. 1
( 12)


This page intentionally left blank
A Linguistic Introduction

Standard German is spoken by approximately 95 million people worldwide.
This book provides an introduction to the linguistic structure of Standard
German that is rich in descriptive detail and grounded in modern linguistic
theory. It describes the main linguistic features: the sounds, structure, and
formation of words, structure of sentences, and meaning of words and sen-
tences. It surveys the history of the language, the major dialects, and German
in Austria and Switzerland, as well as sociolinguistic issues such as style,
language and gender, youth language, and English in¬‚uence on German.
Prior knowledge of German is not required, as glosses and translations of the
German examples are provided. Each chapter includes exercises designed to
give the reader practical experience in analyzing the language. The book is
an essential learning tool for undergraduate and graduate students in German
and linguistics.

sarah m. b. fagan is Professor in the Department of German at the
University of Iowa. Her recent publications include Using German Vocab-
ulary (Cambridge, 2004).
Linguistic Introductions available from Cambridge University Press:

Romani: A Linguistic Introduction Yaron Matras
Yiddish: A Linguistic Introduction Neil G. Jacobs
Portuguese: A Linguistic Introduction Milton Azevedo
Ancient Egyptian: A Linguistic Introduction Antonio Loprieno
Chinese: A Linguistic Introduction Chao Fen Sun
Russian: A Linguistic Introduction Paul Cubberley
Introducci´ n a la ling¨ istica hisp´ nica Jose Ignacio Hualde, Antxon Olarrea, and Anna
o u a
Mar´a Escobar
French: A Linguistic Introduction Zsuzsanna Fagyal, Douglas Kibbee, and Fred Jenkins
Maori: A Linguistic Introduction Ray Harlow
German: A Linguistic Introduction Sarah M. B. Fagan
A Linguistic Introduction

Sarah M. B. Fagan
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: www.cambridge.org/9780521852852
© Sarah M. B. Fagan 2009

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 2009

ISBN-13 978-0-511-54008-0 eBook (EBL)

ISBN-13 978-0-521-85285-2 hardback

ISBN-13 978-0-521-61803-8 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.

List of ¬gures page ix
List of tables x
Acknowledgments xii
Abbreviations xiii

Introduction 1
1 German: speakers and geography 1
2 Objectives 1
3 Organization 2

1 Phonetics and phonology 4
1.1 Phonetics of German 4
1.1.1 Introduction 4
1.1.2 The vowel sounds of German 6
1.1.3 The consonant sounds of German 10
1.2 Phonology of German 14
1.2.1 Introduction 14
1.2.2 The vowel phonemes of German 17
1.2.3 The consonant phonemes of German 18
1.2.4 Phonological rules 20
1.2.5 Phonotactic constraints 32
1.2.6 Stress 38
1.2.7 Intonation 44

2 Morphology 54
2.1 Introduction 54
2.2 In¬‚ection 56
2.2.1 The in¬‚ection of nouns 57
2.2.2 The in¬‚ection of determiners and pronouns 63
2.2.3 The in¬‚ection of adjectives 70
2.2.4 The in¬‚ection of verbs 75
2.3 Derivation 89
2.3.1 Pre¬xation 90
2.3.2 Suf¬xation 93
2.3.3 Circum¬xation 95
2.3.4 Conversion 96
2.3.5 Implicit derivation 97

vi Contents

2.4 Compounding 98
2.4.1 Nominal compounds 100
2.4.2 Adjectival compounds 100
2.4.3 Other compounds 101
2.5 Reductions 102
2.5.1 A typology of reduction types 103
2.5.2 Reductions in word formation 104
2.5.3 Other reduction types and related word-formation processes 106

3 Syntax 115
3.1 Introduction 115
3.2 Noun phrases 117
3.2.1 NP structure 117
3.2.2 Case 119
3.3 Prepositional phrases 123
3.4 Adjective phrases 124
3.5 Adverb phrases 125
3.6 Verb phrases 126
3.7 Sentential phrases 129
3.7.1 Sentences 129
3.7.2 The topological model 138
3.7.3 Word order constraints and freedom 139
3.7.4 Distribution of pronominal elements 142

4 Semantics 149
4.1 Introduction 149
4.2 Lexical semantics 149
4.2.1 Synonymy 149
4.2.2 Antonymy 150
4.2.3 Hyponymy 152
4.2.4 Meronymy 153
4.3 Tense and aspect 153
4.3.1 The present tense 154
4.3.2 The past and the present perfect 155
4.3.3 The future tenses 158
4.3.4 The past perfect 160
4.3.5 Progressive meaning 161
4.4 Modality and evidentiality 164
4.4.1 Epistemic modality 164
4.4.2 Root modality 166
4.4.3 Evidentiality 168
4.5 Thematic roles 169
4.6 Voice 172
4.6.1 The passive 173
4.6.2 The middle 175

5 History of the language 181
5.1 The prehistoric period 181
5.1.1 Proto-Indo-European 181
5.1.2 Germanic 184
Contents vii

5.2 Old High German 187
5.2.1 Introduction 187
5.2.2 Phonology 188
5.2.3 Morphology and syntax 191
5.3 Middle High German 194
5.3.1 Introduction 194
5.3.2 Phonology 195
5.3.3 Morphology and syntax 197
5.4 Early New High German 199
5.4.1 Introduction 199
5.4.2 Phonology 201
5.4.3 Morphology and syntax 203
5.5 New High German 206

6 Regional variation 214
6.1 The standard“colloquial“dialect continuum 214
6.2 Variation in the colloquial 216
6.2.1 Variation in pronunciation 216
6.2.2 Variation in grammar 216
6.2.3 Variation in vocabulary 218
6.3 German in Switzerland 218
6.3.1 Diglossia 218
6.3.2 Swiss Standard German 220
6.4 German in Austria 224
6.4.1 Overview 224
6.4.2 Austrian Standard German 225
6.5 German in the East and West 228
6.6 The German dialects 231
6.6.1 Introduction 231
6.6.2 Upper German dialects 233
6.6.3 Central German dialects 237
6.6.4 Low German dialects 238

7 Sociolinguistic issues 244
7.1 Introduction 244
7.2 Style 245
7.2.1 Introduction 245
7.2.2 Stylistic variation 245
7.3 Address 252
7.3.1 A brief history 252
7.3.2 The address system 253
7.4 Language and gender 255
7.4.1 Equal treatment 255
7.4.2 Achieving linguistic equality 257
7.4.3 Legal language 259
7.4.4 The print media 260
7.5 Jugendsprache 262
7.5.1 Speakers and usage 262
7.5.2 Linguistic features 263
7.6 The German of foreign workers 268
viii Contents

7.6.1 Speakers 268
7.6.2 Linguistic features 269
7.7 Language contact 271
7.7.1 A brief history 271
7.7.2 Recent English in¬‚uence 275

Glossary 281
References 295
Index 310

1.1 The vocal tract page 5
1.2 The larynx 5
4.1 Lexical taxonomy for Mensch 152
6.1 Isoglosses of the High German Consonant Shift 231


