2nd Edition

The Syntax of
Natural Language

Beatrice Santorini, Anthony Kroch, and Michael Diercks

 i

DRAFT manuscript, currently undergoing classroom testing.

Comments welcome at [email protected]
Preface

TBD
Acknowledgements

Many thanks to the following friends and colleagues for help with examples, judgments, and liter-
ature recommendations, for comments on earlier drafts, and for helpful discussion: Rajesh Bhatt,
Michel Degraff, Lars-Olof Delsing, David Embick, Chung-hye Han, Caroline Heycock, Sabine Ia-
tridou, Anne Jenner, Helge Lødrup, Eleni Miltsakaki, Kimiko Nakanishi, Peter Patrick, Don Ringe,
Maribel Romero, Peter Svenonius, Sten Vikner, and Raffaella Zanuttini.
Of course, we alone are responsible for the use we have made of their help.
Thanks also to the students at the University of Pennsylvania for whom this book was
originally developed.
Xiaoxing Yu and Charis Kim were largely responsible for converting this textbook from
its original HTML format to the current typeset form (with regular participation from Diercks).
They proofread carefully and made editorial and expository contributions. They also contributed
to some of the problem sets and answers in the Teacher’s Manual.
Wug drawings (including cover art) are by Laura McLaren and Hannah Lippard. The wug
has become somewhat of an unofficial mascot of linguistics, and is a product of the Jean Berko-
Gleason’s (1958) language acquisition experiment.
Contents

Preface ii

Acknowledgements iii

1 Foundational issues 1
1 Prescriptive versus descriptive grammar . . . . . . . . . . . . . . . . . . . . . . . 2
2 Rule formation and syntactic structure in language acquisition . . . . . . . . . . . 6
2.1 A thought experiment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2 Rule-based word formation . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.3 Question formation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3 More evidence for syntactic structure . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.1 Intuitions about words belonging together . . . . . . . . . . . . . . . . . . 12
3.2 Structural ambiguity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
4 Universal Grammar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
4.1 Formal universals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
4.2 Recursion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
4.3 Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
5 Generative grammar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
5.1 Elementary trees and substitution . . . . . . . . . . . . . . . . . . . . . . . 18
5.2 Grammaticality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
5.3 Grammar versus language . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
6 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
7 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

2 Syntactic categories and constituenthood 32

1 Parts of speech . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
1.1 The trouble with “words” and “parts of speech” . . . . . . . . . . . . . . . 34
1.2 Nouns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
1.3 Adjectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
1.4 Modals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
1.5 Auxiliary verbs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
2 Tests for determining syntactic constituenthood . . . . . . . . . . . . . . . . . . . 54
Contents v

2.1 Substitution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
2.2 Movement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
2.3 Questions and short answers . . . . . . . . . . . . . . . . . . . . . . . . . 57
2.4 It clefts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
3 Some complications for constituency diagnostics . . . . . . . . . . . . . . . . . . . 59
3.1 Mismatches between syntactic structure and other structure . . . . . . . . 59
3.2 False negative results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
4 Representing syntactic constituenthood . . . . . . . . . . . . . . . . . . . . . . . . 67
4.1 Basic terms and relations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
4.2 Derived terms and relations . . . . . . . . . . . . . . . . . . . . . . . . . . 70
5 Regarding scientific models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
6 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
7 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
8 Appendix: Finiteness in English . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
8.1 Verb forms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
8.2 Finiteness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

3 Some basic linguistic relations 85

1 Argumenthood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
1.1 Semantic valency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
1.2 Transitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
2 Modification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
3 Predication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
3.1 Evidence for a subject requirement: Expletive it . . . . . . . . . . . . . . . 93
3.2 Aristotelian versus Fregean predicates . . . . . . . . . . . . . . . . . . . . 94
3.3 More evidence for the subject requirement: Expletive there . . . . . . . . 95
3.4 Some special cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
4 Thematic roles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
4.1 Agent, cause, and instrument . . . . . . . . . . . . . . . . . . . . . . . . . 99
4.2 Experiencer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
4.3 Recipient . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
4.4 Location, path, and goal . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
4.5 Measure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
4.6 Theme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
5 Grammatical relations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
5.1 Subjects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
5.2 Indirect objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
5.3 Direct objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
5.4 Distinguishing properties of grammatical relations . . . . . . . . . . . . . 103
6 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
7 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106

4 Introducing the X′ schema of phrase structure 110

1 The X schema for elementary trees . . . . . . . . . . . . . . . . . . . . . . . . . . 111
′

1.1 Transitive elementary trees . . . . . . . . . . . . . . . . . . . . . . . . . . 111

Contents vi

1.2 The X′ schema . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113

1.3 Intransitive elementary trees . . . . . . . . . . . . . . . . . . . . . . . . . 115
2 Deriving sentences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
2.1 Simple sentences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
2.2 Complex sentences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
3 The adjunct relation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
3.1 Modification is different . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
3.2 The need for an adjunction operation . . . . . . . . . . . . . . . . . . . . . 126
3.3 A typology of syntactic dependents . . . . . . . . . . . . . . . . . . . . . . 129
3.4 More on the distinction between complements and adjuncts . . . . . . . . 129
4 Do we need intermediate projections? . . . . . . . . . . . . . . . . . . . . . . . . . 131
5 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
6 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133

5 Extending the X′ schema 137

1 Noun phrases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
1.1 Parallels and differences between noun phrases and sentences . . . . . . . 138
1.2 Noun phrases as DPs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
1.3 More on determiners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
1.4 Modification and related issues . . . . . . . . . . . . . . . . . . . . . . . . 151
2 Adjective phrases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
3 Prepositional phrases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
3.1 Transitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
3.2 Clausal complements of prepositions . . . . . . . . . . . . . . . . . . . . . 163
4 Crosslinguistic variation in headedness . . . . . . . . . . . . . . . . . . . . . . . . 165
4.1 Head-final vs. head-initial patterns . . . . . . . . . . . . . . . . . . . . . . 165
4.2 Harmonic and disharmonic headedness . . . . . . . . . . . . . . . . . . . 168
4.3 Final thought . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
5 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
6 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174

6 Wh- movement 182

1 Basic descriptive terminology for types of questions . . . . . . . . . . . . . . . . . 183
1.1 Yes-no versus wh- questions . . . . . . . . . . . . . . . . . . . . . . . . . . 183
1.2 Direct versus indirect questions . . . . . . . . . . . . . . . . . . . . . . . . 184
1.3 Information questions versus echo questions . . . . . . . . . . . . . . . . 185
2 A movement analysis of questions . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
2.1 Indirect questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
2.2 Direct questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
3 Topicalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
4 Relative clauses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
4.1 Wh- relative clauses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
4.2 That relative clauses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
4.3 Zero relative clauses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
4.4 Doubly marked relative clauses . . . . . . . . . . . . . . . . . . . . . . . . 201
Contents vii

4.5 Stylistic considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202

4.6 A historical note . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
5 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
6 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204

7 Case theory 209

1 A first look at case . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
1.1 The basic purpose of case . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
1.2 Case government . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
1.3 Synthetic versus analytic case marking . . . . . . . . . . . . . . . . . . . . 213
2 Case features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
3 Case assignment configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
3.1 Head-comp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
3.2 Spec-head . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222
3.3 Head-spec . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
4 Further issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
4.1 Subject movement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
5 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
6 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231

8 Passive 236
1 Characteristics of the passive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
2 A movement analysis of the passive . . . . . . . . . . . . . . . . . . . . . . . . . . 239
2.1 Selectional restrictions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
2.2 Object idiom chunks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
2.3 Passive of simple sentences . . . . . . . . . . . . . . . . . . . . . . . . . . 242
2.4 Passive of ECM sentences . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
3 Further issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
3.1 Impersonal passive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
4 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250
5 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251

9 Subject control versus raising 254

1 Subject control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256
1.1 Evidence for two clauses . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256
1.2 Deriving subject control sentences . . . . . . . . . . . . . . . . . . . . . . 258
2 Raising . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
2.1 A “what if” detour . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
2.2 Nonthematic subject positions . . . . . . . . . . . . . . . . . . . . . . . . . 262
2.3 Deriving raising sentences . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
3 Some special verbs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
3.1 Tend versus occur . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
3.2 Promise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
4 More nonthematic subjects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
4.1 Subject idiom chunks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
Contents viii

4.2 Weather it . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271

4.3 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
5 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272
6 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274

10 The verb raising parameter 280

1 Verb raising: V moves to I in the syntax . . . . . . . . . . . . . . . . . . . . . . . . 282
1.1 The future tense in French . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
1.2 The order of diagnostic adverbs and verbs in French . . . . . . . . . . . . 284
2 Tense lowering: I moves to V in the morphology . . . . . . . . . . . . . . . . . . . 289
2.1 The order of diagnostic adverbs and verbs in English . . . . . . . . . . . . 289
2.2 Do support in English . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290
3 The loss of verb raising in Scandinavian . . . . . . . . . . . . . . . . . . . . . . . . 294
4 Verb raising in the history of English . . . . . . . . . . . . . . . . . . . . . . . . . 297
4.1 The loss of verb raising . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
4.2 A change in the status of not . . . . . . . . . . . . . . . . . . . . . . . . . 299
4.3 The emergence of do support . . . . . . . . . . . . . . . . . . . . . . . . . 300
4.4 The emergence of modals . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
4.5 Remnants of verb raising in modern English . . . . . . . . . . . . . . . . . 304
5 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
6 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309

11 VP shells 314
1 The causative alternation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316
2 Double object sentences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318
3 Double complement sentences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322
3.1 Give and send . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322
3.2 Put . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
3.3 Persuade . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326
4 Object control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327
5 Passive and VP shells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329
6 Further issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 331
6.1 A locality constraint on idioms . . . . . . . . . . . . . . . . . . . . . . . . 331
6.2 Direct vs. indirect causation . . . . . . . . . . . . . . . . . . . . . . . . . . 333
6.3 Semantic nuances: Give, get, send . . . . . . . . . . . . . . . . . . . . . . . 335
6.4 Double object verbs in languages with inherent case . . . . . . . . . . . . 338
7 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344
8 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345

12 The verb-second (V2) phenomenon 350

1 Remnants of V2 in modern English . . . . . . . . . . . . . . . . . . . . . . . . . . 351
1.1 Negative inversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351
1.2 So inversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352
2 V2 in Middle English . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353
3 V2 in German . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355
Contents ix

4 The loss of V2 in English . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 359

5 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 360
6 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 360

13 Constraints on wh- movement 363

1 Ross’s island constraints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 365
2 Subjacency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367
2.1 Two possible derivations for long-distance wh- movement . . . . . . . . . 367
2.2 IP as a barrier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367
2.3 DP as a barrier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 369
3 Further issues and considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . 369
3.1 The status of nonfinite IP . . . . . . . . . . . . . . . . . . . . . . . . . . . . 369
3.2 The coordinate structure constraint revisited . . . . . . . . . . . . . . . . . 370
3.3 The status of DP as a barrier revisited . . . . . . . . . . . . . . . . . . . . . 371
3.4 Argument-adjunct asymmetries . . . . . . . . . . . . . . . . . . . . . . . . 372
3.5 The Comp-trace effect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373
4 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374
5 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375

14 Cross-linguistic variation in wh- movement 377

1 To move or not to move? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378
1.1 Wh- scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378
1.2 Wh- in situ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378
1.3 A copy theory of movement . . . . . . . . . . . . . . . . . . . . . . . . . . 379
1.4 Partial wh- movement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 381
1.5 But what if they are errors? . . . . . . . . . . . . . . . . . . . . . . . . . . 383
2 Where to move? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384
2.1 Hungarian questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384
2.2 Hungarian relative clauses . . . . . . . . . . . . . . . . . . . . . . . . . . . 386
3 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387
4 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 388

15 Binding theory 389

1 Reference and related notions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 390
1.1 Reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 390
1.2 Coreference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 391
1.3 Coindexing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 391
1.4 Grammatical and ungrammatical index assignments . . . . . . . . . . . . 392
2 Hellan 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 393
2.1 The co-argument condition . . . . . . . . . . . . . . . . . . . . . . . . . . 393
2.2 The predication condition . . . . . . . . . . . . . . . . . . . . . . . . . . . 397
2.3 The finite IP condition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 398
2.4 Strict vs. non-strict co-arguments . . . . . . . . . . . . . . . . . . . . . . . 398
3 Extending Hellan’s binding theory to English . . . . . . . . . . . . . . . . . . . . . 401
3.1 The co-argument condition . . . . . . . . . . . . . . . . . . . . . . . . . . 401
Contents x

3.2 The predication condition . . . . . . . . . . . . . . . . . . . . . . . . . . . 403

3.3 The finite IP condition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 403
4 Chomsky (1981) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 403
4.1 Principle A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 404
4.2 Principle B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 408
4.3 Principle C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409
5 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 411

Glossary 412

References 420
 1
Foundational issues

Chapter outline
1 Prescriptive versus descriptive grammar . . . . . . . . . . . . . . . . . . . . . . . 2
2 Rule formation and syntactic structure in language acquisition . . . . . . . . . . 6
2.1 A thought experiment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2 Rule-based word formation . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.3 Question formation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3 More evidence for syntactic structure . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.1 Intuitions about words belonging together . . . . . . . . . . . . . . . . . . 12
3.2 Structural ambiguity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
4 Universal Grammar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
4.1 Formal universals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
4.2 Recursion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
4.3 Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
5 Generative grammar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
5.1 Elementary trees and substitution . . . . . . . . . . . . . . . . . . . . . . 18
5.2 Grammaticality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
5.3 Grammar versus language . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
6 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
7 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Chapter 1: Foundational issues 2

Prescriptive versus descriptive grammar 1

This book is an introduction to generative grammar from a Chomskyan perspective. By the time
you finish this chapter, you will have a clearer understanding of what we mean by this sentence,
and by the time you finish the entire book, your understanding of it will be clearer and deeper still.
But for the moment, you have probably gained the impression that this book is about grammar
of some sort. And right there, we have a problem. The problem is that there is an everyday sense
of the term ‘grammar’ and a quite different sense in which the term is used in linguistics.
In the everyday sense, ‘grammar’ refers to a collection of rules concerning what counts as
socially acceptable and unacceptable language use. Some of these rules, like the ones in (1), make
reference to particular words and apply to both spoken and written language.

(1) a. Don’t use ain’t.

b. Don’t use seen as the past tense of see (as in I seen him at the party last night).

But mainly, the rules in question concern the proper composition of sentences in written
language. You may recall being taught rules at school like those in (2).

(2) Selected prescriptive grammatical rules for English

a. Don’t end a sentence with a linking verb.
b. Don’t end a sentence with a preposition.
c. Don’t split infinitives.
d. Don’t start a sentence with a conjunction.
e. Don’t use a plural pronoun to refer back to a singular noun like everyone, no-one,
someone, and the like.
f. Don’t use an object pronoun for a subject pronoun in a conjoined subject.
g. Don’t use contractions.
h. Don’t use dangling participles.
i. Don’t use sentence fragments.
j. Use whom, not who, as the object of a verb or preposition.

Someone who composes sentences in accordance with rules like those in (2) is said to have
good grammar, whereas someone said to have bad grammar doesn’t apply the rules when they
ought to be applied1 and so produces sentences like (3).

(3) a. Over there is the guy who I went to the party with. violates (2b), (2j)
b. Bill and me went to the store. violates (2f)

From the amount of attention that people devote to rules like those in (1) and (2), it is easy
to get the impression that they are the only linguistic rules there are. But it is also easy to see
Chapter 1: Foundational issues 3

that that can’t be so. The reason is that even people who don’t follow the rules in (1) and (2) don’t
produce rampantly variable, confusing word salad. For instance, even people who invariably
produce sentences like (3) do not produce the likes of (4).

(4) a. Over there is guy the who I went to party the with.
b. Over there is the who I went to the party with guy.
c. Bill and me the store to went.

The sentences in (3) may be instances of bad grammar in the sense of the term “grammar”
that is often used in English classes in elementary and secondary school. But they are nonetheless
still English sentences that people will readily say in conversation and will accept as possible
sentences in their language. By contrast, we don’t need to rely on school rules to tell us that the
examples in (4) are not English sentences—even though they contain exactly the same English
words as the sentences in (3).
Since native speakers of English do not produce a variable mishmash of words of the sort
in (4), there must be another type of rules according to which sentences are composed. We can
determine what some of them are by taking a closer look at the sequences in (4). Why exactly is
it that they are word salad? In (4a), the article the is in the wrong order with respect to the nouns
that it belongs with, guy and party. In (4b), the relative clause (who I went to the party with) is in
the wrong order with respect to the noun that it modifies (guy). In (4c), the preposition to is in
the wrong order with respect to its object (the store). In other words, the sentences in (4) do not
follow the rules in (5).

(5) a. Articles precede nouns.

b. Relative clauses follow the noun that they modify.
c. Prepositions precede their objects.

There’s another rule that’s not followed in (4), which you are asked to formulate in Exer-
cise 1.1.
Rules like those in (5) have a different intention than those in (2). The rules in (2) are
prescriptive; those in (5) are descriptive. Rules of prescriptive grammar have the same sta-
tus as rules of etiquette (like table manners or dress codes) or the laws of society, which divide
the spectrum of possible human behavior into socially acceptable or legal behavior, on the one
hand, and socially unacceptable or illegal behavior, on the other. Rules of prescriptive grammar
make statements about how people ought to use language, and often they are framed as negative
injunctions (“Thou shalt not split infinitives”, on a par with “Thou shalt not steal”). In contrast,
rules of descriptive grammar have the status of scientific observations, and they are intended as
insightful generalizations about the way that speakers use language in fact, rather than about
the way that they ought to use it. Descriptive rules are more general and more fundamental
than prescriptive rules in the sense that all sentences of a language are formed in accordance
with them. Beyond this, most prescriptive rules have a classist and often racist foundation in the
US, in that only certain dominant groups’ language counts as “standard”/acceptable, whereas the
language of socially marginalized communities is itself marginalized as “bad” grammar. More on
that below.
Chapter 1: Foundational issues 4

A useful way to think about the descriptive rules of a language (to which we return in more
detail below) is that they produce, or generate, all the sentences of a language. The prescriptive
rules can then be thought of social control: attempts to filter out some (relatively minute) portion
of the entire output of the descriptive rules as socially unacceptable.
In syntax, as in modern linguistics more generally, we adopt a resolutely descriptive per-
spective concerning language. In particular, when linguists say that a sentence is grammatical,
we don’t mean that it is correct from a prescriptive point of view, but rather that it conforms to
descriptive rules like those in (5). In order to indicate that a sequence of words or morphemes
is ungrammatical in this descriptive sense, we prefix it with an asterisk. Grammatical sentences
are usually not specially marked, but sometimes we prefix them with a checkmark (✔) for clarity.
These conventions are illustrated in (6) and (7).

(6) a. * Over there is guy the who I went to party the with. (= (4a))
b. * Over there is the who I went to the party with guy. (= (4b))

(7) a. ✔ Over there is the guy who I went to the party with. (= (3a))
b. ✔ Over there is the guy with whom I went to the party.

That is to say, by saying that the sentences in (7) are grammatical, we are describing a psycho-
logical phenomenon that is automatic. Nobody who speaks Mainstream U.S. English needs to be
taught that the sentences in (7) are grammatical but the sentences in (6) are not: everyone already
knows this.
Prescriptive grammar is based on the idea that there is a single right way to do things. When
there is more than one way of saying something, prescriptive grammar is generally concerned
with declaring one (and only one) of the variants to be correct. The favored variant is usually
justified as being better (whether more logical, more euphonious, or more desirable on some other
grounds) than the marginalized variant that is being discriminated against. In the same situation
of linguistic variability, descriptive grammar is content simply to document the variants—without
passing judgment on them. For instance, consider the variable subject-verb agreement pattern in
(8).

(8) a. There’s some boxes left on the porch.

b. There are some boxes left on the porch.

In (8a), the singular verb is (contracted to ’s) agrees in number with the preverbal expletive
subject there (in bold), whereas in (8b), the plural verb are agrees with the postverbal logical
subject some boxes (in italics). The color and font of the verb indicates which of the two subjects
it agrees with.
The prescriptive and descriptive rules concerning this pattern are given in (9). The differ-
ences between the two rules are emphasized by underlining.
Chapter 1: Foundational issues 5

(9) In a sentence containing both the singular expletive subject there and a plural logical
subject …
a. Prescriptive rule: … the verb should agree in number with the logical subject.
b. Descriptive rule: … the verb can agree in number with either the expletive subject
or the logical subject.

To take another example, let’s consider the prescriptive rule that says, “Don’t end a sen-
tence with a preposition.”2 A prescriptivist might argue that keeping the preposition (in italics)
together with its object (in boldface), as in (10a), makes sentences easier to understand than does
separating the two, as in (10b).

(10) a. With which friend did you go to the party?

b. Which friend did you go to the party with?

But by that reasoning, (11a), where the verb and its object are adjacent, is preferable to (11b),
where they are not. In fact, however, (11a) is completely ungrammatical in English.

(11) a. * Adopt which cat did your friend?

b. ✔ Which cat did your friend adopt?

Beyond that, the prescriptive rule against sentences ending with prepositions is quite
clearly not a rule that is part of our mental grammars. This is illustrated by the (perhaps apoc-
ryphal) story about Winston Churchill’s response to this rule:

(12) This is the type of arrant pedantry up with which I will not put.

(12) is an absurdly weird sentence: clearly, it’s not a rule of mental grammars of English that
prepositions can’t end sentences.
It is important to understand that there is no semantic or other conceptual reason that
prepositions can be separated from their objects in English, but that verbs can’t. From a de-
scriptive perspective, the grammaticality contrast between (10a) and (11a) is simply a matter of
fact, irreducible to more basic considerations (at least given our present state of knowledge). (13)
highlights the difference between the relevant prescriptive and descriptive rule.

(13) When the object of a preposition appears in a position other than its ordinary one (as in
a question), …
a. Prescriptive rule: … it should be preceded by the preposition.
b. Descriptive rule: … it can either be preceded by the preposition, or stand alone,
with the preposition remaining in its ordinary position.

The contrasting attitude of prescriptive and descriptive grammar towards linguistic varia-
tion has a quasi-paradoxical consequence: namely, that prescriptive rules are never descriptive
rules. The reason for this has to do with the way that social systems (not just language) work.
If everyone in a community consistently behaves in a way that is socially acceptable in some re-
spect, then there is no need for explicit prescriptive rules to ensure the behavior in question. It is
only when behavior that is perceived as socially unacceptable becomes common that prescriptive
Chapter 1: Foundational issues 6

rules come to be formulated to keep the unacceptable behavior in check. For example, if every
customer entering a store invariably wears both a shirt and shoes, there is no need for the store
owner to put up a sign that says “No shirt, no shoes, no service.” Conversely, it is precisely at
illegal dump sites that we observe “No dumping” signs. In an analogous way, in the domain of
language use, rules of prescriptive grammar are only ever formulated in situations where linguis-
tic variation is common. But being prescriptive, they cannot treat all of the occurring variants as
equally acceptable—with the result that they can’t ever be descriptive. So prescriptive rules only
ever exist to attempt to outlaw grammatical constructions that are in the mental grammars of
many speakers of a language.
We can’t emphasize enough that the modern basis of a vast amount of prescriptive gram-
mar teaching is largely classist and often racist: certain kinds of language have been deemed
acceptable, and those deemed unacceptable are those of social groups that have been marginal-
ized in society. This is not to impugn the motives of any particular English teacher: they are
just doing their job as they have been trained to do it. But a system has arisen that centers a
particular variety of English and marginalizes a host of other perfectly legitimate varieties of
English. Lippi-Green (2012) is a good first resource on this topic, though students very inter-
ested in these social phenomena should take a sociolinguistics course, which will cover these
issues in much more depth. In the realm of syntax specifically, the Yale Grammatical Diversity
Project (https://ygdp.yale.edu) is an excellent resource, as it provides linguistic descriptions of
grammatical constructions that are often the target of prescriptive rules.

Rule formation and syntactic structure in language

acquisition 2
As we have just seen, prescriptive and descriptive rules of grammar differ in intention. In ad-
dition, they differ in how they come to be part of a speaker’s knowledge. Prescriptive rules are
taught at school, and because they are taught, people tend to be conscious of them, even if they
don’t actually follow them. By contrast, we follow the rules of descriptive grammar consistently3
and effortlessly, yet without learning them at school. In fact, children have essentially mastered
these rules on their own by first grade. Ordinarily, we are completely unconscious of the de-
scriptive rules of language. If we do become conscious of them, it tends to be in connection with
learning a foreign language whose descriptive grammar differs from that of our native language.
In order to emphasize the difference between the unconscious way that we learn a native language
(or several) in early childhood and the conscious way that we learn a foreign language later on in
life, the first process is often called language acquisition rather than language learning.
As you consider descriptive rules like those in (5), you might not find it all that surprising
that a child raised in an English-speaking community would acquire, say, the rule that articles
precede nouns. After all, you might say, all the child ever hears are articles and nouns in that
order. (Hmmm, though, actually… see the note.)4 So why would it ever occur to such a child to
put the article and the noun in the other order? Isn’t it just common sense that children learn
Chapter 1: Foundational issues 7

their native language by imitating older speakers around them?

Well, yes and no. It is true that children learn some aspects of their native language by
imitation and memorization. Children in English-speaking communities learn English words,
children in Navajo-speaking communities learn Navajo words, children in Swahili-speaking com-
munities learn Swahili words, and so on. But language acquisition isn’t purely a process of mem-
orization. In fact, given current human life spans, it couldn’t possibly be!

2.1 A thought experiment

To see this, let’s consider a toy version of English that contains three-word sentences consisting
of a noun, a transitive verb, and another noun. The toy version contains sentences like (14) that
are sensible given the real world as well as sentences like (15) that aren’t, but that might be useful
in fairy tale or science fiction contexts.

(14) a. Cats detest lemons.

b. Children eat tomatoes.
c. Cheetahs chase gazelles.

(15) a. Lemons detest cats.

(“Secret life of citrus fruits”)
b. Tomatoes eat children.
(“Attack of the genetically modified tomatoes”)
c. Gazelles chase cheetahs.
(“Avenger gazelle”)

Again for the sake of argument, let’s assume a (small) vocabulary of 1,000 nouns and 100
verbs. This gives us a list of 1,000 × 100 × 1,000 (= 100 million) three-word sentences of the
type in (14) and (15). Numbers of this magnitude are difficult to put in human perspective, so
let’s estimate how long it would take a child to learn all the sentences on the list. Again, for
the sake of argument, let’s assume that children can memorize sentences quickly, at a rate of
one sentence a second. The entire list of three-word sentences could then be memorized in 100
million seconds, which comes to 3.17 years. So far, so good. However, the minute we start adding
complexity to Toy English, the number of sentences and the time it would take to memorize them
quickly mushrooms. For instance, adding only 10 adjectives to the child’s vocabulary would cause
the number of five-word sentences of the form in (16) to grow to 10 billion (100 million × 10 ×
10).

(16) a. Black cats detest green peas.

b. Happy children eat ripe tomatoes.
c. Hungry cheetahs chase speedy gazelles.

Even at the quick rate of one sentence per second that we’re assuming, the list of all such
five-word sentences would take a bit over 317 years to learn. Clearly, this is an absurd con-
sequence. For instance, how could our memorious child ever come to know, as every English
Chapter 1: Foundational issues 8

speaker plainly does, that the sentence in (17) is ungrammatical? If grammatical knowledge
were based purely on rote memorization, the only way to determine this would be to compare
(17) to all of the 10 billion five-word sentences and to find that it matches none of them.

(17) * Cats black detest peas green.

And even after performing the comparison, our fictitious language learner still wouldn’t have the
faintest clue as to why (17) is ungrammatical!
In addition to this thought experiment with its comically absurd consequences, there is
another reason to think that language acquisition isn’t entirely based on rote memorization—
namely, that children use what they hear of language as raw material to construct linguistic
rules. How do we know this? We know because children sometimes produce rule-based forms
that they have never heard before.

2.2 Rule-based word formation

One of the earliest demonstrations that children acquire linguistic rules, rather than simply imi-
tating the forms of adult language, is the wug experiment (Berko 1958). In it, the psycholinguist
Jean Berko Gleason used invented words to examine (among other things) how children between
the ages of 4 and 7 form plurals in English. She showed the children cards with simple line draw-
ings of objects and animals and elicited plurals from them by reading them accompanying texts
like ((18)).5

(18) This is a wug. Now there is another one. There are two of them. There are two .

(19) Wugs, for reference

More than 75% of the children pluralized the invented words cra, lun, tor, and wug in exactly
the same way that adults did in a control group: they added the sound -z to the word (Berko
1958, pp. 159–162).6 Since none of the children had encountered the invented words before the
experiment, their response clearly indicates that they had acquired a plural rule and were using
it to produce the novel forms.
Children are also observed to produce novel rule-based forms instead of existing irregular
adult forms (for instance, comed or goed instead of came or went). This process, which is known
as overregularization, is further illustrated in (20) (Marcus et al. 1992, pp. 148–149; based on
Brown 1973).

(20) a. beated, blowed, catched, cutted, doed, drawed, drived, falled, feeled, growed, holded,
maked, sleeped, standed, sticked, taked, teached, throwed, waked, winned
(Adam, between the ages of 2 and 5)
b. drinked, seed, weared (Eve, between the ages of 1½ and 2)
Chapter 1: Foundational issues 9

Overregularized forms don’t amount to a large fraction of the forms that children produce
overall (less than 5% in the case of past tense forms, according to Marcus et al. 1992, p. 35), but they
are important because they clearly show that even the acquisition of words can’t be completely
reduced to rote memorization.

2.3 Question formation

In addition to morphological rules (which concern the structure of words), children also acquire
syntactic rules (which concern the structure of sentences). Some of these rules are of particular
interest because they differ from the corresponding adult rules that the children eventually ac-
quire. At the same time, however, the children’s novel rules don’t differ from the rules of the
adult grammar in completely arbitrary ways. Rather, the children’s rules share certain abstract
properties with the adult rules, even when they differ from them.
To see this, let’s consider how young children form yes-no questions. Some 3- to 5-year-
olds form such questions from declarative sentences by copying the auxiliary element to the
beginning of the sentence, as in (21) (Crain and Nakayama 1987, p. 536). (We use the term ‘aux-
iliary element’ as a convenient cover term for elements that invert with the subject in (adult)
English questions, like forms of the verb to be or modals like can. See § 1.4 of Chapter 2 for more
details.)

(21) a. The girl is tall. → Is the girl is tall?

b. The red pig can stand on the house. → Can the red pig can stand on the house?

In the course of language acquisition, the questions in (21) are eventually replaced by those in
(22), where we can think of the auxiliary element as having been moved rather than copied.

(22) a. Is the girl tall?

b. Can the red pig stand on the house?

But now notice a striking indeterminacy, first pointed out by Chomsky (1971, pp. 26–27).
When children produce questions like those in (22), there is no way of telling whether they are
using the adult rule for question formation in (23a) or the logically possible alternative rule in
(23b).

(23) a. Adult question formation rule:

To form a question from a declarative sentence containing an auxiliary element, find
the subject of the sentence, and invert the subject and the auxiliary.
b. Logically possible alternative:
To form a question from a declarative sentence containing an auxiliary element, find
the first auxiliary element, and move it to the beginning of the sentence.
Chapter 1: Foundational issues 10

Word of Caution
Don’t confuse ‘subject’ with ‘simple subject.’ Subjects, in contrast to simple subjects, are
possible responses to questions like Who is tall? and Who can stand on the house? The
subjects in (22) are the noun phrases the girl and the red pig.

If the subject consists of a single word or a clause, then the simple subject is identical
to the subject; otherwise, the simple subject of a sentence is obtained by stripping the
subject of any modifiers (yielding girl and pig as the simple subjects of (22)). The notion
of subject is basic to syntactic theory, but we will have no further use for the notion of
simple subject.

Both rules in (23) give the same result for simple sentences, which are likely to form most
of the data that young children attend to. Both rules also require children to identify auxil-
iary elements. However, the adult rule additionally requires children to identify the subject of
the sentence by grouping together sequences of words like the girl or the red pig into a single
abstract structural unit. Because of this grouping requirement, the adult rule is called structure-
dependent. By contrast, the alternative rule in (23b) is not structure-dependent, since it requires
the child only to classify words according to their syntactic category (Is this word an auxiliary
element?), but not to group the words into structural units. The rule in (23b) is simpler in the sense
that it relies on fewer, as well as computationally less complex, cognitive operations, and children
might reasonably be expected to experiment with it in the course of acquiring question forma-
tion. Nevertheless, Chomsky (1971) predicted that children would use only structure-dependent
rules in the course of acquisition.
As we mentioned, both rules give the same result for simple sentences. So how could we
possibly tell which of the two rules a child was actually using? Well, forming yes-no questions
is not restricted to simple sentences. So although we can’t tell which rule a child is using in the
case of simple sentences like (21), the rules in (23) give different results for a complex sentence
like (24), which contains a relative clause (who was holding the plate).

(24) The boy who was holding the plate is crying.

In particular, the sentence in (24) contains two auxiliary elements—one (was) for the rel-
ative clause, and another one (is) for the entire sentence (the so-called matrix sentence, which
contains the relative clause). A child applying the structure-dependent question formation rule
to (24) would first identify the subject of the matrix sentence (the boy who was holding the plate)
and then invert the entire subject—including the relative clause and the auxiliary contained
within it (was)—with the matrix auxiliary (is). On the other hand, a child applying the structure-
independent rule would identify the first auxiliary (was) and move it to the beginning of the
sentence. As shown in (25), the two rules have very different results,

(25) a. Structure-dependent rule:

[ The boy who was running ] is crying.
Is [ the boy who was running ] crying?
Chapter 1: Foundational issues 11

b. Structure-independent rule:
The boy who was running is crying.
Was the boy who running is crying?

Recall that Chomsky predicted that children would not use structure-independent rules,
even though they are (in some sense) simpler than structure-dependent ones. This prediction
was tested in an experiment with 3- to 5-year-old children by Crain and Nakayama (1987). In
the experiment, the experimenter had the children pose yes-no questions to a doll (Jabba the Hut
from Star Wars). For instance, the experimenter would say to each child Ask Jabba if the boy who
was holding the plate is crying. This task elicited various responses. Some children produced the
adult question in (25a), whereas others produced the copy question in (26a) or the restart question
in (26b).

(26) a. Is [the boy who was holding the plate] is crying?

b. Is [the boy who was holding the plate], is he crying?

Although neither of the questions in (26) uses the adult rule in (23a), the rules that the
children used to produce them are structure-dependent in the same way that the adult rule is.
This is because children who produced (26a) or (26b) must have identified the subject of the
sentence, just like the children who produced (25a). Out of the 155 questions that the children
produced, none were of the structure-independent type in (25b). Moreover, no child produced
the structure-independent counterpart of (26a), shown in (27), which results from copying (rather
than moving) the first auxiliary element in the sentence.

(27) Was the boy who was holding the plate is crying?

In other words, regardless of whether a child succeeded in producing the adult question in
(25a), every child in the experiment treated the sequence the boy who was holding the plate as a
structural unit, thus confirming Chomsky’s prediction.

More evidence for syntactic structure 3

We have seen that young children are capable of forming and applying both morphological and
syntactic rules. Moreover, as we have seen in connection with question formation, children do
not immediately acquire the rules of the adult grammar. Nevertheless, the syntactic rules that
children are observed to use in the course of acquisition are a subset of the logically possible rules
that they might postulate in principle. In particular, as we have just seen, children’s syntactic
rules are structure-dependent, even when they differ from the target adult rules. Another way of
putting this is that the objects that syntactic rules operate on (declarative sentences in the case
Chapter 1: Foundational issues 12

of the question formation rule) are not simply strings of words, but rather groups of words that
belong together, so-called syntactic constituents.

3.1 Intuitions about words belonging together

Evidence for syntactic structure isn’t restricted to data from child language acquisition. Further
evidence comes from the intuitions that speakers, whether adults or (older) children, have that
certain words in a sentence belong together, whereas others do not. For instance, in a sentence
like (28), we have the strong intuition that the first the belongs with dog, but not with did, even
though the is adjacent to both.

(28) Did the dog chase the cat?

Similarly, the second the in (28) belongs with cat and not with chase. But a word doesn’t
always belong with the following word. For instance, in (29), dog belongs with the preceding the,
not with the following the.

(29) Did the dog the children like chase the cat?

Words that belong together can sometimes be replaced by placeholder elements such as
pronouns, as illustrated in (30).

(30) a. Did the dog chase the cat? → Did she chase him?
b. Did the dog the children like chase the cat? → Did the dog they like chase him?

Extra Info
Strictly speaking, the term ‘pronoun’ is misleading since it suggests that pronouns sub-
stitute for nouns regardless of syntactic context. In fact, pronouns substitute for noun
phrases (as will be discussed in more detail in Chapter 2). A less confusing term would be
‘pro-noun phrase,’ but we’ll continue to use the traditional term.

It’s important to recognize that pronouns don’t simply replace strings of words regardless
of context. Just because a string like the dog is a constituent in (30a) doesn’t mean that it’s
always a constituent. We can see this by replacing the dog by a pronoun in (30b), which leads to
the ungrammatical result in (31).

(31) Did the dog the children like chase the cat?
↓
* Did she the children like chase the cat?

The ungrammaticality in (31) is evidence that the and dog belong together less closely in
(30b) than in (30a). In particular, in (30b), dog combines with the relative clause, and the combines
with the result of this combination, not with dog directly, as it does in (30a). (We will consider
the internal structure of noun phrases more closely in Chapter 5.)
Chapter 1: Foundational issues 13

In some sentences, we have the intuition that words belong together even when they are
not adjacent. For instance, see and who in (32a) belong together in much the same way as see and
Bill do in (32b).

(32) a. Who will they see?

b. They will see Bill.

Finally, we can observe that there are various sorts of ways that words can belong together.
For instance, in a phrase like the big dog, big belongs with dog, and we have the intuition that
big modifies dog. On the other hand, the relation between see and Bill in (32b) isn’t one of
modification. Rather, we have the intuition that Bill is a participant in a seeing event.
In the course of this book, we will introduce more precise ways of expressing and repre-
senting intuitions like the ones just discussed. For the moment, what is important is that we have
strong intuitions that words belong together in ways that go beyond adjacency.

3.2 Structural ambiguity

A particularly striking piece of evidence for the existence of syntactic structure is the phe-
nomenon of structural ambiguity. The classified advertisement in (33) is a humorous illus-
tration.

(33) Wanted: Man to take care of cow that does not smoke or drink.

World knowledge tells us that the intent of the advertiser is to hire a clean-living man to take care
of a cow. But because of the way the advertisement is formulated, it also has an unintentionally
comical interpretation—namely, that the advertiser has a clean-living cow and that the advertiser
wants a man (possibly a chain-smoking alcoholic) to take care of this animal. The intended and
unintended interpretations describe sharply different situations; that is why we say that (33) is
ambiguous (like the two mutually exclusive interpretations of an optical illusion), rather than that
it is vague (like a blurry image). Moreover, the ambiguity of (33) can’t be pinned on a particular
word, as is possible in ambiguous sentences like those in (34).

(34) a. As far as I’m concerned, any gender is a drag. (Patti Smith)

b. Our bikinis are exciting. They are simply the tops.

Sentences like those in (34) are examples of lexical ambiguity; their ambiguity is based on a lexeme
(= vocabulary item) with two distinct meanings. In (33), on the other hand, the words themselves
have the same meanings in each of the two interpretations, and the ambiguity derives from the
possibility of grouping the words in distinct ways. In the intended interpretation, the relative
clause that does not smoke or drink modifies man; in the unintended interpretation, it modifies
cow.
To avoid any confusion, we should emphasize that we are here considering structural ambi-
guity from a purely descriptive perspective, focusing on what it tells us about the design features
of human language and disregarding the practical issue of effective communication. As writers
of advertisements ourselves, we would take care not to use (33), but to disambiguate it by means
Chapter 1: Foundational issues 14

of an appropriate paraphrase. For the ordinary interpretation of (33), where the relative clause
modifies man, we might move the relative clause next to the intended modifier, as in (35a). The
comical interpretation of (33), on the other hand, cannot be expressed unambiguously by moving
the relative clause. If it were the desired interpretation, we would have to resort to a more drastic
reformulation, such as (35b).

(35) a. Wanted: Man that does not smoke or drink to take care of cow.
b. Wanted: Man to take care of nonsmoking, nondrinking cow.

The topic of our investigation in this book is not about editorial questions like when sentences
like (35) should replace sentences like (33), though the skills you build here will surely help with
editorial precision. Rather, our focus is on the more foundational issues: why does human lan-
guage have the characteristics that it does? For example, why do ambiguities like these only arise
in certain contexts and not others?

Universal Grammar 4

4.1 Formal universals

The structure-dependent character of syntactic rules is a general property of the human lan-
guage faculty (the part of the mind/brain that is devoted to language), often also referred to as
Universal Grammar, especially when considered in abstraction from any particular language.
There are two sources of evidence for this. First, as we have seen, the syntactic rules that chil-
dren acquire are structure-dependent (even when they are not the rules that adults use). Second,
even though structure-independent rules are logically possible and computationally tractable,
no known human language actually has rules that disregard syntactic structure as a matter of
course. For instance, (36) gives two examples of computationally very simple rules for question
formation, but no known human language has rules of this type.

(36) a. To form a question, switch the order of the first and second words in the corresponding
declarative sentence.
The girl is tall. → Girl the is tall?
The blond girl is tall. → Blond the girl is tall?
b. To form a question, reverse the order of the words in the corresponding declarative
sentence.
The girl is tall. → Tall is girl the?
The blond girl is tall. → Tall is girl blond the?

The structure-dependent character of syntactic rules (often referred to more briefly as

structure dependence) is what is known as a formal universal of human language—a property
Chapter 1: Foundational issues 15

common to all human languages that is independent of the meanings of words. Formal univer-
sals are distinguished from substantive universals, which concern the substance, or meaning,
of linguistic elements. An example of a substantive universal is the fact that all languages have
indexical elements such as I, here, and now. These words have the special property that their
meanings are predictable in the sense that they denote the speaker, the speaker’s location, and the
time of speaking, but that what exactly they refer to depends on the identity of the speaker.

4.2 Recursion
Another formal universal is the property of recursion. A simple illustration of this property is
the fact that it is possible for one sentence to contain another. For instance, the simple sentence
in (37a) forms part of the complex sentence in (37b), and the resulting sentence can form part
of a still more complex sentence. Recursive embedding is illustrated in (37) up to a level of five
embeddings.

(37) a. She won.

b. The Times reported that
[she won].
c. John told me that
[the Times reported that
[she won]].
d. I remember distinctly that
[John told me that
[the Times reported that
[she won]]].
e. They don’t believe that
[I remember distinctly that
[John told me that
[the Times reported that
[she won]]]].
f. I suspect that
[they don’t believe that
[I remember distinctly that
[John told me that
[the Times reported that
[she won]]]]].
Chapter 1: Foundational issues 16

4.3 Parameters
Formal universals like structure dependence and recursion are of particular interest to linguistics
in the Chomskyan tradition. This is not to deny, however, that individual languages differ from
one another, and not just in the sense that their vocabularies differ. In other words, Universal
Grammar is not completely fixed, but allows some variation. The ways in which grammars can
differ are called parameters.
One simple parameter concerns the order of verbs and their objects. In principle, two orders
are possible: verb-object (VO) or object-verb (OV), and different human languages use either one
or the other. As illustrated in (38) and (39), English and French are languages of the VO type,
whereas Hindi, Japanese, and Korean are languages of the OV type.

(38) a. Peter read the book. [English]

b. Pierre lisait le livre. [French]
Pierre was.reading the book
‘Pierre was reading the book.’

(39) a. पिवत्र ने िकताब पढ़ी । [Hindi]

Pavitr-ne kitaab parh-ii.
Pavitr book read
‘Pavitr read the book.’
b. 漣が 本を 読んだ。 [Japanese]
Ren-ga hon-o yon-da.
Ren book read
‘Ren read the book.’
c. 현서가 책을 읽었다. [Korean]
Hyense-ka chayk-ul ilk-ess-ta.
Hyense book read
‘Hyense read the book.’

Another parameter of Universal Grammar concerns the possibility, mentioned earlier in

connection with prescriptive rules, of separating a preposition from its object, or preposition
stranding. (The idea behind the metaphor is that the movement of the object of the preposition
away from its ordinary position leaves the preposition stranded high and dry.) The parametric
alternative to preposition stranding goes by the name of pied piping,7 by analogy to the Pied
Piper of Hamelin, who took revenge on the citizens of Hamelin for mistreating him by luring the
town’s children away with him. In pied piping of the syntactic sort, the object of the preposition
moves away from its usual position, just as in preposition stranding, but it takes the preposition
along with it. The two parametric options are illustrated in (40). (You’ll note that a single language
can exhibit two parameter settings. We return to this issue later on in the chapter, in the section
on Grammar versus language.)

(40) a. Preposition stranding: ✔ Which house does your friend live in?
b. Pied piping: ✔ In which house does your friend live?
Chapter 1: Foundational issues 17

Just as in English, preposition stranding and pied piping are both grammatical in Swedish.
(In Swedish, it is preposition stranding that counts as prescriptively correct! Pied piping is
frowned upon, on the grounds that it sounds stiff and artificial.)

(41) Swedish
a. ✔ Vilket hus bor din kompis i?
which house lives your friend in
‘Which house does your friend live in?’
b. ✔ I vilket hus bor din kompis?

In other languages, such as French and Italian, preposition stranding is actually ungram-
matical. Speakers of these languages reject examples like (42b) and (43b) as word salad, and accept
only the corresponding pied-piping examples in (42a) and (43a).

(42) a. ✔ Dans quelle maison est-ce que ton ami habite? [French]
in which house is it that your friend lives
‘Which house does your friend live in?’
b. * Quelle maison est-ce que ton ami habite dans?
which house is it that your friend lives in
Intended meaning: ‘Which house does your friend live in?’

(43) a. ✔ In quale casa abita il tuo amico? [Italian]

in which house lives the your friend
‘Which house does your friend live in?’
b. * Quale casa abita il tuo amico in?
which house lives the your friend in
Intended meaning: ‘Which house does your friend live in?’

Generative grammar 5
At the beginning of this chapter, we said that this book was an introduction to generative gram-
mar from a Chomskyan perspective. Until now, we have clarified our use of the term ‘grammar,’
and we have explained that a Chomskyan perspective on grammar is concerned with the formal
principles that all languages share as well as with the parameters that distinguish them. Let’s
now turn to the notion of a generative grammar.

(44) A generative grammar is an algorithm for specifying, or generating, all and only the
grammatical sentences in a language.

What’s an algorithm? It’s simply any finite, explicit procedure for accomplishing some
task, beginning in some initial state and terminating in a defined end state. Computer programs
Chapter 1: Foundational issues 18

are the algorithms par excellence. More concrete examples of algorithms include recipes, knitting
patterns, the instructions for assembling an IKEA bookcase, or a list of steps for balancing your
checkbook.
An important point to keep in mind is that it is often difficult to construct an algorithm for
even trivial tasks. A quick way to gain an appreciation for this is to describe how to tie a bow.
Like speaking a language, tying a bow is a skill that most of us master around school age and that
we perform more or less unconsciously thereafter. But describing (not demonstrating!) how to
do it is not that easy, especially if we’re not familiar with the technical terminology of knot-tying.
In an analogous way, constructing a generative grammar of English is a completely different task
than speaking the language, and much more difficult (or at least difficult in a different way)!
Just like a cooking recipe, a generative grammar needs to specify the ingredients and pro-
cedures that are necessary for generating grammatical sentences. We won’t introduce all of these
in this first chapter, but in the remainder of the section, we’ll introduce enough ingredients and
procedures to give a flavor of what’s to come.

5.1 Elementary trees and substitution

The raw ingredients that sentences consist of are vocabulary items. These belong to various
syntactic categories, like noun, adjective, transitive verb, preposition, and so forth. Depending
on their syntactic category, vocabulary items combine with one another to form constituents,
which in turn belong to syntactic categories of their own. For instance, determiners (a category
that includes the articles a and the and the demonstratives this, that, these and those) can combine
with nouns to form noun phrases, but they can’t (or at least don’t ordinarily) combine with other
syntactic categories like adverbs, verbs, or prepositions.

(45) a. ✔ a house
b. ✔ the cats
c. ✔ those books

(46) a. * a slowly
b. * the went
c. * those of

It’s possible to represent the information contained in a constituent by using labeled

bracketing. Each vocabulary item is enclosed in brackets that are labeled with the appropri-
ate syntactic category. The constituent that results from combining vocabulary items is in turn
enclosed in brackets that are labeled with the constituent’s syntactic category. The labeled brack-
etings for the noun phrases in (45) are given in (47).

(47) a. [NounPhr [Det a ][Noun house ] ]

b. [NounPhr [Det the ][Noun cats ] ]
c. [NounPhr [Det those ][Noun books ] ]

Noun phrases can combine with other syntactic categories, such as prepositions or transitive
verbs. Prepositions combine with a noun phrase to form prepositional phrases. A transitive verb
Chapter 1: Foundational issues 19

combines with one noun phrase to form a verb phrase, which in turn combines with a second
noun phrase to form a complete sentence.

(48) a. [PrepPhr [Prep on ] [NounPhr [Det the] [Noun table ] ] ]

b. [VerbPhr [TrVerb drafted ] [NounPhr [Det a ] [Noun letter ] ] ]
c. [Sentence [NounPhr [Det the ] [Noun secretary ] ] [VerbPhr [TrVerb drafted ] [NounPhr [Det a ]
[Noun letter ] ] ] ]

Noun phrases don’t, however, combine with any and all syntactic categories. For instance, they
can’t combine with determiners (at least not in English).

(49) * the this letter

As constituent structure grows more complex, labeled bracketings very quickly grow dif-
ficult for humans to visually process, as you may have experienced while looking at (48c). It’s
often more convenient to represent constituent structure with tree diagrams. Tree diagrams, or
trees for short, convey exactly the same information as labeled bracketings, but the information
is presented differently. Instead of enclosing an element in brackets that are labeled with a syn-
tactic category, the category is placed immediately above the element and connected to it with a
line or branch. The labeled bracketings that we have seen so far translate into the trees in (50)
and (51).8

(50) a. NounPhr b. NounPhr c. NounPhr

Det Noun Det Noun Det Noun

a house the cats those books

(51) a. PrepPhr b. VerbPhr

Prep NounPhr TrVerb NounPhr

on drafted
Det Noun Det Noun
the table the letter
c. Sentence

NounPhr VerbPhr

Det Noun TrVerb NounPhr

the secretary drafted
Det Noun
the letter

Trees like those in (50) and (51) resemble dishes that are ready to serve; they don’t provide
a record of how they were brought into being. We can provide such a record by representing
Chapter 1: Foundational issues 20

vocabulary items themselves in the form of trees that include combinatorial information. For
example, prepositions and transitive verbs can be represented as trees with empty slots for noun
phrases to fit into, as shown in (52).

(52) a. PrepPhr b. Sentence

Prep »NounPhr« »NounPhr« VerbPhr

on
TrVerb »NounPhr«
drafted

We’ll refer to trees for vocabulary items like those in (52) as elementary trees. The pur-
pose of elementary trees is to represent a vocabulary item’s combinatorial possibilities, and so
they ordinarily contain unfilled nodes. Such nodes are called substitution nodes, and they are
filled by a substitution operation, as shown in (53). In this text, we will indicate substitution
nodes by enclosing them in inverted guillemets (» «). The guillemets “point” to the spot where a
node needs to be filled.

(53) a. PrepPhr b. NounPhr

Prep »NounPhr« Det Noun

on the table

(53a) has a substitution node of The root (topmost) node in (53b) has
some syntactic category. the same syntactic category as the
substitution node in (53a).

c. PrepPhr

Prep NounPhr
on
Det Noun
the table

Substitution occurs when the root

node of (53b) is identified with the
substitution node in (53a).

Elementary trees don’t necessarily contain substitution nodes, though; ones that invariably
play the role of (53b) in the substitution operation don’t. The elementary tree for the noun in (54b)
is an example.
Chapter 1: Foundational issues 21

Notice, by the way, that there are two conceivable ways to arrive at trees for noun phrases
like those cats, depending on whether it is the noun that is taken as the substitution node, as in
(54), or the determiner, as in (55). At this point, there is no reason to prefer one way over the
other, but in Chapter 5, we will adopt a variant of (54).

(54) a. NounPhr b. Noun

cats
Det »Noun«
those

(55) a. Det b. NounPhr

those
»Det« Noun
cats

In summary, a generative grammar as we’ve constructed it so far consists of a set of ele-

mentary trees, which represent the vocabulary items in a language and the range of their com-
binatorial possibilities, and a substitution operation, by means of which the elementary trees
combine into larger constituents and ultimately into grammatical sentences. In Chapter 4, we
will introduce two further formal operations. The first, adjunction, will enable the grammar to
generate sentences containing modifiers, such as adjectives or relative clauses modifying nouns
(the big dog, the dog that the children like). The second, movement, will enable the grammar to
represent, among other things, the similarities and the differences between declarative sentences
(They will see Bill) and questions corresponding to them (Will they see Bill?, Who(m) will they
see?).

5.2 Grammaticality
As we mentioned earlier, the aim of a generative grammar is to generate all and only the gram-
matical sentences of a language. Since the notion of grammaticality is basic to syntactic theory,
it is important to distinguish it from notions with which it is easily confused.
First and foremost, ‘is grammatical’ is not the same thing as ‘makes sense.’ The sentences
in (56) all ‘make sense’ in the sense that it is easy to interpret them. Nevertheless, as indicated
by the asterisks, they are not grammatical.9

(56) a. * Is our children learning?

b. * Me wants fabric.
c. * To where are we be taking thou, sir?
d. * The introduction explained that “the Genoese people, besides of hard worker, are
good eater too, and even ‘gourmand,’ of that honest gourmandise which will not
drive a man to hell but which is, after all, one of the few pleasures that mankind
can enjoy in this often sorrowful world.”
Chapter 1: Foundational issues 22

Conversely, sentences can be grammatical, but not ‘make sense.’ The ‘fairy tale’ or ‘science
fiction’ sentences in (15) are of this type. Two further examples are given in (57). Since the
sentences are grammatical, they aren’t preceded by an asterisk. Their semantic anomaly can be
indicated, if desired, by a prefixed pound sign (hash mark).

(57) a. # Colorless green ideas sleep — cf. Revolutionary new ideas appear
furiously. (Chomsky 1965, p. 149) infrequently.
b. # I plan to travel there last year. — cf. I plan to travel there next year.

Second, ‘grammatical’ must be distinguished from ‘acceptable’ or ‘easily processable by

human beings’. This is because it turns out that certain well-motivated simple grammatical op-
erations can be applied in ways that result in sentences that are virtually impossible for human
beings to process. For instance, it is possible in English to modify a noun with a relative clause,
and sentences containing nouns that are modified in this way, like those in (58), are ordinarily
perfectly acceptable and easily understood. (Here and in the following examples, the relative
clauses are bracketed and the modified noun is underlined.)

(58) a. The mouse [that the cat chased] escaped.

b. The cat [that the dog scared] jumped out the window.

But now notice what happens when we modify the noun within the relative clause in (58a)
with a relative clause of its own.

(59) The mouse [that the cat [that the dog scared] chased] escaped.

Even though (59) differs from (58a) by only four additional words and a single additional
level of embedding, the result is virtually uninterpretable without pencil and paper. The reason
is not that relative clause modification can’t apply more than once, since the variant of (58a) in
(60), which contains exactly the same words and is exactly as long, is perfectly fine (or at any rate
much more acceptable than (59)).

(60) The mouse escaped [that the cat chased] [that the dog scared].

The reason that (59) is virtually uninterpretable is also not that it contains recursive struc-
ture (the relative clause that modifies mouse contains the relative clause that modifies cat). After
all, the structures in (37) are recursive, yet they don’t throw us for a loop the way that (59)
does.
(59) is unacceptable not because it is ungrammatical, but because of certain limitations on
human short-term memory (Chomsky and Miller 1963, pp. 286, 471). Specifically, notice that in
the (relatively) acceptable (60), the subject of the main clause the mouse doesn’t have to ”wait”
(that is, be kept active in short-term memory) for its verb escaped since the verb is immediately
adjacent to the subject. The same is true for the subjects and verbs of each of the relative clauses
(the cat and chased, and the dog and scared). In (59), on the other hand, the mouse must be kept
active in memory, waiting for its verb escaped, for the length of the entire sentence. What is even
worse, however, is that the period during which the mouse is waiting for its verb escaped overlaps
the period during which the cat must be kept active, waiting for its verb chased. What makes (59)
Chapter 1: Foundational issues 23

so difficult, then, is not the mere fact of recursion, but that two relations of exactly the same sort
(the subject-verb relation) must be kept active in memory at the same time. In none of the other
relative clause sentences is such double activation necessary. For instance, in (58a), the mouse
must be kept active for the length of the relative clause, but the subject of the relative clause (the
cat) needn’t be kept active since it immediately precedes its verb chased.

Extra Info
Sentences like (58) and (59) are often referred to as center-embedding structures, and
the dependencies between the subjects and their verbs are said to be nested.

The mouse that the cat chased escaped.

The mouse that the cat that the dog scared chased escaped.

By contrast, the corresponding dependencies in (60) are not nested.

The mouse escaped that the cat chased.

The mouse escaped that the cat chased that the dog scared.

This remains true even if we focus on the dependencies between the relative clauses and
the nouns that they modify.

The mouse escaped that the cat chased that the dog scared.

A final important point to bear in mind is that any sentence is an expression that is paired
with a particular interpretation. Grammaticality is always determined with respect to a pairing
of form and meaning. This means that a particular string can be grammatical under one interpre-
tation, but not under another. For instance, (61) is ungrammatical under an subject-object-verb
(SOV) interpretation (that is, when the sentence is interpreted as Sue hired Tom).

(61) Sue Tom hired.

(61) is grammatical, however, under an object-subject-verb (OSV) interpretation (that is,

when it is interpreted as Tom hired Sue). On this interpretation, Sue receives a special intonation
marking contrast, which would ordinarily be indicated in writing by setting off Sue from the rest
of the sentence by a comma. In other words, the grammaticality of (61) depends on whether its
interpretation is analogous to (62a) or (62b).

(62) a. * She him hired.

b. ✔ Her, he hired. (The other job candidates, he didn’t even call back.)
Chapter 1: Foundational issues 24

5.3 Grammar versus language

We conclude this chapter by considering the relationship between the concepts of grammar and
language. The notion of language seems straightforward because we are used to thinking and
speaking of “the English language,” “the French language,” “the Swahili language,” and so forth.
But these terms are actually much vaguer than they seem at first glance because they cover a
plethora of varieties, including ones that differ enough to be mutually unintelligible. For instance,
Ethnologue distinguishes 32 dialects of English in the United Kingdom alone. In addition, distinct
dialects of English are spoken in former British colonies, including Australia, Canada, India, New
Zealand, South Africa, the United States, and many other African, Asian, and Caribbean nations,
and many of these dialects have subdialects of their own. Similarly, Ethnologue distinguishes 11
dialects of French in France and 10 dialects of Swahili in Kenya, and there are further dialects in
other countries in which these languages are spoken. Moreover, we use terms like “the English
language” to refer to historical varieties that differ as profoundly as present-day English does
from Old English, which is about as intelligible to a speaker of modern English as German—that
is to say, not.
Although the most salient differences between dialects are often phonological (that is,
speakers of different dialects often have different accents), dialects of a so-called single language
can differ syntactically as well. For instance, in standard French, as in the Romance languages
more generally, adjectives ordinarily follow the noun that they modify. But that order is reversed
in Walloon, a variety of French spoken in Belgium. The two parametric options are illustrated in
(63) (Bernstein 1993, pp. 25–26).
(63) a. un chapeau noir [Standard French]
a hat black
b. on neûr tchapê [Walloon]
a black hat
‘a black hat’
Another example of the same sort concerns multiple negation in sentences like (64a) in
varieties of English.
(64) a. The kids didn’t eat nothing.
b. The kids didn’t eat anything.
In present-day Mainstream U.S. English, didn’t and nothing each contribute their negative
force to the sentence, and the overall force of (64a) isn’t negative; rather, the sentence means
that the kids ate something. In many varieties of English, however, (64a) conveys exactly the
same meaning as in Mainstream U.S. English(64b); that is, the sentence as a whole has negative
force. In these varieties, the negation in nothing can be thought of as agreeing with (and rein-
forcing) the negation in didn’t rather than cancelling it; hence the term negative concord for
this phenomenon (‘concord’ is a variant term for ‘agreement’). Negative concord is routinely
characterized as “illogical” by prescriptivists,10 and it is one of the most heavily stigmatized fea-
tures in present-day English.11 However, it was productive in earlier forms of English, and it is
attested in renowned masters of the language such as Chaucer and Shakespeare. Moreover, neg-
ative concord is part of the mainstream forms of languages like Afrikaans, French, modern Greek,
Chapter 1: Foundational issues 25

Hausa, Italian, Lubukusu, Moroccan Arabic, Nweh, Spanish, and Tarifit Berber (just to name a
few). From a descriptive and generative point of view, negative concord is simply a parametric
option of Universal Grammar just like any other, and negative concord is no more illogical than
the noun-adjective order in (63a) or preposition stranding.
In both of the examples just discussed, we have dialects of “the same language” (English
and French, respectively) differing with respect to a parameter. The converse is also possible:
two “different languages” that are parametrically (all but) indistinguishable. For example, the
same linguistic variety spoken on the Dutch-German border may count as a dialect of Dutch or
German depending on which side of the political border it is spoken, and the same is true of many
other border dialects as well. According to Max Weinreich, “a language is a dialect with an army
and a navy.” A striking confirmation of this aphorism concerns the recent terminological history
of Serbo-Croatian. As long as Yugoslavia was a federal state, Serbo-Croatian was considered a
single language with a number of regional dialects. The 14th edition of Ethnologue, published in
2000, still has a single entry for Serbo-Croatian. In the 15th edition, published in 2005, the single
entry is replaced by three new entries for Bosnian, Croatian, and Serbian. What changed in the
meantime wasn’t the languages, but the political situation.
As the previous discussion has shown, the notion of language is based more on sociopo-
litical considerations than on strictly linguistic ones. By contrast, the term ‘grammar’ refers to
a particular set of parametric options that a speaker acquires. For this reason, the distinction
between language and grammar that we have been drawing is also referred to as the distinction
between E-language and I-language (mnemonic for ‘external’ and ‘internal’ language) (Chom-
sky 1986).
As we have seen, the same language label can be associated with more than one grammar
(the label “English” is associated with grammars both with and without negative concord), and
a single grammar can be associated with more than one language label (as in the case of border
dialects). It is important to distinguish the concept of shared grammar from mutual intelligibility.
To a large extent, Mainstream U.S. Englishand many other varieties of American English are mu-
tually intelligible even where their grammars differ with respect to one parameter or another. On
the other hand, it is perfectly possible for two or more varieties that are mutually unintelligible
to share a single grammar. For instance, in the Indian village of Kupwar (Gumperz and Wilson
1971), the three languages Marathi, Urdu, and Kannada, each spoken by a different ethnic group,
have been in contact for about 400 years, and most of the men in the village are bi- or trilingual.
Like the standard varieties of these languages, their Kupwar varieties have distinct vocabularies,
thus rendering them mutually unintelligible to monolingual speakers, but in Kupwar, the consid-
erable grammatical differences that exist among the languages as spoken in other parts of India
have been virtually eliminated. The difference between standard French and Walloon with re-
spect to prenominal adjectives is another instance of this same convergence phenomenon. Here,
too, the adjective-noun order in Walloon is due to language contact and bilingualism, in this case
between French and Flemish, the other language spoken in Belgium; in Flemish, as in the Ger-
manic languages more generally, adjectives ordinarily precede the nouns that they modify.
Finally, we should point out that it is perfectly possible for a single speaker to acquire more
than one grammar. This is most strikingly evident in balanced bilinguals. Speakers can also ac-
quire more than one grammar in situations of syntactic change. For instance, in the course of
 Chapter 1: Foundational issues 26

its history, English changed from an OV to a VO language, and individual speakers during the
transition period (which began in late Old English and continued into Middle English) acquired
and used both parametric options. Speakers can also acquire more than one grammar in situa-
tions of diglossia or stable syntactic variation. For instance, English speakers whose vernacular
grammar has negative concord or preposition stranding might acquire the parametric variants
without negative concord or with pied piping in the course of formal education.

Notes 6
1. It is also possible to overzealously apply rules like those in (2), even in cases where they shouldn’t be applied. This
phenomenon is known as hypercorrection. Two common instances are illustrated in (i).

(i) Hypercorrect example Mainstream target Explanation

Over there is the guy whom I the guy who I think the relative pronoun who is the subject of the
think took her to the party. took her to the party relative clause, not the object (cf. the guy
who, *whom took her to the party)

This is strictly between you and I. between you and me the second pronoun is part of the conjoined
object of the preposition between, not part of a
subject

2. The prescriptive rule is actually better stated as “Don’t separate a preposition from its object,” since the traditional
formulation invites exchanges like (i).

(i) A: Who are you going to the party with?

B: Didn’t they teach you never to end a sentence with a preposition?
A: Sorry, let me rephrase that. Who are you going to the party with, Mr. Know-it-all?

3. As William Labov has often pointed out, everyday speech (apart from false starts and other self-editing phenomena)
hardly ever violates the rules of descriptive grammar.

4. Actually, that’s an oversimplification. Not all the articles and nouns an English-speaking child hears appear in the
article-noun order. To see why, carefully consider the underlined sentence in this note.

5. Since then, wugs have become something of an unofficial mascot for linguistics; it is wugs that adorn the cover of
this book.

6. When children didn’t respond this way, they either repeated the original invented word, or they didn’t respond at
all. It’s not clear what to make of these responses. Either response might indicate that the children were stumped by
the experimental task. Alternatively, repetition might have been intended as an irregular plural (cf. deer and sheep),
and silence might indicate that some of the invented words (for instance, cra) struck the children as phonologically
strange.

7. The term ‘pied piping’ was invented in the 1960s by John Robert Ross (1967), a syntactician with a penchant for
 Chapter 1: Foundational issues 27

metaphorical terminology.

8. Online corpora that are annotated with syntactic structure, such as the Penn Treebank, the Penn Parsed Corpora of
Historical English, and others like them, tend to use labeled bracketing because the resulting files are computationally
extremely tractable. The readability of such corpora for humans can be improved by suitable formatting of the labeled
bracketing or by providing an interface that translates the bracketed structures into tree diagrams.

9. (56a) is from a speech by George W. Bush (linked here). (56b) was the subject line of an email message in response
to an offer of free fabric; the author is humorously attempting to imitate the language of a child greedy for goodies.
(56c) is from “Pardon my French” (Calvin Trillin. 1990. Enough’s enough (and other rules of life). p. 169). (56d) is
from “Connoisseurs and patriots” (Joseph Wechsberg. 1948. Blue trout and black truffles: The peregrinations of an
epicure. p. 127).

10. Two important references concerning the supposed illogicality of negative concord (and of marginalized varieties of
English more generally) are Labov (1972a) and Labov (1972b).
Those who argue that negative concord is illogical often liken the rules of grammar to those of formal logic
or arithmetic, where one negation operator or subtraction operation cancels out another; that is, (NOT (NOT A))
is identical to A, and (−(−5)) = +5. Such prescriptivists never distinguish between sentences containing even and
odd numbers of negative expressions. By their own reasoning, (i.a) should have a completely different status than
(i.b)—not illogical, but at worst redundant.

(i) a. They never told nobody nothing.

b. They never told nobody.

11. Because of the social stigma associated with it, it is essentially impossible to study negative concord in present-
day English. This is because even for those speakers of negative concord varieties who don’t productively control
Mainstream U.S. Englishas a second dialect, the influence of prescriptive grammar is so pervasive that if such speakers
reject negative concord sentences as unacceptable, we don’t know whether they are rejecting them for grammatical
or for social reasons.

Exercises and problems 7

Exercise 1.1
The sentences in (4) violate several descriptive rules of English, three of which were given in (5).
As mentioned in the text, there is a fourth descriptive rule that is violated in (4). Formulate the
rule (you shouldn’t need more than a sentence).
Chapter 1: Foundational issues 28

Exercise 1.2
(1)–(4) illustrate the facts of subject-verb agreement in the variety of English spoken in Belfast,
Ireland (data from Henry 1995, chapter 2). Describe the data as clearly and briefly as you
can.

Take Note

In order to avoid conflating morphological form with semantic function, you can refer
to “is” and “are” as “the i- form” and “the a- form”, rather than as “singular” and “plu-
ral”.

(1) a. ✔ The girl is late. (2) a. * The girl are late.

b. ✔ She is late. b. * She are late.
c. ✔ Is {the girl, she} late? c. * Are {the girl, she} late?

(3) a. ✔ The girls are late. (4) a. ✔ The girls is late.

b. ✔ They are late. b. * They is late.
c. ✔ Are {the girls, they} late? c. * Is {the girls, they} late?

Exercise 1.3
A. Which of the newspaper headlines in (1) are lexically ambiguous, which are structurally
ambiguous, and which are a mixture of both types of ambiguity? Explain.

(1) a. Beating witness provides names

b. Child teaching expert to speak
c. Drunk gets nine months in violin case
d. Enraged cow injures farmer with ax
e. Prostitutes appeal to pope
f. Teacher strikes idle kids
g. Teller stuns man with stolen check

B. At least two of the examples form a subgroup even within their type (lexical, structural,
mixed). Can you find them and explain their similarity?
Chapter 1: Foundational issues 29

Exercise 1.4
In the chapter, we showed that sentences are recursive categories. In other words, one instance
of the syntactic category ‘Sentence’ can contain another instance of the same category. Provide
evidence that noun phrases and prepositional phrases are recursive categories, too.

Word of Caution
Be careful to give examples that are recursive, and not just ones in which the syntactic
category in question occurs more than once. For instance, (1) does not provide the evidence
required in this exercise, because the second prepositional phrase is not contained in the
first. This is clearly shown by the fact that the order of the prepositional phrases can be
switched.

(1) a. The cat jumped [PP onto the table ] [PP without the slightest hesitation ].
b. The cat jumped [PP without the slightest hesitation ] [PP onto the table ].

Exercise 1.5
Which, if any, of the sentences in (1)–(6) are ungrammatical? Which, if any, are semantically or
otherwise anomalous? Briefly explain.

(1) a. They decided to go tomorrow yesterday.

b. They decided to go yesterday tomorrow.

(2) a. They decided yesterday to go tomorrow.

b. They decided tomorrow to go yesterday.

(3) a. Yesterday, they decided to go tomorrow.

b. Tomorrow, they decided to go yesterday.

(4) They decided to go yesterday yesterday.

(5) How long didn’t Rebecca wait?

(6) Pelé came from a much poor family.

Chapter 1: Foundational issues 30

Exercise 1.6
A. The following expressions are structurally ambiguous. For each reading (= interpretation),
provide a paraphrase that is itself unambiguous or a diagnostic scenario (that is, a scenario
that is compatible only with the reading in question).

(1) a. chocolate cake icing

b. clever boys and girls
c. Charis will answer the question precisely at noon.
d. Laila will watch the man from across the street.
e. They should decide if they will come tomorrow.

B. For each reading, provide as much of a labeled bracketing as you can, focusing on distin-
guishing between the readings. In other words, you may not be able to give a full labeled
bracketing, but for each reading, determine which words go together more or less closely.

Exercise 1.7
Can you solve the riddle posed in endnote 4 of this chapter?

Problem 1.1
Syntactic structure and recursion are both formal universals of human language. Are they both
equally basic? Explain in a (possibly very) brief paragraph.

Problem 1.2
A. Can you come up with a sentence (or other expression) that is structurally ambiguous more
than two ways? For each reading, provide an unambiguous paraphrase or a diagnostic
scenario.
B. For each reading, which words go together? Provide a labeled bracketing, with a focus on
distinguishing the interpretations (that is, you don’t necessarily need to label each word).

Problem 1.3
Assume a toy grammar G1 with a vocabulary of lowercase letters and a single syntactic category
X. G1 generates structures like the following:
Chapter 1: Foundational issues 31

X X X X X
y
X X X X X X X X
y g a m p
X X X X X X
t y e p
X X X X
a m m
X X
p h
Another toy grammar G2 generates structures like this:
X X X X X
a
X X X X X X X X
m e m m t
X X X X X X
p h p h
X X X X
p e p
X X
e z
Briefly answer the following questions:
A. What is the difference between G1 and G2?
B. If presented with a string of lowercase letters (for example, “abc”), is it possible to tell
which grammar has generated the string?

Problem 1.4
The grammars of Early Modern English (1500–1710) and present-day English differ enough for
certain Early Modern English sentences to be ungrammatical today. Find several such sentences,
and briefly describe the source of the ungrammaticality as best as you can. Early Modern English
texts that are easily accessible on the Web include Shakespeare’s plays and the Authorized Version
of the Bible (also known as the King James Bible).

Problem 1.5
Can you think of other cognitive domains that exhibit structure-dependence, recursion, and/or
structural ambiguity?
 2
Syntactic categories
and constituenthood

Chapter outline
1 Parts of speech . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
1.1 The trouble with “words” and “parts of speech” . . . . . . . . . . . . . . . 34
1.2 Nouns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
1.3 Adjectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
1.4 Modals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
1.5 Auxiliary verbs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
2 Tests for determining syntactic constituenthood . . . . . . . . . . . . . . . . . . . 54
2.1 Substitution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
2.2 Movement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
2.3 Questions and short answers . . . . . . . . . . . . . . . . . . . . . . . . . 57
2.4 It clefts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
3 Some complications for constituency diagnostics . . . . . . . . . . . . . . . . . . 59
3.1 Mismatches between syntactic structure and other structure . . . . . . . 59
3.2 False negative results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
4 Representing syntactic constituenthood . . . . . . . . . . . . . . . . . . . . . . . 67
4.1 Basic terms and relations . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
4.2 Derived terms and relations . . . . . . . . . . . . . . . . . . . . . . . . . . 70
5 Regarding scientific models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
6 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
7 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
8 Appendix: Finiteness in English . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
8.1 Verb forms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
8.2 Finiteness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

In Chapter 1, we provided evidence for the idea that sentences aren’t just strings of words,
but that they consist of syntactic constituents—groups of words that belong together. In many
cases, native speakers have clear intuitions about whether a particular string is a constituent.
But not all cases are clear, and it is useful to have diagnostic tests at our disposal for determining
constituenthood in such cases. It also sometimes happens that native speakers disagree over
Chapter 2: Syntactic categories and constituenthood 33

whether a string is a constituent. In such cases, the tests can help us to clearly establish that fact
so that we have a firm basis for further exploring the differences between the alternative mental
grammars.
As we will discuss in more detail in Chapters 4 and 5, syntactic constituents come in various
‘flavors’ like noun phrases, verb phrases, adjective phrases, adverb phrases, and several others.
The reason is that the words that make up the constituents themselves belong to different syntac-
tic categories. Traditionally, these are known as parts of speech. However, the traditional notion
of part of speech is flawed in two ways. First, the concept of ‘word’ seems simple and straight-
forward, but isn’t. Second, parts of speech are traditionally defined in terms of meaning, even
though these definitions quickly break down. Our solution to these problems will be to allow
syntactic categories to be associated with morphemes (rather than words) and to define them
in terms of morphological form and morphosyntactic distribution (rather than semantic
function).
The chapter is organized as follows. We begin by clarifying the notion of syntactic cat-
egory itself, as just discussed, and presenting four syntactic categories. We then present the
constituenthood tests and some complications in applying them in connection with certain syn-
tactic categories. Finally, we introduce the graphic representation of constituenthood by means
of syntactic trees.

Parts of speech 1
In the U.S. education system, many people’s first experiences with so-called grammar involves
identifying parts of speech as part of the task of “diagramming sentences.” So in an example like
(1), we can label the words in the sentence based on the category they fall into.

(1) Dolphins play endlessly.

noun verb adverb

Linguists tend to divide lexical items (memorized elements in a language) into two broad
sets of categories: lexical categories, which tend to be open-class elements with real-world se-
mantic content, and functional categories, which tend to mainly serve functions within grammar,
to make sentences fit together, as it were. Rather than using the term parts of speech, we will
use the term syntactic category to capture the same notion.
Chapter 2: Syntactic categories and constituenthood 34

Lexical:
syntactic category Examples
noun kumquat, Darius, thusness
verb synthesize, endure, appear
adjective glad, comely, unequivocal
adverb rather, simply, luckily

Functional:
syntactic category Examples
preposition with, under, beside
article the, a, an
demonstrative this, that, yonder
quantifier every, any, no, few
pronoun me, us, I, who
possessive pronoun hers, mine, theirs
complementizer/subordinator that, whether, if
auxiliary verb be, have, do

Table 2.1: Selected English grammatical categories, both lexical and functional

In some sense, grammatical categories might seem simple or fundamental, and they are
very useful in understanding grammatical patterns. But they can get somewhat complicated to
define, especially given how we’re accustomed to thinking about them.

1.1 The trouble with “words” and “parts of speech”

What kinds of elements are able to be assigned a part of speech? Colloquially we think of ‘words’
having parts of speech, but the concept of a ‘word’ falls apart surprisingly quickly when you
investigate it. The elements we call ‘words’ generally have some degree of phonological and
grammatical independence. So, phonologically, we can see a word standing alone, not requiring
other content to be said.
(2) Q: Which do you like better, coffee or tea?
A: Tea.
Likewise, many ‘words’ tend to be able to combine with other ‘words’ in flexible ways. So in (3) tea
can be used as the subject or the object of the sentence, or as part of a prepositional phrase.
(3) a. Tea is good for you.
b. She doesn’t drink tea.
c. There are beneficial antioxidants in tea.
But these notions start to fall apart fairly quickly once you poke at them a little. The sen-
tences in (4) include the phonetically transcribed [tʰiz].
Chapter 2: Syntactic categories and constituenthood 35

(4) a. Please do not [tʰiz] your younger sibling. (tease)

b. That shop sells [tʰiz] from around the world. (teas)
c. I think the antioxidants are [tʰiz] best property. (tea’s = tea possessive)
d. [tʰiz] good for you. (tea’s = tea is)

Most of us would say without hesitation that that tease in (4a) is a word. And even though
teas in (4b) is made up of two morphemes (tea + pluRal), we’d guess that most of you would still
be inclined to call that a word. But are the instances of [tʰiz] tea’s in (4c) and (4d) words? For
many native English speakers, intuitions about what a ‘word’ is start to fall apart here. Despite
being phonetically identical to the other elements we readily call words, in (4d) we have a sense
that two different grammatical elements are being combined, a noun tea and a verbal element is
(a similar argument can be applied to (4c)). So though we are taught in primary school (we can
call it “4th grade grammar” for simplicity) that ‘words’ have parts of speech, it’s not clear what
a word even is. (The issue becomes even more salient when looking at agglutinative languages,
where entire sentences can be expressed with a single ‘word.’)
So perhaps at bare minimum we need to say that morphemes are assigned parts of speech,
and not ‘words.’ But even a monomorphemic word can be trouble. Consider our 4th grade gram-
mar definitions of major parts of speech, which tend to rely on semantic classifications.

(5) a. Noun
person place thing (or idea!)
b. Verb
an action (or state!)
c. Adjective
modifies a noun
d. Adverb
modifies a verb or adjective

These classifications fall apart fairly quickly on closer inspection. A word like yesterday
in (6a) we might want to say is an adverb, since it modifies a verb (yesterday is when the going
happened). But yesterday in (6b) is clearly not an adverb, being used as a possessor. But yesterday
means the same thing in both instances.

(6) a. We went to the beach yesterday.

b. Yesterday’s beach trip was fun.

Likewise, beer is clearly a noun in (7a) (and typically would be considered a noun), but is used as
a verb in (7b) to mean ‘give (someone) a beer.’

(7) a. Give me a beer!

b. Beer me! (Colloquial Mainstream English, = “give me a beer”)

The morphological properties of ‘words’ further complicates things. Reading in (8a) be-
haves like a verb, yet in (8b) it behaves like a noun. But in each case it’s referring to something
we’d reasonably call the action of reading.
Chapter 2: Syntactic categories and constituenthood 36

(8) a. Maisha is reading an Elephant and Piggie book.

b. Daily reading is recommended for kindergarteners.

The canonical example of this from the literature is destroy, as in (9). A verbal use of destroy is
unambiguous in (9a), but even though in (9b) destruction is a noun, it transparently describes an
action.

(9) a. The cat destroyed the couch.

b. the cat’s destruction of the couch

So what do we do about about our definitions from (5) above? Clearly they’re not sufficient.
In fact, a linguist is much more inclined to write definitions like those in (10).

(10) a. Noun
‘Words’ that do nouny things
b. Verb
‘Words’ that do verby things
c. Adjective
‘Words’ that do adjective-y things
d. Adverb
‘Words’ that do adverby things

The definitions in (10) are of course silly, but they’re actually not far off. What (10a), for
instance, presupposes is that there are such things as ‘nouny things.’ What this is saying is that
syntactic categories are defined by their grammatical distribution, not by semantic properties. So
what are ‘nouny’ sorts of things to do? We can consider both morphological distribution and
syntactic distribution. Syntactically, for example, nouns often serve as the subject or object of a
sentence and they commonly follow determiners and adjectives. Morphologically, they can be
marked as plural and take various derivational affixes that only attach to nouns (like -tion and
-er, for example). In English, elements that behave in this way are consistently nouns.1 The tables
below illustrate (non-exhaustively) some of the distributional properties of major grammatical
categories in English.
Chapter 2: Syntactic categories and constituenthood 37

Category Syntactic Distribution Examples

Noun ▶ subject/object of sentence The small dog ran away.
▶ object of preposition The chew toy is with Lilly.
▶ often follow determiners/adjectives

Verb ▶ takes a subject/object Lilly might run away.

