see
Nielsen 1989 for an overview of early GMC), but a very widely accepted hypothesis is
that GMC first split into a Northwest GMC branch and an East GMC branch (repre-
sented almost solely by Gothic). The differences between the East GMC group and the
Northwest GMC group are partly matters of regional variation. So, for example, IE
final *-ō became -a in Gothic (giba ‘I give’), but -u in Northwest GMC (Old High
German gibu ‘I give’). To some extent, though, they are matters of chronology. Many
differences between East GMC and Northwest GMC reflect the fact that East GMC
separated from the rest of GMC early and was recorded early, and so retains archaic
features lost in the remaining languages (such as passive inflections and reduplicated
verbs), and fails to participate in the later innovations in which those other languages
took part.
Northwest GMC in turn is hypothesized to have split into a North GMC and a West
GMC branch. The existence of a North GMC subgroup is beyond dispute, given the
strong familial resemblance of its member languages to each other; these languages are
the products of a very robust heritage of common innovation in all areas of grammar,
which sets them apart from the rest of GMC, and the resemblances have been further
reinforced by subsequent sustained contacts, with the result that there is still today a
high degree of mutual intelligibility among them. The evidence for a West GMC
genetic subgroup is more problematic, and has been called into question (though see
Voyles 1971). On surer ground is the existence of a strongly innovative subgroup
of West GMC languages, the North Sea Coast, or Ingvaeonic group, consisting of
Anglo-Saxon, Frisian and Old Saxon, which share a number of features to the exclu-
sion of German, their sister West GMC language. Among these features, perhaps the
ones with the greatest systematic significance are the loss of person distinctions in the
81 Introduction
Figure 1.1 The Germanic Family Tree
plural verb, loss of case contrasts in part of the pronominal paradigms, and loss of
GMC reflexive pronouns. Each of these will be discussed in Chapter 4.
The tree shown in Figure 1.1 gives a widely accepted, though not uncontroversial,
picture of the genetic relationships among the GMC languages.
This tree sets forth a hypothesis about genetic relatedness, its branches graphically
representing the order of divergence from a common ancestor (“the tree model”). Such
tree diagrams do not give a complete picture of the interrelationships among them,
though, and must be supplemented by another graphic device, such as the curly
brackets used here. Similarities between languages are not always the result of
common ancestry. Rather, originally separate varieties can converge over time through
borrowing/areal spread of linguistic features across geographical space and linguistic
boundaries (the “wave model”). In addition to shared ancestry, the GMC languages
have remained geographically contiguous, creating the constant possibility of linguis-
tic borrowing, mutual influence, and consequent convergence. For example, note that
there is no single branch of the tree which dominates “German”; the German language
(to the extent that it is a unitary language at all) is the product of centuries of mutual
influence between originally separate West GMC linguistic groups. In a similar way,
the varieties that we label “Low German,” regarded now (in part for political reasons)
as a variety of German, originated as a variant of Ingvaeonic West GMC – Old
Saxon – which originally had more in common with Old English, but which has been
1.2 Divergence and convergence in the Germanic languages9
“Germanized” by successive waves of linguistic influence from the south. For early
GMC, Rosel (1962) and Nielsen (1989) have reconstructed a complex history of
periods of waxing and waning linguistic and cultural affinity between GMC sub-
groups, in order to account for the pattern of shared features. In later GMC, besides
the interactions which gave rise to modern German, we can mention the mutual
influence among the Scandinavian languages (particularly during the period of
Danish hegemony beginning in the fourteenth century and lasting, in the case of
Norwegian and Faroese, into the twentieth century), which resulted in a high degree
of homogeneity at all levels, the possibly profound Norse influence on English
beginning in the Old English period, which has been implicated in many of the features
of Modern English but whose effect on the grammar of English is still awaiting a full
evaluation, and the strong influence of Low German in late medieval times on the
Scandinavian languages during the period of the Hanseatic league. The effect of the
latter appears to been particularly strong in Danish, which in some respects (including
phonology (Section 3.2.2) and syntax (Sections 4.9.4.1.2.2 and 4.9.5.2), for example)
resembles German more than the other Scandinavian languages. In some cases, the
effect of contact has been claimed not to be limited to direct borrowing, but to appear
in grammar simplification/constructional loss, as a result of disrupted transmission of
the language between generations (e.g., Norde 2001: 243; McWhorter 2002).