1.1 Vowel sounds in German page 7
1.2 Consonant sounds in German 10
2.1 Noun plurals 59
2.2 The de¬nite article 64
2.3 The demonstrative dieser 65
2.4 The inde¬nite article ein and negative kein 65
2.5 Possessive determiners (unin¬‚ected) 66
2.6 Personal pronouns in German 67
2.7 Re¬‚exive pronouns in German 68
2.8 The relative pronoun der 69
2.9 The demonstrative pronoun der 70
2.10 The inde¬nite pronoun einer and the negative pronoun keiner 70
2.11 The strong adjective endings 72
2.12 The weak adjective endings 73
2.13 Present tense forms 76
2.14 Present tense forms of the modal verbs and wissen ˜to know™ 79
2.15 The past of weak verbs 79
2.16 The past of strong verbs 80
2.17 Ablaut classes in German 80
2.18 Principal parts of verbs in the mixed class 81
2.19 Principal parts of the modals and wissen ˜to know™ 81
2.20 In¬‚ection of haben ˜to have™, sein ˜to be™, and werden ˜to
become™ 82
2.21 Present Subjunctive I forms 83
2.22 Present Subjunctive II forms 85
2.23 Present Subjunctive II forms of auxiliary verbs 85
2.24 Imperative forms 87
3.1 The topological model of sentence structure 139
5.1 PIE stops 182
5.2 Grimm™s Law 184
5.3 The High German Consonant Shift 188
5.4 The declension of OHG tag ˜day™ and MHG tac ˜day™ 197
List of tables xi

5.5 The preterite indicative of nehmen ˜to take™ in OHG and MHG 198
5.6 The declension of ENHG tag ˜day™ 202
5.7 The declension of “weak” feminine nouns 203
5.8 The declension of i-stem nouns 204
5.9 Principal parts of selected strong verbs in MHG 204
5.10 German replacements of French loanwords 207
6.1 Personal pronouns in northern colloquial non-standard German 217
6.2 Vocabulary in northern German 219
6.3 Vocabulary in southern German 219
6.4 Vocabulary in Swiss Standard German 224
6.5 Vocabulary in Austrian Standard German 228
7.1 Stylistic variation in vocabulary 249

I am very grateful for the help and generosity of number of people who have
contributed in various ways to this project. I owe very special thanks to col-
leagues who read and gave me valuable feedback on individual chapters and
who discussed and helped clarify a number of issues: Michael Jessen, Catherine
Ringen, David Fertig, Wolfgang Ertl, John te Velde, Orrin Robinson, Joseph
Salmons, Robert Howell, Margaret Mills, Glenn Ehrstine, Mark Louden, and
Barbara Fennell. Marc Pierce and Kirsten Kumpf provided valuable comments
on the entire manuscript, for which I am very grateful. Several individuals
helped with the ¬gures in this book. I would like to thank Ingo Titze for
permission to use modi¬ed versions of images from Principles of Voice Pro-
duction, Julie Ostrem for her help in making these images available, and James
Pusack for putting me in touch with Julie. Special thanks to Erin Chrissobolis,
who helped modify and create the images used in the ¬gures. I thank the edi-
tors at Cambridge University Press who have helped see this project through
to completion: Andrew Winnard, Sarah Green, Karl Howe, and my copyeditor,
Rosemary Williams.


A adjective
acc. accusative
Adv adverb
AdvP adverb phrase
AP adjective phrase
ASG Austrian Standard German
Aux auxiliary verb
C complementizer
C consonant
Con conjunction
CP complementizer phrase
dat. dative
Det determiner
DO direct object
E event time
Eng. English
ENHG Early New High German
F foot
fem. feminine
FWG Foreign Worker German
gen. genitive
Gk. Greek
Gmc. Germanic
Goth. Gothic
GSG German Standard German
IA Item and Arrangement
IE Indo-European
In¬‚ in¬‚ection
IO indirect object
IP in¬‚ection phrase
IP intonational phrase
IP Item and Process
xiv Abbreviations

IPA International Phonetic Alphabet
Lat. Latin
masc. masculine
MHG Middle High German
MLG Middle Low German
N noun
neut. neuter
NHG New High German
nom. nominative
NP noun phrase
OE Old English
OHG Old High German
OV object“verb
OVS object“verb“subject
P preposition
PIE Proto-Indo-European
PO prepositional object
PP prepositional phrase
Pro pronoun
PRO the understood subject of an in¬nitive
PS phrase structure
R reference time
REFL re¬‚exive
S speech time
sg. singular
Sk. Sanskrit
SOV subject“object“verb
Spec speci¬er position of CP
SSG Swiss Standard German
SUB subject
SVO subject“verb“object
t trace
V verb
V vowel
VP verb phrase
WP Word and Paradigm
YP phrase (e.g., NP, AP, etc.)
XP phrase (e.g., NP, AP, etc.)

1 German: speakers and geography
Standard German is spoken by approximately 95 million speakers worldwide
(Gordon 2005). It is an of¬cial language in Germany, Austria, Switzerland,
Luxembourg, and Liechtenstein. It is the national (sole of¬cial) language in
Germany, Austria, and Liechtenstein; in Switzerland it is an of¬cial language
along with French, Italian, and Rhaeto-Romansh; in Luxembourg it shares
of¬cial status with French and Luxemburgish (L¨ tzebuergesch), a Mosel Fran-
conian dialect. German is also an of¬cial regional language in Belgium, Italy,
and Denmark, and is spoken in a number of other countries, including the
Czech Republic, Kazakhstan, Poland, and Paraguay.2
Standard German is the variety of German that is described in grammars and
dictionaries. It is the “of¬cial” form of the language; texts written in German
typically follow the spelling and grammar norms of this variety of German.
Standard German is the form that is typically used in school in German-speaking
countries and the variety that is taught to non-native speakers studying German
as a foreign language. Although non-standard varieties of German will also
be treated here (German dialects, Foreign Worker German, etc.), Standard
German is the variety that is the focus of this study. The chapters on the sounds
of German, the structure of German words, the regularities of word order, and
so on all deal with the standard language.

2 Objectives
This book aims to provide an introduction to the linguistic structure of Stan-
dard German that is rich in descriptive detail and grounded in modern linguistic
theory. It includes a history of the language, a description of the major German
dialects, and a discussion of sociolinguistic issues in addition to an analysis of
the basic structural components of the language, namely, phonetics, phonology,
morphology, syntax, and semantics. It is intended for a broad readership. It is
written in such a way as to be accessible to university students in German and
linguistics, teachers of German, and linguists with a variety of interests. Prior
2 German

knowledge of German is not required, as all necessary glosses and/or transla-
tions are provided for the examples in German.3 Professional competence in
linguistics is also not essential; basic linguistic concepts are introduced brie¬‚y
and specialist linguistic terminology is explained. A glossary of technical terms
is also provided.
This book differs in its scope, depth, and focus from other linguistic intro-
ductions to German that are currently available in English.4 It is not concerned
simply with the purely structural aspects of German, but also presents a detailed
view of the language in its historical, regional, and social settings. Where other
texts introduce the reader to linguistics with German as the object of investiga-
tion, this book focuses on the linguistic features of the language and explains
linguistic concepts only brie¬‚y. Emphasis is placed on linguistic detail and the
elucidation of insights into the language afforded by current linguistic research.
The general theoretical framework employed here is that of generative lin-
guistics, the view that a formal and explicit set of rules (a generative gram-
mar) underlies the knowledge that native speakers have of their language. The
components of this grammar, which are all interrelated, include phonetics,
phonology, morphology, syntax, and semantics. In general, a more traditional
approach to theoretical issues is taken instead of one that represents the most
recent directions in the ¬eld. The expectation is that an approach that has the
advantage of time and exposure will be accessible to a wide audience.