▶ can follow auxiliary verbs like be and have Cooper wants Lilly to run away.
▶ can follow modals like might and should
▶ can follow the infinitive marker to
Adjective ▶ often follow a determiner and precede a The very small dog ran away.
noun Lilly looks so small!
▶ often follow linking verbs like be and seem
▶ can follow intensifiers like very and so

Adverb ▶ often precede adjectives The small dog ran away briefly.
▶ often precede or follow verb phrases Luckily it won’t take a week.
▶ some appear at the beginning of a sentence

Table 2.2: Syntactic distribution of selected English grammatical categories

Category Morphological Distribution Examples

Noun ▶ bears plural morphology cats, wishes, children
▶ some derivational affixes are only on nouns prediction, driver
(e.g. -tion, -er)

Verb ▶ take the third-person singular -s suffix loathes, says, tries

▶ bear past tense morphology jumped, dreamt, froze
▶ can bear the affixes -ize, -ate, -ify pixelize, formulate, beautify

Adjective ▶ can bear the affixes -ish, -like childish, childlike

▶ can occur in comparative (-er) and superlative bigger, biggest
(-est) forms (see § 3 for details)

Adverb ▶ often bear the affix -ly briefly, coolly, equally

▶ can occur in comparative (-er/more) and more briefly, cooler, coolest
superlative (-est/most) forms

Table 2.3: Morphological properties of selected English grammatical categories

The fact that distributional properties are primary in our knowledge of grammatical cat-
egories is evident in literature, especially children’s literature, where invented nonsense words
sometimes play a major role. The following is a brief selection from Dr. Seuss’s The Lorax:
Chapter 2: Syntactic categories and constituenthood 38

You won’t see the Once-ler. Don’t knock at his door. He stays in his Lerkim on top of
his store. He stays in his Lerkim, cold under the floor, where he makes his own clothes out
of miff-muffered moof. And on special dank midnights in August, he peeks out of the
shutters and sometimes he speaks and tells how the Lorax was lifted away. He’ll tell you,
perhaps… if you’re willing to pay.

Then he pulls up the pail, makes a most careful count to see if you’ve paid him the proper
amount. Then he hides what you paid him away in his Snuvv, his secret strange hole in
his gruvvulous glove. Then he grunts. “I will call you by Whisper-ma-Phone, for the
secrets I tell you are for your ears alone.”

SLUPP!

Down slupps the Whisper-ma-Phone to your ear and the old Once-ler’s whispers are not
very clear, since they have to come down through a snergely hose, and he sounds as if he
had smallish bees up his nose. “Now I’ll tell you”, he says, with his teeth sounding gray,
“how the Lorax got lifted and taken away… It all started way back… such a long, long time
back…
Each of the bolded words in the selection above were invented by Theodor Geisel (a.k.a.
Dr. Seuss), and if you’ve never read The Lorax, this may be the first time you’ve encountered
them. Sometimes we can get an interpretation: “Down slupps the Whisper-ma-Phone” invokes
an onomatopoeic interpretation of slupp as the sound the phone is making while being lowered.
Others, like gruvvulous, don’t suggest much of an interpretation at all, instead achieving a stylistic
and rhythmic effect. Yet in all cases we have a very clear sense of their syntactic category: slupps
is a verb describing the motion of the Whisper-ma-phone, gruvvulous and snergelly are adjectives
describing glove and hose, respectively, and Lerkim, Snuvv, and Lorax are all nouns. For each, our
ability to readily identify their syntactic category is due to the fact that their use falls within the
distributional criteria for each category, like those illustrated in Tables 2.2 and 2.3.
Again we are left to conclude that the proper defining characteristics of grammatical cat-
egories are not based in semantics: even nonsense words have grammatical categories, despite
not meaning anything. Instead, grammatical categories are defined by grammatical criteria: their
morphological and syntactic distributions. In the sections that follow, we provide additional de-
scription of the nuances around some prominent grammatical categories.
Chapter 2: Syntactic categories and constituenthood 39

1.2 Nouns
In the traditional school definition, a noun “refers to a person, place, or thing”. But as has often
been pointed out, this definition incorrectly excludes nouns like those in (11) (unless the concept
of thing is reduced to near-vacuity).

(11) explosion, glint, mind, moment, thunder

As a result, modern linguists (not just generative ones) define nouns not in semantic
(meaning-based) terms, but in distributional terms—with reference to their occurrence relative to
other syntactic categories in the language. In English, for instance, a useful criterion for whether
a word is a noun is whether it can form a phrase with the determiner the. According to this
criterion, the words in curly brackets in (12a) are nouns, but those in (12b) are not.

(12) a. the { explosion, glint, mind, moment, thunder }

b. * the { explodes, glinted, minded, momentous, thundered }

Count nouns and mass nouns

Nouns are traditionally divided into two major subclasses: count nouns and mass nouns.
Count nouns (in Mainstream U.S. English) can be accompanied by the indefinite article, and they
have a plural form, whereas mass nouns have neither of these properties.

(13) Count noun:

a. a(n) { book, child, couch, explosion, invitation, job, problem, vegetable }
b. books, children, couches, explosions, invitations, jobs, problems, vegetables
(14) Mass noun:
a. * a(n) { advice, furniture, rice }
b. * advices, furnitures, rices

Conversely, singular count nouns cannot follow expressions of quantity like a lot of, much,
and so on, whereas mass nouns can.

(15) Count noun:

a. * a lot of { book, child, couch, explosion, invitation, job, problem, vegetable }
b. * much { book, child, couch, explosion, invitation, job, problem, vegetable }
(16) Mass noun:
a. a lot of { advice, furniture, rice }
b. much { advice, furniture, rice }

Although the distinction between count and mass nouns is generally clear-cut, under spe-
cial circumstances, what are ordinarily mass nouns in English can be used as count nouns—for
instance, when it is possible to impose an interpretation of a kind of X or a salient quantity of X
on the mass noun. Notice, incidentally, that the interpretation of the plural mass nouns in (17b)
depends on whether we refers to a group of coffee shop customers (salient quantity reading) or
to a business owner (salient kind reading).
Chapter 2: Syntactic categories and constituenthood 40

(17) a. Different rices have different cooking times.

b. We ordered four coffees and two teas.

For some mass nouns, the conversion to count noun is very natural.

(18) a. Mass noun use (basic):

a lot of wine, wine from Burgundy, more wine than we can drink tonight
b. Count noun use:
a wine from Alsace, wines from Burgundy, more wines than I had ever heard of

Conversely, it is possible for what are ordinarily count nouns to be pressed into service as mass
nouns.

(19) a. ? This recipe for carrot cake calls for { a lot of carrot, more carrot than I have on hand }
b. ? There’s just not enough couch for all ten of you.
c. 2nd Servant: Pray heauen it be not full of Knight againe.
1st Servant: I hope not, I had liefe as beare so much lead.
(William Shakespeare. Merry Wives of Windsor, Act 4, Scene 2.)
d. The article says that [Mr. Mangetout] once consumed fifteen pounds of bicycle in
twelve days … Just as an example, I have never eaten so much as a pound of bicycle.
… I can see myself acting with considerable restraint at a dinner party at which the
main course, is, say, queen-size waterbed.
(Calvin Trillin. 1983. Third helpings. 3.)

Although the match between count and mass nouns across languages is reasonably good,
mismatches occur. Some examples of nouns that are mass nouns in English, but count nouns in
other languages are given in (20)–(22).

(20) French:
a. un meuble, des meuble-s
a furniture.sg, of.the furniture-pl
‘a piece of furniture, furniture’
b. un renseignement, des renseignement-s
an information.sg, of.the information-pl
‘a piece of information, information’
(21) German:
a. ein Möbel, Möbel
a furniture.sg, furniture.pl
‘a piece of furniture, furniture’
b. ein Ratschlag, Ratschläg-e
an advice.sg, advice-pl
‘a piece of advice, advice’
Chapter 2: Syntactic categories and constituenthood 41

c. eine Nachricht, Nachricht-en

a news.sg, news-pl
‘a piece of news, news’
(22) Italian:
a. un consigli-o, consigl-i
an advice.sg, advice-pl
‘a piece of advice, advice’
b. una notizia, notizi-e
a news.sg, news-pl
‘a piece of news, news’

Not surprisingly, when learning English as a foreign language, speakers from these lan-
guages often produce ungrammatical examples like (23).

(23) a. * They gave me some useful advices.

b. * The room was crowded with too many furnitures.
c. * Before I decide, I need some further informations.

In conclusion, here’s a question for you. Is mail a count noun or a mass noun? And how
about email? Does everyone agree with you, especially about email?

Proper nouns and common nouns

A further distinction is traditionally drawn between proper nouns (also known as proper
names) and common nouns. In contrast to common nouns, proper nouns are conventionally
capitalized in English.

(24) a. Common nouns:

city, collie, country, planet, state
b. Proper nouns:
Philadelphia, Lassie, France, Jupiter, Maryland

Proper nouns refer to particular individual entities. The relation between proper nouns and
individuals is not necessarily one-to-one, however. For instance, the proper noun Sandra may
refer to many different individuals of that name. Individuals with the same proper name needn’t
share any distinguishing properties (other than having the same name). So Athens, Greece and
Athens, Georgia needn’t have any substantive property in common that sets them apart from
other cities, and there is nothing to stop us from giving the name Lassie to a pet without any of
the prototypical qualities of Lassie from the television show (intelligence, loyalty, and so on). Our
Lassie, in fact, needn’t even be a dog. In short, proper nouns function like pure pointers. In the
same way that one and the same pointer can be used to point to different and unrelated items in
a presentation, the same proper noun can be used to refer to different and unrelated individuals.
In contrast, common nouns refer to sets of entities that are related by sharing certain properties.
That is, common nouns have intrinsic semantic content and cannot be used in the relatively
arbitrary way that proper nouns can be.
Chapter 2: Syntactic categories and constituenthood 42

From the point of view of reference, then, proper nouns behave like pronouns, which also
function like pointers to individuals without themselves containing much in the way of semantic
content. At first glance, it might therefore seem that the term ‘proper noun’ is a misnomer, and
that proper nouns could essentially be classified as determiners rather than as nouns, similar to
how pronouns can be aligned with determiners. However, there are two arguments against this
misnomer view of proper nouns.

Proper nouns with and without determiners

First, although proper nouns in English typically appear without a determiner, certain proper
nouns must be accompanied by the definite article. Some examples are given in (25).

(25) the Bronx, the Hague, the Thames, the Titanic, the Soviet Union, the United States, the
White House

Pronouns, on the other hand, can’t be accompanied by an article, as shown in (26).

(26) * the { you, he, she, it, we, they }

Given the contrast between (25) and (26), the proper nouns in (25) can’t themselves be
determiners, but must be nouns, as English does not tolerate sequences of determiners. Even the
misnomer view must therefore admit the existence of nouns that refer to particular individual
entities. But if the proper nouns in (25) are in fact nouns, then there is no conceptual advantage
in assuming that those in (24b) are determiners. Instead, we could simply take the view that
proper nouns are invariably nouns, but that some are complements of the ordinary overt definite
article the, whereas others are complements of a silent counterpart of it. In other words, the
proper nouns in (24b) would be structurally parallel to those in (25), as indicated in (27).

(27) a. Overt determiner:

the Bronx, the Hague, the Thames, the Titanic, …
b. Silent determiner:
[def] Philadelphia, [def] Lassie, [def] France, …

Several additional pieces of evidence, all of the same type, point in the same direction.
Before the breakup of the Soviet Union, the Soviet republic whose capital is Kiev was called
the Ukraine, but after the breakup, the newly independent country began to be called Ukraine,
without the article. Similarly, the United States is generally referred to in Spanish as los Estados
Unidos ‘the United States’, but in recent years, Latin American newspapers have begun to omit
the article los. According to the misnomer view, Ukraine and Estados Unidos would be nouns in
the old usage and turn into determiners in the new usage. Treating the change as affecting the
pronunciation of the determiner is clearly more straightforward.
In French, names of countries and regions must be accompanied by the definite article,
whereas in English, they generally aren’t (the change from the Ukraine to Ukraine thus eliminated
its exceptional status in English). Some examples are shown in (28); la and le are the feminine
and masculine forms of the French definite article, respectively.
Chapter 2: Syntactic categories and constituenthood 43

(28) a. French:
la Bolivie, la Bourgogne, le Brésil, le Canada, le Danemark, la France, le Pérou
b. English:
Bolivia, Burgundy, Brazil, Canada, Denmark, France, Peru

According to the misnomer view, the names of countries and regions would be nouns in
French but determiners in English. Again, it seems more reasonable to pin the difference between
the two languages not on the nouns themselves, but on the status of the determiner.
In standard German, personal names stand alone in the standard language. But in the
vernacular, they can be accompanied by the definite article, as shown in (29); der and die are
masculine and feminine forms of the German definite article, respectively.

(29) a. Standard German:

Annemarie, Eva, Lukas, Peter
b. Vernacular German:
die Annemarie, die Eva, der Lukas, der Peter

According to the misnomer view, proper names would change syntactic category depend-
ing on the speaker’s register, whereas the alternative view again more straightforwardly pins the
variation on the determiner.
Finally, in modern Greek, personal names are normally accompanied by the definite article.
This is shown for the nominative in (30); ο/o and η/i are masculine and feminine forms of the
Greek definite article, respectively. (“Nominative” is the form in which subjects of sentences
appear: see Chapter 7.)

(30) Greek, nominative:

η Ελένη, η Βάλια, ο Παύλος, ο Γιάννης
i Eleni, i Valia, o Pavlos, o Yannis

Modern Greek retains from its ancestor language Indo-European a special vocative case form
that is used when addressing someone. Masculine nouns lose their final -s in the vocative, and
other nouns remain unchanged, but for all nouns, the article is obligatorily absent in the vocative,
as shown in (31).2

(31) Greek, vocative:

(*η) Ελένη, (*η) Βάλια, (*ο) Παύλος, (*ο) Γιάννης
(*i) Eleni, (*i) Valia, (*o) Pavlos, (*o) Yannis

Once again, under the misnomer view, names in Greek would change their syntactic category
depending on their case form—a bizarre consequence.
Chapter 2: Syntactic categories and constituenthood 44

Borrowing
A second type of argument against the misnomer view of proper nouns is based on linguistic
borrowing. Speakers of one language often borrow words from another language, either because
no native words exist for certain concepts or because the borrowed word is perceived as having
a cachet that the native counterparts lack. Proper nouns are easily borrowed in this way; we
need only to think of the many geographical names in English derived from various indigenous
languages. The prolific borrowing of proper nouns is not surprising if proper nouns are a sub-
category of nouns, since nouns are precisely the category that is most readily borrowed. But it is
unexpected under the misnomer view, since pronouns are ordinarily not borrowed at all.
On the basis of the considerations just discussed, we conclude, then, that proper nouns are
a semantically (and often syntactically) special subtype of noun.

1.3 Adjectives
Adjectives versus modifiers
In the teaching of high school English, the terms ‘adjective’ and ‘modifier’ seem to be used
relatively interchangeably. From the point of view of linguistic theory, this usage reflects a
lamentable lack of appreciation for the central linguistic distinction between function and form.
Modifiers are linguistic expressions that serve a certain function—namely, to restrict or qualify
some other expression. Adjectives, on the other hand, are members of a syntactic category that
is defined by certain formal properties. For instance, it is possible to derive adverbs from many
adjectives (heavy, heavily; mere, merely; rough, roughly; sweet, sweetly).
There is no one-to-one relation between modifiers and adjective phrases. Modifiers are not
necessarily adjective phrases, as illustrated in (32).

(32) a. Adjective phrase:

a very aggressive driver
b. Adverb phrase:
They drive very aggressively.
c. Prepositional phrase:
the desk next to the window
They arrived on time.
d. Noun phrase:
They will arrive Monday night

Conversely, adjective phrases are not necessarily modifiers. For instance, the adjective
phrase in (33) is the predicate of a simple clause.

(33) This candidate is very suitable.

See the next section for more discussion of predicative uses of adjectives, and see §3 of Chapter 3
for discussion of the notion of a predicate.
Chapter 2: Syntactic categories and constituenthood 45

Adnominal versus predicative use

In English, adjective phrases that modify a noun typically occur in prenominal position.

(34) a. a very big box, this blue book

b. * a box very big, this book blue

However, there are other languages that typically require adjectival modifiers of nouns to be
postnominal. The examples in (35) are from French.

(35) a. une chaise très confortable, ce livre bleu

a chair very comfortable this book blue
b. * une très confortable chaise, ce bleu livre
a very comfortable chair this blue book

But even in English, adjective phrases can occur postnominally if they are ‘heavy’ enough,
and even relatively short adjectives are sometimes used postnominally to convey elegance or
high style.3

(36) a. ✔ an unlikely story

b. * an unlikely to be true story
(37) a. * a story unlikely
b. ✔ a story unlikely to be true
(38) House Beautiful (upscale interior decorating magazine)

A convenient cover term for ‘prenominal’ and ‘postnominal’ is adnominal. It is worth

noting that certain adjectives can only be used adnominally, and not predicatively.

(39) a. the mere fact, her utter surprise

b. * The fact is mere, Her surprise was utter

The postnominal use of adjective phrases should not be confused with their predicative use, as
illustrated above in (33).

Gradable versus categorical adjectives

Many adjectives have the property of denoting properties that an entity can exhibit to a greater
or lesser degree. Examples of such gradable adjectives are given in (40).

(40) beautiful, heavy, kind, loud, rich

Gradable adjectives can be modified by means of degree adverbs like enough, exceedingly,
overly, too, very, and so on, and they are associated with three degrees of comparison, as illus-
trated in Table 2.4.
Chapter 2: Syntactic categories and constituenthood 46

Positive Comparative Superlative

friendly friendli-er friendli-est
heavy heavi-er heavi-est
compatible more compatible most compatible
intelligent more intelligent most intelligent

Table 2.4: Gradable adjectives in English

English is unusual in having two ways of forming the degrees of comparison, an inflec-
tional way involving the use of the bound morphemes -er and -est, and an analytical way in-
volving the use of the free morphemes more and most. Most languages allow only one option or
the other.4 For instance, German allows only the inflectional option.5

Positive Comparative Superlative Gloss

freundlich freundlich-er freundlich-st- ‘friendly’
schwer schwer-er schwer-st- ‘heavy’
kompatibel kompatibl-er kompatibel-st- ‘compatible’
intelligent intelligent-er intelligent-est ‘intelligent’

Table 2.5: Gradable adjectives in German

The Romance languages, on the other hand, generally only allow the analytical option, as
illustrated in Table 2.6 for French.6

Positive Comparative Superlative Gloss

belle plus belle la plus belle ‘beautiful.f’
long plus long le plus long ‘long.m’
compatible plus compatible le/la plus compatible ‘compatible’
intelligent plus intelligent le plus intelligent ‘intelligent.m’

Table 2.6: Gradable adjectives in French

In contrast to gradable adjectives, some adjectives denote categorical properties—

properties that entities can either have or not have. For instance, integers are either prime or
not. Some further examples of categorical adjectives are given in (41).

(41) dead, married, pregnant, unique

It is possible for adjectives that are basically categorical to have restricted or extended
senses that are gradable. For instance, pregnant has a restricted sense of ‘visibly pregnant’, and
unique has an extended vernacular sense of ‘unusual’, rather than the original strict sense of ‘one
of a kind’. Given these senses, sentences like (42) are expected.
Chapter 2: Syntactic categories and constituenthood 47

(42) a. Lois looks more pregnant than Stacy.

b. A more unique present would be hard to find.

1.4 Modals
Historically, English modals, which are listed in (43), derive from a special class of verbs in Proto-
Germanic (the ancestor of English and the other Germanic languages).

(43) can, could, may, might, must, shall, should, will, would

Modals have always differed from ordinary verbs in Germanic, and over the history of
English, they have diverged even further, to the point where they are now assigned a syntactic
category of their own. Because the modals have meanings that are often expressed in other
languages by verbal inflections, this category has been called I(nflection). Nowadays it is more
often called T(ense), but we will continue to use the older term, as the motivation for using “T”
has become obsolete.
In what follows, we review the ways that modals differ from verbs in English, both mor-
phologically (which forms they exhibit) and syntactically (how they combine with other elements
of a sentence).

Morphology
Modals differ markedly from verbs in the range of forms that they exhibit. English verbs appear
in a number of distinct forms (see § 8.1 for details), whereas modals have a single, invariant
form.
Modals never end in -s. even in sentences with third person singular subjects.

(44) a. ✔ She { can, may } play the piano.

b. * She { can-s, may-s } play the piano.

Modals also lack productive past tense forms. It is true that could, might, should, and would
originated in Germanic as past tense forms of can, may, shall, and will. But today, only could can
serve as the past tense of can, and that only in certain contexts.7
Chapter 2: Syntactic categories and constituenthood 48

Example Potential paraphrase

Nowadays, you can get one for a dollar. = Nowadays, it is possible to get one for a dollar.
Back then, you could get one for a dollar. = Back then, it was possible to get one for a dollar.
We can go there tomorrow. = It is possible for us to go there tomorrow.
We could go there tomorrow. ≠ It was possible for us to go there tomorrow.
You may ask the boss. = You are allowed to ask the boss.
You might ask the boss. ≠ You were allowed to ask the boss.
Shall I pick up some bread? = Is it a good idea for me to pick up some bread?
Should I pick up some bread? ≠ Was it a good idea for me to pick up some bread?

Table 2.7: Lack of productive past tense forms for English modals

Finally, and very strikingly, modals have lost their nonfinite forms (infinitives, -ing forms,
past participles); the missing forms must be paraphrased.

(45) Infinitive after (another) modal:

a. * They must can speak Spanish. (modal)
b. ✔ They must be able to speak Spanish. (paraphrase of modal)
c. ✔ They must speak Spanish. (verb)
(46) Infinitive after infinitival to:
a. * They want to can speak Spanish. (modal)
b. ✔ They want to be able to speak Spanish. (paraphrase of modal)
c. ✔ They want to speak Spanish. (verb)
(47) -ing as gerund:
a. * They like { cann-ing, may-ing } play the piano. (modal)
b. ✔ They like { being able, being allowed } to play the piano. (paraphrase of modal)
c. ✔ They like learning to play the piano. (verb)
(48) Past participle:
a. * They have { cann-ed, may-ed } play the piano. (modal)
b. ✔ They have { been able, been allowed } to play the piano. (paraphrase of modal)
c. ✔ They have learned to play the piano. (verb)
Chapter 2: Syntactic categories and constituenthood 49

Syntax
The loss of nonfinite forms for modals gives rise to three further differences between verbs and
modals, all of them manifestations of an important phenomenon in the grammar of modern Eng-
lish called do support.

Emphasis
In the simplest case, do support affects affirmative sentences whose truth is being emphasized.
It involves replacing the finite verb by the verb’s bare form and adding a form of auxiliary do to
the sentence in the appropriate tense (either present or past tense). This form of do then receives
emphatic stress, as indicated by underlining in (49).

(49) a. ✔ He dances; she sang. (unemphatic, without do support)

b. ✔ He does dance; she did sing. (emphatic, with do support)

By contrast, emphasizing the truth of a sentence with a modal is achieved by simply stress-
ing the modal, and do support with modals is ungrammatical.

(50) a. ✔ He can dance; she will sing. (emphasis without do support)

b. * He does can dance; she does will sing. (emphasis with do support)

Negation
English sentences containing modals are negated by simply adding not (or its contracted form
-n’t) after the modal. Do support is ungrammatical.

(51) a. ✔ He { may, must, should, will, would } not dance. (negation without do support)
b. * He does not { may, must, should, will, would } dance. (negation with do support)

By contrast, negated sentences without modals require do support in modern English. As

in the case of emphasis, the verb appears in its bare form, and an appropriately tensed form of
auxiliary do is added to the sentence, followed by negation.

(52) a. ✔ She { does, did } not dance. (negation with do support)

b. * She not { dances, danced }. (negation without do support)
* She { dances, danced } not.
Chapter 2: Syntactic categories and constituenthood 50

Questions
The final difference between modals and verbs concerns question formation. If a declarative
sentence contains a modal, the corresponding question is formed by inverting the modal with
the subject. Do support is ungrammatical.

(53) a. ✔ { Can, may, must, should, will, would } he dance? (question without do support)
b. * Does he { can, may, must, should, will, would } dance? (question with do support)

Once again, however, in a sentence without a modal, question formation requires do sup-
port. That is, it is an appropriately tensed form of do, rather than the verb itself, that inverts with
the subject.

(54) a. ✔ { Does, did } she dance? (question with do support)

b. * { Dances, danced } she? (question without do support)

1.5 Auxiliary verbs

English has three auxiliary verbs, each homonymous with an ordinary verb: be, do, and
have.

Auxiliary do
The goal of the previous section was to establish the special status of modals, and we used the
facts of do support as a criterion for distinguishing modals from verbs. In this section, we con-
sider some of the same facts, but with a different focus. Rather than focusing on the distinctive
properties of modals, we focus on the morphological and syntactic properties of auxiliary do
itself.
The only difference between auxiliary do and the modals is that it has an -s form. In this
respect, it patterns with ordinary verbs, including its main verb counterpart.

(55) Modal:
a. I can dance the polka.
b. He { can, * can-s } dance the polka.

(56) Auxiliary do (emphasis, negation, question):

a. I do dance the polka. I do not dance the polka. Do you dance the polka?
b. She do-es dance the polka. She do-es not dance the polka. Do-es she dance the polka?
(57) Main verb do:
a. I do the dishes.
b. He do-es the dishes.
(58) Ordinary verb:
a. I dance the polka.
Chapter 2: Syntactic categories and constituenthood 51

b. She dance-s the polka.

In all other respects, auxiliary do resembles a modal rather than an ordinary verb. In par-
ticular, it does not occur in nonfinite contexts. Notice the clear contrast between the judgments
for auxiliary do in (60) and main verb do in (61).

(59) Modal:
a. * They will can dance the polka. (after another modal)
b. * They want to can dance the polka. (after infinitival to)
c. * Their canning dance the polka while blindfolded is unusual. (gerund)
(60) Auxiliary do (emphasis):
a. * They will do dance the polka. (after another modal)
Intended meaning: It will be the case they do dance the polka.
b. * They claim to do dance the polka. (after infinitival to)
Intended meaning: They claim that they do dance the polka.
c. * Their doing dance the polka while blindfolded is unwise. (gerund)
Intended meaning: That they do dance the polka while blindfolded is unwise.

(61) Main verb do:

a. ✔ They will do the dishes. (after another modal)
b. ✔ They want to do the dishes. (after infinitival to)
c. ✔ Their doing the dishes was considerate. (gerund)
(62) Ordinary verb:
a. ✔ They will dance the polka. (after another modal)
b. ✔ They want to dance the polka. (after infinitival to)
c. ✔ Their dancing the polka while blindfolded is unwise. (gerund)

Auxiliary do also behaves like a modal in do support contexts. Double instances of auxiliary
do are ruled out, just like double modals are (see (48a)). Once again, auxiliary do and main verb
do differ sharply, as shown in (64) and (65).

(63) Modal:
a. * They does can dance the polka. (emphasis)
b. * They doesn’t can dance the polka. (negation)
c. * Does he can dance the polka? (question)
(64) Auxiliary do:
a. * She does do dance the polka. (emphasis)
Intended meaning: It is the case that she does dance the polka.
b. * She doesn’t do dance the polka. (negation)
Intended meaning: It isn’t the case that she does dance the polka.
Chapter 2: Syntactic categories and constituenthood 52

c. * Does she do dance the polka? (question)

Intended meaning: Is it the case that she does dance the polka?

(65) Main verb do:

a. ✔ He does do the dishes. (emphasis)
b. ✔ He doesn’t do the dishes. (negation)
c. ✔ Does he do the dishes? (question)
(66) Ordinary verb:
a. ✔ She does dance the polka. (emphasis)
b. ✔ She doesn’t dance the polka. (negation)
c. ✔ Does she dance the polka? (question)

Auxiliary have
Let’s now turn to auxiliary have, which combines with past participles to form the perfect forms
of verbs. Auxiliary have behaves like a V with respect to its morphology and its occurrence
in nonfinite contexts, but like an I with respect to do support. Specifically, auxiliary have, like
auxiliary do, shares all the morphological properties of its main verb counterpart. In addition,
it can appear in nonfinite contexts (unlike auxiliary do). With respect to do support, however,
auxiliary have differs from its main verb counterpart and patterns together with the modals and
auxiliary do. The complex behavior of auxiliary have can be captured by saying that it moves
from V to I in the course of the derivation. (see Chapter 10, § 4.5 for further discussion.)
(67) and (68) show that auxiliary have, like auxiliary do (cf. (56)), behaves morphologically
like its main verb counterpart in having an -s form.

(67) Auxiliary have:

a. I have adopted two cats.
b. He ha-s adopted two cats.
(68) Main verb have:
a. I have two cats.
b. He ha-s two cats.

Auxiliary have differs from auxiliary do (cf. (60)) and resembles main verb have in being
able to appear in nonfinite contexts.

(69) Auxiliary have:

a. ✔ They must have adopted two cats. (after modal)
b. ✔ They claim to have adopted two cats. (after infinitival to)
c. ✔ They do not regret having adopted two cats. (gerund)
(70) Main verb have:
a. ✔ They must have two cats. (after modal)
Chapter 2: Syntactic categories and constituenthood 53

b. ✔ They claim to have two cats. (after infinitival to)

c. ✔ They do not regret having two cats. (gerund)

However, just like auxiliary do (cf. (64)) and in contrast to main verb have, auxiliary have is ruled
out in do support contexts.

(71) Auxiliary have:

a. * He does have adopted two cats. (emphasis)
b. * He doesn’t have adopted two cats. (negation)
c. * Does he have adopted two cats? (question)
(72) Main verb have:
a. ✔ She does have two cats. (emphasis)
b. ✔ She doesn’t have two cats. (negation)
c. ✔ Does she have two cats? (question)

Be (auxiliary and main verb)

The examples in (73)-(75a) illustrate the behavior of auxiliary be, which is used to form the pro-
gressive (is coming, was dancing) and the passive (is abandoned, was sold) in English. Auxiliary
be behaves just like auxiliary have. In particular, it has an -s form (irregular though that form is),
and it can appear in nonfinite contexts, but it is excluded from do support contexts. As a result,
auxiliary be can be treated just like auxiliary have—as belonging to the syntactic category V, but
moving from V to I in the course of a derivation.
Main verb be differs from all other main verbs in English in behaving exactly like its aux-
iliary counterpart. In other words, main verb be is the only main verb in modern English that
moves from V to I.

(73) Auxiliary be:

a. ✔ They are learning Spanish. They are invited to the ceremony. (non-third person)
b. ✔ He i-s learning Spanish. He i-s invited to the ceremony. (third person)
(74) Main verb be:
a. ✔ They are happy. (non-third person)
b. ✔ She i-s happy. (third person)
(75) Auxiliary be:8
a. ✔ They must be learning Spanish. (after modal)
b. ✔ They must be invited to the ceremony.
c. ✔ They claim to be learning Spanish. (after infinitival to)
d. ✔ They claim to be invited to the ceremony.
e. ✔ They don’t regret being invited to the ceremony. (gerund)
(76) Main verb be:
a. ✔ They must be happy. (after modal)
Chapter 2: Syntactic categories and constituenthood 54

b. ✔ They claim to be happy. (after infinitival to)

c. ✔ They don’t regret being syntacticians. (gerund)

(77) Auxiliary verb be:

a. * He does be learning Spanish. He does be invited to the ceremony. (emphasis)
b. * He doesn’t be learning Spanish. He doesn’t be invited to the ceremony. (negation)
c. * Does he be learning Spanish? Does he be invited to the ceremony? (question)
(78) Main verb be:
a. * She does be happy. (emphasis)
b. * She doesn’t be happy. (negation)
c. * Does she be happy? (question)

Summary
Table 2.8 provides a synopsis of the morphological and syntactic properties of the items that we
have discussed, arranged from most to least verb-like. As is evident from the table, the syntac-
tic category of an item depends on whether it is the verb-like or the modal-like properties that
predominate.

Ordinary verb Be and auxiliary have Auxiliary do Modal

Has -s form yes yes yes no
Occurs in nonfinite contexts yes yes no no
Occurs with do support yes no no no
Syntactic category V V I I

Table 2.8: Summary of morphological and syntactic properties of ordinary verbs, auxiliaries
verbs, and modals in English

Tests for determining syntactic constituenthood 2

At first glance, a sentence consists of a string of words arranged in a single dimension—that
of linear order. However, we presented evidence in Chapter 1 for a second dimension that is
less obvious (though no less real!) than linear order—the dimension of syntactic structure.
Whether a particular string of words is a syntactic constituent isn’t always self-evident, and so
several tests have been developed as diagnostics for constituenthood. In this chapter, we review
these tests, along with some of the complications that arise in applying them. We also discuss in
more detail how syntactic structure is represented in tree diagrams of the sort introduced in § 5.1
of Chapter 1.
Chapter 2: Syntactic categories and constituenthood 55

2.1 Substitution
The most basic test for syntactic constituenthood is the substitution test. The reasoning behind
the test is simple. A constituent is any syntactic unit, regardless of length or syntactic category. A
single word is the smallest free-standing constituent belonging to a particular syntactic category.
So if a single word can substitute for a string of several words, that’s evidence that the string is
a constituent.
We mentioned in Chapter 1, § 3.1 that pronouns can substitute for noun phrases. Some
examples are given in (79).

(79) a. The little boy fed the cat. → ✔ He fed her.

b. Black cats detest green peas. → ✔ They detest them.

As already mentioned in Chapter 1, it’s important to understand that a particular string

of words can be a noun phrase in one syntactic context, but not in another. For instance, the
substitution test tells us that the underlined strings are noun phrases in (79), but not in (80).

(80) a. The little boy from next door fed the cat without a tail.
↓ ↓
* He from next door fed her without a tail.

b. These black cats detest those green peas.

↓ ↓
* These they detest those them.

Rather, in these sentences, the noun phrases are the longer underlined strings in (81).

(81) a. The little boy from next door fed the cat without a tail. → ✔ He fed her.
b. These black cats detest those green peas. → ✔ They detest them.

Pronouns are not the only placeholder elements, or pro-forms. For instance, the adverbs
here or there can substitute for constituents that refer to locations or directions. As in the case of
noun phrases, whether a particular string is a constituent depends on its syntactic context.

(82) a. Put it on the table. → ✔ Put it there.

b. Put it over on the table. → ✔ Put it over there.
c. Put it over on the table. → ✔ Put it there.

(83) a. Put it on the table that’s already set. → * Put it there that’s already set.
b. Put it over on the table that’s already set. → * Put it over there that’s already set.
c. Put it over on the table that’s already set. → * Put it there that’s already set.

It’s worth noting that the syntactic category of the pro-forms in (82) and (83) differs from
the category of the underlined strings. As just mentioned, the pro-forms are adverbs, whereas the
underlined strings are prepositional phrases. The mismatch in syntactic category doesn’t affect
Chapter 2: Syntactic categories and constituenthood 56

the validity of the test, however. All that is necessary is that the pro-form can replace the string,
preserving (at least roughly) the original meaning.
The word so can substitute for adjective phrases (here, the most natural-sounding results
are obtained in contexts of comparison or contrast). As usual, the same string sometimes is a
constituent and sometimes isn’t. (The judgments in (84) are ours; for some speakers, (84c) is
grammatical.)

(84) a. ✔ I am very happy, and Linda is so, too.

b. ✔ I am very fond of Lukas, and Linda is so, too.
c. * I am very fond of my nephew, and Linda is so of her niece.

Finally, pronouns and sometimes the word so can substitute for subordinate clauses intro-
duced by that, as in (85) and (86).

(85) a. I { know, suspect } that they’re invited.

b. ✔ I { know, suspect } it.
(86) a. I { imagine, think } that they’re invited.
b. ✔ I { imagine, think } so.

We conclude our discussion of substitution by noting a complicating factor. Nowhere is

it written that every constituent has a corresponding pro-form. For instance, although some
prepositional phrases can be replaced by the pro-forms here or there, others—for instance, ones
referring to purposes or reasons—can’t. As a result, it is perfectly possible for the substitution test
to give false negative results, and for the results of the substitution to disagree with the results
of the other tests presented in what follows. We return to the issue of false negatives later on in
section 3.2.

2.2 Movement
Substitution by pro-forms is not the only diagnostic for whether a string is a constituent. If it is
possible to move a particular string from its ordinary position to another position—typically, at
least in English, to the beginning of the sentence—that, too, is evidence that the string is a con-
stituent. In order to make the result of movement completely acceptable, especially in the case of
noun phrases, it’s sometimes necessary to use a special intonation or to invoke a special discourse
context. In the examples that follow, “ ” indicates the ordinary position that a constituent has
moved from, and appropriate discourse material (enclosed in parentheses) may be added to make
the examples sound natural.

(87) a. I fed the cats.

b. ✔ The cats, I fed . (The dogs, I didn’t.)

(88) a. I fed the cats with long, fluffy tails.

b. ✔ The cats with long, fluffy tails, I fed . (The other cats, I didn’t.)
Chapter 2: Syntactic categories and constituenthood 57

Movement of constituents other than noun phrases is illustrated in (89)–(91).

Adjective phrase:

(89) a. Ali Baba returned from his travels wiser than before.
b. ✔ Wiser than before, Ali Baba returned from his travels .

Adverb phrase:

(90) a. They arrived at the concert hall more quickly than they had expected.
b. ✔ More quickly than they had expected, they arrived at the concert hall .

Prepositional phrase:

(91) a. The cat strolled across the porch with a confident air.
b. ✔ With a confident air, the cat strolled across the porch .

2.3 Questions and short answers

Another diagnostic for constituenthood is whether the string can function as a short answer to a
question. The question itself also functions as a diagnostic test, since we can think of it as being
derived by substituting a question word for a string and subsequently moving the question word
to the beginning of the sentence.9 (92)–(96) illustrate this pair of tests for a variety of constituent
types.
Noun phrase: Adjective phrase:
(92) What do you see? (93) How did the traveler return?
a. The cats. a. Wiser than before.
b. Cats with long, fluffy tails. b. Fairly jet-lagged.
c. The cats with long, fluffy tails.

Adverb phrase: Prepositional phrase:

(94) How did they do? (95) How did the cat stroll across the porch?
a. Not badly. a. With a confident air.
b. Surprisingly well. b. quickly.
c. Much better than they had (96) Where did Ali Baba go?
expected. a. On a long journey.
b. To New York.

Notice, incidentally, that so substitution, discussed earlier, has a variant that is reminiscent
of questions. In addition to just substituting for the string of interest, so can subsequently move
Chapter 2: Syntactic categories and constituenthood 58

to the beginning of the sentence, triggering subject-aux inversion—the same process that turns
declarative sentences into yes-no questions. This variant of so substitution is illustrated in (97) and
(98). (The judgment for (97c) varies across speakers in the same way as the one for (84c).)

(97) a. ✔ I am very happy, and so is Linda.

b. ✔ I am very fond of Lukas, and so is Linda.
c. * I am very fond of my nephew, and so is Linda of her niece.

(98) ✔ I { imagine, think } that they’re invited, and so do they.

2.4 It clefts
The final constituent test that we’ll consider is based on a special sentence type known as it
clefts. We begin by noting that ordinary sentences can often be divided into two parts: a part
that contains background information that is presupposed, the ground, and a part that is intended
to be particularly informative, the focus. In spoken language, this focus-ground partition (also
known as its information structure) is often conveyed by intonation alone in English.10 But
English can also express the focus/ground distinction, via a syntactic frame consisting of it, a
form of the copula to be, and the subordinating conjunction that. In the examples in (99)–(100),
the frame is in standard font, the ground is in italics, and the focus is in bold. Notice that a single
sentence can be partitioned into focus and ground in more than one way, giving rise to more than
one it cleft.

(99) a. Ordinary cats detest the smell of citrus fruits.

b. ✔ It is ordinary cats that detest the smell of citrus fruits.
(100) a. Ordinary cats detest the smell of citrus fruits.
b. ✔ It is the smell of citrus fruits that ordinary cats detest.

If a string can appear as the focus of an it cleft, then it is a constituent. Some examples for
various constituent types other than noun phrase are given in (101)–(103).
Adjective phrase:

(101) a. Ali Baba returned from his travels wiser than before.
b. ✔ It was wiser than before that Ali Baba returned from his travels .

Adverb phrase:

(102) a. They arrived at the concert hall more quickly than they had expected.
b. ✔ It was more quickly than they had expected that they arrived at the concert hall
.

Prepositional phrase:

(103) a. The cat strolled across the porch with a confident air.
b. ✔ It was with a confident air that the cat strolled across the porch .
Chapter 2: Syntactic categories and constituenthood 59

Some complications for constituency diagnostics 3

3.1 Mismatches between syntactic structure and other structure

We mentioned earlier that it is not always self-evident whether a particular sequence of words is
a syntactic constituent. For instance, in reading a sentence like (104) out loud, we can perceive
an intonation break between cat and that (indicated by the slash).

(104) This is the cat / that chased the rat.

Because the intonation break is clearly audible, it is very tempting to equate the sentence’s ab-
stract syntactic structure with its relatively concrete prosodic structure. Specifically, because the
and cat belong to the same prosodic constituent, it is tempting to treat the cat as a syntactic
constituent.
There are two pieces of evidence against doing so. First, as we have already seen in similar
examples, substituting a pronoun for the string the cat is ungrammatical in the context of (104)
(though not in other contexts).

(105) a. This is the cat that chased the rat.

b. * This is it that chased the rat.
(106) a. I petted the cat.
b. ✔ I petted it.

Second, the string cat that chased the rat is shown to be a constituent by the grammaticality
of substituting the pro-form one (one substitution is discussed in more detail in section 1.4 of
Chapter 5).

(107) a. This is the cat that chased the rat.

b. * This is the one.

The facts in (105) and (107) converge to tell us that the word cat first combines with the
relative clause, not with the. Thus, (104) exhibits a mismatch between two types of linguistic
structure: syntactic and prosodic.
It is worth noting that the syntactic structure just described corresponds to the way that
the interpretation of the entire expression the cat that chased the rat is composed from the in-
terpretation of smaller expressions. In a simple semantics, the term cat denotes the set of all
cats. Combining cat with the relative clause yields cat that chased the rat, which denotes a sub-
set of all cats—namely, those with the property of having chased the rat. Further combining cat
that chased the rat with the definite article the yields as a denotation some unique individual
within the rat-chasing subset of cats (exactly which individual this is depends on the discourse
context).11
Chapter 2: Syntactic categories and constituenthood 60

This correspondence of syntactic structure and semantic structure (the step-by-step compo-
sition of the expression’s meaning, as just illustrated for (104)) holds up as a first approximation,
and it is consistent with the correspondence between noun phrases and individuals, between
adjective phrases and properties, between prepositional phrases and locations, directions, etc.,
between verb phrases and events, states, etc., and so on. Nevertheless, mismatches between syn-
tactic structure and semantic structure are possible. For instance, the sentence in (108) has two
distinct meanings, which can be paraphrased as in (109).

(108) Every student knows two languages.

(109) a. For every student, it is the case that they know two languages.
(Langston knows Arabic and Basque, Maisha knows Chinese and Danish, Lucca
knows English and French, …)
b. There are two languages that every student knows.
(Arabic and Basque are known by Langston, Maisha, Lucca, …)

In the interpretation in (109a), the universal quantifier every is said to take scope over the
number two (EVERY > TWO). In the interpretation in (109b), the number takes scope over the
universal quantifier (TWO > EVERY). In either case, though, the ambiguous sentence itself (not
the paraphrases!) has a single syntactic structure. This is evident from the syntactic constituent-
hood tests in (110)–(111), where the question and short answer pair are compatible with either
scope interpretation.

(110) a. Every student knows two languages.

b. Who knows two languages? Every student.
(111) a. Every student knows two languages.
b. What does every student know? Two languages.

Other mismatches are also possible. Recall from the section on it clefts that one and the
same sentence can be associated with more than one information structure. Finally, mismatches
between syntactic and morphological structure are common (we discuss two important cases in
Chapters 10 and 11).

3.2 False negative results

In a perfect world for syntacticians, the relation between a string’s being a constituent and its
passing the constituenthood tests (also called constituency tests) would be biconditional, as in-
dicated in the “Dream world” column in Table 2.9. The world, however, is not made to order
for syntacticians, and it turns out to be possible for constituents to fail one or more constituent-
hood tests. The actual state of affairs is thus as indicated in the “Real world” column of Table
2.9.12
Chapter 2: Syntactic categories and constituenthood 61

If a string… then it… Dream world Real world

passes the constituency tests, is a constituent. TRUE TRUE
is a constituent, passes the constituency tests. TRUE sometimes FALSE

Table 2.9: False negatives in constituency tests

In other words, the failure of a string to pass a constituenthood test can be a false negative
result. In what follows, we present three such cases—constituents that fail at least some of the
constituenthood tests
▶ because they are words rather than phrases,
▶ because they contain finite verbs, or
▶ because they are contained within so-called syntactic islands.

Phrasal versus lexical constituents.

Since single words are indivisible units, they are constituents by definition.13 Nevertheless, they
don’t necessarily behave on a par with multiword constituents. For instance, cats passes the
constituenthood tests reviewed earlier in (112), but not in (113).

(112) Cats are not social animals. → They are not social animals.

(113) a. The cats are hungry. → * The they are hungry.

b. Tabby cats are quite common. → * Tabby they are quite common.
c. Cats without tails are relatively rare. → * They without tails are relatively rare.

The reason for the grammaticality contrast in (112) and (113) is a systematic difference
between the syntactic contexts in these examples. In (113), the word cats is accompanied by a
determiner or a modifier of some sort, indicated by italics. In such contexts, cats combines with
these other words to form a noun phrase, but it isn’t a noun phrase in its own right. By contrast,
cats in (112) is a bare, or unmodified, noun. As such, it functions as a noun and as a noun phrase
at the same time. In other words, there are two levels of constituenthood: the lexical level,
where single words are constituents by definition, and the phrasal level, where single words
don’t necessarily behave on a par with multiword constituents.
The constituenthood tests reviewed earlier turn out to be diagnostic only for phrasal con-
stituents. Moving, questioning, and it-clefting lexical constituents, rather than phrasal ones,
yields ungrammatical results, as illustrated in (114)–(125). As before, the relevant lexical con-
stituent is underlined, and any material belonging with it to the same phrasal constituent is in
italics.
Movement:

(114) a. I fed the cats.

b. * Cats, I fed the .
Chapter 2: Syntactic categories and constituenthood 62

(115) a. Ali Baba returned from his travels wiser than before.
b. * Wiser, Ali Baba returned from his travels than before.

(116) a. They arrived at the concert hall more quickly than they had expected.
b. * Quickly, they arrived at the concert hall more than they had expected.

(117) a. The cat strolled across the porch with a confident air.
b. * With, the cat strolled across the porch a confident air.

Question/short answer:

(118) a. * What did you see the ?

b. * Cats.

(119) a. * How did Ali Baba return from his travels than before?.
b. * Wiser.

(120) a. * How did they arrive at the concert hall more than they expected?
b. * Quickly.

(121) a. * How did the cat stroll across the porch a confident air?
b. * With.

It cleft:

(122) a. Ordinary cats detest the smell of citrus fruits.

b. * It is smell that ordinary cats detest the of citrus fruits.

(123) a. Ali Baba returned from his travels wiser than before.
b. * It was wiser that Ali Baba returned from his travels than before.

(124) a. They arrived at the concert hall more quickly than they had expected.
b. * It was quickly that they arrived at the concert hall more than they had expected.

(125) a. The cat strolled across the porch with a confident air.
b. * It was with that the cat strolled across the porch a confident air.

By contrast, the examples in (126)–(134) illustrate the grammatical results of moving, ques-
tioning, and it-clefting phrasal constituents that happen to consist of a single word (notice the
absence of italicized material in this case). Examples for prepositional phrases are missing because
prepositions, at least the ones in the examples that we have been using, require an object.
Movement:

(126) a. I like cats.

b. ✔ Cats, I like .
Chapter 2: Syntactic categories and constituenthood 63

(127) a. Ali Baba returned from his travels wiser.

b. ✔ Wiser, Ali Baba returned from his travels .

(128) a. They arrived at the concert hall quickly.

b. ✔ Quickly, they arrived at the concert hall .

Question/short answer:

(129) a. ✔ What do you like ?

b. ✔ Cats.

(130) a. ✔ How did Ali Baba return from his travels ?.

b. ✔ Wiser.

(131) a. ✔ How did they arrive at the concert hall ?

b. ✔ Quickly.

It cleft:

(132) a. Ordinary cats detest citrus.

b. ✔ It is citrus that ordinary cats detest .

(133) a. Ali Baba returned from his travels wiser.

b. ✔ It was wiser that Ali Baba returned from his travels .

(134) a. They arrived at the concert hall quickly.

b. ✔ It was quickly that they arrived at the concert hall .

Finiteness
Testing for the constituenthood of verb phrases is more complicated than is testing for the con-
stituenthood of other syntactic categories. First, there are no simple pro-forms for verb phrases.
The best we can do is to use the periphrastic forms do so for substitution and do what for ques-
tions.14 (Notice that it’s only what, rather than the entire pro-form do what, that moves to the
beginning of a question.)

(135) Substitution:
a. She will write a book. → ✔ She will do so.
b. The two boys could order tuna salad sandwiches. → ✔ The two boys could do so.

(136) Question/short answer:

a. What will she do? → ✔ Write a book.
b. What could the two boys do? → ✔ Order tuna salad sandwiches.
Chapter 2: Syntactic categories and constituenthood 64

Second, and more importantly, given our present focus on false negative results, verbs and
the verb phrases that contain them come in two varieties, finite and nonfinite. Roughly speak-
ing (see §8 for details), finiteness refers to whether the verb bears tense and agreement (finite) or
not (nonfinite). Finite verbs can function on their own as the core of an independent sentence,
whereas nonfinite verbs cannot. Now, two of the constituenthood tests—substitution and the
question/short answer test—yield grammatical results regardless of a verb phrase’s finiteness, as
shown in (137) and (138).

(137) a. Substitution, nonfinite verb phrase:

She will write a book. → ✔ She will do so.
b. Substitution, finite verb phrase:
She wrote a book. → ✔ She did so.

(138) a. Question/short answer, nonfinite verb phrase:

What will she do? → ✔ Write a book.
b. Question/short answer, finite verb phrase:
What did she do? → ? Wrote a book.

But the results from the other two tests are more complex. Movement of nonfinite verb
phrases is grammatical,15 but movement of finite ones is not.

(139) Movement, nonfinite verb phrase:

a. (She says that) she will write a book, → ✔ (and) write a book, she will .
b. though she may write a book → ✔ write a book though she may
(140) Movement, finite verb phrase:
a. (She says that) she wrote a book, → * (and) wrote a book, she .
b. though she wrote a book → * wrote a book though she

In it clefts, nonfinite verb phrases are marginally acceptable in focus, whereas finite verb
phrases are again clearly ruled out.

(141) It cleft, nonfinite verb phrase:

She will write a book. → ⁇ It is write a book that she will .
(142) It cleft, finite verb phrase:
She wrote a book. → * It is wrote a book that she .

To summarize: we have good evidence that nonfinite verb phrases are constituents. In
the case of finite verb phrases, we have evidence for constituenthood from two of the four con-
stituenthood tests. Given this slightly complex state of affairs, we will proceed as follows. We
will make the simplifying assumption that the ungrammaticality of moving or focusing finite verb
phrases has nothing to do with their constituenthood, but that it is due to some other reason, yet
to be determined. (We give you a stab at solving the problem in Problem 10.2.) Having made this
assumption, we are free to treat finite verb phrases as constituents on a par with their nonfinite
Chapter 2: Syntactic categories and constituenthood 65

counterparts even though the syntactic behavior of the two types of verb phrases is not identical
in all respects.
Chances are that you are a bit leery of the simplifying assumption just described. If so,
think of it as comparable to taking out a loan. True, taking out a loan is risky, and taking out
loans in a careless or irresponsible way can lead to financial disaster. Nevertheless, the credit
market is a necessary and productive part of any modern economy. In a similar way, making
simplifying assumptions in science can help us to make progress where we would otherwise be
stumped by the complexity of the phenomena that we are investigating. Of course, we have to be
careful about what simplifying assumptions we make. Otherwise, we end up fooling ourselves
into believing that we are making progress, when in fact we are working on such a distorted
model of reality that our work is worthless.
Apart from this wrinkle concerning finiteness, verb phrases behave just as we have come
to expect from other constituent types. The tests yield grammatical results for verb phrases, but
not for verbs.

(143) Substitution:
She will write a book. → * She will do so a book.

(144) Movement:
a. (She says that) she will write a book, → * and write, she will a book.
b. though she may write a book → * write though she may a book

(145) Question/short answer:

a. * What will she do a book?
b. * Write.

(146) It cleft:
She will write a book. → * It is write that she will a book.

And once again, particular strings can be phrasal constituents in one syntactic context, but
not in another. For instance, write isn’t a phrasal constituent when it combines with a direct
object, but it is when used on its own. This is the source of the grammaticality contrast between
(143)–(146) and (147)–(150).

(147) Substitution:
She will write. → ✔ She will do so.

(148) Movement:
a. (She says that) she will write, → ✔ and write, she will .
b. though she may write → ✔ write though she may

(149) Question/short answer:

a. What will she do?
b. ✔ Write.
Chapter 2: Syntactic categories and constituenthood 66

(150) It cleft:
She will write. → ? It is write that she will .

As we proceed, we will develop a model of phrase structure that readily explains the apparently
variable behavior of verbs as constituents (or not).

Islands
In (151a), the doctors is a constituent, as is evident from the possibility of substituting a pronoun
for the string, as in (151b).

(151) a. We should invite the lawyers and the doctors.

b. We should invite the lawyers and them. (pointing to the doctors)

But although the doctors passes the substitution test, the other three tests yield ungram-
matical results.

(152) * The doctors, we should invite the lawyers and .

(153) a. * Who should we invite the lawyers and ?
b. * The doctors.
(154) * It is the doctors that we should invite the lawyers and .

Take Note

Earlier, we pointed out that question formation can be thought of as a combination of

substitution and movement. The parallel between movement and question formation in
(152), (153) and it clefting in (154) suggests that the latter, too, involves movement in some
way. In the remainder of this section, we will therefore use the term ‘movement’ in a broad
sense to include all three constituenthood tests in (152)–(154).

Notice that there is nothing semantically or conceptually ill-formed about (152)–(154), since
it is possible to paraphrase the intended meaning grammatically as in (155)–(157).

(155) ✔ The doctors, we should invite together with the lawyers.

(156) a. ✔ Who should we invite together with the lawyers?
b. ✔ The doctors.
(157) ✔ It is the doctors that we should invite together with the lawyers.

Taken together with the grammaticality of (151b), the contrast between (152)–(154) and
(155)–(157) shows that movement of the noun phrase the doctors is somehow prevented by the
specific syntactic configuration in (152)–(154). Ross (1967) introduced the metaphorical term
island for configurations in which movement is blocked where it is expected to be possible. The
conceit underlying the term is that the constituents that might be expected to move, but can’t,
are stranded on an island like castaways.16
Chapter 2: Syntactic categories and constituenthood 67

Ross identified several types of islands, including conjoined phrases like those in (152)–
(154), and his work has given rise to an enormous body of literature. Our purpose here is neither
to catalog the types of islands nor to pursue the proper linguistic analysis of them (we return to
the topic in Chapter 13), but simply to point out that constituenthood tests based on movement
will yield false negative results for phrasal constituents if they happen to be contained within
islands.

Representing syntactic constituenthood 4

In Chapter 1, § 5.1, we introduced tree diagrams as a convenient way of representing syntactic
structure. For mathematicians working in the field of graph theory, the formal properties of tree
diagrams are interesting in their own right, but for syntacticians, the interest of trees lies in the
fact that they are representations, or models, of constituent structure. In other words, the
graphic structure of a tree on the page is intended as a statement (or at least, a hypothesis) about
the way that speakers group together syntactic elements in their minds. In a good model, the
properties of the model correspond straightforwardly to the properties of the domain of inquiry.
Such a close correspondence allows us to state observations and generalizations about the domain
of inquiry without undue complication. Moreover, if we’re lucky, we might even be able to use
our understanding of the model’s formal properties as a sort of conceptual lever to generate
hypotheses and to discover facts and generalizations about the domain of inquiry that would
otherwise escape notice.
In light of these considerations, let’s consider the sentence in (158), focusing particularly
on the constituenthood of the underlined string.

(158) The secretary drafted the letter.

According to the two tests that apply to finite verb phrases, the string draft the letter is a
constituent.
Substitution:

(159) The secretary drafted the letter. → The secretary did so.

Question/short answer:

(160) a. What did the secretary do?

b. Draft the letter.

Having established this fact, let’s now consider two alternative representations of the sen-
tence. We’ve already encountered (161a) in Chapter 1, § 5.1. (161b) is an alternative, ‘flatter’
tree.
Chapter 2: Syntactic categories and constituenthood 68

(161) a. Sentence b. Sentence

NounPhr VerbPhr NounPhr TrVerb NounPhr

drafted
Det Noun TrVerb NounPhr Det Noun Det Noun
the secretary drafted the secretary the letter
Det Noun
the letter

At first glance, the flatter tree might be argued to be preferable on the grounds that it is
simpler in the sense of containing fewer nodes. But let’s focus on the question of which tree
is a better representation of the sentence as it exists in a speaker’s mind. Specifically, let’s ask
whether either of the trees in (161) has some graphic property that corresponds to the results of
the constituenthood tests in (159) and (160). In (161a), the answer is ‘yes’, since there is a single
node (the one labeled VerbPhr) that dominates all and only the words in the string drafted the
letter. We say that the node exhaustively dominates the string (for more details, see §4.2. By
contrast, the tree in (161b) lacks such a node and has no other graphic property corresponding to
the string’s constituenthood. Clearly, then, (161a) is a more useful model of the sentence, because
it follows the natural convention in (162).

(162) Syntactic constituents are represented graphically as nodes in a tree.

4.1 Basic terms and relations

Dominance
It is convenient to represent syntactic structure by means of graphic structures called trees; these
consist of a set of nodes connected by branches. It is sometimes useful to distinguish between
two types of nodes: terminal nodes, which are labeled with vocabulary items, and nonterminal
nodes, which are labeled with syntactic categories. In a very simple tree like (163), the only
terminal node is labeled Zelda, and the two nonterminals are labeled N and NP.

(163) NP

Zelda

In (163), no node has more than one branch emanating from it. The nodes in such a simple
tree are related to one another by a single relation, the dominance relation. Dominance is a
theoretical primitive; in other words, it is an irreducibly basic notion, comparable to a mathemat-
ical concept like point. Dominance is represented graphically in terms of top-to-bottom order.
That is, if a node A dominates a node B, A appears above B in the tree. In (163), for instance, NP
dominates N and Zelda, and N dominates Zelda. The node that dominates all other nodes in a
tree, and is itself dominated by none, is called the root node.
Chapter 2: Syntactic categories and constituenthood 69

Dominance is a transitive relation (in the logical sense of the term, not the grammatical
one). In other words, if A dominates B, and if B dominates C, then it is necessarily the case that
A dominates C.
Does a node A dominate itself? If the answer to this question is defined to be yes, then the
dominance relation is reflexive (again, in the logical sense of the term, not the grammatical one);
if not, then it is irreflexive. In principle, it is possible to build a coherent formal system based
on either answer. From the point of view of syntactic theory, it is preferable to define dominance
as reflexive because it simplifies the definitions of linguistically relevant derived relations such
as c-command and binding.
An important subcase of dominance is immediate dominance. This is the case where the
two nodes in question are connected by a single branch without any intervening nodes. More
formally, immediate dominance is defined as in (164).

(164) A immediately dominates B if and only if

a. A dominates B, and
b. there is no node C, distinct from A and B, such that A dominates C and C dominates
B.

Unlike dominance, immediate dominance is not a transitive relation. This is apparent from
even a simple structure like (163), where NP immediately dominates N, and N immediately dom-
inates Zelda, but NP does not immediately dominate Zelda.

Precedence
In general, trees are more complex than the very simple case in (163), and they contain nodes
that have more than one branch emanating from them, as in (165) (=Chapter 1, (51c)).

(165) Sentence

NounPhr VerbPhr

Det Noun TrVerb NounPhr

the secretary drafted Det Noun

the letter

In such trees, two nodes are related either by dominance or by a second primitive relation,
precedence. Precedence is represented graphically in terms of left-to-right order. Dominance
and precedence are mutually exclusive. That is, if A dominates B, A cannot precede B, and con-
versely, if A precedes B, A cannot dominate B. Like dominance, precedence is a transitive relation,
and just as with dominance, there is a nontransitive subcase called immediate precedence. The
definition of immediate precedence is analogous to that of immediate dominance; the term dom-
inates in (164) is simply replaced by precedes. The difference between precedence, which is tran-
sitive, and immediate precedence, which isn’t, can be illustrated in connection with (165). The
Chapter 2: Syntactic categories and constituenthood 70

first instance of Noun (the one that immediately dominates secretary) both precedes and
immediately precedes TrVerb, and TrVerb in turn both precedes and immediately precedes
the second instance of NounPhr (the one that dominates the letter). The first instance of
Noun precedes the second instance of NounPhr, but not immediately.

4.2 Derived terms and relations

Kinship terminology
Certain relations among nodes are often expressed by using kinship terms. If A dominates B,
then A is the ancestor of B, and B is the descendant of A. If A immediately dominates B, then
A is the parent of B, and B is the child of A. If A immediately dominates B and C, then B and
C are siblings. Often, the female kinship terms mother, daughter, and sister are used for
the corresponding sex-neutral ones. In (165), Sentence is the ancestor of every other node
in the tree. The terminal secretary is the child of the first Noun. The first NounPhr and
VerbPhr are sisters, and so are TrVerb and the second NounPhr, but drafted and the
second instance of the are not (they don’t have the same mother). Notice, incidentally, that
syntactic trees are single-parent families. Most theories of syntax do not allow nodes with more
than one parent.

Branching
Depending on the number of daughters, nodes are classified as either nonbranching (one daugh-
ter) or branching (more than one daughter). A more detailed system of terminology distin-
guishes nodes that are unary-branching (one daughter), binary-branching (two daughters),
and ternary-branching (three daughters). Nodes with more than three daughters are hardly
ever posited in syntactic theory. Indeed, according to an influential hypothesis (Kayne 1984),
Universal Grammar allows at most binary-branching nodes. According to this hypothesis, it is a
formal universal of human language that no node can have more than two branches.

Exhaustive dominance
Some node A exhaustively dominates two or more nodes B, C, … if and only if A dominates
all and only B, C, … For instance, A dominates the string B C in (166a)–(166c), but exhaustively
dominates it only in (166a). In (166b) and (166c), A fails to exhaustively dominate B and C be-
cause it runs afoul of the only condition (it dominates too much material). In (166d), A fails to
exhaustively dominate B and C for the opposite reason. A doesn’t dominate C, because it runs
afoul of the all condition (it dominates too little material).
Chapter 2: Syntactic categories and constituenthood 71

a. X b. Y c. Y d. X

A Y A X A X A C
(166)
B C B C D D E B

B C

As is evident from (166b)–(166d), dominance is a necessary but not a sufficient condition

for exhaustive dominance.

Regarding scientific models 5

We conclude this discussion of the model character of syntactic representations by emphasizing
that models are just that—models, and not the actual domain of inquiry itself. The purpose of any
model is to help us understand some part of reality that is too complex to understand in all of its
detail, at least all at once. This means that models are partial in two respects. A first way is that
they leave out properties of a phenomenon that aren’t relevant from a particular point of view.
This fact is often stated as the maxim “Don’t mistake the map for the territory”. For instance, a
mountaineer’s map might show topographical information in great detail, but completely ignore
political boundaries, whereas just the reverse might be true of a diplomat’s map. Analogously,
in linguistics, syntactic models leave out many important properties of language, such as real-
world plausibility, pragmatic felicity, the location of intonation breaks, variation based on social
factors like race, gender, age, and other aspects of identity, and so on. These are the focus of other
subdisciplines of linguistics, which in turn might ignore syntactic structure. Our attention on the
aspects of language we are working on should not be interpreted as trivializing the aspects we
are not working on.
A second way that models are partial is that they are subject to revision as our understand-
ing of a particular domain improves and deepens. Another way of putting this—which sounds
more impressive—is to say that scientific progress is possible.
 Chapter 2: Syntactic categories and constituenthood 72

Notes 6
1. These distributional lists are built based on Nagelhout et al. (2018)

2. For the (*) notation, see the glossary entry asterisk (*).

3. Originally, English, like the other Germanic languages, allowed adjective phrase modifiers only in prenominal po-
sition. After the Norman Conquest in 1066, Norman French became the official language of England and remained
so for several centuries. In Norman French, as in the other Romance languages, adjective phrase modifiers were
primarily postnominal. The word order variation illustrated in (34) and (36) and the stylistic connotations of the
postnominal order are thus the result of language contact. See also Note 4.
The situation in English is the reverse of that in Walloon, a variety of French spoken in Belgium. Unlike
in standard French, adjective phrases modifying nouns are prenominal in Walloon. This is the result of language
contact with Flemish, a variety of Germanic.

4. Once again, the unusual situation in English is a result of the complicated history of the language (see Note 3).
Originally, English, like the other Germanic languages, allowed only inflectional comparison. Contact with Norman
French introduced the option of analytical comparison. In the modern language, most adjectives are associated with
one or the other of the options. As a rule of thumb, the inflectional option is used for short adjectives and the analytic
option for long ones. But what exactly counts as “short” and “long” is variable across speakers and sometimes even
within the same individual.

5. The word-final hyphen on the superlative forms is intended to indicate that further endings must be added before
they are full-fledged German words.

6. Certain adjectives in Romance do require inflectional comparison. These are irregular adjectives whose comparison
involves suppletion (that is, the stems from which the comparative and/or superlative degrees are formed are
etymologically unrelated to the positive). Some examples of inflectional comparison in Romance are French bon
‘good’ versus mieux ‘better’ and mal ‘bad’ versus pire ‘worse’. Notice that the English counterparts show suppletion
as well. Suppletion is not restricted to adjective comparison; the relation between go and its past tense form went is
also suppletive (went is etymologically related to the wend of to wend one’s way).

7. It is useful to distinguish between form (morphology) and meaning (reference). Ordinarily, forms with past-tense
morphology are used to refer to an event or state prior to the time of speaking. However, it is possible in English
to use past-tense forms to refer to events or states contemporaneous with a reported time of speaking; this is the
so-called sequence-of-tense phenomenon in reported speech, illustrated in (i).

(i) a. He said, “They know a lot”. (direct speech)

b. He said (that) they knew a lot. (reported speech)

As is evident from (ii) and (iii), can, may, shall, and will continue to maintain a productive morphological
relationship with could, might, should, and would, respectively, in sequence-of-tense contexts.

(ii) Direct speech:

a. She said, “That may be so”.
b. He told us, “You won’t last long”.
(iii) Reported speech:
 Chapter 2: Syntactic categories and constituenthood 73

a. She said (that) that might be so.

b. He told us (that) we wouldn’t last long.

Nevertheless, in keeping with the point made in the body of the text, the morphological relationship between
the modals in (ii) and their counterparts in (iii) is purely formal, lacking the referential underpinning evident in
(6).

(iv) a. They know a lot { ✔ now, * then }.

b. They knew a lot { * now, ✔ then }.

We thank Aaron Dinkin for drawing our attention to the sequence-of-tense phenomenon.

8. For reasons not well understood, gerunds of progressive forms, as in (i), are unacceptable.

(i) * I like be-ing learn-ing Spanish.

9. Under certain discourse conditions, English allows the question word to remain in situ (that is, in the place where
it substitutes before it moves), as illustrated in (i). See Chapter 6 for more discussion.

(i) a. Who did you see? (information question or echo question)

b. You saw who? (only echo question)

10. It is worth pointing out that focus-ground partitioning is relevant not just for it clefts, but also for questions and
(short) answers. The focus in a question is the unknown information expressed by the question word. A short answer
to a question consists of a focus and no other material. Repeating the ground of the question yields a full sentence
(that is, includes material apart from the focus).

(i) a. What animals detest the smell of citrus fruits? (question)

b. Ordinary cats. (short answer)
c. Ordinary cats detest the smell of citrus fruits. (full answer)
(ii) a. What do ordinary cats detest? (question)
b. The smell of citrus fruits. (short answer)
c. Ordinary cats detest the smell of citrus fruits. (full answer)

11. Notice that the resultant interpretation is distinct from the one that would result from first combining cat and the
and then combining the cat with the relative clause. The denotation of the cat is a unique member of the set of cats.
Combining the cat with the relative clause would attribute to this unique entity the property of having chased the
rat. Given that the cat already denotes a unique entity, the property of having chased the rat wouldn’t be a defining
property of the cat in question; it would simply be an additional, more or less accidental one. The interpretation in
question is possible semantically, and it can be expressed by using a non-restrictive relative clause, as in (i).

(i) This is the cat, which (by the way) chased the rat.

But (i) is not synonymous with (104), where the rat-chasing property is restrictive, that is, defining.

12. In contrast to the statement in row 2 of Table 2.9, the statement in (i), derived from the statement in row one of Table
2.9 by the modus tollens rule of propositional logic, is true.

(i) If a string isn’t a constituent, then it doesn’t pass the constituenthood tests.

13. For simplicity, we ignore syntax-morphology mismatches of the sort alluded to earlier.
 Chapter 2: Syntactic categories and constituenthood 74

14. For completeness, we should mention that do so substitution and the question test for verb phrases are subject to a
semantic restriction. Specifically, do so and do what cannot substitute for verb phrases with so-called stative verbs
like know or want.

(i) a. They { know her parents, want the cookies }.

b. * They do so.
(ii) a. What do they do?
b. * { Know her parents, want the cookies }.

As their name implies, stative verbs refer to states (rather than to activities or accomplishments), and a reasonably
reliable diagnostic for them is their inability to appear in the progressive construction.

(iii) a. * They are { knowing her parents; wanting the cookies }. (stative verb)
b. ✔ They are { meeting her parents; eating the cookies }. (nonstative verb)

Since do is the prototypical activity verb, it is not surprising that expressions containing it, like do so and do what,
give rise to a semantic clash when they substitute for verb phrases containing stative verbs.

15. It is true that movement of nonfinite verb phrases in out-of-the-blue contexts, as in (i), is not very felicitous:

(i) Write a book, she will .

But it is clearly grammatical given appropriate discourse contexts, as the examples in the text show.

16. The island metaphor is not perfect. Although constituents can’t move out of an island, islands as a whole are able to
move, as shown in (i)–(iii).

(i) ✔ The doctors and lawyers, we should invite .

(ii) a. ✔ Who should we invite ?
b. ✔ The lawyers and the doctors.
(iii) ✔ It is the lawyers and the doctors that we should invite .

Exercises and problems 7

Exercise 2.1
Using the constituenthood tests reviewed in this chapter, determine whether the underlined
strings in the following sentences are (phrasal) constituents. Some of the sentences below have
more than one interpretation. Begin by focusing on the ordinary interpretation, and then con-
sider any more unusual readings.
Chapter 2: Syntactic categories and constituenthood 75

Take Note

Here and throughout the exercises in the book, your answer should include not just your
conclusions, but also the supporting evidence. What we mean by the term evidence is one
or more linguistic expressions (sentences or phrases) together with an indication of their
grammaticality (checkmark or asterisk). Sometimes providing the evidence is all that is
necessary. At other times, a bit of discussion is required to explain the relevance of the
evidence to the question at hand.

(1) a. They put the car in the garage.

b. They put the car in the garage.
c. They put the car in the garage.
(2) a. They know the guy with the fedora.
b. They know the guy with the fedora.
(3) a. They threw the towel in the closet.
b. They threw in the towel.

Exercise 2.2
How well does each of the trees in (1) and (2) represent the syntactic structure of the sentence it
is intended to represent? Here are key questions to consider:
• Are any strings represented as constituents that shouldn’t be?
• Are any strings not represented as constituents that should be?
• Are any of the trees misleading in other respects?

Background Info

Abbreviations for syntactic categories:

Det - determiner (roughly speaking, article or demonstrative pronoun), NounPhr - noun

phrase, PrepP - prepositional phrase, TrVerb - transitive verb, VerbPhr - verb phrase
Chapter 2: Syntactic categories and constituenthood 76

(1) a. Sentence

NounPhr VerbPhr

Pronoun TrVerb NounPhr

they put
Det NounPhr
the
Noun PrepPhr
car
Prep NounPhr
in
Det Noun
the garage
b. Sentence

NounPhr VerbPhr

Pronoun TrVerb NounPhr

they put
Det Noun PrepPhr
the car
Prep NounPhr
in
Det Noun
the garage
c. Sentence

NounPhr VerbPhr

Pronoun TrVerb NounPhr PrepPhr

they put
Det Noun Prep NounPhr
the car in
Det Noun
the garage
Chapter 2: Syntactic categories and constituenthood 77

(2) a. Sentence

NounPhr VerbPhr

Pronoun TrVerb NounPhr

they know
Det NounPhr
the
Noun PrepPhr
guy
Prep NounPhr
with
Det Noun
the fedora
b. Sentence

NounPhr VerbPhr

Pronoun TrVerb NounPhr

they know
Det Noun PrepPhr
the guy
Prep NounPhr
with
Det Noun
the fedora
c. Sentence

NounPhr VerbPhr

Pronoun TrVerb NounPhr PrepPhr

they know
Det Noun Prep NounPhr
the guy with
Det Noun
the fedora
Chapter 2: Syntactic categories and constituenthood 78

Exercise 2.3
Particle verbs in Germanic languages pose a number of interesting syntactic puzzles. This exercise
investigates a limited component of those puzzles. In English, like other Germanic languages,
many constructions have a preposition-like particle following the verb.

(1) a. I messed up the assignment.

b. I climbed up the stairs.
c. I looked after my friend.

As it turns out, there are a number of sub-categories of these particle verbs, some of which
can be identified based on their constituent structure. Specifically looking at the constituency
of the potential prepositional phrase following the verb, note the difference between climb up
and mess up in how they behave when testing for a PP constituent using the it-cleft constituency
diagnostic.

(2) a. It is up the stairs that I climbed.

b. *It is up the assignment that I messed.

This suggests that in an example like (1b) there is a PP constituent, but there is no PP constituent
in an example like (1a).

(3) a. Climb [up the stairs]

b. *Mess [up the assignment]

For the sake of this exercise, we can use the term prepositional verb for a particle verb construc-
tion where the P forms a PP constituent with the following noun phrase, and the term phrasal
verb for a particle verb construction where the P does not form a constituent with the following
noun phrase.

(4) a. Prepositional Verb: [VP V [PP P NP ] ]

b. Phrasal Verb: [VP V P NP ] ]

For this exercise, you will look at the English phrase turn on. This phrase is multiply-ambiguous,
as we illustrate below.

(5) a. Alex turned on College Avenue. (physical turning onto a road)

b. Alex turned on the light. (activation)
c. Alex turned on Jay.
i. sexual arousal – Alex aroused Jay
ii. betrayal – Alex betrayed Jay

Your task is to use constituency diagnostics (or other differences you note) to say whether the
ambiguities here are lexical ambiguities or structural ambiguities. That is, do these interpretations
all have the same syntactic structure and just mean different things (lexical ambiguity)? Or are
there different syntactic structures involved (syntactic ambiguity)? Or is it some of both?
Chapter 2: Syntactic categories and constituenthood 79

If you are a native speaker of English, you can provide the judgments for the sentences you create
on your own. If you are not a native speaker of English, you will need to work with someone
who is, or ask someone who is for judgments on the diagnostic sentences you construct.
We specifically recommend testing for the potential PP constituent, though you can note other
patterns that you find as well.

Exercise 2.4
As we mentioned in this chapter, children’s stories and poetry are often excellent instances of
nonce words (nonsense words) being used creatively for expressive effect. Lewis Carroll’s poem
“Jabberwocky” is another such instance, and gives an opportunity to explore our intuitions
about these invented words.

Jabberwocky, by Lewis Carroll

’Twas brillig, and the slithy toves

Did gyre and gimble in the wabe:
All mimsy were the borogoves,
And the mome raths outgrabe.

“Beware the Jabberwock, my son!

The jaws that bite, the claws that catch!
Beware the Jubjub bird, and shun
The frumious Bandersnatch!”

He took his vorpal sword in hand;

Long time the manxome foe he sought—
So rested he by the Tumtum tree
And stood awhile in thought.

And, as in uffish thought he stood,

The Jabberwock, with eyes of flame,
Came whiffling through the tulgey wood,
And burbled as it came!

One, two! One, two! And through and through

The vorpal blade went snicker-snack!
He left it dead, and with its head
He went galumphing back.

“And hast thou slain the Jabberwock?

Come to my arms, my beamish boy!
Chapter 2: Syntactic categories and constituenthood 80

O frabjous day! Callooh! Callay!”

He chortled in his joy.

’Twas brillig, and the slithy toves

Did gyre and gimble in the wabe:
All mimsy were the borogoves,
And the mome raths outgrabe. (Through the Looking Glass, Lewis Carroll, 1871)

Your task in this exercise is identify the syntactic categories of each of the words bolded in
the poem. For list items with phrases containing multiple words, identify the syntactic category
for each. For each word, explain the diagnostic evidence that supports your conclusion. This
will include evidence from both morphological and syntactic distribution, but can also include
additional intuitions that you have, just be as specific as possible about where you think the
intuition comes from.
There are not specifically correct answers here, in the sense that these words are clearly
made up. But for some of them, native English speakers will have very clear judgments about
what syntactic category they must be (which arise from specific properties of their distribution).
For other words here, there may be multiple different (“right”) answers, in the sense that there
are multiple parses/structures of the sentence that are possible. The point is not to get a right
answer, but to build a strong argument, based on specific evidence.

Problem 2.1
The substitution test discussed in detail in this chapter and the substitution operation introduced
in Chapter 1 are not identical, but they are related. In a few sentences, explain how.

Appendix: Finiteness in English 8

Take Note

A number of points in the constituency discussion in this chapter run into questions about
how finite verbs pattern differently from nonfinite verbs in English. This appendix details
finiteness in English in additional detail.
Chapter 2: Syntactic categories and constituenthood 81

8.1 Verb forms

English verbs can take the various forms listed in Table 2.10.

Name Description Examples

bare Default form in present tense They play together.
sentences. I see.
Also appears in various nonfinite I want to play.
contexts, such as in to infinitive They need to see you.
clauses,
after modals, They may play.
We will see.
and in connection with do support. They don’t play lacrosse.
Do you see?
-s Special form used in the present Lukas runs for miles.
tense to mark agreement with a third The cat enjoys treats.
person singular subject.
-ing As present participle, combines The cat is playing with the yarn.
with auxiliary be to express various I was seeing her until a week ago.
aspectual nuances.
Also occurs on its own as the Playing with landmines is dangerous.
gerund. We always enjoy seeing you.
past tense The morphologically simplest way of The cat played with the yarn.
expressing past tense. We saw a deer.
past participle Combines with auxiliary be to form Baseball is played all over the world.
passive forms. She was last seen off Mozambique.
Combines with auxiliary have to They have never played lacrosse.
form perfect forms. I had never seen it before.

Table 2.10: Verb forms in English

▶ For all verbs, the -ing form is predictable from the bare form, being derived from it by the
affixation of -ing (play-ing, see-ing, hav-ing, be-ing).
▶ The -s form is similarly predictable for most verbs, with major (be, is) or minor (have, has)
exceptions.
▶ The past tense and past participle forms are predictable from the bare form in some cases, but
not in others. With regular verbs, the past tense and past participle forms are homonymous
and are formed by affixing -ed to the bare form. Why bother distinguishing between the
two forms? That is, why not just posit a single past form? The reason is that the past tense
and the past participle are distinct for irregular verbs such as go (went, gone), see (saw, seen),
sing (sang, sung), or write (wrote, written).
Chapter 2: Syntactic categories and constituenthood 82

▶ The past tense form is sometimes called the -ed form (read “ee dee form”) because -ed is the
default expression of the past tense morpheme.
▶ The past participle form is sometimes called the -en form (read “ee enn form”) because -en
is a common expression of the past participle morpheme (though not the default expression,
which, as just mentioned, is homonymous with the default past tense morpheme).
▶ A verb’s bare form, past tense, and past participle (in other words, exactly the forms that
aren’t generally predictable) are known as its principal parts. (1) shows some examples.

(1) a. play, played, played

b. come, came, come
c. stand, stood, stood

8.2 Finiteness
Finiteness of verbs
The verb forms just discussed are classified into two categories: finite and nonfinite. Finite
verbs can function on their own as the core of an independent sentence, whereas nonfinite verbs
cannot. Rather, nonfinite verbs must ordinarily combine with a modal, an auxiliary verb, or the
infinitival particle to.
A verb’s -s form and past tense form are always finite, and the two participles (the -ing and
-en forms) are always nonfinite.

(2) Finite verb:

a. ✔ He gives a performance every week.
b. ✔ He gave a performance every week.
(3) Nonfinite verb:
a. ✔ She is giving a performance here tonight.
b. ✔ She has given a performance here in the past.

To complicate matters a bit, a verb’s bare form can be either finite or nonfinite. Bare forms
that occur on their own are finite and express present tense. Otherwise, bare forms are nonfinite.
Examples are given in (4) and (5).