1.2.1Germanic languages and Standard Average European
Such convergence by diffusion of linguistic features across boundaries is possible even
when the languages in question are not related, or only remotely related. Vennemann,
in a series of papers (Vennemann 2003a,b,c), has hypothesized such external influ-
ences from the very earliest period of GMC prehistory (see also Schrijver 2003). As a
result of such contacts with neighboring languages, the GMC languages in modern
times have been claimed to have become, in greater or lesser degrees, part of a group
of “Standard Average European (SAE) languages,” which share with other languages
of north central Europe (notably Romance languages) a cluster of linguistic construc-
tions to the exclusion of geographically more distant languages on the European
periphery. Haspelmath (1998) discusses the eleven most compelling features of
SAE, though suggesting that there are other, weaker ones. These are:
a.Definite and indefinite articles. This is, in fact, problematic as an SAE
feature. While the GMC languages all exhibit definite articles, at least in
an embryonic form (see Section 4.3), they share these not only with
Romance, but with the Celtic languages, which are not part of the SAE
cluster. On the other hand, while most of the modern GMC languages
101 Introduction
have indefinite articles in common with the Romance languages, these
are demonstrably of late origin. They are not found in the earliest attested
versions of these languages, nor yet in Modern Icelandic.
b. Have-perfects. All of the extensively attested GMC languages except
Gothic have a periphrastic perfect formed with have plus a past participle –
which they share with all of the Romance languages, as well as Czech and
some Balkan languages.
c. All GMC languages, including Gothic, have a periphrastic passive
formed with the past participle plus a verb of being or becoming
(Section 5.4.1). They share this feature exclusively with the Romance
and Slavic languages, according to Haspelmath.
d. Anticausative prominence. Languages make use of various means for
deriving verbs from other verbs while changing their valency. In Gothic,
for example, transitive/causative verbs sometimes involve additional
morphology, relative to their intransitive counterparts (wakan ‘to be
awake’ / wak-j-an ‘to waken someone’), but sometimes intransitives
are morphologically more complex than their transitive/causative coun-
terparts. Included here are inchoatives with the inherited -nan suffix
(gaskaidnan ‘to divorce (intrans)’ skaidan ‘to separate (trans)’) and
the apparently innovated reflexive middle verbs (sik laisjan ‘to learn’ –
literally, ‘to teach oneself’). Haspelmath claims that the derivation of
intransitives from transitives (through the addition of “anticausative”
morphology) is most frequent in German, French, Romanian, Russian,
Greek and Lithuanian, while “causativization” – the derivation of transi-
tives from intransitives – is more usual in neighboring non-SAE languages.
e. Nominative experiencers. In some languages the semantic argument roles
of agent or actor are assigned to nominative subjects, and for semantic
roles other than agent/actor (including roles such as experiencer and
possessor) are represented by non-nominative nominal phrases (see
Section 4.2.1.4.2). Thus, for example, in Scottish Gaelic, I have a book is
expressed as Tha leabhar agam ‘Is a book with-me’, and I like the book
is expressed as Is toil leam an leabhar ‘Is pleasing with-me the book’.
According to Haspelmath, the SAE languages, to a greater extent than
neighboring languages, tend to realize these experiencer and possessor argu-
ments, too, as nominative subjects, as English does. The fit of this feature
with other hypothesized SAE features is quite loose, however. On the one
hand, Basque and Turkish – not SAE languages by other standards – have a
high proportion of nominative experiencers. On the other hand, Icelandic
and Faroese have low ratios, and are thus excluded from the SAE fold.
1.2 Divergence and convergence in the Germanic languages11
f.Dative external possessors. In Romance, continental West GMC, Gothic,
Balto-Slavic, Hungarian, Greek and Armenian, the possessor is expressed
as a dative in certain kinds of possessive constructions (Gothic afmaimait
imma auso ‘he-cut-off him-Dat ear’). This construction is missing in
non-SAE languages, but also in Modern English (see Section 4.4.6 and
McWhorter 2002). Among the IE languages, it is lacking only in English
and Insular Celtic, according to Vennemann (2003b: 356).
g.Many of the SAE languages, including Romance languages, GMC and
Albanian, allow negated nominal phrases to carry the force of sentential
negation (Gothic ni waihts im ‘I am nothing’). The distribution of this
possibility in GMC is explored in Section 6.2.5. It is found in all GMC
languages except Old Saxon. Non-SAE languages tend to require a
sentential negator, typically attached to the verb.
h.Particle comparatives. The SAE languages – including GMC (Section
4.7.1), Romance, Balto-Slavic, Balkan languages, Hungarian, Finnish
and Basque, according to Haspelmath – characteristically make use of
a “comparative particle” which is not a preposition, and thus does not
affect the case of the following nominal (Gothic frijondans wilja seinana
mais þau guþ ‘loving their own will more than God-Acc’). Non-SAE
languages use a variety of different devices.
i.A-and-B conjunction. Most European languages, according to Haspelmath,
including the SAE languages, make use of an A-and-B construction for
conjunction, as opposed to a variety of other devices (e.g., A-and B-and,
A B-and) found elsewhere.
j.Relative clauses. Among the many strategies available for forming rela-
tive clauses (Section 6.5.1.5.1), the use of a strategy involving relative
pronouns which occur at the front of the relative clause and encode the
case number and person features of the relativized argument (Gothic
sunus meins, in þuzei waila galeikada ‘my son, in whom I am well
pleased’) is claimed by Haspelmath to be unique to SAE languages.
k.Verb fronting in polar questions. In GMC, Romance and Slavic lan-
guages and modern Greek, yes/no questions are formed by fronting the
verb (Section 6.3.2), rather than by intonation or the use of a question
particle.