3 Organization
The ¬rst four chapters deal with the major structural components of the
language. Chapter 1 treats the phonetics and phonology of German, includ-
ing phonological processes, phonotactic constraints, stress, and intonation.
Chapter 2, which presents the morphology of German, deals with in¬‚ection as
well as the word-formation processes of derivation, compounding, and reduc-
tion. The discussion of the syntax of German in chapter 3 includes a description
of the phrase structure of the language, from noun phrases to sentential phrases,
and highlights the salient characteristics of German word order. Chapter 4,
which deals brie¬‚y with lexical semantics, focuses on issues of sentence-level
semantics: tense and aspect, modality, and voice.
The ¬nal three chapters of the book treat variation in the language, from
diachronic and regional to social. Chapter 5 presents a history of German,
beginning with a description of its Indo-European and Germanic ancestors
and then presenting the important phonological, morphological, and syntactic
characteristics of three of the major periods of the language. Chapter 6 deals
with regional variation. It addresses regional variation in the colloquial language
and presents the characteristics of the major German dialects. It also describes
the varieties of German spoken in Switzerland and Austria and deals with the
Introduction 3

linguistic differences in Germany between the East and the West. Chapter 7
treats the sociolinguistic issues of style, forms of address, language and gender,
youth language, the speech of foreigners, and the in¬‚uence of English on
Each chapter includes exercises that are intended to give the reader practical
experience in analyzing the language and an opportunity to put to use the
information presented in that chapter. Solutions to the exercises can be found
in the online answer key at www.cambridge.org/fagan.

1 See chapter 6 for further discussion of German dialects and the relationship of these
dialects to Standard German.
2 See Gordon 2005 for additional information on the countries in which German is
3 A number of the exercises do require a basic knowledge of German.
4 These include Russ 1994, Johnson 1998, Boase-Beier and Lodge 2003, and Fox 2005.
Phonetics and phonology

1.1 Phonetics of German

1.1.1 Introduction
The sub¬eld of linguistics known as phonetics deals with the sounds of
human speech. There are three branches of phonetics: articulatory phonet-
ics, which is concerned with how the human vocal tract produces speech
sounds; acoustic phonetics, which investigates the physical properties of the
sound waves produced when we speak; and auditory phonetics, which deals
with the way that speech sounds are perceived by listeners. This discussion of
German phonetics focuses on the articulatory characteristics of the sounds of
Speech sounds are produced when an airstream is put into motion. In German,
as in most languages, speech sounds are produced by pushing air from the lungs
out of the body through the vocal tract. A diagram of the vocal tract is provided
in Figure 1.1. To produce the different sounds of a language, the airstream is
modi¬ed in various ways by manipulating the larynx (voice box), the velum
(soft palate), the tongue, and the lips.
The larynx is made up of cartilages and muscle (see Figure 1.2). The vocal
cords, two pairs of folds of muscle and ligament, are attached to the inner
sides of the thyroid cartilage (the Adam™s apple) at the front of the larynx and
to the two arytenoid cartilages at the back of the larynx. The lower pair of
folds can be spread apart or brought together by movement of the arytenoid
cartilages. When the vocal cords are spread apart and the airstream passes
freely through the space between them, the glottis, the sound produced is
characterized as voiceless. The sound produced when pronouncing the s in
das ˜the™, for example, is voiceless. When the vocal cords are brought close
together, but not completely closed, and the air passing through them causes
them to vibrate, the sound that is produced is voiced. An example of a voiced
sound is the vowel a in ja ˜yes™.
If the velum (the soft area at the back of the roof of the mouth) is raised
against the back of the throat (pharynx), only allowing the airstream to pass

Phonetics and phonology 5

Figure 1.1 The vocal tract. Modi¬ed from I. R. Titze, “Principles of Voice
Production: Second Printing.” Copyright 2000 by the National Center for
Voice and Speech, Iowa City IA 52242.

Figure 1.2 The larynx. Modi¬ed from I. R. Titze, “Principles of Voice
Production: Second Printing.” Copyright 2000 by the National Center for
Voice and Speech, Iowa City IA 52242.

through the mouth, the sound produced is oral. All the sounds in the word Lippe
˜lip™ are oral. If the velum is lowered and air is allowed to pass through the
nasal passages as well as through the mouth, the sound produced is nasal. The
m in Mutter ˜mother™ is a nasal sound.
The tongue is a highly ¬‚exible organ of speech and plays an important role
in the modi¬cation of the airstream. It can be lowered, raised, moved forward
in the mouth, pulled back, and so on, so that it approaches or touches various
surfaces in the mouth. Important parts of the upper surface of the vocal tract
are the teeth, the alveolar ridge (the ridge behind the upper front teeth), the
6 German

(hard) palate (the roof of the mouth), the velum, and the uvula (the small
piece of soft tissue that hangs down from the rear portion of the velum); see
Figure 1.1.
The lips, which like the tongue are very ¬‚exible, are the ¬nal articulators to
modify the airstream as it leaves the body. The lips can be used together with
the teeth, as in the production of f in ¬nden ˜to ¬nd™. They can also be used
by themselves, in a rounded position, for example, as in the production of u in
Mutter ˜mother™.
Linguists use special phonetic alphabets to represent speech sounds in order
to ensure a one-to-one correspondence between sound and symbol. The symbols
of conventional spelling systems typically do not have this characteristic. The
symbol e in the German spelling system, for example, can represent three
different vowel sounds. Compare the conventional spelling of the following
words with their phonetic transcriptions (phonetic symbols are placed between
square brackets): nett [nµt ] ˜nice™, lebt [le¦pt ] ˜lives™, Sache [zax™] ˜thing™.
The symbols of the International Phonetic Alphabet (IPA) are used to represent
speech sounds in this book.

1.1.2 The vowel sounds of German
Vowels are those speech sounds that are produced without a closure of the
mouth or a narrowing of the speech organs to a degree that would produce
audible friction when the airstream passes through the mouth. Five parameters
are necessary to distinguish the different vowel sounds of German: tongue
height, tongue position, lip position, length, and tenseness.
There is a direct correlation between the tongue height of a given vowel and
the degree to which the mouth is open during the articulation of that vowel.
The i sound in Miete [mi¦t ™] ˜rent™ is a high vowel; the mouth is almost closed
during the articulation of this vowel. The e sound in Fee [fe¦] ˜fairy™ is a mid
vowel; the mouth is half open (the jaw is lowered somewhat) to pronounce this
vowel. The a sound in Saal [za¦l] ˜hall™ is a low vowel; the mouth is open wide
(the jaw is quite low) during the production of this vowel.
The parameter of tongue position refers to the location of the highest point
of the tongue, from front to back in the mouth. The i sound in Miete [mi¦t ™]
is a front vowel; the highest part of the tongue during the articulation of this
vowel is at the front of the mouth, under the palate. The e sound in Mitte [m©t ™]
˜middle™ is a central vowel; the highest part of the tongue is somewhat further
back in the mouth than for i. The u sound in Fuß [fu¦s] ˜foot™ is a back vowel;
the highest point of the tongue is in the back of the mouth under the velum.
The difference in tongue position between front and back vowels is especially
apparent if you pronounce a front vowel like i [i¦] and then a back vowel like u
Phonetics and phonology 7

Table 1.1 Vowel sounds in German

front central back

unrounded rounded unrounded rounded unrounded rounded

high tense i¦ y¦ u¦
©  …
mid tense e¦ ø¦ o¦
µ µ¦ ™ ”
lax “
low a a¦

[u¦]; you can feel the tongue being pulled back when it moves from the i to the
u position.
During the articulation of a vowel, the lips can be rounded, as in the artic-
ulation of o in so [zo¦] ˜so™. The lips can also be spread apart (unrounded), as
in the pronunciation of e in See [ze¦] ˜lake™. Lip position is thus a matter of lip
rounding. We say that [o¦] is a rounded vowel; [e¦] is an unrounded vowel.
The length of a vowel is the duration of that vowel relative to the duration
of other vowels. The a in Staat [ʃta¦t ] ˜country™ is a long vowel (length is
represented using the length mark, ¦). The a in Stadt [ʃtat ] ˜city™ is a short
The parameter of tenseness is commonly described as involving the degree
of muscular tension necessary to produce a vowel. Tense vowels are said to
be produced with greater muscular tension than lax vowels. Although this has
not been veri¬ed experimentally, there are phonetic differences between tense
and lax vowels. In German, tense vowels are produced further from the mid-
central position of the vowel area (the natural, relaxed position for the tongue)
and are higher than their lax counterparts. The i in Miete [mi¦t ™] ˜rent™ is
a tense vowel. It is somewhat higher and further forward than the i in Mitte
[m©t ™] ˜middle™, which is a lax vowel. Tense vowels in German are long
when they appear in a stressed syllable; they are short when unstressed. The
i in Musik [mu"zi¦k ] ˜music™ is tense and long; in musikalisch [muzi"k a¦l©ʃ]
˜musical™ it is tense but short because the syllable in which it occurs does not
bear primary stress. (The raised vertical stroke, ", indicates that the following
syllable bears primary stress. Stress will be indicated in transcriptions only
when relevant to the discussion. See section 1.2.6 for further discussion of
Table 1.1 lists the vowel sounds of German. Only the long variants of the
tense vowels are included in this table. Examples of words that contain these
vowels are provided in examples (1) through (3).
8 German