(4) Finite verb:

a. ✔ We give performances all the time.
(5) Nonfinite verb:
a. ✔ We will give a performance tomorrow.
b. ✔ We promised to give a performance soon.

Table 2.11 summarizes the above discussion.

Chapter 2: Syntactic categories and constituenthood 83

Example Verb form Finite?

play-s, see-s 3sg present tense
play-ed, saw past tense yes
bare (present tense)
play, see
bare (otherwise)
play-ing, see-ing -ing no
play-ed, see-n past participle

Table 2.11: Verb forms and finiteness in English

Finiteness of clauses
Finiteness is a property not only of verbs, but also of clauses. Ordinary clauses—ones that can
stand alone—are finite. All finite clauses contain exactly one finite element per clause. In the
simplest case, this finite element is a finite verb, as illustrated in (6).

(6) They helped us. (finite clause, finite main verb)

It is also possible for a finite clause’s finite element to be a finite auxiliary verb or a modal
(finite by definition). In this case, illustrated in (7), the entire clause is then finite, the auxiliary
verb or modal is finite, but the main verb is nonfinite.

(7) They did help us. (finite clause, finite auxiliary verb, nonfinite main verb.)

Clauses can also be nonfinite. The verb of a nonfinite clause is always nonfinite. (8) il-
lustrates the case that is easiest to recognize—the case where the nonfinite clause (enclosed in
square brackets) contains the particle to.

(8) They agreed [ to help us ]. (nonfinite clause, nonfinite main verb.)

You might be wondering why we classify the subordinate-clause verb in (8)—help—as non-
finite. After all, help is a bare verb form, and so in principle might be finite (recall Table 2.11). The
reason is that the verb neither expresses present tense nor agrees with a third-person singular
subject. We can see this clearly by changing the past tense verb of the main clause to present
tense and the main-clause subject to third-person singular. As shown in (9), help does not change
to help-s, as we would expect if it were finite.

(9) He always agrees [ to { help, *help-s } us ].

The same reasoning extends to cases where the nonfinite clause has a subject of its own.
The examples in (10) and (11) show that neither a subordinate-clause nor a main-clause third-
person subject causes the bare verb to change form.

(10) a. We expect [ her to { help, *help-s } us ].

b. She expects [ him to { help, *help-s } us ].
Chapter 2: Syntactic categories and constituenthood 84

(11) a. We make [ her { help, *help-s } us ].

b. She makes [ him { help, *help-s } us ].

We conclude from this that the bare forms are nonfinite, as are the clauses containing
them.
Table 2.12 summarizes the above discussion. The relevant clauses are uniformly delimited
by square brackets. In the general case, the yes/no values agree across columns; it is only the
second row that contains a mismatch.

Example Is the clause Does the clause Is the main verb

finite? contain a finite finite?
element?
[ They helped us ] . yes yes yes
[ They did help us ] . yes yes no
He always agrees [ to help us ] . no no no
She expects [ him to help us ] . no no no
He makes [ her help us ] . no no no

Table 2.12: Clauses and finiteness in English

 3
Some basic linguistic
relations

Chapter outline
1 Argumenthood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
1.1 Semantic valency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
1.2 Transitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
2 Modification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
3 Predication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
3.1 Evidence for a subject requirement: Expletive it . . . . . . . . . . . . . . . 93
3.2 Aristotelian versus Fregean predicates . . . . . . . . . . . . . . . . . . . . 94
3.3 More evidence for the subject requirement: Expletive there . . . . . . . . 95
3.4 Some special cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
4 Thematic roles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
4.1 Agent, cause, and instrument . . . . . . . . . . . . . . . . . . . . . . . . . 99
4.2 Experiencer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
4.3 Recipient . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
4.4 Location, path, and goal . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
4.5 Measure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
4.6 Theme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
5 Grammatical relations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
5.1 Subjects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
5.2 Indirect objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
5.3 Direct objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
5.4 Distinguishing properties of grammatical relations . . . . . . . . . . . . . 103
6 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
7 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106

In Chapters 1 and 2, we presented various pieces of evidence for the existence of syntactic
structure in human language. The facts presented there raise a simple question—what is the basis
of syntactic structure? In this chapter, we introduce three fundamental linguistic relationships
that underlie syntactic structure, which will lay the foundation for discussions that follow. Two
of these—argumenthood and modification—are at bottom semantic relationships (although the
Chapter 3: Some basic linguistic relations 86

expression of argumenthood is more constrained in natural language than purely semantic con-
siderations would dictate). The third—predication—is arguably purely syntactic. The chapter also
presents and illustrates two further basic linguistic concepts that are related but not identical:
thematic roles and grammatical relations.

Argumenthood 1

1.1 Semantic valency

Background Info

Semantics borrows a lot of formalisms from math and logic: one we will refer to here is
a function. For those of us who are allergic to math, a function relates an input to an
output: you can think of it metaphorically as a machine that takes some input and for
each input, generates some output.

A very rudimentary function, for example, is a function that adds 2 to an input.

Input Relationship Output
1 +2 3
5 +2 7
10 +2 12
20 +2 22
100 +2 102

The most obvious factor that determines how vocabulary items combine in sentences has
to do with their meaning, a point most conveniently illustrated with verbs. From the point of
view of a simple formal semantics, the verb laugh is a function from entities to truth values, as
illustrated in (1). Entities that laugh are associated with the value T(rue); entities that don’t with
the value F(alse). In the possible world described in (1), Beatrice, Alex, Charis, and Xiaoxing
laugh, and Maisha and Langston don’t.

Extra Info
By convention, entities (i.e. individual elements) are indicated by boldface, sets of enti-
ties are enclosed in curly brackets ({ }), and ordered pairs are enclosed in angle brackets
(⟨ ⟩). It is also conventional to indicate denotations (i.e. the formal semantic properties) of
expressions by enclosing the expressions in special double square brackets (⟦ ⟧).
Chapter 3: Some basic linguistic relations 87

⎧
⎪ Beatrice → T, ⎫
⎪
⎪ ⎪
⎪
⎪ Maisha → F, ⎪
⎪
⎪ ⎪
⎪ Langston → F, ⎪
(1) ⟦ laugh ⟧ = ⎨ ⎬
⎪
⎪ Alex → T, ⎪
⎪
⎪
⎪ Charis → T, ⎪
⎪
⎪
⎪ ⎪
⎪
⎩ Xiaoxing → T ⎭
Laugh can combine with a single argument, which denotes an entity. Intuitively, we can
think of arguments as the central participants in a situation/event: a laughing event necessarily
includes a laugher (and doesn’t necessarily include any other participant), so we say that laugh
takes one argument. Combining laugh with an argument (say, Alex) has a syntactic effect and a
corresponding semantic effect. The syntactic effect is to yield the sentence in (2a). (The semantic
contribution of the past tense morpheme -ed is irrelevant for present purposes, and so we will
ignore it in what follows.) The corresponding semantic effect is to apply the function in (1) to the
argument; that is, to select the relevant the argument in the function and to return the associated
value. In the example at hand, the sentence comes out as true.

(2) a. Alex laughed.

b. ⟦ Alex laughed ⟧ = T

But combining Maisha with laughed yields Maisha laughed, which is false in the imaginary (and
simplified) world under consideration.
In addition to denoting simple functions, verbs can also denote recursive functions. For
instance, a transitive verb denotes a function from entities to a second function, the latter of
the same type as just described for the intransitive verb laugh (a function from entities to truth
values). So the transitive verb invite might denote the function in (3).
⎧
⎪ Tre → ( Maisha → T ), ⎫
⎪
⎪ ⎪
⎪
⎪ Langston → ( Maisha → T ), ⎪
⎪
⎪ ⎪
⎪ Ruth → ( Maisha → F ), ⎪
(3) ⟦ invite ⟧ = ⎨
⎪
⎪ Tre → ( Alex → F ), ⎬
⎪
⎪
⎪
⎪ Langston → ( Alex → F ), ⎪
⎪
⎪
⎪ ⎪
⎩ Ruth → ( Alex → T) ⎪ ⎭
Combining invite with a theme argument like Langston—that is, the participant who is
being invited—has the syntactic effect of yielding the phrase in (4a). As before, the corresponding
semantic effect is to select the entity denoted by the argument in (3) and to return the associated
values, as shown in (4b).

(4) a. invited Langston.

{ }
( Maisha → T ),
b. ⟦ invited Langston ⟧ =
( Alex → F)

Further combining invited Langston with an agent argument (say, Maisha) yields the sen-
tence in (5a) and the truth value in (5b). This second step in the derivation of a transitive sentence
is exactly equivalent to the first and only step that is necessary in an intransitive sentence.
Chapter 3: Some basic linguistic relations 88

(5) a. Maisha invited Langston.

b. ⟦ Maisha invited Langston ⟧ = T

It is important to understand that the order of the arguments in (3) reflects derivational
order (the order in which the arguments combine structurally), not their superficial linear or-
der. Given purely semantic considerations, it is equally easy to write functions in which the
derivational order is congruent with the linear order. Exercise 3.1 asks you to do so.
Verbs like laugh and invite are instances of one-place and two-place predicates, respec-
tively. The term predicate here refers to a vocabulary item, with a focus on its capacity to
combine with one or more arguments. The number of arguments that a predicate requires is its
semantic valency.
The relations denoted by predicates can involve more than two arguments. An example
of a three-place predicate is give, which denotes the relation among a set of givers, a set of gifts,
and a set of recipients. Even more complex relations are possible. For instance, rent is a five-
place predicate denoting a relation among property owners, tenants, rental property, amounts of
money, and lengths of time (lease terms).
It will be useful to have terminology for the different kinds of arguments that we will
encounter: we summarize the main kinds of verbal arguments in §4 below.

1.2 Transitivity
In principle, a predicate’s valency might completely determine the syntactic structure that it
appears in. The ungrammaticality of the sentences in (6) would fall out directly from such a
system.

(6) a. * Lukas laughed the train. (one-place predicate; superfluous argument)

b. * Andy invited. (two-place predicate; missing theme argument)

The actual situation, however, is more complex. For instance, eat denotes a relation be-
tween eaters and food. It is therefore a two-place predicate, like invite. However, unlike invite,
eat has both a transitive and an intransitive use, as illustrated in (7).

(7) a. Transitive:
The children have eaten their supper.
b. Intransitive:
The children have eaten.

Notice that the semantic properties of eat remain constant in (7). In other words, (7a) and (7b)
are both interpreted as involving the ingestion of food, even though there is no explicit mention
of food in (7b). In view of the mismatch between the semantic and syntactic properties of eat
in sentences like (7b), it is useful to distinguish between semantic and syntactic arguments. As
mentioned earlier, we can think of semantic arguments as central participants in a situation. Syn-
tactic arguments, on the other hand, are constituents that appear in particular syntactic positions
Chapter 3: Some basic linguistic relations 89

(see Chapter 4 for further discussion). Semantic arguments are typically expressed as syntactic
arguments, but the correspondence between the two is not perfect, as (7b) shows.
We will use the term transitivity to refer to the number of syntactic arguments that a verb
combines with, and we can then divide verbs into three subcategories as in Table 3.1.

Word of Caution
We are using the term ‘transitivity’ in a slightly unorthodox way. Traditionally, the term
refers to the number of a verb’s objects, which is one less than the number of its argu-
ments. Thus, as the terms imply, an intransitive takes no objects, and a ditransitive takes
two.

Degree of transitivity Number of syntactic arguments

Intransitive 1
Transitive 2
Ditransitive 3

Table 3.1: Transitivity is the number of syntactic arguments a verb takes

Because of mismatches as in (7), it turns out to be quite rare for verbs to belong to just one
syntactic subcategory. (8)–(10) shows some two-place verbs besides eat that can be used either
transitively or intransitively. The slashes separate the arguments from the predicate and each
other.

(8) a. He / interrupted / the meeting. (basically transitive)

b. He / interrupted. (intransitive use)
(9) a. Amy / knits / sweaters. (basically transitive)
b. Amy / knits. (intransitive use)
(10) a. They / are reading / a book. (basically transitive)
b. They / are reading. (intransitive use)

Conversely, certain one-place verbs can be used not only intransitively, but also transi-
tively, as illustrated in (11)–(13). Notice that the verb and its object in the transitive examples
are etymologically related, or cognate. For this reason, the transitive use of one-place verbs as in
(11)–(13) is known as the cognate object construction.

(11) a. Charlie / died. (basically intransitive)

b. Charlie / died / a peaceful death. (transitive use)
(12) a. Maisha / laughed. (basically intransitive)
b. Maisha / laughed / an infectious laugh. (transitive use)
(13) a. Mona Lisa / was smiling. (basically intransitive)
b. Mona Lisa / was smiling / a mysterious smile. (transitive use)
Chapter 3: Some basic linguistic relations 90

Further, it is possible to use some basically three-place verbs not just ditransitively, but
transitively and even intransitively.

(14) a. We / teach / college students / syntax. (basically ditransitive)

b. We / teach / college students. (transitive use)
c. We / teach / syntax. (transitive use)
d. We / teach. (intransitive use)
(15) a. He / told / me / the whole story. (basically ditransitive)
b. He / told / me. (transitive use)
c. He / told / the whole story. (transitive use)
d. He / better not tell. (intransitive use)

Finally, it is possible to use basically two-place verbs ditransitively.

(16) a. I / baked / a delicious cake. (basically transitive)

b. I / baked / my friends / a delicious cake. (ditransitive use)
(17) a. She / sang / a lullaby. (basically transitive)
b. She / sang / her baby / a lullaby. (ditransitive use)

Many transitive predicates in Mainstream U.S. English allow for the addition of a recipient of the
event, as in (16) and (17), which is always placed between the verb and the theme object. See §4
for more discussion of what kinds of thematic roles are relevant for syntax.

Modification 2
Events are associated with more or less central participants and properties. The central partici-
pants are the semantic arguments just discussed. Properties of a situation typically taken to be
less central, such as manner, place (location, origin, destination, path), time (point in time, dura-
tion, frequency), reason (cause, purpose), and so on, can be expressed by modifiers. Semantic
arguments are closely associated with specific predicates; modifiers, less so. For examples, not
all situations are associated with themes, so themes are arguments. But situations are all located
in time, so temporal expressions are modifiers.
Chapter 3: Some basic linguistic relations 91

Take Note

Arguments and modifiers both introduce restrictions on the denotation of a predicate,

and the relationships of argumenthood and modification do not differ in this respect. For
instance, the situations denoted by invite Langston (where Langston is an argument) are
a subset of those denoted by invite, just as the situations denoted by laugh uproariously
(where uproariously is a modifier) are a subset of those denoted by laugh. This means
that the distinction between arguments and modifiers is not a semantic distinction, but a
syntactic one.

Modifiers of verbs are typically adverb phrases or prepositional phrases, but noun phrases
can serve as verbal modifiers as well (Exercise 3.3 asks you to find examples). In the following
examples, the modifier is in italics, and the verb that it modifies is underlined.

(18) a. Manner:
He read the letter carefully.
b. Location:
We met the students in my office.
c. Origin:
We departed from Bangalore.
d. Destination:
We arrived in Benares.
e. Path:
We followed along the path.
f. Point in time:
They discussed the proposal in the afternoon.
g. Duration:
She kept their books for five years.
h. Frequency:
They read the Times quite often.
i. Cause:
He threw it away out of spite.
j. Purpose:
You should send a message (in order) to warn everyone.

Because of their semantically peripheral character, modifiers are syntactically optional. The
converse is not true, however. Not all syntactically optional constituents are modifiers; recall from
(7b) that semantic arguments aren’t always expressed.
Verbs are not the only category that can be modified. For instance, nouns are often modified
by adjective phrases, prepositional phrases, and relative clauses, and contrary to what the name
might suggest, even adverb phrases can modify nouns.
Chapter 3: Some basic linguistic relations 92

(19) a. a very important period

b. a period of great import
c. the car that just turned the corner
d. the instructions below

Moreover, adjective phrases, adverb phrases, and preposition phrases, which typically
serve as modifiers (though they are not restricted to that function), can themselves contain mod-
ifiers.

(20) a. very proud of the children

b. surprisingly good to eat
(21) a. almost there
b. right below
(22) a. very much in the dark
b. two feet under the shed

There are cases where it is difficult to tell which constituent is the modifier and which
the modifiee. For instance, in an expression like early in the morning, is the adverb modifying
the preposition, as indicated in (23a), or is the prepositional phrase modifying the adverb, as in
(23b), or is either of these possibilities grammatical (that is, is the expression structurally ambigu-
ous)?

(23) a. early in the morning

b. early in the morning

Predication 3
The two linguistic relations discussed so far—argumenthood and modification—are basically se-
mantic notions that are optionally expressed in the syntax. In this section, we introduce a third
relation, predication, which differs from argumenthood and modification in being an irreducibly
syntactic relation. By this, we mean that some cases of predication are not semantically moti-
vated.
As a very basic introduction, predication is essentially the attribution of a property to some
individual/element (the ‘subject’). The predicates are bracketed in (24), where ‘dog-ness’ is at-
tributed to Lilly in (24a) and happiness is attributed to Lilly in (24b).

(24) a. Lilly is [ a dog ].

b. Lilly is [ happy ].
Chapter 3: Some basic linguistic relations 93

3.1 Evidence for a subject requirement: Expletive it

In order to establish that predication cannot be reduced to semantic considerations, we begin by
distinguishing between two kinds of it: ordinary referential it and expletive it. As the name
implies, referential it has some referent (for more on referents and reference, see §1 of Chapter 15).
As a result, it can be replaced by a more complex description of that referent, as in (25).

(25) a. It bit the zebu.

b. The tsetse fly bit the zebu.

Referential it can also receive stress (indicated by underlining) in answers to questions, as shown
in (26).

(26) a. What bit the zebu?

b. (pointing) It did.

By contrast, expletive it doesn’t refer, so it can’t be replaced by a description, as shown in

(27).

(27) a. It seems that the manuscript has been found.

b. * The fact seems that the manuscript has been found.

This lack of content of the expletive is also evident in the lack of a content question that
replaces the expletive with what:

(28) *W
hat seems that the manuscript has been found?

Expletive it is unable to receive stress. In fact, as (29) shows, a situation designed to elicit an
answer analogous to (26b) is ungrammatical. So if (29a) is uttered based on some evidence, it’s
not possible to utter (29b) in response in order to invoke the idea that some different sort of
evidence is relevant.

(29) a. It seems that the manuscript has been found.

b. * (pointing) No, IT does.

The point is that expletive it in examples like (27a) is a different sort of element from referential
it in examples like (25a).
Returning now to the main thread of the discussion, we note that the italicized that clause
in (30a) functions as the sole syntactic argument of the adjective evident, on a par with the noun
phrase in (30b). (For simplicity, we ignore the copula as semantically vacuous.)

(30) a. That they are corrupt is evident.

b. Their corruption is evident.

An indication of the semantic equivalence of the two expressions is the fact that they can both
serve as a short answer to the question in (31a).
Chapter 3: Some basic linguistic relations 94

(31) a. What is evident?

b. That they are corrupt.
Their corruption.

(32) shows a syntactic variant of (30a) where the that clause appears at the end of the entire
sentence. The original preverbal position of the that clause is occupied by expletive it.

(32) It is evident that they are corrupt.

Given that the that clause satisfies the semantic requirement of evident for an argument in both
(30a) and (32), the presence of the expletive pronoun in (32) is apparently superfluous. From a
semantic point of view, one might therefore expect it to be optional. But this is not the case, as
the ungrammaticality of (33) shows.

(33) * Is evident that they are corrupt.

The ungrammaticality of (33) leads us to conclude that there exists a purely syntactic well-
formedness condition requiring all clauses to have a subject.
Earlier, we saw that it is possible for arguments to be semantically motivated and yet not
appear overtly in the syntax. Expletive subjects represent roughly the converse of this situation,
being cases where an expression that is not motivated by semantic considerations is nevertheless
obligatory in the syntax.

3.2 Aristotelian versus Fregean predicates

We will refer to the requirement just mentioned as the subject requirement. According to
it, every clause consists of a subject and a predicate (independently of semantic requirements).
The term ‘predicate’ as used here has a different sense than in our earlier discussion concerning
argumenthood; it refers to what remains of a clause when its subject is removed.
For clarity, we can use the term ‘Aristotelian predicate’ for this sense, since the obser-
vation that all sentences consist of a subject and a predicate goes back (at least) to Aristotle
(384–322 BCE). Predication is the relation between a subject and an Aristotelian predicate. So
(34a) and (34b) are two alternative ways of stating the subject requirement.

(34) a. Every clause has a subject.

b. Every clause is an instance of predication.

The sense of ‘predicate’ that we used earlier, to refer to a single vocabulary item, is much
more recent and can be attributed to one of the founders of modern logic, the mathematician and
philosopher Gottlob Frege (1848–1925). Accordingly, we can use the term ‘Fregean predicate’ for
this sense. What Frege recognized is that Aristotle’s division of a clause into subject and predicate
is simply the first of a potential series of such bifurcations. Just as it is possible to peel off, as it
were, the subject of a clause, leaving the Aristotelian predicate, it is possible to further peel off
any arguments (and modifiers) contained within the Aristotelian predicate, yielding in the final
instance a single vocabulary item, the Fregean predicate. So in a sentence like the children chased
Chapter 3: Some basic linguistic relations 95

the dogs with water balloons, the Aristotelian predicate is chased the dogs with water balloons,
whereas the Fregean predicate is chase.
Fruitful as Frege’s analytic insight is, we should not let it obscure the key difference be-
tween subjects and other constituents of a clause: namely, that constituents contained within
an Aristotelian predicate must be licensed by semantic considerations (in other words, these
constituents are present because of semantic considerations), whereas the subject, which is ex-
ternal to the Aristotelian predicate and combines with it, is required independently of semantic
considerations.
In (35), the Aristotelian predicate of the largest clause is in italics, and its Fregean predicate
is underlined. As the increasingly complex sentences show, Aristotelian predicates are recursive
categories. Fregean predicates, on the other hand, not being phrases, are not.

(35) a. The tabby cat enjoys catnip immensely.

b. They have found that the tabby cat enjoys catnip immensely.
c. My downstairs neighbor suspects that they have found that the tabby cat enjoys catnip
immensely.

3.3 More evidence for the subject requirement: Expletive there

In English, further evidence for the purely syntactic character of the subject requirement comes
from the expletive there construction. As with it, English there comes in two variants: locative
and expletive. The pro-form there has inherently locative meaning in (36); this is evident in the
fact that the adjuncts (right) here and (over) there render (36a) and (36b) self-contradictory and
redundant, respectively.

(36) a. # There comes the train (right) here.

b. ✔ There comes the train (over) there.

It is possible to stress locative there, regardless of its position in a sentence.

Expletive there has no locative meaning, and so both sentences in (37) are completely ac-
ceptable. In contrast to locative there, expletive there can never bear stress.

(37) a. ✔ There is a clean shirt (right) here.

b. ✔ There is a clean shirt (over) there.

Expletive there behaves like an ordinary subject in ways that clause-initial locative there doesn’t.
For instance, it inverts with the verb in direct questions, and it is grammatical in so-called small
clauses (in this instance, selected by despite: see §3.4 for more on small clauses).

(38) a. ✔ Is there a problem?

b. * Is (over) there our train?
(39) a. ✔ despite [there being a problem]
b. * despite (over) [there being our train]
Chapter 3: Some basic linguistic relations 96

(40) illustrates an ordinary sentence and its expletive there counterpart.

(40) a. Several vexing questions remain.

b. There remain several vexing questions.

Just as expletive it occupies the position that would otherwise be occupied by a clausal subject,
expletive there occupies the position that would otherwise be occupied by a noun phrase subject.
And just as with expletive it, omitting expletive there results in ungrammaticality.

(41) * Remain several vexing questions.

It should be pointed out that not every English sentence has an expletive there counter-
part. Rather, expletive there is subject to a licensing condition (a necessary condition for its
occurrence) that can be stated roughly as in (42).

(42) Licensing condition on expetive there:

For expletive there to be grammatical as the subject of a clause, the (Fregean) predicate of
that same clause must be a verb of existence or coming into existence.

In the following examples, the (Fregean) predicate licensing the expletive there is in blue bold-
face.

(43) a. After their military defeat, there arose among the Plains tribes a powerful spiritual
movement.
b. There is a problem.
c. There began a reign of terror.
d. In the end, there emerged a new caudillo.
e. There ensued a period of unrest and lawlessness.
f. There exists an antidote.
g. There has occurred an unfortunate incident.
h. There remains a single course of action.

Predicates that aren’t verbs of (coming into) existence don’t license expletive there. This
is the reason for the ungrammaticality of the following examples; the non-licensing (Fregean)
predicates are underlined in red.

(44) a. * There came more than sixty dignitaries.

b. * There continued the same problem.
c. * There rang the mail carrier.
d. * There sang an impressive choir from Russia.
e. * There strutted a poodle into the room.
Chapter 3: Some basic linguistic relations 97

3.4 Some special cases

Nonfinite clauses
The instances of predication provided so far have all been finite clauses like those in (45) (i.e.
clauses that bear tense).

(45) a. He laughed uproariously.

b. It will seem that they won the game.
c. There is a problem.

Nonfinite clauses like those in (46) are also instances of predication; the clauses at issue are set
off by brackets.

(46) a. We expected [ him to laugh uproariously ].

b. We expected [ it to seem that they won the game ].
c. We expected [ there to be a problem ].

At first glance, it might seem preferable to treat the italicized noun phrases in (47) as objects
of expected, rather than as subjects of the embedded nonfinite clause the way we have done.
However, such an approach faces at least two difficulties. First, the relation between the italicized
and underlined constituents in the all of nonfinite embedded clauses in (46) is the same as the
relation between the undoubted subjects and predicates of the finite clauses in (45) and their
embedded counterparts in (47).

(47) a. We expected that [ he would laugh uproariously ].

b. We expected that [ it would seem that they won the game ].
c. We expected that [ there would be a problem ].

Second, in (46a), the thematic relation of agent that the noun phrase him bears to laugh is exactly
the same as that between the subject he and laughed in (45a) and (47a). If him were the object of
expected rather than the subject of the nonfinite clause, that fact would not be captured directly.
Moreover, we would be forced to admit the otherwise unprecedented pairing of the grammatical
relation of direct object with the thematic role of agent (see §4 and §5).

Small clauses
Because of the parallel between nonfinite and finite embedded clauses illustrated in (46) and (47),
it makes sense to assign to in to-infinitive clauses to the same category as modals (see Chapter 4,
§2.1). There also exist instances of predication without any instantiation of I whatsoever. These
are called small clauses (the idea behind the name is that the absence of I makes them smaller
than an ordinary clause).1 (48)–(51) provide some examples; the captions indicate the syntactic
category of the small clause’s (Aristotelian) predicate, which is underlined.

(48) Adjective phrase:

a. We consider [ the proposed solution completely inadequate ].
Chapter 3: Some basic linguistic relations 98

b. They proved [ the theory false ].

(49) Noun phrase:
a. They called [ the actor a traitor ].
b. I consider [ Mark Judy’s closest collaborator ].
(50) Prepositional phrase:
a. They made [ him into a star ].
b. I want [ everyone off the boat ].
(51) Verb phrase:
a. God let [ there be light ]. (bare verb)
b. I hear [ the cat scratching at the door ]. (gerund)

Small clauses are typically arguments of verbs, but they can also be arguments of (certain)
prepositions—notably with—as illustrated in (52).

(52) a. Adjective phrase:

With the weather much less turbulent, flights were able to resume for the first time in
days.
b. Noun phrase:
With her husband an airline industry lobbyist, the senator’s support for the bailout
was hardly surprising.
c. Prepositional phrase:
With all three of their kids in college, their budget is pretty tight.
d. Verb phrase (gerund):
With the parade passing right in front of her house, Jenny could not have asked for a
better view of it.

Imperatives
Imperative sentences like (53) appear to lack a subject.

(53) Come over here.

There is reason to believe, however, that they contain a second-person subject comparable to the
pronoun you except that it is silent (the “you understood” of traditional grammar). For one thing,
(53) has a variant in (54) where the subject is explicitly expressed.

(54) You come over here.

Another reason to assume that all imperatives contain a silent, yet syntactically active subject is
that the grammaticality pattern in (55), where the subject is overt, has an exact counterpart in
(56).

(55) a. You shave { yourself, yourselves }.

Chapter 3: Some basic linguistic relations 99

b. * You shave you.

c. * You shave themselves.
(56) a. Shave { yourself, yourselves }.
b. * Shave you.
c. * Shave themselves.

Thematic roles 4
This section provides an elementary overview of concepts and terminology related to thematic
roles.

4.1 Agent, cause, and instrument

Agents are arguments that bring about a state of affairs.2 The line between agents, on the one
hand, and causes or instruments, on the other, can be fuzzy, but agents are (or are perceived to
be) conscious or sentient, in a way that causes or instruments aren’t. Some examples are given
in (57)–(59).

(57) Agent:
a. The lions devoured the wildebeest.
b. The boys caught some fish.
c. My mother wrote me a letter.
(58) Cause:
a. Hurricane-force winds demolished much of the town.
b. An epidemic killed off all of the tomatoes.
c. An economic downturn put thousands of workers out of work.
(59) Instrument:
a. This key opens the door to the main office.
b. They must have used indelible ink.
Chapter 3: Some basic linguistic relations 100

4.2 Experiencer
Experiencers are arguments that undergo a sensory, cognitive, or emotional experience.

(60) Experiencer:
a. The rhesus monkey had never seen snow before.
b. Many people fear snakes.
c. Their resourcefulness struck her as admirable.

4.3 Recipient
Recipients are arguments that receive something (whether good or bad) in a situation.

(61) Recipient:
a. They gave the workers a raise.
b. I paid my landlord the rent.
c. He spared me his usual sob story.

4.4 Location, path, and goal

Locations are points (or regions) in space.

(62) Location:
a. We always eat breakfast in the kitchen.
b. The cork has been bobbing under the bridge for an hour.

Paths connect locations.

(63) Path:
a. Lucky raced down the driveway.
b. The boat passed under the bridge so quickly I missed seeing it.
c. We drove the scenic route.

When locations serve as endpoints of paths, we generally refer to them as goals.

(64) Goal:
a. We traveled to Paris quite a bit in those days.
b. Lucky raced to the edge of the woods.
c. I’d like to send this package to France.

Recipients can also serve as the endpoint of paths, and the distinction between goals and
recipients can be difficult. (The issue is addressed in detail in §3 of Chapter 11.)

(65) I’d like to send this package to my cousin.

Chapter 3: Some basic linguistic relations 101

Locations and paths can also be spatial or temporal.

(66) a. Point in time:

Let’s start the meeting at two.
b. Goal in time:
The meeting will last until two.

4.5 Measure
Measure (a.k.a. amount) arguments express extensions along some dimension (length, duration,
cost, and so on).

(67) Measure:
a. They rowed for three days.
b. The book costs ten dollars.

4.6 Theme
Finally, the thematic role of theme is something of a catch-all. According to one definition,
‘theme’ refers to an argument undergoing motion of some sort, including motion in a metaphori-
cal sense, such as a change of state. As is usual in the syntactic literature, we will also use the term
for arguments that are most ‘affected’ in a situation or for the content of an experience.

(68) Theme:
a. The lions devoured the wildebeest.
b. This key opens the front door.
c. Hurricane-force winds demolished much of the town.
d. They gave the workers a raise.
e. I’d like to send this package to France.
f. Many people fear snakes.

Grammatical relations 5
Grammatical relations must be carefully distinguished from thematic roles. In what follows, we
illustrate three basic grammatical relations: subject, indirect object, and direct object. The appar-
ent contradiction between the existence of sentences with two objects and the binary-branching
hypothesis of phrase structure is addressed in Chapter 11.
Chapter 3: Some basic linguistic relations 102

5.1 Subjects
Subjects are ordinarily the only argument to precede the predicate in English. As the examples
in (69) illustrate, subjects can express many different thematic roles.

Word of Caution
Take care not to confuse the grammatical relation of subject with the thematic role of
agent. The existence of passive sentences is a clear indication that the two notions are not
synonymous (compare (69a) with (69h) and also (69i)).

(69) a. The lions devoured the wildebeest. (agent)

b. Hurricane-force winds demolished much of the town. (cause)
c. This key opens the door to the main office. (instrument)
d. The children liked all their Christmas presents. (experiencer)
e. The workers got a raise. (recipient)
f. An unpaved road led up to the shanty. (path)
g. The summit wasn’t attained until years later. (goal)
h. The wildebeest was devoured by the lions. (theme)
i. The ball rolled down the hill. (theme)

5.2 Indirect objects

Indirect objects occur in conjunction with direct objects, as illustrated in (70).

(70) The gave the workers a raise. (recipient)

They are thematically very restricted—namely, to recipients. Nevertheless, there is no one-to-

one correspondence between the grammatical relation of indirect object and the thematic role
of recipient. This is because recipients needn’t be indirect objects, but can also be expressed as
subjects, as in (69e).

5.3 Direct objects

Direct objects include a verb’s remaining noun phrase arguments. Thematically, they are almost
as flexible as subjects, as shown in (71).

(71) a. You should use this key for the door to the main office. (instrument)
b. The children’s drawings pleased their parents to no end. (experiencer)
c. We drove the scenic route. (path)
d. We reached our hotel after a short subway ride. (goal)
e. The performance lasted two hours. (measure)
f. The lions devoured the wildebeest. (theme)
g. We rolled the ball down the hill. (theme)
h. They gave the workers a raise. (theme)
Chapter 3: Some basic linguistic relations 103

5.4 Distinguishing properties of grammatical relations

Linear order
In many languages, word order correlates to some degree with grammatical relations, in that
a language will typically place subjects and objects in some typical position. In most of the
world’s languages, subjects precede their predicates, regardless of whether a language is verb-
initial (SVO) or verb-final (SOV). The ordinary position for direct objects is adjacent to the verb—
immediately following it in verb-initial languages (V-DO) and immediately preceding it in verb-
final ones (DO-V). However, indirect objects ordinarily intervene between the verb and the direct
object in verb-initial languages, yielding V-IO-DO order. In verb-final languages, the correspond-
ing order often is not, as might be expected, the mirror image order. Rather, the indirect object
continues to precedes the direct object, yielding the order IO-DO-V.
These are of course drastic oversimplifications: there are exceptions to these generaliza-
tions, and there is a lot of nuance within languages affecting canonical word orders (as reported
above) vs. non-canonical word orders that are often used for various kinds of discourse functions.
The point, of course, is that in many languages word order has a relatively direct correlation with
grammatical relations.

Case
However, in many languages, word order is not a diagnostic property of grammatical roles. In
many such languages, it is instead a property called case that is a more consistent identifier of
which noun phrases represent which grammatical relation in the sentence. We discuss this in
much more depth in Chapter 7, but we will introduce it briefly here as is relevant for our current
concerns.
Basic aspects of case are already familiar to English speakers, which has case-marking of
nominals in its pronominal system. In English, pronominal subjects of a sentence appear in a
distinct form from pronouns that appear elsewhere.

(72) a. They like her.

b. She likes them.

(73) a. *Them like she.

b. *Her likes they.

We use the term nominative to refer to the case form that sentence subjects appear in,
and accusative to refer to the form that sentence objects appear in. So the nominative forms of
the pronouns above are they and she, and the accusative forms are them and her.3
There are many languages, however, where case-marking is more thorough, appearing
on most or all nominals in the language, and differentiating much more than subjects vs. non-
subjects. The Turkish example in (74) distinguishes the morphological form of direct objects
(accusative case) from indirect objects (dative case)
Chapter 3: Some basic linguistic relations 104

(74) Mehmet-∅ adam-a elma-lar-ı ver-di

Mehmet-nom man-dat apple-pl-acc give-pst.3sg
‘Mehmet gave the apples to the man.’ (Blake 2004, p. 1)

Turkish marks nominals with six distinct cases based on their grammatical function in a sentence.
These forms (and their functions) are described by the table in (75).

(75) Case-marking in Turkish (Blake 2004, p. 2)

Case Grammatical form of the word
function adam- ‘man’
nominative subject adam
accusative direct object adamı
genitive possessor adamın
dative indirect object adama
locative location adamda
ablative source location adamdan

There are two main (distinct) systems of case-marking that appear across human languages:
what is known as nominative-accusative case alignment (as illustrated above), and ergative-
absolutive alignment, which slices up the case-marking pie in a different sort of way (marking
agentive subjects differently from non-agentive subjects, for example). Case-marking in human
languages is extraordinarily complex cross-linguistically, so this is only a rudimentary introduc-
tion: the point is that case-marking is one of the ways that grammatical relations are morphosyn-
tactically represented in language.

Verbal inflectional morphology

Another way that morphology encodes grammatical relations in syntax is via agreement and
other kinds of inflectional morphology, where the morphological form of one element changes
based on the properties of some other element. The version of this that most language learners
are relatively familiar with is verbal agreement (i.e. verb conjugation). Inflection on Swahili verbs
appears in the template in (76).

(76) Subject.Agreement - Tense - Object.Agreement - Verb

So in an example like (77a), the plural subject agreement agrees with wazazi ‘parents,’ and the
singular object agreement agrees with mtoto ‘child.’ In Swahili it’s possible to leave the subject
and object unpronounced as in (77b) so that just the agreement morphemes themselves specify
the relevant grammatical relations of subject and object.

(77) a. Wazazi wa- li- mu- ona mtoto.

parents subj.agR.pl- pst- obj.agR.sg- see child
‘The parents saw the child.’
b. wa- li- mu- ona
subj.agR.pl- pst- obj.agR.sg- see
‘They (pl) (parents) saw her/him/them(sg) (child)
 Chapter 3: Some basic linguistic relations 105

c. *mu- li- wa- ona

obj.agR.sg- pst- subj.agR.pl- see

Example (77c) attempts to switch the position of the subject and object agreement, but this is
ungrammatical. As shown in (78), it is possible to switch the singular and plural agreements
(subject agreement for the singular form has a different form), but this switches the grammatical
functions: now the child is the one seeing the parents.

(78) a- li- wa- ona

subj.agR.sg- pst- obj.agR.pl- see
‘She/he/they(sg) (child) saw them (pl) (parents).’

So we see that languages may encode grammatical relations in multiple ways: word or-
der, nominal inflection (case), and verbal inflection (agreement). And many languages use more
than one of these strategies, at times marking the same grammatical relation repeatedly, such as
marking subjecthood with word order, nominative case-marking, and subject agreement on the
verb. Each of these strategies is itself a large area of inquiry, but for our purposes we are simply
using these to understand grammatical relations.

Notes 6
1. Small clauses are exceptional in another regard: they are a further instance of constituents where the constituenthood
tests of Chapter 2 yield false negative results (at least for most speakers).

(i) a. What do you like?

b. ⁇ My coffee hot.

2. Here and in the rest of the section, our usage is a bit sloppy. Specifically, we ignore the distinction between linguistic
expressions and the discourse entities that they refer to (for some discussion, see §1 of Chapter 15). For instance, in
(57a), the agent of the devouring is a group of live animals, not the two words referring to them). So a more precise
formulation would be something like “We use the term ‘agent’ to refer to linguistic expressions that (in turn) refer to
entities that bring about a state of affairs.” This mild imprecision is hopefully tolerable in service of a more natural
exposition.

3. In fact, those forms are probably default forms rather than specifically accusative forms, since they are used much
more broadly than just accusative contexts.
Chapter 3: Some basic linguistic relations 106

Exercises and problems 7

Exercise 3.1
A. Imagine a language Hsilgne that is exactly like English except that transitive predicates
combine first with the agent, and then with the theme. Does (1) mean the same thing in
Hsilgne as it does in English? Explain, using the discussion in connection with (3)–(5) in
the text as a model.

(1) Alex invited Langston.

B. Are you sure that you speak English and not Hsilgne? What’s the evidence for your con-
clusion?

Exercise 3.2
A. In your own words, discuss the difference between the terms ‘modify’ and ‘refer’. (See
Chapter 15 for discussion of the term ‘refer’.) Feel free to use illustrative examples, but be
as concise as you can.
B. In your own words, discuss the difference between ‘modify’ and ‘predicate’. Feel free to
use illustrative examples, but be as concise as you can.
C. This part of the exercise focuses on a fundamental distinction in linguistics—that of form
vs. function. In everyday usage, the term ‘modifier’, which refers to any expression that
adds information about some other expression, is sometimes used interchangeably with
the term ‘adjective’. By contrast, linguists distinguish carefully between the two terms and
define adjectives on the basis of morphosyntactic form rather than on the basis of function
(see Chapter 2, § 1.3 for details). Assuming the linguistic definition of ‘adjective’, give an
example of your own of 1) an adjective (or adjective phrase) that is not a modifier, and 2) a
modifier that is not an adjective.
D. This part of the exercise is intended to help you get a better feel for the concept of ‘small
clause’, which many students find difficult. Make up some example sentences that you
think might contain small clauses, but that you’re not sure about. Clearly indicate the
string that you think is the small clause.
Chapter 3: Some basic linguistic relations 107

Exercise 3.3
A. This part of the exercise focuses on a fundamental distinction in linguistics—that of form vs.
function. The function of modification is typically expressed by constituents of a certain
form—namely, adverb phrases or prepositional phrases. But modifiers can also take the
form of noun phrases, as in (1).

(1) Manner: They solved the problem another way.

Provide examples of noun phrase modifiers for the other types of modification illustrated
in the chapter in (18), repeated for convenience as (2). (This is not necessarily possible for
every type.)

(2) a. Manner:
He read the letter carefully.
b. Location:
We met the students in my office.
c. Origin:
We departed from Bangalore.
d. Destination:
We arrived in Benares.
e. Path:
We followed along the path.
f. Point in time:
They discussed the proposal in the afternoon.
g. Duration:
She kept their books for five years.
h. Frequency:
They read the Times quite often.
i. Cause:
He threw it away out of spite.
j. Purpose:
You should send a message (in order) to warn everyone.

B. Here are some questions concerning the relationship between thematic roles and gram-
matical functions.
• Is there a one-to-one relations between thematic roles and grammatical relations?
• Are there any thematic roles that must be expressed by a particular grammatical re-
lation? For instance, are agents invariably expressed as subjects?
• Conversely, are there grammatical relations that are restricted to expressing a partic-
ular thematic role? For instance, do direct objects always express themes?
Chapter 3: Some basic linguistic relations 108

• Are there typical or common mappings between thematic roles and grammatical re-
lations?
• Can you imagine a prepositional phrase expressing the thematic role of theme (not
necessarily in English)?

Exercise 3.4
Given the licensing condition on expletive there, repeated in (1), determine whether each of the
grammaticality judgments in (2) is expected. The brackets indicate clause boundaries (of to in-
finitive clauses or small clauses).

(1) For expletive there to be grammatical as the subject of a clause, the (Fregean) predicate of
that same clause must be a verb of existence or coming into existence.
(2) a. ✔ Feynman suspected [ there to be a problem with the O-ring ].
b. * Feynman suspected [ there a problem with the O-ring ].
c. ✔ There was suspected [ to be a problem with the O-ring ].
d. * There was suspected [ a problem with the O-ring ].

Exercise 3.5
Using concepts introduced in this chapter, discuss the syntactic difference(s) between the two
sentences in (1).

(1) a. Winston considered the judges careful.

b. Winston considered the judges carefully.

(The sentences come from this source, but the discussion there of (1a) is confusing. Thanks to
Rosemary George for tracking down the archived link.)

Exercise 3.6
A. The following headlines each have an intended reading and (possibly more than one) un-
intended humorous reading. Using the concepts introduced in this chapter, explain the
differences between the intended and unintended readings.

(1) a. Greeks Fine Hookers

b. Lawmen from Mexico Barbecue Guests
c. Lawyers Give Poor Free Legal Advice
d. Lung Cancer in Women Mushrooms
(2) Q: What did the Zen master say to the guy at the hot dog stand?
A: Make me one with everything.
(3) The comedian and civil rights activist Dick Gregory tells of walking up to a lunch
counter in Mississippi during the days of racial segregation. The waitress said to
him, “We don’t serve colored people”. “That’s fine,” he replied, “I don’t eat colored
people. I’d like a piece of chicken.”
Chapter 3: Some basic linguistic relations 109

B. Recall Exercise 1.3. (Re)formulate (and possibly improve/generalize) your answer to Part B
in terms of the concepts introduced in this chapter.

Problem 3.1
Nonfinite clauses like the bracketed sequence in (1) are prima facie counterexamples to the subject
requirement.

(1) I promised [ to come on time ].

Provide as much evidence as you can for the existence of a silent subject in (1) and nonfinite
clauses like it.

Problem 3.2
The acceptability of (1) in vernacular usage might tempt one to conclude that the subject require-
ment is not absolute in English.

(1) Seems like they’re finally getting somewhere.

Provide as much evidence as you can bearing on this conclusion.

Problem 3.3
The sentences in (1) apparently violate the licensing condition on expletive there, yet they are
still acceptable (though quite formal). Explain.

(1) a. At the end of the intermission, there sounded a silvery bell.

b. Then the curtain rose, and there waltzed onto the stage a strangely, but exquisitely
dressed apparition.
Introducing the X′
4
schema of phrase
structure

Chapter outline
1 The X′ schema for elementary trees . . . . . . . . . . . . . . . . . . . . . . . . . . 111
1.1 Transitive elementary trees . . . . . . . . . . . . . . . . . . . . . . . . . . 111
1.2 The X′ schema . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
1.3 Intransitive elementary trees . . . . . . . . . . . . . . . . . . . . . . . . . 115
2 Deriving sentences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
2.1 Simple sentences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
2.2 Complex sentences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
3 The adjunct relation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
3.1 Modification is different . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
3.2 The need for an adjunction operation . . . . . . . . . . . . . . . . . . . . 126
3.3 A typology of syntactic dependents . . . . . . . . . . . . . . . . . . . . . . 129
3.4 More on the distinction between complements and adjuncts . . . . . . . 129
4 Do we need intermediate projections? . . . . . . . . . . . . . . . . . . . . . . . . . 131
5 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
6 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133

As was mentioned in Chapter 1 we can represent the individual vocabulary items of a

language as small pieces of syntactic structure, or elementary trees. The idea is to generate
phrases and sentences by composing (and possibly otherwise manipulating) these elementary
trees in mathematically well-defined ways. In this view, vocabulary items are comparable to
the atoms of physical matter. Atoms do not combine into molecules just because they happen
to be next to each other; rather, their combinatorial possibilities are governed by their internal
structure (for instance, the number of electrons on an atom’s outermost shell and the relative
number of protons and electrons).
Accordingly, in the first part of this chapter, we consider the internal structure of elemen-
tary trees. As in the last chapter, we begin by focusing on how verbs combine with their argu-
ments to form larger phrases. For the time being, we will treat noun phrases and prepositional
phrases as unanalyzed chunks, postponing discussion of their internal structure until Chapter 5.
We then generalize the approach developed for verbs and their arguments to the point where
Chapter 4: Introducing the X′ schema of phrase structure 111

we can build sentences—both simple ones and complex ones containing subordinate clauses. In
order to derive sentences, we will find it necessary to introduce a formal operation called move-
ment, which allows us to represent the fact that constituents can have more than one function
in a sentence.
In the second part of the chapter, we turn to the representation of the modification relation
(already familiar from Chapter 3, § 2). As we will show, it is not possible to combine modifiers with
elementary trees by the substitution operation introduced in Chapter 1. Besides substitution and
movement, we therefore introduce a third and final formal operation called adjunction.
This chapter repeatedly references structural relationships in trees. If the jargon becomes
difficult to follow, you may find yourself wanting to revisit Chapter 2, § 4, where the terms are
introduced.

The X′ schema for elementary trees 1

1.1 Transitive elementary trees

We begin our investigation of the internal structure of elementary trees by considering how a
transitive verb like ate combines with its two arguments in a sentence like (1).

(1) The children ate the pizza.

From the possibility of pronoun substitution, as in (2), we know that the two arguments are
constituents (specifically, noun phrases).

(2) They ate it.

In principle, the verb could combine with its two noun phrase arguments in either order, or
with both at once. The three possibilities are represented by the structures in (3) (we address
the question of which syntactic category to assign to the nodes labeled by question marks in a
moment).

(3) a. ? b. ?

NounPhr ? ? NounPhr

the children V NounPhr NounPhr V the pizza

ate ate
the pizza the children
Chapter 4: Introducing the X′ schema of phrase structure 112

c. ?

NounPhr V NounPhr
ate
the children the pizza
However, as we already know from the substitution test introduced in Chapter 2, § 4, transitive
verbs form a constituent with their object, as shown in (4).

(4) The children ate the pizza. → ✔ The children did so.

In other words, only (3a) is an adequate representation of the sentence.

What is the syntactic category of the constituents that result from these two combinations?
In principle, the result of combining a verb with a noun phrase might be a phrase with either
verbal or nominal properties. But clearly, a phrase like ate the pizza doesn’t have the distribution
of a noun phrase. For instance, it can’t function as the object of a preposition (even a semantically
vacuous one like of ). Nor does it pattern like a noun phrase in other respects. As we have just
seen, the appropriate pro-form for it is not a pronoun, but a form of do so, just as would be the
case if the predicate of the sentence were an intransitive verb. In other words, for the purposes
of do so substitution, the combination of a verb and its object is equivalent to an intransitive verb
(say, intransitive eat); cf. (4) with (5).

(5) The children ate. → ✔ The children did so.

So how about if we assign the syntactic category V to the verb-object combination? That
won’t do, though, because then do so should be able to substitute for eat regardless of the presence
of an object. But that isn’t the case, as shown in (6), which contrasts with (5).

(6) The children ate the pizza. → * The children did so the pizza.

Notice furthermore that the syntactic category of the verb-object constituent is distinct from the
syntactic category of the constituent that includes the subject. This is evident from the con-
trast in (7), which would be unexpected if both constituents belonged to the same syntactic cat-
egory.

(7) a. ✔ We saw the children eat the pizza.

b. * We saw eat the pizza.

In order to represent the facts in (4)–(7), the following notation has been developed. Verbs
are said to project three bar levels, conventionally numbered from zero to two. The lowest bar
level, V◦ , is a syntactic category for vocabulary items; it is often indicated simply by V without a
superscript. The next bar level is V′ (read as ‘V-bar’),1 the syntactic category of a transitive verb
and its object. The highest bar level is V′′ (read as ‘V double bar’), which is the result of combining
a V′ with a subject. For a transitive verb, each bar level corresponds to the number of arguments
with which the verb has combined.
Somewhat confusingly for the novice, the verb’s second projection, V′′ , is more often than
not labeled VP. In early work in generative grammar, the label VP was intended as a mnemonic
Chapter 4: Introducing the X′ schema of phrase structure 113

abbreviation for the verb phrase of traditional grammar and did indeed correspond to that cate-
gory. In current phrase structure theory, however, the label that corresponds to the traditional
verb phrase is V′ , whereas VP includes a verb’s subject, which the traditional verb phrase does
not. The idea is that the highest bar level projected by a verb contains all of its arguments. For
clarity, we will avoid using the term ‘verb phrase’ if possible, but if we do use it, we mean the
traditional verb phrase that excludes the subject (that is, V′ , not VP). Conversely, when we say
VP, we always mean the projection that contains all of the verb’s arguments, not the verb phrase
of traditional grammar.
The fully labeled structure for (1), with the standard labels for the three verbal projections,
is given in (8).
(8) VP

NounPhr V′

the children V◦ NounPhr

ate
the pizza
Given (8), we can ‘un-substitute’ the two arguments. This yields the elementary tree for
ate in (9).
(9) VP

»NounPhr« V′

V◦ »NounPhr«
ate

1.2 The X′ schema

As we show in § 2 later on in this chapter and in Chapter 5, the basic form of the elementary
tree in (9) can be extended to other syntactic categories. In other words, (9) is an instantiation
of a general phrase structure template, shown in (10) and known as the X′ schema (read: X-bar
schema) of phrase structure. X, Y, and Z are variables over syntactic categories, meaning any
kind of syntactic category will have a structure like in (10).
(10) XP

»ZP« X′

X◦ »YP«
morpheme
A number of standard terms are used in connection with the X′ schema. The vocabulary
item dominated by X—the element that projects the entire syntactic structure—is known, follow-
ing traditional terminology, as the head. X (also written X◦ ) is the head’s lexical projection, X′
Chapter 4: Introducing the X′ schema of phrase structure 114

the intermediate projection, and XP (also written X′′ ) the maximal projection (sometimes
also called phrasal projection).

Word of Caution
The terms ‘intermediate’ and ‘phrasal’ are somewhat misleading, since they suggest that
the syntactic status of intermediate projections is somehow intermediate between lexical
and phrasal constituents. This is not the case. Intermediate projections are full-fledged
phrases, and ‘intermediate’ simply refers to the position of the projection in the tree struc-
ture.

The three projections of the head form what we will call the spine of the elementary tree.
The correspondence between projections and the bar levels mentioned earlier is summarized in
Table 4.1.

Label Alternate label Projection Bar level

X X◦ Lexical 0
X′ Intermediate 1
XP X′′ Maximal (phrasal) 2

Table 4.1: Correspondence between projections and bar levels

Again following traditional terminology, the sibling of the head—YP in (10)—is called its
complement. As we discuss in the next subsection, elementary trees need not include a com-
plement position.

Word of Caution
Note the spelling of complement with e (not i). The idea is that complements complete the
meaning of the head.

The child of the maximal projection—ZP in (10)—is called the specifier.

Word of Caution
The term “specifier” suggests that constituents in that position somehow specify the re-
mainder of the tree, and this might lead you to confuse specifiers with modifiers (dis-
cussed later on in this chapter). At one point, constituents in specifier position were in-
deed thought to have this function, and that is how the name arose. We no longer believe
this, but the name has stuck (what we might call terminological inertia). In order to mini-
mize this confusion, syntacticians often use the abbreviation “spec” (pronounced “speck”
and sometimes spelled with an initial capital).

Each elementary tree has at most one specifier, and elementary trees can lack a specifier
altogether, as we will see later on in this chapter. The specifier and complement positions of a
Chapter 4: Introducing the X′ schema of phrase structure 115

head are its (syntactic) argument positions. In summary, an elementary tree consists of a spine
and from zero to two argument positions.

Take Note

The terms ‘specifier’, ‘complement’, and ‘argument’ can be used to refer to the structural
positions just defined or to the constituents that substitute into those positions. (This is
analogous to the way we can use a nontechnical term like ‘bowl’ to refer either to the
container itself (Hand me the bowl) or to its contents (I’d like a bowl (of soup)). If it is nec-
essary to avoid confusion between the two senses, we can distinguish between ‘specifier
position’ and ‘constituent in specifier position’ (and analogously for ‘complement’ and
‘argument’).

An important question that arises in connection with the X′ schema in (10) is how to rep-
resent predicates with more than two semantic arguments (say, rent or give). The most obvious
approach is to allow elementary trees with more than two complements. Plausible as this ap-
proach may seem, however, it is now widely assumed that syntactic structure is at most binary-
branching. In cases where we have evidence from linguistic judgments concerning the issue,
we repeatedly find that binary-branching structures correctly represent our judgments, whereas
ones with more branches don’t. It is this empirical basis that leads us to hypothesize that binary-
branchingness is a formal universal. If a predicate has more than two semantic arguments, there
are two ways in which the additional arguments can be integrated into syntactic structure. In
some cases (as with rent), the supernumerary arguments are integrated into syntactic structure
by adjunction, an operation distinct from substitution that we introduce later on in this chapter.
This case involves a syntax-semantics mismatch, since a semantic argument ends up occu-
pying a position that is not a syntactic argument position. In other cases (as with give), the
apparently atomic predicate is decomposed semantically and syntactically into more than one
head, thus yielding a total of more than two argument positions. This second case is discussed in
detail in Chapter 11.

1.3 Intransitive elementary trees

So far, we have discussed the internal structure of the elementary trees required for transitive
verbs (and transitive categories more generally). In this section, we address the internal struc-
ture of the elementary trees required for intransitive verbs—for instance, intransitive eat. Two
structures come to mind as possibilities—shown in (11).

(11) a. VP b. VP

»NounPhr« V′ »NounPhr« V◦
ate
V◦
ate

The trees differ in the presence of an intermediate projection, and (11b) might at first glance seem
Chapter 4: Introducing the X′ schema of phrase structure 116

preferable because it is simpler (in the sense of containing fewer nodes). However, (11b) violates
the X′ schema, and adopting it would complicate the grammar as a whole, which consists not just
of the elementary trees, but of the rules and definitions stated over them. For instance, adopting
(11a) allows us to summarize the facts concerning do so substitution illustrated in (4)–(6) by means
of the succinct generalization in (12).

(12) Do so substitutes for instances of V′ .

Given (11b), (12) would need to be reformulated as the more cumbersome disjunctive statement
(that is, a statement containing or) in (13).

(13) Do so substitutes for instances of V′ or of V without a complement.

A second, similar reason to prefer (11a) is that it permits the succinct definition of the notion of
specifier in (14a) rather than the disjunctive statement in (14b).

(14) a. Specifiers are siblings of intermediate projections.

b. Specifiers are siblings of intermediate projections or of lexical projections without a
complement.

Notice that we allow verbs that can be used either transitively or intransitively, such as eat,
to be associated with two elementary trees. More generally, we will allow a vocabulary item of
any syntactic category to project one or more elementary trees, as required by its combinatorial
properties.

Deriving sentences 2

2.1 Simple sentences

We are almost at the point of being able to construct X′ -compliant representations of complete
sentences, but before we can do so, we need to address the syntactic representation of tense.
The following discussion relies on the notion of do support and on the status of modals and
auxiliary do as members of the same syntactic category I(nflection); see Chapter 2, § 1.4 for
detailed discussion.
In a sentence like (15), the verb waited contains the bound morpheme -ed, which expresses
past tense.

(15) They waited.

If tense morphemes were invariably expressed on the verb in this way, then complete struc-
tures for full sentences could be derived by substituting appropriate structures into the argument
Chapter 4: Introducing the X′ schema of phrase structure 117

positions of the verb’s elementary tree. But this is not a general solution, because tense is not
always expressed as a bound morpheme. For instance, in (16), the future tense counterpart of
(15), the future tense is expressed by a free morpheme, the modal will.

(16) They will wait.

Even more strikingly, the past tense in English, though ordinarily expressed as a bound
morpheme, must be expressed by a free morpheme in do support contexts, as shown in (17).

(17) a. Emphasis: b. Negation: c. Question:

They did wait. They didn’t wait. Did they wait?
The morphologically variable expression of tense as a free or a bound morpheme raises
two related syntactic questions. First, what is the representation of sentences like (16) and (17a),
where tense is expressed as a free morpheme? (We postpone the treatment of questions and
negated sentences to Chapters 6 and 10, respectively.) Second, and more generally, how can
we represent all sentences in a syntactically uniform way, regardless of how tense is expressed
morphologically? There are two reasons for wanting a uniform representation. First, from a
semantic point of view, both past and future are semantically parallel functions, taking situations
(denoted by VPs) as input and returning as output situations that are located in time, either before
or after the time of speaking.2 Second, and very generally, when similar objects are represented in
a uniform way, it is easier for the mind to manipulate them, and they are computationally more
tractable than when they are not so represented. Mathematicians, logicians, computer scientists,
and others are therefore fond of finding normal forms (also known as canonical forms) for
the abstract objects they deal with. An example from daily life is that we impose a normal form
on the set of letters in the alphabet (namely, the conventional order in the alphabet song). That
way, looking up words in the dictionary is much quicker than it would be if the words were not
sorted with reference to the alphabet’s normal form. The X′ schema of phrase structure under
discussion in this chapter is a normal form, and our earlier preference for (11a) over (11b) can be
framed as a preference for a representation that is in normal form over one that isn’t.
Returning to the problem at hand, we begin by answering the first question in several steps.
First, it is clear that (16) and (17a) share a common predicate-argument structure (predicate
used here in the sense of Fregean predicate). That is, both of these sentences denote a situation
in which someone is waiting, with the sentences differing only as to which point in time the
situation holds. We can capture this commonality by taking the elementary tree for the verb wait
in (18a) and substituting an argument constituent in the specifier position, yielding (18b).

(18) a. VP b. VP

»NounPhr« V′ NounPhr V′

V◦ they V◦
wait wait
Second, in accordance with the general approach to syntactic structure that we have been devel-
oping, modals and auxiliaries, like all vocabulary items, project elementary trees. The elementary
Chapter 4: Introducing the X′ schema of phrase structure 118

trees for will and auxiliary did are shown in (19).

Extra Info
In this book, we give the syntactic category for modals, auxiliary do, and the silent tense
morphemes that we introduce presently the name ‘I’, short for “inflection”. (The auxiliary
verbs be and have are introduced in connection with the passive in Chapter 8.)

Nowadays, the usual name for this category is T, short for “tense”. Both names designate
exactly the same category. We prefer I, because the category includes modals, which do
not express tense, at least not in a “pure” way. The reason for the other name is historical.
The abbreviation I was current at one point, but then there were thought to be reasons to
“split” it into two separate heads: T(ense) and Agr(eement). This split was later abandoned,
and Agr(eement) became obsolete. Although there was no reason not to bring back the
old name I, the new label T stuck. Another instance of terminological inertia!

(19) a. IP b. IP

»NounPhr« I′ »NounPhr« I′

I◦ »VP« I◦ »VP«
will did

We can substitute the structure in (18b) into each of the elementary trees in (19), yielding
3
(20).

(20) a. IP b. IP

»NounPhr« I′ »NounPhr« I′

I◦ VP I◦ VP
will did
NounPhr V′ NounPhr V′

they V◦ they V◦
wait wait

The structures in (20) neatly reflect the semantic relation between tense and situations.
The element in I corresponds to the tense function, the complement of I (= VP) corresponds
to the function’s input (the situation), and the maximal projection of I (= IP) corresponds to the
function’s output (the situation located in time). There remains a problem, however: the I element
and the subject of the sentence are in the wrong order in (20). This problem can be solved by
introducing a movement operation that transforms the structures in (20) into those in (21).
Chapter 4: Introducing the X′ schema of phrase structure 119

(21) a. IP b. IP

NounPhri I′ NounPhri I′

they I◦ VP they I◦ VP
will did
ti V′ ti V′

V◦ V◦
wait wait

A few remarks are in order about this operation, but first, let’s introduce some descriptive
terminology. Movement always leaves a trace in the moved constituent’s original position. A
constituent and its traces (there can be more than one) are called a chain, and the elements of a
chain are its links. As we will see, even simple sentences can contain more than one chain, so
chains need to be distinguished from one another. Therefore, all links of a chain share an index.
We will use the lowercase letters (i, k, m, …) as movement indices.4 (We will generally omit the
indices j and l, since they are easily confused with i at small font sizes.) Higher links in a chain
are called the antecedents of lower ones. Finally, the highest and lowest links in a chain are
called the chain’s head and tail, respectively. It is perfectly possible, by the way, for a chain to
consist of a single constituent.5 This happens when a constituent hasn’t moved. The chain then
contains a single link, which is simultaneously the chain’s head and its tail.

Word of Caution
Don’t confuse the movement chain sense of the term ‘head’ with the X′ schema sense.
The head of an X′ structure is the vocabulary item that projects the structure. The head of
a movement chain is the highest constituent in a chain, regardless of its X′ status. Which
sense is meant is generally clear from the context.

Is the head of the movement chains in (21) necessarily a head in the X′ sense? No, because
the subject of a sentence can be a multi-word expression like the students in the class.

The movement operation just introduced is not intended to represent a psycholinguistic

process. Rather, it is conveniently understood as a way of representing mismatches between var-
ious aspects of linguistic structure. Specifically, in the case at hand, they satisfies two distinct
functions. First, it is a semantic argument of the verb wait. Second, it is the subject of the entire
sentence, whose syntactic head is I, not V. It is important to understand that the thematic role
is distinct from the grammatical function. This is clearly shown by the existence of passive sen-
tences. For instance, in the active sentence in (22a), it is the agent argument that functions as the
subject, whereas in its passive counterpart in (22b), it is the theme argument.
Chapter 4: Introducing the X′ schema of phrase structure 120

Background Info

In (22), grammatical functions are indicated above the boxed element and thematic roles
are indicated below. The following abbreviations are used:

sbj - subject, obj - object, pp - prepositional phrase, a - agent, thm - theme

sbj obj
(22) a. The committee drafted the letter.
a thm
sbj pp
b. The letter was drafted (by the committee).
thm a
But despite the double function of they, it is pronounced in the sentence only once. In
other words, there is a mismatch (a lack of one-to-one correspondence) between properties of
the syntax and properties of the phonology. It is this mismatch that is resolved by assuming
movement. Now there are two links, one corresponding to each syntactic property (argument,
subject), but only one of them is pronounced, which is consistent with the phonology.
We are now in a position to answer the second question posed earlier—namely, how can
sentences be represented in a syntactically uniform way regardless of the morphological expres-
sion of tense? A simple answer to this question is possible if we assume that English has tense
elements that are structurally analogous to auxiliary do, but not pronounced, as shown in (23).
In this book, we adopt the convention of enclosing such silent elements in square brackets.

(23) a. IP b. IP

»NounPhr« I′ »NounPhr« I′

I◦ »VP« I◦ »VP«
[pRs] [pst]
Elementary trees like (23) make it possible to derive structures for sentences in which tense
is expressed as a bound morpheme along the same lines as for sentences containing a modal
or auxiliary do. In other words, they make it possible to impose IP as the normal form for all
sentences. In (24), we illustrate the derivation of They waited.

(24) a. VP b. VP

»NounPhr« V′ NounPhr V′

V◦ they V◦
waited waited

Select elementary tree for verb Substitute argument

Chapter 4: Introducing the X′ schema of phrase structure 121

c. IP d. IP

»NounPhr« I′ NounPhri I′

I◦ VP they I◦ VP
[pst] [pst]
NounPhr V′ ti V′

they V◦ V◦
waited waited

Substitute predicate-argument Move subject from Spec(VP) to

structure in (24b) in elementary tree Spec(IP)
for tense

2.2 Complex sentences

This section is devoted to the derivation of sentences that contain complement clauses (also
known as clausal complements). Some examples are given in (25); the complement clauses
are in italics.

(25) a. We will ask if they left.

b. We believe that they came.

Although sentences with complement clauses can become unboundedly long (recall the
instances of recursion in Chapter 1, deriving structures for them proceeds straightforwardly along
the lines already laid out. If and that are both complementizers, so called because they have the
effect of turning independent sentences into the complements of a matrix verb, and they project
the elementary trees in (26).

(26) a. CP b. CP

C′ C′

C◦ »IP« C◦ »IP«
if that

Given elementary trees like (26), we can derive the italicized complement clause in (25a) as
in (27). The steps in (27a)–(27d) are analogous to (24a)–(24d).
Chapter 4: Introducing the X′ schema of phrase structure 122

(27) a. VP b. VP

»NounPhr« V′ NounPhr V′

V◦ they V◦
left left

Select elementary tree for verb Substitute argument

c. IP d. IP

»NounPhr« I′ NounPhri I′

I◦ VP they I◦ VP
[pst] [pst]
NounPhr V′ ti V′

they V◦ V◦
left left

Substitute predicate-argument Move subject from Spec(VP) to

structure in (27b) in Spec(IP)
elementary tree for tense

e. CP

C′

C◦ IP
if
NounPhri I′

they I◦ VP
[pst]
ti V′

V◦
left

Add complementizer layer

Chapter 4: Introducing the X′ schema of phrase structure 123

The structure in (27e) in turn allows us to derive the entire matrix clause, as in (28). For
readability, we collapse the internal structure of the complement clause in what follows.

(28)
a. VP b. VP

»NounPhr« V′ NounPhr V′

V◦ »CP« we V◦ CP
ask ask
if they left

Elementary tree for matrix clause Substitute arguments, including

verb clausal complement (27e)

c. IP

»NounPhr« I′

I◦ VP
will
NounPhr V′

we V◦ CP
ask
if they left

Substitute (28b) in for modal (19a)

d. IP

NounPhri I′

we I◦ VP
will
ti V′

V◦ CP
ask
if they left

Move subject from Spec(VP) to Spec(IP)

Chapter 4: Introducing the X′ schema of phrase structure 124

Given representations like (28d), we can now formally characterize recursive structures
as in (29).

(29) a. A structure is recursive if and only if it contains at least one recursive node.
b. A node is recursive if and only if it dominates a node distinct from it, but with the
same label.

Word of Caution
The recursive nodes in (28d) are the higher IP, I′ , VP, and V′ nodes (and no others). The
lower IP, I′ , VP, and V′ nodes (hidden in the triangle in (28d), but visible in (27e)) are not
recursive nodes, since they don’t dominate another instance of the same category.

Note that a recursive node need not be the root node of a tree and that it can be any
projection level (XP, X′ , or X).

For a node to be recursive, it is not enough that the tree contains a second instance of the
category somewhere. For instance, none of the NounPhr nodes in (28d) is recursive, even
though the sentences contains many of them. The first node has to dominate (though not
necessarily immediately dominate) the second one.

The adjunct relation 3

3.1 Modification is different

The elementary trees introduced so far allow us to represent two of the three basic linguistic
relations discussed in Chapter 3—namely, argumenthood and predication. As we have seen, se-
mantic arguments of a verb can be represented by substituting syntactic arguments at one of
the two argument positions in the verb’s elementary tree: either the complement position or the
specifier position. VPs and IPs can be treated as arguments (specifically, as complements) of I and
C, respectively. And finally, although predication is not reducible to argumenthood (recall from
Chapter 3 that expletive subjects are required independently of a verb’s semantic requirements),
subjects occupy specifier positions regardless of whether they are semantic arguments or not. In
other words, predication does not require a special structural relation uniquely associated with
it.
An important remaining question is how to represent the modification relation using the X′
schema developed so far. In principle, modification might resemble predication in not requiring
a structural relation of its own. As it turns out, however, neither of the two head-argument
relations (head-specifier, head-complement) adequately represents the relation between a head
and its modifier. As we have seen, when a verb combines with a complement, the category of the
Chapter 4: Introducing the X′ schema of phrase structure 125

resulting constituent (V′ ) is distinct from that of the verb (V) (recall the contrast between (4) and
(6)), and when the verb and the complement in turn combine with the specifier, the category of the
resulting constituent (VP) is distinct yet again (recall the contrast in (7)). By contrast, modifying
a verb-complement combination like ate the pizza in (30) does not change the syntactic category
of the resulting constituent, which remains V′ (the modifier is in italics).

(30) a. The children ate the pizza.

b. The children ate the pizza with gusto.

This is evident from the do so substitution facts in (31), where do so can replace either the
unmodified or the modified verb-complement combination.

(31) a. The children ate the pizza with gusto. → ✔ The children did so with gusto.
b. The children ate the pizza with gusto. → ✔ The children did so.

The same pattern holds for intransitive verbs that combine with a modifier.

(32) a. The children ate with gusto. → ✔ The children did so with gusto.
b. The children ate with gusto. → ✔ The children did so.

The do so substitution facts in (31) and (32) motivate the syntactic structure for (30b) that
is given in (33) (for clarity, we focus on the internal structure of the VP).

(33) VP

NounPhr V′

the children V′ PP

V◦ NounPhr with gusto

ate
the pizza

The structural relation of the modifier with gusto to the spine of the V projection is known
as the adjunct relation, and the modifier itself is said to be an adjunct. Modifiers are always
represented as adjuncts. As a result, ‘modifier’ and ‘adjunct’ tend to be used somewhat inter-
changeably. In this book, however, we will distinguish between the two terms as follows. We
will use ‘modifier’ when we want to highlight a phrase’s semantic function of qualifying or re-
stricting the constituent being modified. For instance, as we mentioned in Chapter 3, a verb like
laugh denotes the set of entities that laugh. Combining the verb with a modifier like uproariously
yields the expression laugh uproariously, which denotes a subset of the set denoted by laugh.
We will use the term ‘adjunct’ when focusing on a constituent’s structural position in a tree. As
we will see later on in this chapter, it is possible for semantic arguments to be represented as
syntactic adjuncts. This does not change the semantic argument into a modifier, however!
Chapter 4: Introducing the X′ schema of phrase structure 126

3.2 The need for an adjunction operation

The structure in (33) raises the question of what elementary tree for transitive ate is involved
in its derivation. ‘Unsubstituting’ both arguments and the modifier, as we did in (9) for trees
containing only arguments, yields the structure in (34).

(34) VP

»NounPhr« V′

V′ PP ← not a possible elementary tree!

V◦ NounPhr
ate

Is the structure in (34) a satisfactory elementary tree? Clearly, allowing it means that our
grammar now contains two elementary trees for transitive ate. At first glance, this doesn’t seem
like a serious problem, since we already allow two elementary trees for ate, the intransitive and
transitive ones in (35).

(35) a. VP b. VP

»NounPhr« V′ »NounPhr« V′

V◦ V◦ »NounPhr«
ate ate

But (34) differs in a crucial respect from the structures in (35): it is a recursive structure.
This has an extremely undesirable consequence: namely, that if we were to derive structures like
(33) by means of elementary trees like those in (34), there would be no principled way to avoid
an unbounded number of such elementary trees. For instance, the derivations of the sentences
in (36), with their increasing number of modifiers, would each require a distinct elementary tree
for drink, and each additional modifier would require an additional elementary tree.

(36) a. We would drink lemonade.

b. We would drink lemonade / in summer.
c. We would drink lemonade / in summer / on the porch.
d. We would drink lemonade / in summer / on the porch / with friends.
e. We would drink lemonade / in summer / on the porch / with friends / for fun.

But the whole point of a generative grammar is to generate an unbounded set of sentences from
a finite set of elementary expressions and operations. Given this aim, elementary trees must be
non-recursive structures, with the consequence that adjuncts cannot be integrated into larger
syntactic structures by substitution. Accordingly, we introduce a further tree operation called
adjunction. The specific operation of interest to us is sometimes called Chomsky-adjunction,
Chapter 4: Introducing the X′ schema of phrase structure 127

to distinguish it from Joshi-adjunction, a different formal operation that plays a central role in
Tree-Adjoining Grammar (Joshi, Levy, and Takahashi 1975).
For the moment, we will use the adjunction operation to integrate modifiers into syntactic
structures. As we will see in Chapter 6, the adjunction operation is also used for other purposes.
Whatever its linguistic purpose, however, it is always the same formal (that is, graph-theoretical)
operation: namely, a two-step process that targets a particular node. When the purpose of ad-
junction is to integrate a modifier into a larger structure, as it is here, the target of adjunction
is an intermediate projection, indicated by the red box in (37a). The first step in carrying out
adjunction is to make a clone of the target of adjunction that immediately dominates the original
node, as in (37b). The second step is to attach the tree for the modifier as a child of this higher
clone, as in (37c).

(37) a. VP b. VP

»NounPhr« V′ »NounPhr« V′

V◦ »NounPhr« V′
ate

V◦ »NounPhr«
ate

Select target of adjunction Clone target of adjunction

c. VP

»NounPhr« V′

V′ PP

V◦ »NounPhr« with gusto

ate

Attach modifier as child of higher

clone

Deriving the rest of the structure for the entire sentence proceeds as outlined in § 2.1 earlier, as
shown in (38).
Chapter 4: Introducing the X′ schema of phrase structure 128

(38) a. VP

NounPhr V′

the children V′ PP

V◦ NounPhr with gusto

ate
the pizza

Substitute arguments

b. IP

»NounPhr« I′

I◦ VP
[pst]
NounPhr V′

the children V′ PP

V◦ NounPhr with gusto

ate
the pizza

Substitute (38a) in elementary tree for tense

c. IP

NounPhri I′

the children I◦ VP
[pst]
NounPhr V′

the children V′ PP

V◦ t i with gusto
ate

Move subject
Chapter 4: Introducing the X′ schema of phrase structure 129

For expository reasons, we have illustrated the derivation of the sentence with adjunction
preceding substitution and movement. However, the order of adjunction with respect to the other
operations is irrelevant.

3.3 A typology of syntactic dependents

A convenient cover term for complements, specifiers, and adjuncts is syntactic dependent. Each
type of syntactic dependent stands in a unique structural relation to the head and the spine that it
projects. Complements and adjuncts are both children of intermediate projections, but they differ
in that complements are siblings of heads, whereas adjuncts are siblings of the next higher pro-
jection level. As siblings of intermediate projections, adjuncts resemble specifiers. But adjuncts
are children of intermediate projections, whereas a specifier is the child of a maximal projection.
These structural relations and distinctions are summarized in Table 4.2, which also includes the
formal operations that fill or create the positions in question.

Relation to head Sibling of… Child of… Formal operation

Complement Head Intermediate projection Substitution
Adjunct Intermediate projection Intermediate projection Adjunction
Specifier Intermediate projection Maximal projection Substitution

Table 4.2: Summary of structural relations and formal operations

3.4 More on the distinction between complements and adjuncts

Given Table 4.2, it is easy to tell whether a constituent is represented in a particular tree diagram
as a complement or as an adjunct. However, it is not always self-evident whether a phrase is a
complement or an adjunct as a matter of linguistic fact.

Word of Caution
Remember that tree structures are models of linguistic facts, that can be correct or in-
correct. Just because it is possible to build a tree that represents a certain phrase as a
complement doesn’t mean that the phrase actually is a complement.

The most reliable way to determine the relation of a particular phrase to a verb is to use
do so substitution. If a phrase need not be included as part of the sequence being replaced by do
so, then it is an adjunct. If it must be included, then it is a complement. Using this test, we find
that phrases specifying cause or rationale, time, location, or manner are generally adjuncts, even
if they are noun phrases. Some examples, including the results of do so substitution, are given in
(39); the adjuncts are in italics.

(39) a. Rationale:
They waited for no good reason, but we did so for a very good one.
b. Duration:
They waited (for) a day, but we did so (for) a month.
Chapter 4: Introducing the X′ schema of phrase structure 130

c. Location:
They waited in the parking lot, but we did so across the street.
d. Manner:
They waited patiently, but we did so impatiently.

In the examples that we have seen in this book so far, semantic arguments are expressed
as syntactic arguments (or not at all). It is possible, however, for semantic arguments to be ex-
pressed in the syntax as adjuncts (this is one of the the mismatch cases mentioned earlier in
connection with binary-branchingness). For example, as we mentioned in Chapter 3, § 1, rent,
from a semantic point of view, is a five-place predicate, with arguments denoting property owner,
tenant, rental property, amount of money, and lease term. Some of these semantic arguments are
expressed as syntactic arguments. For instance, in (40), the phrase denoting the rental property
is a complement, as is evident from the results of do so substitution.

(40) Dennis rented the apartment to Lois. → * Dennis did so the apartment to Lois.

By contrast, do so substitution shows that the phrase denoting the lease term is an adjunct, even
though lease terms are semantic arguments of rent on a par with rental properties.

(41) Dennis rented Lois the apartment for two months. → ✔ Dennis did so for two months.

A final word should be said about the correlation between a syntactic dependent’s obliga-
tory or optional character and its status as a complement or adjunct. It is tempting to assume the
biconditional relationship in the “Dream world” column in Table 4.3.6

If a syntactic dependent is… then it is… Dream world Real world

obligatory a complement. TRUE TRUE
a complement obligatory. TRUE sometimes FALSE

Table 4.3: Correlation between obligatoriness and complement status

But as the rightmost column indicates, the biconditional relationship doesn’t hold. It is true that
obligatory syntactic dependents are complements. For instance, the contrast in (42) is evidence
that the noun phrase following devour is a complement, a conclusion that is borne out by do so
substitution in (43).

(42) Every time I see him…

a. * …he’s devouring.
b. ✔ …he’s devouring a six-inch steak.
(43) * He devoured a hamburger and french fries, …
a. * …and she did so six samosas.
b. ✔ …and she did so, too.
Chapter 4: Introducing the X′ schema of phrase structure 131

But not all complements are obligatory. The grammaticality of both (44a) and (44b) shows
that the phrase French fries in (44b) is optional. But the ungrammaticality of (44c) shows that it
is nevertheless a complement.

(44) a. ✔ She ate, and he did so, too.

b. ✔ She ate French fries, and he did so, too.
c. * She ate French fries, and he did so three samosas.

Although the second row of Table 4.3 is false, the first row does have the valid consequence
in (45) (derived by the modus tollens rule of propositional logic).

(45) If a syntactic dependent is not a complement, it is not obligatory.

The two valid generalizations in the first row of Table 4.3 and (45) can be summarized
succinctly as in (46).

(46) a. Obligatory syntactic dependents are complements.

b. Adjuncts are optional.

Do we need intermediate projections? 4

We conclude this chapter by raising a question concerning the simplicity of the theory of phrase
structure that we have been developing. Given that words (or syntactic atoms of some sort) com-
bine with one another to form phrases, any theory of syntax must assume heads and phrases.
But making a further distinction between two types of phrases (intermediate projections ver-
sus maximal projections) seems inelegant, and attempts have therefore been made to eliminate
intermediate projections, along with the possibility of adjunction to them. For instance, given
our current assumptions, sentences like (47) force us to allow adjunction to intermediate projec-
tions.

(47) a. [IP They [I′ never [I′ will agree to that ] ] ].

b. God let [VP there [V′ suddenly [V′ be light ] ] ].

However, if the IP and the small clause VP in such sentences were ‘split up’ into two sep-
arate projections, it would be possible to eliminate the intermediate projections and to adjoin
the modifiers to maximal projections instead. This is illustrated in (48), where IP has been split
into Agr(eement)P and T(ense)P, and the small clause VP has been split into Pred(ication)P and a
lower VP.

(48) a. [AgrP They [TP never [TP will agree to that ] ] ].

 Chapter 4: Introducing the X′ schema of phrase structure 132

b. God let [PredP there [VP suddenly [VP be light ] ] ].