This SAE hypothesis must be regarded with caution for a number of reasons. First, it
is clear that the different hypothesized SAE features do not pick out the same subsets
of languages. Second, they are attributable to widely different time periods. One of
them – the external dative construction – is probably of IE date, as Haspelmath notes.
12 1 Introduction
Others date from different prehistoric eras (Gothic, for example, has a periphrastic
passive but no periphrastic perfect), while at least some others date from postmedieval
times. This is true, for example, of indefinite articles, and the events which led to
Icelandic and Swedish being on opposite sides of the fence with respect to nominative
experiencers. The SAE constructions also sometimes differ in detail in the languages
that share them, in ways which undermine the likelihood of common origin. GMC, for
example, uses ‘become’ as the cardinal passive auxiliary, while Romance uses ‘be’,
and it is not unimaginable that the two could have originated independently, starting
out from copular sentences in which the predicate was a participle functioning as a
stative adjective. Similarly, the core GMC relative pronouns are based on demonstra-
tive pronouns while the relative pronouns of Romance are based on interrogative
pronouns, and the syntactically comparable relative pronouns of ancient Greek appear
to have been relative pronouns from the beginning (Fortson 2004: 130). Thus, we
might be dealing here with three independent responses to the breakdown of
the inherited correlative syntax of IE (see Section 6.5.1), especially in view of the
difficulty of defending on other linguistic or historical evidence the existence of one or
more eras of contact sufficient for diffusion of the construction to have taken place.
Other attempts have been made as well to account for features of subgroups of
GMC, by diffusion of constructions across major language boundaries. Lindstedt
(2000: 371) claims that the loss of simple past tense in some varieties of spoken
German and the extension of present perfect to the expression of (imperfective) past
tense is an areal development, encompassing southern German as well as northern
Italian and spoken French.
1.2.2 Typological classification
In addition to genetic groupings, it is sometimes profitable to group the GMC
languages typologically, according to structural properties – groupings which often
crosscut genetic groupings, as well as each other. For example, it is proposed in
Section 4.9.5 that those GMC languages which have developed V(erb)–O(bject)
syntax have a wider range of “noun-phrase raising” constructions than OV languages,
and in Section 5.7.1.1 it is observed that “verb-raising” is characteristic of the OV
languages of GMC. It also appears that those languages which have lost morphological
case contrasts in the noun phrase, whether VO (English, Swedish, Danish, Norwegian)
or OV (Dutch, West Flemish, Frisian, Afrikaans), also have some properties in
common to the exclusion of GMC languages with richer morphological case (see
Section 5.7.2.1). Such instances are often viewed as the results of implicational
linkages between grammatical properties, of the sort pursued in “principles-and-
parameters” models of linguistic variation. Thus, in addition to sharing features by
1.3 A survey of the Germanic languages 13
virtue of common inheritance, and by virtue of language contact/areal spread, some
features shared by subsets of the GMC languages arguably arise as by-products of
other typological commonalities.
English is in many respects a typological outlier in the GMC family. McWhorter
(2002: 265) claims that “where a subset of GMC languages have departed sharply from
the original GMC ‘typology,’ English never fails to be a member,” and enumerates ten
features with respect to which English deviates from all or virtually all the rest of
GMC. These include the loss of inherent reflexives (in middle voice constructions, for
example), the absence of external possessors (Section 4.4.6), the absence of gender
(Section 4.1.2.2), the loss of GMC prefixes, the absence of a perfect construction with
‘be’, the absence of passives with ‘become’, differences in verb-second syntax
(Section 6.3), the absence of a distinctively singular form of the second person
pronoun (Section 4.8.1.5), and the loss of the indefinite pronoun man. McWhorter
attributes these changes to disruption of intergenerational transmission during the
Danish conquest.
1.3A survey of the Germanic languages
Another dimension along which the GMC languages can be divided is the distinction
between state and non-state languages. The GMC languages, relative to other language
families of the world, include a high proportion of national languages. These include
some varieties of English (hereafter EN), German (GE), Dutch (DU), Swedish (SW),
Danish (DA), Norwegian (NO), Faroese (FA), Icelandic (IC), Afrikaans (AF) and
Luxembourgish, as well as some English-based creoles (Tok Pisin in Papua New Guinea
and Bislama in Vanuatu). Other GMC languages are non-state languages. These include
Yiddish (YI), Pennsylvania German (PG), Schwyzertütsch, East Frisian (EF) and all of
the varieties regarded as nonstandard “dialects” of the state languages (including some
sufficiently remote from the standard variety that they would count as separate
languages under different political circumstances). Intermediate in status is West
Frisian (WF), which is officially the second language of the Netherlands, though
according to Gorter et al. (2001: 111), this status has “only entailed moderate promo-
tion by the state.” Some varieties, such as Limburgisch in the Netherlands and North
Frisian (NF) in Schleswig-Holstein, have official standing as “regional languages.”
This accidental distinction between state and non-state languages is not unconnected
with the internal shape of the language, s 第1页 / 共524页 第2页 / 共524页 第3页 / 共524页 ©版权说明： 本文档由用户提供并上传，收益归属内容提供方，若内容存在侵权，请进行举报或认领！ |
发表于 2022-6-21 22:18:03