(1) High vowels
[i¦] liegen ˜to lie™ [y¦] lugen ˜to ¬b™
[©] Kiste ˜box™ [] Kuste ˜coast™
[u¦] spuken ˜to haunt™ […] spucken ˜to spit™
(2) Mid vowels
[e¦] lesen ˜to read™ [ø¦] l¨ sen ˜to solve™
[µ] kennen ˜to know™ [“] k¨ nnen ˜to be able to™
[µ¦] K¨ se ˜cheese™
[™] bitte ˜please™ [ɐ] bitter ˜bitter™
[o¦] Ofen ˜oven™ [”] offen ˜open™
(3) Low vowels
[a] Stadt ˜city™ [a¦] Staat ˜country™

Table 1.1 tells us that [i¦] is a high, front, tense, long, unrounded vowel. It
differs from [y¦] only in lip position; [y¦] is a high, front, tense, long, rounded
vowel. If you say [i¦] and then round your lips, the resulting sound will be [y¦]. If
you consider all the front vowels in Table 1.1, you will notice that there are three
additional pairs of vowels that differ from each other in this way (lip rounding):
[©]/[], [e¦]/[ø¦], and [µ]/[“] in addition to [i¦]/[y¦]. The rounded vowels of these
¨ ¨
pairs ([y¦], [], [ø¦], and [“]; found in words like suß ˜sweet™, funf ˜¬ve™, Ol
˜oil™, and K¨ ln ˜Cologne™, respectively) are particularly interesting because
they have no counterparts in English: English has no front rounded vowels.
A sound that stands out in Table 1.1 is [µ¦]; it is the only vowel in German that
is both lax and long. [µ¦] is typically represented by orthographic a or ah: V¨ ter
˜fathers™, ahnlich ˜similar™. Although the [µ¦] pronunciation in such words is
considered standard, speakers in northern and central Germany substitute [e¦]
instead: [fe¦t ɐ], [e¦nl©c].
The low vowels, [a] and [a¦], are not marked for tenseness, and both are
considered central vowels. Although some studies treat the two different a
sounds in German as qualitatively different (e.g., differing in tongue position),
they are treated here (following Mangold 2005, among others) as differing only
in duration.
It turns out that there is an interesting relationship between these two vow-
els. When long a is umlauted (in the formation of plurals, the derivation of
er-nominals, etc.), the resulting vowel is [µ¦], as the pairs of words in (4)
(4) Zahn [tsa¦n] ˜tooth™ Z¨ hne [tsµ¦n™] ˜teeth™
J¨ ger [jµ¦kɐ] ˜hunter™
jagt [ja¦kt ] ˜hunts™ a
j¨ hrlich [jµ¦ɐ8l©c] ˜yearly™
Jahr [ja¦ɐ8] ˜year™ a ¸
Phonetics and phonology 9

As expected, when short a is umlauted, the resulting vowel is [µ], a sound that
differs from [µ¦] only in length:1
(5) Fall [fal] ˜case™ F¨ lle [fµl™] ˜cases™
alter [µlt ɐ] ˜older™
alt [alt ] ˜old™
k¨ mmt [k µmt ] ˜combs (3rd person sg.)™
Kamm [k am] ˜comb™ a

Two vowel sounds in German never occur in stressed syllables: [™] (schwa)
and [ɐ]. Both are mid central vowels that are lax, short, and unrounded.
[ɐ] is somewhat lower than [™]. In adjective and verb endings, <e> is
pronounced as [™]; [ɐ] is typically the pronunciation of <er> when this
sequence occurs at the end of words (angle brackets, <>, indicate orthographic
alte [alt ™] (Leute) ˜old (people)™
(6) (ich) lese [le¦z™] ˜(I) read™
alter [alt ɐ] (Freund) ˜old (friend)™ Leser [le¦zɐ] ˜reader™

The sounds we have discussed so far are monophthongs, vowels that do not
show a change in quality (tongue height, tongue position) within a syllable.
Vowels that do change in quality during a syllable (because of movement of
the tongue during their articulation) are diphthongs. German has the three
diphthongs illustrated in (7).
(7) [a©] mein ˜my™ Mai ˜May™ Bayern ˜Bavaria™
[a…8] Haus ˜house™ Couch ˜couch™ Clown ˜clown™
[”©] neu ˜new™ M¨ use ˜mice™
a Boykott ˜boycott™

Diphthongs are represented phonetically by two vowel symbols, the beginning
and end points of the vowel articulation.2 One of the sounds in a diphthong
is more prominent than the other. This is indicated by placing the diacritic 8
under the less prominent of the two vowels. In the diphthongs in (7), the second
vowel is less prominent.
In German, diphthongs also arise phonetically from two additional sources.
A vocalic pronunciation of the r-sound, [ɐ], results in a number of different
diphthongs. Some examples are [i¦ɐ8] in Tier [t i¦ɐ8] ˜animal™, [u¦ɐ8] in Uhr [u¦ɐ8]
˜clock™, and [e¦ɐ8] in Meer [me¦ɐ8] ˜sea™.3 A second source is non-native (but well-
integrated) words that end in -ion or -ation, in which we ¬nd the diphthong
[io¦]: Emotion [emotsio¦n] ˜emotion™; Operation [op ™ʁatsio¦n] ˜operation™.
In this diphthong, unlike the others discussed here, the ¬rst vowel is less
German also has nasalized vowels (indicated with the diacritic )) in words
that have been borrowed from French. Some of the nasalized vowels that occur
in German are illustrated in (8).
10 German

Table 1.2 Consonant sounds in German


t k
Stop voiced p#

voiceless p t k
voiceless aspirated p t k
Fricative voiced v z j
voiceless f s c
¸ x h
Nasal m n N
Lateral l

Trill r

B = bilabial; LD = labiodental; A = alveolar; PA = postalveolar; P = palatal;
V = velar; U = uvular; G = glottal

(8) [o)¦] Balkon ˜balcony™
[“)¦] Parfum ˜perfume™
[µ)¦] Teint ˜complexion™
[a)¦] Restaurant ˜restaurant™

A vowel is nasalized when the velum is lowered during its articulation and air is
allowed to escape through the nasal cavity. Many of these words with nasalized
vowels also have alternative pronunciations with an oral vowel followed by a
nasal consonant in place of the nasalized vowel:4

(9) [”N], [o¦n] Balkon ˜balcony™
[y¦m] Parfum ˜perfume™

Because nasalized vowels do not play a major role in the sound system of
German (they occur in a relatively small number of loanwords from French),
they will not be treated further.