A useful way to frame the issue is as a trade-off between two options. The first option
buys a relatively small set of familiar syntactic categories at the cost of assuming intermediate
projections. The second option buys an intuitively appealing two-level phrase structure scheme at
the cost of a proliferating and increasingly abstract set of syntactic categories. In this introductory
textbook, we will continue to assume the classic X′ schema in (10) with its three bar levels.
Given this choice, we know from (47) that adjunction must be able to target intermediate
projections. As we will see in Chapter 6, adjunction must also be able to target heads. Assuming
three bar levels, can adjunction also target maximal projections? The simplest answer (taking
‘simple’ to mean ‘maximally general’) answer is ‘yes’. In this textbook, however, we will see only
examples of adjunction to heads and intermediate projections.7

Notes 5
1. Why is V′ read as V-bar when it contains not a bar (V), but a prime symbol? The reason is that when the idea of
bar levels was introduced in the 1970s, the various levels were distinguished by horizontal bars over a syntactic
category. The lowest level had no bars, the first level one, and the second two. But back in the days of typewriters,
such overbars were cumbersome to type (you typed the symbol –*, rolled up the platen a bit, backspaced, typed
an overbar *–, repeated from –* to *– for each overbar, and then rolled the platen down again the right amount).
Overbars are also expensive to typeset, and even today, they aren’t part of the standard character sets for HTML
documents. Therefore, it was and continues to be convenient to substitute prime symbols for overbars. However,
linguists have failed to update their terminology (terminological inertia again!), and so the old term ‘bar’ is still with
us.

2. The semantics of tense we are assuming here is oversimplified, but sufficient for our purposes.

3. The representations in (20) look like appropriate representations for the questions Will he wait? and Did he wait?
But they can’t be, since they contain unfilled substitution nodes. Moreover, as we will see in Chapter 6, there is
reason to postulate a projection above IP in the representations of questions.

4. In the syntax literature, it is standard to use alphabetical subscripts not just to track movement but also to indicate
reference and coreference (see § 1 of Chapter 15). For clarity, this book uses lowercase letters as movement indices
exclusively. If we need referential indices, we use the natural numbers.

5. Mathematicians would call a chain with a single constituent a degenerate case.

6. A similarly tempting biconditional relationship (and false there, too) was discussed in §3.2 of Chapter 2.

7. Sentences like (i) appear to require adjunction to maximal projections (IP, given our assumptions).

(i) Tomorrow, we will eat pizza.

However, in Chapter 6, we will give such examples a structure in which the clause-initial phrase moves to the
Chapter 4: Introducing the X′ schema of phrase structure 133

specifier position of the projection of a (silent) head higher than I. Analogous reasoning would extend to examples
like (ii), where the silent head would have to be even higher.

(ii) Tomorrow, what will we eat?

Exercises and problems 6

Exercise 4.1
What is the X′ status of Fregean and of Aristotelian predicates? You should be able to answer in
one or two brief sentences.

Exercise 4.2
The trees in (1) fail to correctly account for certain grammaticality judgments. What are those
judgments?

(1) a. IP

NounPhri I′

Bertie I◦ VP
[pst]
ti V′

V′ PP

V′ NounPhr in one sitting

V◦ a dozen apples
consumed
Chapter 4: Introducing the X′ schema of phrase structure 134

b. IP

NounPhri I′

Jeeves I◦ VP
will
ti V′

V◦ NounPhr PP
mix
his special concoction for Bertie

Exercise 4.3
A. Are the italicized phrases in (1) syntactic arguments or adjuncts? Explain. There is no need
for extensive discussion beyond your one-word conclusion and the evidence from do so
substitution on which you base it.

(1) a. They waited for us.

b. This program costs twenty dollars.
c. We drove to Denver.
d. We worded the letter carefully.
e. They may behave very inconsiderately.
f. This volcano might erupt any minute.

B. Build structures for the sentences in (1). Needless to say, the structures you build should
be consistent with the evidence you gave in (A).

Exercise 4.4
A. Build structures for the sentences in (1). Motivate your attachment of the lowest argument
or adjunct in each sentence (in other words, provide the do so substitution evidence that
leads you to attach the phrase in question the way that you do).

(1) a. They demolished the house.

b. Mona Lisa called the other neighbor.
c. Mona Lisa called the other day.
d. You will recall that her smile amazed everyone.
e. Most people doubt that Mona Lisa lives in Kansas.
f. My friend wondered if Mona Lisa would come to his party.

B. Indicate all recursive nodes in the structures that you build for (1).
Chapter 4: Introducing the X′ schema of phrase structure 135

Exercise 4.5
A. (1) is structurally ambiguous. Paraphrase (or otherwise distinguish) the two relevant in-
terpretations. (Focus on the structural ambiguity, ignoring the referential ambiguity of
they.)

(1) They claimed that they paid on the 15th .

B. Build a structure for each of the interpretations, indicating which structure goes with which
interpretation. We encourage you to build chunks and indicate how they fit together dif-
ferently for the two interpretations.
C. Indicate all recursive nodes in the structures that you build for (1).

Exercise 4.6
A. Make up a sentence with two adjuncts. Provide syntactic evidence that the adjuncts are
adjuncts rather than syntactic arguments. Then build the structure for the sentence. Finally,
switch the linear order of the adjuncts, and build the structure for the resulting word order
variant of your original sentence.
B. Make up a simple sentence in which one of the semantic arguments of the verb is expressed
in the syntax as an adjunct. Provide evidence that the adjunct is one. Finally, build the
structure for your sentence.

Exercise 4.7
In connection with introducing the formal operation of adjunction, we insisted that “the whole
point of a generative grammar is to generate an unbounded set of sentences from a finite set of
elementary expressions and operations”. What problem arises if we allow a generative grammar
to include a potentially unbounded set of elementary trees or operations?

Exercise 4.8
Is it possible for two adjuncts to be siblings? Explain.

Problem 4.1
Is it possible for an adjunct to precede a complement? How about immediately precede?
What if you are allowed to “swivel” at X′ in the X′ schema, so that YP precedes X? (The
resulting head-final structures are discussed in more detail in Chapter 5, § 4.)
Chapter 4: Introducing the X′ schema of phrase structure 136

(1) a. XP b. XP

»ZP« X′ »ZP« X′
(spec) (spec)
X◦ »YP« »YP« X◦
head (comp) (comp) head

Head-initial Head-final

Problem 4.2
In the chapter, we define adjunction as an operation that clones a target node and attaches a
phrase as the child of the higher clone. Imagine an operation that clones the target node and at-
taches a phrase as the child of the lower clone. Would such an operation be useful? Explain.

Problem 4.3
As mentioned in the text, it is attractive to assume a theory of phrase structure that does away
with the notion of intermediate projection. For cases like (47), it is possible to eliminate the dis-
tinction between intermediate and maximal projections by means of extending the set of syntactic
categories.
But there are further difficulties that cannot be solved by this means. What are they? Can
you think of a way of solving them?
Hint: Redo Table 4.2 using two bar levels.
 5
Extending the X′
schema

Chapter outline
1 Noun phrases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
1.1 Parallels and differences between noun phrases and sentences . . . . . . 138
1.2 Noun phrases as DPs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
1.3 More on determiners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
1.4 Modification and related issues . . . . . . . . . . . . . . . . . . . . . . . . 151
2 Adjective phrases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
3 Prepositional phrases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
3.1 Transitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
3.2 Clausal complements of prepositions . . . . . . . . . . . . . . . . . . . . . 163
4 Crosslinguistic variation in headedness . . . . . . . . . . . . . . . . . . . . . . . . 165
4.1 Head-final vs. head-initial patterns . . . . . . . . . . . . . . . . . . . . . . 165
4.2 Harmonic and disharmonic headedness . . . . . . . . . . . . . . . . . . . 168
4.3 Final thought . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
5 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
6 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174

In Chapter 4, we introduced a normal form for phrase structure, the X′ schema, according
to which lexical items project an elementary tree consisting of a spine and up to two argument
positions. In this chapter, we extend the X′ schema to syntactic categories other than V, I, or C,
to include N(oun), D(eterminer), A(djective), and P(reposition). The final section of the chapter
illustrates crosslinguistic variation with regard to the order of heads and complements.
Chapter 5: Extending the X′ schema 138

Noun phrases 1

1.1 Parallels and differences between noun phrases and

sentences
The X′ schema of phrase structure that we introduced in Chapter 4 is a specific expression of a
more general idea—namely, that lexical items of different syntactic categories show significant
cross-categorial parallels. In the history of generative grammar, this idea was primarily based
on the cross-categorial parallels between noun phrases and sentences (Chomsky 1970). In what
follows, we review these parallels as well as some differences between the two categories.

Argument structure
Early in the history of generative grammar (Lees 1960), it was observed that sentences like (1a)
and noun phrases like (1b) share several important properties.

(1) a. The army destroyed the city.

b. the army’s destruction of the city

The semantically central element of the sentence in (1a) is the verb destroyed, and its semantic
arguments, the agent the army and the theme the city, are both expressed as syntactic arguments
of the sentence. In a parallel way, the semantically central element in the noun phrase in (1b) is
the nominal counterpart of destroy, the noun destruction. Like the verb, the noun is associated
with an agent argument and a theme argument that are both overtly expressed—in this case, as
the possessive expression the army’s and the prepositional phrase of the city.
The correspondence in (1) is supported by that between the passive sentence in (2a) and its
passive-like noun phrase counterpart in (2b).

(2) a. The city was destroyed (by the army).

b. the city’s destruction (by the army)

In both of these examples, the argument preceding the head is now the theme the city(’s), and the
agent argument is expressed by an optional by phrase.
Chapter 5: Extending the X′ schema 139

Modification
A further parallel between sentences and noun phrases is that in both categories, the semantically
central element—the verb or the noun—can be modified in similar ways, as illustrated in (3) and
(4).

(3) Adverb: She regularly gives money to the organization.

Prepositional phrase: She gives money to the organization on a regular basis.

(4) Adjective: her regular gifts of money to the organization.

Prepositional phrase: her gifts of money to the organization on a regular basis

Some cross-categorial differences

Sentences and noun phrases also exhibit certain differences. First, arguments and modifiers are
not always expressed in exactly the same way across the two categories. For instance, the agent
argument is expressed as an ordinary noun phrase in a sentence like (1a), but as a possessive noun
phrase in a noun phrase like (1b). In a sentence, the theme argument is expressed as a noun phrase,
but in a noun phrase, it must be part of a prepositional phrase, usually an of phrase. Finally,
although verbs and nouns can both be modified by prepositional phrases, verbs are modified by
adverbs, whereas nouns are modified by adjectives (with some exceptions in both directions).
In connection with this last difference, notice that adverbs can precede or follow the verb they
modify, whereas adjectives (in English) are ordinarily restricted to prenominal position.

(5) Adverb
a. The kids regularly donate their old toys.
b. The kids donate their old toys regularly.

(6) Adjective
a. the kids’ regular donation of their old toys.
b. * The kids’ donation their old toys regular.

A further and even more fundamental difference between sentences and noun phrases con-
cerns the subject requirement. As we saw in Chapter 3, all sentences require a syntactic subject,
even when it does not correspond to a semantic argument, as is evident from the contrast between
(7) and (8).

(7) a. ✔ It appears that the manuscript has been found.

b. ✔ There exists a solution.

(8) a. * Appears that the manuscript has been found.

b. * Exists a solution.

By contrast, noun phrases never require a subject. For instance, the agent argument of a noun
can be expressed, but it needn’t be, as shown in (9).1
Chapter 5: Extending the X′ schema 140

(9) a. ✔ the committee’s criticism of the proposal

b. ✔ the criticism of the proposal

What is even more striking is that sentences with expletive subjects have no noun phrase
counterparts. As (10) shows, the very expletive expressions that are obligatory in (7) are ungram-
matical in noun phrases.2

(10) a. * it(s) appearance that the manuscript has been found

b. * there(’s) existence of a solution

In summary, noun phrases resemble sentences in that their core categories—nouns and
verbs, respectively—have semantic arguments that can be expressed as syntactic arguments in
partly similar ways. Nouns and verbs can also be modified in largely similar fashion. We do
not deny that the two categories differ fundamentally with respect to the subject requirement.
However, in the remainder of this part of the chapter, our focus will be on how to represent the
parallel aspects of noun phrases and sentences.

1.2 Noun phrases as DPs

A striking fact about nouns is that, in many constructions in many languages, they cannot in
general function as arguments on their own, but must be accompanied by a determiner.

(11) a. * Assignment is not difficult.

b. * You should hand in assignment.

(12) a. { The, this, that } assignment is not difficult.

b. You should hand in { the, this, that } assignment.

We conclude from this that noun phrases are the result of composing two projections, one headed
by the noun and the other by the determiner, as shown in (13).

(13) a. DP b. NP c. DP

D′ N′ D′

D◦ »NP« N◦ D◦ NP
the assignment the
N′

N◦
assignment
Chapter 5: Extending the X′ schema 141

Take Note

Given the structure in (13c), the traditional term ‘noun phrase’ is a misnomer since noun
phrases are maximal projections of D rather than of N. Because the term ‘noun phrase’
is firmly established in usage, we continue to use it as an informal synonym for ‘DP’.
However, in order to avoid confusion, we will use the term ‘NP’ only to refer to the sub-
constituent of a noun phrase that is the complement of a determiner. We will never use
‘NP’ to refer to an entire noun phrase (that is, a DP).

In the simplest case, the elementary tree for a noun consists of just a spine, as in (13b). But
as we have just seen, nouns, like verbs, can have both complements and specifiers. For instance,
depending on which of the noun phrases in (14) it appears in, criticism is associated with one of
the elementary trees in (15).

(14) a. They refuted the criticism.

b. They refuted the criticism of the proposal.
c. They refuted the committee’s criticism.
d. They refuted the committee’s criticism of the proposal.

(15) a. NP b. NP

N′ N′

N◦ N◦ »PP«
criticism criticism

c. NP d. NP

»DP« N′ »DP« N′

N◦ N◦ »PP«
criticism criticism

In (14a), the phrase the criticism is derived in exactly the same way as the assignment in
(13)—by substituting the NP in (15a) as the complement of the determiner. In (15b), the noun
phrase containing criticism is derived as in (16). (The internal structure of PPs is covered later on
in the chapter; for the moment, we ignore it.)
Chapter 5: Extending the X′ schema 142

(16) a. NP b. NP
N′ N′
N◦ »PP« N◦ PP
criticism criticism
of the proposal

Elementary tree for N (15b) Substitute theme argument

c. DP
D′
D◦ NP
the
N′
N◦ PP
criticism
of the proposal

Substitute (16b) in elementary tree

for D

Noun phrases containing possessive noun phrases, like (14c), require a possessive head ’s,
which contains two argument positions. The elementary tree for this head is shown in (17).

(17) DP
»DP« D′
D◦ »NP«
’s

Notice that the possessive head is not a free morpheme. Recall from Chapter 4 that we posited
elementary trees headed by silent bound tense morphemes in order to allow us to represent Eng-
lish past and present tense sentences analogously to their future tense counterparts. In the same
spirit, we allow elementary trees headed by overt bound morphemes like ’s. In general, modern
syntactic theory is not terribly concerned with whether the heads of elementary trees are bound
or free morphemes, or silent or overt, as long as the trees allow us to provide maximally similar
representations for linguistically related phenomena.
The derivation of the entire noun phrase proceeds as shown in (18). The elementary tree
for the noun is given in (18a). The agent argument substitutes into that tree’s specifier position.
To simplify the representation, we omit the argument’s internal structure, which is analogous to
(13c). The resulting NP then substitutes into the complement position of the possessive head, and
Chapter 5: Extending the X′ schema 143

finally, the argument in Spec(NP) moves to Spec(DP) in a manner analogous to subject movement
in sentences.

(18) a. NP b. NP
»DP« N′ DP N′
N◦ »PP« the committee N◦
criticism criticism

Elementary tree for N (15c) Substitute agent argument

c. DP d. DP
»DP« D′ DPi D′
D◦ NP the committee D◦ NP
’s ’s
DP N′ ti N′
the committee N◦ N◦
criticism criticism

Substitute (18b) in (17) Move specifier from Spec(NP) to

Spec(DP)

Finally, deriving (14d) involves substituting both the agent and theme arguments in the
elementary tree in (15d). The remainder of the derivation is identical to that of (14c) and is shown
in (19).

(19) a. NP b. NP

»DP« N′ DP N′

N◦ »PP« the committee N◦ PP

criticism criticism
of the proposal
Elementary tree for N (15d) Substitute agent and theme arguments
Chapter 5: Extending the X′ schema 144

c. DP

»DP« D′

D◦ NP
’s
DP N′

the committee N◦ PP
criticism
of the proposal

Move specifier from Spec(NP) to Spec(DP)

d. DP

DPi D′

the committee D◦ NP
’s
ti N′

N◦ PP
criticism
of the proposal

Substitute (18c) in (17)

Before proceeding, we draw attention to the fact that the orthographic word committee’s is
not a constituent in the syntactic structures in (18) and (19). It is not easy to determine whether
this mismatch can be eliminated for all possessive noun phrases. If not, then we have an addi-
tional mismatch between syntax and morphology on a par with the syntax-prosody and syntax-
semantics mismatches discussed in Chapter 2. We will see undoubted syntax-morphology mis-
matches in Chapters 10 and 11. Until then, we put the issue aside.
In (20), we repeat the tree for the noun phrase in (18d) followed by the tree for a sentential
counterpart. We maximize the parallel by choosing overt heads for both D and I, and we omit the
internal structure of DPs and PPs in order to highlight the topological parallel between the two
trees. As is evident, apart from the labels for the syntactic categories, the two-layered structure
for noun phrases (NP, DP) presented here is analogous to the two-layered structure for sentences
from Chapter 4 (VP, IP).
Chapter 5: Extending the X′ schema 145

(20) a. DP

DPi D′

the committee D◦ NP
’s
ti N′

N◦ PP
criticism
of the proposal
b. IP

DPi I′

the committee I◦ VP
will
ti V′

V◦ DP
criticize
the proposal

Given the structural parallels between (20a) and (20b), it is convenient to generalize the
notion of subject to include both Spec(IP) and Spec(DP). Accordingly, we will use the term ‘subject
movement’ to subsume both movement from Spec(VP) to Spec(IP) and movement from Spec(NP)
to Spec(DP).
The structural parallel between the two trees in (20) is further supported by the following
semantic parallel. In a formal semantics that is simple but sufficient for our purposes, an NP
constituent denotes a set of individuals. For instance, the NP dominating woman denotes the
set of all women, and the NP dominating president of the United States denotes the set of all
presidents of the United States—past, present, and future. Combining an NP with a determiner
like this or those has the syntactic effect of yielding a DP and the semantic effect of picking out
a particular individual (or individuals, in the case of a plural noun) from the set denoted by the
NP. Which particular individuals are actually picked out depends not just on the meaning of the
NP and the determiner, but also on the particular discourse context in which the DP is used.
This is what allows a noun phrase like the cat to refer to different cats in different discourse
contexts. In a similar way, we can think of VPs as denoting situations. For instance, a VP like
these cats jump onto the dresser denotes the set of all situations in which the individuals denoted
by these cats jump onto the piece of furniture denoted by the dresser. Combining a VP with a
tense morpheme in I then picks out one of these situations. For instance, the tensed IP These cats
jumped onto the dresser picks out one of the situations that occurred before the time of speaking
(how the past tense morpheme combines with the verb to yield jumped is discussed in detail in
Chapter 10). Once again, the particular situation picked out depends in part on the discourse
Chapter 5: Extending the X′ schema 146

context, so that the same sentence can be used to refer to more than one situation. In the future
tense counterpart, The cats will jump onto the dresser, the future tense morpheme will doesn’t
combine morphologically with jump into a single word, but the semantic effect of substituting
the VP into the elementary tree for will continues to be picking out a particular situation from a
set—only in this case, situation being picked out is after the time of speaking. Roughly speaking,
we can think of the difference between past and future as analogous to the difference between
the determiners this and that.
In concluding this section, let us draw attention to the fact that in the noun phrases that
we have considered so far, any constituents in Spec(DP) have expressed arguments of the noun
and have hence undergone subject movement. But subjects of noun phrases don’t necessarily
originate in the NP projection. Many nouns, notably ones referring to objects or entities rather
than to events, are best treated as lacking semantic arguments. The elementary trees for such
nouns will therefore not contain any substitution nodes. This is illustrated for the noun book in
(21a). When such nouns co-occur with a possessor, the possessor is best treated as substituting
directly into Spec(DP), rather than moving there from Spec(NP). This is illustrated for the noun
phrase the student’s book in (21b). (For simplicity, we omit the internal structure of the possessor
DP, which again is analogous to (13c).)

(21) a. NP b. DP

N′ DP D′

N◦ the student D◦ NP
book ’s
N′

N◦
book

1.3 More on determiners

Subcategories of determiners
Like verbs and nouns, determiners have different degrees of transitivity. For instance, the definite
article the and the indefinite article a(n) are obligatorily transitive, whereas the demonstratives
this and that are optionally so.

(22) a. I’ll buy { the, a } book.

b. * I’ll buy { the, a }.

(23) a. I’ll buy { this, that } book.

b. I’ll buy { this, that }.

Certain ordinary pronouns pattern just like demonstratives, as shown in (24), and so we will treat
them, too, as optionally transitive determiners.
Chapter 5: Extending the X′ schema 147

(24) a. we Americans, you fool(s)

b. we, you

Finally, some ordinary pronouns behave like obligatorily intransitive determiners, as shown in
(25).

(25) a. I, he, she, it, they

b. * I idiot, he fool, she linguist, it piece of junk, they traitors

In this connection, recall the warning in Chapter 1 that the term ‘pronoun’ is potentially
misleading. It suggests that pronouns are a subclass of nouns. If that were so, then pronouns
should combine with articles and demonstratives in the same way that other nouns do. In fact,
however, pronouns behave exactly like complete noun phrases in this regard, as shown in (26).
The facts in (26) thus provide strong evidence for the analysis of pronouns as determiners just
presented.

(26) a. D + noun: the people, this woman, that addressee

b. D + noun phrase (= DP): * the these people, this the woman, that the addressee
c. D + pronoun (= pro-DP): * the they, this she, that you

Elementary trees for the various types of determiners that we have just discussed are given
in (27).

(27)
a. DP b. DP c. DP

D′ D′ D′

D◦ »NP« D◦ D◦ »NP«
an this this

Obligatorily transitive D Optionally transitive D

d. DP e. DP f. DP

D′ D′ D′

D◦ D◦ D◦ »NP«
he we we

Obligatorily intransitive D Optionally transitive D

Chapter 5: Extending the X′ schema 148

Silent determiners
As shown in (28), plural indefinite count nouns and indefinite mass nouns are apparently not
accompanied by an article in English, in contrast to their singular or definite counterparts.

(28) a. cars, apples; rice

b. a car, an apple; the rice

However, we assume for conceptual reasons that the examples in (28a) contain a silent article
that is semantically roughly comparable to the unstressed some in I would like some apples and
some rice. The elementary tree is shown in (29). We will sometimes also use indef(inite) for this
silent determiner.

(29) DP

D′

D◦ »NP«
∅some

We have two reasons for assuming the existence of silent determiners. First, this assump-
tion allows us to minimize the difference between English and a language like Spanish, where
the indefinite article is overt. The resulting correspondence between English and Spanish deter-
miners is shown in Table 5.1; the plural indefinite articles are in boldface. For simplicity, we give
only the masculine forms of the Spanish determiners.

English Spanish
sg pl sg pl
Demonstrative this these este estos
that those ese esos
Definite article the the el los
Indefinite article a(n) ∅indf.pl un unos

Table 5.1: English and Spanish determiners

Second, assuming the silent determiner allows us to maintain that all noun phrases are DPs. Sen-
tences like (30) can then all be derived using the single elementary tree for brought in (31).

(30) a. Some butlers brought some tea.

b. Some butlers brought tea.
c. Butlers brought some tea.
d. Butlers brought tea.
Chapter 5: Extending the X′ schema 149

(31) VP

»DP« V′

V◦ »DP«
brought

We show the full structure for the apparently articleless noun phrase butlers in (32). The
structure for tea is analogous.

(32) DP

D′

D◦ NP
∅some
N′

N◦
butlers

(33) combines the trees for all four sentences in (30) into a single representation.

(33) IP

DPi I′

D′ I◦ VP
[pst]
{ D◦ } NP ti V′
some
∅some N′ V◦ DP
brought
N◦ D′
butlers
{ D◦ } NP
some
∅some N′

N◦
tea

In principle, we could take an alternative tack. If it were our goal to assign the least pos-
sible amount of structure (that is, the structures with the fewest nodes) to each sentence in (30),
we would reject the silent determiner in (29) and represent butlers as a bare NP, as in (34) (and
analogously for tea).
Chapter 5: Extending the X′ schema 150

(34) NP

N′

N◦
butlers

The alternative structures under discussion for (30d) are given in (35). (35a) simply repeats
the appropriate variant from (33).

(35) a. IP

DPi I′

D′ I◦ VP
[pst]
D◦ NP ti V′
∅some
N′ V◦ DP
brought
N◦ D′
butlers
D◦ NP
∅some
N′

N◦
tea
b. IP

NPi I′

N′ I◦ VP
pst
N◦ ti V′
butlers
V◦ NP
brought
N′

N◦
tea

Clearly, the tree in (35b) is simpler than its counterpart in (35a) in the sense of containing
fewer nodes. However, this simplicity comes at the price of a veritable explosion in the number
Chapter 5: Extending the X′ schema 151

of elementary trees in the grammar, since every argument position that can be filled by a noun
phrase would need to be associated with two elementary trees (one with a DP substitution node,
and one with an NP substitution node). For instance, instead of the single elementary tree for
brought in (36a), we would need the three additional trees in (36b-36d).

(36) a. VP b. VP

»DP« V′ »DP« V′

V◦ »DP« V◦ »NP«
brought brought

c. VP d. VP

»NP« V′ »NP« V′

V◦ »DP« V◦ »NP«
brought brought

More generally, obligatorily intransitive verbs would require two elementary trees rather
than one, obligatorily transitive verbs—the case just illustrated—would require four (2 × 2) rather
than one, and optionally transitive verbs would require six (4 + 2) rather than two (1 + 1). This
result seems unappealing on computational grounds. But more fundamentally, the whole idea
of simplifying the representations of individual sentences is inconsistent with the Chomskyan
paradigm of language. Why? From a Chomskyan perspective, what syntactic theory attempts
to model and understand is the mental capacity to generate sentences. A reasonable working
hypothesis is that the best model for this capacity is the simplest possible grammar. So striv-
ing to cut down on the numbers of nodes in the representations of sentences at the expense of
complicating the grammar itself is missing the whole point of generative grammar in the first
place!

1.4 Modification and related issues

N′ as target of adjunction
As we noted in our introductory review of the parallels between noun phrases and sentences in
§1.1, nouns and verbs can be modified in similar ways. In (37), for instance, the same prepositional
phrase in the hospital modifies the noun stay and the morphologically related verb stayed.

(37) a. Mike’s stay in the hospital

b. Mike stayed in the hospital.

Extending the approach to representing modification introduced in Chapter 4, (§ 3.1) we can

derive the structure for the noun phrase in (37a) as in (38). (For simplicity, we omit the internal
structure of the possessive proper noun in the specifier (see §1.2 for discussion).
Chapter 5: Extending the X′ schema 152

(38) a. NP b. NP

»DP« N′ DP N′

N◦ Mike N◦
stay stay

Elementary tree for N ‘stay’ Substitute argument

c. DP d. DP

»DP« D′ DPi D′

D◦ NP Mike D◦ NP
’s ’s
DP N′ ti N′

Mike N◦ N◦
stay stay

Substitute (38b) in elementary tree Move subject

for possessive ’s

e. DP f. DP

DPi D′ DPi D′

Mike D◦ NP Mike D◦ NP
’s ’s
ti N′ ti N′

N◦ N′ PP
stay
N◦ in the hospital
stay

Select N′ as target of adjunction Adjoin PP at target of adjunction in

(38e)
Chapter 5: Extending the X′ schema 153

Apart from the category labels, the resulting structure in (38f), repeated for convenience
as (39a), is analogous to the structure for the corresponding sentence in (39b).

(39) a. DP b. IP

DPi D′ DPi I′

Mike D◦ NP Mike I◦ VP
’s [pst]
ti N′ ti V′

N′ PP V′ PP

N◦ in the hospital V◦ in the hospital

stay stay

Leftward adjunction
So far, we have discussed modifiers that follow the head, whose representation involves rightward
adjunction. Structures for examples like (40), where the modifier precedes the head it modifies,
can be derived by leftward adjunction, as shown in (41).

(40) a. the chimpanzee’s nervous grimace

b. The chimpanzee nervously grimaced.

(41) a. DP

DPi D′

the chimpanzee D◦ NP
’s
ti N′

AdjP N′

nervous N◦
grimace
Chapter 5: Extending the X′ schema 154

b. IP

DPi I′

the chimpanzee I◦ VP
[pst]
ti V′

AdvP V′

nervously V◦
grimaced

One substitution
As discussed in Chapter 4, do so substitution allows us to distinguish between complements and
adjuncts in the verbal system. A similar diagnostic is available in the nominal system—one sub-
stitution, which is illustrated in (42).

(42) a. this book on the floor and that one

b. this book on the floor and that one on the table

In the most natural interpretation of (42a), one is interpreted as book on the floor. In (42b), on the
other hand, one is interpreted as simply book. We can represent these facts by assuming that the
first conjunct in both cases has the structure in (43).

(43) DP

D′

D◦ NP
the
N′

N′ PP

N◦ on the floor
book

According to (43), the noun book has no complement, and the PP on the floor is an adjunct. The
pro-form one substitutes for instances of N′ , just as do so substitutes for instances of V′ . One
substitutes for the higher N′ in (42a), and for the lower N′ in (42b).
Chapter 5: Extending the X′ schema 155

As in the case of V′ , adjunction to N′ can apply more than once, yielding multiply recursive
structures like (44).

(44) DP

D′

D◦ NP
this
N′

N′ PP

N′ PP with the blue cover

N◦ on the floor
book

Restrictions on one substitution

A cautionary note is in order about one substitution. Although in principle one can substitute for
any instance of N′ , it is subject to two restrictions, which are important to keep in mind when
using one substitution as a diagnostic for syntactic structure. The first restriction, which makes
sense given the etymological connection of the pro-form to the homonymous number word, is
that one can substitute only for count nouns, as illustrated in (45).

(45) a. I prefer this { bed, chair, sofa, table }, and you prefer that one.
b. * I prefer this furniture, and you prefer that one.

A second restriction is that one cannot immediately follow the indefinite article, a cardinal num-
ber, a possessive noun phrase, or (for many speakers) the plural demonstratives these and those.
Whatever the exact source of this restriction is (in some cases it is difficult to derive from the
etymological connection just mentioned), the restriction is very superficial, since an intervening
word renders the ungrammatical (a) examples in (46)–(49) grammatical.3

(46) Indefinite article

a. * I bought a book, and you bought a one, too.
b. I bought a blue book, and you bought a red one.

(47) Cardinal number

a. * I bought { two , ten } books, and you bought { two, ten } ones, too.
b. I bought { two, ten } blue books, and you bought { two, ten } red ones.
Chapter 5: Extending the X′ schema 156

(48) Possessive
a. * I like { Mary’s, her } book, and you like { John’s, his } one.
b. I like { Mary’s, her } blue shirt, and you like { Mary’s, her } red one.

(49) Plural demonstrative

a. * I like these books, and you like those ones.
b. I like these blue books, and you like those red ones.

Structural ambiguity
Having introduced N′ as a possible target of modification, we are now in a position to associate
structurally ambiguous sentences like (50) with two distinct syntactic representations.

(50) They ate the pizza in the living room.

(50) has two interpretations, which can be paraphrased as in (51).

(51) a. Verbal modifier interpretation:

It was in the living room that they ate the pizza (though the pizza may have started
out elsewhere).
b. Nominal modifier interpretation:
It was the pizza in the living room that they ate (though perhaps they then took it and
ate it elsewhere).

On the interpretation in (51a), the prepositional phrase in the living room modifies the verb ate,
and (50) has the structure in (52a). On the interpretation in (51b), the prepositional phrase mod-
ifies the noun pizza, and the sentence has the structure in (52b).
Chapter 5: Extending the X′ schema 157

(52) a. IP

DPi I′

they I◦ VP
[pst]
ti V′

V′ PP

V◦ DP in the living room

ate
D′

D◦ NP
the
N′

N◦
pizza
b. IP

DPi I′

they I◦ VP
[pst]
ti V′

V◦ DP
ate
D′

D◦ NP
the
N′

N′ PP

N◦ in the living room

pizza

The structures in (52) are consistent with the results of relevant constituenthood tests. For
instance, substituting the ordinary pronoun it for the pizza and substituting did so for ate the
pizza yields (53a) and (53b), respectively.
Chapter 5: Extending the X′ schema 158

(53) a. They ate it in the living room.

b. They did so in the living room.

In both sentences, the prepositional phrase is unambiguously interpreted as a verbal modifier, as

expected given that the pizza and ate the pizza are represented as constituents in (52a), but not in
(52b).
Conversely, in the question-answer pair in (54), the prepositional phrase is unambiguously
associated with a nominal modifier interpretation. Again, this is expected, since the pizza in the
living room is represented as a constituent in (52b), but not in (52a).

(54) What did they eat? The pizza in the living room.

The complement-adjunct distinction in the nominal system.

Given the semantic parallel between the sentence in (55a) and the noun phrase in (55b), it is
reasonable to treat the of phrase in (55b) as a complement of the noun author.

(55) a. This woman authors murder mysteries.

b. this author of murder mysteries

That is, the elementary tree for author needed to derive (55b) is as in (56a), and the structure for
the entire noun phrase is (56b).

(56) a. NP b. DP

N′ D′

N◦ »PP« D◦ NP
author this
N′

N◦ PP
author
of murder mysteries

Since one is analogous to do so in substituting for intermediate rather than for lexical pro-
jections, we expect the contrast between (57) and (58), and this accurately reflects the judgment
of many speakers.

(57) a. ✔ This woman authors murder mysteries, and that man does so, too.
b. ✔ this author of murder mysteries and that one

(58) a. * This woman authors murder mysteries, and that man does so nature guides.
b. * this author of murder mysteries and that one of nature guides
Chapter 5: Extending the X′ schema 159

Some speakers, however, accept (58b), or at least do not completely reject it. How can we
make sense of this variation among speakers’ judgments? Recall that complements of nouns, un-
like those of verbs, are always expressed as prepositional phrases. This means that the evidence
whether a particular phrase is a complement or an adjunct is murkier in the case of nouns than in
the case of verbs, both for children acquiring the language and for adult speakers. A further, prob-
ably related, complication is that even nouns that are morphologically derived from obligatorily
transitive verbs are themselves optionally intransitive (for instance, compare consume, destroy,
employ with consumer, destroyer, employer). Moreover, the intransitive use of these nouns might
be more frequent than their transitive use. As a result, the mental grammar of some speakers
might include only the intransitive elementary tree in (59a), and not the transitive elementary
tree in (56a). Such speakers would have no way of deriving the structure in (56b), but they would
be able to derive the alternative structure in (59b) by adjoining the of phrase, rather than by
substituting it.

(59) a. NP b. DP

N′ D′

N◦ D◦ NP
author this
N′

N′ PP

N◦ of murder mysteries
author

For such speakers, author in (58b) would be an N′ , rather than an N, and so they would accordingly
accept (58b).
Notice furthermore that the intransitive elementary tree in (59a) is available even for speak-
ers whose mental grammar includes the transitive elementary tree in (56a), since all speakers of
English accept (60).

(60) ✔ this author and that one

If some of these speakers allow the of phrase to adjoin into the intransitive elementary tree in
addition to substituting into the transitive one, then they, too, would judge (58b) to be acceptable
(at least marginally so).
Both complements and adjuncts function semantically as restrictors, that is, content that
constrains what/who the DP refers to (the set of authors of murder mysteries is a subset of the
set of authors). Therefore, there won’t be an obvious semantic clue for speakers whether their
grammar differs from that of other speakers. The only clue will come from the difference with
respect to one substitution judgments, and any such differences are not going to be salient in
everyday life. It’s only in syntax classes that they are the focus of attention!
Chapter 5: Extending the X′ schema 160

Adjective phrases 2
In this section, we discuss the structure of adjective phrases, beginning with examples like those
in (61) and (62), where the prepositional phrase following the adjective is optional.

(61) a. They are proud.

b. They are proud of their grandchild.

(62) a. They are happy.

b. They are happy with their car.

Recall from Chapter 2 that the pro-form so substitutes for adjective phrases. More specifi-
cally, examples like those in (63) and (64) allow us to conclude that the of phrase is a complement
of proud in (63), but that the with phrase is an adjunct of happy in (64).

(63) a. ✔ They are proud, and we are so, too.

b. ✔ They are proud of their grandchild, and we are so, too.
c. * They are proud of their grandchild, and we are so of him, too.

(64) a. ✔ They are happy, and we are so, too.

b. ✔ They are happy with their car, and we are so, too.
c. ✔ They are happy with their car, and we are so with our bikes.

We can represent these facts by associating the two adjectives with the elementary trees in
(65) and by stating that so substitutes for instances of A′ . (We generally use ‘A’ for ‘adjective’. We
always use ‘Adv’ for ’adverb’, and we use ‘Adj’ where the distinction between the two categories
is at issue.)

(65)
a. AP b. AP c. AP

A′ A′ A′

A◦ A◦ A◦ »PP«
happy proud proud

Obligatorily intransitive Optionally transitive

Most adjectives in English, like the two just discussed, are optionally or obligatorily intran-
sitive. A rare case of an obligatorily transitive adjective is fond.4 The contrast in (66) is evidence
for the complement status of the of phrase (recall from Chapter 4 that obligatory syntactic de-
pendents are complements), and that status is confirmed by the results of so substitution.
Chapter 5: Extending the X′ schema 161

(66) a. * They are fond.

b. ✔ They are fond of their grandchild.

(67) a. * They are fond of their grandchild, and we are so of him, too.
b. ✔ They are fond of their grandchild, and we are so, too.

In view of the facts in (66) and (67), fond is associated with the single elementary tree in (68).

(68) AP

A′

A◦ »PP«
fond

Obligatorily
transitive

Prepositional phrases 3
The syntactic category P corresponds closely to the traditional part of speech of preposition, but is
not identical to it. We address two differences between the syntactic category and the traditional
part of speech in the next two subsections.

Take Note

Following standard usage in the syntax literature, we sometimes use the term ‘preposi-
tion’ to refer to the syntactic category P in contexts where the difference is either clear or
immaterial.
Chapter 5: Extending the X′ schema 162

3.1 Transitivity
The etymology of the term ‘preposition’ ( < Latin prae ‘before’ and positio ‘position’) implies that
all prepositions should precede a complement, and English does in fact have a number of obliga-
torily transitive Ps, some of which are illustrated in (69). The asterisk (*) outside the parenthesized
material is a convention to indicate that the parenthesized material is obligatory.

(69) a. They drove from *(Boston).

b. He’s the inventor of *(that gizmo).
c. She dove into *(the water).
d. They jumped onto *(the bandwagon).

But X′ theory leads us to expect that there should also be intransitive Ps, and as the exam-
ples in (70) show, this expectation is fulfilled.

(70) a. I’ve never seen him before (this meeting).

b. Are you for (the proposal) or against (it)?
c. The bird flew { in, out } (the window).
d. It’s time to get { off, on } (the train).
e. They jumped over (the ditch).
f. We’ve been fast friends ever since (that time).
g. She came to (her senses).
h. Have you looked underneath (the sombrero)?

In traditional grammar, Ps that are used intransitively are known as adverbs or particles,
rather than as prepositions, but this terminology goes against the spirit of X′ theory, which seeks
to maximize the parallels among categories. From our point of view, there is as little reason for
the syntactic category of a lexical item to depend on its transitivity in the case of a P like since as
there is in the case of a V like eat. In both cases, the intransitive variant has a semantic argument
that is not expressed in the syntax, but is supplied in the course of interpretation, based on the
discourse context.
The elementary trees for of and over are shown in (71), and the full structures for the PPs
headed by them in (69) and (70) are shown in (72). (72c) comes out as identical to (71c).

(71) a. PP b. PP c. PP

P′ P′ P′

P◦ »DP« P◦ »DP« P◦
of over over
Chapter 5: Extending the X′ schema 163

(72) a. PP b. PP c. PP

P′ P′ P′

P◦ DP P◦ DP P◦
of over over
that gizmo the ditch

3.2 Clausal complements of prepositions

As we saw in Chapter 4, verbs can take either noun phrase complements or clausal complements.
(73) gives a further example.

(73) a. They reported the monkey’s dislike of camphor.

b. They reported that the monkey dislikes camphor.

The examples of transitive Ps discussed so far have all had noun phrase complements, but given
the parallel between verbs and prepositions concerning transitivity, we might expect Ps to allow
clausal complements as well. Once again, this expectation is borne out, as shown in (74).

(74) a. Noun phrase complement: { after, before, since } the war

b. Clausal complement: { after, before, since } the war ended

In traditional grammar, Ps that take clausal complements are classified as subordinating

conjunctions (along with if and that), but as in the case of intransitive Ps, we again reject the
traditional approach. First, it is conceptually uneconomical. Specifically, it expresses the differ-
ence between (74a) and (74b) in terms of the syntactic category of the heads (preposition versus
subordinating conjunction), which redundantly encodes the difference in the syntactic category
of the complement (noun phrase versus clause). Second, the items in (74) share roughly the same
semantic content, regardless of the categorial status of their complement. In contrast, if and that
are relatively contentless and give the impression of functioning purely as ’grammatical glue’. In
the approach that we are advocating, the distinction between P and C corresponds to the distinc-
tion between contentful and contentless subordinating conjunctions.
The question now arises of what syntactic category clausal complements of prepositions
belong to. At first glance, examples like (74b) suggest that the answer to this question is IP. The
elementary tree for after in (74b) would then be as in (75), and the elementary trees for before
and since would be analogous.
Chapter 5: Extending the X′ schema 164

(75) PP

P′

P◦ »IP«
after

Not like this!

There is good reason to believe, however, that clausal complements (specifically, finite
clausal complements) of P are CPs rather than IPs. As illustrated in (76a), the clausal complement
of after and prepositions like it would be headed by a silent counterpart of the complementizer
that, resulting in (76b) as the structure for after the war ended (for simplicity, the internal structure
of IP is omitted).

(76) a. PP b. PP

P′ P′

P◦ »CP« P◦ CP
after after
C◦ IP
∅that
the war ended

There are several empirical arguments for preferring the elementary tree in (76a) over the
one in (75). First, at least one preposition in English allows—indeed, requires—CP complements
headed by an overt complementizer, as shown in (77).

(77) They differ in *(that) they hold sharply opposing views on educational reform.

A second reason for preferring (76a) over (75) is that sentences like (78), with an overt
complementizer, occurred freely in Middle English (and are still acceptable for some speakers of
Modern English). Some naturally-occurring examples are given in (79).

Extra Info
The thorn character (þ) was borrowed from Old Norse (it is still used in Icelandic) and
used in Old and Middle English where we use ‘th’ today. The yogh character (ȝ) was used
in Middle English where we use ‘g’ or ‘y’.

(78) * { after, before, since } that the war ended

(79) a. And after þat þis bataile was done, þe Britons assemblede ham
‘and after this battle was over, the Britons assembled (themselves)’
(PPCME2, cmbrut3,100.3080)
Chapter 5: Extending the X′ schema 165

b. Ȝit bifore that Dauith cam to Jerusalem, a new debate roos bitwixe the men of Israel
and the men of Juda
‘Yet before David came to Jerusalem, a new debate arose between the men of Israel
and the men of Judah’ (PPCME2, cmpurvey,I,11.415)
c. Now, sith that i have toold yow of which folk ye sholde been conseilled, now wol I
teche yow which conseil ye oghte to eschewe.
‘Now, since I have told you what kind of people you should be counseled by, now I
will teach you which advice you ought to eschew’ (PPCME2, cmctmeli,223.C1.247)

Finally, analyzing clausal complements of prepositions as CPs allows us to treat preposi-

tions in English in the same way as prepositions in other languages such as French, where it is
clear that the clausal complements are CP complements.5

(80) a. ✔ {après, depuis, pendant} la danse

after since during the dance
‘{ after, since, during } the dance’
b. ✔ {après, depuis} que l’ enfant a dansé; pendant que l’ enfant dansait
after since that the child has danced while that the child was.dancing
‘{ after, since } the child danced; while the child was dancing’
c. * { après, depuis } l’enfant a dansé; pendant l’enfant dansait

Crosslinguistic variation in headedness 4

4.1 Head-final vs. head-initial patterns

As illustrated in (81)–(87), heads in English precede their complements, and English is therefore
said to be a head-initial language. The headedness of a lexical item (or a category or an entire
language) always refers to the order of heads and complements. In other words, headedness
is determined with respect to the intermediate projection of elementary trees, not with respect
to the maximal projection. For instance, English determiners can be medial in their maximal
projection (the possessive morpheme ’s must be preceded by a DP in Spec(DP)), and English verbs
and modals must be medial in their maximal projections, but they all count as head-initial because
they are the leftmost elements in the intermediate projections of their elementary trees.

(81) English head-initial verbs

a. They [V′ pursued [DP their goal ] ].
b. She [V′ submitted [DP her application ] ].

(82) English head-initial Infl

Chapter 5: Extending the X′ schema 166

a. They [I′ should [VP pursue their goal ] ].

b. She [I′ could [VP submit her application ] ].

(83) English head-initial complementizers

a. They agreed [C′ that [IP they should pursue their goal ] ].
b. She wondered [C′ if [I′ she could submit her application ] ].

(84) English nouns precede their complements

a. the [N′ pursuit [PP of their goal ] ]
b. the [N′ submission [PP of her application ] ]
c. Lisa’s [N′ pride [PP in her work ] ]

(85) English determiners precede associated NPs

a. [D′ the [NP pursuit of their goal ] ]
b. [D′ the [NP submission of her application ] ]
c. Lisa [D′ ’s [NP pride in her work ] ]

(86) English adjectives precede their complements

a. She is [A′ proud [PP of her work ] ].
b. He is [A′ fond [PP of his children ] ].

(87) English prepositions precede their complements

a. [P′ over [DP the next five years ] ]
b. [P′ with [DP great fanfare ] ]

Universal Grammar by no means prescribes head-initial phrase structure. Rather, many

languages exhibit consistently head-final phrase structure; two such languages are Japanese
and Korean. The examples in (88)–(93) are from Korean. In order to avoid using ‘head-final
preposition’, which is an etymological contradiction in terms, linguists have coined the term
postposition for the Ps in (93). The term adposition is a cover term for prepositions and post-
positions (that is, for Ps regardless of headedness). Examples for I and D are missing because
Korean has neither overt modals of the English sort nor overt articles; the abbreviations in the
glosses are explained in the notes,6 but are not crucial for present purposes.
(88) Korean head-final verb phrases
a. 그들 -은 목적 -을 추구하 -였 -다
kutul -un [V′ [DP mokcek -ul ] chukwuha -yess -ta ] .
they -top goal -acc pursue -pst -decl
‘They pursued their goal.’

b. 그 -는 지원서 -를 제출하 -였 -다
ku -nun [V′ [DP ciwonse -lul ] ceychwulha -yess -ta ] .
3.sg -top application -acc submit -pst -decl
‘They.sg submitted their.sg application.’
Chapter 5: Extending the X′ schema 167

(89) Korean head-final complementizer

그들 -은 목적 -을 추구해야 한 다고 통의하 -였 -다.
[C′ [IP kutul -un mokcek -ul chukwuhayya ha-n ] tako ] tonguyha -yess -ta.
they -top goal -acc pursue must-pRs that agree -pst -decl
‘They agreed that they should pursue their goal.’

(90) Another Korean head-final complementizer

그 -는 지원서 -를 제출해도 되 는지 궁금하 -였 -다.
ku -nun ciwonse -lul ceychwulhayto toy nunci kwungkumha -yess -ta
3.sg -top application -acc submit be-able if wonder -pst -decl
‘They.sg wondered if they.sg could submit their.sg application.’

(91) Korean nouns following complements

a. 그들 -의 목적의 추구
kutul -uy [N′ [DP mokcek -uy ] chukwu ]
they -gen goal.gen pursuit
‘their pursuit of their goal’
b. 그 -의 지원서 -의 제출
ku -uy [N′ [DP ciwonse -uy ] ceychwul
3.sg -gen application -gen submission
‘their.sg submission of their.sg application’
c. 경림 -의 일 -에 대한 자부심
kyenglim -uy [N′ [PP il -ey tayhan ] capwusim ]
Kyenglim -gen work -in regarding pride
‘Kyenglim’s pride in their.sg work’

(92) Korean head-final PPs

a. 다음 오 년 동안
[P′ [DP taum o nyen ] tongan ]
next five years over
‘over the next five years’
b. 대 팡파르 -와 함께
[P′ [DP tay phangphalu -wa ] hamkkey ]
big fanfare-with with
‘with big fanfare’

(93) Korean adjectives following head nouns

a. 일 -이 자랑스러 -운 사람
[A′ [DP il -i ] calangsule -un ] saram
work -nom proud -mod person
‘a person proud of their.sg work’
Chapter 5: Extending the X′ schema 168

b. 아이들 -이 좋 -은 사람
[A′ [DP aitul -i ] coh -un ] saram
children -nom fond -mod person
‘a person fond of their.sg children’

4.2 Harmonic and disharmonic headedness

Languages tend to be harmonic with respect to headedness; that is, they tend to be consistently
head-initial or head-final across syntactic categories. This is what is exemplified by the English
and Korean examples above, and it is very common for languages to pattern like this. However,
in certain languages, some syntactic categories project head-initial trees and others project head-
final ones.

Mixed headedness in Tunen

Such mixed phrase structure is found, for instance, in many languages in West Africa. One ex-
ample is Tunen, a Bantu language spoken in Cameroon (Zone A).7

(94) Context: You enter the room and see a broken window. Someone announces…
Biə́ lɛ a ná [VP ɛtúbə́ sana ] Tunen
1.Pierre sbj.agR pst window break
‘Pierre broke the window.’ (Kerr 2023)

Even this single example gives a very good illustration of mixed-headedness. We have bracketed
the verb phrase in the example above. If we assume that the inflection of past tense emerges on
Infl, we can see that the verb phrase follows tense: that is, Infl is head-initial, where the head
(bearing tense) precedes its complement (the verb phrase).

(95) IP

DPi I′

Biə́ lɛ I◦ VP
a ná
t i ɛtúbə́ sana

At the same time, the story is different inside the verb phrase. Clearly in (94), the head (the
verb sana ‘break’) follows its complement (the object ɛtúbə́ ‘window’). The Tunen verb phrase
therefore appears to be head-final:

(96) VP

ti V′

DP V◦
sana
ɛtúbə́
Chapter 5: Extending the X′ schema 169

Future tense illustrates the same pattern, with the future tense morpheme preceding its VP com-
plement.

(97) Samuɛ́lɛ a ŋɔ [VP ɔ-lɛ́sa nɛ́á-aka ] Tunen

1.Samuel agR fut 3-rice eat-duR
‘Samuel will eat rice tomorrow.’ (this future tense is the ‘tomorrow’ future)
(Kerr 2023)

Both demonstratives and prepositions precede their respective complements, so we see

Dem-N word order (98) and P-DP word order (99).

(98) tɔ́ɔ́yɛ tɔ-banána tɔ-tɛ̂tɛ́á tɔ-fítitiə tɔ-ʜ fandɛ

dem nc-banana agR-small agR-black agR-two
‘these two small black bananas’ (Kerr 2023)

(99) Context: Where are you?

mɛ lɛ́á [PP ɔ nɛ-oní ]
agR be pRep nc-market
‘I am at the market.’ (Kerr 2023)

Head nouns, however, tend to precede modifiers of those nouns, as we can see in (98).
Head nouns likewise precede their complements in Tunen, distinct from verbs following their
complements.

(100) Context: “They - they are - I see three people here.”

mmh bá lɛ́á ɔ ɛ-ndínə́ yɛ bɔ-lɛa
mmh agR be pRep nc-foot assoc nc-tree
‘Mmh. They’re at the foot of the tree.’ (Elizabeth Kerr, personal communication)

Complement clauses introduce some additional complexity. Inside the complement clause,
we see the familiar head-initial Infl and head-final VP. But in the main clause, the CP comple-
ment of manya ‘know’ follows the verb, in an exception to the head-final VP pattern we saw
above.
ʜ
(101) mɛ ndɔ manya [CP ɔwá Matɛ́ŋɛ a ka hɛ-əfulə fana-aka ]
sm.1sg pRes know comp Martin agR past book read-duR
‘I know that Martin has read the book.’ (Kerr 2023)

And in the embedded CP itself (bracketed in (101)), we see that the complementizer precedes the
InflP it is associated with, suggesting that CP itself is head initial, in contrast to Korean CPs in
(89) and (90) above.
Tunen shows much more complexity in word order based on what is being emphasized and
de-emphasized in a sentence (as is the case for many languages). But as an initial description, it
is clear that headedness need not be completely harmonic among all categories in a language. It
is certainly common to see patterns like Korean or English that are relatively consistent in their
headedness, but it is also not surprising among human languages to find patterns like in Tunen,
where some projections are head-initial but others are head-final.
Chapter 5: Extending the X′ schema 170

Mixed headedness in Germanic languages

Two other languages with mixed headedness are Dutch and German. The examples in (102)–(107)
are from German. (The reason that they are all subordinate clauses is that finite verbs in main
clauses move away from their basic position in German (and indeed in all Germanic languages
other than modern English); we discuss this so-called verb-second phenomenon in Chapter 12.)
As the examples show, V and A are head-final in German, whereas N, P, D, and C are head-initial.
I is missing from the examples because German lacks modals of the English type, so that the
position of I is difficult to establish.

(102) V
a. dass sie [V′ [DP ihr Ziel ] verfolgten ]
that they their goal pursued
‘that they pursued their goal’
b. ob sie [V′ [DP ihre Bewerbung ] einreichte ]
if she her application submitted
‘if she submitted her application’

(103) A
a. [A′ [DP seinen Prinzipien ] treu ]
his principles loyal
‘loyal to his principles’
b. [A′ [PP auf die Kinder ] stolz ]
on the children proud
‘proud of the children’

(104) N
a. die [N′ Verfolgung [DP ihr-es Ziel-s ]]
the pursuit their-gen goal-gen
‘the pursuit of their goal’
b. diese [N′ Treue [PP zu seinen Prinzipien ] ]
this loyalty to his principles
‘this loyalty to his principles’

(105) P
a. [P′ über [DP die nächsten fünf Jahre ] ]
over the next five years
‘over the next five years’
b. [P′ mit [DP grossem Trara ] ]
with great fanfare
‘with great fanfare’

(106) D
Chapter 5: Extending the X′ schema 171

a. [D′ die [NP Verfolgung ihr-es Ziel-s ]]

the pursuit their-gen goal-gen
‘the pursuit of their goal’
b. [D′ diese [NP Treue zu seinen Prinzipien ] ]
this loyalty to his principles
‘this loyalty to his principles’

(107) C
a. [C′ dass [IP sie ihr Ziel verfolgten ] ]
that they their goal pursued
‘that they pursued their goal’
b. [C′ ob [IP sie ihre Bewerbung einreichte ] ]
if she her application submitted
‘if she submitted her application’

To complicate matters yet further, German allows postpositions, as in (108).

(108) [P′ [DP den Fluss ] entlang ]

the river along
‘along the river’

And finally, to really liven things up, certain adpositions in Dutch and German can either
precede or follow their complements. This is illustrated in (109) and (110), again for German; the
(a) and (b) examples share the same meaning.

(109) a. [P′ wegen [DP des Wetter-s ]]

because.of the.gen weather-gen
‘because of the weather’
b. [P′ [DP des Wetters ] wegen ]

(110) a. [P′ gegenüber [DP der Kirche ] ]

across.from the church
‘across from the church’
b. [P′ [DP der Kirche ] gegenüber ]

Dutch, too, allows such variation between head-initial and head-final adpositions, and in
that language, it is even accompanied by a systematic meaning difference. Specifically, when
adpositions with variable headedness are postpositions, their meaning is always directional, but
when they are prepositions, their meaning is generally locative. This is illustrated in (111) and
(112) (Kroch 1994).

(111) a. Ik fiets in de straat.

I bike in the street
‘I ride my bike in the street.’ (locative)
Chapter 5: Extending the X′ schema 172

b. Ik fiets de straat in.

I bike the street in
‘I ride my bike into the street.’ (directional)

(112) a. Ik klim in de boom.

I climb in the tree
‘I climb in(to) the tree.’ (locative or directional)
b. Ik klim de boom in.
I climb the tree in
‘I climb into the tree.’ (only directional)

In case the German and Dutch examples just discussed sound exotic, it is worth noting that
English sports two postpositions of its own, as illustrated in (113).8

(113) a. They searched the whole world over.

b. They work the whole week through.

It is not uncommon for languages to undergo phrase structure change. For instance, the
phrase structure of Old English (ca. 800–ca. 1100 C.E.) is reminiscent of that of modern German
and Dutch; in particular, verbs were head-final for most of the Old English period. (The three
languages are closely related historically, so the syntactic similarity is not surprising.) The first
instances of verb-initial phrase structure appeared in late Old English. Early Middle English was
characterized by rampantly variable headedness in the verb phrase (Kroch and Taylor 2000b), but
by ca. 1350, the change from head-final to head-initial verb phrases was essentially complete in all
dialects of Middle English. Since Chaucer lived from 1342 to 1400, his language is already modern
in this respect, though his syntax differs quite strikingly from that of the modern language in
other ways, as we will discuss in Chapters 10 and 12. In the modern language, only isolated
relics of the old verb-final phrase structure survive, like the saying Indictments do not a conviction
make.9

4.3 Final thought

In conclusion, we note that this section has focused on the order of heads and complements in the
X′ . Languages also display variation with respect to the order of specifiers and the intermediate
projection. It is very much more common for Spec(XP) to precedes its X′ than to follow. This
contrasts with the variation one level down, where head-final order prevails over head-initial
order by approximately 60% to 40%.
 Chapter 5: Extending the X′ schema 173

Notes 5
1. It is not just the expression of agent arguments that is freer in noun phrases than in sentences. As the contrast
between (i) and (ii) shows, the same is true of theme arguments.

(i) a. ✔ The mills employed thousands; their practices damaged the environment.
b. * The mills employed; their practices damaged.
(ii) a. ✔ an employer of thousands; the damage to the environment
b. ✔ an employer; the damage

This has potential consequences for the acquisition of one substitution, as we discuss later on in the chapter.

2. Notice also the related contrast in (i); the construction in (i.a)—so-called raising—is discussed in Chapter 9.

(i) a. ✔ The manuscript appears to have been found.

b. * the manuscript’s appearance to have been found.

3. More evidence for the idiosyncratic character of the constraint against (52a) comes from the acceptability of (5) (at
least in formal registers) (thanks for Sonali Mishra for drawing our attention to such examples).

(i) such a one

4. Strictly speaking, this statement is true only of fond in predicative position, not in prenominal position.

(i) Predicative: *Their parents are fond.

(ii) Prenominal: ✔their fond parents

5. In (78), some French speakers prefer or require the subjunctive form of the auxiliary (ait) rather than the indicative
form (a). For present purposes, this variation, which is comparable to that found in English between If I was a rich
man and If I were a rich man, is irrelevant.

6. acc = accusative (case of direct object),

gen = genitive (possessive case),
decl = declarative clause,
mod = modifier,
pRs = present tense,
ps = person,
sg = singular,
top = topic of sentence.

7. The data discussed in this section come from Kerr (2023), with the foundational work on Tunen coming from Dugast
(1971) and Mous (1997), Mous (2003), Mous (2005), and Mous (2014). Additional overview discussion of the word
order of languages of this region can be found in Sande, Baier, and Jenks (2019). Glosses have been simplified for
expository purposes; see original sources for more precise glossing. The pattern reported here was first addressed
in the generative literature by Hilda Koopman for Vata and other Kru languages (Koopman 1984).

8. A further apparent instance of a postposition in English is ago. Although its synchronic status is unclear, ago did
not originate as a preposition.
 Chapter 5: Extending the X′ schema 174

9. Cf. the parallel token in (i).

(i) There had appeared to him something rather fine in his policy of refusing to identify himself in any way with
Sedleigh, a touch of the stone-walls-do-not-a-prison-make sort of thing.
(P.G. Wodehouse. 1974. The world of Psmith. London: Barrie & Jenkins. 114.)

Exercises and problems 6

Exercise 5.1
A. What is the syntactic difference between mainstream U.S. English them and the variant
illustrated in (1), which is a part of many varieties of English?

(1) This is definitely one of them jobs, man, if you’re one of them worriers…
(Overheard on the southwest corner of 34th Street and Walnut Street, Philadelphia,
PA, 31 August 1999)

B. Can you think of obligatorily transitive adjectives other than fond?

C. Can you think of plausible candidates for obligatorily intransitive prepositions?
D. What is the syntactic difference between the prepositions in (2a) and (2b)? A syntactic
difference, not a semantic one.

(2) a. at, despite, from, of

b. along, besides, between, by, under

E. Does with belong with the prepositions in (2a) or in (2b)? (For fun, you might ask a few of
your friends whether they agree with you. Make sure to pick native speakers of English!

Exercise 5.2
Formally, the structures in (1) are consistent with X′ theory. Empirically, however, they are
unsatisfactory representations because they are inconsistent with certain linguistic judgments.
What are those judgments?

Take Note

In the evidence that you provide, you should feel free to replace the with other determiners
and plural nouns with singular nouns.
Chapter 5: Extending the X′ schema 175

(1) a. DP b. PP

D′ P′

D′ PP P′ PP

D◦ NP P′ P◦ DP P′
the on
N′ P◦ DP D′ P◦ DP
on with
N◦ D′ D◦ NP D′
houses the
D◦ NP N′ D◦ NP
the a
N′ N◦ N′
corner
N◦ N◦
corner sign

Exercise 5.3
A. Build structures for the (a) examples in (1)–(3). The other examples are there to guide you;
you’re not being asked to build separate structures for them.

(1) a. the monster’s mother’s lair

b. the monster
c. the monster’s mother

(2) a. the hero of the poem’s name

b. the hero
c. the hero of the poem

(3) a. the mother of the monster’s dislike of the poem’s hero

b. the monster
c. the mother of the monster
d. the poem
e. the poem’s hero

B. Build structures for the noun phrases in (4).

(4) a. yesterday’s lecture

b. this week’s unseasonably high temperatures

C. Build structures for the noun phrases in (5), and indicate the recursive nodes. (Treat my as
I plus ’s.)
Chapter 5: Extending the X′ schema 176

(5) a. the neighbor’s doctor’s dog

b. my father’s mother’s watch

Exercise 5.4
A. What is the syntactic category of hiring in (1a) and (1b)? Provide evidence for your con-
clusions.

(1) a. We dislike Jess’s impulsive hiring of incompetents.

b. We dislike Jess’s impulsively hiring incompetents.

B. Build structures for the underlined gerund phrases in (1). Give the full structure for incom-
petents, but feel free to use triangle notation for the proper noun Jess. If you’re curious
about the internal structure, see §1.2 of Chapter 2.

Exercise 5.5
A. The noun phrase in (1) has two distinct interpretations, which can be paraphrased as in (2).

(1) the houses on the corner with a sign

(2) a. the houses on the corner that have a sign

b. the houses on the corner that has a sign

Build two structures for (1), clearly indicating which structure goes with which interpreta-
tion in (2).
B. Give paraphrases for (or otherwise distinguish between) the two interpretations available
for (3), and then build the relevant structures, clearly indicating which structure goes with
which interpretation.

(3) I enthusiastically recommend this candidate with no qualifications.

C. Repeat (B) for (4).

(4) the mother of the bride’s father

Exercise 5.6
Thanks to Amy Forsyth for the template for the examples in Exercise 5.6.
A. The sentences in (1) are many-ways ambiguous. Find as many interpretations as you can,
clearly describing the relevant situations you have in mind (see (2) for example descrip-
tions). Build trees for each interpretation that you find, clearly indicating which tree is
associated with which interpretation.
Hint: Work backwards. First construct possible trees, and then figure out the interpreta-
tion from the syntactic structure.
Chapter 5: Extending the X′ schema 177

(1) a. The trainer tapped the seal with the ball on its nose.
b. The woman chased the cat with the mouse on the bicycle.

B. Many-ways ambiguous though the sentences in (1) are, they cannot be used to describe the
situations in (2), respectively.

(2) a. There is a seal balancing on its nose, and the trainer uses a ball to tap this seal.
b. There is a cat riding a bicycle, and the woman is using a mouse to chase this cat.

Exercise 5.7
This exercise harks back to Exercise 1.6 from Chapter 1. Relying on your now more sophisticated
knowledge of phrase structure, build structures for the possible interpretations of the sentences
in (1).

(1) a. Charis will answer the question precisely at noon.

b. Laila will watch the man from across the street.
c. They should decide if they will come tomorrow.

Exercise 5.8
Build subtrees for the matrix and the subordinate clauses in (1), and indicate how they fit to-
gether.

(1) a. The students will solve the problem, though it seems difficult.
b. The students will solve the problem, though we acknowledge that it seems difficult.

Exercise 5.9
As noted at the end of this chapter, English has undergone a phrase structure change in the course
of its history. In early Old English, V and I were both consistently head-final. Over the course
of Old English, I became head-initial. In other words, in addition to old structures in which I
followed VP, new ones became available in which I preceded VP. By the beginning of Middle
English (about 1100), I had become exclusively head-initial, but V continued to be variably head-
final or head-initial. Finally, by about 1350, V had become consistently head-initial.
Given this historical sketch, build trees for all of the phrase structure variants of (1) that
were possible during the course of the history of English. (For the non-modern stages, just use
modern vocabulary items, but arranged according to the relevant parameter settings.)

Take Note

For simplicity, assume that the elementary trees for the nouns in the sentences below have
no substitution nodes.

Ignore the internal structure of proper nouns; just represent them using triangle notation.
If you are curious about the internal structure, see §1.2 of Chapter 2.
Chapter 5: Extending the X′ schema 178

(1) a. Beowulf will slay Grendel.

b. The hero will slay the monster.
c. The hero of the poem will slay the monster’s mother.

Exercise 5.10
This exercise assumes that you are familiar with the terms ‘relative clause’ and ‘noun complement
clause’. If not, see the end of the exercise for a tutorial.
A. Provide evidence whether relative clauses, illustrated in (1), are arguments or adjuncts of
the noun they modify.

(1) Relative clauses

a. The idea that Columbus was working with was incorrect.
b. The fact that they have discovered is important.

B. Subordinate clauses of the type illustrated in (2) are traditionally called noun complement
clauses.

(2) Noun complement clauses

a. The idea that Columbus was the first European to discover America is incorrect.
b. The fact that they are wrong is lost on them.

Provide evidence whether noun complement clauses are syntactic arguments of the noun or ad-
juncts. In other words, given the way that the term ‘complement’ is used in X′ theory, is the term
‘noun complement clause’ for these clauses a misnomer?
Chapter 5: Extending the X′ schema 179

Background Info

If you are not familiar with the distinction between noun complement clauses and that
relative clauses, here are two simple diagnostics for distinguishing between them. Our
baseline examples of noun complement clauses and relative clauses are listed here:

(i) Relative clauses

(3) The idea that Columbus was working with was incorrect.
(4) The fact that they have discovered is important.

(ii) Noun complement clauses

(5) The idea that Columbus was the first European to discover America is in-
correct.
(6) The fact that they are wrong is lost on them.

▶ First, stripping away the complementizer that leaves a complete sentence in the case
of a noun complement clause, but something incomplete in the case of a relative
clause (it feels like there is a gap, as indicated by the underlining).

(iii) a. ✔ Columbus was the first European to discover America.

b. ✔ They are wrong.

(iv) a. * Columbus was working with .

b. * They have discovered .

▶ Second, the complementizer that can generally be replaced by a wh- word in a relative
clause, but never in a noun complement clause.

(v) a. * The idea which Columbus was the first European to discover America is
incorrect.
b. * The fact which they are wrong is lost on them.

(vi) a. ✔ The idea which Columbus was working with was incorrect.
b. ✔ The fact which they have discovered is important.
Chapter 5: Extending the X′ schema 180

Problem 5.1
If a given string is structurally ambiguous (that is, it has more than possible structural represen-
tation), is it necessarily associated with more than one meaning? Explain, giving examples.
Hint: Exercise 5.5, B.

Problem 5.2
The structures in (1) are intended to represent the second conjunct in (2). def in (1c) stands for
‘definite’.

(1) a. DP b. DP

D′ D′

D′ PP D′ NP

D◦ on the shelf D◦ N′
that that
N′ PP

N◦ on the shelf
∅one
c. DP

D′

D′ NP

D◦ N′
[def]
N′ PP

N◦ on the shelf
that

(2) the book on the table, and that on the shelf

Discuss the relative merits of the three structures in (1). In other words, what considerations
(whether empirical or conceptual) make each of the structures attractive or unattractive?
Chapter 5: Extending the X′ schema 181

Problem 5.3
In (55) of Chapter 1, we mentioned an alternative approach to noun phrase structure than the one
presented in this chapter. According to the alternative approach (updated to be in accordance
with the X′ schema), all determiners are intransitive, and all nouns have substitution nodes in
the specifier position for (possibly silent) determiners, as shown in (1).

(1) a. DP b. NP c. NP

D′ »DP« N′ »DP« N′

D◦ N◦ N◦ »XP«

Discuss the relative merits of this approach compared to the one presented in the text. In
other words, what considerations (whether empirical or conceptual) would lead you to adopt or
reject this alternative approach to noun phrase structure?

Problem 5.4
Wouldn’t it be simpler to say that determiners (and possessive pronouns) are modifiers, just like
adjectives? Then we could continue say that noun phrases are the maximal projections of nouns,
just like in the olden days. What do you think?

Problem 5.5
For all speakers of English, the sentence in (1) can have either of the interpretations in (2).

(1) Jane has a big black dog, and Jean has a small one.

(2) a. Jane has a big black dog, and Jean has a small dog.
b. Jane has a big black dog, and Jean has a small black dog.

On the other hand, (3) means only (4a) for most speakers of English. However, some speak-
ers are able to interpret (3) as (4b) (Radford 1988) (such variable judgments among different speak-
ers are conventionally indicated by a percent symbol.

(3) Jane has a big black dog, and Jean has a brown one.

(4) a. Jane has a big black dog, and Jean has a brown dog.
b. % Jane has a big black dog, and Jean has a big brown dog.
A. Which of the interpretations of (1) and (3) is problematic? Explain.
B. Can you think of a way of resolving the problem you laid out in your answer to (A)?
 6
Wh- movement

Chapter outline
1 Basic descriptive terminology for types of questions . . . . . . . . . . . . . . . . . 183
1.1 Yes-no versus wh- questions . . . . . . . . . . . . . . . . . . . . . . . . . . 183
1.2 Direct versus indirect questions . . . . . . . . . . . . . . . . . . . . . . . . 184
1.3 Information questions versus echo questions . . . . . . . . . . . . . . . . 185
2 A movement analysis of questions . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
2.1 Indirect questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
2.2 Direct questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
3 Topicalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
4 Relative clauses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
4.1 Wh- relative clauses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
4.2 That relative clauses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
4.3 Zero relative clauses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
4.4 Doubly marked relative clauses . . . . . . . . . . . . . . . . . . . . . . . . 201
4.5 Stylistic considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
4.6 A historical note . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
5 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
6 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204

In this chapter, we discuss the syntactic structure of questions, a sentence type that we in-
troduced in Chapter 2 as a test for constituenthood. We will pursue an insight already mentioned
there—namely, that forming wh- questions is conveniently understood as a combination of re-
placing a constituent by a suitable wh- phrase and moving it to the beginning of the sentence.
For obvious reasons, we call this movement wh- movement. The first section of the chapter
gives a basic introduction to the syntax of questions. The next section presents the syntactic
derivation of wh- questions, which extends straightforwardly to yes-no questions. The second
section then extends the analysis of questions yet further to another sentence type from Chap-
ter 2, which we there called movement, but which we will refer to from now on as topicalization
in order to distinguish it from other instances of movement. The final section covers the various
types of relative clauses found in English. As we will show, each of these three phenomena
Chapter 6: Wh- movement 183

involves movement of a constituent to Spec(CP). In the case of questions and relative clauses, the
constituent is a wh- phrase; in the topicalization, it is not. Despite this difference, the syntax lit-
erature often follows Chomsky (1977) in using the term wh- movement to include topicalization
in order to highlight the shared properties of all three phenomena, and we will follow that usage.
If need be, it is always possible to distinguish between wh- movement in the wide sense (where
it includes topicalization) versus in the narrow sense (where it does not).

Basic descriptive terminology for types of questions 1

1.1 Yes-no versus wh- questions

Questions can be divided into yes-no questions (also known as polar questions) and wh- ques-
tions (also known as constituent questions), according to the expected answer. As the name
implies, the answer to a yes-no question is either ‘yes’ or ‘no’. And again, as the name implies,
wh- questions begin with a wh- expression (who, what, which, where, when, why). How counts as
a wh- expression by virtue of its meaning, even though it doesn’t begin with wh-.
The distinction between the two question types is illustrated in (1) and (2).

(1) Yes-no question:

Q: Has he called?
A: { Yes, no }.
(2) Wh- question:
a. Q: Who just came in?
A: The boy from next door.
b. Q: Who(m) did you invite?
A: All my friends.
c. Q: When did she call?
A: After dinner.
d. Q: Why did he do that?
A: Out of ignorance.
e. Q: How did you fix it?
A: With the right tool.
Chapter 6: Wh- movement 184

1.2 Direct versus indirect questions

Another distinction that can be drawn is between direct questions and indirect questions.
Direct questions are main clauses, whereas indirect questions are part of a larger matrix sentence
(possibly a question itself). Direct questions are generally used to elicit information. They are as-
sociated with characteristic intonation contours, which are represented in standard orthography
by a question mark. Indirect questions are generally used to report about direct questions and
are not associated with a special intonation.
The questions in (1) and (2) are all direct questions. The corresponding indirect questions
are given in (3) and (4). Here and in what follows, indirect questions are enclosed in square
brackets.

(3) Indirect yes-no question:

I can’t remember [ { whether, if } he has called ].
(4) Indirect wh- question:
a. I can’t remember [ who just came in ].
b. I can’t remember [ who(m) you invited ].
c. I can’t remember [ when she called ].
d. I can’t remember [ why he did that ].
e. I can’t remember [ how you fixed it ].

(5) gives examples of various syntactic contexts in which indirect questions occur.

(5) a. Complement of verb:

She asked [ whether they are coming ].
b. Complement of adjective:
I’m not sure [ whether they are coming ].
c. Complement of preposition:
The question of [ whether they are coming ] remains unresolved.
d. Subject:
[ Whether they are coming ] remains up in the air.

Finally, indirect questions can be finite or nonfinite, as shown in (6) and (7) (see §8 of
Chapter 2 on the nature of finiteness). Notice that if, in contrast to whether, requires finite com-
plements.

(6) Indirect yes-no question:

a. They can’t remember [ { whether, if } they should turn off the lights ]. (finite)
b. They can’t remember [ whether to turn off the lights ]. (nonfinite)
(7) Indirect wh- question:
a. They can’t remember [ what they should pay attention to ]. (finite)
b. They can’t remember [ what to pay attention to ]. (nonfinite)
Chapter 6: Wh- movement 185

1.3 Information questions versus echo questions

Ordinarily, speakers use questions to elicit information, potentially initiating a discourse with
them, as in (8).

(8) a. How do you do?

b. What’s up?

But questions can also be used to signal a failure to understand the previous move in a con-
versation. The failure to understand can be genuine or feigned (calculated to express surprise,
disapproval, outrage, and so on). Accordingly, we can distinguish between information ques-
tions and echo questions (also known as reprise questions); the latter two terms underline the
response character of the second type of question. Echo questions can have the same syntactic
form as information questions, but they are associated with a melody that is quite distinct from
that of information questions. Speakers can also further mark the special discourse function of
echo questions by giving them a special syntactic form in which the wh- phrase does not undergo
wh- movement. The wh- phrase is said to remain in situ. (9) and (10) illustrates the two forms
that echo questions can take (with or without wh- movement).

(9) A: Over break, I ended up visiting my [unintelligible]

B: Who did you end up visiting? (wh- movement)
You ended up visiting who? (wh- in situ)
(10) A: Her parents burned all of her clothes.
B: What did they do⁈ (wh- movement)
They did what⁈ (wh- in situ)

The association of the wh- in situ form with the echo function is not universal. Some lan-
guages do not have wh- movement at all (see Chapter 14 for discussion), and some wh- movement
languages allow wh- in situ questions to serve as either information questions or echo questions.
This is illustrated for French in (11) (Engdahl 2006, p. 104, (33)–(34)).

(11) a. A: Ton fils, il lit quoi? (information question)

your son he reads what
‘What does your son read?’
B: Des bandes desinées.
of.the comics
‘Comics.’
b. A: Mon fils, il lit [inaudible]?
my son he reads
‘My son reads …’
B: Il lit quoi? (echo question)
he reads what
‘He reads what?’
Chapter 6: Wh- movement 186

A movement analysis of questions 2

2.1 Indirect questions

We begin our analysis of questions by considering indirect questions. This may seem counter-
intuitive, but it is actually a more direct path to understanding the structure of questions than if
we were to begin with direct questions. As the presence of the complementizer if in (12) shows,
the verb wonder takes a CP complement.

(12) They wonder if the lions will devour the zebra.

The elementary trees for wonder and if are given in (13a)–(13b), and the entire tree for (12) is
given in (13c).

(13) a. VP b. CP

»DP« V′ C′

V◦ »CP« C◦ »IP«
wonder if
Chapter 6: Wh- movement 187

c. IP

DPi I′

they I◦ VP
[pRs]
ti V′

V◦ CP
wonder
C′

C◦ IP
if
DPk I′

the lions I◦ VP
will
tk V′

V◦ DP
devour
the zebra

Now consider the indirect question in (14), which begins with a wh- phrase (a maximal
projection) rather than with a complementizer (a head).

(14) They wonder which zebra the lions will devour.

Let’s adopt the null hypothesis that wonder is associated with the same elementary tree in (14) as it
is in (12)—namely, with (13a). Since (14) contains no overt complementizer, the CP that substitutes
into the complement node of wonder must then be the projection of a silent complementizer. For
reasons to be given shortly, we take this complementizer to be a silent counterpart of that. In
deriving the tree for (14), a further difficulty arises concerning the wh- phrase which zebra. On the
one hand, the wh- phrase must be the object of devour, just as in (12), because devour is obligatorily
transitive. But on the other hand, the wh- phrase precedes the subject of the subordinate clause
rather than following the verb. As usual when we are confronted with a mismatch of this sort,
we invoke movement in order to allow a single phrase to simultaneously play more than one role
in a sentence. Specifically, we will first substitute the wh- phrase into the complement node of
wonder and then move it to Spec(CP). This allows us to accommodate the word order in (14), while
maintaining the status of devour as a transitive verb regardless of which clause type (declarative or
interrogative) it happens to occur in. The requisite elementary tree for the silent complementizer
is shown in (15); by contrast to the elementary tree for if, it contains a specifier node for the wh-
phrase. The full structure of the wh- phrase that moves to that specifier node is shown in (15b).
The resulting structure for (14) is shown in (15c).
Chapter 6: Wh- movement 188

(15) a. CP b. DP

»XP« C′ D′

C◦ »IP« D◦ NP
∅that which
N′

N◦
zebra
c. IP

DPi I′

they I◦ VP
[pRs]
ti V′

V◦ CP
wonder
DPk C′

which zebra C◦ IP
∅that
DPm I′

the lions I◦ VP
will
tm V′

V◦ tk
devour
The argument just presented is based on the obligatorily transitive character of devour,
but it extends straightforwardly to other syntactic relations—for instance, modification. Recall
that the way that we have chosen to represent the modification relation is to adjoin the modifier
at the intermediate projection of the head that is being modified. In (16a), the adverb phrase
unbelievably quickly modifies the verb devour, and so it adjoins at V′ . In (16b), the corresponding
wh- phrase how quickly also modifies devour, so it needs to adjoin to V′ as well, but it precedes
the subject.

(16) a. The lions will devour the zebra unbelievably quickly.

b. They wonder how quickly the lions will devour the zebra.

Again, we resolve the mismatch between the position where the phrase is interpreted and where
Chapter 6: Wh- movement 189

it is pronounced by first adjoining the modifier and then moving it to Spec(CP). (17a) shows the
full structure of the phrase that undergoes movement, and (17b) shows the structure of the entire
question.

(17) a. AdvP

Adv′

AdvP Adv′

Adv′ Adv◦
quickly
Adv◦
how
b. IP

DPi I′

they I◦ VP
[pRs]
ti V′

V◦ CP
wonder
AdvPk C′

how C◦ IP
quickly ∅that
DPm I′

the I◦ VP
lions will
tm V′

V′ tk

V◦ DP
devour
the
zebra
Chapter 6: Wh- movement 190

Why a silent complementizer?

Let’s delve a bit deeper into why we treat the complementizer in (15) and in (17) as a silent coun-
terpart of that. There are several reasons. First, Middle English (1150–1500) routinely allowed
(though it did not require) overt that as the head of indirect wh- questions. The examples in (18)
are all from texts by Chaucer that are part of the Penn-Helsinki Parsed Corpus of Middle English;
(18c)–(18d) show that that alternated with its silent counterpart even within a single sentence by
a single author.