1.1.3 The consonant sounds of German
Consonants are those speech sounds that are produced by impeding the ¬‚ow of
air in some way. The consonant sounds of German can be described in terms
of manner of articulation (how the airstream is impeded), place of articulation
(where the airstream is impeded), and state of the vocal cords. Table 1.2 lists
the consonant sounds of German.
Stops are those consonants that are produced by a complete closure in the
vocal tract. The articulation of a stop involves closure, a build-up of pressure
during closure (because the airstream is trapped), and then release of the closure,
resulting in an “explosion” of air (stops are also called plosives). The p sound
Phonetics and phonology 11

in Pass ˜passport™, for example, is a stop. It is bilabial (both lips are used to
form the closure) and voiceless (the vocal cords are apart during the closure).
The t sound in tief ˜deep™ is an alveolar stop (the closure is formed by placing
the tip and front part of the tongue against the alveolar ridge). The k sound
in Kuh ˜cow™ is a velar stop (the closure is formed by placing the back of the
tongue against the velum).
There are three kinds of stop sounds in German: voiced stops, voiceless
stops, and voiceless aspirated stops. Stops are voiced when the vocal cords are
together and vibrating during the period of closure; they are voiceless when
they are apart and not vibrating during closure. A voiceless stop is aspirated if
its release is followed by a period of voicelessness accompanied by a burst of
air. The p in Panne, for example, is aspirated (indicated by a superscript h); the
p in Spanne, on the other hand, is not.
(10) [p ] Panne [p an™] ˜breakdown™
[p] Spanne [ʃpan™] ˜span of time™
You can see the difference between these two p sounds if you dangle a sheet
of paper in front of your mouth and say Panne and then Spanne. The burst of
air that accompanies the aspirated p in Panne causes the paper to move. This
movement of the paper is absent when you say Spanne, since the p in this word
is not aspirated. This same contrast can be found in English and demonstrated
with word pairs like pin (with aspirated p) and spin (with unaspirated p).
Examples of the three kinds of stop sounds in German are given in (11)
through (13); see also Table 1.2. For each type of stop, there are three places
of articulation: bilabial, alveolar, and velar.
Voiced stops5
[p#] Fieber [fi¦p#ɐ] ˜fever™
[t] wieder [vi¦tɐ] ˜again™
[k] logisch [lo¦k©ʃ] ˜logical™
(12) Voiceless stops
[p] (ich) spare [ʃpa¦ʁ™] ˜(I) save™
[t] Stahl [ʃta¦l] ˜steel™
[k] Skalen [ska¦l™n] ˜scales™
(13) Voiceless aspirated stops
[p ] Paare [p a¦ʁ™] ˜pairs™
[t ] Tal [t a¦l] ˜valley™
[k ] (die) Kahlen [k a¦l™n] ˜(the) bald people™
There is an additional sound in German that is referred to as a “glottal
stop” or “glottal plosive,” [”]. It is produced by closing the vocal cords, hold-
ing them tightly together along their entire length, and then releasing them
12 German

suddenly into the articulation of a following vowel. The glottal stop is the
sound that one typically hears before a stressed vowel at the beginning of a

(14) [”] Ol ["”ø¦l] ˜oil™, offen ["””fn] ˜open™

In English, a glottal stop can be heard before the vowels in utterances like uh-uh
(used to express disagreement or say “no”) or uh-oh (used to express surprise).
Although the glottal stop, [”], is phonetically a state of the glottis rather than
an articulatory stop like [p], [t], and [k], for example (Ladefoged 1971:16), it
will be treated here as a stop.
Fricatives are those consonants that are produced by placing two articulators
close together to create a narrow passage through which air is forced, producing
a turbulent air¬‚ow. The f sound in Fieber ˜fever™, for example, is a labiodental
fricative. The two articulators that create the narrow passage in the production
of this fricative are the lower lip and the upper teeth. The s sound in es ˜it™
is an alveolar fricative. It is produced by raising the blade of the tongue (the
area right behind the apex “ the tip) close to the alveolar ridge. The sch sound
in sch¨ n ˜beautiful™ is postalveolar, produced by placing the front part of the
tongue behind the alveolar ridge, close to the front part of the hard palate. The
lips are also strongly rounded, more so than in the pronunciation of English sh.
The ch sound in Licht ˜light™ (known in German as the “ich-Laut”) is palatal,
produced by raising the front of the tongue towards the hard palate, with the tip
of the tongue touching the lower front teeth. The ch sound in lachen ˜to laugh™,
on the other hand (known in German as the “ach-Laut”), is a velar fricative,
produced by raising the back of the tongue towards the velum (the soft palate).
One of the pronunciations of the r sound in Rede ˜speech™ is the uvular fricative,
produced by raising the back of the tongue towards the uvula. The h sound in
haben ˜to have™, characterized here as a glottal fricative, is produced by slightly
narrowing the glottis (Mangold 2005:52).6
There are two basic kinds of fricatives in German: voiced and voiceless.
Examples of these two types of fricatives are provided in (15) and (16);
see also Table 1.2. There are seven relevant places of articulation for frica-
tives in German: labiodental, alveolar, postalveolar, palatal, velar, uvular, and

(15) Voiced fricatives
[v] Wein [va©n] ˜wine™
[z] Reise [ʁa©z™] ˜trip™
[ ] Marge [maɐ8 ™] ˜margin™
[j] jener [je¦nɐ] ˜that™
[ʁ] Rede [ʁe¦t™] ˜speech™
Phonetics and phonology 13

(16) Voiceless fricatives
[f] fein [fa©n] ˜¬ne™
[s] (ich) reiße [ʁa©s™] ˜(I) rip™
[ʃ] Marsch [maɐ8ʃ] ˜march™
[c] Chemiker [ce¦mik ɐ] ˜chemist™
¸ ¸
[x] Nacht [naxt ] ˜night™
[h] Hut [hu¦t ] ˜hat™

There are three nasal consonants in German. Nasals are produced by forming
a complete closure in the vocal tract and lowering the velum so that air escapes
through the nasal passage. Nasal consonants are similar to stops in that they
involve a closure of the vocal tract. In German, nasals have the same places
of articulation as stops: bilabial, alveolar, and velar. All German nasals are

(17) Nasal consonants
[m] Mehl [me¦l] ˜¬‚our™
[n] Nase [na¦z™] ˜nose™
[N] lange [laN™] ˜long™

The l sound in German can be classi¬ed as an approximant, a sound produced
by bringing two articulators close together without producing a turbulent air-
¬‚ow. It is a lateral sound, articulated by placing the blade of the tongue against
the alveolar ridge and allowing air to escape on either side of the tongue.
Because the point of contact with the tongue is the alveolar ridge, l is classi¬ed
as alveolar. It is a voiced sound. We will refer to it simply as a lateral.

(18) [l] Milch [m©lc] ˜milk™

Many varieties of English have two different l sounds, in contrast to Standard
German: a “clear” l, as in Standard German, and a “dark” or velarized l, [ ],
which is produced by raising the back of the tongue towards the velum at the
same time the blade of the tongue is making contact with the alveolar ridge.7
In those variants of English that have both types of l, clear l is typically found
in prevocalic position, whereas dark l is found postvocalically:8

(19) English l
[l] lip, leave, late, lap, look
[ ] hill, ball, fold, self, ¬lm

Trills are the ¬nal type of consonant sound in German. A trill is produced by
holding an articulator loosely close to another articulator, so that the airstream
sets it in vibration. An alveolar trill, [r], a realization (pronunciation) of German
14 German

r typically found in Bavaria and Austria, is produced by the tip of the tongue
vibrating against the alveolar ridge. A uvular trill, [], another realization of
German r, is produced by the uvula vibrating against the back of the tongue.
Both types of trills are voiced.
[r] Rede [re¦t™] ˜speech™
[] Rede [e¦t™] ˜speech™