(18) a. he wiste wel hymself what that he wolde answere (cmctmeli.m3, 219.C1.75)
‘he himself knew well what he would answer’
b. for ye han ful ofte assayed … how wel that I kan hyde and hele thynges (cmctmeli.m3,
221.C1.149)
‘for you have very often determined how well … I can hide and conceal things’
c. I wolde fayn knowe how that ye understonde thilke wordes and what is youre sentence
(cmctmeli.m3, 227.C2.408)
‘I would like to know how you understand these same words and what your judgment
is’
d. And forther over, it is necessarie to understonde whennes that synnes spryngen, and
how they encreessen (cmctpars.m3, 296.C1b.355)
‘And moreover, it is necessary to understand where sins come from and how they
increase’
e. Now shal ye understonde in what manere that synne wexeth or encreesseth in man.
(cmctpars.m3, 297.C2.393)
‘Now you shall understand in what manner sin grows or increases in man.’
f. The fifthe circumstaunce is how manye tymes that he hath synned … and how ofte that
he hath falle. (cmctpars.m3, 323.C1.1502)
‘The fifth circumstance is how many times he has sinned … and how often he has
fallen.’

Second, contemporary Belfast English (the variety of English that is the focus of Exercise
1.2) resembles Middle English in this respect (Henry 1995, p. 107).

(19) a. I wonder which dish that they picked.

b. They didn’t know which model that we had discussed.

Third, wh- phrases followed by that continue to be attested in the unplanned usage of
speakers of modern Mainstream U.S. English (Radford 1988, p. 500). A few of the examples that
we have collected over the years are shown in (20); the entire collection is here.

(20) a. I realized how interesting that it was.

(Clara Orsitti, interviewed by Vicky Barker, World Update, National Public Radio, 25
January 1999)
b. Most of my colleagues were amazed how quickly that I recovered.
Chapter 6: Wh- movement 191

(advertisement for Temple University Hospital, WRTI, 24 November 1999)

c. “These recounts will determine how much of a pick-up that we will have”, said Demo-
cratic National Committee Chairman Joe Andrew.
(https://www.cnn.com/2000/ALLPOLITICS/stories/11/08/house.races/)

Finally, sequences of wh- phrase and overt complementizer in indirect questions occur in
languages other than English. The complementizer in question is generally the counterpart of
that, but the counterpart of if is attested as well.

(21) Bavarian (Bayer 1983, p. 212, (8a–d)):

a. I woass ned wer dass des toa hod.
I know not who that that done has
‘I don’t know who did that.’
b. … wos dass ma toa soin.
what that we do should
‘…what we should do.’
c. … wann dass da Xaver kumt.
when that the comes
‘…when Xaver is coming.’
d. … wiavui dass a kriagt.
how.much that he gets
‘…how much he gets.’
(22) Dutch (Besten 1989, p. 23, (21b)):
a. welk boek (of ) hij wil lezen
which book if he wants read
‘…which book he wants to read’

2.2 Direct questions

Direct wh- questions
Having argued that wh- phrases move to Spec(CP) in indirect questions, we assume for uniformity
that the same is true for direct questions like (23).

(23) a. [DP Which zebra ] will the lions devour?

b. [AdvP How quickly ] will the lions devour the zebra?
c. [AdjP How experienced ] should they be?
d. [PP Under which shell ] will they hide the pea?

As is evident from comparing the direct questions in (23) with their indirect question coun-
terparts, wh- movement in direct questions is accompanied by movement of the modal from its
ordinary position after the subject to a position immediately preceding the subject. For obvious
reasons, we will refer to this type of movement as head movement. The particular instance of
head movement at issue here is often called subject-aux inversion. From our point of view, the
Chapter 6: Wh- movement 192

term is slightly misleading since subject-aux inversion affects not just the auxiliary verbs be, do,
and have, but also modals. Nevertheless, we will sometimes use the term when it is useful for
expository purposes. In what follows, we will focus on questions with modals. Questions with
auxiliary and ordinary verbs involve complications that are not relevant here; see Chapter 10 and
especially Exercise 10.5.
We will begin by assuming that the CP involved in the derivation of direct questions is
headed by a morpheme expressing interrogative (= information-seeking) force. Despite the for-
mal (= structural) similarity between direct and indirect questions, indirect questions do not have
this same basic pragmatic force. In some languages, such as Japanese, the question morpheme
is overt, as illustrated in (24). (The question morpheme ka, glossed as “q”, appears in final posi-
tion because Japanese is a consistently head-final language, as noted in Chapter 5, § 4; the case
particles separated from their nouns by hyphens can be ignored for present purposes.)

(24) 誰が 寿司を 食べました か？

Dare-ga sushi-o tabe-mashi-ta ka?
who-nom sushi-acc eat-pol-pst q
‘Who ate sushi?’

In contrast to Japanese, the question morpheme in English is silent. This gives us (25a) as
the counterpart to if in (13b), and (25b) as the direct-question counterpart to the indirect question
in (13c), pending subject-aux inversion.

(25) a. CP

»XP« C′

C◦ »IP«
[q]
b. CP

DPk C′

which zebra C◦ IP
[q]
DPi I′

the lions I◦ VP
will
ti V′

V◦ tk
devour
Chapter 6: Wh- movement 193

Head movement
The structure in (25b) actually turns out to be a possible structure for direct questions in African-
American English, where subject-aux inversion in direct questions is optional. In that variety,
direct questions without subject-aux inversion still differ from the corresponding indirect ques-
tions in intonation; that is the contribution of the question morpheme. However, in mainstream
varieties of English, direct questions require inversion. In the two cases of movement that we
have considered so far (subject movement in Chapter 4 and wh- movement in this chapter), the
“landing site” of movement has been a substitution node in an elementary tree. But this option is
not available in the case of subject-aux inversion. The modal in I cannnot substitute in C, because
C is already filled by the question morpheme. We can’t simply allow the modal to overwrite the
question morpheme, since then the tree would not contain the information necessary to gener-
ate the proper question intonation. We also want to continue to say that elementary trees are
projections of morphemes rather than of syntactic categories. In other words, we do not want to
allow elementary trees as in (26), which consist solely of substitution nodes.

(26) CP

C′ ← not a possible elementary tree!

C◦ »IP«

As a result of these considerations, we will implement movement of the modal to the pre-
subject position via adjunction rather than with substitution. We use the same formal definition
of adjunction in Chapter 4, § 3. That is, we first by select the target of adjunction, then clone that
target, and finally attach the adjoining element (here, the modal undergoing movement) as the
daughter of the higher clone. (27) shows the operation on the relevant heads in isolation.

(27) a. C◦ b. C◦ c. C◦
[q]
C◦ I◦ C◦
[q] modal [q]

Select target of Clone target of Attach modal

adjunction adjunction

(28) shows the final structure for the direct question. Just like the moved phrases, the
moved head bears a unique index linking it to its trace.
Chapter 6: Wh- movement 194

(28) CP

DPi C′

which zebra C◦ IP

I◦k C◦ DPm I′
will [q]
the lions t k VP

tm V′

V◦ ti
devour

Although the formal operation of adjunction is defined in exactly the same way regardless
of the use to which we put it, the two uses are associated with differences. First and foremost,
the operation represents different linguistic relations in the two cases. As we already know from
Chapter 4, § 3, adjunction can represent the modification relation, which corresponds to a se-
mantic subset relation. In the case of head movement, adjunction represents an entirely different
linguistic relation. In this case, we can think of the structure formed by adjunction as a mor-
phologically complex head—or to put it more colloquially, as a compound word. In Chapter 10,
we will invoke head movement to combine verb stems with tense suffixes to form finite verbs.
In that case, the position of the tense suffix tells us that the verb left-adjoins to I. By analogy,
we assume that I in the case at hand left-adjoins also; in other words, we are treating the silent
question morpheme as a suffix, just like any other inflectional morpheme in English. Second,
the projection level of the target node differs; it is an intermediate projection for modification
versus a head for head movement. Third, the projection level of the adjoining element differs.
It is a maximal projection for modification versus a head for head movement. Finally, for mod-
ification, the adjoining element is not yet part of the structure, and adjunction is “pure” in the
sense that it doesn’t involve concomitant movement. For head movement, the adjoining element
is already part of the structure, and hence movement is involved. Table 6.1 summarizes the above
discussion.

Use of adjunction for … Modification Head movement

Adjunction structure represents … semantic restriction morphological relation
Target of adjunction intermediate projection head
Adjoining node maximal projection head
Adjoining node already part of structure? no yes
(= Movement involved?)

Table 6.1: Comparison of adjunction in modification and head movement

Chapter 6: Wh- movement 195

Yes-no questions
As we have already seen, the elementary trees required to derive indirect questions can either
include a specifier, as in (15a), or not, as in (13b). As expected given the generality of the X′
schema, the same is true of the elementary trees for direct questions. In particular, the question
morpheme in (25a) has a variant without a specifier, as shown in (29).

(29) CP

C′

C◦ »IP«
[q]

Using this elementary tree allows us to derive yes-no questions like (30), whose structure is shown
in (31).

(30) Will the lions devour the zebra?

(31) CP

C′

C◦ IP

I◦i C◦ DPk I′
will [q]
the lions t i VP

tk V′

V◦ DP
devour
the zebra

The direct question in (30) corresponds to the indirect yes-no question in (12) that we began
our investigation with. So we now have structures for all four question types that result from
crossing direct versus indirect questions with wh- versus yes-no questions.
In concluding this section, we note that traditional grammar distinguishes three types of
illocutionary force: declarative (statements), interrogative (questions), and imperative (com-
mands). The word order of English questions forces us to add a CP projection to the IP projection
that we have been using to represent declaratives. In the interests of uniformity—recall the dis-
cussion of normal form from Chapter 4—it would be sensible to represent ordinary declarative
sentences as CPs also, with a head encoding the declarative illocutionary force. We could rep-
resent the head as [decl], or we could even use punctuation marks to represent the three three
illocutionary forces—[.], [?] (replacing [q]), and [!]. However, despite our love for normal forms,
Chapter 6: Wh- movement 196

we will bow to standard practice and not represent declarative sentences as CPs. We will only do
so where forced to do so by the word order (we will see examples in Chapter 12).

Topicalization 3
We turn now to topicalization, which we already know from Chapter 2, § 2.2, though under
another name, as already mentioned in the introduction to the chapter. From a descriptive point
of view, the term is not completely felicitous, since topicalization is not restricted to topics (=
previously mentioned discourse entities). Nevertheless, we will use the term because it is standard
in the syntax literature. (32) gives some examples

(32) a. [DP The smallest zebra, ] the lions will devour.

b. [AdvP Unbelievably quickly, ] the lions will devour the zebra.
c. [AdjP Quick as a wink, ] the cat hid under the covers.
d. [PP Over the next few days, ] the snow will melt.

It is very clear from (32a) that topicalization must involve movement because devour is
obligatorily transitive, and so the position where the object is pronounced (at the beginning of
the sentence) is not where it is interpreted (as the complement of the verb). As in wh- ques-
tions, the subject is preceded by some other constituent. But in contrast to direct wh- questions,
the examples in (32) do not exhibit subject-aux inversion. All of these properties, topicalization
shares with indirect wh- questions, so the obvious thing to do is to extend the analysis of indirect
questions to topicalization. There is an outstanding question, though—what morpheme heads the
CP projection? Since the instances of topicalization in (32) are not subordinate clauses, it is not
attractive to treat them as projections of (silent) that. Once again, Japanese comes to the rescue.
Just as Japanese has an overt question morpheme ka, so, too, does it have an overt topic mor-
pheme -wa. Some examples are given in (33). (Subjects and direct objects are marked in Japanese
with the particles -ga and -o, respectively, but both can be overriden by -wa.)

(33) a. 子供は りんごを 食べました。

Kodomo-wa ringo-o tabe-mashi-ta.
child-top apple-acc eat-pol-pst
‘As for the child, it ate the apple.’
b. りんごは 子供が 食べました。
Ringo-wa kodomo-ga tabe-mashi-ta.
apple-top child-nom eat-pol-pst
‘As for the apple, the child ate it.’

As we did with the direct question morpheme earlier, we can say that English has a silent
morpheme that corresponds to an overt Japanese morpheme. The structure for (32) is then as in
Chapter 6: Wh- movement 197

(34), where top (short for “topicalization”) is the silent English topic morpheme.

(34) CP

DPi C′

the smallest zebra C◦ IP

[top]
DPk I′

the lions I◦ VP
will
tk V′

V◦ ti
devour

We invite you to build the structures for the remaining instances of topicalization in (32b)–
(32d) in Exercise 6.3.

Relative clauses 4

4.1 Wh- relative clauses

As (35) and (36) show, there is a striking parallel in English between wh- relative clauses and
indirect questions: both are introduced by wh- phrases.

(35) Indirect question:

a. We can’t remember who moved in next door.
b. We can’t remember who(m) you met.
c. We can’t remember where you met them.
d. We can’t remember which you prefer.
e. We can’t remember whose parents you met.
(36) Relative clause:
a. the people who moved in next door.
b. the people who(m) you met.
c. the place where you met them.
d. the movie which you prefer.
Chapter 6: Wh- movement 198

e. the friend whose parents you met.

This parallel follows straightforwardly if we assume that wh- relative clauses are structurally
parallel to indirect questions. The structure of the relative clause in (36b) is given in (37a). In
contrast to indirect questions, which are ordinarily complements, relative clauses are always
modifiers. We therefore integrate the relative clause into the surrounding syntactic structure
by adjunction, as in (37b) (where we collapse the internal structure of the relative clause for
simplicity).

(37) a. CP

DPi C′

who(m) C◦ IP
∅that
DPk I′

you I◦ VP
[pst]
tk V′

V◦ t i
met
b. DP

D′

D◦ N′
the
N′ CP

N◦ who(m) you met

people

4.2 That relative clauses

In addition to wh- relative clauses, English also has that relative clauses, as illustrated in
(38).1

(38) a. the people that moved in next door.

b. the people that you met.
c. the place that you met them.
d. the movie that you prefer.

Structurally, that relative clauses are completely parallel to wh- relative clauses. But in
contrast to wh- relative clauses, it is the complementizer that which is overt in that relative
Chapter 6: Wh- movement 199

clauses, and the wh- phrase that is silent. The structures corresponding to the ones in (37) are
given in (39).

(39) a. CP

DPi C′

∅who(m) C◦ IP
that
DPk I′

you I◦ VP
[pst]
tk V′

V◦ t i
met
b. DP

D′

D◦ N′
the
N′ CP

N◦ that you met

people

4.3 Zero relative clauses

Given the availability of silent wh- elements and silent complementizers in English, we would
expect to find relative clauses that are not introduced by any overt element at all. Such zero
relative clauses (also known as contact relative clauses) are indeed possible in English, as
shown in (40).2

(40) a. the people you met.

b. the place you met them.
c. the movie you prefer.

Given the grammaticality of (40), we would expect the zero relative counterparts of subject
relative clauses like (40a)–(40b) to be fully grammatical and acceptable, but speakers tend to reject
examples like (40c), especially when they are presented in isolation.

(41) a. ✔ The people who moved in next door were from New York.
b. ✔ The people that moved in next door were from New York.
Chapter 6: Wh- movement 200

c. ⁇ The people moved in next door were from New York.

Nevertheless, structurally analogous examples, as in (42), are attested.

(42) a. Everybody lives in the mountains has an accent all to theirself. (Christian and
Wolfram 1976, front matter)
b. Three times a day some nurse looks like Pancho Villa shoots sheep cum into my
belly. (Hiaasen 1995, pp. 248–249)

(43)–(46) give further examples from several varieties of English, classified by linguistic environ-
ment.

(43) Existential there clause:

a. There’s a shortcut takes you to the shops. (Henry 1995, p. 125)
b. “Thanks for the hurricane, there’s a hundred fifty thousand houses in Dade County
need new roofs”, he began. (Hiaasen 1995, pp. 110–111)
c. … it might be worthwhile to mention that there’s a train leaves Pangbourne, I
know, soon after five … (Jerome 1889, p. 207)
(44) It cleft:
a. It was John told us about it. (Henry 1995, p. 125)
b. ’Tis grace hath brought me safe thus far (“Amazing Grace”)
c. It was the quickness of the hand deceived the eye. (Jerome 1889, p. 221)
(45) Copular construction:
a. John is the person could help you with that. (Henry 1995, p. 125)
b. You’re the second guy this month wants to take out trade in this bizzare fashion.
(Wagner 1986, p. 119)
c. He’s the one inspected the damn things. (Hiaasen 1995, p. 5)
(46) Introduction of discourse entity in object position:
a. I met a man can speak five languages. (Henry 1995, p. 125)
b. how come we have … a pink-haired punk granddaughter got the manners of a
terrorist? … Wears somethin’ makes the garage door flap up? (Wagner 1986,
p. 81)

The proper analysis of zero subject relative clauses is still being debated and goes beyond
the scope of this textbook. On the one hand, Henry (1995) argues that such clauses are not true
relative clauses. On the other hand, it has been argued that zero subject relative clauses are
grammatical, but avoided for psycholinguistic reasons (Bever and Langendoen 1971, especially
Section 5; see also Doherty 1993). The idea is that zero subject relative clauses are hard to process
in real time; in particular, they are liable to be misparsed as the predicate of the matrix clause,
at least when they modify subjects, as they do in (41c) and (42). In the sentence-processing
literature, constructions that invite misparsing are known as garden-path constructions. The
classic example is shown in (47).
Chapter 6: Wh- movement 201

(47) The horse raced past the barn fell.

When presented with a word-by-word written version of (47), experimental subjects register
surprise at the last word fell, because they have mistakenly taken raced to be the predicate of the
main clause, as it is in The horse raced past the barn. In other words, they have been led down
the proverbial garden path and must now recover by backtracking to raced and revising their
interpretation so that they can properly process fell as the predicate of the main clause.
In the classic example, the sequence raced past the barn is what is called a reduced relative
clause (“reduced” because it doesn’t contain a finite verb), the analysis of which is again beyond
the scope of this book. However, the same garden path effect arises in the zero subject relative
clause variant of (47) in (48).

(48) The horse was raced past the barn fell.

Once again, the relative clause was raced past the barn is first taken as the main clause predicate
(The horse was raced past the barn) until fell makes that interpretation impossible.
It is worth noting that the sentence-processing approach is consistent with the fact that
all of the examples in (42)–(46) are either spoken or from works intending to represent spoken
examples, where intonation would help hearers to avoid the garden path interpretation. Note
also that it is only in (42) that the zero subject relative clause modifies the subject of the main
clause. In the other examples, the zero subject relative clause comes later in the sentence—indeed,
at the very end, thus virtually eliminating a garden path interpretation.

4.4 Doubly marked relative clauses

Finally, given the discussion so far, we would expect to find the converse of zero relative clauses—
namely, ones with an overt wh- element in Spec(CP) combined with an overt complementizer, as
in (49).

(49) The people who(m) that you met

Such doubly marked relative clauses are judged to be unacceptable in mainstream varieties of
modern English. However, just like doubly marked indirect questions, they are attested in Middle
English, as shown in (40) (once again, the examples are from the Penn-Helsinki Parsed Corpus of
Middle English).

(50) a. thy freend which that thou has lorn (cmctmeli.m3, 218.C1.31)
‘your friend that you have lost’
b. the conseil which that was yeven to yow by the men of lawe and the wise folk (cmct-
meli.m3, 226.C2.373)
‘the counsel that was given to you by the men of law and the wise people’
c. the seconde condicion which that the same Tullius addeth in this matiere (cmct-
meli.m3, 228.C1.429)
‘the second condition that this same Tullius adds in this matter’
Chapter 6: Wh- movement 202

d. for hire olde freendes which that were trewe and wyse (cmctmeli.m3, 237.C2.799)
‘for her old friends who were loyal and wise’
e. the fire of angre and of wratthe, which that he sholde quenche (cmctpars.m3,
308.C2.859)
‘the fire of anger and wrath, which he should quench’

Doubly marked relative clauses are also attested in vernacular varieties of other languages.
(51) gives examples from Bavarian German.

(51) Bavarian (Bayer 1983, p. 23, (10a, b)):

a. der Hund der wo gestern d’ Katz bissn hod
the dog who that yesterday the cat bitten has
‘the dog that bit the cat yesterday’
b. die Frau dera wo da Xaver a Bussl g’gem hod
the woman who.dat that the a kiss given has
‘the woman that Xaver gave a kiss to’

4.5 Stylistic considerations

What these facts suggest is that doubly marked wh- movement constructions (both relative
clauses and indirect questions) are grammatical (= well-formed from a purely structural point
of view). However, their subordinate clause status is signalled redundantly by the absence of
subject-aux inversion as well as by the presence of the overt complementizer, and perhaps a
stylistic constraint has developed in the history of mainstream English and other languages that
deprecates such redundant marking. It is clear that stylistic constraints of the sort that we postu-
late exist. For instance, formal contexts favor the presence of an overt complementizer in contexts
like (52), where the grammar allows the complementizer to be silent (Kroch and Small 1978).

(52) I think (that) they are coming.

It might be objected that the stylistic constraint required for (52) favors the presence of
that, whereas what is at issue for wh- relative clauses in modern English is the absence of that.
Very true. The stylistic economy principle as in (53) covers both cases.

(53) Stylistic economy principle:

Mark subordination (and constructions more generally) in a way that is explicit but not
redundant.

If you like, you can think of (53) as an Goldilocks principle (not too little marking, and not too
much, either—but just right). The Goldilocks principle also seems to be at work in the proscrip-
tion that developed in the history of English against negative concord, discussed briefly in Chap-
ter 1.
There is one final loose end—namely, that zero relative clauses violate the Goldilocks prin-
ciple as it stands. We can make sense of this by assuming that language users have a stylistic
monitor, whose job it is is to guard against violations of (53) and other shibboleths. Now notice
Chapter 6: Wh- movement 203

that redundant marking gives the stylistic monitor an overt element to detect, whereas that is not
true for implicit marking, which might therefore be more likely to fly under the stylistic moni-
tor’s radar, so to speak. In other words, we can take the principle as absolute, but its application
by the stylistic monitor as subject to (human) error.
When the Goldilocks principle operates at the stylistic level, as we are proposing in connec-
tion with the cases just discussed, it works as a filter on structures that are otherwise grammatical.
In formal contexts, it might be set to high alert, but to a lower level otherwise. This fits nicely with
the idea of vernacular usage being “unfiltered”, and it also explains the observation mentioned
earlier in connection with (52).

4.6 A historical note

As we have just seen, English in its modern form, and even more so over the course of its history,
exhibits several different options for forming relative clauses—more options than are common
in a single language. The reason for this state of affairs is the rich and complicated history of
the English language. Old English, the oldest recorded stage of the language, already had several
different types of relative clauses, notably one corresponding to that relative clauses and another
corresponding roughly to doubly marked relative clauses, except that what combined with overt
that was a demonstrative pronoun rather than a wh- word. So instead of who that, Old English
had se þe ‘this that’.
Towards the end of the Old English period, beginning in 793 and ending with a peace
treaty in 878, England was invaded by Vikings, who eventually occupied about half of England,
the so-called Danelaw. For our purposes, what is important is that the Vikings spoke Old Norse,
the ancestor of the modern Scandinavian languages, which in contrast to Old English had silent
that in ordinary complement clauses and relative clauses. Once the Vikings invaders settled in
England permanently, they learned English as adults and carried over this silent that into their
non-native English. English-speaking children growing up in communities in the Danelaw then
acquired this feature natively, and over time it became a feature of English even outside the
Danelaw.
The Vikings not only invaded and settled in England. They also invaded and settled in the
north of France, where they were known as Normans (literally “North men”). Eventually, they
became dukes of Normandy, and one of their dynasty (William the Conqueror) invaded England,
won a decisive battle at Hastings, and was crowned king of England on Christmas Day in 1066.
This chain of events is known as the Norman Conquest, and it is conventionally used as an (ap-
proximate) cutoff point between Old English and Middle English. Until 1362, the official language
of England was French (or more specifically Anglo-Norman, the variety of French originally spo-
ken in Normandy, but as used in England), and members of the governing class in England were
bilingual in English and French. In fact, many were trilingual because they also spoke and wrote
Latin. For our purposes, the Norman Conquest and the multilingualism of England during the
Middle English period of the language is important because relative clauses in French (and Latin
and its descendant Romance languages more generally) are formed with question words. In Ro-
mance, these words generally begin with qu- (qui ‘who’, que ‘what’, quand ‘when’, and others),
but it is standard in the syntax literature to call them wh- words, and so we will, too. It is plausible
to attribute the presence, or least the prevalence, of wh- relative clauses in English to this influ-
 Chapter 6: Wh- movement 204

ence from French. Recall that Old English had a type of relative clause formed with a pronoun
followed by a complementizer (se þe ‘this the’). Replacing the demonstrative pronoun with the
corresponding wh- word (as well as replacing Old English þe with Middle English that) yields the
doubly marked relative clauses typical of Middle English (who that).
Because French (and Latin) were high-status languages, the originally Romance (or at least
Romance-influenced) wh- relative clauses were and still are evaluated as fancy. By contrast, the
originally Old Norse silent complementizer was and still is evaluated as not fancy, since it was
more typical of the north than of the high-status south (the location of London, the capital).

Notes 5
1. The that counterpart to (36e) is missing; it would look as in (i).

(i) the friend ’s parents that you met

The intended interpretation is as in (36e). In other words, (i) might go on as in (ii.a), but not as in (ii.b), where the
relative clause is that you met, not the intended ’s parents that you met.

(ii) a. The friend [ ’s parents that you met ] is coming over for dinner.
b. The friend’s parents [ that you met ] are coming over for dinner.

2. Once again, the counterpart to (36e) is missing for reasons analogous to those in Note 1; just delete that in the
examples there.

Exercises and problems 6

Exercise 6.1
A. Build structures for both of the direct questions in (12). Treat auxiliary do as a modal. You
can use a triangle for personal pronouns, but build full structures for the wh- phrases.

(1) a. In which house do you live? (pied piping)

b. Which house do you live in? (preposition stranding)

B. In your trees for (A), you have attached the PP as either a complement or an adjunct. Pro-
vide the relevant evidence.
Chapter 6: Wh- movement 205

C. Build structures for the complex sentences in (2), which contain indirect questions. From
now on, you can also collapse the wh-phrases using triangle notation.

(2) a. I forget [ in which house you live ]. (pied piping)

b. I forget [ which house you live in ]. (preposition stranding)

D. Build structures for the noun phrases in (3).

(3) a. the house [ in which you live ]. (pied piping)

b. the house [ which you live in ]. (preposition stranding)

Exercise 6.2
A. The question in the reporter’s telegram in (1a) is structurally ambiguous. Paraphrase the
two interpretations.

(1) a. Reporter’s telegram: How old Cary Grant?

b. Cary Grant’s reply: Old Cary Grant fine.

B. Build a structure for the interpretation of (1a) that the reporter intended. Here and in
(C), treat Cary Grant as a compound noun with a silent determiner, and do not attempt to
implement verb movement to C. (The absence of do support is unexpected. The issue is
addressed in Chapter 10.)
C. Build the structure for the other interpretation—the one that Cary Grant cleverly exploits.

Exercise 6.3
A. Using (17a) in the text as a model, build full structures for the topicalized phrases in (1a)–
(1c) (= (32b)–(32d)). Treat as as P, and next and few as Adj. Treat quick as a wink as a
modifier.

(1) a. [AdvP Unbelievably quickly ] the lions will devour the zebra.
b. [AdjP Quick as a wink ], the cat hid under the covers.
c. [PP Over the next few days ], the snow will melt.

B. Using (17b) in the text as a model for when to use triangles, build full structures for the
entire sentences in (1a)–(1c).
Chapter 6: Wh- movement 206

Exercise 6.4
A. The noun phrase (1) is structurally ambiguous. Paraphrase or otherwise distinguish be-
tween the two interpretations.

(1) the cat that ate the rat that ate the cheese

B. Build the structure for the first relative clause. There is no need to build a structure for the
second one, as it is identical except for the last word.
C. Build structures for the two interpretations, indicating clearly which structure goes with
which interpretation. For simplicity, you can use triangle notation for the relative clauses
when it is not necessary to give the full structure.

Exercise 6.5
The following quote (from The Prime of Miss Jean Brodie) exists in several different forms. The
one in (1) is the most abstruse.

(1) It’s the kind of thing that people who like that kind of thing like.
A. Build the full structure for the relative clause who like that kind of thing. Here and through-
out, assume for simplicity that of phrases are complements.
B. Build the structure for the relative clause that people who like that kind of thing like. Use
triangle notation for the embedded relative clause that you’ve already built in (A).
C. Build the structure for the remainder of the sentence (that is, for the matrix sentence ex-
cluding both relative clauses).
D. Describe how the structures in (A)–(C) fit together.

Exercise 6.6
English conditional clauses are ordinarily introduced by if, as in (1), but English also allows such
clauses to be marked by subject-aux inversion, as in (2).

(1) a. I will call if I have time.

b. If I have time, I will call.
(2) a. I will call should I have time.
b. Should I have time, I will call.
A. Build two structures for (1a)—a full structure for the conditional clause (no triangles) and
a structure for the entire sentence. For simplicity, collapse the noun phrases and the con-
ditional clause in the second structure.
B. Build a structure for the full sentence in (1b). Once again, collapse the noun phrases and
the conditional clause.
C. Repeat (A) and (B) for (2). Propose a suitable C head to accommodate the inversion of the
subject with the modal.
Chapter 6: Wh- movement 207

Exercise 6.7
This exercise extends Exercise 5.8.
English has two sorts of though clauses: ordinary ones that do not involve movement, as in
(1a), and ones that do, as in (1b). We will refer to the construction in (1b) as the though preposing
construction.

Word of Caution
The term though preposing is potentially confusing. What is preposed is not though it-
self, but some constituent of the though clause. In other words, though preposing is XP
preposing that is licensed by though.

(1) a. We will solve the problem, though it seems difficult. (ordinary)

b. We will solve the problem, difficult though it seems. (though preposing)
A. Build two structures for (1a)—the full structure for the though clause (no triangles), and a
structure for the entire sentence. For simplicity, collapse the noun phrases and the though
clause in the second structure. Treat though as a P, as argued in Chapter 5, § 3.2.
B. Build the full structure for the subordinate clause in (1b).
C. How do the elementary trees for though needed in (A) and (B) differ?
D. Though preposing is grammatical in (2a), but not in (2b). Explain. You should not need
more than two sentences.

(2) a. ✔ Difficult though the problem seems, …

b. * Difficult though the problem seems very, …

Exercise 6.8
A. Based on the data in (1) and (2), how do why questions differ from how come questions in
Mainstream English?

(1) a. ✔ Why are they making such a fuss?

b. * Why they are making such a fuss?
(2) a. * How come are they making such a fuss?
b. ✔ How come they are making such a fuss?

B. Based on the data in (3), how would you characterize the difference between direct ques-
tions in African American English (AAE) and in Mainstream American English (MAE)?

(3) a. Why you didn’t tell me that?

(overheard at Market and 22nd Street, Philadelphia, PA, 26 September 1998)
b. What you bought tickets for?
(overheard at 30th Street Station, Philadelphia, PA, 26 November 1998)
Chapter 6: Wh- movement 208

c. Where you was at?

(overheard at Rittenhouse Square, Philadelphia, PA, 20 July 2001)
d. Where you went?
(overheard at Chestnut and 36th Street, Philadelphia, PA, 13 August 2001)
e. What I told you?
(Willie Perdomo. From where a nickel cost a dime. Real News. April 2002. 28)

Problem 6.1
Build the full structure for the noun phrase in (1). For the purposes of this exercise, treat has as
a modal.

(1) a product that’s time has come

Hint: Some speakers have reanalyzed that in relative clauses as a relative pronoun.

Problem 6.2
A. Given the discussion in the chapter as it stands, is it possible to derive the statement by
Yoda in (1)? Explain.

(1) When 900 years you reach, look as good you will not.

B. If not, what adjustments would have to be made to accommodate the sentence?

Extra Info
We discuss the syntax of not in Chapter 10. For the moment, ignore not in your an-
swer.

If you’re curious, 900 is a Num(ber), a subtype of Adj, and as is a Deg(ree) word, a subtype
of Adv. However, for this problem, you can ignore the internal structure of 900 years and
look as good.
 7
Case theory

Chapter outline
1 A first look at case . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
1.1 The basic purpose of case . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
1.2 Case government . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
1.3 Synthetic versus analytic case marking . . . . . . . . . . . . . . . . . . . . 213
2 Case features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
3 Case assignment configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
3.1 Head-comp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
3.2 Spec-head . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222
3.3 Head-spec . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
4 Further issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
4.1 Subject movement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
5 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
6 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231

This chapter is devoted to case, a more or less abstract property of noun phrases. Not every
human language expresses this property visibly in terms of morphological distinctions, and so
its existence in the world’s languages is a bit of a mystery. However, in languages with rich
morphological case systems, it is clear that case signals (or at least helps signal) a noun phrase’s
grammatical relation to the rest of the sentence (for instance, whether a noun phrase is a subject or
object). In languages with relatively poor case morphology, like English, this guidepost function
of case is less evident; case carries out its function indirectly by regulating the surface order of
noun phrases (specifically, which links in a noun phrase chain can be expressed overtly). The role
of case in these languages is reminiscent of the rules of a board game, which have no intrinsic
meaning, and can seem mechanical to the point of being downright unsatisfying. Nevertheless,
we will see that case is useful in describing and understanding the phenomena covered in the
next two chapters (passive and raising versus control).
As we have just said, English is a language with few visible case distinctions, and so we
set the stage by reviewing the richer system of Indo-European, the ancestor language of English
and many other (though by no means all) languages of Europe and parts of Asia. The older de-
Chapter 7: Case theory 210

scendants of Indo-European tended to express case synthetically as affixes on words, whereas

the more modern languages tend to express case analytically, by co-opting relatively content-
less prepositions as case markers or by signalling function via word order (as just mentioned).
Languages are not necessarily “pure” in the use of these strategies; it is common to find case
expressed synthetically on pronouns, but analytically on full noun phrases (as is true for Eng-
lish). In order to make it easier to think about case independently of these two case-marking
strategies, we introduce the notion of case feature. We then describe the distribution of case
forms in terms of heads assigning these case features to noun phrases in three well-defined case
configurations. The chapter concludes with some issues raised by the discussion in the main
text but not central to an understanding of it.

A first look at case 1

1.1 The basic purpose of case

In order to understand the role of case in human language, it is useful to consider languages in
which constituent order is not as fixed as it is in English. In German, for instance, unlike English,
the subject of an ordinary declarative clause doesn’t need to precede the verb, as shown in (1)
and (2) (we discuss the structure of German sentences in more detail in Chapter 12; for now, only
the variable constituent order is of interest). In the examples, boldface indicates the subject, and
italics indicates the object.

(1) German:
a. Der Mann sieht den Hund.
the man sees the dog
‘The man sees the dog.’
b. Den Hund sieht der Mann.
the dog sees the man
same as (1a), not the same as (2a)
(2) a. Der Hund sieht den Mann.
the man sees the dog
‘The dog sees the man.’
b. Den Mann sieht der Hund.
the man sees the dog
same as (2a), not the same as (1a)

Since German speakers can’t rely on constituent order to identify subjects and objects, how
is it possible for them to keep track of which noun phrase expresses which grammatical relation
(and more generally, which sentence expresses which meaning)? The answer is that grammatical
Chapter 7: Case theory 211

relations are encoded in German by various morphological case forms. In particular, the subjects
of finite clauses in German appear in a particular form called the nominative case, whereas
objects generally appear in the accusative. (3) gives a morphological analysis of the noun phrases
in (1) and (2).

(3) a. d -er Mann, d -er Hund

the -nom man the -nom dog
b. d -en Mann, d -en Hund
the -acc man the -acc dog
Notice that in (3), the distinction between nominative and accusative case is marked once: on the
head of the noun phrase (the determiner).
In certain exceptional cases in German, case distinctions are marked redundantly: on the
determiner as well on the noun. This is illustrated in (4). The empty set symbol ‘∅’ explicitly
indicates a zero (unpronounced) nominative suffix; -(en) is the accusative suffix that we know
from (3), but here, on the noun, it is optional, as indicated by the parentheses.

(4) a. d- er Bär -∅, d -er Student -∅ (nominative)

the -nom bear -nom, the -nom student -nom
b. d -en Bär (-en), d -en Student (-en) (accusative)
the -acc bear (-acc) the -acc student (-acc)
The redundant case marking in (4) is a historical relic from an earlier stage of German where this
pattern was more extensive. In some languages, redundant case marking on the determiner and
the noun is the norm. This is illustrated for modern Greek in (5).

(5) Modern Greek:

a. Ο άνδρ-ας βλέπει τ-ο σκύλο.
O andr-as vlepi t-o skil-o
the.nom man-nom sees the-acc dog-acc
‘The man sees the dog.’
b. Ο σκύλ-ος βλέπει τ-ον άνδρ-α.
O skil-os vlepi t-on andr-a
the.nom dog-nom sees the-acc man-acc
‘The dog sees the man.’

Finally, case can be marked solely on the noun. This is the case for Latin, as shown in (6),
for the simple reason that Latin had no articles.

(6) Latin:
a. Fēmin-a can-em videt.
woman-nom dog-acc sees
‘The woman sees the dog.’
b. Can-is fēmin-am videt.
dog-nom woman-acc sees
‘The dog sees the woman.’
Chapter 7: Case theory 212

To summarize the discussion in this section: noun phrases can be case-marked either on the
determiner, or on the noun, or redundantly on both. But regardless of the particular pattern, case
marking has the same basic purpose: it visibly expresses a noun phrase’s grammatical function
in a sentence.

1.2 Case government

In many languages, a noun phrase’s morphological case depends not only on its function in the
entire sentence, but also on the head that it is most closely associated with. For instance, in
German, the object in a sentence can appear in the accusative, as we have already seen, or in the
dative,1 depending on the verb. Some examples are shown in (7) and (8).

(7) a. ✔ { d-en Hund, d-ie Frau } unterstützen (accusative)

the-acc dog the-acc woman support
‘to support the { dog, woman }’
b. * { d-em Hund, d-er Frau } unterstützen (dative)
the-dat dog the-dat woman support
(8) a. * { d-en Hund, d-ie Frau } helfen (accusative)
the-acc dog the-acc woman help
‘to support the { dog, woman }’
b. ✔ { d-em Hund, d-er Frau } helfen (dative)
the-dat dog the-dat woman help

In traditional grammar, the verb is said to govern the case of the object. So in the examples,
unterstützen ‘support’ governs the accusative, whereas helfen ‘help’ governs the dative.
Case government in Latin is illustrated in (9). As in German, each particular verb governs
the case of its object, but in Latin, the choice of case ranges over three cases—dative, accusative,
and ablative.

(9) a. { fēmin -ae, *fēmin -am, *fēmin -ā } { sub- venīre, suc- currere } (dative)
woman -dat -acc -abl under- come under- run
‘to help the woman’
b. { fēmin -am, *fēmin -ae, *fēmin -ā } ad- iuvāre (accusative)
woman -acc -dat -abl to- support
‘to support the woman’
c. { fēmin -ā, *fēmin -ae, *fēmin -am } fruī (ablative)
woman -abl -dat -acc enjoy
‘to enjoy the company of the woman’

In both German and Latin, prepositions resemble verbs in governing the case of their com-
plement. In German, prepositions govern the accusative, the dative, or (rarely) the genitive; in
Latin, they govern the ablative or the accusative.

(10) a. German:
Chapter 7: Case theory 213

durch d-ie Tür, bei d-er Kirche, während d-es Krieges

through the-acc door, by the-dat church, during the-gen war
‘through the door, by/near the church, during the war’
b. Latin:
dē sapienti-ā, ad rīp-am
about wisdom-abl, to shore-acc
‘about wisdom, to the shore’

Finally, in both German and Latin, certain prepositions can govern more than one case.
In such cases, the accusative marks direction, and the other case (dative in German, ablative in
Latin) marks location.

(11) German:
a. in { d-ie, *d-er } Bibliothek schicken
in the-acc the-dat library send
‘to send into the library’
b. in { d-er, *d-ie } Bibliothek arbeiten
in the-dat the-acc library work
‘to work in the library’
(12) Latin:
a. in { bibliothēc -am, *bibliothēc -ā } mittere
in library -acc -abl send
‘to send into the library’
b. in { bibliothēc -ā, *bibliothēc -am } labōrāre
in library -abl -acc work
‘to work in the library’

1.3 Synthetic versus analytic case marking

Proto-Indo-European (PIE) is a language that was spoken thousands of years ago before the in-
vention of writing; it has been reconstructed through a painstaking comparison of languages for
which there are written records or which are still spoken today. It is the ancestor of the Indo-
European language family, which includes many (though not all) of the languages of parts of
Asia and Europe, including English. The nominative marked the subject of finite clauses, the
accusative and dative (and perhaps other cases) marked objects (depending on the head, as just
discussed), and the genitive indicated possession. The PIE ablative indicated the source of move-
ment (as in I drove from Chicago), the locative was used for locations (as in I used to live in Chicago),
and the instrumental marked instruments or means (as in They cut it with a pocketknife). Finally,
the vocative was used to address persons (as in Hey, Chris, come on over here).
The original PIE case system is essentially preserved in Sanskrit, although the distinction
between the ablative and the genitive is somewhat obscured because the relevant forms were
often homophonous in Sanskrit. Such homophony among two or more case forms is called case
syncretism. Among living languages, the PIE system is best preserved in the Baltic languages
(Latvian and Lithuanian) and some Slavic languages (Czech and Ukrainian). In these languages,
Chapter 7: Case theory 214

the genitive and the ablative have merged completely, leaving seven cases. In other words, in
the history of these languages, case syncretism extended to all forms of the genitive and the
ablative, not just to some of them, and so children learning the language no longer had any
evidence anywhere in the language for distinguishing between the two cases. Several other Slavic
languages, including Russian, have in addition almost completely lost the vocative, leaving six
cases. In Latin, the PIE ablative, instrumental, and locative merged into a single case, called the
ablative, which serves all three functions. With a few exceptions, the vocative is mostly merged
with the nominative, so this also leaves six cases. In Ancient Greek, the ablative, instrumental,
and locative were lost, but the vocative was not (it persists into modern Greek), leaving five cases.
Old English had five cases as well, having lost the ablative, locative, and vocative; in addition, the
instrumental had mostly merged with the dative. Another Germanic language, modern German,
retains four cases: nominative, dative, accusative, and an increasingly moribund genitive. These
developments are summarized in Tables 7.1 and 7.2.

PIE, Sanskrit Baltic, Other Slavic Latin

some Slavic
Nominative retained retained retained retained
Dative retained retained retained retained
Accusative retained retained retained retained
Genitive retained merged as merged as retained
Ablative retained genitive genitive
merged as
Locative retained retained retained ablative
Instrumental retained retained retained
mostly
Vocative retained retained lost merged with
nominative
Number of distinct cases 8 7 6 6

Table 7.1: Summary of historical case developments in Indo-European languages

Chapter 7: Case theory 215

Ancient Greek Old English German

Nominative retained retained retained
Dative retained retained retained
Accusative retained retained retained
Genitive retained retained retained
Ablative lost lost lost
Locative lost lost lost
Instrumental lost mostly merged with dative lost
Vocative retained lost lost
Number of distinct cases 5 5 4

Table 7.2: Continued summary of historical case developments in Indo-European languages

Table 7.3 shows the complete case paradigms for the Latin nouns fēmina ‘woman’ and avus
‘grandfather’. These two nouns are each representative of two distinct declensions, or word
classes. Latin had a total of five such word classes, each of which was characterized by unique
endings for combinations of case and number. For instance, dative singular is marked by -ae
on fēmina and by -ō on avus. In the remaining three declensions, this combination happens to
be marked by the same suffix, namely -ī (but there are other distinctions in the paradigms that
motivate distinguishing the declensions, rather than collapsing them into one).

a-declension o-declension
Latin ‘woman’ ‘grandfather’
sg pl sg pl
Nominative fēmin-a fēmin-ae av-us av-ī
Genitive fēmin-ae fēmin-ārum av-ī av-ōrum
Dative fēmin-ae fēmin-īs av-ō av-īs
Accusative fēmin-am fēmin-ās av-um av-ōs
Ablative fēmin-ā fēmin-īs av-ō av-īs
Vocative fēmin-a fēmin-ae av-e av-ī

Table 7.3: Complete case paradigms for Latin fēmina ‘woman’ and avus ‘grandfather’

As Table 7.3 shows, Latin exhibited some case syncretism. In particular, the genitive and
the dative singular are homophonous for fēmina ‘woman’, the dative and ablative singular are
homophonous for avus ‘grandfather’, and the dative and the ablative plural are homophonous
for both nouns.
In the descendants of Latin, the Romance languages, case continues to be expressed syn-
thetically on pronouns. For instance, the distinction between dative and accusative pronouns is
Chapter 7: Case theory 216

illustrated for French in (13). Note that unstressed pronouns in French are clitics; unlike full noun
phrases, they precede the verb they are construed (= belong) with.

(13) a. Je veux leur parler.

I want 3pl.dat talk
‘I want to talk to them.’
b. Je veux les voir.
I want 3pl.acc see
‘I want to see them.’

With full noun phrases, however, the same distinction is expressed analytically by the presence
or absence of the dative case marker à.

(14) a. Je veux parler à vos voisins.

I want talk dat your neighbors
‘I want to talk to your neighbors.’
b. Je veux voir vos voisins.
I want see your neighbors
‘I want to see your neighbors.’

This case marker has its historical origin in the spatial preposition à ‘to’, but is distinct from it
in the modern language. This is evident from the fact that the pro-form for phrases where à is a
spatial preposition is not leur, as in (13a) (or lui in the singular), but y, just as it is for other spatial
prepositions like dans ‘in’ or sur ‘on’.

(15) a. Nous avons envoyé le vin à Toulouse; mon amie habite à Paris.
we have sent the wine to Toulouse my friend lives in Paris
‘We sent the wine to Toulouse; my friend lives in Paris.’
b. Nous y avons envoyé le vin; mon amie y habite.
we there have sent the wine my friend there lives
‘We sent the wine there; my friend lives there.’
(16) a. Le cadeau se trouve dans mon sac; nous avons mis le cadeau sur la table.
the present Refl finds in my bag we have put the present on the table
‘The present is (literally, ‘finds itself’) in the bag; we put the present on the table.’
b. Le cadeau s’ y trouve; nous y avons mis le cadeau.
the present Refl there finds we there have put the present
‘The present is there; we put the present there.’

As mentioned earlier, Old English had five cases, which are illustrated in Table 7.4 for three
declensions. As is evident, case syncretism is more extensive in Old English than in Latin.
Chapter 7: Case theory 217

m f n
Old English ‘fox’ ‘learning’ ‘animal’
sg pl sg pl sg pl
Nominative fox fox-as lar lar-a deor deor2
Genitive fox-es fox-as lar-e lar-a deor-es deor-a
Dative fox-e fox-e lar-e lar-um deor-e deor-um
Instrumental fox-e fox-e lar-e lar-um deor-e deor-um
Accusative fox fox-as lar-e lar-a deor deor

Table 7.4: Complete case paradigms for OE fox ‘fox’, lar ‘learning’, and deor ‘animal’

In the course of Middle English (1150–1500), the old genitive case suffixes were lost, and
their function was taken over by a syntactic head—the possessive determiner ’s (in the plural,
the possessive is spelled out as a silent determiner that is orthographically represented as an
apostrophe). The old synthetic genitive case is illustrated in (17). Recall that the thorn character
(þ) corresponds to modern English ‘th’.

(17) þe king-es suster of France (cmpeterb, 59.593)

the king-gen sister of France
‘the king of France’s sister’

Although the change itself is not yet fully understood, it is clear that the modern possessive
marker is no longer a synthetic case suffix on a noun (N) (king), as it is in (17), but rather analyt-
ically case-marks an entire noun phrase (DP) (the king of France). This is clear from the fact that
it follows postnominal material like the prepositional phrase of France in modern English (cf. the
translation of (17)). The difference between the synthetic and analytic strategies emerges even
more sharply from the contrast in (18), where the possessive determiner obligatorily follows an
element that is not even a noun. For clarity, the noun phrase that is case-marked by the posses-
sive determiner is underlined in (18b); the entire sequence in (18b) from the to cat is of course
also a noun phrase.

(18) a. * the guy’s that I used to go out with cat

b. the guy that I used to go out with’s cat

Word of Caution
It is tempting to think of the Modern English ending ’s on singular noun phrases as an
instance of synthetic case marking. But that can’t be, since (17) is no longer grammatical.
Under the analysis just given, the nominative, possessive, and objective case forms of full
noun phrases are all homophonous in Modern English, and ’s in the king’s is an analytic
case marker on a par with the preposition of in of the king (though unlike the preposition,
it is a bound morpheme).

Beginning in late Old English (ca. 1000 ce), the distinction between the dative and the
accusative weakened, and the distinction was lost completely in the course of Middle English
Chapter 7: Case theory 218

(1150–1500). In what follows, we will refer to the case that resulted from the merger as the
objective.
As in French, case is no longer visibly marked on full noun phrases in Modern English, but
continues to be marked synthetically on pronouns. This is evident from Table 7.5. The original
forms for second person singular were replaced in early Modern English by the corresponding
plural form; we include the old forms for completeness.3

Nominative Objective Possessive

1 sg I me my
2 sg thou thee thy
3 sg m, f, n he, she, it him, her, it his, her, its
1 pl we us our
2 pl you you your
3 pl they them their

Table 7.5: Synthetic case in Modern English pronouns

As the table shows, you and it have lost the distinction between nominative and objective, so
in this respect these pronouns resemble full noun phrases. Finally, it is worth noting that de-
spite the efforts of prescriptive grammarians to keep a distinction alive between nominative who
and objective whom, the two forms have merged as who in the vernacular (with whom surviving
marginally and used for stylistic rather than grammatical reasons). James Thurber has a diaboli-
cally witty essay on the topic.

Case features 2
The next two sections introduce some general concepts and conditions that enable us to derive
the distribution of case forms in English (which we focus on) and other languages. We begin by
introducing the notion of case feature.
Consider the contrast between (19) and (20).

(19) a. ✔ They will help her.

b. ✔ She will help them.
(20) a. * Them will help she.
Intended meaning: (19a)
b. * Her will help they.
Intended meaning: (19b)
Chapter 7: Case theory 219

Why are the sentences in (20) ungrammatical? The answer is that noun phrases in English
are subject to the requirements in (21).

(21) a. Subjects of finite clauses appear in the nominative.

b. Objects appear in the objective.

As is evident, both of the subjects in (20) are objective forms, and both of the objects are nom-
inative forms. Each of the sentences in (20) therefore maximally violates the requirements in
(21).
Now compare the examples in (19) and (20) with those in (22).

(22) a. You will help her.

b. She will help you.

As we saw in Table 7.5, they and she exhibit distinct forms for the nominative and objective,
whereas you doesn’t. But because case syncretism between the nominative and the objective is
not complete in English (in other words, because at least some pronouns still have distinct forms
for the two cases), we will treat you as a nominative form in (22a), equivalent to they and she,
but as an objective form in (22b), equivalent to them and her. Moreover, we will treat the noun
phrase my big brother as a nominative form in (23a) and as an objective form in (23b) for exactly
the same reason.

(23) a. My big brother will help her.

b. She will help my big brother.

In order to disambiguate instances of case syncretism like you and my big brother, it is useful
to associate noun phrases with case features. Each case feature has a value that is selected from
among all of the various case forms in that language (regardless of whether the case forms are
expressed synthetically or analytically). In English, for instance, a case feature can assume one of
three values (nominative, objective, possessive). In Russian, a case feature has a choice among six
values (nominative, genitive, dative, accusative, locative, instrumental). If we need to explicitly
represent a noun phrase’s case feature, we can do so by means of labels as in (24) and (25).

(24) a. [DP-nom They ] will help [DP-obj her ].

b. [DP-nom You ] will help [DP-obj her ].
c. [DP-nom My big brother ] will help [DP-obj her ].
(25) a. [DP-nom She ] will help [DP-obj them ].
b. [DP-nom She ] will help [DP-obj you ].
c. [DP-nom She ] will help [DP-obj my big brother ].
Chapter 7: Case theory 220

Case assignment configurations 3

Earlier, we said that the purpose of case is to encode a noun phrase’s function in the sentence.
In order to implement this insight in structural terms, we will require each noun phrase (or more
precisely, each noun phrase chain) to be associated with a syntactic head that assigns a case
feature to it.4 In English, case assigners are either verbs or prepositions (with very few exceptions),
but there are languages (usually ones with rich overt case morphology) where the set of case-
assigning heads includes adjectives and nouns. We can imagine the case feature originating
on the head and being transferred to the noun phrase in the course of the derivation. This fits
nicely with the idea of different heads governing different cases. It also immediately raises the
question (at least for a syntactician!) of whether the paths that case features take on their journey
from head to noun phrase are constrained in any way. Below, we will answer this question in
the affirmative and present three licit paths, or case assignment configurations: the head-
comp configuration, the spec-head configuration, and the head-spec configuration. Recall from
Chapter 4 that ‘spec’ (read as ‘speck’) refers to the specifier node in the X′ schema of phrase
structure. By analogy, we abbreviate ‘complement’ as ‘comp’.
The configurations are illustrated in schematic form in (26); nodes that are irrelevant for
case assignment are omitted.

(26) a. XP b. XP c.
XP
X ′
DP X′
X′
X◦ DP noun phrase
head X◦ X◦ YP
head head
noun phrase
DP Y′

noun phrase

Head-comp Spec-head Head-spec

(local, sideways) (local, up) (nonlocal, down)

The remainder of the section discuses concrete examples. The linear order of ‘spec’ and
‘head’ in the last two configurations correlates (we hope, mnemonically) with their height in
the tree. We draw your attention to the fact that the case-assigning head and the noun phrase
belong to the same maximal projection in the first two configurations, but not in the third. (More
precisely, no maximal projection intervenes between the case-assigning head and the noun phrase
Chapter 7: Case theory 221

in the first two structures, but there is such a intervening node in the third.) In this sense, case
assignment in the first two configurations, but not in third, is local. Finally, if you don’t find
our technical names mnemonic, you can think of the configurations as the head assigning case
sideways, up, or down.
Before proceeding, we briefly note that from the 1990s on, proposals have been made to
simplify the theory of case assignment by identifying a single configuration (in other words,
by reducing the set of three configurations in (26) to one). We touch on a relevant issue under
§ 4.1 Subject movement, but refrain from reviewing this work in detail, since some of it assumes
abstract syntactic heads not covered in this book. In this connection, it is important to keep in
mind that simplifications to case theory do not necessarily come for “free”, but may have to be
“bought” at the “cost” of complications elsewhere in the grammar, so that they may not in fact
turn out to be as much of a conceptual bargain as they seem at first.

3.1 Head-comp
The first case assignment configuration is the one that corresponds to the traditional notion of
case government. It is also the simplest in the sense that it contains the fewest number of nodes.
In English, it is used in connection with the assignment of objective case by P or V, as in (27). The
first two examples are simple. The third shows a more complex structure with two heads, each
of which assign case to its DP complement.

(27) a. expecting us
b. with them
c. expecting a storm with high winds

(28) gives the associated structures, with the endpoints of the path highlighted. (The higher
DP positions in (28a), (28c) are irrelevant here.)

(28) a. VP b. PP

DP V′ P′

V◦ DP P◦ DP
expecting with
us them
Chapter 7: Case theory 222

c. VP

DP V′

V◦ DP
expecting
D′

D◦ NP
a
N′

N′ PP

N◦ P′
storm
P◦ DP
with
high winds

3.2 Spec-head
As we have just seen, heads can assign case to their complements. Given the X′ schema, we might
expect heads to also assign case to the other argument position—in the spec-head configuration.5
Indeed, since subjects of sentences start out life as specifiers of verbs, this seems at first blush like
an attractive way to assign case to the subject of sentences. According to this idea, what assigns
nominative case to They (or more precisely, to its trace in Spec(VP)) in sentences like (29) is the
finite verb laughed in (29a) and its nonfinite counterpart laugh in (29b).

Word of Caution
In what follows, it is important to distinguish carefully between finite clauses and finite
verbs. See the appendix to Chapter 2 (§8: Appendix: Finiteness in English) for detailed
discussion.

(29) a. ✔ They laughed. (finite clause, finite verb)

b. ✔ They will laugh. (finite clause, nonfinite verb)

Although we will end up rejecting this approach, let us play it out for the moment in order
to show why it is unsatisfactory. (30) shows the structures just described. (We assume that DPs
are free to move on after being assigned case to other positions in the sentence.)
Chapter 7: Case theory 223

(30) a. IP b. IP

DPi I′ DPi I′

they I◦ VP they I◦ VP
[pst] will
ti V′ ti V′

V◦ V◦
laughed laugh

Finite clause, finite verb Finite clause, nonfinite verb

(to be revised!) (to be revised!)

Now if verbs were able to assign nominative case regardless of their finiteness, nonfinite
laugh should be able to do so to they in the structures in (31). (In (31b), the lower subject moves
on from Spec(VP) to Spec(IP), but that can’t be the reason for the ungrammaticality, because that
movement is grammatical in (30).)

(31) a. * IP

DPi I′

they I◦ VP
[pst]
ti V′

V◦ VP
made
DP V′

they V◦
laugh
Chapter 7: Case theory 224

b. * IP

DPi I′

they I◦ VP
[pst]
ti V′

V◦ IP
expected
DPk I′

they I◦ VP
to
tk V′

V◦
laugh

But these structures are completely ungrammatical, a fact impossible to attribute to semantic
anomaly, since the intended meaning can be expressed grammatically by simply replacing the
nominative by the objective, as in (32).

(32) a. * We made [ they laugh ]. (small clause)

Intended meaning: We made them laugh.
b. * We expected [ they to laugh ]. (to clause)
Intended meaning: We expected them to laugh.

We must therefore reject the idea that it is V that assigns nominative case. However, we
don’t have to give up the idea that nominative case is assigned in the spec-head configuration.
All we need to do is revise our idea about which head it is that assigns case head and have finite
I be the case-assigner. This then gives (33), which supersedes (30).
Chapter 7: Case theory 225

(33) a. IP b. IP

DPi I′ DPi I′

they I◦ VP they I◦ VP
[pst] will
ti V′ ti V′

V◦ V◦
laughed laugh

Finite clause, finite verb Finite clause, nonfinite verb

(revised) (revised)

The contrast between (29) and (32) now follows straightforwardly since I is finite in (29), but
missing or nonfinite in (32)6, 7 .
At the beginning of this chapter, we said that case is assigned to noun phrases and that
every noun phrase needs case. Here, we are at a convenient point to make that formulation more
precise. What actually gets and case is a noun phrase chain, and case-assigning heads and chains
stand in a one-to-one relationship. This fact will become more salient when we put our theory
of case assignment to work in order to describe the grammatical phenomena under discussion in
the next two chapters.
Nominative case is not the only case in English to be assigned in the spec-head configu-
ration. So is possessive case. Here, the case-assigning head is the possessive determiner (’s or
its silent plural variant). Three examples, along with their structures, are given in (34) and (35).
For the possessive pronoun, we have to assume that the specifier and the head combine in the
morphology as the synthetic form their.

(34) a. the county’s revenues

b. the counties’ revenues
c. their revenues
Chapter 7: Case theory 226

(35) a. b. c.
DP DP DP

DP D′ DP D′ DP D′

the county D◦ NP the counties D◦ NP their D◦ NP

’s ’ [poss]
N′ N′ N′

N◦ N◦ N◦
revenues revenues revenues

Extra Info
In possessive constructions, there are two DPs: a lower one (the possessor) and a higher
one (the entire noun phrase containing both the possessor and the thing possessed). (35)
highlights case assignment to the possessor DP. The containing DP gets case from some
other head.

3.3 Head-spec
The final case assignment configuration is the head-spec configuration. It is more unusual than
the first two in two respects. First, as already mentioned in connection with (26), it is the only
configuration in which a maximal projection intervenes between the case-assigning head and the
noun phrase. Second, in English, it allows V to assign objective case to a directly following noun
phrase in sentences like (36).

(36) They expected us to leave.

Sentences of this type are not very common cross-linguistically and were for this reason at
one time called Exceptional Case-Marking (ECM) sentences. Over time, opinions have gone
back and forth over exactly how exceptional Exceptional Case-Marking is. Perhaps as a result of
this, or perhaps only because it is shorter, the abbreviation is favored nowadays.
At first glance, it seems like we could avoid having to postulate this third configuration by
giving (36) the structure in (37a) rather than that in (37b). We treat to as a nonfinite counterpart of
would (cf. the paraphrase in (38)). For expository reasons, we omit the higher I projection.
Chapter 7: Case theory 227

(37) a. VP b. VP

»DP« V′ »DP« V′

V◦ DP IP V◦ IP
expected expected
us I′ DP I′
I◦ VP
to us I◦ VP
V′ to
V′
V◦
leave V◦
leave

Not like this Like this

But there are several reasons to reject (37a). First, (36) has roughly the same meaning as
(38).

(38) They expected that we would leave.

In both sentences, what is expected is a proposition (a state of affairs). This proposition is ex-
pressed as a single IP in (38), as indicated by the underlining, so that there is a direct correspon-
dence between semantic and syntactic constituenthood. This parallel is preserved in (37b), but
not in (37a).
A second, related, argument for rejecting (37a) comes from sentences containing expletive
there. Recall from Chapter 3, § 3.3 that expletive there is licensed as the subject of a clause con-
taining a verb of (coming into) existence. If we treat the DP following expect in ECM sentences
as a subject, as we do in (37b), then (39a) is expected to be grammatical on a par with (39b).

(39) a. They expect there to be a recession.

b. They expect that there will be a recession.

By contrast, treating that same DP as a complement of the matrix verb, as in (37a), would require
complicating the licensing condition on expletive there. And of course, even if we were successful
in revising the licensing condition, we would still forfeit the simple representation of the parallel
between (36) and (38) just discussed.
There is a third objection against (37a), which we invite you to find in Exercise 7.3.
As you may have noticed, the structure in (37a) violates the X′ schema and hence the binary
branching hypothesis, and so you might wonder why we have not adduced this as an argument
Chapter 7: Case theory 228

against the structure. Such an argument would not be based on evidence as we are using the
term in this book, but it would still be a valid conceptual argument. For the moment, however,
we have preferred to think of the X′ schema and the binary branching hypothesis as following
from facts like those just discussed, rather than as treating them as axiomatic.
The term ‘ECM’ is often restricted to sentences like (36), where the complement of the case-
assigning head is a nonfinite IP. In a slightly wider sense, the term can be used in connection with
small clauses like (51a) from Chapter 3, repeated here as (40) and given the structure in (41).

(40) God let [ there be light ].

(41) IP

DPi I′

God I◦ VP
[pst]
ti V′

V◦ VP
let

DP V′

there V◦ DP
be
light

In this wider sense, ‘ECM’ is essentially synonymous with ‘case assignment in the head-spec
configuration’.
We have now reached the point where we have covered all of the case assignment con-
figurations necessary for cross-linguistic description, and have also covered most of the case-
assigning heads necessary for English. At this point, it would be convenient to include a sum-
mary table for reference. For pedagogical purposes, we invite you to produce such a summary
table yourself in Exercise 7.1, A. Further refinements are the topic of Exercise 7.8 and Problem
7.1.
 Chapter 7: Case theory 229

Further issues 4

4.1 Subject movement

Earlier, we concluded that nominative case is assigned by finite I in the spec-head configuration.
If this conclusion is correct, then subject movement in English follows from considerations of
case assignment. In other words, the subject moves from Spec(VP) to Spec(IP) because it needs
case but can’t get it in its original position from either V or finite I. But given the possibility of
case assignment in the head-spec configuration, there is nothing in principle to stop finite I from
assigning nominative case downwards to Spec(VP). If this possibility could be generalized to other
cases of head-spec case assignment, then at least the number of case assignment configurations
could be reduced from three to two—a step in the direction of simplifying the theory of case
assignment.
Subject movement would then have to be derived in some other way—from some other
principle. A possible candidate is predication (that is, the subject requirement), if it is understood
as requiring subjects to be realized outside of their Aristotelian predicates. Given the word order
facts of English, it is very difficult to decide between the two case assignment alternatives. We
do know that nominative case assignment in the head-spec configuration is possible because of
languages like Irish, where finite verbs are clause-initial, preceding the subject, yielding VSO
order. But then it is not clear how predication as just defined is satisfied in languages of this
type. It might be suggested that predication is satisfied at the VP level in Irish and at the IP level
in English (that is, what counts as an Aristotelian predicate differs between the languages—V′
in Irish versus I′ in English. But that is actually just a restatement of the problem and begs the
question of why predication can’t be satisfied at the VP level in English also. What’s even worse
is that predication is satisfied at the VP level in English in small clauses headed by V (recall (36)
and especially (39a)).
As a result of these considerations, we continue to assume that case can be assigned in the
head-spec configuration.

Notes 5
1. A very few German verbs govern a third case, the genitive. These are felt to be archaic by present-day speakers, so
we ignore them here.

2. The -s-less plural of deer, which is exceptional in modern English, goes back to Old English, where it was simply the
ordinary plural form for the declension to which deor ‘animal’ belonged.
 Chapter 7: Case theory 230

3. There is currently a shift occurring in American English where singular they is being extended and popularized
as a singular (ungendered) animate pronoun, leveling the distinctions between singular and plural, but retaining
case-related distinctions (they vs. them).

4. In addition to case government, which holds between a head and a noun phrase, there is also the phenomenon of case
agreement, where constituents within a noun phrase all share the same case. If case agreement exists in English,
it is invisible, but we know it exists in other languages from examples like (4) and (5). The mechanisms for case
assignment that we discuss below would have to be extended to cover case agreement. In this book, however, we
will put the issue aside.

5. The spec-head configuration is sometimes referred to as spec-head agreement on the grounds that subjects and
verbs of sentences, which are in the spec-head configuration in the VP, agree in number (the dog runs/*run; the dogs
run/*runs). We avoid the term because there is no one-to-one correspondence between the topological configura-
tion and agreement morphology. For instance, neither Spec(DP) nor Spec(NP) exhibit agreement with their heads.
Conversely, determiners do agree with the head of their NP complement. Moreover, in many languages, adjectives
agree with the nouns they modify and with the subjects that they are predicated of.

(i) { that, *those } house; { those, *that } houses

(ii) a. un-e maison { grand -e, *grand -∅ } (French)
a-f house.f big -f -m
‘a big house’
b. ένα { μεγάλ -ο, *μεγάλ -ος, μεγάλ -η } σπίτι (modern Greek)
en-a { megal -o, *megal -os, *megal -i } spiti
a-n big -n -m -f house
‘a big house’
(iii) a. La maison est { grand -e, *grand -∅ } . (French)
the.f house.f is big -f -m
‘The house is big.’
b. Το σπίτι είναι { μεγάλ -ο, *μεγάλ -ος, μεγάλ -η } (modern Greek)
To spiti ine { megal -o, *megal -os, *megal -i } .
the.n house.n is big -n -m -f
‘The house is big.’

But the configurations between the trigger of agreement (noun) and the agreeing constituent (determiner, adjective)
is different for each of these three phenomena and far from being the spec-head configuration.

6. The higher finite I in the clauses in (31) can’t assign case to the lower subject, since the path connecting the two is
nothing like the requisite spec-head configuration. Instead, it assigns nominative to the higher subject (its “own”
subject, as it were).

7. One might attempt to salvage part of the original proposal by replacing the current licensing condition in (i.a) with
the logically possible alternative in (i.b).

(i) a. Nominative case is assigned by finite I.

b. Nominative case is assigned by finite V where possible and by finite I otherwise.

Although there is no empirical argument against (i.b), we reject it because it violates conceptual economy
(or Occam’s razor). Our reasoning is as follows. Finite V implies finite I (in the form of a silent tense morpheme),
but finite I is consistent with either finite or nonfinite V (as exemplified in (30)). The set of clauses with finite I is
therefore a proper superset of the set of clauses with finite V. As a result, (i.a) and (i.b) are empirically equivalent
(the “otherwise” clause in (i.b) is guaranteed to kick in). But then the statement in (i.b) is redundant and therefore
less preferable.
Chapter 7: Case theory 231

Exercises and problems 6

Exercise 7.1
A. Table 7.6 summarizes the case assignment facts for English discussed in the chapter. A few
values remain to be filled in based on the evidence in (1).

Head Case Configuration

Adj n/a
D, possessive poss spec-head
I, finite nom spec-head
I, nonfinite (to)
N n/a
P obj head-comp
V obj head-comp, head-spec

Table 7.6: Incomplete summary of case assignment

(1) a. * { Us, we, our } to leave now would not be helpful.

b. For { ✔ us, *we, *our } to leave now would not be helpful.

B. Why is there no entry (not even a “n/a” entry) for ordinary (= nonpossessive) D?

Exercise 7.2
According to the analysis in the text, why are the sentences in (1) ungrammatical? You don’t
need to be concerned with the structure of the entire sentence; just treat the underlined part as
an IP headed by nonfinite I.

(1) a. * They claim to they understand Hegel.

b. * They claim they to understand Hegel.
Chapter 7: Case theory 232

Exercise 7.3
How is the thematic relation between leave and its argument us represented in (37b)? How about
in (37a)?

Exercise 7.4
A. Build structures for the sentences in (1). You can use triangle notation for all DPs. As-
sume that for can take IP complements. Assume further that for clauses can be subjects of
sentences; in other words, PPs can substitute in Spec(VP) and move to Spec(IP).

(1) a. That should be them now.

b. They waited for us.
c. We waited for there to be a sale.
d. For daycare to be available for the children would be convenient.
e. For daycare to be available would be convenient for the parents.
f. I suspect the class to be difficult.

B. How do each of the DPs in (1) get case? Which case is assigned by what head in what
configuration? Group similar cases.

Exercise 7.5
Can you think of a way of assigning objective case to the adjuncts in (1), without introducing a
new case assignment configuration?

(1) a. We discussed the proposal the very next week.

b. We kept their books five years.
c. We don’t read the Times every single day

Exercise 7.6
In the chapter, we stated that nouns and adjectives aren’t case-assigners in English. Provide
evidence for that statement. One piece of evidence for each category is sufficient.

Exercise 7.7
This exercise extends Exercise 5.4.
A. If you have not already done so for Exercise 5.4, build structures for the underlined DPs in
(1). Here, you can use triangle notation for both constituent DPs (Kim and incompetents).

(1) a. We dislike Kim’s impulsive hiring of incompetents.

b. We dislike Kim’s impulsively hiring incompetents.

B. How do the constituent DPs get case? Which case is assigned by what head in what con-
figuration?
Chapter 7: Case theory 233

Exercise 7.8
This exercise extends Exercise 7.1, A.
A. English has a silent nonfinite gerund head ([geR]) with the elementary tree in (1).

(1) IP

»DP« I′

I◦ »VP«
[geR]

The VP complement of this head must be headed by a verb form in -ing. Based on the
evidence in (2), determine this head’s case assignment properties.

(2) { ✔Them, *they, *their } being late, we cancelled the meeting.

Exercise 7.9
A. The verb assigns case to us in the head-spec configuration in both (1a) (= (37b)) and (1b), but
(1b) is completely ungrammatical. Why is that? The intended meaning of (1b) is expected
our departure.

(1) a. ✔ VP

»DP« V′

V◦ IP
expected

DPk I′

us I◦ VP
to
tk V′

V◦
leave
Chapter 7: Case theory 234

b. * VP

»DP« V′

V◦ DP
expected

DPi D′

us D◦ NP
the
ti N′

N◦
departure

B. In (2), V assigns objective case to the lower DP us in the head-spec configuration, as indi-
cated by the boxes, and the possessive head assigns possessive case to the higher DP, as
indicated by the circles. Recall that possessive case is invisible on full DPs. Again, the in-
tended interpretation is expected our departure. But again, as in (1b), the result is completely
ungrammatical. What has gone wrong?

(2) * VP

»DP« V′

V◦ DP
expected

DPi D′

us
D◦ NP
[poss]
ti N′

N◦
departure
Chapter 7: Case theory 235

Problem 7.1
This exercise extends Exercise 7.1, A.
English has (now very archaic) PPs of the type in (1). Assume that there is a DP, just like
expletive there (even though the there at issue here is referential).

(1) thereafter (= after that), therefrom (= from that), therewith (=with that)

What would you suggest as the elementary tree shared by all of the relevant prepositions,
and in what configuration do they assign case to there?
Note: There are several solutions.

Problem 7.2
In English, there are no case-assigning determiners without a specifier. Is such a determiner
possible in principle?

Problem 7.3
From a purely graph-theoretical point of view, the head-spec configuration is a mirror image of
the spec-head configuration in the following sense. In both, a case feature travels from a head X
to X′ . At X′ it either moves up a node to XP or down a node to Comp(XP). In both cases, it then
continues down a node, arriving at either Spec(XP) or Spec(Comp(XP). Given this relation, the
mirror image of the head-comp configuration might be expected to be a fourth case assigment
configuration. In the head-comp configuration, the head feature travels from X up to X′ and then
down to Comp(XP). In the mirror image, it would move up to XP instead. Why is this expectation
not actually sensible?

Problem 7.4
In the examples discussed in the text, there is always a one-to-one correspondence between case-
assigning heads and noun phrases. What would happen if we relaxed this condition?
A. In particular, give an example of a linguistic expression that would be grammatical if we
allowed a single head (say, V) to assign case to two noun phrases in the configurations that
we have discussed?
B. Conversely, give an example of a linguistic expression that would be grammatical if we
allowed a single noun phrase (or more precisely, a single chain) to get case from more than
one head?
 8
Passive

Chapter outline
1 Characteristics of the passive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
2 A movement analysis of the passive . . . . . . . . . . . . . . . . . . . . . . . . . . 239
2.1 Selectional restrictions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
2.2 Object idiom chunks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
2.3 Passive of simple sentences . . . . . . . . . . . . . . . . . . . . . . . . . . 242
2.4 Passive of ECM sentences . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
3 Further issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
3.1 Impersonal passive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
4 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250
5 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251

In this chapter, we first present some general characteristics of the passive. We then moti-
vate a movement analysis of it based on considerations of selectional restrictions and object
idiom chunks, beginning with simple sentences and then covering the ECM sentences that we
introduced in the previous chapter in connection with spec-head case assignment. We conclude
with a quick look at impersonal passives, which are available in languages with richer case mor-
phology than English.

Characteristics of the passive 1

In English, as in many other languages, active sentences like (1a) have passive counterparts like
(1b).
Chapter 8: Passive 237

Background Info

In the remainder of this section, grammatical functions are indicated above the boxed ele-
ment and thematic roles are indicated below. The following abbreviations are used:

sbj - subject, obj - (direct) object, a - agent, thm - theme

sbj obj
(1) a. We approved them .
a thm
sbj pp
b. They were approved (by us) .
thm a
Passivization has a number of effects. First and foremost, the agent argument, which is
expressed as the subject of the active sentence, appears in the passive as an optional by phrase.
Second, the now vacant subject position is taken over by the theme argument. In other words, the
agent and theme arguments are linked to different grammatical relations in the passive than
in the active. If we think of subject as outranking object on a hierarchical scale of grammatical
relations (as is often assumed), then the passive demotes the agent argument and promotes the
theme argument. Third, passive past participles, unlike their homonymous active counterparts,
can’t assign objective case.

(2) a. ✔ We approved them. (active)

b. * Itexpl was approved them (by us). (passive)
Intended meaning: ‘They were approved (by us).’

Word of Caution
English has a class of verbs called inchoative vebrs, discussed in Chapter 11 and illustrated
in (i), that have a passive flavor, but are not passive verbs.

(i) The vase dropped.

As with passive sentences, sentences containing these verbs have subjects that are themes
rather than agents. However, unlike passives, they do not allow the expression of an agent,
as shown in (ii).1

(ii) * The vase dropped by the child. (ungrammatical on agentive reading of by-phrase)

In English, the passive is expressed analytically by a combination of the past participle and
auxiliary be. Other languages allow the passive to be expressed synthetically, as illustrated in (3)
for Korean and in (4) for Latin.

(3) a. Active:
Chapter 8: Passive 238
Chapter 0: 2

✔ sbj obj
철수가 계획을 허락하였다.
Chelswu-ka kyehoyk-ul helak-ha-yessta.
Chelswu-nom plans-acc approve-act-pst
a thm
Chapter
‘Chelswu 0:
approved the plans.’ 3
b. Passive:
✔ sbj pp
계획이 (철수에 의하여) 허락되었다.
Kyehoyk-i Chelswu-eyuyhaye helak-toy-yessta.
plans-nom Chelswu-by approve-pass-pst
thm a
Chapter
‘The plans0:were approved (by Chelswu).’ 4
(4) a. Active:
✔ sbj obj
Puer-ī port-am claud-unt.
boy-nom.pl.m door-acc.sg.f close-3pl.pRs
a thm
Chapter
‘The boys0:
are closing the door.’ 5
b. Passive:
✔ sbj obj
Port-a (a puer-īs) claud-it-ur.
door-nom.sg.f by boy-abl.pl.m close-3sg.pRs-pass
thm pp
‘The door is being closed (by the boys).’

In these languages, it is the bound morphemes -toy- and -ur that result in the effects of
passivization mentioned above. As in English, the grammatical relations of the agent and theme
arguments differ in the active and the passive, and passive verb forms cannot assign the case that
active verb forms can.

(5) a. * 계획을 (철수에 의하여) 허락되었다.

Kyehoyk-ul (Chulswu-eyuyhaye) helak-toy-essta.
plans-acc Chulswu-by approve-pass-pst
Intended meaning: ‘The plans were approved (by Chulswu).’
b. * Port-am (a puer-īs) claud-it-ur.
door-acc.sg.f by boy-abl.pl.m close-3sg.pRs-pass
Intended meaning: ‘The door is being closed by the boys.’

Latin (among other languages) also uses the analytic strategy for the passive in certain
linguistic contexts. In particular, the passive voice in the so-called perfect tenses is formed by
combining the passive participle (ptcp.pass) with the appropriate form of esse ‘be’, as illustrated
in (6).
Chapter 8: Passive 239
Chapter 0: 6

(6) ✔ sbj obj

Port-a (a puer-īs) claus-a { est, erat, erit }.
door-nom.sg.f by boy-abl.pl.m close-ptcp.pass.nom.sg.f is was will.be
thm pp
‘The door { has, had, will have } been closed (by the boys).’

It is worth noting that the synthetic passive form in (4b) agrees with the subject in number,
but that the participle in the analytic form in (6) shows additional case and gender agreement.
Case and gender are ordinarily features of the nominal system (nouns and their canonical mod-
ifiers, adjectives), so participles share features of both verbs and adjectives. Indeed, the term
‘participle’, derived from Latin particeps ‘sharing’, implies as much.

A movement analysis of the passive 2

2.1 Selectional restrictions

Before proceeding, we need to introduce the concept of selectional restriction. Selectional
restrictions are conditions that a (Fregean) predicate imposes on one or more of its arguments,
depending on its meaning. For instance, drink imposes a selectional restriction on its theme
argument to the effect that the theme argument must refer to a kind or amount of liquid.

(7) a. Amy drank the { lemonade, #sandwich }.

b. Lukas drank a whole { quart, #piece }.

The predicate elapse selects a subject that refers to an amount of time.

(8) { Two hours, the remainder of the shift, #two liters, #Larry } elapsed without further inci-
dent.

The felicitous use of the verb murder requires (among other conditions) that both the agent and
the theme arguments refer to humans (and arguably, also other species—see below for related
discussion). By contrast, kill imposes weaker selectional restrictions, requiring only that the
theme argument refer to a living being.

(9) The { ✔ paramilitary, #bomb, #avalanche } murdered { ✔their parents, #the olive tree, #their
house }.
(10) The { ✔ paramilitary, ✔bomb, ✔avalanche } killed { ✔their parents, ✔the olive tree, #their
house }.

Two points are important to keep in mind concerning selectional restrictions. First, notice
that we are using pound signs, rather than asterisks, in (7)–(10); in other words, we are treating the
Chapter 8: Passive 240

ill-formedness of sentences that violate selectional restrictions as semantic/pragmatic deviance,

not as ungrammaticality. This approach is consistent with the fact that selectional restrictions
can be flouted for special effect. For example, the ordinary (literal) meaning of lap up is ‘to
drink, usually eagerly, using the tongue, hence especially of animals’. Based on this meaning, we
would expect lap up to select a nonhuman animate agent and a liquid theme. But although both
restrictions are violated in (11), the sentence does not come across as deviant.

(11) The students lapped up their teacher’s praise.

Rather, the violation of the selectional restrictions signals to the hearer that the sentence is in-
tended to be taken not literally, but figuratively—as an instance of metaphor. (12) summarizes the
kind of reasoning that a hearer of (11) would go through; the reasoning process itself is ordinarily
not explicit, but subconscious and lightning-quick.

(12) The students lapped up their teacher’s praise⁇ Whoa there, that’s complete nonsense!
It’s only nonhuman animals that lap up things. And even then, whatever they’re lapping
up has to be liquid, not something abstract like praise.
But the speaker seems to know English and be compos mentis, so what could they have
possibly meant by what they said?
Oh, maybe what they mean is that the attitude of the students towards their teacher’s
praise resembles the eagerness with which a thirsty animal laps up some welcome liquid.

Word of Caution
In distinguishing figurative from literal uses of language, don’t let yourself be confused by
the fact that in the vernacular, the adverb literally is routinely used to draw attention to
a statement’s figurative character, as when people say things like My boss literally hit the
roof. In other words, literally has come to mean figuratively!