1.2 Phonology of German

1.2.1 Introduction
My goal here is to present a descriptive phonology of German, not a theoretical
account that aims to compete with the most recent theoretical treatments. I do,
however, take into account the recent literature and incorporate facts and ideas
that are new and go beyond the traditional accounts of German phonology found
in previous monographs and textbook descriptions. I express phonological
generalizations in the form of rules and derivations rather than simply relying
on prose formulations, since the formalization of rules helps to ensure accuracy
and can reveal generalizations that might not be apparent in prose accounts.
Because of length limitations I focus more on segmental than on prosodic
phonology, and more on the word level than on the sentence level. I also focus
more on “regular” than “irregular” phonology and do not treat those rules that
are morphologically conditioned.9
Phonology is the sub¬eld of linguistics that deals with the sound patterns
of language, the regularities that underlie the sound systems of language. For
example, if we consider the way in which r is pronounced in German, we
see that there are two phonetically distinct pronunciations that are possible, a
consonantal pronunciation, [ʁ], and a vocalic, [ɐ8], and if we look at the distri-
bution of these two pronunciations, we see a pattern. We ¬nd the consonantal
pronunciation when r occurs at the beginning of words, the vocalic when it
occurs at the end of words:10
(21) Consonantal r
Rat [ʁa¦t ] ˜advice™, Rippe [ʁ©p ™] ˜rib™, rund [ʁ…nt ] ˜round™
(22) Vocalic r
Bier [pi¦ɐ8] ˜beer™, leer [le¦ɐ8] ˜empty™, Uhr [u¦ɐ8] ˜clock™
This pattern in the distribution of the r sounds in German is a characteristic of
the language that is treated in the phonology. It is a characteristic of the sound
system of German.
The minimal unit in the sound system of any language is the phoneme.
Phonemes are more abstract units than the actual speech sounds (or phones)
Phonetics and phonology 15

in a language. Although the two different pronunciations of German r, for
example, are phonetically distinct, at some level both of these pronunciations
are the same; they are both perceived as r by native speakers of German.
They are two different phonetic realizations of the r sound, or r phoneme in
German, /r/. Phonemes are written between slash brackets, //, to distinguish
them from their various phonetic realizations, or allophones, which are placed
between square brackets, []. At the phonemic level of representation, a word
like leer ˜empty™, for example, is transcribed as /le¦r/. At the phonetic level, it
is transcribed as [le¦ɐ8], since the r phoneme occurs at the end of this word and
thus has a vocalic pronunciation.
Phonemes are the distinctive sounds of a language. They are the sounds that
have the potential to bring about contrasts in meaning. If we substitute r for l
in the word Lippe ˜lip™, for example, this yields a completely different word,
Rippe ˜rib™. A pair of words like Lippe and Rippe is called a minimal pair,
two words that differ in meaning and that are identical in form except for one
sound that occurs in the same place in each word. Given a minimal pair like
Lippe and Rippe, we can say that /l/ and /r/ are phonemes of German. The
discovery of minimal pairs is thus crucial in determining the phonemes of a
Sounds are considered to be allophones of the same phoneme if they are
phonetically similar and in complementary distribution (they never occur in
the same environment) or in free variation (the substitution of one sound
for the other does not cause a change in meaning). The phoneme /i¦/, for
example, has two allophones, [i¦] and [i], which are in complementary dis-
tribution: [i¦] occurs only in stressed syllables; [i] occurs only in unstressed

Fabrik [fa"p#ʁi¦k ] ˜factory™ Fabrikant [fap#ʁi"k ant ] ˜industrialist™
Maschine [ma"ʃi¦n™] ˜machine™ maschinell [maʃi"nµl] ˜mechanical™
Musik [mu"zi¦k ] ˜music™ musikalisch [muzi"k a¦l©ʃ] ˜musical™

The two allophones of /r/, [ʁ] and [ɐ8], are in complementary distribution in all
environments except after short vowels, where they are in free variation. After
short vowels, /r/ may be realized as [ʁ] or [ɐ8], with no change in meaning.

(24) hart [haʁt ] or [haɐ8t ] ˜hard™
Herr [hµʁ] or [hµɐ8] ˜sir™
dort [t”ʁt ] or [t”ɐ8t ] ˜there™

The smallest unit in the analysis of phonological structure is the feature.
Features are the characteristics of segments; they are the units that make up
individual speech sounds. Features such as [round], [tense], and [high] play
a role in the description of German vowels. The vowel [i¦], for example, has
16 German

the features [’round], [+tense], and [+high], among others ([+] means that
a feature is present and [’] means that it is absent). Other features that are
relevant in the description of speech sounds in German are those such as
[voice], [continuant], [spread glottis], etc. These as well as other features will
be presented and explained as needed in the discussion of phonological rules
in section 1.2.4.
A feature that is capable of distinguishing one phoneme from another (or
one set of phonemes from another set) is a distinctive feature. For example, the
feature [round] is a distinctive feature of German because it distinguishes the
phoneme /y¦/ from /i¦/. The single difference between these two phonemes is lip
rounding: /y¦/ is [+round]; /i¦/ is [’round]. Features can be used to characterize
not just individual sounds, but also classes of sounds. For example, the two
features [’back] and [+round] describe the class of front rounded vowels in
German. A class of sounds that share a feature or features is a natural class.
Other examples of natural classes are voiced fricatives, nasal consonants, and
back vowels. As we will see in section 1.2.4, many of the regularities that
underlie the phonology of German can be expressed as rules that apply to
natural classes.
The syllable plays an important role in the phonology of German. The dis-
tribution of the allophones of certain phonemes, for example, is best described
in terms of where they occur in a syllable. A syllable is made up minimally
of a nucleus, usually a vowel (a monophthong or a diphthong), which forms
the core of the syllable. The onset of a syllable is made up of the segment or
segments that precede the nucleus; the coda consists of the segment or segments
that follow the nucleus. The word schlank [ʃlaNk ] ˜slim™ is monosyllabic. The
single vowel in this word, [a], is the nucleus; the ¬rst two segments [ʃl] form
the onset; the last two segments [Nk ] form the coda. The word Physik [fy.zi¦k ]
˜physics™ is disyllabic (the period marks a syllable boundary). Its ¬rst syllable
consists of the onset [f] and the nucleus [y]; its coda is empty. The second
syllable consists of the onset [z], the nucleus [i¦], and the coda [k ]. In a word
like kaufen [k a…8.fn] ˜to buy™, the diphthong [a…8] forms the nucleus of the ¬rst
syllable (its coda is empty).11 The nasal [n] forms the nucleus of the second
syllable (its coda is also empty). The diacritic is placed under the [n] to indi-
cate that it is syllabic and thus forms the nucleus. The placement of syllable
boundaries in the discussion of phonological rules in section 1.2.4 will simply
be given. The principles that determine the placement of syllable boundaries
will be discussed in section 1.2.5.
The foot also plays an important role in the phonology of German. A foot is
a stressed syllable and any following unstressed syllables that intervene before
the next stressed syllable. The phrase widerlicher Geruch ˜disgusting smell™
for example, with six syllables, two of which are stressed, consists of two feet:
F ["widerlicher Ge]F ["ruch]
Phonetics and phonology 17

1.2.2 The vowel phonemes of German
On the basis of the minimal pair test, we can assume the following vowel
phonemes for German:
(25) Vowel phonemes in German and their allophones
/i¦/ Musik [i¦] ˜music™, musikalisch [i] ˜musical™
/©/ bitte [©] ˜please™
/y¦/ Physiker [y¦] ˜physicist™, Physik [y] ˜physics™
// mussen [] ˜to have to™
/u¦/ Jubel [u¦] ˜jubilation™, jubilieren [u] ˜to jubilate™
/…/ Mutter […] ˜mother™
/e¦/ leben [e¦] ˜to live™, lebendig [e] ˜lively™
/µ¦/ Pr¨ tor [µ¦] ˜praetor™, pr¨ torisch [µ] ˜praetorial™
a a
/µ/ Bett [µ] ˜bed™
/ø¦/ Goethe [ø¦] ˜Goethe™, Goetheana [ø] ˜works by and about Goethe™
/“/ k¨ nnen [“] ˜to be able to™
/™/ bitte [™] ˜please™
/o¦/ Probe [o¦] ˜test™, probieren [o] ˜to try™
/”/ Gott [”] ˜god™
/a¦/ Drama [a¦] ˜drama™, dramatisch [a] ˜dramatic™
/a/ Tanne [a] ˜¬r™
/a©/ nein [a©] ˜no™
/a…8/ Baum [a…8] ˜tree™
/”©/ treu [”©] ˜loyal™
Given the minimal pair bieten [pi¦tn] ˜to offer™ and bitten [p©tn] ˜to ask™, for
example, we can posit the phonemes /i¦/ and /©/. With the minimal pair spielen
[ʃpi¦ln] ˜to play™ and sp¨ len [ʃpy¦ln] ˜to rinse™, we can add the phoneme /y¦/
to the list. The minimal pair Kiste [k ©st ™] ˜box™ and K¨ ste [k st ™] ˜coast™
yields the additional phoneme //. Minimal pairs can be found for each of the
phonemes listed in (25).
Although [™] never occurs in stressed syllables, it contrasts with other
unstressed vowels and can thus be considered a phoneme of German. Con-
sider, for example, the minimal pair Rebellen [ʁep#µl™n] ˜rebels™ and Rebellin
[ʁep#µl©n] ˜female rebel™ and the near-minimal pair fehlend ˜missing™ [fe¦l™nt ]
and elend ˜wretched™ [e¦lµnt ].
Minimal pairs can also be found showing that [™] contrasts with [ɐ]: Lehre
[le¦ʁ™] ˜teaching™ and Lehrer [le¦ʁɐ] ˜teacher™; Reife [ʁa©f™] ˜ripeness™ and
reifer [ʁa©fɐ] ˜riper™. However, notice that [ɐ] is always associated with the
sequence <er>: the -er used to form nouns from verbs (Fahrer ˜driver™); the
comparative -er (kleiner ˜smaller™); the -er adjective ending (ein großer Tisch
˜a large table™), etc. In spite of the minimal pairs contrasting [™] and [ɐ],
18 German

we will not posit [ɐ] as a phoneme of German. We will instead represent it
phonemically as the sequence /™r/ (see, for example, Benware 1986, Hall 1992,
Kohler 1995, Mangold 2005). In section we will discuss the rules that
determine when this sequence will be realized phonetically as [ɐ] (as in reifer
/ra©f™r/ [ʁa©fɐ]) and when it will be realized as [™ʁ] (as in reifere /ra©f™r™/
Notice that all the tense/long vowel phonemes (not just /i¦/, as mentioned
in section 1.2.1) have long and short allophones. The long allophones occur
in stressed syllables; the short allophones occur in unstressed syllables. This
distribution of allophones can be expressed as a rule; it will be treated in
section 1.2.4.
Although the diphthongs in German are represented phonetically as a
sequence of vowels, they will be treated here, following common practice
(Benware 1986, Kohler 1995, Mangold 2005), as single phonemes. As the fol-
lowing minimal pairs demonstrate, the diphthongs contrast with single vowels,
both long and short.12

(26) Mais [ma©s] ˜corn™ mies [mi¦s] ˜lousy™ muss [m…s] ˜must™
faul [fa…8l] ˜lazy™ viel [fi¦l] ˜much™ voll [f”l] ˜full™
Meute [m”©t ™] ˜pack™ Miete [mi¦t ™] ˜rent™ Mitte [m©t ™] ˜middle™

1.2.3 The consonant phonemes of German
The consonant phonemes that can be posited for German are listed in (27)
through (31).

(27) Stop phonemes in German and their allophones
/p/ Bass [p] ˜bass™, rauben [p#] ˜to rob™, Raub [p ] ˜robbery™
/t/ Deich [t] ˜dike™, leiden [t] ˜to suffer™, Leid [t ] ˜sorrow™
/k/ Gabel [k] ˜fork™, lagen [k] ˜(they) lay™, lag [k ] ˜(I) lay™,
ruhig [c] ˜calm™
/p / Pass [p ] ˜passport™, Raupen [p] ˜caterpillars™ ([ʁa…8pm])
/t / Teich [t ] ˜pond™, leiten [t] ˜to lead™ ([la©tn])
/k / Kabel [k ] ˜cable™, Laken [k] ˜sheet™ ([la¦kŋ])

Affricate phonemes in German and their allophones13
/pf/ Pfund [pf] ˜pound™
/ts/ Zunge [ts] ˜tongue™
/tʃ/ Cello [tʃ] ˜cello™
/t / Manager [t ] ˜manager™, das Dschungel¬eber [tʃ] ˜the yellow
Phonetics and phonology 19

(29) Fricative phonemes in German and their allophones
/f/ falsch [f] ˜false™
/s/ Wasser [s] ˜water™
/ʃ/ Schule [ʃ] ˜school™
/x/ Nacht [x] ˜night™, nicht [c] ˜not™
/h/ haben [h] ˜to have™
/v/ kurven [v] ˜to circle™, kurvt [f] ˜circles™
/z/ reisen [z] ˜to travel™, gereist [s] ˜traveled™
/ / Garage [ ] ˜garage™
/j/ ja [j] ˜yes™
(30) Nasal phonemes in German and their allophones
/m/ machen [m] ˜to make™
/n/ neu [n] ˜new™
/N/ singen [N] ˜to sing™
Liquid phonemes in German and their allophones14
/l/ lachen [l] ˜to laugh™
/r/ leeren [ʁ] ˜to empty™, leer [ɐ8] ˜empty™, bitter [ɐ] ˜bitter™15

Minimal pairs can be found in which the glottal stop contrasts with other

(32) neben [ne¦p#m] ˜beside™ eben [”e¦p#m] ˜even™
mein [ma©n] ˜my™ ein [”a©n] ˜one™
dich [t©c] ˜you™
¸ ich [”©c] ˜I™

However, the distribution of the glottal stop is predictable (see section
Furthermore, its presence is optional. The words eben, ein, and ich, for example,
can be pronounced without a glottal stop with no change in meaning. Thus, the
glottal stop is not given the status of a phoneme in German.
The phonemic status of the affricates in (28) is not uncontroversial. Some
studies treat such stop“fricative sequences as sequences of two separate
phonemes (Moulton 1962, Heike 1972, Benware 1986, Kohler 1995); others
treat them as monophonemic (Hall 1992, Mangold 2005).
Another area of controversy involves the phonemic status of [N]. Following
studies such as Benware 1986, Kohler 1995, and Mangold 2005, /N/ is treated
here as a phoneme of German. Many studies, however (e.g., Seiler 1962,
Vennemann 1970, Hall 1992, Wiese 1996), do not give [N] phonemic status,
but treat it instead as derived from the cluster /N / or /Nk/, where /N/ is a nasal
that is unspeci¬ed for place of articulation.
20 German