Don’t, by the way, conclude from examples like (11) that selectional restrictions are in force
only intermittently (in force when language is used literally, but not in force when language is
used figuratively). Rather, it is precisely the fact that selectional restrictions are always in force
that prompts a hearer of (11) to go through the reasoning process outlined in (12) and to come
up with an interpretation in which the selectional restrictions are satisfied at the level of the
metaphor.
A second point to keep in mind concerning selectional restrictions is that the criteria for
set membership that the restrictions are based on are not always crystal-clear. In other words,
sets like liquid things, animate beings, murderers, or murder victims, and so on, are somewhat
fuzzy around the edges. Speakers might disagree, for instance, about whether the sentences in
(13) are deviant; the disagreement would concern whether the selectional restrictions on murder
might, on the basis of recent advances in the understanding of animal intelligence, be broadened
to include members of species other than Homo sapiens.

(13) a. The { chimpanzee, dolphin } murdered the explorer.

b. The explorer murdered the { chimpanzee, dolphin }.
Chapter 8: Passive 241

Fortunately, for our purposes in this chapter, locating the exact boundary between cases
that meet selectional restrictions and ones that violate them will not be necessary. The important
thing is that selectional restrictions exist, and that there are sentences in which they are clearly
met and ones in which they are clearly violated.
As we have just seen in (9), subjects or objects can violate a verb’s selectional restrictions.
In what follows, we focus on the objects. For convenience, the relevant facts are repeated in
(14).

(14) a. The paramilitary murdered { ✔their parents, #the olive trees, #their houses }.
b. The paramilitary killed { ✔their parents, ✔the olive trees, #their houses }.

Example (15) gives the passive versions of these sentences. As is evident, the acceptability of the
subjects in (15) matches that of the objects in (14).

(15) a. { ✔their parents, #the olive trees, #their houses } were murdered by the paramilitary.
b. { ✔their parents, ✔the olive trees, #their houses } were killed by the paramilitary.

This fact would follow straightforwardly if the theme subject started out in Comp(V), where it
would be have to satisfy the verb’s selectional restrictions in the same was in the active. Having
been cleared, so to speak, it could then move to the subject position, where we hear it. This is the
intuition that we will pursue below.

2.2 Object idiom chunks

Additional evidence pointing in the same direction comes from so-called object idiom chunks,
which generally appear as the complements of the verbs that license them (the examples in this
section are taken from or modeled on Radford 1988, pp. 422–423). Some examples are given in
(16). The object idiom chunks are underlined, and the licensing verbs are italicized in green.

(16) a. They { gave, paid } hardly any heed to our proposal.

b. The Prime Minister paid due homage to the dead.
c. The government keeps close tabs on their operations.

The restriction of object idiom chunks to the complement position of the licensing verb is
thrown into striking relief by the contrast between nearly synonymous expressions, such as heed
and attention. The variants with the ordinary expressions are grammatical, but those with the
idiom chunks are not, since they are not licensed by the verbs bolded in red.

(17) a. The younger one is always trying to attract my { *heed, ✔attention }.

b. According to the government, the situation requires { *tabs on, ✔monitoring of } the
activists.

Given their licensing requirements, it isn’t surprising that object idiom chunks are gener-
ally ungrammatical in subject position.

(18) a. * More heed to maintenance would result in lower repair bills.

Chapter 8: Passive 242

b. * Tabs won’t affect me.

They are, however, able to occur in subject position under one condition—in passive sen-
tences where the passive participle is that of the licensing verb. This is illustrated by the contrast
between (19) and the passive counterparts of (17).

(19) a. ✔ Hardly any heed was { given, paid } to our proposal.

b. ✔ Due homage was paid by the Prime Minister to the dead.
c. ✔ Close tabs were kept on their operations by the government.

2.3 Passive of simple sentences

Our analysis of the passive is based on the premise that theme arguments originate in the same
structural position in both the active and the passive. This means that the elementary trees for
active and passive participles both contain a complement position. However, the relevant ele-
mentary trees differ in two important ways. First, in the active, the agent argument is obligatorily
linked to (= expressed in) Spec(VP), whereas in the passive, it is linked to an optional adjunct
by phrase. We will visually underline this by omitting Spec(VP) in our elementary trees for pas-
sive participles. Second, passive participles in English (and passive verbs more generally) cannot
assign objective case; recall the ungrammaticality of (2b). (We will generally not represent this
property in the elementary tree explicitly.) The elementary trees we propose for active and pas-
Chapter
sive participles 0: as shown in (20).
are thus Chapter 0: 7

(20) a. VP b. VP

»DP« V′ V′

V◦ »DP« V◦ »DP«
active verb passive verb

Active verb: assigns case Passive verb: cannot assign case

In what follows, we illustrate the derivation of a passive sentence like (21).

(21) The proposal was adopted.

First, we substitute the theme argument the proposal in the elementary tree for the passive main
verb in (20b). This yields (22a). Then, we substitute (22a) as the complement of the passive
auxiliary verb be, as in (22b). In order to distinguish the auxiliary verb (be) from the main verb
(adopted), we give it the syntactic category Aux, but we could just as easily treat it as another
instance of V. For simplicity, we will treat auxiliary verbs as lacking specifier positions. Finally,
we substitute the resulting structure as the complement of I, as in (22c).2
Chapter 8: Passive 243
Chapter 0: Chapter 0: 9

(22) a. VP b. AuxP

V′ Aux′

V◦ DP Aux◦ VP
adopted was
the proposal V′

V◦ DP
adopted
the proposal

Substitute theme argument in (20b) Substitute (22a) as complement

Chapter 0: of passive auxiliary 11

c. IP

»DP« I′

I◦ AuxP
[pst]
Aux′

Aux◦ VP
was
V′

V◦ DP
adopted
the proposal

Substitute (22b) as complement of I

Because of the inability of the passive participle to assign objective case, the theme argu-
ment cannot receive case in the complement position. Since every DP must receive case, the
theme argument must move to the closest position to which case can be assigned. This position
is Spec(IP), where it is possible for the theme argument to receive nominative case from finite I.
The resulting final structure is shown in (23).
Chapter 8: Passive 244
Chapter 0: 12

(23) IP

DPi I′

the proposal I◦ AuxP

[pst]
Aux′

Aux◦ VP
was
V′

V◦ ti
adopted

2.4 Passive of ECM sentences

This section focuses on the passive of ECM verbs like expect, which were introduced in Chapter 7
in connection with head-spec licensing. (24a) shows an active ECM verb sentence, and (24b)
shows the corresponding passive.

(24) a. ✔ Your folks expect you to call.

b. ✔ You are expected to call.

The elementary tree for expect in (24a) is given in (25a). In accordance with the previous
discussion, the elementary tree for the passive participle expected is as in (25b). The difference
between the two trees is analogous to that between the trees in (20); the only difference is the
syntactic category of the complement (DP in the case of ordinary verbs, IP in the case of ECM
verbs). As in (20b), the elementary tree (25b) is missing a specifier position, and the verb lacks
Chapter
the ability to 0:
assign objective case. Chapter 0: 13

(25) a. VP b. VP

»DP« V′ V′

V◦ »IP« V◦ »IP«
expect expected

Active ECM verb: Passive ECM verb: cannot assign

assigns case case

In what follows, we illustrate the derivation of (24b). The derivation of the complement
clause is shown in (26); we assume that the complement subject moves from Spec(VP) to Spec(IP)
Chapter 8: Passive 245

to provide the complement clause with a subject (recall the subject requirement discussed in
Chapter 3). Chapter 0: Chapter 0: Chapter 0: 15

(26) a. VP b. IP c. IP

DP V′ »DP« I′ DPi I′

you V◦ I◦ VP you I◦ VP
call to to
DP V′ ti V′

you V◦ V◦
call call

Complement VP Substitute (26a) as Move subject of

complement of complement clause
nonfinite I

Chapter
The subsequent steps0:of the derivation involving the matrix
Chapter
clause 0:
are as shown in (22). 18

(27) a. VP b. AuxP
V′ Aux′
V◦ IP Aux◦ VP
expected are
DPi I′ V′
you I◦ VP V◦ IP
to expected
ti V′ DPi I′
V◦ you I◦ VP
call to
ti V′
V◦
call

Substitute (26c) as complement of Substitute (27a) as complement of

passive participle of ECM verb passive auxiliary
Chapter 8: Passive 246
Chapter 0: 20

c. IP
»DP« I′
I◦ AuxP
[pRs]
Aux′
Aux◦ VP
are
V′
V◦ IP
expected
DPi I′
you I◦ VP
to
ti V′
V◦
call

Substitute (27b) as complement of matrix I

In (26c), the passive participle of the ECM verb, being passive, cannot assign case to the subject
of the complement clause, which must therefore move to the nearest case position. This is the
matrix Spec(IP),
Chapter where
0: it receives nominative case from finite I. The resulting structure is shown 21
in (28).

(28) IP
DPi I′
you I◦ AuxP
[pRs]
Aux′
Aux◦ VP
are
V′
V◦ IP
expected
ti I′
I◦ VP
to
ti V′
V◦
call
Chapter 8: Passive 247

Further issues 3

3.1 Impersonal passive

In languages with morphologically richer case systems than English, there is evidence for a dis-
tinction between two types of case: structural and inherent. For instance, the following Ger-
man examples show that the active participle unterstützt ‘supported’ assigns accusative case, but
that the homonymous passive participle cannot. Instead, in the passive, finite I assigns nomina-
tive case to the theme argument in subject position, as in English.

Extra Info
The ß character (known as ‘sharp s’) represents ‘ss’ in certain contexts.

As usual, the German examples are in the form of subordinate clauses in order to avoid the
complications associated with the verb-second phenomenon covered in Chapter 12.

(29) Active:
✔ daß wir dies-en Kandidat-en unterstützt haben
that we this-acc candidate-acc supported have
‘that we have supported this candidate’
(30) Passive:
a. ✔ daß dies-er Kandidat unterstützt wurde
that this-nom candidate.nom supported became
‘that this candidate was supported’
b. * daß dies-en Kandidat-en unterstützt wurde
that this-acc candidate-acc supported became
Intended meaning: ‘that this candidate was supported’

As we know from Chapter 7, there are also German verbs that assign the dative. Such
verbs—for instance, helfen ‘help’—continue to assign the dative even in passive sentences, as
shown in (31) and (32).3

(31) Active:
✔ daß wir dies-em Kandidat-en geholfen haben
that we this-dat candidate-dat helped have
‘that we have helped this candidate’
(32) Passive:
a. * daß dies-er Kandidat geholfen wurde
that this-nom candidate.nom helped became
Intended meaning: ‘that this candidate was helped’
Chapter 8: Passive 248

b. ✔ daß dies-em Kandidat-en geholfen wurde

that this-dat candidate-dat helped became
‘that this candidate was helped’

The accusative and the dative are both assigned in the head-comp configuration by the
verbs unterstützen ‘support’ and helfen ‘help’, respectively, but they differ in that the accusative
alternates with the nominative, whereas the dative doesn’t. What we mean by ‘alternate’ is
simply that an accusative object in the active corresponds to a nominative subject in the passive,
whereas a dative object in the active remains dative in the passive. The alternating cases (nom-
inative, accusative) are known as structural cases (because which one occurs depends on the
structural position of the argument bearing the case), whereas the nonalternating case (dative) is
known as an inherent case.
The grammaticality of (32b) raises the question of what element receives nominative case
from finite I. It is generally assumed that German has a silent expletive element (∅expl ), corre-
sponding to English expletive it or there, that occupies Spec(IP) and receives case there. The
structure of (32b)0:is then as in (33). (The structure in (33) ignores that the Aux moves and adjoins
Chapter 22
to I; that movement, described in Chapter 10, is irrelevant for present purposes.)

(33) CP

C◦ IP
daß
DP I′

∅expl AuxP I◦
[pst]
Aux′

VP Aux◦
wurde
V′

DP V◦
geholfen
diesem Kandidaten

Just like its overt counterpart es ‘it’,4 the silent expletive bears a number feature with the
default value ‘singular’. Evidence for this comes from the number agreement pattern in (34): the
overt plural subject in (34a) agrees with a plural verb form, whereas the silent expletive in (34b)
agrees with the corresponding singular verb form.

(34) a. daß dies-e Kandidat-en unterstützt { wurd-en, *wurd-e }

that these-nom.pl candidates-nom.pl supported became-pl became-sg
‘that these candidates were supported’
Chapter 8: Passive 249

b. daß ∅expl dies-en Kandidat-en geholfen { wurd-e, *wurd-en }

that these-dat.pl candidates-dat.pl helped became-sg became-pl
‘that these candidates were helped’

Like German, Latin is a verb-final language with rich morphological case, and it patterns
like German with regard to structural versus inherent case. As in German, the structural cases
are nominative and accusative. The inherent cases include the remaining cases (apart from the
vocative)—that is, the dative, genitive, and ablative. (35)–(38) illustrate the parallel with German,
using the dative as a representative for the inherent cases.

(35) Active:
✔ Avuncul-us puer-ōs adiuva-t.
uncle-nom boys-acc help-3sg.pRs
‘The uncle is helping the boys.’
(36) Passive:
a. ✔ Puer-ī adiuva-nt-ur.
boys-nom help-3pl.pRs-pass
‘The boys are being helped.’
b. * Puer-ōs adiuva-nt-ur.
boys-acc help-3pl.pRs-pass
Intended meaning: ‘The boys are being helped.’
(37) Active:
✔ Avuncul-us puer-īs subveni-t.
uncle-nom boys-dat help-3sg.pRs
‘The uncle is helping the boys.’
(38) Passive:
a. * Puer-ī subveni-unt-ur.
boys-nom help-3pl.pRs-pass
Intended meaning: ‘The boys are being helped.’
b. ✔ Puer-īs subvenī-t-ur.
boys-dat help-3sg.pRs-pass
‘The boys are being helped.’

Passives with alternating case are sometimes called personal passives, and ones with
inherent case, impersonal. German allows even more radical impersonal passives than the ones
with helfen just presented, going so far as to allow passives of intransitive verbs like tanzen ‘dance’
and others, as illustrated in (38).

(39) daß ∅expl { getanzt, gelaufen, gefochten } wurde

that danced run fought became
‘that there was dancing, running, fencing (lit. fighting)’

This is also the case in Latin (Allen and Greenough 1903, § 207).
 Chapter 8: Passive 250

(40) Curre-ba-t-ur, pugna-ba-t-ur ‘There was { running, fighting }.’

run-pst-3sg-pass fight-pst-3sg-pass

It is striking, however, that not every intransitive verb can be passivized. Verbs with sub-
jects that are already themes, like the inchoative verbs mentioned at the beginning of the chapter,
cannot. We give examples for German only; we have no native speakers of Latin to consult, and
the examples, being ungrammatical, would of course be unattested in the texts that have come
down to us. Note that gefallen is lexically ambiguous between ‘fall’ and ‘please’; both meanings
are relevant here.

(41) * daß ∅expl { aufgefallen, gefallen, gefallen } wurde

that been.noticeable fallen pleased became
Intended meaning: ‘that there was being noticeable, falling, pleasing’

Notes 4
1. By the criterion just mentioned, participial constructions as in (i) are true passive constructions, despite the absence
of a passive auxiliary.

(i) a. The shrubs planted next to the fence (by our neighbors) are azaleas.
b. Planted in the shade (by an ignorant gardener), the cactus failed to thrive.

Note the close parallel with the constructions in (ii), which do contain a passive auxiliary.

(ii) a. The shrubs that were planted next to the fence (by our neighbors) are azaleas.
b. Having been planted in the shade (by an ignorant gardener), the cactus failed to thrive.

2. As we will see in Chapter 10, auxiliary be and have move from their original position to I, where they form a complex
verb with the tense morpheme. As this movement is irrelevant for present purposes, we ignore it here and throughout
the chapter.

3. Certain archaic German verbs govern the genitive, which patterns as an inherent case, as shown in (i) and (ii), which
are parallel to (31) and (32).

(i) Active:

✔ daß wir der Ahn-en gedenken

that we the.gen.pl ancestors-gen.pl commemorate
‘that we commemorate the ancestors’

(ii) Passive:

a. * daß die Ahn-en gedacht werd-en

that the.nom.pl ancestors-nom.pl commerated become-pl
Intended meaning: ‘that the ancestors are commemorated’
 Chapter 8: Passive 251

b. ✔ daß der Ahn-en gedacht wird

that the.gen.pl ancestors-gen.pl commemorated becomes.sg
‘that the ancestors are commemorated’

4. In case you are wondering whether the silent expletive in the impersonal passive can be replaced with the overt one,
the answer is ‘no’. The fact is clear, but the reason for it is not yet fully understood.

Exercises and problems 5

Exercise 8.1
Which, if any, of the following sentences are true? Support your conclusions with evidence.

(1) a. All subjects are agents.

b. All agents are subjects.
c. All subjects get nominative case.
d. All noun phrases that get nominative case are subjects.

Exercise 8.2
Build structures for the sentences in (1).

(1) a. The puppeteers might get arrested.

b. The puppeteers got arrested.

Exercise 8.3
A. Build structures for the sentences in (1) and (2).

(1) a. We expect them to make headway.

b. We expect headway to be made.
c. Headway is expected to be made.
(2) a. The media expect the guerrillas to release the journalist.
b. The media expect the journalist to be released by the guerrillas.
c. The journalist is expected to be released by the guerrillas.

B. One of the sentences in (2) is structurally ambiguous. Which one is it, and why?
Chapter 8: Passive 252

Exercise 8.4
This exercise extends Exercise 3.4.
Explain the grammaticality pattern in (1), including how expletive there is licensed. Do
noun phrases (DPs) differ from clauses (CPs, IPs) with respect to case assignment? Assume for
simplicity that expletive it substitutes directly into Spec(IP).

(1) a. It was suspected that there was a problem with the O-ring.
b. That there was a problem with the O-ring was suspected.
c. There was suspected to be a problem with the O-ring.
d. * There to be a problem with the O-ring was suspected.

Exercise 8.5
The active sentence in (1) has two conceivable passive counterparts in (2): the grammatical (2a)
and the completely ungrammatical (2b) (intended to have the same meaning as (2a)). Why is (2b)
ungrammatical?

(1) The administration will expect us to hire them.

(2) a. ✔ We will be expected to hire them.
b. * { They, them } will be expected { we, us } to hire.

Problem 8.1
Explain the following pattern of judgments.

(1) ✔ The carpenter measured the boards.

(2) ✔ The boards were measured by the carpenter.
(3) a. ✔ The tickets cost 100 dollars.
b. ✔ The concert lasted two hours.
c. ✔ The carpenter measure 5′10″.
d. ✔ Louis XIV reigned 72 years.
e. ✔ The trip took two hours.
(4) a. * 100 dollars are cost by the tickets.
b. * Two hours were lasted by the concert.
c. * 5′10″ is measured by the carpenter.
d. * 72 years were reigned by Louis XIV.
e. * Two hours were taken by the trip.
Chapter 8: Passive 253

Problem 8.2
Consider the following judgments for the sentences with expect in (1) and (2).

(1) a. ✔ They expected there to be a solution.

b. ✔ There was expected to be a solution.
(2) a. * They expected for there to be a solution.
b. * There was expected for to be a solution.

Now consider the corresponding judgments for sentences with want in (3) and (4).

(3) a. ✔ They wanted there to be a solution.

b. * There was wanted to be a solution.
(4) a. ✔ They wanted for there to be a solution.
b. * There was wanted for to be a solution.
A. The judgments for the sentences with expect are as expected given our discussion in this
and the previous chapter. The judgments for the sentences with want can be made to be
consistent with the discussion as well. How so?
B. For some speakers, (2a) is grammatical. The remaining judgments remain the same. Revise
your answers to (A) to account for this pattern of judgments.
Subject control versus
9
raising

Chapter outline
1 Subject control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256
1.1 Evidence for two clauses . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256
1.2 Deriving subject control sentences . . . . . . . . . . . . . . . . . . . . . . 258
2 Raising . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
2.1 A “what if” detour . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
2.2 Nonthematic subject positions . . . . . . . . . . . . . . . . . . . . . . . . 262
2.3 Deriving raising sentences . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
3 Some special verbs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
3.1 Tend versus occur . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
3.2 Promise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
4 More nonthematic subjects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
4.1 Subject idiom chunks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
4.2 Weather it . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
4.3 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
5 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272
6 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274

So far in this book, we have come across three types of clausal complements, summarized in
the following table. (By ‘clausal complement’, we mean any complement that contains a subject
and a predicate.) The clausal complement is bracketed, and the complement’s I head, if any, is in
boldface.

Type of complement I is… Matrix Complement

Finite complement modal or tense We heard [ that the children [pst] danced ].
ECM complement nonfinite to We expected [ the children to dance ].
Small clause absent We saw [ the children dance ].

Table 9.1: Summary of clausal complements up to this point

Chapter 9: Subject control versus raising 255

Despite their superficial diversity, the complement clause types in Table 9.1 all have one property
in common—the presence of an overt subject (here, the children). But English also has nonfinite
complement clauses where a subject is understood, but not expressed overtly. For instance, dance,
the verb in the apparently subjectless complement clauses in (1), has an understood agent.

(1) a. Subject control:

The children agreed [ to dance ].
b. Raising:
The children seemed [ to dance ].

In both sentences, this agent is interpreted as being identical to the referent of the matrix subject
the children. Yet unlike the second row of Table 9.1, where the matrix clause and the complement
clause each have their own subject (we, the children), the sentences in (1) contain only a single
overt subject, the one in the matrix clause. In this chapter, we argue that the nonfinite comple-
ments in (1) contain a structural subject position, just as in Table 9.1, which is filled by a silent
element, and we argue further that this silent element is not the same in the two examples. Rather,
we distinguish between subject control and raising (sometimes called subject-to-subject rais-
ing), as already indicated by the labels in (1).

Background Info

Some of the discussion here uses terms like reference, refer, and corefer (among others).
If these are not clear from the glossary and our explanations here, you can consult §1 of
Chapter 15 for a more complete exposition.

In a subject control sentence like (1a), the complement subject position is filled by a silent
pronominal element PRO, which corefers with the referent of the matrix subject. In other words,
we give (1a) the structure in (2a), by analogy to that in (2b). (The numerical subscripts on the
children, PRO, and they represent coreference, meaning that they all refer to the same real-world
entity (here, the children).)

(2) a. [ The children ]1 agreed [ ∅ PRO1 to dance ].

b. [ The children ]1 agreed [ that they1 would dance ].

Extra Info
The idea behind the term ‘subject control’ is that the matrix subject fixes, or controls, the
reference of PRO. Notice that the parallel between PRO and overt pronouns in (2) is not
complete. PRO in (2a) must corefer with the matrix subject, whereas the overt pronoun in
(2b) needn’t. This is succinctly summarized by the notation in (i).

(i) a. [ The children ]1 agreed [ ∅ PRO1,*2 to dance ].

b. [ The children ]1 agreed [ that they1,2 would dance ].
It’s for this reason that only sentences where the clausal complement is nonfinite can
count as instances of subject control.
Chapter 9: Subject control versus raising 256

Raising sentences differ from subject control sentences in that the matrix subject does not
start out in the matrix clause, but rather moves there from the complement clause. Omitting
unnecessary details for the moment, the derivation is illustrated in (3).

(3) a. seemed [ the children to dance ].

b. [ The children ]i seemed [ t i to dance ].

The assumption that the matrix subject position starts out empty is supported by the fact that
when seem takes a finite counterpart, that same position is filled by expletive it.

(4) Itexpl seemed [ that the children danced ].

Extra Info
Analogously to subject control, for a sentence to count as an instance of raising, the com-
plement clause must be nonfinite, as in (3), since it is only then that the complement
subject raises into the matrix clause. (4) contains the same matrix predicate as (3), but it is
not an instance of raising because the complement subject remains within its clause.

A note on terminology. We will refer to the class of (Fregean) predicates to which agree
belongs as subject control predicates. Similarly, we refer to the class of predicates like seem
as raising predicates. This latter term is potentially confusing. It is important to understand
that it is not the predicate that is raising. Rather, as schematically indicated in (3) and as we will
see in more detail below, it is the complement subject that raises. A better term for the class
of predicates in question would be ‘raising triggers’. But we continue to use the term ‘raising
predicate’ because it is standard in the literature.
A leading role in distinguishing raising from subject control is played by expletive there. We
conclude the chapter by showing that expletive there belongs to a larger class of nonthematic
subjects.

Subject control 1

1.1 Evidence for two clauses

This section relies on the notion of selectional restrictions introduced in the previous chapter.
Here, we will use the concept to show that subject control sentences contain two separate clauses,
each with their own subject. We begin by showing that in finite-complement counterparts of
subject control sentences like (5), which incontrovertibly contain two clauses, both the matrix
and the complement verbs (or predicates more generally) impose separate selectional restrictions
on their respective subjects. (For simplicity, we omit referential indices in what follows; unless
Chapter 9: Subject control versus raising 257

otherwise noted, the intended interpretation is always the one where the complement subject
corefers with the matrix subject.)

(5) The children agreed that they would dance.

Consider (6), where we have taken care to satisfy the selectional requirements of the com-
plement clause (wet selects some physical object as its argument, and get imposes no further
selectional restrictions of its own). We can therefore be sure that the acceptability contrast in (6)
is due to the selectional restriction imposed by the matrix verb agree, which selects humans as
subjects.

(6) a. The children agreed [ that they would get wet ].

b. # The { trees, rocks } agreed [ that they would get wet ].

Conversely, in (7), we have taken care to satisfy the selectional restriction imposed by agree.
Here, the acceptability contrast is due to the selectional restrictions imposed by the complement
predicates. In particular, elapse selects stretches of time, and evaporate selects liquids. Neither
selects humans.

(7) a. The children agreed [ that they would speak Twi ].

b. # The children agreed [ that they would { elapse, evaporate } ].

If subject control sentences contain two clauses, as we are proposing, each with their own
subjects, they ought to pattern analogously to (6) and (7), and this is in fact exactly what we find
in (8) and (9).1

(8) a. The children agreed [ PRO to get wet ].

b. # The { trees, rocks } agreed [ PRO to get wet ].
(9) a. The children agreed [ PRO to speak Twi ].
b. # The children agreed [ PRO to { elapse, evaporate } ].

One last thing. Not all subject control predicates have a finite-complement variant.

(10) a. The children tried [ PRO to learn Twi ].

b. * The children tried [ that they would learn Twi ].

But even for such subject control predicates, we assume a biclausal structure with a PRO subject
for the lower clause. This is because, just as in (8) and (9), the subject control verb and the lower
predicate impose separate selectional restrictions on their respective subjects. You are invited to
provide the relevant evidence in Exercise 9.1, A.
Chapter 9: Subject control versus raising 258

1.2 Deriving subject control sentences

Given the facts just discussed, it comes as no surprise that the syntactic representation of subject
control sentences is straightforwardly analogous to that of their finite complement counterparts.
The elementary tree for agree is the same for both cases and is given in (11).

(11) VP

»DP« V′

V◦ »CP«
agree

Substituting a finite CP complement headed by that at the CP substitution node would

yield structures for sentences like (5). Substituting a nonfinite CP complement headed by a silent
complementizer yields structures for subject control sentences like (1a). In what follows, we
illustrate the derivation of (1a) in detail.

(12) a. VP b. IP

DP V′ DPi I′

PRO V◦ PRO I◦ VP
dance to
ti V′

V◦
dance

Substitute PRO in specifier Substitute (12a) as complement

position of lower verb of to and move subject
Chapter 9: Subject control versus raising 259

c. CP

C′

C◦ IP
∅
DPi I′

PRO I◦ VP
to
ti V′

V◦
dance

Substitute (12b) as complement

of silent C

Substituting the structure in (12c) as the complement of the control verb yields (13a), which in
turn becomes the complement of the matrix I element, yielding (13b).

(13) a. VP

DP V′

the children V◦ CP
agreed
C′

C◦ IP
∅
DPi I′

PRO I◦ VP
to
ti V′

V◦
dance
Chapter 9: Subject control versus raising 260

b. IP

»DP« I′

I◦ VP
[pst]
DP V′

the children V◦ CP
agreed
C′

C◦ IP
∅
DPi I′

PRO I◦ VP
to
ti V′

V◦
dance

Finally, moving the matrix subject yields (14a). The structurally analogous tree for the finite
complement counterpart of (14a) (= (5)) is shown for comparison in (14b). As is evident, the trees
differ only in a few terminal nodes.

(14) a. IP
DPj I′
the children I◦ VP
[pst]
tj V′
V◦ CP
agreed
C′
C◦ IP
∅
DPi I′
PRO I◦ VP
to
ti V′
V◦
dance
Chapter 9: Subject control versus raising 261

b. IP
»DP« I′
I◦ VP
[pst]
DP V′
the children V◦ CP
agreed
C′
C◦ IP
that
DPi I′
they I◦ VP
would
ti V′
V◦
dance

In concluding our discussion of subject control, let us point out that we have not addressed
an important question—namely, how PRO gets case. In English, PRO and overt noun phrases
are in complementary distribution (at least almost perfectly so).2 In other words, the positions
where PRO can appear are ones from which overt noun phrases are barred, and vice versa. It
has therefore been proposed that PRO receives a case unique to it—so-called null case—which is
assigned by nonfinite I in the spec-head configuration. However, there is evidence from languages
like Icelandic (see Note 1) and German that PRO can be assigned the same case features as overt
subjects are, which contradicts the null case idea. We will not attempt to resolve the issue here,
and we will ignore it in what follows.

Raising 2

2.1 A “what if” detour

Let us turn now to raising sentences like (1b), repeated here as (15).

(15) The children seemed [ to dance ].

At first glance, it seems as if we could treat such sentences as instances of subject control. But
that would leave us without an explanation for the contrast in (16)—in particular, for the gram-
maticality of (16b).
Chapter 9: Subject control versus raising 262

(16) a. * There agreed to be a problem.

b. ✔ There seemed to be a problem.

The analysis in the previous section does correctly rule out (16a), to which we assign the
structure indicated schematically in (17).

(17) * There agreed [ PRO ] to be a problem.

Given (17), the sentence is ruled out for two reasons. First, expletive there fails to satisfy the
selectional restriction of agree, which, as we saw earlier, selects humans as subjects. Second, and
conversely, expletive there is itself not licensed; it neither occupies the specifier position of a verb
of (coming into) existence nor can it have originated in such a position. (This is why we mark
(16a) as ungrammatical rather than as just semantically deviant.)
It is true that the sentence contains the there licenser be, but there never substitutes into
its specifier position. The predicate whose specifier position there does substitute into, namely
agree, is not a there licenser.

(18) * There agreed some students.

So (17) correctly rules out (16a) as ungrammatical. But by the same token, if we give (16a)
the analogous structure in (19), we incorrectly expect (16b) to be ungrammatical on a par with
(16a).

(19) Incorrect structure: There seemed [ PRO to be a problem ].

The reason is that there in the representation in (19), once again, is not licensed, as shown by the
ungrammaticality of (20).

(20) * There seemed a problem.

It is important to understand that although agree is a subject control predicate and seem is not,
both verbs share the property of not being there licensers.

2.2 Nonthematic subject positions

At this point, notice that the representations in (17) and (19) are both ruled out because there isn’t
licensed in the matrix clause. However, only in (17) is the matrix verb’s selectional restriction
violated. It turns out that a crucial difference between agree and seem is that seem imposes no
selectional restrictions. We can see this by replacing agree in (8) with seem, as we do in (21); the
acceptability contrast between (8a) and (8b) then disappears. (By contrast, replacing agree with
seem in (9) has no effect on the contrast in acceptability. After reading this section to the end,
you will be able to explain this fact, and we invite you to do so in Exercise 9.1, B.)

(21) a. The children seemed to get wet.

b. The { trees, rocks } seemed to get wet.
Chapter 9: Subject control versus raising 263

Another noteworthy property of seem is that its specifier position is not (and, in fact, must
not be) associated with any thematic role. It is true that seem takes an argument: what for
lack of a better term we will call the proposition argument.3 However, for some reason, this
argument cannot substitute into the specifier position, as we clearly see in the finite complement
counterparts of raising sentences.4

(22) a. It seems that the problem was too hard.

b. * That the problem was too hard seems.

To summarize: the subject position of seem is semantically deficient in the sense that it
is associated neither with selectional restrictions nor with a thematic role. We will call such a
subject position nonthematic.

2.3 Deriving raising sentences

Of course, despite being superfluous from a semantic point of view, nonthematic subject positions
are nevertheless syntactically obligatory (recall the subject requirement introduced in Chapter 3,
§ 3). This makes possible the following analysis of subject raising sentences, which we illustrate
by deriving (16b). We begin by deriving the complement clause. Note how the eventual matrix
subject there is licensed in (23a) as the specifier of main verb be. We assume that the subject moves
from Spec(VP) to Spec(IP) in (23b) in order to provide the complement clause with a subject in
accordance with the subject requirement.

(23) a. VP b. IP

DP V′ DPi I′

there V◦ DP there I◦ VP
be to
a problem DP V′

there V◦ DP
be
a problem

Substitute eventual matrix subject as Substitute (23a) as complement of to

specifier of lower verb and subject movement

We now substitute (23b) into the elementary tree for seemed in (24a).
Chapter 9: Subject control versus raising 264

(24) a. VP b. VP

»DP« V′ »DP« V′

V◦ »IP« V◦ »CP«
seemed seemed

Elementary tree for nonfinite Elementary tree for finite

complement, cf. (29a) complement, cf. (29b)

Before proceeding with the derivation, a few words about the elementary trees in (24) are in order.
First, notice that both elementary trees in (24) contain a specifier position. Though semantically
unnecessary, as discussed above, this position is motivated by the obligatoriness of expletive it
in small clauses.

(25) a. They made [ it seem that there was a problem ].

b. * They made [ seem that there was a problem ].

Second, the syntactic category of the clausal complement is IP in (24a), whereas it is CP in (24b),
which is the elementary tree we would use in order to derive the finite complement counterpart
of (16b) (It seemed that there was a problem). The reason that raising predicates, in contrast to
subject control predicates, require different elementary trees depending on the finiteness of their
complement has to do with certain structural conditions that must be satisfied by traces of move-
ment (but not by PRO). We simply mention the existence of these conditions here (violating them
results in “improper movement”); their motivation and exact character go beyond the scope of
this textbook.
Substituting the clause in (23b) as the complement of the elementary tree for raising seem
in (24a) yields (26a), which in turn becomes the complement of the matrix I element, yielding
(26b).
Chapter 9: Subject control versus raising 265

(26) a. VP

»DP« V′

V◦ IP
seemed
DPi I′

there I◦ VP
to
ti V′

V◦ DP
be
a problem
b. IP

»DP« I′

I◦ VP
[pst]
»DP« V′

V◦ IP
seemed
DPi I′

there I◦ VP
to
ti V′

V◦ DP
be
a problem

At this point in the derivation, the option arises in principle of substituting expletive it in
the matrix Spec(VP) and moving it to the matrix Spec(IP). In fact, this is what we would do if the
complement of seem were finite. In the case of a nonfinite complement, however, this step yields
the ungrammatical (27).

(27) * Itexpl seemed there to be a problem.

Why is (27) ungrammatical? The equally ungrammatical (28a) provides a hint; the ungram-
maticality of the unambiguously objective case form them is reminiscent of (2b) in Chapter 8,
repeated here as (28b).
Chapter 9: Subject control versus raising 266

(28) a. * Itexpl seemed them to dance.

b. * Itexpl was approved them (by us).

The parallel ungrammaticality of all three sentences leads us to conclude that seems is deficient in
not being able to assign case. (Recall that it is also deficient in not licensing a thematic specifier
position.) Thus, both there in (27) and them in (28) (more precisely, their chains) don’t receive case.
Let’s run through the possibilities. The chains can’t receive case via the trace in the complement-
clause Spec(VP) because verbs can’t assign case to their specifiers. The chains can’t receive case
from nonfinite I, because nonfinite I doesn’t assign case. And finally, they can’t receive case from
seems in the head-spec configuration because, as we have just concluded, seems (and raising verbs
in general) don’t assign case either.
Since the complement-clause subject cannot get case within its own IP, it moves on via
the matrix Spec(VP) to the matrix Spec(IP), as shown in (29a). Here, it can finally get case from
finite I in the spec-head configuration. It is this movement of the subject out of the complement
clause in (29a) that we refer to as raising. For comparison, the structure of the finite complement
counterpart is shown in (29b); here, each of the two subjects gets nominative case from its own
finite I.

(29) a. IP

DPi I′

there I◦ VP
[pst]
ti V′

V◦ IP
seemed
ti I′

I◦ VP
to
ti V′

V◦ DP
be
a problem
Chapter 9: Subject control versus raising 267

b. IP

DPi I′

it I◦ VP
[pst]
ti V′

V◦ CP
seemed
C′

C◦ IP
that
DPk I′

there I◦ VP
[pst]
tk V′

V◦ DP
was
a problem

Some special verbs 3

3.1 Tend versus occur

As was the case with subject control verbs, certain raising verbs are able to take finite comple-
ments in addition to nonfinite ones. But this is not true for all of them. Seem, as we have seen,
belongs to the first type. Tend belongs to the second type, as shown in (30)–(31).

(30) a. ✔ There seems (to me) to be a solution.

b. ✔ It seems (to me) that there is a solution.
(31) a. ✔ There tends to be a traffic jam at rush hour.
b. * It tends that there is a traffic jam at rush hour.

Despite the contrast between examples (30b) and (31b), seem and tend are both raising
verbs; what is crucial is that examples (30a) and (31a), where the subject of the complement
clause leaves its clause, are both grammatical.
Chapter 9: Subject control versus raising 268

There are also verbs that exhibit the converse pattern of tend.

(32) a. ✔ It occurred to me that there is a solution.

b. * There occurred to me to be a solution.

Such verbs have a nonthematic subject position, just like seem and tend. But as was mentioned
in the introduction, they are not raising verbs, since (for whatever reason) their complement
subjects cannot leave the clause they originate in.

3.2 Promise
Promise has the noteworthy property of behaving either as a subject control predicate or as a
raising predicate. As a subject control predicate, promise means something like ‘vow’ and selects
human subjects. This promise can take either finite or nonfinite complements.

(33) a. The { children, #horses } promised [ to eat their oatmeal ].

b. The { children, #horses } promised [ that they would eat their oatmeal ].

On this interpretation, the matrix clause can contain manner adjuncts that modify promise (notice
how promise can be replaced by vow in the examples).

(34) a. The children obediently promised [ to eat their oatmeal ].

b. The children softly promised [ to eat their oatmeal ].

In (34a), it could be argued that obediently is licensed as a property of the children (✔The children
were obedient). The grammaticality of (34b) is therefore particularly important, since the only
licenser for softly is promise (*The children were soft).
But promise can also have a ‘weaker’ meaning; on this interpretation, a sentence like (35a)
can be paraphrased as (35b).

(35) a. The weather promises to be lovely tomorrow

b. All available evidence indicates that the weather will be lovely tomorrow.

On this interpretation, modifying the matrix predicate by a manner adverb as in (36) is as deviant
in the original promise sentence as in the paraphrase.

(36) a. # The weather { confidently, softly } promises to be lovely tomorrow.

b. # All available evidence { confidently, softly } indicates that the weather will be lovely
tomorrow.

Notice, moreover, that in the alternative paraphrase in (37), the presence of expletive there in
matrix subject position suggests that the position is nonthematic.

(37) There is every indication that the weather will be beautiful tomorrow.

The nonthematic character of the matrix subject position for this interpretation of promise
is borne out by the grammaticality of (38).
Chapter 9: Subject control versus raising 269

(38) There promises to be a new version by spring.

From these facts, we conclude that the proper representation for a sentence like (38) must
be the raising structure schematically indicated in (39).

(39) Therei promises [ t i to be a new version by spring ].

Promise sentences with thematic subjects, on the other hand, are ambiguous between a
raising analysis and a subject control analysis (as long as they contain no disambiguating ad-
verbs). Which reading is prominent depends, as always, on the discourse context. In a sentence
like (33a), the prominent interpretation, and the only one considered so far, is the subject control
interpretation represented in (40).

(40) The { children, #horses } promised [ PRO to eat their oatmeal ].

However, (33a) also has the raising interpretation represented in (41a), which can be paraphrased
as in (41b).

(41) a. [ The { children, horses } ]i promised [ t i to eat their oatmeal ].

b. There was every indication that the { children, horses } would eat their oatmeal.

As expected, under this interpretation, the acceptability contrast between children and horses that
is due to the selectional restrictions imposed by subject control promise disappears.
There is at least one other verb in English that clearly has the same property as promise—
namely, threaten. You are invited to provide evidence for this assertion in Exercise 9.1, C.

More nonthematic subjects 4

In distinguishing between subject control and raising in this chapter, we have relied heavily on
the grammaticality or ungrammaticality of sentences containing expletive there. As it turns out,
expletive there is not the only nonthematic subject (= subject that is not associated with a the-
matic role). In this section, we present two further instances of nonthematic subjects: subject
idiom chunks and weather it.
Chapter 9: Subject control versus raising 270

4.1 Subject idiom chunks

In Chapter 8, § 2.2, we introduced the notion of object idiom chunks—noun phrases that are
ordinarily restricted to object position. The notion of subject idiom chunk is related (though not
exactly identical). We use it to refer to subjects that occur in sentence idioms like (42); the subject
idiom chunks are underlined.

(42) a. The cat is out of the bag.

b. The fur will fly.
c. The jig is up.
d. The pot is calling the kettle black.
e. The shit hit the fan.

Subject idiom chunks share two important properties with expletive there. First, just as
expletive there must be licensed as the specifier of a verb of existence, the subjects in (42) have
whatever idiomatic force they have only in connection with the rest of the idiom, but not oth-
erwise.5 In particular, the subject idiom chunk must start out as the specifier of the verb in the
idiom. For instance, neither cat in (42a) nor pot in (42d) have a metaphorical sense of secret or
hypocrite, respectively, in other syntactic contexts. So sentences like those in (43) have only literal
interpretations.

(43) a. The cat is safe with them.

(only literal meaning; can’t have the idiomatic meaning: ‘The secret is safe with
them.’)
b. # Your neighbor sounds like a real pot.
(literal meaning is semantically deviant; also can’t have the idiomatic meaning:
‘Your neighbor sounds like a real hypocrite.’)

Second, presumably because they are not interpreted literally, subject idiom chunks don’t
seem to refer or be associated with any thematic role, and so they can occupy the nonthematic
subject position of raising predicates. As a result of these two properties, contrasts as in (44) are
expected.

(44) a. # The cat agreed [ PRO to be out of the bag ].

b. [ The cat ]i seems [ t i to be out of the bag ].

(44a) is ruled out both on a literal and an idiomatic reading. Agree selects humans as subjects and
is therefore incompatible with the cat either as a literal or as an idiomatic (nonthematic) subject.
In addition, the cat isn’t licensed as an idiom chunk in the representation in (44a) because it is
never a specifier of the lower verb (be, the verb heading the rest of the idiom). By contrast, both
readings, and in particular the idiomatic one, are possible in (44b). This is expected, since the
matrix subject originates in Spec(VP) of the complement clause, forming a constituent with the
rest of the idiom.
Note that examples like (45) do not invalidate the diagnostic value of subject idiom chunks
in distinguishing between subject control and raising predicates.
Chapter 9: Subject control versus raising 271

(45) The cat wanted [ PRO to be out of the bag ]. (only literal interpretation)

Here, the selectional restrictions that want imposes on its subject are less strict than in the case
of agree. Since the subject of want isn’t required to be human (but only to have a reasonably
well-developed nervous system), the sentence is grammatical, unlike (44a). However, unlike in
(44b), the matrix subject position isn’t nonthematic and the matrix subject doesn’t move out of
the lower clause, so the sentence has only a literal interpretation.

4.2 Weather it
The third type of nonthematic subject is weather it, the subject of verbs of precipitation.

(46) ✔ It is { hailing, pouring, raining, sleeting, snowing }.

As with subject idiom chunks and their predicates, the licensing relationship between weather it
and their predicates is mutual: not only is weather it licensed by weather verbs, but the weather
verbs are in turn themselves licensed by weather it, as shown in (47).

(47) * The { air, atmosphere, environment, precipitation, sky, weather } is { hailing, pouring,
raining, sleeting, snowing }.

Given the nonthematic character of weather it, contrasts as in (48) are expected.

(48) a. # It decided to rain at night.

b. It tends to rain at night.

Unexpectedly, given what we have said so far, sentences like (49) do occur.

(49) It’s trying to rain; it finally managed to rain.

What is going on here? A plausible explanation is that the selectional restrictions of try
and manage are being flouted (recall the discussion in Chapter 8 of the students lapping up their
teacher’s praise), prompting the hearer to conceptualize the weather metaphorically as an ani-
mate being.6

4.3 Summary
We have seen that expletive there is licensed as the subject of verbs of (coming into) existence.
In a similar way, subject idiom chunks and their predicates stand in a mutual licensing relation-
ship, as does weather it with predicates denoting precipitation. Because of their special licensing
requirements, none of these subjects is licensed by subject control predicates. Conversely, sub-
ject control predicates select humans as subjects, so that their selectional restrictions are not met
by nonthematic subjects. By contrast, raising predicates neither interfere with the licensing of
nonthematic subjects (which takes place in a lower clause) nor do they impose selectional restric-
tions that nonthematic subjects cannot meet. It is precisely because of their semantic deficiency
or neutrality that they are able to act as grammatical catalysts, allowing licensing relations that
are normally confined to the same clause to extend across an IP boundary.
 Chapter 9: Subject control versus raising 272

As we have seen, the special properties of nonthematic subjects make them useful diagnos-
tics for distinguishing subject control predicates from raising predicates. The relevant judgment
patterns are summarized in Table 9.2; for completeness, we also include the judgments for manner
adverbs discussed in connection with promise.

Subject control Raising

Expletive there * ✔
Subject idiom chunk # (or only literal) ✔ (idiomatic or literal)
Weather it # (or metaphorical) ✔
Manner adverb ✔ *

Table 9.2: Judgments of subject control and raising sentences with nonthematic subjects.

Notes 5
1. It might occur to a careful reader that an alternative approach to the facts in (8) and (9) is possible, according to
which control sentences contain a single subject, which must simultaneously satisfy the selectional restrictions of
both the higher and the lower verbs. Regardless of whether such an approach might be worked out for English, we
do not adopt it, since it cannot be extended to subject control universally. In particular, the approach in question
fails for Icelandic.
In contrast to English (and most other languages), Icelandic has certain verbs whose subjects appear in some
non-nominative case (genitive, dative, or accusative), even in finite clauses. The analysis of these so-called ‘quirky
case’ subjects is beyond the scope of this textbook, but it is well established that they are true subjects (despite the
lack of subject-verb agreement) (see Zaenen, Maling, and Thraínsson 1985; Sigurðsson 1991, and the many references
therein). (i) gives examples of Icelandic finite clauses with an ordinary nominative subject and with a quirky case sub-
ject. The subjects are in boldface. Note that the underlined quantifiers agree in case with the subjects; this fact will
be important directly. (How exactly this case agreement is implemented is irrelevant for present purposes.)

(i) a. Ordinary nominative subject:

Strákarn-ir komust all-ir í skóla.
boys-def.nom got all-nom.pl.m in school
‘The boys all got to school.’ ( = Sigurðsson’s (6a))
b. ‘Quirky dative’ subject:
Strákarn-um leiddist öll-um í skóla.
boys-def.dat was.bored all-dat.pl.m in school
‘The boys were all bored in school.’ ( = Sigurðsson’s (6c))

(ii) shows that the clauses in (i) can be embedded under a subject control verb (here, vonast til ‘hope for’). As
in English, the subject of the embedded clauses is silent, but note that the quantifiers continue to exhibit the same
case that they did in (i).
 Chapter 9: Subject control versus raising 273

(ii) a. Embedded ordinary subject:

Strákarn-ir vonast til að PRO komast all-ir í skóla
boys-def.nom hope for comp get all-nom.pl.m in school
‘The boys hope to all get to school.’ ( = Sigurðsson’s (8a))
b. Embedded quirky subject:
Strákarn-ir vonast til að PRO leiðast ekki öll-um í skóla
boys-def.nom hope for comp be.bored not all-dat.pl.m in school
‘The boys hope to all not be bored in school.’ ( = Sigurðsson’s (8c))

In particular, in (ii.b), the quantifier must appear in the dative. From this, we conclude that the silent subject of the
lower clause in (ii.b) is assigned quirky dative case in (ii.b), just as it is in the finite clause in (i.b). The fact that the
matrix and embedded subjects are each assigned a different case in (ii.b) provides conclusive evidence that control
constructions are indeed biclausal, since a single noun phrase cannot be assigned more than one case (even if it were
to satisfy more than one selectional restriction at the same time).

2. In fact, the complementary distribution is not perfect. There is a bit of overlap in English—for instance, in the subject
position of gerunds.

(i) a. [ PRO going out with them ] would bother me.

b. * [ { Kim’s, Kim } going out with them ] would bother me.

3. For simplicity, we ignore the verb seem’s optional experiencer argument (It seems to me that you’ve solved the problem),
since it has no effect on our conclusions.

4. Clear evidence that the specifier position at issue is the Spec(VP) associated with seem (and not, say, some higher
specifier position, such as Spec(IP)) comes from the grammaticality contrast in (i).

(i) a. They made [ it seem [ that the problem was hard ] ].

b. * They made [ [ that the problem was hard ] seem ].

5. The licensing relation between subject idiom chunks and their predicates is actually even tighter than that between
expletive there and its licensers. Existential there must be licensed by verbs of (coming into) existence, but these verbs
can occur independently of existential there. In other words, the licensing relationship is only one-way. By contrast,
neither the subjects nor the predicates of idiomatic sentences can be interpreted idiomatically independently of each
other. So in this case, as also for weather it and predicates of precipitation, the licensing relationship is mutual.

6. This metaphorical animate being can urinate, as is evident from the (vulgar) slang terms for ‘to rain’ illustrated in
(i); cf. also (ii).

(i) a. It’s pissing.

b. Es schifft (wie die Sau). (German)
it pisses like the sow
‘It’s raining heavily.’
(ii) Il pleut comme vache qui pisse. (French)
it rains like cow that pisses
‘It’s raining heavily.’
Chapter 9: Subject control versus raising 274

Exercises and problems 6

Exercise 9.1
A. As mentioned in the text, certain subject control predicates, like try, cannot take finite com-
plements. Provide evidence that in subject control sentences containing these predicates,
both the matrix predicate and the complement predicate impose their own selectional re-
strictions, thus motivating a biclausal analysis.
B. Explain the acceptability contrast in (1).

(1) a. The children seemed to learn Twi.

b. # The children seemed to { elapse, evaporate }.

C. Show that threaten is both a subject control verb and a raising verb.

Exercise 9.2
A. You have been asked to review an article for Linguistic Inquiry by Professor A.B.C.
Gerneweis, who concludes on the basis of the contrast in (1) that volunteer is a control
verb. What is wrong with the argument?

(1) a. ✔ They volunteered to do the job.

b. * There volunteered to do the job.

B. In your review, you graciously provide the conclusive evidence that Professor Gerneweis
failed to provide.

Exercise 9.3
A. Using the examples below as a model, determine whether the verbs in (4) are raising pred-
icates, subject control predicates, or neither. In each example below, the evidence takes
the form of grammaticality judgments about the example sentences, and the conclusions
are annotated below each example. For the purposes of this exercise, use only active verb
forms.

(1) ✔ There chanced to be an opening

→ Conclusion: Chance is a raising verb.

(2) a. * There resolved to be a solution.

→ Conclusion:Resolve is not a raising verb.
b. ✔ They resolved to solve the problem.
→ Conclusion: Resolve is a subject control verb.
Chapter 9: Subject control versus raising 275

(3) a. * There is possible to be a solution.

→ Conclusion: Possible is not a raising adjective (despite the grammaticality
of It is possible that there is a solution.).
b. * They are possible to solve the problem.
→ Conclusion: Possible doesn’t allow a nonfinite complement, so it is not a
subject control adjective either.

(4) agree, aspire, attempt, be, beg, cease, choose, claim, come, commence, continue,
dare, demand, deserve, desire, determine, elect, end up, endeavor, expect, fail, forget,
happen, have, hope, intend, look, mean, need, neglect, plan, pledge, prefer, presume,
pretend, proceed, prove, purport, remember, request, start, strive, swear, tend, train,
try, volunteer, vow, wish, yearn

B. Subject control predicates and raising predicates can belong to other syntactic categories
than V. The predicates in (5) are adjectives, those in (6) are participles, and about in (7) is a
preposition. Repeat (A) for these predicates. It’s not always clear whether the participles
are verbs or adjectives. If in doubt, treat them as adjectives; the exact category is not crucial
for the purposes of the exercise.
Note: Subject control predicates and raising predicates that aren’t verbs need to be inte-
grated into the sentence using some form of the verb be. See Exercise 9.4, (1d)–(1f) for
examples.

(5) afraid, anxious, apt, certain, content, eager, ecstatic, evident, fortunate, glad, happy,
hesitant, liable, likely, lucky, possible, ready, reluctant, sorry, sure, unlikely
(6) bound, delighted, destined, determined, embarrassed, excited, fated, going, inclined,
itching, jonesing, prepared, scared, (all) set, supposed, thrilled
(7) about

Exercise 9.4
A. Using the examples in (1)-(3) in Exercise 9.3 as a model, determine whether the underlined
predicates in (1) are subject control or subject raising predicates.

(1) a. They failed to be on time.

b. They aspire to graduate early.
c. They remembered to do the laundry.
d. They are hesitant to move.
e. They are destined to get the job.
f. They are about to hit the jackpot.

B. Build structures for the sentences in (1).

Chapter 9: Subject control versus raising 276

Exercise 9.5
A. As was mentioned in this chapter, raising verbs are logically distinct from there licensers.
There are, however, some verbs that belong to both classes (this is comparable to a single
person belonging to two distinct clubs). Can you think of any?
B. Tend is a raising verb that cannot take finite complements. Can you think of other verbs
like it?
C. Expect is an ECM verb. Which other class(es) of verbs discussed in this chapter does it
belong to?

Exercise 9.6
A. The sentence in (1) is ambiguous. Explain, taking into account the facts in (2).

(1) They were determined to be U.S. citizens.

(2) a. ✔ They were very determined to be U.S. citizens.
b. ✔ They were determined to be U.S. citizens by the INS.
c. * They were very determined to be U.S. citizens by the INS.

B. Build the trees for the two interpretations of (1).

Exercise 9.7
A. What verb class does prove belong to in (1a) and in (1b)?

(1) a. They proved there to be an error in the calculation.

b. There proved to be an error in the calculation.

B. Which case does each instance of there get? How can you tell? What’s the case-assigning
head, and what’s the case assignment configuration?
C. Build the trees for both sentences in (1).

Exercise 9.8
Superficially, raising looks quite different than ordinary subject movement, the simple passive,
and the passive of ECM verbs, but it shares properties with all three. Complete Table 9.3 to better
understand the similarities and the differences. By ‘local movement’ is meant movement where
the DP head of the movement chain doesn’t leave its IP.
Chapter 9: Subject control versus raising 277

Name Raising Passive of ECM Ordinary Simple passive

subject
movement
Tail of chain Spec(VP), low
Intermediate Spec(IP), low;
links (if any) Spec(VP), high
Head of chain Spec(IP), high
Case assigner Finite I
Case Spec-head
configuration
Local? No

Table 9.3: Case assignment in raising and other constructions

Exercise 9.9
A. (1) is grammatical, but both sentences in (2), with the same intended meaning, are com-
pletely ungrammatical. Why is that?

(1) It seems that Jackie has solved the problem.

(2) a. * It seems Jackie to have solved the problem.
b. * Jackie seems that has solved the problem.

B. Repeat (A) for (3) and (4).

(3) It seems that they like caviar.

(4) a. * There seems that they like caviar.
b. * They seem that they like caviar.
c. * Theyi seem that t i like caviar.
d. * Caviari seems that they like t i .

Exercise 9.10
English indirect questions can be finite, as in (1), or nonfinite, as in (2).

(1) a. ✔ They know who they should invite.

b. ✔ They know which topic they should talk about.
c. ✔ They know who should speak.
(2) a. ✔ They know who to invite.
b. ✔ They know which topic to talk about.
c. * They know who to speak.
Chapter 9: Subject control versus raising 278

A. Using what you’ve learned in this chapter, build structures for (2) that are as parallel as pos-
sible to those for (1a). For simplicity, build a full structure just for (2a) and only structures
for the indirect questions in (2b), (2c).
B. Why is (2c) ungrammatical?

Exercise 9.11
The sentences in (1) and (2) are structurally ambiguous. Build partial structures for them, as indi-
cated by the slashes, and explain how those structures fit together to yield the various readings.
Focus on the intended reading and the most salient funny reading, but feel free to address other
unintended readings as well.
In (2), you will need to come up with a structure for the gerund clause introduced by meet-
ing by extending material covered in this this chapter.
Treat freak accident, North Koreans and Oval Office as single words.

(1) Gibson plays a man

/ who develops the ability to understand
/ what women are thinking
/ after a freak accident.
(Source: https://www.imdb.com/title/tt0207201/news?year=2000, accessed 18 March
2012)
(2) I remember meeting a mother of a child
/ who was abducted by the North Koreans
/ right here
/ in the Oval Office.
(Source: https://www.slate.com/id/76886/, accessed 8 April 2010)

Problem 9.1
Provide as insightful a description of the following facts as you can. What patterns do you notice?
Can you state any generalizations?

(1) a. ✔ a person who we can hire

b. ✔ a person with whom we can consult
c. ✔ a person who(m) we can consult with
d. ✔ a person who can fix the sink

(2) a. * a person who to hire

b. ✔ a person with whom to consult
c. ✔ a person who(m) to consult with
d. * a person who to fix the sink
Chapter 9: Subject control versus raising 279

(3) a. ✔ a person that we can hire

b. * a person with that we can consult
c. ✔ a person that we can consult with
d. ✔ a person that can fix the sink

(4) a. ? a person for us to hire

b. * a person with for us to consult
c. ? a person for us to consult with
d. % a person for to fix the sink

(5) a. ✔ a person we can hire

b. * a person with we can consult
c. ✔ a person we can consult with
d. % a person can fix the sink

(6) a. * a person us to hire

b. * a person with us to consult
c. * a person us to consult with

(7) a. ✔ a person to hire

b. * a person with to consult
c. ✔ a person to consult with
d. ✔ a person to fix the sink
 10
The verb raising
parameter

Chapter outline
1 Verb raising: V moves to I in the syntax . . . . . . . . . . . . . . . . . . . . . . . . 282
1.1 The future tense in French . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
1.2 The order of diagnostic adverbs and verbs in French . . . . . . . . . . . . 284
2 Tense lowering: I moves to V in the morphology . . . . . . . . . . . . . . . . . . . 289
2.1 The order of diagnostic adverbs and verbs in English . . . . . . . . . . . . 289
2.2 Do support in English . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290
3 The loss of verb raising in Scandinavian . . . . . . . . . . . . . . . . . . . . . . . . 294
4 Verb raising in the history of English . . . . . . . . . . . . . . . . . . . . . . . . . 297
4.1 The loss of verb raising . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
4.2 A change in the status of not . . . . . . . . . . . . . . . . . . . . . . . . . 299
4.3 The emergence of do support . . . . . . . . . . . . . . . . . . . . . . . . . 300
4.4 The emergence of modals . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
4.5 Remnants of verb raising in modern English . . . . . . . . . . . . . . . . . 304
5 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
6 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309

As we saw in Chapter 4, tense in English can be expressed in one of two ways. The future
tense is expressed by will, which precedes the verb and is a free morpheme; that is, it can be
separated from the verb and stand alone.

(1) a. * We watch will that show.

b. ✔ We will never watch that show.
c. Q: Will you watch that show?
A: ✔ Yes, we will.

By contrast, the past tense is expressed by a bound morpheme, ordinarily the suffix -ed, which
combines with the verb to form a morphologically complex word.1

(2) a. ✔ We watch-ed that show.

b. * We-ed never watch that show.
Chapter 10: The verb raising parameter 281

c. Q: Did you watch that show?

A: *Yes, we -ed.

This dual expression of tense is typical of the Germanic language family, to which English belongs.
In all of these languages, the future is expressed analytically (as two separate words), whereas
the past is mostly expressed synthetically (as a single morphologically complex word).2
Treating sentences as projections of I, as in Chapter 4, § 2.1, provides a structural locus for
free tense morphemes and is therefore straightforwardly compatible with the analytic expression
of tense. The analogous semantic contribution of free and bound tense morphemes, in both Eng-
lish and other languages, is a good reason to extend the IP analysis to sentences with synthetic
tense forms. But the extension does raise the question of how tense and the verb combine to form
a complex word when tense is expressed synthetically. Following an approach called Distributed
Morphology that goes back to Halle and Marantz (1993), we assume that word formation takes
place in the morphology, a component of the grammar that operates on structures generated by
the syntax and that associates the terminal nodes in those structures with words in the traditional
sense. In Distributed Morphology, this association is called spellout. For instance, play and past
tense are spelled out as the regular form played, whereas sing and past tense are spelled out as
the irregular form sang. In this chapter, we present evidence that in some languages, the verb
moves up to I before the structure is handed over to the morphology, whereas in others, the verb
remains in situ (that is, it does not move). Instead, tense moves down to V in the morphology.
We will refer to the choice between V-to-I movement in the syntax and I-to-V lowering in the
morphology as the verb raising parameter. As we will show, the parametric variation can be
detected on the basis of the relative order of inflected verbs and certain adverbs.
It has been proposed that verb raising is correlated with the ‘strength’ of a language’s
subject-verb agreement morphology (the number of distinct person-number endings in the verbal
paradigm for, say, the present tense). We do not present this proposal here, as it is not clear
that the correlation holds up under close scrutiny (Bobaljik 2002; Heycock and Sundquist 2017;
Heycock and Wallenberg 2013).
In the remainder of the chapter, we first contrast verb raising in French with tense lowering
in English. Our description of English also includes discussion of a closely related and important
topic in the grammar of the modern language: the do support that is obligatory in negated sen-
tences (They don’t like okra vs. *They not like okra). (Do support is also obligatory in direct ques-
tions (Do they like okra? vs. *Like they okra?); this is the focus of Exercise 10.5.) We then briefly
review the parametric variation attested in the modern Scandinavian languages and the loss of
verb raising in all but one of them, Icelandic. The chapter concludes with a detailed overview of
issues related to the verb raising parameter in the history of English. As we will show, the intri-
cate web of facts concerning do support in modern English is the result of a complex diachronic
interplay of general principles with several unrelated and contingent language-particular devel-
opments.
Chapter 10: The verb raising parameter 282

Verb raising: V moves to I in the syntax 1

1.1 The future tense in French

As just mentioned, in certain languages, the verb moves up to I in the syntax. One such language
is French, and we begin our discussion of the verb raising parameter by considering the future
tense in French, which is formed by attaching suffixes to a verb’s infinitive.

Future tense of chanter ‘sing’ Present tense of avoir ‘have’

je chanter-ai ‘I will sing’ j’ai ‘I have’
tu chanter-as ‘you.sg will sing’ tu as ‘you.sg have’
il, elle chanter-a ‘he, she will sing’ il, elle a ‘he, she has’
nous chanter-ons ‘we will sing’ nous avons ‘we will have’
vous chanter-ez ‘you.pl will sing’ vous avez ‘you.pl have’
ils, elles chanter-ont ‘they will sing’ ils, elles ont ‘they have’

Table 10.1: Future tense formation in French

As is evident from Table 10.1, the future tense affixes are nearly identical to the present tense
forms of the verb avoir ‘to have’, the only difference being that the affixes are truncated in the
first and second person plural by comparison to the full two-syllable forms of avoir ‘have’. This
correspondence suggests that the future tense in French developed via a semantic shift from ‘they
have to V’ to ‘they will V’.3 In addition, and more immediately relevant for the present discussion,
the originally free forms of avoir were reanalyzed as bound morphemes.4 The analytic roots of
the synthetic French future tense thus indicate that the two ways of expressing tense (analytic or
synthetic) are not just semantically parallel, but that they are also morphologically more closely
related than might appear at first glance. (3) shows the elementary tree for the French future
suffix in the first-person plural, which ends up being spelled out as -ons in the morphology.

(3) IP
»DP« I′

I◦ »VP«
fut
[ 1pl ]

Given elementary trees like (3), sentences like (4) can be derived as follows.

(4) Nous chanter -ons une chanson.

we sing -fut.1pl a song
‘We will sing a song.’
Chapter 10: The verb raising parameter 283

We begin with the elementary tree for the verb chanter in (5a) and substitute it as the complement
of (3) to yield the structure in (5b).

(5) a. VP b. IP

»DP« V′ »DP« I′

V◦ »DP« I◦ »VP«
chanter- fut
[ 1pl ] »DP« V′

V◦ »DP«
chanter-

Elementary tree for chanter Substitute (5a) in elementary tree (3)

The verb then moves to I. Because I is already filled, the verb cannot directly substitute in I,
but must adjoin to it, as shown in (6). The head movement that we invoke here is exactly the
same formal operation that was introduced in Chapter 6, § 2.2 in connection with subject-aux
inversion.

(6) a. b.
IP IP

»DP« I′ »DP« I′

I◦ VP I◦ VP
fut
[ 1pl ] »DP« V′ »DP« V′
V◦i I◦
chanter- fut
V◦ »DP« t i »DP«
[ 1pl ]
chanter-

Select I as target of adjunction Left-adjoin V to I

The remaining steps of the derivation, shown in (7), are identical to the ones that would be
required to derive the corresponding English sentence We will sing a song.
Chapter 10: The verb raising parameter 284

(7) a. IP
»DP« I′

I◦ VP

V◦i I◦ DP V′
chanter- fut nous t i DP
[ 1pl ]
une chanson

Substitute subject and object

b. IP
DPk I′
nous I◦ VP

V◦i I◦ tk V′
chanter- fut ti DP
[ 1pl ]
une chanson

Move subject

Finally, the structure in (7b) is handed over to the morphological component, where the
combination of chanter and the future tense affix is spelled out as chanterons.

1.2 The order of diagnostic adverbs and verbs in French

Extra Info
In reading the following section, bear in mind that our focus is not on the distribution of
adverbs per se. In particular, we are not claiming that all, or even most, adverbs left-adjoin
to V′ in French; in fact, there are many that right-adjoin. Rather, the idea is to use the
particular subset of adverbs that must left-adjoin to V′ as a diagnostic tool to determine
the structural position of finite verbs in French.

The facts of French presented so far are consistent with a verb raising analysis, but do
not provide conclusive evidence in favor of it. In other words, nothing in what we have said
so far prevents the French verb from remaining in situ and not combining with tense until the
morphology. In this section, we present conclusive evidence in favor of the verb raising analysis
that is based on the order of verbs and certain diagnostic adverbs (Emonds 1978).
Chapter 10: The verb raising parameter 285

As illustrated in (8)–(10), there are certain adverbs in French (underlined) that ordinarily
precede the main verb of a sentence (in boldface), rather than follow it. (Strictly speaking, à peine
is a PP; what is relevant for the purposes of the argument is not its syntactic category, but rather
its syntactic distribution.)

(8) a. ✔ Elle avait à peine travaillé trois heures.

she had hardly worked three hours
‘She had hardly worked three hours.’
b. ✔ Mon ami a complètement perdu la tête.
my friend has completely lost the head
‘My friend completely lost his head.’
c. ✔ J’ avais presque oublié mon nom.
I had almost forgotten my name
‘I had almost forgotten my name.’
(9) a. * Elle avait travaillé à peine trois heures.
b. * Mon ami a perdu complètement la tête.
c. * J’ avais oublié presque mon nom.
(10) a. * Elle avait travaillé trois heures à peine.
b. * Mon ami a perdu la tête complètement.
c. * J’ avais oublié mon nom presque.

Word of Caution
As highlighted by the grammaticality contrast in (i), constraints in adverb placement in
one language don’t necessarily carry over to their translation equivalents in another.

(i) a. * … perdu la tête complètement = (10b)

b. ✔ … lost one’s head completely

The negative adverb jamais ‘never’ and the simple negative marker pas ‘not’ behave alike
in this regard in French.5

(11) a. ✔ Les enfants d’ aujourd’hui (n’) ont jamais écouté la radio.

the children of today ne have never listened.to the radio
‘Today’s children have never listened to the radio.’
b. * Les enfants d’aujourd’hui (n’)ont écouté jamais la radio.
c. * Les enfants d’aujourd’hui (n’)ont écouté la radio jamais.
(12) a. ✔ Je (n’) ai pas fait la vaisselle.
I ne have not done the dishes
‘I haven’t done the dishes.’
b. * Je (n’) ai fait pas la vaisselle.
c. * Je (n’) ai fait la vaisselle pas.
Chapter 10: The verb raising parameter 286

The word order facts in (8)–(12) follow straightforwardly if the adverbs in question must
adjoin to the left of V′ , as shown schematically in (13), rather than to the right.

(13) VP

»DP« V′

AdvP V′

diagnostic adverb V◦ »DP«

nonfinite verb

But now consider an unexpected fact: when the main verb of the sentence is finite, the
obligatory adverb-verb order that we have just seen is ungrammatical.

(14) a. * Elle à peine travailler-a trois heures.

she hardly work-fut.3sg three hours
Intended meaning: ‘She will hardly work three hours.’
b. * Mon ami complètement perdr-a la tête.
my friend completely lose-fut.3sg the head
Intended meaning: ‘My friend will completely lose his head.’
c. * Je presque oublier-ai mon nom.
I almost forget-fut.1sg my name
Intended meaning: ‘I will almost forget my name.’
d. * Les enfants d’ aujourd’hui (ne) jamais écouter-ont la radio.
the children of today ne never listen.to-fut.3pl the radio
Intended meaning: ‘Today’s children will never listen to the radio.’
e. * Je (ne) pas fer-ai la vaisselle.
I ne not do-fut.1sg the dishes
Intended meaning: ‘I won’t do the dishes.’

Instead, the adverb must follow the verb, although it still cannot follow the entire V′ .

(15) a. ✔ Elle travaillera à peine trois heures.

b. ✔ Mon ami perdra complètement la tête.
c. ✔ J’oublierai presque mon nom.
d. ✔ Les enfants d’aujourd’hui (n’)écouteront jamais la radio.
e. ✔ Je (ne) ferai pas la vaisselle.
(16) a. * Elle travaillera trois heures à peine.
b. * Mon ami perdra la tête complètement.
c. * J’oublierai mon nom presque.
d. * Les enfants d’aujourd’hui (n’)écouteront la radio jamais.
e. * Je (ne) ferai la vaisselle pas.
Chapter 10: The verb raising parameter 287

Table 10.2 summarizes all the facts just presented.

Finiteness of main verb AdvP > verb … verb > AdvP … verb > XP > AdvP
nonfinite, as in (8)–(13) ✔ * *
finite, as in (14)–(16) * ✔ *

Table 10.2: Adverb placement in French by finiteness of main verb

As already noted, the adverb placement facts for nonfinite verbs in the first row are ex-
pected under the assumption that diagnostic adverbs obligatorily left-adjoin to V′ . This assump-
tion also explains the rightmost judgment for finite verbs (the blue star in the second row). The
judgments highlighted in red, which are the opposite of their green counterparts in the row above,
are puzzling at first glance. But they too follow straightforwardly if we assume that finite verbs
obligatorily move to I in French, as in (17a).

(17) a. ✔ IP

»DP« I′

I◦ VP

V◦i I◦ »DP« V′
verb [tense]
AdvP V′

diagnostic adverb t i »DP«

Verb raising yields finite V > Adv (grammatical in French)

b. * IP

»DP« I′

I◦ VP
[tense]
»DP« V′

AdvP V′

diagnostic adverb V◦i »DP«

verb

No verb raising yields *Adv > finite V (ungrammatical in French)

Chapter 10: The verb raising parameter 288

If French did not require finite verbs to move to I, as in the hypothetical scenario represented in
(17b), it is difficult to see how the contrast between the green and the red cells in Table 10.2 could
be derived in a principled way.
As (18)–(20) show, the adverb facts for other simple tenses are parallel to those for the
future tense (cf. (14)–(16)), and it is therefore natural to extend the verb raising analysis to them
as well.

(18) a. * Elle à peine travaill-ait trois heures.

she hardly work-impf.3sg three hours
Intended meaning: ‘She used to hardly work three hours.’
b. * Mon ami complètement perd la tête.
my friend completely lose.pRs.3sg the head
Intended meaning: ‘My friend is completely losing his head.’
c. * Je presque oublie mon nom.
I almost forget-pRs.1sg my name
Intended meaning: ‘I am almost forgetting my name.’
d. * Les enfants (ne) jamais écout-aient la radio.
the children ne never listen.to-impf.3pl the radio
Intended meaning: ‘The children never listened to the radio.’
e. * Je (ne) pas fais la vaisselle.
I ne not do-pRs.1sg the dishes
Intended meaning: ‘I don’t do the dishes.’
(19) a. ✔ Elle travaillait à peine trois heures.
b. ✔ Mon ami perd complètement la tête.
c. ✔ J’oublie presque mon nom.
d. ✔ Les enfants d’aujourd’hui (n’)écoutaient jamais la radio.
e. ✔ Je (ne) fais pas la vaisselle.
(20) a. * Elle travaillait trois heures à peine.
b. * Mon ami perd la tête complètement.
c. * J’oublie mon nom presque.
d. * Les enfants d’aujourd’hui (n’)écoutaient la radio jamais.
e. * Je (ne) fais la vaisselle pas.

In concluding our discussion of French, we draw your attention to the fact that verb move-
ment, just like other instances of movement, allows us to accommodate mismatches between an
item’s expected position given its thematic or semantic relations and the position in which it is
pronounced. In the case at hand, assuming verb movement allows us to maintain a simple gener-
alization concerning diagnostic adverbs (they obligatorily left-adjoin) regardless of the finiteness
of the verb they modify. Perhaps even more importantly, we can maintain—regardless of verb
finiteness or presence of adverbs—the idea encoded in the X′ schema that verbs and their com-
plements are sisters. Nonfinite verbs presents no difficulty in this regard. But even in the case of
finite verbs, where the verb-complement adjacency expected under sisterhood is interrupted by
an intervening adverb, the expected structural relation is preserved via the trace of the verb.
Chapter 10: The verb raising parameter 289

Tense lowering: I moves to V in the morphology 2

2.1 The order of diagnostic adverbs and verbs in English

Having established that French exhibits verb raising in the syntax, we now investigate the corre-
sponding English facts, using exactly the same tool that we used in French—namely, the position
of diagnostic adverbs. As in French, certain adverbs in English obligatorily precede nonfinite
verbs.

(21) a. They will { almost, hardly, never } fail.

b. They have { almost, hardly, never } failed.
(22) a. * They will fail { almost, hardly, never }.
b. * They have failed { almost, hardly, never }.

But unlike in French, these adverbs precede the main verb of a sentence even when the
verb is finite.

(23) a. They { almost, hardly, never } failed.

b. * They failed { almost, hardly, never }.

The ungrammaticality of (23b) means that the verb raising analysis that is successful for French is
exactly wrong for English. Instead, in English all verbs remain in situ in the syntax and any tense
morphemes lower and adjoin to V in the morphology. The input to the morphology is therefore
(17b), repeated here as (24), and the order of the adverb and the verb remains unchanged after
tense lowering (which we do not show explicitly).

(24) IP

»DP« I′

I◦ VP
[tense]
»DP« V′

AdvP V′

diagnostic adverb V◦i »DP«

verb

Morphological tense lowering yields Adv > finite V

(grammatical in English regardless of finiteness of V)
Chapter 10: The verb raising parameter 290

2.2 Do support in English

We now turn to an apparently idiosyncratic and quirky consequence of the fact that English
lacks verb raising—namely, the do support that is necessary in sentences negated with not. In
order to highlight the conditions under which do support takes place, we will contrast sentences
containing not with ones containing never.
In vernacular English, never can function as a sentence negator equivalent to simple ‘not’,
without its literal meaning of ‘not ever’.

(25) Q: Did you get a chance to talk to Tom last night?

A: Nope, I never got a chance.

But despite their functional equivalence in contexts like (25), not and never differ from each
other in a striking way: not obligatorily triggers do support, whereas never doesn’t. (All forms
of do in this section are to be read without emphatic stress.)

(26) a. * They not applied.

b. ✔ They { did not, didn’t } apply.
(27) a. ✔ They never applied.
b. * They did never apply.

In order to explain this puzzling fact, we present an analysis of do support that relies on two main
ideas: first, that never and not are integrated into the structure of English sentences in different
ways, and second, that tense (and heads more generally) can lower in the morphology only under
certain structural conditions.6

A syntactic difference between never and not

As shown in (28), never is intransitive and hence a maximal projection in its own right, whereas
not is transitive and hence not a phrase on its own.

(28) a. AdvP b. NegP

Adv′ Neg′

Adv◦ Neg◦ »XP«

never not

We present two pieces of evidence for this distinction. The first comes from negative inversion.
(29a) shows an ordinary negative sentence, and (29b) shows its negative inversion counterpart,
where the negative constituent (in boldface) has moved to the beginning of the sentence, and
modal (in italics) precedes the subject (underlined). (We give the structure for such a negative
inversion sentence directly; for the moment, it is sufficient to understand that (29b) is structurally
analogous to the corresponding direct wh- question What would they accept more happily?)

(29) a. They would accept no present more happily.

Chapter 10: The verb raising parameter 291

b. No present would they accept more happily.

An important property of this construction is that the material preceding the modal must be a
maximal projection. Thus, in contrast to the full DP no present in (29b), the head of that DP, the
negative determiner no, cannot undergo negative inversion on its own, as shown in (30).

(30) * No would they accept present more happily.

Bearing in mind this fact about negative inversion, consider the canonical and negative
inversion sentences in (31).

(31) a. They will never tolerate this mess.

b. Never will they tolerate this mess.

(32) gives the structures for (31a) and (31b), which has the same structure as the corresponding
wh- question When will they tolerate this mess? For present purposes, what is important is that
never is a maximal projection and therefore a suitable candidate for negative inversion.

(32) a. IP

DPi I′

they I◦ VP
will
ti V′

AdvP V′

never V◦ DP
tolerate
this mess
b. CP

AdvPk C′

never C◦ IP

I◦m C◦ DPi I′
will [neg]
they I◦ VP
tm
ti V′

tk V′

V◦ DP
tolerate
this mess

Now consider the not variant of (31a) in (33).

Chapter 10: The verb raising parameter 292

(33) They will not tolerate this mess.

Under the reasonable assumption that Infl can take either NegP or VP complements, (33) has the
structure in (34).

(34) IP

DPi I′

they I◦ NegP
will
Neg′

Neg◦ VP
not
ti V′

V◦ DP
tolerate
this mess

In this structure, not on its own is not a maximal projection, and so not, like no but unlike never,
should not be able to undergo negative inversion. This expectation is confirmed by the ungram-
maticality of (35).

(35) * Not they will tolerate this mess.

A second piece of evidence for the status of not (and its variant n’t) as a transitive head
comes from the fact that it optionally adjoins to I, forming a complex head that can exhibit
morphological irregularities. For instance, n’t moves to will, the result is spelled out as won’t.
Moreover, various non-mainstream English varieties allow the combination of n’t with various
forms of the aspectual auxiliaries be and have to be spelled out as ain’t. Such irregularity is the
hallmark of two heads combining (whether in the syntax or in the morphology). Comparable
examples arise in connection with irregular past tense forms in English, where the combination
of two heads like sing and past tense is spelled out as irregular sang. Other well-known examples
from languages other than English include the idiosyncratic spellouts for preposition-determiner
combinations like those in (36).

(36) a. à + le > au; à + les > aux; de + le > du; de + les > des (French)
to the.m.sg to the.m.pl of the.m.sg of the.m.pl
b. an + dem > am; in + dem > im; zu + dem > zum; zu + der > zur (German)
to the.m.dat.sg in the.m.dat.sg to the.m.dat.sg to the.f.dat.sg
c. con + il > col; in + il > nel; su + il > sul (Italian)
with the.m.sg in the.m.sg on the.m.sg
d. por + o > pelo; por + a > pela; em + o > no; em + a > na (Portuguese)7
by the.m.sg by the.f.sg in the.m.sg in the.f.sg
Chapter 10: The verb raising parameter 293

A constraint on tense lowering in the morphology

We turn now to the second piece of our solution to the puzzle presented by the contrast between
(26) and (27), repeated here as (37) and (38).

(37) a. * They not applied.

b. ✔ They { did not, didn’t } apply.
(38) a. ✔ They never applied.
b. * They did never apply.

The idea is that tense lowering in the morphology (but not verb raising in the syntax!) is subject
to the locality condition in (39).

(39) When a head A lowers onto a head B in the morphology, A and B must in a local relation
in the sense that no projection of a head distinct from A and B intervenes on the path of
branches that connects A and B.

The notion of intervene is defined as in (40).

(40) An element C, C distinct from (projections of) A and B, intervenes between A and B if
and only if (some projection of) A dominates C and (some projection of) C dominates B.

It is important to understand that intervention is defined not in terms of linear precedence,

but in terms of the structural relation ‘dominate’. This means that the place to look for whether
the locality condition in (39) is satisfied or violated is not the string of terminal nodes beginning
with A and ending with B, but the path of branches that connects A with B in the tree.
The structure for (38a) is given in (41). In this structure, tense lowering is consistent with
the locality condition in (39), since adjoining never at V′ results in the adverb being too low in the
tree to intervene between I and V. (In other words, AdvP isn’t on the path from I to V, indicated
by the boxed nodes.)

(41) IP

DP I′

I VP
[pst]
DP V′

AdvP V′

never V
apply
Chapter 10: The verb raising parameter 294

By contrast, in the structure in (42a) tense lowering would violate the locality condition
because the path between I and V is interrupted by NegP and Neg′ . As a result, only the do
support variant of (42a) is grammatical, which is shown in (42b). It’s true that NegP and Neg′
intervene between I and V in (42b) as well, but forms of do are free morphemes. Therefore,
unlike tense affixes, they don’t need to undergo tense lowering onto V to form a well-formed
morphological word. Since (39) is a constraint on tense lowering, not a constraint on syntactic
trees in general, (42b) does not violate it.