1.2.4 Phonological rules
The relationship between the phonemes of German and their phonetic manifes-
tations can be expressed as rules, as statements that tell us how these phonemes
are realized phonetically. For example, one of these rules tells us when the
phoneme /i¦/ is realized as [i¦] and when it is realized as [i]. Phonological rules
do not just apply to single phonemes; they also apply to classes of phonemes.
The rule that describes when /i¦/ is realized as [i] applies to all tense vowels,
not just to /i¦/.
Phonological rules relate the phonemic level of analysis to the phonetic.
They derive the phonetic realization of words from their underlying or phone-
mic representation. Given the phonemic representation of a word like Musik
˜music™, for example, /mu¦"zi¦k /, the phonological rules of German will yield
the phonetic representation [mu"zi¦k ]. These rules will tell us that /u¦/ is short
in this word (because it is not stressed) and that /k / is aspirated (because it
occurs before a pause).
Rules can generally be expressed using the following form of notation:
A ’ B / X _____ Y
A in this notation stands for an element in underlying (phonemic) representation
(a phoneme or a class of phonemes) and B represents the change this element
undergoes (how it is realized phonetically). The focus bar, _____, indicates
the position of the segment undergoing a change, and X and Y describe the
environment in which the segment must be located to undergo the change. The
rule in (33) can thus be read as “A becomes B when it occurs between X and
Y.” X or Y may be absent. If X, for example, is absent, the rule is read as “A
becomes B when it occurs before Y.” Any additional symbols employed in the
following discussion of the phonological rules of German will be explained as
they occur. We capitalize the names of the rules formulated in this and other
chapters (which apply to German) to distinguish them from terms used for
general processes that may apply in other languages. Vowel Shortening The rule of Vowel Shortening in Ger-
man states that vowels are shortened when they are unstressed (V stands for
(34) Vowel Shortening
V ’ [’long]/ _____
Long vowels that are stressed are unaffected by the rule and thus retain their
length. This rule yields the short allophones of the long vowel phonemes;
the long allophones occur when the rule fails to apply. This rule allows us
Phonetics and phonology 21

to represent the form meaning ˜music™ as /mu¦zi¦k / in both Musik ˜music™
and musikalisch ˜musical™ and account for the fact that the i in this form is
pronounced as long and tense in Musik, but short and tense in musikalisch.
/mu¦"zi¦k / ’ [mu"zi¦k ]
/mu¦zi¦"k a¦l©ʃ/ ’ [muzi"k a¦l©ʃ]
Notice that Vowel Shortening also accounts for the fact that the u in both words
is short and tense (it is unstressed in both words). Voicing Assimilation The contrast between the two sets of
stop phonemes in German has traditionally been viewed as a one of voicing.
Under the traditional view, the fortis stops (represented here as /p t k /) are
[’voice]; the lenis stops (represented here as /p t k/) are [+voice] (W¨ nglera
1960, Moulton 1962, Wurzel 1970, Rubach 1990, Hall 1992, Wiese 1996).
(I use the traditional terms “fortis” and “lenis” here simply as a means of
distinguishing the two sets of stops.) An alternative position suggests that
the relevant feature is [tense] or [spread glottis], not [voice] (Kloeke 1982,
Meinhold and Stock 1982, Iverson and Salmons 1995, Jessen and Ringen 2002).
Following Jessen and Ringen (2002), who provide experimental data in support
of their position, we will assume that the contrast is one of [spread glottis]. Both
sets of stop phonemes are [’voice].16 The fortis phonemes, unlike the lenis,
are [+spread glottis]. Sounds that are [+spread glottis] have an active glottal
opening gesture; in stops, the feature [+spread glottis] is often signaled by
aspiration, although aspiration may be absent due to the environment in which
the stop occurs (Jessen and Ringen 2002:192). A [+spread glottis] stop that
occurs before a syllabic nasal or lateral, for example, is not aspirated. Compare
the careful pronunciation of leiten ˜to lead™, with schwa and a non-syllabic nasal
following /t / (/la©t ™n/ ’ [la©t ™n]) and the pronunciation without schwa and
a syllabic nasal (/la©t ™n/ ’ [la©tn]).
The lenis stops are underlyingly (phonemically) [’voice] and are often
realized phonetically as voiceless stops:17
/tax/ ’ [tax]
(36) Dach ˜roof™
/tas tax/ ’ [tas tax]
das Dach ˜the roof™
/ha…8stax/ ’ [ha…8stax]
Hausdach ˜house roof™
When they occur between voiced sounds, however, they tend to be voiced. That
is, lenis stops assimilate to their voiced surroundings.
/o¦t™r/ ’ [o¦tɐ]
(37) oder ˜or™
/a©n tax/ ’ [a©n tax]
ein Dach ˜a roof™
Although voicing in this environment is typical, some speakers do not show
voicing consistently, which suggests that voicing is phonetically conditioned,
22 German

that is, that it has a phonetic explanation (Jessen and Ringen 2002:205).18
Following Jessen and Ringen, we treat voicing assimilation as phonetic, not
phonological (voicing assimilation is not expressed as a phonological rule).
However, lenis stops that occur between two voiced sounds are transcribed
phonetically as voiced, since they do tend to be voiced in this environment. Fortition There are some instances, however, where lenis
stops occur between voiced segments yet are realized as [’voice]. At ¬rst
glance, it appears that the lenis stops in the word pairs in (38) through
(40) appear in roughly identical environments (all occur between [+voice]

/ne¦pl©k/ ’ [ne¦p#l©c]
(38) neblig ˜foggy™ ¸
/µrhe¦pl©x/ ’ [µɐ8he¦p l©c]
erheblich ˜considerable™ ¸
/hantl…N/ ’ [hantl…N]
(39) Handlung ˜action™
/hantl©x/ ’ [hant l©c]
handlich ˜handy™ ¸
/n“rkl©k/ ’ [n“ɐ8kl©c]
(40) n¨ rglig ˜cranky™
o ¸
/k µrkl©x/ ’ [k µɐ8k l©c]
k¨ rglich ˜sparse™
a ¸

However, if we consider the structure of these words, we see a difference
between the two members of each pair:

(41) nebl-ig ˜foggy™ erheb-lich ˜considerable™
Handl-ung ˜action™ hand-lich ˜handy™
n¨ rgl-ig ˜cranky™
o k¨ rg-lich ˜sparse™

The lenis stop in the second member of each pair occurs immediately before
a suf¬x (-lich); the lenis stop in the ¬rst member of each pair does not (the
consonant /l/ occurs between the stop and the following suf¬x). This difference
in structure can explain the difference in the realization of the lenis stop. Jessen
and Ringen (2002:212) explain forms like handlich by assuming a constraint
that requires stops at the end of a phonological word to be [+spread glottis].
Their account requires that one consider handlich (but not Handlung) to be two
phonological words, an assumption that is independently motivated (Wiese
1996, Jessen and Ringen 2002).19
Beckman et al. (to appear) argue that this same constraint applies to fricatives
as well as stops. It accounts for the fact that fricatives in word-¬nal position
are voiceless, as the following examples demonstrate.

/kra¦z/ ’ [kʁas]
(42) Gras ˜grass™
/krµ¦z™r/ ’ [kʁµ¦zɐ]
Gr¨ ser ˜grasses™
/kra¦zra©x/ ’ [kʁa¦sʁa©c]
grasreich ˜full of grass™ ¸
Phonetics and phonology 23

We will express this constraint as the rule of Fortition (following Wiese 1996,
curly brackets are used to indicate the boundaries of phonological words):

(43) Fortition

[’spread glottis] [+spread glottis] / _____}

This rule states that sounds that are [’spread glottis] (lenis stops and voiced
fricatives) become [+spread glottis] (fortis stops and voiceless fricatives) when
they occur at the end of a phonological word. It accounts for the differences
in pronunciation of the lenis stops in the word pairs in (41) and the voiced
fricatives in the examples in (42). Aspiration As mentioned in section, not all [+spread
glottis] stops are aspirated. There is much variation in the degree of aspiration,
which is dependent on factors such as boundaries, stress, place of articulation,
and so on (Jessen and Ringen 2002:192). For example, aspiration is stronger
before stressed vowels than it is before schwa, an unstressed vowel, but it is
present in this environment (Jessen 1998:93“94). The degree of aspiration is an
issue of phonetics, however, and will not concern us here. The presence versus
absence of aspiration, on the other hand, is an issue that we will address. The
presence of aspiration is predictable, and can be accounted for by the following

. 1
( 12)