(42) a. b. IP
IP
»DP« I′
»DP« I′
I◦ NegP
did
I NegP Neg′
[pst]
Neg◦ VP
Neg′ not
»DP« V′
Neg◦ VP
not V◦
apply
»DP« V′

V
apply

The loss of verb raising in Scandinavian 3

Before turning to the complex history of do support in the history of English, we review one
of its ingredients—the loss of verb raising—in the history of Scandinavian, where it occurred
without any complicating factors. The modern Scandinavian languages are descendants of Old
Norse (the language of the Vikings), which split into two main branches with names that are self-
explanatory: Mainland Scandinavian (Danish, Norwegian, Swedish) and Insular Scandinavian
(Faroese, Icelandic). Old Norse had verb raising, which is preserved in Icelandic, is in its very
final stages in Faroese, and has been lost in the Mainland Scandinavian languages (at least in their
standardized forms). The examples in this section are from or modeled after those in Holmberg
and Platzack (1995) and Platzack (1988).
The Icelandic facts are illustrated in (43). As usual with examples from the Germanic lan-
Chapter 10: The verb raising parameter 295

guages other than English, we give examples in the form of subordinate clauses in order to avoid
complications associated with aspects of main clause word order (see Chapter 12). As the exam-
ples show, Icelandic patterns with French.

Extra Info
The Icelandic characters eth (capital Ð, lowercase ð) and thorn (capital Þ, lowercase
þ) represent the voiced and voiceless ‘th’ sounds in this, eth and thin, thorn, respec-
tively. The Danish character ø and the Swedish character ö stand for the same mid-front
rounded vowel (more or less like the vowel in ‘day’, except that the lips are rounded as for
‘do’).

(43) Icelandic:
a. ✔ að Jón keypti { ekki, aldrei, raunverulega } bókina
that Jón bought not never actually book.def
‘that Jón { didn’t buy, never bought, actually bought } the book’
b. * að Jón { ekki, aldrei, raunverulega } keypti bókina

(44) and (45) illustrate the corresponding facts for modern Danish and Swedish. As is evident,
the pattern is the converse of the Icelandic one.

(44) Danish:
a. * at Ulf købte { ikke, aldrig, faktisk } bogen
that Ulf bought not never actually book.def
Intended meaning: ‘that Ulf { didn’t buy, never bought, actually bought } the book’
b. ✔ at Ulf { ikke, aldrig, faktisk } købte bogen
(45) Swedish:
a. * att Ulf köpte { inte, aldrig, faktiskt } boken
that Ulf bought not never actually book.def
Intended meaning: ‘that Ulf { didn’t buy, never bought, actually bought } the book’
b. ✔ att Ulf { inte, aldrig, faktiskt } köpte boken

The loss of verb raising in the Mainland Scandinavian languages has been the subject of
detailed investigation (Falk 1993; Platzack 1988; Sundquist 2003). For instance, in Swedish, the
earliest tense lowering examples are from the late 1400s. During a transition period from 1500 to
1700, both verb raising and tense lowering are attested, sometimes even in the same text (as in
examples (46b) and (47b)).

(46) a. at Gudz ord kan ey vara j honom

that God’s word can not be in him
‘that God’s word cannot be in him’
b. när thet är ey stenoghth
when it is not stony
‘when it is not stony’
Chapter 10: The verb raising parameter 296

(47) a. om den dristigheten än skulle wara onågigtt uptagen

if that boldness yet would be amiss taken
‘if that boldness would yet be taken amiss’
b. om annar sywkdom ey krenker nokon
if another illness not ails someone
‘if someone isn’t afflicted with another illness’
Finally, after 1700, verb raising in Swedish dies out completely in the standardized variety. How-
ever, it is preserved in at least one regional dialect, that of Älvdalen.
(48) Älvdalen Swedish:
a. ✔ um du for int gar ita ia firi brado
if you get not done this before breakfast
‘if you don’t get this done before breakfast’
b. ✔ fast die uar int ieme
if they were not home
‘if they weren’t home’
c. ✔ ba fo dye at uir uildum int fy om
just because that we would not follow him
‘just because we wouldn’t follow him’
Faroese, spoken on the Faroes, a group of islands roughly midway between Iceland and
mainland Europe, is at the very tail end of losing verb raising Heycock, Sorace, and Z. Hansen
(2010) and Heycock, Sorace, Z. S. Hansen, et al. (2011). Speakers do not ordinarily produce verb
raising sentences, but when asked to give grammaticality judgments, many speakers accept both
word orders in (49), characterizing the verb raising variant in (49a) as archaic.
(49) Faroese:
a. Verb raising (archaic):
Hann spur, hvi tad eru ikki fleiri tilikar samkomur.
he asks why there are not more such gatherings
‘He asks why there aren’t more such gatherings.’
b. Tense lowering (vernacular):
Hann spur, hvi tad ikki eru fleiri tilikar samkomur.
In concluding this section, we draw your attention to the absence of do support in (44b),
(45b), and (47b). As the availability of negative inversion in (50a) shows,8 the translation coun-
terparts of not have a different status in Scandinavian than they do in English. They are ordinary
intransitive adverbs, and therefore tense lowering is possible in modern Mainland Scandinavian.
(Icelandic negative inversion in (50b) is shown for completeness; the verb would be able to move
up to I even if negation were a head.)
(50) a. Swedish:
Inte vet jag var hon bor.
not know I where she lives
‘I don’t know where she lives.’
Chapter 10: The verb raising parameter 297

b. Icelandic:
Ekki veit ég hvar hún býr.
not know I where she lives

Verb raising in the history of English 4

This section provides a review of the loss of verb raising and the rise of do support in the history
of English.9 As we will see, this part of the grammar of modern English is the culmination of one
of the most complicated chapters in the history of the language. It reflects several distinct but
interlocking developments, which include:
• the loss of verb raising,
• a change in the status of not from intransitive (like never) to transitive,
• the emergence of do support,
• the emergence of a distinct class of modals, and
• the retention of verb raising in the case of two lexical items (have and be).

4.1 The loss of verb raising

Just like French, Middle English (1100–1500) exhibited verb raising. Examples with several di-
agnostic adverbs are shown in (51)–(53); they and the others in this section are from Kroch and
Taylor (2000a)

Extra Info
The characters eth (capital Ð, lowercase ð) and thorn (capital Þ, lowercase þ) were bor-
rowed from Old Norse and used in Old and Middle English where we use ‘th’ today. The
yogh character (ȝ) was used where we use ‘g’ or ‘y’ today.

(51) always:
a. he weneth alwey that he may do thyng that he may nat do. (cmctmeli,222.C1.193)
‘he always thinks that he can do things that he can’t do’
b. for þe Britons destroiede alwai þe cristen peple þat seynt Austyne hade baptisede
(cmbrut3,98.2951)
‘for the Britons always killed the Christians that St. Austin had baptized’
c. þe ȝong man resortyd alwey to þe preste (cmkempe,57.1270)
‘the young man always resorted to the priest’
(52) never:
Chapter 10: The verb raising parameter 298

a. for God … ȝeueþ neuer two tymes to-geder (cmcloud,20.115)

‘for God … never gives two times together’
b. and y ne10 seiȝ neuer þe ryȝtful for-saken (cmearlps,44.1880)
‘and I have never seen (lit. ‘not saw never’) the righteous forsaken’
c. he thought he sawe never so grete a knyght (cmmalory,180.2434)
‘he thought he had never seen so great a knight’
d. for þey synneden neuere. (cmwycser,234.204)
‘For they never sinned.’
(53) not:
a. This emperour Claudius was so obliuiows þat, sone aftir he had killid his wyf, he asked
why sche cam not to soper. (cmcapchr,49.534)
‘This emperor Claudius was so oblivious that, soon after he had killed his wife, he
asked why she didn’t come to supper.’
b. He mad eke a precept þat no Jew into Jerusalem schuld entre, but Cristen men he
forbade not þe entre. (cmcapchr,52.604-605)
‘He also made a law that no Jew should enter into Jerusalem, but he did not forbid
Christians from entering (lit. ‘the entry’).’
c. Ich ne hidde nouȝt þy mercy (cmearlps,49.2107)
‘I did not hide (literally, ‘not hid not’) your mercy’
d. Bott I sawe noght synne. (cmjulnor,60.289)
‘But I did not see sin.’
e. but he wythdrowe not hir temptacyon (cmkempe,16.321)
‘but he did not withdraw her temptation’
f. but Balyn dyed not tyl the mydnyghte after. (cmmalory,69.2361)
‘But Balyn did not die till the midnight after.’

In the course of Middle English, several syntactic developments took place that eventually
culminated in the complex grammar of modern English. First, between 1475 and 1525, the fre-
quency of verb raising dropped from roughly 65% to 10%. In the case of adverbs, this loss simply
led to the modern word order adverb > finite verb, as is evident from the translations for (51)
and (52). But the effects in the case of negation were more complicated and involved two further
changes: a change in the status of not and the emergence of do support. We discuss these changes
in turn.
Chapter 10: The verb raising parameter 299

4.2 A change in the status of not

Negative inversion
There is good evidence that in early Middle English not was an ordinary adverb on a par with
never and French and Scandinavian negation. Like never and negative phrases throughout the
history of English, it could undergo negative inversion.

(54) a. & nohht ne stannt itt stille (cmorm,I,125.1079)

and not ne stood it still
‘and it didn’t stand still’
b. Acc nohht ne mihht itt oppnenn hemm Þe ȝate off heoffness blisse
and not ne might it open them the gate of heaven’s bliss
(cmorm,I,142.1171)
‘and it could not open the gate of heaven’s bliss for them’

In the absence of further developments, the loss of verb raising would have resulted in
a word order change from verb > not to not > verb, as happened in mainland Scandinavian.
However, unlike Scandinavian (or French) negation, not in the course of Middle English went
from being an ordinary intransitive adverb to being a transitive head. As a result, not cannot
undergo negative inversion, as we already showed earlier in the chapter, and the modern English
counterparts of (54) are ungrammatical, as shown in (55).

(55) a. * Not did it stand still.

b. * Not could it open the gates of heaven’s bliss for them.

Adjunction to I′
There is a further piece of evidence that not changed from an intransitive head to a transitive a
head in the course of Middle English. In early Middle English, not could adjoin not just to V′ , but
also to I′ .

(56) a. Þatt Jesuss nohht ne wollde Ben borenn nowwhar i þe land (cmorm,I,122.1052)
that Jesus not ne wanted be born nowhere in the land
‘that Jesus did not want to be born anywhere in the land’
b. ða þinges ðe hie naht ne scolden ȝiuen (cmorm,I,142.1171)
the things that they not ne should give
‘the things that they shouldn’t give’

In this respect, not resembled never and other adverbs, which have preserved this ability
to this day, as shown in (57).11

(57) a. Middle English:

he swore þat Saxones neuer shulde haue pees ne reste (cmbrut3,69.2088)
‘he swore that the Saxons never should have peace or rest’
b. Modern English:
Chapter 10: The verb raising parameter 300

He { always, never } will admit his shortcomings.

However, as it developed from an intransitive to a transitive head, not lost the ability to
adjoin to I′ in the course of Middle English, with the result that the Modern English counterparts
of (56) are ungrammatical, as shown in (58).

(58) a. * that Jesus not would be born anywhere in the land.

b. * the things that they not should give

This is consistent with the elementary tree for Modern English not in (28b), where it is a transitive
head that can take a VP complement, forcing it to appear lower in the tree than required to
generate the word order in (58).

4.3 The emergence of do support

The reanalysis of not from an ordinary adverb to a head was essentially complete by 1400,12 and
shortly thereafter, the first examples of the contracted form n’t are attested, as we might expect.
What consequences did this have for children acquiring sentences containing not at that time?
On the one hand, verb raising was being lost. On the other hand, the new status of not as a head
ruled out tense lowering in negative sentences. In other words, the children were caught between
a rock and a hard place. In the absence of any other developments, ordinary negative sentences
would have become ineffable (that is, there would have been no way to say negative sentences
in English).
One can imagine a number of different resolutions to such an impasse, each of them repre-
senting a particular possible accident of history. For instance, speakers might have begun using
the adverb never to take over the function of the negative head not, as happened in Middle Por-
tuguese (Marlyse Baptista, personal communication). In fact, this is attested in the vernacular
modern language, as we saw in (25), but it never became the only way of expressing negation.
Alternatively, children might have managed to acquire verb raising solely based on the word or-
der in sentences containing not. This is what happened with be and auxiliary verb have. But with
ordinary verbs, what actually happened was something that depended on an unrelated develop-
ment in the language that had taken place in the 1200s: the development of the verb do into a
modal.
Like many languages, Middle English had a construction (no longer available in Modern
English) involving a causative verb and a lower verb, in which the lower verb’s agent could be
left unexpressed.13 We illustrate this construction for French and German in (59). The causative
verb is in boldface, and the lower verb is underlined.

(59) a. French:
Édouard a fait assembler une grande armée.
Edward has made assemble a great army
‘Edward had a great army assembled.’ (lit. ‘Edward had (someone) assemble a great
army.’)
b. German:
Chapter 10: The verb raising parameter 301

Eduard liess ein grosses Heer versammeln.

Edward let a great army assemble
‘Edward had a great army assembled.’

In Middle English, two different causative verbs were used in this construction depending
on the dialect (Ellegård 1953). The East Midlands dialect used do, as illustrated in (60), whereas the
West Midlands dialect used make. In other words, the West Midlands equivalent of (60a) would
have been (using modern spelling) Edward made assemble a great host.

(60) Middle English (East Midlands):

a. Kyng Edwarde dede assemble a grete hoste (cmbrut3,122.3377)
‘King Edward had a great army assembled.’
(lit. ‘King Edward had (someone) assemble a great army.’)
b. This Constantin ded clepe a gret councel at Constantinople (cmcapchr,81.1483)
‘This Constantine had a great council called at Constantinople’
(lit. ‘This Constantine had (someone) call a great council at Constantinople’)
c. He ded make ferþingis and halfpenies, whech were not used before
(cmcapchr,128.2962)
‘He had farthings and halfpennies made, which weren’t used before’
(lit. ‘He had (someone) make farthings and halfpennies’)

Now in many discourse contexts, causative sentences like The king had a great army as-
sembled are used more or less interchangeably with simple sentences like The king assembled a
great army. (In the first case, we zoom in, as it were, on the details of the situation—we explicitly
describe that the king is getting someone else to do the actual legwork of assembling the army,
whereas in the second case, we zoom out, ignoring the fact that the king isn’t running around
himself.) As a result, in situations of dialect contact, it was possible for West Midlands speakers
(those with causative make) to misinterpret sentences with causative do from the East Midlands
dialect as an alternative way of saying the corresponding simple sentence. Based on this misin-
terpretation, they might then themselves have begun to use do, but as an auxiliary verb bleached
of its causative content rather than as a causative verb (for which they would have continued to
use their own make). Since the border between the East and West Midlands dialects ran diago-
nally through the whole of England, the chances of dialect contact and of the reinterpretation and
adoption of do as an auxiliary verb were good. In any event, however it came to pass, it is West
Midlands speakers who first used do as an auxiliary verb. Once the auxiliary use was established,
it could then have spread to other dialects and especially the dialect of London, where people
came from many different regional backgrounds and where dialect distinctions were leveled as a
result.
What is important from a syntactic point of view is that auxiliary do occurred rarely before
1400. However, when agreement weakened and verb raising began to be lost, auxiliary do was
increasingly pressed into service since it allowed the ever-increasing number of speakers with
the tense lowering grammar to produce negative sentences with not.
Chapter 10: The verb raising parameter 302

4.4 The emergence of modals

In modern English, the do of do support is a modal (I) rather than an auxiliary verb (V) (see
Chapter 2, § 1.4 for further details concerning the distinction between these categories). We
know that auxiliary do entered the language as an auxiliary verb (Ecay 2015), but it must have
been reanalyzed as a modal early on, since as an auxiliary verb, it would have to combine with
tense and would thus, in the tense lowering grammar, run afoul of exactly the locality constraint
that it actually helped to circumvent. In any event, the do of do support was one of a growing
number of modals in Middle English that developed out of an earlier class of auxiliary verbs.
Historically, many members of this class exhibited morphological peculiarities, and some of them
were already syntactically special from the very beginning of Middle English. For instance, must
and shall never occur as nonfinite forms in Middle English. Children acquiring these vocabulary
items (what we might call premodals) would therefore never have encountered evidence that they
moved from V to I (unlike in the case of Modern English auxiliary have and be). The children
might instead have assigned these items to the syntactic category I from the get-go, and so they
might already have been modals in early Middle English.
Consider now the effect of the loss of verb raising on the status of any premodals that were
still members of the syntactic category V. In particular, consider a structure like (61) (we assume
that the premodals, just like modals, took VP complements).

(61) IP

»DP« I′

I◦ NegP
[tense]
Neg′

Neg◦ VP
not
»DP« V′

V◦ VP
premodal
… main verb …

In the outgoing verb raising grammar, a premodal can combine with tense even in the
presence of the new negation (where it is a head rather than an ordinary adverb) because verb
raising is not subject to the locality constraint on morphological tense lowering. In the incoming
tense lowering grammar, structures containing the new not (the head rather than the adverb) are
ordinarily rescued by do support, as shown in (62).
Chapter 10: The verb raising parameter 303

(62) IP

»DP« I′

I◦ NegP
do
Neg′

Neg◦ VP
not
»DP« V′

V◦ VP
premodal
… main verb …

But in contrast to sentences containing ordinary verbs, do support in a structure like (62)
might plausibly have been ruled out on the grounds that modal do inherited a constraint from
causative do that is given in (63).

(63) The complement of a causative construction must be headed by a ‘true’ verb (that is, not
a premodal, modal, or auxiliary verb like have or be).

The constraint on causative verbs that we are postulating is not specific to Middle English; it
is independently motivated by effects in modern English and German, as illustrated in (64) and
(65).

(64) a. ✔ The coach had the players run. (no auxiliary)

b. * The coach had the players be running. (auxiliary)
c. * The coach had the players have run.
(65) a. ✔ Der Trainer liess die Spieler laufen. (no auxiliary)
the coach had the players run.
‘The coach had the players run.’
b. * Der Trainer liess die Spieler am Laufen sein. (auxiliary)
the coach had the players at.the running be.
‘The coach had the players be running.’
c. * Der Trainer liess die Spieler gelaufen sein.
the coach had the players run.pst.ptcp be.
‘The coach had the players have run.’
d. * Der Trainer liess die Spieler laufen wollen.
the coach had the players run want.
‘The coach had the players want to run.’

Again, various ways out of this impasse are conceivable. For instance, the constraint in
(63) might have been relaxed for modal do. What actually happened, however, was that any
Chapter 10: The verb raising parameter 304

remaining premodals were reanalyzed as modals along the lines of must and shall, yielding the
modern structure in (66).

(66) IP

»DP« I′

I◦ NegP
modal
Neg′

Neg◦ VP
not
… main verb …

After this reanalysis, sentences like (67), with nonfinite forms of premodals like cunnen and
mowen, both meaning ‘be able to’, ceased to be possible in English (at least in mainstream va-
rieties).

(67) a. he schuld cun best rede þe booke (cmkempe,4.52)

‘He should be able to read the book best.’
b. I shal not conne wel go thyder (cmreynar,14.261)
‘I won’t be able to go there easily.’
c. and hij shul nouȝt mow stonde (cmearlps,19.765)
‘and he shall not be able to stand.’
d. Noo man shall mow resyst thy power in all thy life. (cmfitzja,A3R.28)
‘No man shall be able to resist your power in all your life.’

4.5 Remnants of verb raising in modern English

Despite the overall loss of verb raising in the history of English, verb raising is still possible with
two verbs in Modern English—namely, have and be. These two verbs, which did not belong to
the premodals, have functioned as both auxiliary verbs and main verbs throughout the history of
the language (as they do in Germanic and Romance more generally). The two uses are illustrated
for modern English in (68) and (69); auxiliaries are underlined and main verbs are in boldface.
(For more detailed discussion of the morphological and syntactic properties of have and be, see
Chapter 2, § 1.4.)

(68) Auxiliary verb:

a. I have read that chapter. (perfect)
b. I am reading that chapter. (progressive)
c. That material is treated in the next chapter. (passive)
(69) Main verb:
a. I have that book.
Chapter 10: The verb raising parameter 305

b. This chapter is difficult.

We begin by considering these verbs as auxiliaries in structures like (70) (we assume for
simplicity that the elementary trees for auxiliary verbs don’t have specifiers, but the assumption
isn’t crucial in what follows).

(70) IP
»DP« I′
I◦ NegP
[tense]
Neg′
Neg◦ VP
not
»DP« V′
V◦ VP
auxiliary verb
… main verb …

As just discussed in connection with modals, tense lowering is impossible in a structure

like (70) because not intervenes between tense and the verb, nor can the structure be rescued by
do support given the constraint in (63). This is exactly the situation in which the premodals were
reanalyzed as instances of I. In the case of the premodals, this reanalysis was possible because
they so rarely occurred as nonfinite forms. But an analogous reanalysis in the case of auxiliary
verbs couldn’t happen because nonfinite auxiliary have and be were productive in Middle English
(as they continue to be in Modern English). Some examples are given in (71) and (72); again, the
auxiliary verbs are underlined and the main verbs are in boldface. In addition, the element in I
(modal or premodal), which guarantees the nonfiniteness of the auxiliary verb, is in italics.

(71) a. y shulde haue axede of here no more (cmbrut3,19.562)

‘I should have asked no more of her’
b. and after he wolde haue conquerede al Scotland and Walys (cmbrut3,23.686)
‘and afterwards he would have conquered all Scotland and Wales’
c. And Gutlagh wolde haue went into his countree (cmbrut3,25.728)
‘And Gutlagh would have gone into his country’
(72) a. Bot euensang sal be saide wid foure salmes (cmbenrul,18.626)
‘But evensong shall be said with four psalms’
b. the wordes of the phisiciens sholde been understonden in this wise
(cmctmeli,226.C2.365)
‘the words of the physicians should be understood in this way’
c. A sone, Josias bi name, schal be born to the hous of Dauith (cmpurvey,I,13.518)
‘A son, Josias by name, shall be born to the house of David’
Chapter 10: The verb raising parameter 306

As in the earlier case of the development of the premodals to modals, various ways of re-
solving this impasse are conceivable. Again, for instance, the constraint against do occurring
with auxiliary elements might have been relaxed. However, what actually happened in the his-
tory of English is that children acquired the verb raising option with precisely these two lexical
items. As a result, the order of auxiliary have and be with respect to negation in modern English
parallels that in French.

English French
Verb raising We have not read the book. Nous (n’) avons pas lu le livre.
we ne have not read the book
We are not invited. Nous (ne) sommes pas invités.
we ne are not invited
No verb raising * We (do) not have read the book. * Nous (ne) pas avons lu le livre.
* We (do) not be invited. * Nous (ne) pas sommes invités.

Table 10.3: Order of auxiliary have and be with respect to negation in modern English and French

Let us now turn to the main verb uses of have and be. For a time, main verb have be-
haved syntactically like auxiliary have, raising from V to I and otherwise exhibiting the syntactic
behavior of a modal, as illustrated in (73).

(73) a. Negation without do support:

They haven’t any money. You haven’t any wool.
b. Question formation without do support:
Have they any money? Have you any wool?

However, in present-day American English, the pattern in (73) is archaic and has been replaced
by the pattern in (74), where main verb have exhibits the syntax of an ordinary verb.

(74) a. Negation with do support:

They don’t have any money. You don’t have any wool.
b. Question formation without do support:
Do they have any money? Do you have any wool?

The replacement of (73) by (74) in American English took place from about 1800 to 1950 (Zim-
mermann 2017). British English usage, which was more conservative during this time, is now to
some extent falling in line with American English.14
Finally, main verb be is the only main verb in English that continues to raise to I.

(75) No do support:
a. This chapter isn’t difficult.
b. Is this chapter difficult?
 Chapter 10: The verb raising parameter 307

(76) Do support:
a. * This chapter doesn’t be difficult.
b. * Does this chapter be difficult?

Even here, there are some indications that main verb be may one day follow in the footsteps
of main verb have. In imperatives, be, like all other verbs, allows do support.

(77) a. (Do) sit down.

b. (Do) have a seat.
c. (Do) be quiet.

And in so-called wh- imperatives, do support is obligatory even with main verb be. (In contrast
to true questions, these questions are only felicitous if the main verb refers to a property under
the addressee’s control.)

True question Wh- imperative

Why aren’t you quiet? ✔ *
Why don’t you be quiet (for a change)? * ✔
Why aren’t you Rh-positive? ✔ *
Why don’t you be Rh-positive (for a change)? * #

Table 10.4: Do support with main verb be in 𝑤ℎ- imperatives

Conceivably, even auxiliary have and be will one day allow or require do support. Sentences
like (78) may be paving the way.

(78) What kind of time frame do they be lookin’ at?

(overheard on trolley between 22nd and 30th Streets, Philadelphia, PA, 19 May 2012)

Notes 5
1. In what follows, we focus on the past tense since the present tense is not overtly marked in English. The -s of the
third person singular expresses subject agreement rather than present tense (Kayne 1989).

2. Yiddish and the southern German dialects from which it developed are exceptions in this regard. In these languages,
the synthetic simple past has been completely replaced by the analytic present perfect (Middle High German ich
machte ‘I made’ > Yiddish ikh hob gemakht, lit. ‘I have made’).

3. A comparable shift occurred in English from ‘they have to V’ to ‘they must V’. Such semantic shifts, with concomitant
changes in morphological status (see Note 4), are very common across languages.
 Chapter 10: The verb raising parameter 308

4. Such reanalysis might be the source of much, if not all, inflectional morphology. In many cases, especially in lan-
guages that are not written, the sources of the inflections would be obscured by further linguistic changes, primarily
phonological reduction. Consider, for instance, the development of the future tense in Tok Pisin, an English-based
contact language that originated in the 1800s and that has become the national language of Papua New Guinea. In
current Tok Pisin, particularly among speakers who learn it as a first language, the future marker is the bound mor-
pheme b-. We are fortunate to have written records of Tok Pisin from the late 1800s, and so we happen to know that
this morpheme is the reflex of the adverb phrase by and by, which the earliest speakers of Tok Pisin used to express
future tense. The modern form of the marker developed via reduction (by and by > baimbai > bai > b-). Without
written records, the derivation of the modern form would be speculation at best.

5. Historically, the negative marker in French was ne, and jamais and pas (an intensifier literally meaning ‘step’) had
no negative force of their own. Modern English has comparable intensifiers, as in I don’t want to do it one bit, at all.
In the course of the history of French, ne, being phonologically weak, was often elided in speech, and jamais and pas
were reanalyzed as carrying negative force. In modern French, ne is characteristic of the formal language, and in
some spoken varieties, such as Montreal French, ne hardly ever occurs. In the present discussion, we disregard ne,
treating it as a semantically meaningless particle and glossing it as ne.

6. Do support and the syntax of negation raises some thorny problems, and no completely satisfactory analysis of it
exists as yet. So although our analysis is adequate to explain the contrast between (26) and (27), it is not intended to
solve many other puzzles that have been discovered in connection with these phenomena.

7. See Hutchinson, Lloyd, and Sousa (2018), §§ 3.4, 3.8.

8. For some reason, Danish ikke ‘not’ cannot invert, perhaps because it cannot bear prosodic stress.

9. The discussion in this section is based on data and ideas in Frisch (1997), Kroch (1989), and Roberts (1993).

10. Early Middle English had a negative particle ne, etymologically cognate with French ne (see Note 5) and syntactically
comparable to it. The Middle English particle was lost between 1200 and 1400.

11. The possibility of adjoining adverbs to I′ complicates the assignment of structures to sentences with adverb-verb
word order once verb raising begins to be lost. This is because they could be instances of the old verb raising
grammar, with the adverb adjoined at I′ , or instances of the new grammar without verb raising, with the adverb
adjoined at either I′ or V′ . In any particular sentence, it isn’t possible to tell which is the right structure. But in
a corpus of sentences, it is possible to correct for the complication introduced by the possibility of adjunction to
I′ , because the frequency of adjunction to I′ has remained relatively stable from Early Middle English until today
(10–15% with never). This means that in a situation of grammar variation and change, frequencies of adverb-verb
order in a corpus that appreciably exceed the diachronically stable rate can reasonably certainly be attributed to the
tense lowering grammar.

12. Not continued to be available as an adverb with a low frequency into the 1600s. The evidence for this is the existence,
though rare, of negative sentences in Early Modern English of the modern mainland Scandinavian type, with not
preceding a finite verb, as in the examples from Shakespeare in (i).

(i) a. they deafe mens’ eares, but not edify.

b. he that filches from me my good name robs me of that which not enriches him.
c. Safe on this ground we not fear today to tempt your laughter by our rustic play.

These sentences are linguistic hybrids in the sense that they contain the adverbial not characteristic of early Middle
English, yet have lost verb raising, just like modern English. As adverbial not finally dies out completely in the 1600s,
so do sentences of this type.

13. The agentless construction discussed in the text was also attested with verbs of perception, as illustrated in (i).
 Chapter 10: The verb raising parameter 309

(i) a. They heard say that the English had won the battle of Agincourt.
‘They heard someone say that …, they heard it said that …’
b. They heard tell of the wages of sin.
‘They heard someone tell of the wages of sin.’

A fossilized form of the construction is the nominalization hearsay.

14. The replacement of (73) by (74) is complicated by the existence in both American and British English of the have
got pattern illustrated in (i), where have serves as an auxiliary verb rather than as the main verb. Sutherland (2000)
studies the competition among all three variants (have with and without do support and have got) in both varieties
of English.

(i) a. They haven’t got any money. You haven’t got any wool.
b. Have they got any money? Have you got any wool?

Exercises and problems 6

Exercise 10.1
Build structures for the Middle English sentences in (1).

Extra Info
A note on spelling: u and v were interchangeable in Middle English.

The data raise various issues beyond the ones concerning verb raising. For instance, is
never before a constituent? Is such a determiner or an adjective? What about numerals?
Solve the issues as best you can, and briefly describe the issues and justify your solutions.
Assume that you can bring evidence from Modern English (or other languages, for that
matter) to bear on the structures you are building for the Middle English sentences.

(1) a. Engist knew neuer before þat lande.

‘Engist never before knew that land.’
b. she saide she had neuer company of man worldely.
‘She said that she never had the company of any worldly man.’
c. Seynt Edmond vsyd euer after that prayer to his lyvys ende
‘Saint Edmund afterwards always used that prayer till the end of his life (lit. ‘to his
life’s end’)’
d. sche had euyr mech tribulacyon tyl sche cam to Iherusalem.
‘She always had much tribulation till she came to Jerusalem.’
Chapter 10: The verb raising parameter 310

e. I knewe never such two knyghtes.

‘I never knew two such knights.’
f. thes two gyauntes dredde never knyght but you.
‘These two giants never feared any knight but you.’

Exercise 10.2
A. Explain the grammaticality contrast in (2), assuming the judgments as given. If necessary,
invent a new syntactic category for so to belong to.

(1) A (challenging B): You’re lying; you didn’t go to the movies.

(2) a. B (responding to the challenge): I did so go to the movies.
b. B (responding to the challenge): * I so went to the movies.

B. Some speakers accept (2b) as a response to (1). How does the grammar of such speakers
differ from the grammar of speakers with the contrast in (2)?

Exercise 10.3
A. Build a structure for (1). (Don’t build structures for the material in parentheses.)

(1) They didn’t only write the letter (but they sent it).

B. Now build a structure for (2), making sure that it is consistent with the locality constraint
on head movement from the chapter.

(2) They not only wrote the letter (but they sent it).

C. There turn out to be two structures for (2). They are topologically distinct, but there is no
semantic difference between them. What’s the difference between the structure that you
came up in (B) and the alternative?

Exercise 10.4
Given the discussion in the chapter and the notes below, exactly one of the Korean sentences in
(1) is ungrammatical. Which sentence is it, and why is it ungrammatical?
Chapter 10: The verb raising parameter 311

Background Info

Assume that 하였다 hayessta is a morphologically simple head of category I (despite ap-
parently containing the same bound morpheme 었다 -essta as 먹었다 mek-essta). This is
exactly parallel to the way that we treat tense on auxiliary do in English. Korean allows
only left adjunction.

The nominative (nom) and accusative (acc) case morphemes explicitly indicating the gram-
matical functions subject and object, respectively, are included for completeness. They are
not important for the purposes of the exercise.

The data for this exercise have been simplified to ignore a syntactic process called scram-
bling. As a result, native speakers of Korean will find more than one sentence in (1) unac-
ceptable.

(1) a. 철수가 밥을 먹었다.

Chulswu-ka pap-ul mek-essta.
Chulswu-nom meal-acc eat-pst
‘Chulswu ate the meal.’
b. 철수가 밥을 먹지 아니하였다.
Chulswu-ka pap-ul mekci ani-hayessta.
Chulswu-nom meal-acc eat neg1-did
‘Chulswu did not eat the meal.’
c. 철수가 밥을 먹었다 아니.
Chulswu-ka pap-ul mek-essta ani.
Chulswu-nom meal-acc eat-pst neg1
‘Chulswu did not eat the meal.’
d. 철수가 안 밥을 먹었다.
Chulswu-ka an pap-ul mek-essta.
Chulswu-nom neg2 meal-acc eat-pst
‘Chulswu did not eat the meal.’

Exercise 10.5
A. The chapter discusses do support in negative sentences. In Modern English, do support is
also obligatory in the formation of direct questions from sentences with finite main verbs,
as shown in (1) and (2).

(1) They will apply. → ✔ Will they apply?

(2) a. They applied. → * Applied they?
b. They applied. → ✔ Did they apply?

Given the basic structure in (3), can you explain the contrast in (2)?
Chapter 10: The verb raising parameter 312

(3) CP

C′

C◦ IP
[q]
DPi I′

they I◦ VP
{ [pst], did }
ti V′

V◦
apply

B. Under what circumstances would the question in (2a) be grammatical?

C. Build the structure for (4).

(4) Won’t they apply?

Exercise 10.6
Build structures for the following expressions (adapted from the Shakespeare quotations in
Note 12).

(1) a. They not edify mens’ eares.

b. that which not enriches him
c. We not fear to tempt your laughter.

Problem 10.1
In both (1) and (2), the (a) examples are unambiguous, with an “also” reading associated with
the agent argument. The (b) examples are also unambiguous, with a “contrary to expectation”
reading. The (c) examples are structurally ambiguous between an “also” reading associated with
the activity and an “also” reading associated with the theme argument.

(1) a. They too will read the novel.

(They—in addition to an unspecified set of readers—will read the novel.)
b. They will too read the nvoel.
(Contrary to expectation, they will read the novel.)
c. They will read the novel too.
(They will—in addition to an unspecified set of activities—read the novel.)
(They will read—in addition to an unspecified set of readable things—the novel.)
(2) a. They too read the novel.
Chapter 10: The verb raising parameter 313

b. They did too read the novel.

c. They read the novel too.
A. Provide the two elementary trees for too required to account for the above facts. For clarity,
you can distinguish between tooalso and toocontrast .
B. How are the elementary trees from (A) integrated into larger structures?
C. Given your answers to (A) and (B), why doesn’t (2a) have a “contrary to expectation” read-
ing? (After all, too immediately precedes read in (2a) just as it does in (1b) and (2b).)
D. Given your answers to (A) and (B), why don’t (2a) or (2b) have an “also” reading associated
with the activity?

Problem 10.2
Explain the facts in (1) and (2). Ignore the first conjunct and the conjunction and.

They said they would finish the project, and …

(1) ✔ finish the project they { will, did }.
(2) a. * { will, did } finish the project they.
b. * finished the project they.
 11
VP shells

Chapter outline
1 The causative alternation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316
2 Double object sentences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318
3 Double complement sentences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322
3.1 Give and send . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322
3.2 Put . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
3.3 Persuade . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326
4 Object control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327
5 Passive and VP shells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329
6 Further issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 331
6.1 A locality constraint on idioms . . . . . . . . . . . . . . . . . . . . . . . . 331
6.2 Direct vs. indirect causation . . . . . . . . . . . . . . . . . . . . . . . . . . 333
6.3 Semantic nuances: Give, get, send . . . . . . . . . . . . . . . . . . . . . . . 335
6.4 Double object verbs in languages with inherent case . . . . . . . . . . . . 338
7 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344
8 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345

In Chapter 4, § 3.1, we mentioned the binary-branching hypothesis—the idea that syn-

tactic nodes have at most two daughters. At first glance, this hypothesis is threatened by the
existence of double object sentences in natural language, illustrated for English in (1).

(1) Malia will give Langston the receipts.

In such sentences, the verb appears to be associated with three semantic arguments (agent, recip-
ient, theme), and it looks like the recipient (Langston) and the theme (the receipts) must both be
represented as complements of the verb. This is the view that underlies traditional grammatical
terminology, where the verb is said to be ditransitive and to take two objects, the indirect object
(the recipient) and the direct object (the theme). In this chapter, we present a proposal for how
to make double object sentences consistent with the binary-branching hypothesis. The proposal
hinges on the fact that ditransitive verbs can be semantically decomposed into a causative part
and a second part whose meaning differs according to the verb in question.
Chapter 11: VP shells 315

The simplest case of causative decomposition is the causative alternation illustrated in

(2).

(2) a. The vase will drop.

b. The children will drop the vase.

It is easy to see that (2b) is roughly synonymous with The children will make (or let) the vase
drop. In order to capture this fact, we give the sentences in (2) the basic structures in (3). For
simplicity, we show only the relevant parts of the structure. The postverbal position of the vase
is the result of moving and adjoining the lower verb to the higher [caus] head (shown in (3b) by
the arrow, for example (2b)).

(3) a. VP b. VP

DP V′ DP V′

the vase V◦ the children V◦ VP

drop [caus]
DP V′

the vase V◦
drop

Extending this approach allows us to relate (1) to the semantically more basic Langston will
get the receipts, as in (4).

(4) a. VP b. VP

DP V′ DP V′

Langston V◦ DP Malia V◦ VP
get [caus]
the receipts DP V′

Langston V◦ DP
get
the receipts
Recursive structures like (3b) and (4b) (where a V◦ selects another VP) are known as VP
shells.1 In principle, the term could also apply to ordinary small clauses, where the higher verb
is let or make, but in practice, it is restricted to structures where the higher V head is silent. What
is important for our purposes is that VP shell structures contain two verbal heads, so that they
can accommodate three arguments without either of the heads being associated with more than
one complement—exactly as required by the binary-branching hypothesis.
We go on to provide a VP shell analysis of further verb classes, where the lower verb takes
not a DP complement as in (4b), but a PP, AdvP, or CP complement. A striking consequence
is that object control sentences like We persuaded the children to dance come out as causative
Chapter 11: VP shells 316

variants of subject control sentences (in other words, as equivalent to We caused the children
to agree to dance). Another welcome consequence is that the passive of verbs with causative
semantics patterns with the passive of ECM verbs. It is important to understand that this is not
because ECM verbs themselves have causative semantics (they don’t), but because the structures
with ECM verbs and with verbs with causative semantics are topologically analogous.
In the final section of the chapter, we show that in addition to maintaining the binary-
branching hypothesis, VP shells also allow us to preserve a generalization concerning idioms—the
idea that expressions with idiomatic meanings are invariably constituents. We then distinguish
the semantic notions of direct and indirect causation and relate the notion of direct causation to
the development of do support covered in the previous chapter. Among other considerations, the
distinction between direct and indirect causation is also helpful in helping to understand certain
differences between three verbs with otherwise similar meanings: give, get, and send. The chap-
ter concludes with a brief discussion of VP shells in Japanese and German, two languages with
inherent case. The case-marking facts of these languages, especially those of Japanese, provide
striking support for the VP shell analysis of verbs with causative semantics.
Before proceeding, a note of clarification is in order. According to the VP shell analysis,
the terms ‘ditransitive’, ‘double object verb’, ‘double complement verb’, and so on are misnomers.
Indeed, even calling the causative variant of drop in (3b) a transitive verb is a misnomer, since
the apparent object (the vase) remains the subject of the lower verb in the VP shell structure.
Nevertheless, for convenience, we will continue to use the familiar terms of traditional grammar
as purely descriptive terms, without intending to imply corresponding structural analyses.

The causative alternation 1

English has a class of verbs expressing movement or other changes of state that are known as
inchoative verbs. These verbs are intransitive, and their subject is a theme argument. (In this
respect, inchoative verbs resemble the passive of ordinary verbs, but recall from Chapter 8 that
unlike passives, they are incompatible with an agent by phrase.) Some examples are given in (5)
and (6); the theme argument is in boldface.

(5) a. The apple dropped.

b. The ball rolled (down the hill).
c. The boat sank.
(6) a. The water { boiled, froze }.
b. The ice melted.
c. The door opened.

Many (though not all) inchoative verbs also have a transitive use, where the subject is
Chapter 11: VP shells 317

an agent or cause initiating the change of state, and the theme argument appears in postverbal
position, as illustrated in (7) and (8).

(7) a. The children dropped the apple.

b. The children rolled the ball (down the hill).
c. The { explosion, sailors } sank the boat.
(8) a. The chemist { boiled, froze } the water.
b. The { cook, sun } melted the ice.
c. The { janitor, wind, key } opened the door.

The alternation between the intransitive and transitive uses of these verbs is known as the
inchoative-causative alternation (or causative alternation for short).
The causative meaning of the transitive variants suggests giving them the same structure
as ordinary causative sentences like (9).

(9) The children { let, made } the apple drop.

The structure for (9) is given in (10a); the verbs expressing causation (let or make) take a
small clause complement headed by the intransitive inchoative verb. In (10b), the analog to let or
make is an abstract (= silent) verbal head [caus]. For simplicity, we show only the relevant parts
of the structures.

(10) a. VP b. VP

DP V′ DP V′

agent V◦ VP agent V◦ VP
{ let, make } [caus]
DP V′ DP V′

theme V◦ theme V◦
inchoative inchoative

The structural parallel in (10) captures the semantic parallel between (7a) and (9), but the
word order is not yet right: in (7a), the verb precedes the theme. This is a garden-variety mismatch
between where an element is interpreted and where it is pronounced, and we solve it in the usual
way by movement. Specifically, the inchoative verb adjoins to the causative verbal head, as shown
in (11).
Chapter 11: VP shells 318

(11) VP

DP V′

agent V◦ VP

V◦i V◦ DP V′
inchoative [caus]
theme t i

We left-adjoin the verb by analogy to English causative suffixes like -ify and -ize and the
Japanese causative suffix -sase discussed later on in this chapter). The complex verbal head is
spelled out as a homonym of the simple inchoative head.

Double object sentences 2

The inchoative verbs in the previous section are all intransitive, but there is nothing to prevent
the lower VP in a VP shell structure from being transitive. For instance, sentences with transitive
get have corresponding ditransitive counterparts, as illustrated in (12).

(12) a. Langston got the receipts.

b. Malia got Langston the receipts.

In the transitive variant, a theme and a recipient substitute in the complement and specifier
position of get, respectively. As in the causative alternation, the ditransitive variant introduces
an additional argument—an agent. The schematic structures for the sentences in (12) are given
in (13).

(13) a. VP
DP V′
recipient V◦ DP
get
theme
Chapter 11: VP shells 319

b. VP
DP V′
agent V◦ VP
[caus]
DP V′
recipient V◦ DP
get
theme

Double object get, before head

movement

c. VP
DP V′
agent V◦ VP
V◦i V◦ DP V′
get [caus]
recipient t i DP
theme

After head movement

A careful consideration of the meaning of (12a) suggests a further decomposition of tran-

sitive get into a head that explicitly expresses change of state and a verb expressing the state of
possession. In other words, we are proposing structures corresponding to the following para-
phrases.

(14) a. Langston got the receipts.

= It came about that Langston has/had the receipts.
b. Malia got Langston the receipts.
= Malia made it come about that Langston has/had the receipts.

The structures themselves are given in (15). For simplicity, we omit the non-thematic specifier of
come (in other words, we are treating come like a raising verb).
Chapter 11: VP shells 320

(15) a. VP b. VP

V′ V′

V◦ VP V◦ VP
come
DP V′ V◦i V◦ DP V′
have ∅come
recipient V◦ DP recipient t i DP
have
theme theme

Transitive get, with change-of-state After head movement

head, before head movement

(16) a. VP

DP V′

agent V◦ VP
[caus]
V′

V◦ VP
come
DP V′

recipient V◦ DP
have
theme

Double-object get, with change-of-state

head, before head movement
Chapter 11: VP shells 321

b. VP

DP V′

agent V◦ VP

V◦k V◦ V′
[caus]
V◦i V◦ tk VP
have ∅come
DP V′

recipient t i DP

theme

After head movement

However, even with this simplification, these three-level VP shell structure are a bit daunting,
and so we will generally use the simpler two-level structures in (13) in what follows.
The VP shell analysis of double object get can be extended to other double object verbs, but
unlike in the case of get, the ditransitive and transitive variants aren’t spelled out as homonyms.
Verbs of transfer like bequeath, hand, lend, sell, send, and others are all basically causative variants
of get, with various additional semantic nuances as indicated informally in Table 11.1.

Double object verb Semantic nuance

bequeath get from owner after owner’s death
hand get by hand-to-hand contact
lend get from owner temporarily
sell get in exchange for money
send get by indirect means over some fairly long distance

Table 11.1: Some causative variants of get and their semantic nuances

It is worth noting that in Old English, sell was just the ordinary word for give; the ‘exchange
for money’ bit of meaning was added later. The difference between get and give is difficult to state;
we return to the issue later on in this chapter. There are also other double object verbs where
the head of the lower VP shell is not (some variant of) get, but rather some other transitive verb.
Table 11.2 shows some examples.
Chapter 11: VP shells 322

Transitive verb Double object verb

eat feed
hear tell
learn teach
see show

Table 11.2: Some causative variants of transitive verbs

Double complement sentences 3

3.1 Give and send

Many double object sentences have a double complement counterpart in which the order of the
recipient (red, bold) and theme (blue, italics) arguments is reversed and the recipient is expressed
as a PP rather than as a DP.

(17) a. Malia gave Langston the receipts.

b. Malia gave the receipts to Langston.

At first glance, double complement sentences seem to be completely synonymous with

their double object counterparts and to stand in a one-to-one correspondence with them. In-
deed, early on in generative grammar, it was held that any double complement sentence could
be transformed into a double object sentence by an operation known as Dative Shift (in many
languages, recipients are marked by dative case morphology or dative case particles). However,
certain semantic restrictions on the two sentence types have led this view to be abandoned (Green
1974; Oehrle 1976; Jackendoff 1990). For instance, recipients in double object sentences, but not
in double complement sentences, are constrained to be animate.

(18) Double object sentence:

a. Malia sent Langston the receipts.
b. * Malia sent the post office box the receipts.
(19) Double complement sentence:
a. Malia sent the receipts to Langston.
b. Malia sent the receipts to the post office box.

This effect is so strong that noun phrases that can be interpreted as inanimate in a double com-
plement sentence are coerced into an animate interpretation in the corresponding double object
sentence, if that is possible. For instance, in (20b), Philadelphia cannot be interpreted as a location,
Chapter 11: VP shells 323

as is possible in (20a), though it can be interpreted metonymically as ‘people at the Philadelphia

office’.2

(20) a. Malia sent the receipts to Philadelphia.

(ambiguous between location and metonymy reading)
b. Malia sent Philadelphia the receipts.
(only metonymy reading)

What the facts in (18)–(20) suggest is that ascribing exactly the same thematic role (that
of recipient) to the first DP in a double object sentence and to the PP in a double complement
sentence is not quite correct. Rather, the PP headed by to denotes a path or direction along which
the theme moves, and the complement of to denotes the path’s endpoint or goal, which can be
either a recipient at that location, as in (19a), or a pure location, as in (19b). We give the structures
that we are assuming shortly.
This move of carefully distinguishing between recipients and locations is supported by the
parallel between (18)–(20) on the one hand and the corresponding simple get and go sentences in
(21) and (22) on the other. (It is also possible to replace get by have or receive without changing
the judgments.)

(21) Parallel to double object sentence:

a. Langston got the receipts.
b. * The post office box got the receipts.
c. Philadelphia got the receipts.
(only metonymy reading)
(22) Parallel to double complement sentence:
a. The receipts went to Langston.
b. The receipts went to the post office box.
c. The receipts went to Philadelphia.
(ambiguous between metonymy and location reading)

Schematic structures for (21) and (22) are given in (23). ((23a) repeats (13a) for convenience.)
In order to rule out (21b), we need to assume that get imposes a selectional restriction on its
specifier according to which it must be filled by a recipient and not by a pure location.
Chapter 11: VP shells 324

(23) a. VP b. VP

DP V′ DP V′

recipient V◦ DP theme V
get
theme V◦ PP
go
P′

P◦ DP
to
goal

Embedding the structures in (23) under [caus] yields the facts in (18)–(20). In particular, the
ill-formedness of (18b) is directly related to the ill-formedness of (21b).
In the double complement examples presented so far, the path complement is headed by a
transitive P. The X′ schema leads us to expect that semantically suitable intransitive heads should
also be able to serve as complements of go, and this expectation is borne out in (24), where here
and there are AdvPs without arguments of their own.

(24) a. Malia sent the receipts { here, there }.

b. The receipts went { here, there }.

Notice that here and there unambiguously refer to locations or to paths with locations as end-
points. Therefore, neither (24a) nor (24b) have metonymy readings, in contrast to (20a) and (22c),
respectively. Moreover, the ill-formedness of (25a) and (25b) follows from the fact that these sen-
tences violate the selectional restriction imposed by get on its specifier (compare the absence of
location readings in (20b) and (22b)).3

(25) a. * Malia sent { here, there } the receipts.

b. * { Here, there } got the receipts.

From what we have said so far, it is clear that not every double complement sentence has
a double object counterpart. Specifically, double complement sentences with goals that are lo-
cations rather than recipients have no double object counterpart. However, since goals are not
required to be pure locations, but can instead be recipients, it might still be the case that every
double object sentence has a double complement counterpart. But this turns out not to be true
either. The reason is that in a double complement structure, the preposition to imposes a seman-
tic requirement on the theme: namely, that the theme travel (or at least be able in principle to
travel) to the goal denoted by the complement of to along some path. By contrast, themes in
double object sentences, which lack to, aren’t subject to such a requirement (at least not in prin-
ciple; we return to this issue later on in this chapter). For instance, since it is perfectly possible
for ideas or migraines to be the result of certain causes, the double object sentences in (26) are
acceptable.
Chapter 11: VP shells 325

(26) Double object sentence:

a. The scandal gave the reporter an idea.
b. Flickering lights give some people a migraine.
(27) Double complement sentence:
a. * The scandal gave an idea to the reporter.
b. * Flickering lights give a migraine to some people.

The reason that the corresponding double complement sentences in (27) are unacceptable is
that the idea and the migraine are conceptualized as arising within somebody’s mind or brain as
the result of a cause, but without having traveled there along some path. One way of putting this
in terms of thematic roles is to say that the subject of get in sentences like these is an experiencer
rather than an ordinary recipient. As expected, the simple get and go sentences in (26) and (27)
are parallel to (28) and (29).

(28) Parallel to double object sentence:

a. The reporter got an idea.
b. Some people get a migraine headache.
(29) Parallel to double complement sentence:
a. * An idea went to the reporter.
b. * A migraine headache goes to some people.

Contagious diseases, incidentally, seem not to be conceptualized as traveling along a path.

Instead, they are conceptualized as spreading (occupying their original location in addition to the
new location). This explains the contrast between (30) and (31).

(30) a. Lisa gave Joel a bad cold.

b. Joel got a bad cold.
(31) a. * Lisa gave a bad cold to Joel.
b. * A bad cold went to Joel.

3.2 Put
Another double complement verb is put, which can be decomposed into a VP shell structure
where [caus] takes a VP complement headed by be, which in turn takes a complement denoting
a location.
Chapter 11: VP shells 326

(32) VP

DP V′

agent V◦ VP
[caus]
DP V′

theme V◦ XP
be
location

Unlike give or send, put cannot be associated with a recipient argument. Even human or
animate complements of be receive a purely locative interpretation. As a result, put appears in
double complement sentences, but (at least in English) not in double object sentences, as shown
in (33).4

(33) a. Amy put the books { on the shelf, there }.

b. * Amy put { the shelf, there } the books.

3.3 Persuade
In the VP shells that we have considered so far, the complements in the lower VP shell have been
DP (double object verbs) or PP (double complement verbs, put). In persuade, we have the case of
a VP shell where the lower verb takes a clause (CP) complement, as shown in (34).

(34) a. We will persuade them that they should dance.

b. We will [caus] them [agree] that they should dance.

The relevant VP shell structure is shown schematically in (35). (We omit head movement
for simplicity.)

(35) VP

DP V′

we V◦ VP
[caus]
DP V′

them V◦ CP
agree
that they should dance
Chapter 11: VP shells 327

Object control 4
Persuade takes nonfinite complement clauses as well, as shown in (36).

(36) a. We will persuade them to dance.

b. We will [caus] them [agree] to dance.

According to the causative decomposition in (36b), the lower VP shell in such sentences is headed
by an abstract subject control verb (Larson 1988). The VP shell structure is shown in (37).

(37) a. VP
DP V′
we V◦ VP
[caus]
DP V′
them V◦ CP
agree
C′
C◦ IP
∅
PRO to dance

Object control, before head movement

b. VP
DP V′
we V◦ VP

V◦i V◦ DP V′
agree [caus]
them t CP
i

C′
C◦ IP
∅
PRO to dance

After head movement

Chapter 11: VP shells 328

In (37), the reference of PRO is controlled by the subject of [agree]. But unlike in the cases
of subject control discussed in Chapter 9, § 1, where the subject of the control verb is assigned
nominative case, here the subject in question—them—is assigned objective case in the head-spec
configuration. The case assigner is the complex verbal head (spelled out as persuade), which
inherits its case-assigning power from [caus]. Superficially, therefore, them looks like an object,
and so verbs like persuade are known as object control verbs. The superficial difference between
subject control verbs and object control verbs is highlighted in (38). With subject control verbs,
the verb immediately precedes the nonfinite complement; with object control verbs, an apparent
object intervenes.

(38) a. They agreed [ to dance ]. (subject control)

b. We persuaded them [ to dance ]. (object control)

The analysis just presented is straightforwardly consistent with the contrast between ordi-
nary ECM verbs and object control verbs illustrated in (39).

(39) a. We expected there to be a problem

b. * We persuaded there to be a problem

The relevant partial structures for the two sentences are shown in (40). Head movement is omitted
in (40b) to clarify the relevant licensing relationships.

(40) a. VP

DP V′

we V◦ IP
expect
DPi I′

there I◦ VP
to
ti V′

be a problem
Chapter 11: VP shells 329

b. VP

DP V′

we V◦ VP
[caus]
DP V′

there V◦ CP
agree
C′

C◦ IP
∅
DPi I′

PRO I◦ VP
to
ti V′

be a problem

In (40a), expletive there is licensed by originating as the specifier of main verb be in the comple-
ment clause. (40b), on the other hand, is ruled out for exactly the same reasons as (41).

(41) * There agreed [ PRO to be a problem ].

First, expletive there is not licensed, since neither ordinary nor abstract agree are verbs of exis-
tence. Second, both ordinary and abstract agree select rational beings as subjects, and expletive
there fails to satisfy this selectional restriction.

Passive and VP shells 5

As just noted, [caus] assigns objective case in the head-spec configuration to the subject of its
complement. Like any other verb, [caus] loses its case-assigning ability when passivized. The
relevant VP shells are shown in (42).
Chapter 11: VP shells 330

(42) a. VP b. VP

DP V′ V′

agent V◦ VP V◦ VP
[caus] caus
[ pass]
»DP« V′ »DP« V′

… …

Although the labels for the syntactic categories differ, the configurations in (42) are analogous to
those for an ECM verb like expect in (43).

(43) a. VP b. VP

DP V′ V′

agent V◦ VP V◦ VP
expect expect
[ pass ]
»DP« I′ »DP« I′

… …

As a result, the parallels between ECM verbs and the VP shell verbs in Table 11.3 are ex-
pected.5
Chapter 11: VP shells 331

Active Passive
ECM verb We expected the children to find → The children were expected to
the Easter eggs. find the Easter eggs.
Double object verb We gave them a present. → They were given a present.
Double complement We sent them to the park. → They were sent to the park.
verb
Object control verb We persuaded the children to find → The children were persuaded to
the Easter eggs. find the Easter eggs.
ECM verb We expected the children to find → * The Easter eggs were expected
the Easter eggs. the children to find.
Double object verb We gave them a present. → * A present was given the children.
Object control verb We persuaded the children to find → * The Easter eggs were persuaded
the Easter eggs. the children to find.

Table 11.3: Passivization patterns in ECM and VP shell verbs

Further issues 6

6.1 A locality constraint on idioms

Idioms are expressions whose meaning does not follow straightforwardly from the individual
parts, as in (44). (The idiomatic meaning can sometimes be traced back to an etymological source,
but even if that is possible, the source is unknown to most of the idiom’s users.)

(44) a. red tape ‘bureaucratic difficulties’

b. the Big Apple ‘New York City’
c. kick the bucket ‘die’
d. let the chips fall where they may ‘disregard the consequences of one’s actions’

It has been traditional in generative grammar (Marantz 1984) to (attempt to) impose a lo-
cality constraint on idioms along the lines of (45) (locality constraints are so called because they
make reference to relatively small, or local, domains).

(45) All parts of an idiomatic expression must together form a constituent.

The examples in (44) trivially satisfy this constraint: red tape is an NP, the Big Apple is a DP, and
kick the bucket and let the chips fall where they may are instances of V′ .
Chapter 11: VP shells 332

The motivation for (45) is to account for the absence of theoretically possible idioms like
the made-up example in (46), where blue and hopping, though adjacent, don’t form a con-
stituent.

(46) a. They’ve bred a strain of blue hopping drosophila.

Intended meaning: ‘They’ve bred a strain of drosophila that is unusually large.’
b. The great apes all have blue hopping brains.
Intended meaning: ‘The great apes all have unusually large brains.’
c. She’s a blue hopping child for her age.
Intended meaning: ‘She’s an unusually large child for her age.’

Since idioms consisting of discontinuous chunks should not exist, idioms like those in (47)
at first glance seem to pose a problem for the locality constraint.

(47) a. give someone the creeps ‘make someone uneasy’

b. throw someone to the wolves ‘sacrifice someone’

However, just as the VP shell analysis allows us to preserve the binary-branching hypothesis
in the face of prima facie counterevidence, it also allows us to preserve the locality constraint
on idioms in the face of apparently discontinuous idioms. This is because the VP shell analysis
allows us to say that what is idiomatic in (47) are the underlined instances of V′ in (48).

(48) a. [caus] someone [get] the creeps

b. [caus] someone [go] to the wolves (by throwing)

Strong evidence for the decomposition in (48a) is the existence of the related idiom in
(49).

(49) get the creeps ‘become uneasy’

We attribute the absence of an idiomatic interpretation for go to the wolves to the absence of the
salient manner specification; compare the corresponding idiomatic passive form get thrown to the
wolves ‘be sacrificed’. (You are invited to explain the unavailability of noncausative throw to the
wolves in Problem 11.6.)
Since heads form constituents with their complements but not with their specifiers, poten-
tial idioms such as those in (50) are predicted not to be possible.

(50) a. the { creeps, wolves } [get] someone

b. the { creeps, wolves } [go] to someone

This straightforwardly explains the unacceptability of sentences like (51) and (52) (on their in-
tended idiomatic interpretation).

(51) a. * The creeps got me.

b. * The wolves went to Felix.
(52) a. * Oscar threw the wolves Felix. (=[caus] the wolves [get] Felix)
b. * Crazy people give the creeps to me. (=[caus] the creeps [go] to me)
Chapter 11: VP shells 333

6.2 Direct vs. indirect causation

Although [caus] is syntactically analogous to overt causative verbs like let and make, we do not
claim that it is completely synonymous with them. Overt causative verbs express what has been
called indirect causation, where the causing event and the caused event are conceptualized as
two separate events. By contrast, [caus] expresses direct causation, and the causing event is not
conceptualized as distinct from the caused event. It is important to realize that direct and indirect
causation are conceptual categories that speakers impose on the universe of discourse; they do
not themselves reflect distinctions inherent in that universe. In other words, the very same event
in the real world can be conceptualized as involving either indirect or direct causation. We can
think of this choice as corresponding to a camera zooming in or out on a given scene. For instance,
a bridge-building event could be described as involving indirect causation. In (53a), the agent of
the caused event is expressed as the subject of an active verb, whereas in (53b), the agent of the
caused event is expressed as a by phrase modifying a passive verb. Despite the difference in voice
(active vs. passive), the agent of the caused event (the two legions) is expressed explicitly in both
sentences.

(53) a. Caesar had two legions build a bridge.

b. Caesar had a bridge built (by two legions).

It is also possible to zoom out, treating some of the complexity associated with the bridge-
building as not at issue, and to describe the same event as in (54).

(54) Caesar built a bridge.

From this zoomed-out perspective, the bridge-building is an event with a single agent. The legions
can no longer be integrated into this sentence as a subordinate agent, but only as an instrument
wielded by the sole remaining agent.

(55) a. * Caesar built a bridge by two legions.

b. ✔ Caesar built a bridge { using, with } two legions.

A naturally-occurring example of zooming is given in (56), where the speaker elaborates on a

possibly misleading description of a movie scene.

(56) And he put the rope around his neck,

made him put the rope around his neck.
(Kendall and Farrington 2021, DCA_se2_ag2_m_03_xmin=848.83082)

It is worth noting that the zoomed-in and zoomed-out perspectives on event complexity
correlate with the complexity of the surface morphology in an arguably iconic manner. The
zoomed-in, more detailed perspective in (53) is expressed by two surface verbs (causative have
and build), whereas the zoomed-out, less detailed perspective in (54) is expressed by a single
surface form (build).
Although this chapter focuses on causative decomposition of only certain verb classes, we
would like to point out that the availability of [caus] opens the possibility of treating all (or at
Chapter 11: VP shells 334

least most) transitive verbs as containing [caus]. Pursuing this approach, apparently simple build
would be derived from [caus], the ‘come’ head mentioned earlier, and a verbal head meaning
something like ‘be-in-a-built-state’. A sentence like (54) would then be derived from a structure
like (57a), corresponding to a paraphrase like (57b).
(57) a. VP

DP V′

Caesar V◦ VP
[caus]
V′

V◦ VP
come
DP V′

a bridge V◦
be-in-built-state
b. Caesar caused it to come about that a bridge was in a built state.
If it were possible to generalize this approach to all transitive heads, the notion of ‘complement’
(along with the notion of ‘intermediate projection’) could be eliminated from the theory of phrase
structure. The price of this simplification is the inclusion in the theory of abstract heads like
[caus]. Much recent work in morphosyntax adopts the approach just sketched, and Problem 11.3
invites you to evaluate this idea.
The distinction we have just drawn between direct and indirect causation allows us to
revisit a point that we made in connection with the rise of do support in the history of English.
Recall from Chapter 10, § 4.3, that Middle English allowed a variant of (53a) where the subordinate
verb exhibits active voice, but where the subordinate agent is expressed with an optional by
phrase, as in the passive.
(58) Caesar { did, had, let, made } build a bridge (by two legions).
Recall further that the causative verb was make in certain dialects of Middle English and
do in others. For native speakers of both dialects, their own causative verb expressed indirect
causation. However, Ellegård (1953) surmises that in a situation of dialect contact, speakers of
the make dialect misanalyzed the do of the do dialect in sentences like (58) as an overt expression
of direct causation. This mis- or reanalysis would then have taken hold as a way of circumventing
the ineffability of simple negative sentences with structures violating the locality constraint on
tense lowering, at least in sentences containing verbs with causative semantics. In modern Eng-
lish, do has gone one step further, having developed into a modal that is compatible even with
verbs without causative semantics (Ecay 2015). This is comparable to the development of other
modals; for instance, the modern English future auxiliary will originally had the meaning ‘want’,
but has developed into a pure tense marker, able to co-occur with entities that are incapable of
wanting.
Chapter 11: VP shells 335

6.3 Semantic nuances: Give, get, send

Double object verbs like get, give, and send all share a basic meaning of ‘cause to (come to) have’.
Nevertheless, they are not interchangeable with each other in every context, and even where
they can appear in the same context, they do not always have the same meaning. In this section,
we attempt to shed some light on these differences. Because of our focus here on the differences
rather than on the similarities among the various verbal heads that combine with [caus], we need
a convention to distinguish among them. In the syntax literature, the convention is to put the
eventual spellout form in all caps under a mathematical root sign (the idea is that the element
√
in question is the root of the eventual spellout form). For instance, give indicates the root
that after adjoining to [caus] in the syntax will be spelled out with it as a form of give in the
morphology.
We begin by recalling that the subject of have is a recipient, but cannot be a location. This
explains the contrast between (59) and (60).

(59) Lisa { ✔gave, ✔got, ✔sent } Joel a book.

(60) Lisa { *gave, *got, *sent } there a book.
√
(61) shows that send imposes a selectional restriction on its complement (the theme ar-
gument), which to a first approximation requires the theme to be able to undergo transfer along
a path.

(61) Lisa { ✔gave, ✔got, #sent } Joel a { raise, promotion }.

As always when selectional restrictions are at issue, figurative language use complicates the pic-
ture. For instance, if raise or promotion are used metonymically for letters containing confirma-
tion of a raise or promotion (as in So, have all the the promotions gone out?), then using sent (61)
becomes (more) felicitous. And love or hugs are certainly often equated metaphorically with a
substance that can be enclosed in an envelope or conveyed along an electronic pipeline, explain-
ing uses like I send my love or I’m sending a big hug.
(62) is also consistent with the proposed selectional restriction. In (62a) and (62b) job is
ambiguous between ‘piece of work’ and ‘way of earning a living’. (62c), however, is unambiguous
and has only the ‘piece of work’ meaning.

(62) a. Lisa gave Joel a job. (ambiguous)

b. Lisa got Joel a job. (ambiguous)
c. Lisa sent Joel a job. (unambiguous)

So far, give and get have been patterning alike, but that is not always the case. For instance,
(63a) is ambiguous between Lisa handing Joel an idea (written down, say, on a piece of paper) vs.
Lisa somehow (possibly unbeknownst to herself) giving rise to an idea in Joel’s mind. In the first
scenario, Lisa is an agent, and Joel is a recipient. In the second scenario, Lisa is a cause, and Joel
is an experiencer. This ambiguity is lost in (63b) and (63c), which have only the agent-recipient
sense.

(63) a. Lisa gave Joel an idea. (ambiguous)

Chapter 11: VP shells 336

b. Lisa got Joel an idea. (unambiguous)

c. Lisa sent Joel an idea. (unambiguous)

It is worth noting that the noncausative alternant with get—Joel got an idea—can have the
meaning ‘An idea arose in Joel’s mind’; in fact, that is its ordinary meaning. This fact suggests
√
that when not embedded under [caus], gave is spelled out as a form of get.
The distinction between the agent-recipient and the cause-experiencer senses is highlighted
in (64), where the subject, being inanimate, cannot be an agent. Subtracting the agent-recipient
reading, as it were, from the corresponding examples in (63) makes give in (64) unambiguous and
got/sent in (64) infelicitous.

(64) Lisa’s haircut { ✔gave, #got, #sent } Joel an idea.

Two further examples of the same type are (65) and (66).

(65) { Lisa, the nasty weather } haircut { ✔gave, #got, #sent } Joel a cold.
(66) The flickering lights { ✔gave, #got, #sent } my friend a migraine.

The interpretation of the send examples in (64)–(66) follows straightforwardly from the selectional
√
restriction that send imposes on its theme. But let’s consider the infelicity of the get examples
more closely. Recall that abstract [caus] expresses direct causation—the causing event and the
caused event are conceptualized as non-distinct. When the caused event is a phenomenon that
arises inherently in an experiencer, direct causation gives rise to a weird interpretation—one
where an agent initiates a causing event that has a magic effect on the experiencer along the
lines of “Poof, now you’ve got an idea, a cold, a migraine, …” The interpretation is weird in a way
that “Poof, now you’ve got a job, a car, …” isn’t (assuming that the agent has a job or car to give
away).
Reconceptualizing the external cause and the internal experience as distinct (in other
words, as involving indirect causation) and accordingly expressing the event chain with sepa-
rate overt causative verbs cancels the weird magic effect, and the resulting sentences become
acceptable, as in (67).

(67) a. ✔ Lisa made Joel get a cold (by sneezing at him repeatedly).
b. ✔ The nasty weather made Joel get a cold.
c. ✔ The bright lights made my friend get a migraine (by causing the neurons in her
brain to misfire).
√
For reasons that remain mysterious, the combination of [caus] and give patterns like (67). In
other words, even though give is synthetic just like get, it behaves in (64)–(66) as if it were analytic
like make … get.
We turn now to the double-complement counterparts of the above sentences. These all
involve a lower go head, which imposes a selectional restriction on its theme argument (already
√
mentioned earlier for send). In other words, the theme must be an entity capable of undergoing
√
transfer along a path (with the usual caveats concerning figurative usage). In addition, the give
Chapter 11: VP shells 337

√ √
variant of go, unlike get or send, requires the goal to be a recipient. Both of these selectional
restrictions together explain the pattern of judgments in (68) and (69).

(68) a. We { ✔gave, ✔got, ✔sent } the tickets to Gillian.

b. We { #gave, #got, #sent } a headache to Gillian.
(69) We { #gave, ✔got, ✔sent } the tickets there.

As expected, (70a) has only a metonymy reading (where Hawaii means ‘people at the
Hawaii office’), whereas (70b) and (70c) also have the ordinary locative reading.

(70) a. We gave the tickets to Hawaii. (unambiguous)

b. We got the tickets to Hawaii. (ambiguous)
c. We sent the tickets to Hawaii. (ambiguous)

In examples like (71a), the theme argument (the noise) is conceptualized as moving along a
path from a source location where it doesn’t affect or bother the experiencers at a goal location
(them) to one where it finally does. The causative variant in (71b) remains infelicitous because of
the “weird genie” effect discussed above.

(71) a. ✔ The noise finally got to them.

b. # We finally got the noise to them.

To explain the difference with the apparently analogous example in (72), we invoke metaphor,
overtly signalled by the presence of the particle through. The argument at the goal location is
conceptualized as a recipient rather than as an experiencer.

(72) a. ✔ Our idea finally got through to them.

b. ✔ We finally got our idea through to them.

We conclude this section with two unresolved issues. First, for some speakers, get and send
in (73) and (74) both entail (75); for other speakers, that is true only for get.

(73) a. We got the knitters the yarn.

b. We got the yarn to the knitters.
(74) a. We sent the knitters the yarn.
b. We sent the yarn to the knitters.
(75) The knitters got/have the yarn.

For (74b), we could say that some speakers allow to to include the meaning ‘towards’,
whereas other speakers don’t. But that doesn’t explain why the loose sense of to is restricted
to sentences with send, and it is also not clear how to extend the idea to the double object sen-
tences, which don’t contain a preposition. What is clear is that send implies a more complicated
path than get, with more room for the transfer of the theme to go astray. In other words, we
can think of send as a zoomed-in version of get. But how to apply this intuition to explain the
Chapter 11: VP shells 338

variation in judgments in a principled way (that is, in a way that doesn’t simply restate the facts)
is not clear, and we leave the issue unresolved here.
A second unresolved issue concerns the constraint according to which recipients in dou-
ble object sentences must be animate. This constraint is actually not quite correct, as shown in
(76).

(76) a. The carpenter { gave, ?got, #sent } the cabinet a fancy handle.
b. The sellers { gave, ⁇got, #sent } the house a new coat of paint.

Note the roughly corresponding judgments for the noncausative alternants in (77), but also the
bleached sense of accept in (78).

(77) a. The cabinet { got, received, #accepted } a fancy handle.

b. The house { got, received, #accepted } a new coat of paint.
(78) This fare machine doesn’t accept coins.

The sentences in (76) and (77) suggest that inanimate objects can be metaphorically conceptu-
alized as animate, but with limits that are not easy to define. Similar issues arise in connection
with sentences like (79).

(79) The movers { ✔gave, *got } the couch a shove.

6.4 Double object verbs in languages with inherent case

Case in Japanese analytic and synthetic causative sentences
As discussed in Chapter 7, some languages have richer case morphology than does modern Eng-
lish. For instance, Japanese has a distinction between dative and accusative that English once
had, but lost in medieval times. Strikingly, in Japanese, the VP shell analysis of double object
causatives that we have argued for above on semantic grounds is supported by facts from case
morphology.
We begin by considering case marking in simple intransitive and transitive Japanese sen-
tences. As shown in (80), subjects and direct objects are marked with the nominative case particle
-ga and the accusative particle -o, respectively.6

(80) a. 太郎が 走った こと

Tarō-ga hashit-ta koto
-nom run-pst that
‘(the fact) that Tarō ran’
b. 太郎が りんごを 食べた こと
Tarō-ga ringo-o tabe-ta koto
-nom apple-acc eat-pst that
‘(the fact) that Tarō ate an apple’

Embedding a sentence under the causative suffix -(s)ase has the following effects on case
marking. (The suffix is spelled out as -sase or -ase depending on whether the lower verb that it
Chapter 11: VP shells 339

combines with ends in a vowel or a consonant.) When the complement sentence is intransitive,
the matrix subject is marked with -ga, as usual, but the subject of the complement clause is marked
with -o, as shown in (81). (This is analogous to what happens in English when we embed They ran
under a causative verb and end up with We made them run, with objective case on the embedded
subject.)

(81) 花子が 太郎を 走らせた こと

Hanako-ga Tarō-o hashir-ase-ta koto
-nom -acc run-caus-pst that
‘(the fact) that Hanako made Tarō run’

Given (81), one might expect embedding a transitive sentence under a causative to result in
the case-marking pattern in (82), where the object of the lower clause is marked with -o because
it is an object, as in (80b), and the subject of the lower clause is also marked with -o, by analogy
to (81) (cf. English They chased him and We made them chase him).

(82) * 花子が 太郎を りんごを 食べさせた こと

Hanako-ga Tarō-o ringo-o tabe-sase-ta koto
-nom -acc apple-acc eat-caus-pst that
Intended meaning: ‘(the fact) that Hanako made Tarō eat an apple’

As it turns out, however, the case-marking pattern in (82) is ungrammatical, violating what is
known in the Japanese syntax literature as the double -o constraint, which prohibits the occur-
rence of more than one -o-marked noun phrase per surface verb. Rather, when a transitive sen-
tence is embedded under a causative verb, the subject of the lower clause must be marked with
the dative case marker -ni, as in (83).7

(83) 花子が 太郎に りんごを 食べさせた こと

Hanako-ga Tarō-ni ringo-o tabe-sase-ta koto
-nom -dat apple-acc eat-caus-pst that
‘(the fact) that Hanako made Tarō eat an apple’

Strikingly, double object sentences exhibit the very same nominative-dative-accusative

case-marking pattern. (84) shows the simple sentences underlying both the ordinary small clause
causatives in (85) and their double object counterparts in (86).

(84) a. 太郎が 本を 見た こと
Tarō-ga hon-o mi-ta koto
-nom book-acc see-pst that
‘(the fact) that Tarō saw the book’
b. 太郎が 手紙を 受け取った こと
Tarō-ga tegami-o uketot-ta koto
-nom letter-acc receive-pst that
‘(the fact) that Tarō received the letter’
Chapter 11: VP shells 340

(85) a. 花子が 太郎に 本を 見させた こと

Hanako-ga Tarō-ni hon-o mi-sase-ta koto
-nom -dat book-acc see-caus-pst that
‘(the fact) that Hanako made Tarō see the book’
b. 花子が 太郎に 手紙を 受け取らせた こと
Hanako-ga Tarō-ni tegami-o uketor-ase-ta koto
-nom -dat letter-acc receive-caus-pst that
‘(the fact) that Hanako made Tarō receive the letter’
(86) a. 花子が 太郎に 本を 見せた こと
Hanako-ga Tarō-ni hon-o mise-ta koto
-nom -dat book-acc show-pst that
‘(the fact) that Hanako showed Tarō the book’
b. 花子が 太郎に 手紙を 送った こと
Hanako-ga Tarō-ni tegami-o okut-ta koto
-nom -dat letter-acc send-pst that
‘(the fact) that Hanako sent Tarō the letter’

The identical pattern in (85) and (86) is exactly what the VP shell analysis leads us to ex-
pect, since the relevant structures are completely analogous. The structures before and after verb
movement are shown in (87).

(87) a. VP

DP V′

agent VP { V◦ }
-sase
DP V′ caus

recipient DP V◦
verb
theme

Before head movement

Chapter 11: VP shells 341

b. VP

DP V′

agent VP V

DP V′ V◦i { V◦ }
verb -sase
recipient DP ti caus

theme

After head movement

The verbal heads in (87) are spelled out differently in the morphology. The spellout is either
analytic, with each syntactic head corresponding to a transparently identifiable morphological
form (mi-, uketor- and -(s)ase), or it is synthetic, with the two heads corresponding to a single
morphological item (mise-, okut-).
The Japanese causative exhibits a further important property that corroborates the VP shell
analysis. Unlike English let or make, Japanese -(s)ase is a bound morpheme. This fact provides
independent cross-linguistic evidence for treating the silent causative morpheme in English as a
bound morpheme, as we have been doing.

Similarities and differences in case marking between Japanese and German

Like Japanese, double object sentences in German exhibit nominative-dative-accusative case
marking. As in Japanese, marking the recipient as accusative is completely ungrammatical. (As
usual, we give German examples in the form of subordinate clauses in order to avoid the compli-
cations associated with the verb-second phenomenon covered in Chapter 12.)

(88) dass wir { ✔ihnen, *sie } die ganze Geschichte erzählten

that we.nom them.dat them.acc the.acc whole story told
‘that we told them the whole story’

The ungrammaticality of the accusative pronoun in (88) is reminiscent of the double-o con-
straint. But when causation is expressed analytically, as in (89), the case-marking pattern is
reversed.

(89) dass wir { *ihnen, ✔sie } die ganze Geschichte hören liessen
that we.nom them.dat them.acc the.acc whole story hear let.pst
‘that we let them hear the whole story’

We attribute the difference between Japanese and German to the different morphological
status of the overt causative morpheme. In Japanese, -(s)ase is a bound morpheme just like its
silent counterpart. When it combines with the lower verb, the resulting verb is a single word,
Chapter 11: VP shells 342

and apparently the case array associated with single words can’t include the same morphological
case twice. (That’s the double -o constraint.) But German liessen ‘let’ in (89) is a separate word,
and so gets to have its “own” case, just like let does in English.

The passive of double object sentences in German

As discussed earlier in this chapter, modern American English allows the recipient of a double
object sentence to become the subject of a passive sentence, but not the theme. For convenience,
we repeat the relevant facts in (90) and (91).

(90) Active:
We gave them a present.
(91) a. Recipient passive:
✔ They were given a present.
b. Theme passive:
* A present was given them.

In German, the pattern is exactly reversed, as shown in (95) and (96). (The same goes for
Japanese.)

(92) Active:
dass wir d-en Nachbarn ein-en Roman gegeben haben
that we.nom the-dat neighbors a-acc novel given have
‘that we gave the neighbors a novel’
(93) a. Recipient passive:
* dass die Nachbarn ein-en Roman gegeben wurden
that the.nom neighbors a-acc novel given became.3pl
Intended meaning: ‘that the neighbors were given a novel’
b. Theme passive:
✔ dass ein Roman d-en Nachbarn gegeben wurde
that a-nom novel the-dat neighbors given became.3sg
‘that the neighbors were given a novel’

In English, passive removes the ability of the [caus] head to assign objective case. In Ger-
man, the [caus] head assigns dative case when the lower VP shell is headed by a transitive verb.
As we know from Chapter 8, § 3.1 dative is an inherent case, which remains unaffected by pas-
sivization. So what apparently happens in German is that passive removes the case-assigning
ability of the next available case-assigning head—the head of the lower VP shell structure. Rather
than the recipient, it is then the complement of this head—the theme argument—that has to move
to the nearest empty position where it can be assigned case.
Schematic configurations for the active and passive for both languages are shown in (94)
and (95), where the argument that is forced to move is highlighted by a box. The only purpose
of the numerical indices on the two instances of objective case in the English examples in (94)
Chapter 11: VP shells 343

is to make it clear which head assigns case to which DP. The structures for German in (95) are
represented as head-initial for ease of comparison with English.

(94) English:

a. b. VP
VP
V′
DP V′
V◦ VP
◦
agent V VP [caus]
[caus] DP V′
DP V′
+obj-1 ◦
recipient V◦ DP recipient √ V DP
√ give
+obj-1 give theme
theme +obj
+obj-2 +obj
+obj-2

Passive removes [+obj-1]

Highest instance of structural case
([+obj-1]) on higher verb

(95) German:

a. b. VP
VP
V′
DP V′
V◦ VP
◦
agent V VP [caus]
[caus] +dat DP V′
DP V′
+dat
recipient
recipient V◦ DP +dat V◦ DP
√ √
+dat give give
theme
+acc theme
+acc

Passive removes [+acc]

Highest instance of structural case
([+acc]) on lower verb
 Chapter 11: VP shells 344

Notes 7
1. The idea underlying the VP shell analysis goes back to Chomsky (1955) and was taken up in Larson (1988) and
Larson (1990) (see also Jackendoff 1990). The treatment in this chapter is indebted to that in Harley (2002), though
not identical to it in all details.

2. Metonymy is the traditional term for various types of figurative language use, notably including the one relevant
here, where an expression that literally refers to a location is used to refer instead to a group of people typically
located there. Common examples include the White House (broadly ‘the U.S. executive’, more narrowly ‘the U.S.
president along with close staf’), the Kremlin (‘the Russian government’), Westminster (‘the U.K. parliament’), and
so on.

3. The alternation in (i)—specifically, the well-formedness of (b)—is only apparently problematic for what we say in the
text.

(i) a. ✔ Amy sent the mail { back, off }.

b. ✔ Amy sent { back, off } the mail.

Back and off are so-called particles, which can behave like ordinary PPs, as in (a), but also more like bound affixes,
as in (b). A detailed analysis of the syntax of particles is beyond the scope of this chapter, but evidence for their
differing syntactic status in (i) comes from contrasts like (ii).

(ii) a. ✔ Amy sent the mail right { back, off }.

(cf. ✔ right to the dean)
b. * Amy sent right { back, off } the mail.

4. Again, alternations as in (i) are only apparent exceptions to the statement in the text and reflect the status of on and
back as particles; see Note 3.

(i) a. Amy put her sweater (right) { back, on }.

b. Amy put (*right) { back, on } her sweater.

5. Some speakers of British English allow the theme passive in A present was given the children. Theme passives are
possible in languages that distinguish between structural and inherent case, and English was once such a language.
In American English, the theme passive has been entirely replaced by the recipient passive (Bacovcin 2017), whereas
British English is not quite as far along in the replacement process.

6. In addition to marking grammatical relations like subject or direct object, Japanese also marks discourse functions
such as topic. In Japanese main clauses, topic marking with -wa overrides subject marking with -ga. It is therefore
customary to illustrate -ga marking by using subordinate clauses, as we do in what follows.

7. Like French à, Japanese -ni is ambiguous between a true dative marker and a postposition meaning ‘to’, but unlike in
French, there is no way to distinguish the two uses, as the French clitics lui, leur vs. y allow us to do (see Chapter 7,
§ 1.3). Fortunately, though, the exact status of -ni is irrelevant for present purposes since the parallel between
ordinary small clause causatives and VP shell causatives in (83)–(89) holds up even if ni is a postposition.
Chapter 11: VP shells 345

Exercises and problems 8

Exercise 11.1
A. Find five double object or double complement verbs not mentioned in the chapter and
suggest a semantic decomposition for them.
B. Can you find further double complement verbs of the put type?

Exercise 11.2
A. Provide a paraphrase for the two interpretations of the newspaper headline in (1) and the
VP shell structure corresponding to each paraphrase.

(1) Lawyers Give Poor Free Legal Advice

B. Repeat (A) for (2).

(2) I will make you my favorite sandwich.

C. Repeat (A) for the two salient interpretations of the punchline in (3).

(3) Q: What did the Zen master say to the guy at the hotdog stand?
A: Make me one with everything.

Exercise 11.3
Using (1) and (2) as a model, determine whether the verbs in (3) are ECM verbs or object con-
trol verbs (or neither!). Use only active verb forms. For the purpose of this exercise, do not
be concerned about how you would semantically decompose any object control verbs that you
find.
Judgments for assume:

(1) a. ✔ I assumed the meeting to be over.

b. ✔ I assumed there to be a problem.
c. * I assumed the meeting that it was over.
d. ✔ I assumed that there was a problem.

Conclusion for assume:

The grammaticality of (1b) shows that assume is an ECM verb. The judgments in (1c)- (1d)
are not required, but they corroborate the ECM analysis.

Judgments for convince:

Chapter 11: VP shells 346

(2) a. ✔ I convinced my friend to take the job.

b. * I convinced there to be a problem.
c. ✔ I convinced my friend that he should take the job.
d. * I convinced that there was a problem.

Conclusion for convince:

The ungrammaticality of (2b) shows that convince is an object control verb. The judgments
in (2c)- (2d) are not required, but they corroborate the object control analysis.

(3) acknowledge, advise, allow, anticipate, ask, beg, blackmail, challenge, command, con-
sider, corral, dare, deem, determine, discover, encourage, enjoin, expect, fear, find, forbid,
get, help, instruct, invite, know, order, perceive, permit, predict, pressure, prompt, prove,
provoke, remind, report, request, require, tell, tempt, urge, warn

Exercise 11.4
Explain the pattern of interpretations in (1) and (2), or as much of it as possible. Assume the
judgments as given even if they are not your own.
Hint: Find the section in the chapter that discusses differences between give and get.

(1) a. The finger { got, went } to the patient. (unambiguously literal)

b. The surgeon gave the finger to the patient. (unambiguously literal)
c. The surgeon got the finger to the patient. (unambiguously literal)
(2) a. The patient got the finger (from the surgeon). (ambiguous between literal and id-
iomatic reading)
b. The surgeon gave the patient the finger. (ambiguous between literal and idiomatic
reading)
c. The surgeon got the patient the finger. (unambiguously literal)

Exercise 11.5
Is the example in (1) expected or not? Explain.

(1) Tennessee coach Pat Summitt, the longtime head of the Lady Vols, gives an earful to Alexis
Hornbuckle during their win over Texas Tech.
(Daily Pennsylvanian, 28 March 2005, p. 9)
Chapter 11: VP shells 347

Exercise 11.6
A. Get is one of the most versatile verbs in English. In addition to taking DP and PP/AdvP com-
plements, it also takes AdjP complements. Give the VP shell structures for the sentences
in (1).

(1) a. The children got wet.

b. The rain got the children wet.

B. Give VP shell structures for the sentences in (2). As much as possible, avoid assuming silent
heads.

(2) a. The red cloth enraged the bull.

b. The new fiscal policies enriched the king.
c. The news saddened me.

C. Build VP shell structures for the sentences in (3).

(3) a. They kept the president’s arrival almost completely secret.

b. They kept the president’s arrival a complete secret.

Exercise 11.7
Propose structures for each of the following expressions, focusing on the parallels between the
‘die’ and ‘kill’ expressions. Assume that German and Latin are head-final.

(1)
‘die’ ‘kill’
German um- kommen um- bringen
around come around bring
Latin inter- ire inter- facere
between go between make
Latin per- ire per- dere
through go through give

Exercise 11.8
Modern English has causative verbs that are morphologically related to either other verbs, as in
(1a), or to other syntactic categories, as in (1b).

(1) a. drink, fall, lie (as in ‘lie down’, not ‘prevaricate’), rise, sit
b. full, gold

The causatives are not spelled out as homonymous with the same form, but at one point in history
the two forms were related by a regular phonological rule (specifically, ablaut or umlaut—see the
Chapter 11: VP shells 348

Wikipedia entries for these terms for more information). But historical sound changes has since
obscured those regularities. Over time, the meaning of the originally causative form has also
sometimes drifted away from a strict causative meaning. Can you figure out the causative verbs
in question?
Hint: Consult etymonline (see a word’s “Related entries”) or the Oxford English Dictionary On-
line.

Problem 11.1
Can you think of double object or double complement verbs that are not amenable to the causative
decomposition proposed in the chapter?

Problem 11.2
Explain the pattern of acceptability judgments in (1)–(3).

(1) a. ✔ They gave the people some money.

b. ✔ They gave some money. (recipient unexpressed)
c. * They gave the people.
(theme unexpressed; ✔ on unintended reading, where the people is theme)
(2) a. ✔ They gave some money to the people.
b. ✔ They gave to the people. (theme unexpressed)
c. * They gave some money to. (recipient unexpressed)
(3) a. ✔ They gave. (recipient and theme unexpressed, as in I already gave at the office).
b. * They gave to. (recipient and theme unexpressed)

Problem 11.3
As mentioned in the main text, it is possible to give at least some transitive verbs like build
structures that correspond to the paraphrase ‘cause to come to be built’. Is this decomposition
possible for all transitive verbs?

Problem 11.4
A. Sentences like (1) are ungrammatical in adult English.

(1) You’re giggling me!

Intended meaning: ‘You’re making me giggle!’

But many children go through a stage of producing such sentences with activity verbs such
as dance, giggle, laugh, and others. What is the difference between the children’s grammar
and the adult grammar?
B. If necessary, modify your answer to (A) in light of the judgments of (2)–(4) for adult English.
Chapter 11: VP shells 349

(2) a. The horses { walked, trotted, galloped, jumped }.

b. The trainers { walked, trotted, galloped, jumped } the horses.
(3) a. The athletes { walked, ran, marched } around the track.
b. The coach { walked, ran, marched } the athletes around the track.
(4) a. * At some point, all new parents want to sleep their baby.
b. ✔ I have one set to show the desktop, and another to sleep the display. … Hit
Command-Option-H to Hide Others. This works in almost any app, letting
you quickly disappear all apps other than the frontmost one.
c. ✔ They were known as the ‘desaparecidos’, because the Argentinian junta
would ‘disappear’ them—for instance, by throwing them out of planes over
the Andes.

Problem 11.5
Can you think of a reason for why (1b) lacks a ‘fetch’ interpretation?

(1) a. They got the package.

Interpretation 1: ✔ They received the package.
Interpretation 2: ✔ They fetched the package.
b. We got them the package.
Interpretation 1: ✔ We made it come about that they received the package.
Interpretation 2: * We made it come about that they fetched the package.

Problem 11.6
Unlike get in (1), verbs of transfer like hand, kick, throw, and others don’t participate in the
causative alternation, as shown in (2). Can you think of a principled reason?

(1) a. ✔ The package got to Julie.

b. ✔ We got the package to Julie.
(2) a. * The ball { handed, kicked, threw, … } to Julie.
b. ✔ We { handed, kicked, threw, … } the ball to Julie.
The verb-second (V2)
12
phenomenon

Chapter outline
1 Remnants of V2 in modern English . . . . . . . . . . . . . . . . . . . . . . . . . . 351
1.1 Negative inversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351
1.2 So inversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352
2 V2 in Middle English . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353
3 V2 in German . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355
4 The loss of V2 in English . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 359
5 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 360
6 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 360

By historical descent, English belongs to a language family called Germanic, which at one
time had three branches: East Germanic, North Germanic, and West Germanic. East Germanic
(Gothic) became extinct in the 1600s, but the other two branches are still extant today. Over time,
the North Germanic branch developed into the modern Scandinavian languages (Danish, Faroese,
Icelandic, Norwegian, and Swedish), and the West Germanic branch developed into Afrikaans,
Dutch, English, Frisian, German, and Yiddish. In this chapter, we discuss a noteworthy syntactic
property shared by all of the Germanic languages, except for modern English. In these languages,
the finite verb in a declarative main clause comes in second position, regardless of whether the
first phrasal constituent is the subject. For this reason, the word order phenomenon is called
verb-second, or V2 (read: vee two) for short. Another way of putting this is to say that ordinary
declarative main clauses in V2 languages have the same syntactic structure as direct questions
do in English.1
Modern English still has some remnants of V2, and we begin our discussion with them.
We then illustrate V2 in Middle English and German, where the phenomenon is productive. The
chapter concludes with a brief discussion of the loss of V2 in the history of English.
Chapter 12: The verb-second (V2) phenomenon 351

Remnants of V2 in modern English 1

Recall from Chapter 6 that direct and indirect wh- questions are formed in English by moving
a wh- phrase from the position where it is interpreted to Spec(CP), where it is pronounced. In
direct wh- questions, this movement is obligatorily accompanied by I moving to C. Recall further
that English exhibits the related phenomenon of topicalization, where it is a non wh- phrase
that moves to Spec(CP), without accompanying head movement. Finally, there is a fourth pos-
sible combination, where movement of a non wh- phrase occurs together with head movement.
We have already briefly encountered this possibility in Chapter 10, § 6 under the name of neg-
ative inversion. As illustrated in the next section, inversion is also possible with so. The four
possibilities are summarized in Table 12.1.

XP moves to Spec(CP)
I moves to C Wh- Non-wh-
Yes Direct wh- question Negative inversion
So inversion
No Indirect wh- question Topicalization

Table 12.1: Possible syntactic outcomes of movement to Spec(CP)

1.1 Negative inversion

Each pair in (1)–(4) shows a direct wh- question along with a corresponding example of negative
inversion.

(1) a. What would they accept more willingly?

b. No present would they accept more willingly.
(2) a. Where can you see the Southern Cross?
b. Nowhere in the northern hemisphere can you see the Southern Cross.
(3) a. When is that door left unattended?
b. { Never, At no time } is that door left unattended.
(4) a. Under what circumstances would they agree to a settlement?
b. Under no circumstances would they agree to a settlement.

When negative phrases like those in (1)–(4) are specified more fully with but or except phrases,
they can be paraphrased by only phrases, and these too allow inversion.

(5) a. { Never except on Sundays, Only on Sundays } do we see them.

Chapter 12: The verb-second (V2) phenomenon 352

b. { No-one but computational linguists, Only computational linguists } would we consider

for that position.
c. { Nowhere except in my neighborhood, Only in my neighborhood } are the rents so low.

1.2 So inversion
Another remnant of V2 is a variant of the substitution test for adjective phrases mentioned
in Chapter 5, § 2. The original test and its V2 variant are illustrated in (6a) and (6b), respec-
tively.

(6) a. Ordinary so substitution: We are happy, and they are so, too.
b. So substitution with inversion: We are happy, and so are they.

We show schematic structures for V2 and topicalization sentences in English in (7). These
structures correspond to the righthand column in Table 12.1. As is standard in the syntax lit-
erature, we use the term ‘topic’ for the clause-initial phrase in both cases. The [v2] head in C
“attracts” the contents of I, whereas [top] doesn’t. Therefore, (7a) is identical to the structure
for a direct question, except that Spec(CP) would be filled by a wh- phrase, and C would be the
[q] morpheme corresponding to Japanese ka. (7b) is identical to the structure for an indirect
question, except that Spec(CP) would again be filled by a wh- phrase, and C would be [that] (and
optionally overt in Middle English).

(7) a. CP

XPi C′

topic C◦ IP

I◦m C◦ DPk I′
[tense] [v2]
subject t m VP

tk V′

verb … t i …

V2 — I moves to C
Chapter 12: The verb-second (V2) phenomenon 353

b. CP

XPi C′

topic C◦ IP
[top]
DPk I′

subject I◦ VP
[tense]
tk V′

verb … t i …

Topicalization — I remains in situ

V2 in Middle English 2
(8) gives some examples of Middle English V2 clauses from The Travels of Sir John Mandeville,
a bestselling travel book from the 1300s; the examples are taken from Mossé (1968, p. 279), a
convenient compendium of Middle English texts.

Extra Info
As in previous chapters, eth (capital Ð, lowercase ð) and thorn (capital Þ, lowercase þ)
appear where we use ‘th’ today. Yogh (ȝ) was used where we use ‘g’ or ‘y’.

Notice the two instances of overt that in (8b), where modern English disallows them.

(8) a. þanne wolde he make hem to drynken of a certeyn drynk.

‘Then he would make them drink of a certain drink.’
b. After that ȝit scholde he putten hem in a fayrere paradys where þat þei scholde see
God.
‘Yet after that he would put them in a more beautiful paradise where they would see
God.’
c. þan wolde he schewe hem his entent.
‘Then he would show them his intent.’
d. þere scholde þei dwellen with the most fairest damyselles þat myghte be, and pley
with hem everemore.
Chapter 12: The verb-second (V2) phenomenon 354

‘There they would dwell with the fairest damsels that there might be and play with
them for ever more.’

(9) gives some further examples from the Canterbury Tales of Geoffrey Chaucer (ca. 1340–ca.
1400), written near the end of his life.

(9) a. Wel koude he rede a lessoun or a storie. (Mossé 1968: 303, 709)
‘He could read a lesson or a story well.’
b. This tresor hath Fortune unto us yeven. (Mossé 1968: 307, 779)
‘This treasure, Fortune has given us.’
c. Thy profit wol I telle thee anon. (Mossé 1968: 308, 809)
‘I will tell you right away what’s in it for you.’
d. And thanne shal al this gold departed be, my deere frend, bitwixen me and thee.
‘And then all this gold will be divided up, my dear friend, between me and you.’
(Mossé 1968: 309, 831–832)
e. Thanne may we bothe oure lustes all fulfille. (Mossé 1968: 309, 833)
‘Then both of us can fulfill all our desires.’
f. Into the blisse of hevene shul ye gon. (Mossé 1968: 311, 912)
‘You will enter the bliss of heaven.’

In the above examples, the finite head in second position is a modal. But since Middle
English exhibited verb movement to I, ordinary finite verbs can end up in C as well, as shown in
(10).

(10) [T]his contrey know I well. (PPCME2, cmmalory-m4,181.2463)

‘I know this country well.’

Notice that in such sentences, it is not possible to tell if the verb moves from a head-initial or
head-final position (recall that Middle English allowed both verb-initial and verb-final projec-
tions).
Schematic structures for V2 sentences in Middle English with finite main verbs are shown
in (11), assuming head-initial VP. Because Middle English has verb movement to I, the complex
head in C resulting from verb movement has one more layer than it does in the vestigial V2
sentences in modern English, so we show the structures both before and after verb movement for
clarity. Direct questions in Middle English have identical structures, except that what moves to
Spec(CP) would be a wh- phrase and C would be [q].
Chapter 12: The verb-second (V2) phenomenon 355

(11) a. CP
XPi C′
topic C◦ IP
[v2]
DPk I′
subject I◦ VP
[tense]
tk V′
verb … t i …

Middle English V2 before verb

movement

b. CP
XPi C′
topic C◦ IP
I◦m C◦ DPj I′
[v2]
V◦k I◦ subject t m VP
verb [tense]
tj V′
tk … ti …

After movement of V to I to C

V2 in German 3
Even more striking than in Middle English is the V2 phenomenon in German and Dutch. As
already mentioned in Chapter 5, § 4, these languages have head-initial C but head-final V. These
phrase structure parameters are straightforwardly apparent in German sentences like (12). For
simplicity, we assume that I is head-final as well.

(12) a. ✔ dass die Kinder dieses Buch lesen werden

that the children this book read will
‘that the children will read this book’
b. * dass die Kinder werden dieses Buch lesen
Chapter 12: The verb-second (V2) phenomenon 356

c. * dass die Kinder werden lesen dieses Buch

Matrix declarative clauses, however, exhibit the V2 order illustrated in (13). The finite head
occupies the second position in the clause; any nonfinite verbs continue to be head-final.

(13) a. * Die Kinder dieses Buch lesen werden.

b. ✔ Die Kinder werden dieses Buch lesen.
c. ✔ Dieses Buch werden die Kinder lesen.
d. ✔ { Morgen, Im Wohnzimmer } werden die Kinder dieses Buch lesen.
tomorrow in.the living.room will the children this book read
‘The children will read this book { tomorrow, in the living room }.’

Once again, in sentences with finite main verbs, it is that verb (the only one in the clause) that
moves, and again, in such sentences, V2 movement obscures the verb-final character of the lan-
guage.

(14) * Die Kinder dieses Buch lesen.

the children this book read
Intended meaning: ‘The children are reading this book.’
(15) a. ✔ Die Kinder lesen dieses Buch.
b. ✔ Dieses Buch lesen die Kinder.
c. ✔ { Morgen, Im Wohnzimmer } lesen die Kinder dieses Buch.

A further bit of word order variation in German provides strong support for the idea that
finite heads move to C in V2 sentences. As shown in (16) and (17), object pronouns in German
regularly move from their expected position to a position right after the complementizer.

(16) ✔ dass die Kinder ihren Eltern das Buch zeigen werden
that the children their parents the book show will
‘that the children will show their parents the book’
(17) a. ✔ dass es die Kinder ihren Eltern zeigen werden
that it the children their parents show will
‘that the children will show it to their parents’
b. ✔ dass ihnen die Kinder das Buch zeigen werden
that them the children the book show will
‘that the children will show them the book’

In V2 sentences, these same pronouns appear right after the fronted verb.

(18) a. ✔ Die Kinder werden es ihren Eltern zeigen.

the children will it their parents show
‘The children will show it to their parents.’
b. ✔ Ihren Eltern werden es die Kinder zeigen.
their parents will it the children show
‘The children will show it to their parents.’
Chapter 12: The verb-second (V2) phenomenon 357

c. ✔ { Morgen, im Wohnzimmer } werden es die Kinder ihren Eltern zeigen.

tomorrow in.the living.room will it the children their parents show
‘The children will show it to their parents { tomorrow, in the living room }.’
(19) a. ✔ Die Kinder werden ihnen das Buch zeigen.
the children will them the book show
‘The children will show them the book.’
b. ✔ Das Buch werden ihnen die Kinder zeigen.
the book will them the children show
‘The children will show them the book.’
c. ✔ { Morgen, im Wohnzimmer } werden ihnen die Kinder das Buch zeigen.
tomorrow in.the living.room will them the children the book show
‘The children will show them the book { tomorrow, in the living room }.’

These word order facts are parallel to the ones for wh- questions in German, as expected if the
verb moves to C in both sentence types.

(20) a. ✔ Wer wird es den Eltern zeigen?

who will it the parents show
‘Who will show it to the parents?’
b. ✔ Wem werden es die Kinder zeigen?
who will it the children show
‘Who will the children show it to?’
c. ✔ { Wann, Wo } werden es die Kinder ihren Eltern zeigen?
when where will it the children their parents show
‘{ When, where } will the children show it to their parents?’
(21) a. ✔ Wer wird ihnen das Buch zeigen?
who will them the book show
‘Who will show them the book?’
b. ✔ Was werden ihnen die Kinder zeigen?
what will them the children show
‘What will the children show them?’
c. ✔ { Wann, Wo } werden ihnen die Kinder das Buch zeigen?
when where will them the children the book show
‘{ When, where } will the children show them the book?’

The schematic structures for German V2 sentences before and after movement of a finite
main verb are shown in (22). They are analogous to those for Middle English, except that the V
and I projections are head-final. Again, the structures for direct questions are identical, with the
same provisos noted for Middle English (that is, Spec(CP) and C would be filled by a wh- phrase
and [q], respectively).
Chapter 12: The verb-second (V2) phenomenon 358

(22) a. CP

XPi C′

topic C◦ IP
[v2]
DPk I′

subject VP I◦
[tense]
tk V′

… t i … verb

German V2 before verb movement

b. CP

XPi C′

topic C◦ IP

I◦m C◦ DPj I′
[v2]
V◦k I◦ subject VP tm
verb [tense]
tj V′

… ti … tk

After movement of V to I to C

In conclusion, it is worth noting that German allows Spec(CP) to be filled not only by
movement, but also by direct substitution of expletive es ‘it’. This yields sentences like (23).

(23) ✔ Es werden einige Kinder das Buch lesen.

it will some children the book read
‘Some children will read the book.’

Thus, in addition to the subject requirement that it shares with English, German (and V2 lan-
guages more generally) have an additional topic requirement. In both cases, the requirement is
purely syntactic. In other words, if the requirement is not satisfied by a semantically contentful
element, it must be satisfied by an expletive element (it or there in the case of the English sub-
ject requirement, and es ‘it’ in the case of the German topic requirement). The purely syntactic
Chapter 12: The verb-second (V2) phenomenon 359

character of the topic requirement is apparent from the ungrammaticality of variants of (23) like
(24).

(24) a. * Das Buch werden es einige Kinder lesen.

the book will it some children read
Intended meaning: ‘Some children will read the book.’
b. * dass es einige Kinder das Buch lesen werden
that it some children the book read will
Intended meaning: ‘that some children will read the book’

The loss of V2 in English 4

V2 in Old and Middle English was a more complex phenomenon than in any of the other V2
languages. In particular, there is evidence that Old English (700–1100) exhibited not just one,
but two types of V2. The first type invariably involved verb movement to C, as just discussed,
and was found in direct questions, imperatives, certain clauses with negative force, and clauses
introduced by certain adverbs (notably, þa ‘then’). The second type of V2 was structurally distinct
from the first type (it involved verb movement to a head lower than C), and it was also superficially
less transparent. In particular, in topic-initial sentences, subjects followed the finite verb if they
were full noun phrases, as in the first type. But unstressed pronoun subjects preceded the verb,
resulting in verb-third (V3) word orders, as summarized in Table 12.2.

Full DP subject Pronoun subject

V2 (topic verb subject …) ✔ *
V3 (topic subject verb …) * ✔

Table 12.2: Distribution of V2 and V3 word orders in Old English

Modern English still has a relic of this system in examples like (25) and (26).2

(25) a. ✔ Here comes the bus; there goes the bus.

b. * Here comes it; there goes it.
(26) a. ? Here the bus comes; there the bus goes.
b. ✔ Here it comes; there it goes.

In the south of England, this complicated V2 type continued to hold sway into Middle
English times. Beginning in Old English, though, the north of England was repeatedly attacked
and invaded by Vikings, who eventually settled there and mixed with the Old English-speaking
 Chapter 12: The verb-second (V2) phenomenon 360

population. The Vikings spoke Old Norse (the ancestor of all of the Scandinavian languages),
which exhibited the simpler type of V2, and this simpler type spread in the north and became
prevalent, if not dominant.
In the course of Middle English, speakers of the two dialects came into increasing contact
with one another, and it has been proposed that northern speakers moving to the south or children
growing up in a linguistically mixed environment misanalyzed sentences like (26b) as non-V2
(Kroch, Taylor, and Ringe 2000). Such speakers would then have begun to produce sentences like
(26a). It is this linguistic innovation—previously ungrammatical in both dialects—that eventually
drove out both types of V2 in declarative clauses.

Notes 5
1. The oldest Germanic language for which we have more than very short inscriptions is Gothic, an East Germanic
language, which is attested in a Bible translation from the 4th century ce and which shows no signs of V2. The
other Germanic languages begin to be set down in writing a few centuries later, and they already exhibit at least the
first signs of V2, so V2 must have developed in the interim. Bizarre though the word order may seem to speakers
of modern English and modern varieties of Romance, V2 spread from Germanic to the medieval versions of French,
Italian, Spanish, and Portuguese. The way this would have happened is by large numbers of native speakers of a
V2 Germanic language learning a non-V2 Romance language imperfectly as adults (for instance, Franks, Lombards,
and Vandals learning the precursors of French, Italian, and Spanish, respectively). These adult learners would have
produced Romance V2 sentences in error, but if there were enough of them in the population, children growing up
in such an environment would have acquired the V2 order natively. In the course of their further history, however,
the major Romance languages have all lost V2. It is still found in Sursilvan, a Romance dialect spoken in the Grisons,
a canton in southeastern Switzerland bordering on Italy.

2. A detailed analysis of this second type of V2 is beyond the scope of this introductory textbook; for discussion, see
Ringe and Taylor (2014, chapter 8).

Exercises and problems 6

Exercise 12.1
The sentences in (1) are superficially analogous, but their structures differ. Provide the struc-
tures.

(1) a. English:
Chapter 12: The verb-second (V2) phenomenon 361

✔ The children will play.

b. German:
✔ Die Kinder werden spielen.
the children will play
‘The children will play.’

Exercise 12.2
A. Build structures for the modern English V2 clauses in (1), including full structures for the
topics.

(1) a. Never in their lives have they seen such orchids.

b. Under no circumstances will they agree.

B. Build structures for the topics in (2). Treat only as an adverbial modifier.

(2) a. Only in Brazil have they seen such orchids.

b. Only under those circumstances will they agree.

Exercise 12.3
Build structures for the Middle English sentences in (1) (adapted from the examples in (9)). For
simplicity, use triangle notation for DPs, PPs and AdvPs.

(1) a. A vernycle hadde he sowed upon his cappe. (Mossé 1968: 303, 685)
‘He had sewn a vernicle on his cap.’
b. Wel koude he rede a storie. (Mossé 1968: 303, 709)
‘He could read a story well.’
c. This tresor hath Fortune unto us yeven. (Mossé 1968: 307, 779)
‘Fortune has given this treasure to us.’

Exercise 12.4
The Middle English sentence in (1) was structurally ambiguous between the two interpretations
in (2). Explain.

(1) Thys rememberede thei nevyr to doo.

(2) a. They never remembered to do this.
b. They remembered never to do this.

Extra Info
For simplicity, assume that never modifies the same syntactic category in both interpre-
tations. (Hint: see Chapter 10, § 4.2.)
Chapter 12: The verb-second (V2) phenomenon 362

Exercise 12.5
A. Build the structure for the German clause in (1). For simplicity, use triangle notation for
DPs, PPs, and AdvPs, and treat hat as the head of IP.

(1) Mein Onkel hat bei seinen Nachbarn dreimal aus der Schweiz angerufen.
my uncle has at his neighbors three.times from the Switzerland called
‘My uncle called his neighbors three times from Switzerland.’

B. Given the discussion in the chapter, which (if any) of the sentences in (2) are expected to
be grammatical? Explain.

(2) a. Angerufen hat mein Onkel aus der Schweiz dreimal.

b. Aus der Schweiz hat mein Onkel dreimal angerufen.
c. Dreimal aus der Schweiz hat mein Onkel bei seinen Nachbarn angerufen.

Problem 12.1
A. Assume that the Middle English sentence in (1) is a V2 sentence. What does it mean? Is it
structurally ambiguous? (That is, who is greeting whom?) Does it matter whether the V′
is head-initial or head-final?

(1) The king greeted the queen.

B. Repeat (A) for the Middle English sentence in (2).

(2) The king will the queen greet.

C. Repeat (A) for the Middle English sentence in (3). Assume that VP shells are consistently
head-initial or consistently head-final.

(3) The queen was the palfrey given.

 13
Constraints on wh-
movement

Chapter outline
1 Ross’s island constraints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 365
2 Subjacency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367
2.1 Two possible derivations for long-distance wh- movement . . . . . . . . . 367
2.2 IP as a barrier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367
2.3 DP as a barrier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 369
3 Further issues and considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . 369
3.1 The status of nonfinite IP . . . . . . . . . . . . . . . . . . . . . . . . . . . 369
3.2 The coordinate structure constraint revisited . . . . . . . . . . . . . . . . 370
3.3 The status of DP as a barrier revisited . . . . . . . . . . . . . . . . . . . . 371
3.4 Argument-adjunct asymmetries . . . . . . . . . . . . . . . . . . . . . . . . 372
3.5 The Comp-trace effect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373
4 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374
5 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375

In Chapter 6, we argued that wh- phrases in questions, relative clauses, and various other
constructions move from their canonical position (the position where they are interpreted) to
Spec(CP). In that chapter, the source IP and the target CP for wh- movement were neighbors, so
to speak, but examples like (1) suggest that the structural distance that a wh-phrase can move is
unlimited, often called unbounded (that is, unbounded apart from performance considerations
such as limitations on memory). Movement in (1a) is called local movement, as distinct from
long-distance movement in (1b)–(1e) (also known as long or nonlocal movement).

(1) a. [CP Whati were they reading t i ] ?

b. [CP Whati did he say
[CP that they were reading t i ] ] ?
c. [CP Whati does she believe
[CP that he said
[CP that they were reading t i ] ] ] ?
d. [CP Whati are they claiming
Chapter 13: Constraints on wh- movement 364

[CP that she believes

[CP that he said
[CP that they were reading t i ] ] ] ] ?
e. [CP Whati do you think
[CP that they are claiming
[CP that she believes
[CP that he said
[CP that they were reading t i ] ] ] ] ] ?

Surprisingly, however, even very short-distance movement can be stunningly ungrammat-

ical, as shown in (2).

(2) a. They read Jonathan’s book.

b. * Whosei did they read t i book? (cf. ✔ Whose book did they read?)

Accordingly, beginning in the 1960s, attempts were made to understand the conditions under
which wh- movement is and isn’t possible. Ross (1967) catalogued various contexts where wh-
movement isn’t possible, which he gave the descriptive blanket term islands (the idea behind
the metaphor being that a phrase is marooned—it can move within the island, but not escape off
of it). Ross’s work represents significant empirical progress over what was known at the time
about wh- movement, but from a theoretical point of view, islands are simply a list of syntactic
contexts, sharply raising the question of whether it is possible to characterize them and to derive
the constraints on movement associated with them in a principled way. With this aim in mind,
Chomsky (1973) proposed the so-called subjacency condition, a relatively simple, purely struc-
tural condition on wh- movement. Along with Ross’s original work and later additions to the list
of islands, Chomsky’s proposal has given rise to a vast literature with two main wings. One wing
extends Chomsky’s guiding idea that constraints on wh- movement are structural, whereas the
other wing attempts to derive the same constraints from other sources, notably pragmatics and
psycholinguistic processing.
This chapter is not intended to summarize the decades of work since Ross’s and Chomsky’s
pioneering work. Rather, we illustrate Ross’s island constraints and present the subjacency con-
dition as formulated in Chomsky (1973), along with a very few considerations motivating further
work. An excellent survey of the current state of research is Liu et al. (2022), which should be
accessible after reading this chapter.
Chapter 13: Constraints on wh- movement 365

Ross’s island constraints 1

As mentioned in the introduction to the chapter, Ross (1967) discovered a number of contexts out
of which wh- movement is unacceptable:
• noun complement clauses,
• relative clauses,1
• sentential subjects,
• indirect questions,
• possessive noun phrases, and
• coordinate structures.
We begin by illustrating the island character of noun complement clauses. First, recall
from (1b)–(1e) that long wh- movement out of that complement clauses to verbs is grammati-
cal. By contrast, wh- movement out of that complements to nouns is not, as shown in (3) and
(4). For clarity, in the next few examples, the heads associated with the complement clauses are
underlined.

Extra Info
In the rest of this section and for much of the remaining chapter, we indicate unaccept-
ability with asterisks, in accordance with Ross’s and Chomsky’s assumption that these
sentences violate structural constraints. But when discussing approaches that attempt to
derive such examples from extragrammatical considerations, we use ‘#’ instead.

Noun complement:
(3) a. They made the claim [ that they know the mayor ].
b. * Whoi did they make the claim [ that they know t i ]?

(4) a. They mentioned the fact [ that they had met the mayor ].
b. * [ Which public figure ]i did they mention the fact [ that they had met t i ]?

Particularly striking is the contrast between (3b) and (4b) on the one hand and the essentially
synonymous examples in (5) on the other.

(5) a. ✔ Whoi did they claim [ that they know t i ]?

b. ✔ [ Which public figure ]i did they mention [ that they had met t i ]?
Chapter 13: Constraints on wh- movement 366

The remaining examples in this section illustrate the other islands.

Relative clause:
(6) a. They met someone [ who knows the president ].
b. * [ Which public figure ]i did they meet someone [ who knows t i ]?

Sentential subject:
(7) a. That they personally know the president seems improbable.
b. * Whoi does [ that they personally know t i seem improbable? ]

(8) illustrates the island character of indirect questions. In connection with (8b), it is im-
portant to distinguish the two instances of wh- movement: that of which problem and that of how.
Which problem moves from its original position as complement of solve to the Spec(CP) of the
complement clause. It is this movement—grammatical on its own—that creates an island for any
further wh- movement, preventing how from moving “off island” to the Spec(CP) of the matrix
clause.

Word of Caution
Be sure to interpret How in (8b) as modifying the complement verb solve, as indicated by
the trace, not the matrix verb forgotten. In other words, a possible answer to (8b) is by
Fourier analysis, but not by succumbing to Alzheimer’s.

Indirect question:
(8) a. They have forgotten which problem they should solve by Fourier analysis.
b. * Howi have they forgotten which problem they should solve?

Finally, (9) (= (2)) and (10) illustrate the island character of possessive noun phrases and of
coordinate structures.

Possessive noun phrase:

(9) a. They read Jonathan’s book.
b. * Whosei did they read t i book?

(10) a. They ordered tiramisu and espresso.

b. * [ Which dessert ]i did they order t i and espresso?

c. * [ Which beverage ]i did they order tiramisu and t i ?

Chapter 13: Constraints on wh- movement 367

Subjacency 2

2.1 Two possible derivations for long-distance wh- movement

When we consider examples of long-distance wh- movement like (1b), repeated in (11), two pos-
sible derivations come to mind.

(11) What did he say that they were reading?

On the one hand, the wh- phrase might move in one fell swoop from the position where it is
interpreted, however deeply embedded that is, to the sentence-initial Spec(CP) position, yielding
a wh- movement chain with two links, as in (12).

(12) [CP Whati [IP did he say [CP t i that [IP they were reading t i ] ] ] ] ?

On the other hand, wh- movement might take place in several steps. The first step takes the
moved constituent from its original position to the nearest Spec(CP), and each subsequent step
takes it to the next higher Spec(CP). This derivation of (11), which involves two steps and yields
a chain with three links, is shown in (13).

(13) [CP Whati [IP did he say [CP t i that [IP they were reading t i ] ] ] ] ?

Derivations as in (13) are known as cyclic (the idea being that each successively higher CP
forms a separate cycle in the derivation of the entire sentence). Cyclic derivations decompose
apparently nonlocal movement into sequences of local movement. Derivations as in (12), which
do not decompose nonlocal movement, are accordingly known as noncyclic.

2.2 IP as a barrier
Given only grammatical instances of long-distance wh- movement like (1b)–(1e), repeated here
as (14), it is impossible to decide which of the two alternatives just presented is correct, or even
whether a choice must be made between them. (The parentheses indicate intermediate traces
that are posited in a cyclic, but not in a noncyclic, derivation.)

(14) a. [CP Whati did he say [CP (t i ) that they were reading t i ] ] ?
b. [CP Whati does she believe [CP (t i ) that he said [CP (t i ) that they were reading t i ] ] ] ?
c. [CP Whati are they claiming [CP (t i ) that she believes [CP (t i ) that he said [CP (t i ) that
they were reading t i ] ] ] ] ?
d. [CP Whati do you think [CP (t i ) that they are claiming [CP (t i ) that she believes [CP
(t i ) that he said [CP (t i ) that they were reading t i ] ] ] ] ] ?
Chapter 13: Constraints on wh- movement 368

However, the existence of syntactic islands forces us to choose the cyclic alternative. For
instance, consider the derivation of the ungrammatical question in (8b), repeated as (15). (Remem-
ber to interpret how in (15) as modifying solve, not forgotten, as indicated by the trace.)

(15) * Howi have they forgotten [CP which problem they should solve t i ] ?

If wh- movement were able to occur in one fell swoop, there would be nothing to stop long-
distance wh- movement in (15), and the question should be grammatical, contrary to fact. But
the ungrammaticality of the question can be made to follow if we assume that wh- movement is
cyclic. Specifically, let’s assume, following Chomsky (1973), that wh- movement is subject to the
condition in (16), and that IPs form barriers to movement.

(16) Subjacency condition:

In a chain formed by movement, the path connecting two neighboring links must not
contain more than one barrier (in other words, on the path bewteen A and B, there is at
most one barrier C such that A c-commands C and C dominates B).

Given (16), a wh- phrase can move out of an IP that dominates it just in case an empty
local Spec(CP) is available as an intermediate landing site. We allow CPs to “sprout” specifiers
as needed; more precisely, we allow complementizers to have elementary trees with specifiers as
needed, as shown for that in (17).

(17) CP

»XP« C′

C◦ »IP«
that

In the absence of an intermediate landing site, wh- movement is ruled out as ungrammati-
cal. The barriers relevant in the derivation of (15) are shown in (18).2

(18) * [CP Howi have [IP they forgotten [CP which problem [IP they should solve t i ] ] ] ] ?
* ✔

Extra Info
In determining whether subjacency is violated, we can consider the relevant movement
chains in either top-down or bottom-up fashion. In what follows, we choose bottom-up. In
other words, we follow the wh- phrase as it attempts to move from its canonical position to
the target Spec(CP). The lowest barrier in each chain is indicated by a check mark, because
movement past it is consistent with subjacency; subsequent higher barriers are indicated
by asterisks.
Chapter 13: Constraints on wh- movement 369

2.3 DP as a barrier
Given the structural parallels between indirect questions and relative clauses, the island character
of relative clauses illustrated in (6) follows directly. However, under the assumption that only IP
counts as a barrier, subjacency fails to rule out the other island types. (19) shows the relevant
derivations, to be revised, according to which the examples should be grammatical. (In (19b), the
CP brackets the sentential subject.)

(19) Derivations to be revised!

a. * Whoi did [IP they make the claim [CP t i that [IP they know t i ] ] ] ?
✔ ✔
b. * Whoi does [IP [CP t i that [IP they know t i personally ] ] seems improbable ] ?
✔ ✔
c. * Whosei did [IP they buy t i book ] ?
✔
Chomsky therefore proposed that the set of barriers includes not only IPs, but DPs as well.
In that case, the examples are all correctly ruled out, as shown in (20).

(20) Correct Derivations

a. * Whoi did [IP they make [DP the claim [CP t i that [IP they know t i ] ] ] ] ?
* ✔ ✔
b. * Whoi does [IP [DP [CP t i that [IP they know t i personally ] ] ] seems improbable ] ?
* ✔ ✔
c. * Whosei did [IP they buy [DP t i book ] ] ?
* ✔
Notice that in (20b), we need to assume that the sentential subject is dominated by DP. It is not
exactly clear how the relationship between DP and CP can be made consistent with the X′ schema,
and this could be taken as a weakness of the subjacency approach.

Further issues and considerations 3

3.1 The status of nonfinite IP

In addition to allowing finite complements, forget is a subject control verb. As expected, replacing
the finite IP in (15) by its nonfinite counterpart has no effect on grammaticality, as shown in
(21).
Chapter 13: Constraints on wh- movement 370

(21) * [CP Howi have [IP they forgotten [CP which problem [IP PRO to solve t i ] ] ] ] ?
* ✔
But if nonfinite IPs are barriers just like finite ones, the questions in (22) should to be ungram-
matical too.

(22) a. ✔ [CP Whoi are [IP they likely [IP to know t i already ] ] ] ?

b. ✔ [CP Howi do [IP they expect [IP to solve the problem t i ] ] ] ?

The grammaticality of these questions forces us to conclude that ECM IPs (that is, sisters of V or
Adj) do not count as barriers.

3.2 The coordinate structure constraint revisited

Only a single IP barrier intervenes between the moved phrase in Spec(CP) and its trace in both
(23) and (24), yet wh- movement in (23b), (23c), (24b), and (24c) is completely unacceptable. (Take
care to read (23b) and (24b) without an intonation break before the conjunction.)

(23) a. [AdvP When and where ]i were [IP you born [AdvP t i ] ] ?

b. * [AdvP When ]i were [IP you born [AdvP t i and where ] ] ?

c. * [AdvP Where ]i were [IP you born [AdvP when and t i ] ] ?

(24) They say they will get out the vote and win the election, and …
a. [VP get out the vote and win the election ]i , [IP they will [VP t i ] ] .

b. * [VP get out the vote ]i , [IP they will [VP t i and win the election ] ] .

c. * [VP win the election ]i , [IP they will [VP get out the vote and t i ] ] .

A further example of this sort is shown in (25b).

(25) a. They have [VP peeled the cucumbers and chopped up the onions ] .
b. * [ Which vegetables ]i have [IP they [VP peeled t i and chopped the onions ] ] ?

As (26) shows, (25b) violates a parallelism constraint known as the across-the-board

(ATB) constraint, according to which movement of a conjunct out of a coordinate structure must
affect all co-conjuncts simultaneously.

(26) [ Which vegetables ]i have [IP they [VP peeled t i and chopped t i ] ] ?
Chapter 13: Constraints on wh- movement 371

The exceptional character of wh- movement out of coordinate structures (consistent with
subjacency, but still ungrammatical/unacceptable) and the special restriction imposed upon them
(the ATB constraint just mentioned) strongly suggest that the coordinate structure constraint is
sui generis and that it should not be grouped together with the other island constraints. Accord-
ingly, most proposals to reduce the island constraints to more general structural principles make
no attempt to include the coordinate structure constraint.

3.3 The status of DP as a barrier revisited

As we have just seen, some examples of wh- movement are unacceptable even though they respect
subjacency. The converse situation occurs as well. Chomsky’s definition of barrier to include both
IP and DP prohibits wh- movement out of all noun phrases even though such movement is often
completely acceptable, as shown in (27).

(27) a. ✔ Whoi did [IP you take [DP a picture of t i ] ] ?

b. ✔ [ { Which, how many } states ]i do [IP you know [DP the capitals of t i ] ] ?

(28) gives some naturally-occurring examples. In (28a), the struck-out which is included
for clarity; it is silent in the original. Nominal attempt is treated as taking a subject control
complement by analogy to the homonymous verb.

(28) a. When I was a little boy, he teased me about a temporary but intense devotion I had to
Gene Autry, the singing cowboy—a devotion whichi [IP I would make [DP some lame
attempt [CP t i [IP to justify t i … ] ] ] ]
(Calvin Trillin. 1996. Messages from my father. New York: Farrar, Strauss, and Giroux.
43–44)
b. arcane technical wisdom [ of which ]i [IP he has scarcely [DP a glimmering of [DP an
understanding t i ] ] ]
(Jeremy Campbell. 1982. Grammatical man. Information, entropy, language, and life.
Simon and Schuster. 260–261)

The acceptability of (27) and (28) suggests that perhaps DP is not a barrier after all—at
least not in all contexts. It is facts like these that have prompted attempts to derive the island
constraints, or at least some of them, from extragrammatical considerations, as we now briefly
illustrate. We begin by noting that any question is associated with a so-called existential presup-
position. The presupposition expresses the backdrop of knowledge against which the question is
raised, and the question itself solicits information that is missing in the questioner’s knowledge
store. For instance, the question in (29a) is associated with the presupposition in (29b).

(29) a. Question:
{ What, which book } are they reading?
b. Presupposition:
There is { something, some book } that they are reading.
Chapter 13: Constraints on wh- movement 372

Consider now the pattern of acceptability judgments for the questions in (30a) and for their
associated presuppositions in (30b), which range from ordinary to implausible.

(30) a. Question:
What did they { ✔read, ✔write, ?sell, #drop } a paper about?
b. Presupposition:
There is a topic that they are { ✔reading, ✔writing, ?selling, #dropping } a paper about.

If we make the natural assumption that, all other things being equal, the acceptability of a ques-
tion matches the plausibility of its associated presupposition, then questions that are perfectly
well-formed from a grammatical point of view might nevertheless be judged as unacceptable if
they are associated with a highly implausible presupposition (Kroch 1998). This, then, would
straightforwardly account for the successively decreasing acceptability of the variants of the
question in (30a), despite their structural parallelism.
It is worth noting that this attempt to derive the judgments in (30) from pragmatic consid-
erations is consistent with the judgments in (31).

(31) a. Idiomatic reading:

✔ What did they drop a hint about?
(cf. ✔There is a topic such that they hinted at it.)
b. Literal reading:
# What did they drop a hint about?
(cf. #There is a topic such that they dropped a piece of paper with the hint written
on it.)

3.4 Argument-adjunct asymmetries

Ross (1967) already observed that not all island violations are equally unacceptable. In particu-
lar, long movement of arguments is more acceptable than long movement of adjuncts. Subjects
(which are also non-complements) behave like adjuncts, giving the pattern in (32).

Extra Info
The pattern in (32) is generally known as the argument-adjunct asymmetry, though
it would be more accurate to call it an asymmetry between complements and non-
complements.

(32) a. ? [ Which problem ]i have they forgotten howk they should solve t i t k ? (The last one
on the page.)
b. * Howi have they forgotten [ which problem ]k they should solve t k t i ? (By Fourier
analysis.)
c. * [ Which problem ]i have they forgotten howk t i should be solved t k ? (The last one
on the page.)
Chapter 13: Constraints on wh- movement 373

The pragmatic approach just discussed in connection with (30) and (32) carries over natu-
rally to the contrast between (32a) and (32b). It does not fare so well, however, with (32c). The
relevant presuppositions are given in (33); indeed, it might be argued that the presupposition for
(32c) is identical to the one in (32a), which would render the contrast between those two examples
even more mysterious.

(33) a. ? There is a problem such that they have forgotten how they should solve that prob-
lem.
b. # There is a manner such that they have forgotten which problem they should solve
in that manner.
c. ? There is a problem such that they have forgotten how that problem should be solved.

3.5 The Comp-trace effect

The final constraint on wh- movement that we wish to discuss is the so-called Comp-trace effect.
(The name of the effect comes from a time when the term ‘complementizer’ was abbreviated as
‘Comp’—rather than as ‘C’, as it is today.) We begin by noting that long-distance wh- movement
of complements and adjuncts is unaffected by whether the complement clause is headed by an
overt complementizer or a silent one (indicated in the following examples by ∅that ).

(34) [ Which friends ]i did they say { ✔that, ✔ ∅ } they had invited t i ?

By contrast, long-distance movement of subjects is possible only with a silent complementizer.

Replacing the silent complementizer by the overt one is ungrammatical for many speakers; hence
the name of the effect.

(35) [ Which friends ]i did they say { ✔ ∅that , %that } t i had invited them?

As indicated by the percent sign, there is some variation among English speakers with
regard to the status of (35) (Sobin 1987). But even speakers who judge (35) to be acceptable report
a Comp-trace effect in connection with movement out of indirect questions. As we have just
seen, indirect questions are islands, and so long wh- movement out of them is not expected to be
acceptable to begin with. However, indirect questions introduced by whether or if tend to give
rise to relatively weak island effects; in (36), this weak effect is indicated by ‘?*’.

(36) ?* [ Which friends ]i did they forget whether they had invited t i ?

But analogous long subject movement, illustrated in (37), is completely unacceptable for all speak-
ers and thus provides evidence for the existence of a Comp-trace effect even for those speakers
who accept (36).

(37) * [ Which friends ]i did they forget whether t i had invited them?
 Chapter 13: Constraints on wh- movement 374

Relative clauses also exhibit the Comp-trace effect. The examples in (38) are parallel to
those in (34)–(37).

(38) a. the friends who(m)i they said { ✔that, ✔ ∅that } they had invited t i

b. the friends whoi they said { ✔ ∅that , %that } t i invited them

c. ?* the friends who(m)i they forgot whether they had invited t i

d. * the friends whoi they forgot whether t i had invited them

A puzzling fact is that local movement in that relative clauses, as in (39a), does not result
in a Comp-trace effect, unlike long wh- movement in (39b).

(39) a. the people [CP [ who ]i that t i moved in next door ]

b. % the people [CP [ who ]i that I think [CP that t i moved in next door ] ]

This contrast between local and nonlocal movement has given rise to many attempts at explana-
tion, but none that we know of is entirely satisfactory.

Notes 4
1. Ross (1967) treated relative clauses together with noun complement clauses, subsuming the noun phrases containing
them under the category of complex noun phrases. We follow more modern treatments by distinguishing the two
cases.

2. The representation in (18) assumes that which problem moves before how does, thereby preventing the complement
Spec(CP) from serving as an escape hatch for how. It is also necessary to rule out an alternative derivation, according
to which it is how that moves first. In that case, the complement Spec(CP) is empty, and how can move through it
up to the matrix Spec(CP) without violating subjacency. But then, of course, the intermediate trace of how blocks
movement of which problem into the lower Spec(CP).
Chapter 13: Constraints on wh- movement 375

Exercises and problems 5

Exercise 13.1
At first glance, the structure in (1a) seems preferable to that in (1b) on the grounds that it is simpler
in the sense of postulating fewer nodes. It is standardly argued, however, that the structure in
(1b) with a silent complementizer is preferable. Provide evidence for the standard view.

(1) a. They believe [IP we have read War and Peace ].

b. They believe [CP ∅that [IP we have read War and Peace ] ].

Exercise 13.2
According to the analysis of relative clauses presented in in Chapter 6, that relative clauses like
(1) have the structure in (2a). But since the wh- phrase is silent, an alternative analysis is possible
in principle, according to which the wh- phrase remains in its original position, as shown in (2b).
Provide evidence that the movement analysis in (2a) is preferable.

(1) the people that you met

(2) a. the people [CP whoi that [IP you met t i ] ] (movement)
b. the people [CP that [IP you met whoi ] ] (no movement)

Exercise 13.3
This exercise extends Exercise 6.7. Refer to that exercise for information concerning though
preposing.
A. The examples in (1)—and in particular, the though clauses—are completely acceptable. (The
remaining material is given to provide context; it should be ignored for the purposes of the
exercise.)

(1) a. ✔ Though I wonder why the problem seems difficult, my friends don’t wonder—
they know exactly.
b. ✔ Though the students enjoy problems which are difficult, they can’t always
solve them completely.

But the corresponding though preposing variants in (2) are not.

(2) a. * Difficult though I wonder why the problem seems, …

b. * Difficult though the students enjoy problems which are, …

Explain the ill-formedness of the though clauses in (2) in terms of subjacency.

B. Given subjacency, is the acceptability judgment for (3b) expected? Explain.
Chapter 13: Constraints on wh- movement 376

(3) a. Though we regret that the problem seems difficult, we will assign it anyway.
b. * Difficult though we regret that the problem seems, …

Exercise 13.4
Does the relative clause in (1) respect subjacency? Explain.

(1) a letter of which every line was an insult

(Jane Austen. 1981. The complete novels. New York: Gramercy. 84. From Chapter 7 of
Sense and sensibility.)

Exercise 13.5
Consider the sentence in (1a) and the simplified subpart of it in (1b).

(1) a. Already Agassiz had become interested in the rich stores of the extinct fishes of Eu-
rope, especially those of Glarus in Switzerland and of Monte Bolca near Verona, of
which at that time, only a few had been critically studied.
(Encyclopædia Britannica Online, Agassiz, (Jean) Louis (Rodolphe). Accessed 27 Au-
gust 1999.)
b. the fishes of Europe, of which only a few had been studied
A. Does the relative clause in (1b) respect subjacency? Explain.

Extra Info
(B) and (C) address issues unrelated to islands or subjacency.

B. Is of Europe a complement or an adjunct of fishes? How can you tell?

C. Does had in (1b) undergo V-to-I raising? How can you tell?

Exercise 13.6
A. Build structures for the two readings of (1).

(1) What would you like the power to do?

B. Briefly explain how your structures from (A) are consistent with subjacency.
Cross-linguistic 14
variation in wh-
movement

Chapter outline
1 To move or not to move? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378
1.1 Wh- scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378
1.2 Wh- in situ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378
1.3 A copy theory of movement . . . . . . . . . . . . . . . . . . . . . . . . . . 379
1.4 Partial wh- movement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 381
1.5 But what if they are errors? . . . . . . . . . . . . . . . . . . . . . . . . . . 383
2 Where to move? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384
2.1 Hungarian questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384
2.2 Hungarian relative clauses . . . . . . . . . . . . . . . . . . . . . . . . . . . 386
3 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387
4 Exercises and problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 388

In this chapter, we discuss some important properties of wh- movement in various lan-
guages other than English. As we have seen, wh- phrases undergo movement to Spec(CP) in
English. In many languages, however, wh- phrases occur in the same position as the correspond-
ing ordinary phrases. In the first part of the chapter, we present a way of unifying the treatment
of languages with and without wh- movement. In the course of doing so, we discuss a variant
of wh- movement known as partial wh- movement (Who do you think who is coming?) which
occurs in various adult languages as well as in child English. The second part of the chapter fo-
cuses on Hungarian, which resembles English in exhibiting wh- movement, but differs from it in
that wh- movement targets different positions in questions and relative clauses.
Chapter 14: Cross-linguistic variation in wh- movement 378

To move or not to move? 1

1.1 Wh- scope

For the purposes of this chapter, it is convenient to introduce the notion of wh- scope. For
simplicity, we restrict our attention for the moment to direct questions, and define the notion as
in (1).

(1) The scope of a wh-phrase in a question is determined with respect to some full answer to
the question.
The wh- phrase’s scope in the question is the IP that corresponds to the root in the full
answer.

In (2), the wh- phrases are underlined in red, and their scope is in bold.1

(2) a. Question: Wheni did [IP the visitors arrive t i ] ?

Full answer: [IP The visitors arrived at noon ].
b. Question: Whoi do [IP the parents think that the children saw t i ] ?
Full answer: [IP The parents think that the children saw the teacher ].

It is useful to think of the wh- phrases in (2) as marking their scope in the sense that they
move from inside their scope to the edge of it. In structural terms, a wh- phrase moves to the
specifier of the lowest projection that dominates its scope. We will refer to this position as the
wh- phrase’s scope-marking position or scope position.

1.2 Wh- in situ

In English, wh- phrases generally come to occupy their scope position by wh- movement. In
other languages, however, wh- phrases do not undergo wh- movement, but instead remain in
situ (that is, in the position where the wh- phrase enters the derivation of the question, whether
by substitution or adjunction). Such wh- in situ languages include Chinese, Hindi, Korean, and
Japanese, among many others. (3) gives the Japanese counterparts of (2).2
Once again, the wh- phrases are underlined in red, and their scope indicated in bold. As is
evident, a wh- phrase in a wh- in situ language is contained within its scope, rather than moving
to the edge of it. (We assume that the question marker か -ka is the overt counterpart to the silent
complementizer that we have been assuming for English questions; this is why it is represented
outside of the IP, in the clause-final position that is expected given the head-final character of
Japanese phrase structure.)
Chapter 14: Cross-linguistic variation in wh- movement 379

Extra Info
The abbreviations used in the glosses are: accusative (acc), locative (loc), nominative
(nom), plural (pl), present tense (pRs), Question (q), topic (top).

(3) a. Question:
[IP 見学者は いつ 着きました ] か。
[IP kengakusha-wa itsu tsuki-mashita ] ka?
visitor-top when arrive-pst q
‘When did the visitors arrive?’
Full answer:
[IP 見学者は 昼に 着きました ] 。
[IP kengakusha-wa hiru-ni tsuki-mashita ] .
visitor-top noon-at arrive-pst
‘The visitors arrived at noon.’
b. Question:
[IP 両親は 子供たちが 誰を 見た と 思います ] か。
[IP ryōshin-wa kodomo-tachi-ga dare-o mi-ta to omoi-masu ] ka?
parents-top child-pl-nom who-acc see-pst that think-pRs q
‘Who do the parents think that the children saw?’
Full answer:
[IP 両親は 子供たちが 先生を 見た と 思います ] 。
[IP ryōshin-wa kodomo-tachi-ga sensei-o mi-ta to omoi-masu ] .
parents-top child-pl-nom teacher-acc see-pst that think-pRs
‘The parents think that the children saw the teacher.’

1.3 A copy theory of movement

Given the aim of generative syntax of minimizing the differences among languages, it has been
proposed to derive wh- in situ questions by wh- movement—contrary to superficial appearances.
This becomes possible by slightly revising our ideas about movement. Instead of saying that
movement leaves a trace, let us say that it leaves a copy and that Universal Grammar allows vari-
ation concerning which copy is pronounced. In English, it is the highest copy of wh- movement
that is pronounced;3 in wh- in situ languages like Japanese, the lowest copy. In what follows, we
indicate silent copies by striking them through.

(4) a. [CP Which movie did [IP the children watch which movie ] ] ?
b. The parents asked [CP which movie [IP the children watched which movie ] ].
(5) a. [CP どの 映画を [IP 子供たちは どの 映画を 見ました ] か ] 。
[CP dono eiga-o [IP kodomo-tachi-wa dono eiga-o mi-mashi-ta ] ka ] .
child-pl-top which movie-acc see-pst q
‘Which movie did the children watch?’
Chapter 14: Cross-linguistic variation in wh- movement 380

b. 両親は [CP どの 映画を [IP 子供たちが どの 映画を 見た ] か ]

ryōshin-wa [CP dono eiga-o [IP kodomo-tachi-ga dono eiga-o mi-ta ] ka ]
parents-top child-pl-nom which movie-acc see-pst q
聞いた。
kii-ta.
ask-pst
‘The parents asked which movie the children watched.’

Word of Caution
For convenience, we will continue to use the terms ‘wh- movement’ and ‘wh- in situ’ to
refer to languages, questions, and so on, in which the highest and lowest copies of a wh-
phrase are pronounced, respectively.

Independent evidence in favor of the copy theory of movement comes from the judgments
of speakers of certain languages who do not allow ordinary long-distance movement. For in-
stance, some speakers of German reject the examples in (6), while accepting the corresponding
wh- copy questions in (7).

(6) a. % Wen glauben die Besucher, dass sie gesehen haben?

who-acc believe the visitors that they seen have
‘Who do the visitors believe that they saw?’
b. % Wen glaubst du, dass Martin meint, dass du magst?
who-acc believe you that thinks that you like
‘Who do you believe that Martin thinks that you like?’
(7) a. % Wen glauben die Besucher, wen sie gesehen haben?
who-acc believe the visitors who-acc they seen have
‘Who do the visitors believe that they saw?’
b. % Wen glaubst du, wen Martin meint, wen du magst?
who-acc believe you who-acc thinks who-acc you like
‘Who do you believe that Martin thinks that you like?’

The overt repetition of the wh- phrase in (7) lends strong support to the idea that wh-
phrases move through Spec(CP) in long-distance movement. In addition, it supports the idea that
grammars can differ as to which copies of movement are pronounced. (8) and (9) give isomorphic
structures for (6) and (7), which differ only in which links are pronounced: only the highest one in
ordinary long-distance wh- movement, or all links except the lowest in the coresponding wh- copy
questions. ∅dass indicates the silent counterpart of the complementizer dass ‘that’; like modern
standard English, the varieties of German under discussion allow either Spec(CP) or C, but not
both, to be filled with overt material in subordinate clauses.

(8) a. [CP Wen glauben die Besucher, [CP wen dass sie wen gesehen haben ] ] ?
b. [CP Wen glauben die Besucher, [CP wen ∅dass sie wen gesehen haben ] ] ?
(9) a. [CP Wen glaubst du, [CP wen dass Martin meint, [CP wen dass du wen magst ] ] ] ?
b. [CP Wen glaubst du, [CP wen ∅dass Martin meint, [CP wen ∅dass du wen magst ] ] ] ?
Chapter 14: Cross-linguistic variation in wh- movement 381

1.4 Partial wh- movement

Certain languages exhibit yet a further variant of wh- movement in which the wh- phrase un-
dergoes wh- movement, but to a position lower than its scope position. The scope position it-
self is occupied by a distinct wh- phrase (generally the language’s counterpart of what). This
phenomenon, known as partial wh- movement, is illustrated for German in (10).4 The true (=
contentful) wh- phrase is underlined in red, and the scope-marking wh- phrase is in blue bold-
face.

(10) a. Was glauben die Besucher, wen ∅dass sie wen gesehen haben?
what believe the visitors who-acc they seen have
‘Who do the visitors believe that they saw?’
b. Was glauben die Besucher, mit wem ∅dass sie mit wem gesprochen haben?
what believe the visitors with who-dat they spoken have
‘Who do the visitors believe that they talked with?’

For comparison, (11) shows the full wh- movement counterparts of (10), where the wh- phrase
functions as a scope marker on its own.5

(11) a. Wen glauben die Besucher, wen ∅dass sie wen gesehen haben?
who-acc believe the visitors they seen have
‘Who do the visitors believe that they saw?’
b. Mit wem glauben die Besucher, mit wem ∅dass sie mit wem gesprochen
with who-dat what believe the visitors they spoken
haben?
have
‘Who do the visitors believe that they talked with?’

The scope of the wh- phrase needs to be indicated in one of the two ways just presented.
The absence of scope marking results in severe ungrammaticality, as shown in (12).

(12) a. * Glauben die Besucher, wen ∅dass sie wen gesehen haben?
believe the visitors who-acc they seen have
Intended meaning: (11a)
b. * Glauben die Besucher, mit wem ∅dass sie mit wem gesprochen haben?
believe the visitors with who-dat they spoken have
Intended meaning: (11b)

The scope-marking wh- phrase in (10) is often called a wh- expletive. The idea is that the
relationship between it and the true wh- phrase is comparable to the relationship between the
expletive subject and the logical subject in the expletive there construction. In that construction,
the logical subject occupies a VP-internal position marking its function as an argument of a verb
(we will remain agnostic about the exact location of the position), and expletive there occupies
Spec(IP), marking the subject’s function as the subject of predication. It is this higher position
that the logical subject occupies in an ordinary sentence without expletive there. As (13) shows,
the higher position cannot be left empty.
Chapter 14: Cross-linguistic variation in wh- movement 382

(13) a. There remain several vexing problems. — analogous to (10)

b. Several vexing problems remain several vexing problems. — analogous to (11)
c. * Remain several vexing problems. — analogous to (12)

The analysis of partial wh- movement that we will propose relies on the copy theory of
the movement as well as on an idea introduced in earlier chapters—namely, that the elements
manipulated by the syntax are not words per se, but rather features or bundles of features that
are spelled out by the morphology. Two examples are given in (14).

(14) Features Spellout

⎡ person: 3 ⎤
⎢ ⎥
number: pl⎥
a. ⎢ → their
⎢ gender: m ⎥
⎢ case: poss ⎥
⎣ ⎦
root: sing
b. → sang
[tense: pst]

In the present case, let us assume that wh- phrases consist of a wh- feature and some remaining
substantive content. This syntactic dichotomy corresponds semantically to a so-called wh- quan-
tifier (which takes scope over some proposition) and a restriction on the quantifier. For instance,
if the wh- quantifier is restricted to humans, the bundle consisting of the features [wh: +] and
[human: +] is spelled out as who. Otherwise, the wh- quantifier is spelled out by default as what.
In ordinary wh- movement, the wh- feature and the features associated with it in its in-situ po-
sition always move together. In partial wh- movement, however, the feature bundle moves only
as far as the next highest Spec(CP). After that, the wh- feature moves on alone. When the sen-
tence is spelled out, lone wh- features are spelled out as the language’s default wh- word. Under
this analysis, partial wh- movement is partial in two distinct (though related) senses. First, the
true wh- phrase moves only part of the way to its scope position. Second, the wh- phrase in the
scope position contains only part of the features of the true wh- phrase. The derivations that we
are proposing for the partial movement in (10a) and its full movement counterpart in (11a) are
indicated schematically in (15).

wh: + wh: +
(15) a. [wh: +] glauben die Besucher, ∅dass sie gesehen haben?
[ human: +] [human: +]
wh: + wh: + wh: +
b. glauben die Besucher, dass sie gesehen haben?
[human: +] [human: +] [human: +]

In instances of partial wh- movement with more than two clauses, the true wh- phrase can
move to the lowest Spec(CP) or to the intermediate one. The two options are shown in (16a) and
(16b). (16c) gives the variant with full wh- movement by way of comparison. For simplicity, we
show the questions after the features have been spelled out.

(16) a. [CP Was glaubst du, [CP was ∅dass Martin meint, [CP wen ∅dass du wen magst ] ] ]
b. [CP Was glaubst du, [CP wen ∅dass Martin meint, [CP wen dass du wen magst ] ] ]
Chapter 14: Cross-linguistic variation in wh- movement 383

c. [CP Wen glaubst du, [CP wen dass Martin meint, [CP wen dass du wen magst ] ] ]
who-acc believe you that thinks that you like
‘Who do you believe that Martin thinks that you like?’

1.5 But what if they are errors?

Although adult English exhibits neither wh- copy questions or partial wh- movement, both types
of questions have been reported for child English (Thornton 1995, p. 187), as illustrated in (17)
and (18).

(17) a. Who do think who Grover wants to hug?

b. Who do you think who’s in there, really, really, really?
(18) a. What do you think what Cookie Monster eats?
b. What do you think what the baby drinks?

The availability of partial wh- movement in child English raises a subtle issue. We have been
assuming that the various constructions under discussion (ordinary long-distance wh- movement,
wh- copy questions, and partial wh- movement) are generated by slightly distinct grammars.
But the fact that wh- copy questions and partial wh- movement occur in acquisition suggests
the possibility that these variants aren’t generated by separate grammars, but that they reflect
a failure to properly implement structures that are generated by the adult standard grammar.
We know that speakers make performance errors under taxing conditions, and long-distance
wh- movement imposes a greater processing load than does local wh- movement, especially for
speakers with immature cognitive resources. Under this latter view, wh- copy questions and
partial wh- movement are comparable to the restart questions mentioned in Chapter 1, § 2.3.

(19) Is the boy who was holding the plate, is he crying?

As it turns out, both views are correct—for different speakers (Crain and Thornton 1998,
pp. 196–198). Some children produce wh- copy questions and partial wh- movement only rarely
and under special conditions (for instance, at the beginning of an experimental session, when they
are more nervous than later on). For such children, it is reasonable to assume that the nonadult
questions are in fact performance errors. But other children produce the nonadult questions con-
sistently and independently of various factors that favor performance errors (such as beginning
of session, young age, and distance between the scope position and the in-situ position, whether
measured in terms of numbers of words or number of intervening clauses). For such children,
it is reasonable to assume that the nonadult questions are generated by grammars that are con-
sistent with Universal Grammar, although they happen not to be the grammars in use in the
child’s language community. Analogous reasoning of course extends to adult usage, leading us
to assign a different status to the one-time production of a wh- copy question by a highly embar-
rassed English speaker than to the consistent production of such questions by an entire speech
community.
A final note is in order. Even if it had turned out that wh- copy questions and partial wh-
movement were performance errors every time they occurred for all speakers, these phenomena
would still provide insight into the speakers’ underlying grammars. The reason is that even
Chapter 14: Cross-linguistic variation in wh- movement 384

undoubted performance errors respect the structures generated by the grammar and therefore
provide evidence concerning them.6 The canonical example is slips of the tongue, like the classic
examples in (20). Such errors provide clear evidence that syllables consist of onsets and rhymes
(as indicated by hyphenation), since they involve switching onsets.7

(20) You have h-issed all my m-ystery lectures … in fact, you have t-asted the whole w-orm.

Speech errors like the hypothetical example in (21), though conceivable in principle, never occur
because they do not respect syllable structure.

(21) * … in fact, you have ter-sted the whole wa-me.

Analogously, there are hypothetical errors concerning wh- movement that are attested in
neither child English nor German. For instance, the examples in (22) involve pronouncing copies
of the wh- phrase right after each subject.

(22) a. * Do [IP you what think that [IP Cookie Monster what likes ] ]
Intended meaning: ‘What do you think that Cookie Monster likes?’
b. * Glauben [IP die Besucher wen, dass [IP sie wen gesehen haben ] ]
believe the visitors who-acc that they who-acc seen have
Intended meaning: ‘Who do the visitors believe that they saw?’

Given the absence of such “crazy” errors, we can conclude that wh- copy questions and partial
wh- movement provide evidence for cyclic wh- movement to Spec(CP), variation concerning the
spellout of traces, and the existence of syntactic features—independently of whether these con-
structions are performance errors.

Where to move? 2

2.1 Hungarian questions

(23) gives a Hungarian sentence with so-called neutral order.8

(23) Emöke látta Attilá-t tegnap este.

saw -acc yesterday evening
‘Emöke saw Attila last night.’

In such neutral sentences, which are felicitous as answers to the question What happened?, the
subject occupies clause-initial position and precedes the verb. By contrast, in wh- questions like
(24), the clause-initial position is occupied by the wh- phrase, and the subject follows the finite
verb.
Chapter 14: Cross-linguistic variation in wh- movement 385

(24) a. Ki-t látott Emöke tegnap este?

who-acc saw yesterday evening
‘Who(m) did Emöke see last night?’
b. Mikor látta Emöke Attilá-t?
when saw -acc
‘When did Emöke see Attila?’

It is noteworthy that answers to such questions preserve the constituent order of the ques-
tion when the answer is intended as an exhaustive answer to the question. In fact, under such an
interpretation, the variant without subject-verb inversion is ungrammatical.

(25) a. Attilá-t látta Emöke tegnap este.

-acc saw yesterday evening
‘Emöke saw Attila (and no one else) last night.’
b. Tegnap este látta Emöke Attilá-t.
yesterday evening saw -acc
‘Emöke saw Attila last night (and only then).’
(26) a. * Attilá-t Emöke látta tegnap este.
b. * Tegnap este Emöke látta Attilá-t.

As expected if focused phrases (including questioned phrases) move to the same syntactic posi-
tion, questions with focused constituents are ungrammatical in Hungarian.

(27) a. * Ki-t tegnap este látta Emöke?

who-acc yesterday evening saw
Intended meaning: ‘Who(m) did Emöke see last night?’
b. * Mikor Attilá-t látta Emöke?
when -acc saw
Intended meaning: ‘When did Emöke see Attila?’

The constituent order in (24) and (25) is reminiscent of English questions, and so it is natural
to propose that focused phrases move to Spec(CP) in Hungarian, with concomitant movement
of the verb to C. But contrary to what that proposal predicts, the complementizer hogy ‘that’
precedes the wh-/focus position in various types of subordinate clauses: in ordinary complement
clauses with a focused constituent, in indirect yes/no questions with a focused constituent, and
in indirect wh- questions. (Hungarian is a so-called null subject language, so the subjects of finite
clauses can be silent, as they are in the matrix clauses in the following examples.)

(28) a. Azt mondják, hogy Attilá-t látta Emöke tegnap este.

it say.3pl that -acc saw yesterday evening
‘They say (it) that Emöke saw (only) Attila last night.’
b. Azt mondják, hogy tegnap este látta Emöke Attilá-t.
it say.3pl that yesterday evening saw -acc
‘They say (it) that Emöke saw Attila last night (only).’
Chapter 14: Cross-linguistic variation in wh- movement 386

(29) a. Azt kérdeztük, hogy Attilá-t látta-e Emöke tegnap este.

it asked.1pl that -acc saw-q yesterday evening
‘We asked (it) whether Emöke saw (only) Attila last night.’
b. Azt kérdeztük, hogy tegnap este látta-e Emöke Attilá-t.
it asked.1pl that yesterday evening saw-q -acc
‘We asked (it) whether Emöke saw Attila last night (only).’
(30) a. Azt kérdeztük, hogy ki-t látott Emöke tegnap este.
it asked.1pl that who-acc saw yesterday evening
‘We asked (it) who(m) Emöke saw last night.’
b. Azt kérdeztük, hogy mikor látta Emöke Attilá-t.
it asked.1pl that when saw -acc
‘We asked (it) when Emöke saw Attila.’

The examples in (28)–(30) clearly show that the target of focus movement in Hungarian
cannot be Spec(CP), but rather must be some lower position. Accordingly, it has been proposed
that the clausal structure of Hungarian is as indicated schematically in (31).

(31) [CP [C hogy [FocP Focus [Foc Verb ] [IP Subject … Verb … Focus … ] ] ]

2.2 Hungarian relative clauses

In addition to Spec(FocP), wh- movement in Hungarian also targets Spec(CP), but it does so not
across the board, as in English (as we have just demonstrated), but only in relative clauses. We
can’t tell this directly since relative pronouns, as in English, can’t co-occur with the complemen-
tizer.

(32) a. a férfi aki-t (*hogy) Emöke látott tegnap este

the man who-acc that saw yesterday evening
‘the man who(m) Emöke saw last night’
b. a nap amikor (*hogy) Emöke látta Attilá-t
the day when that saw -acc
‘the day when Emöke saw Attila’

However, focus movement is possible in relative clauses (though not in questions—recall

the ungrammaticality of (27)). This is as expected if relative pronouns move to a position distinct
from Spec(FocP).

(33) a. a férfi aki-t tegnap este látott Emöke

the man who-acc yesterday evening saw
‘the man who(m) Emöke saw last night (only)’
b. a nap amikor Attilá-t látta Emöke
the day when -acc saw
‘the day when Emöke saw (only) Attila’
 Chapter 14: Cross-linguistic variation in wh- movement 387

Notes 3
1. In questions with a single wh- phrase, the scope of the wh- phrase coincides with the question’s ground (recall the
discussion of it clefts in Chapter 2, § 2.4). However, the notions of scope and ground aren’t identical, as they don’t
coincide in more complicated questions with multiple wh- phrases (questions of the type Who saw what?).

2. In Japanese, subjects and topics of sentences are marked by が -ga and は -wa, respectively. Topics are generally
restricted to matrix clauses. When subjects function as topics, as they often do, topic marking overrides subject
marking (in other words, the sentence contains no ga-marked phrase). When nominative case marking is at issue,
linguists use subordinate clauses to avoid the masking effect of topic marking, but we are not interested in case
marking here, so we are free to use matrix clauses.

3. Even languages that ordinarily require wh- movement allow wh- in situ under certain circumstances. For instance,
English echo questions commonly exhibit wh- in situ (see Chapter 6 for examples), and all but one wh- phrase is
required to remain in situ in English multiple wh- questions like (i).

(i) a. Who who saw what?

b. * Who what who saw what?

4. A classic reference for partial wh- movement is McDaniel (1989), which discusses the phenomenon in German and
Romani. Lutz, Müller, and Stechow (2000) is a collection of papers providing a cross-linguistic survey of partial wh-
movement and wh- scope marking more generally.

5. As noted in the previous section, not all German speakers accept long-distance wh- movement as in (6).

6. Performance errors also provide insight into the structure of various cognitive domains other than language. For
instance, if an instructor misremembers the names of Jeannine and Mark as Jennifer and Michael, that tells us that
names are organized in that person’s memory by first letter. Another person might misremember Jeannine as Evelyn
because Jeannine resembles Evelyn; for such a person, names would be organized by facial similarity.

7. For those unfamiliar with phonology, the onset is the beginning consonant (cluster) of a syllable, while the rhyme
consists of the vowel and consonant(s) at the end of that syllable. In (20), we specifically address speech errors
involving changes in the onset of a word’s first syllable.

8. Thanks to Eva Banik for the Hungarian examples in this section. For detailed discussions of relevant facts, see also
Kiefer and Kiss (1994) and Kiss (2002).
Chapter 14: Cross-linguistic variation in wh- movement 388

Exercises and problems 4

Exercise 14.1
Given what you know about wh- movement, which (if any) of the following sentences are ex-
pected to be grammatical in Hungarian?

(1) Relative out of question:

a. a férfi aki-nek kérdezik, hogy mi-t mondott Emöke
the man who-dat asked-3pl that what-acc told
‘the man such that they asked what Emöke told him’
b. a titok ami-t kérdezték, hogy ki-nek mondta Emöke
the secret which-acc asked-3pl that who-dat told
‘the secret such that they asked to whom Emöke told it’
(2) Relative out of relative:
a férfi aki-t Emöke hívta a nöt aki ismeri
the man who-acc called the woman who knows
‘the man such that Emöke called the woman who knows him’
(3) Question out of relative:
Ki-t hívott Attila a nöt aki ismer?
who-acc called the woman who knows
‘Who is the person such that Attila called the woman who knows that person?’
(4) Question out of question:
a. Ki-nek kérdezik, hogy mi-t mondott Emöke?
who-dat asked-3pl that what-acc told
‘Who is the person such that they asked what Emöke told that person?’
b. Mi-t kérdezték, hogy aki-nek mondta Emöke?
what-acc asked-3pl that who-dat told
‘What is it such that they asked to whom Emöke told it?’
 15
Binding theory

Chapter outline
1 Reference and related notions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 390
1.1 Reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 390
1.2 Coreference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 391
1.3 Coindexing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 391
1.4 Grammatical and ungrammatical index assignments . . . . . . . . . . . . 392
2 Hellan 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 393
2.1 The co-argument condition . . . . . . . . . . . . . . . . . . . . . . . . . . 393
2.2 The predication condition . . . . . . . . . . . . . . . . . . . . . . . . . . . 397
2.3 The finite IP condition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 398
2.4 Strict vs. non-strict co-arguments . . . . . . . . . . . . . . . . . . . . . . . 398
3 Extending Hellan’s binding theory to English . . . . . . . . . . . . . . . . . . . . . 401
3.1 The co-argument condition . . . . . . . . . . . . . . . . . . . . . . . . . . 401
3.2 The predication condition . . . . . . . . . . . . . . . . . . . . . . . . . . . 403
3.3 The finite IP condition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 403
4 Chomsky (1981) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 403
4.1 Principle A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 404
4.2 Principle B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 408
4.3 Principle C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409
5 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 411

This chapter is devoted to binding theory, the part of syntactic theory that is concerned
with how the interpretation of noun phrases is constrained by syntactic considerations. For the
purposes of binding theory, it is useful to distinguish several types of noun phrases: full noun
phrases (the question, the student that asked the question, and so on), ordinary pronouns (I, you,
they, and so on), reflexive pronouns (myself, yourself, themselves, and so on) and the reciprocal
pronoun each other.
We begin by presenting background concepts concerning reference and related notational
conventions. We then discuss two alternative approaches to binding theory. The first approach,
due to Hellan (1988), was proposed on the basis of Norwegian, a language with an unusually
Chapter 15: Binding theory 390

rich set of pronouns. We first discuss the Norwegian facts and then extend Hellan’s analysis to
English. The second approach, due to Chomsky (1981), was proposed on the basis of English and
does not cover the full range of Norwegian facts. However, it includes an important condition on
the distribution of ordinary noun phrases that is missing from Hellan’s binding theory.

Reference and related notions 1

1.1 Reference
The preeminent function of noun phrases like Benjamin Franklin, my two cats, the king of France,
Santa Claus, or colorless green ideas is to refer—that is, to stand for a particular discourse entity,
the referent. As the examples show, referents can be entities in the actual world, entities in some
possible world, or even entities that could not possibly exist in principle. One of the characteristic
features of human language is the absence, in general, of a one-to-one relation between noun
phrases and referents. On the one hand, it is possible to use different noun phrases to refer to the
same referent. The classic example for this is the fact that the expressions the morning star and
the evening star both have the same referent, the planet Venus (which is not a star at all!). On the
other hand, the same noun phrase can have different referents. For instance, my two cats, used
either by the same person at different points in time or by different persons at the same point in
time, can refer to distinct feline individuals. Similarly, my current checking account balance can
refer to vastly differing amounts of money.
In general, then, determining the intended referent of an expression requires recourse to a
particular discourse context (who is speaking when, to whom, and so on). The interpretation of
certain expressions, however, is particularly context-dependent. The expressions in question are
pronouns. For instance, it is perfectly natural to introduce a new topic in a conversation with a
friend using (1a) (provided that the speaker and the friend have in common a unique (or at least
uniquely salient) acquaintance by the name of Vanessa). But replacing the proper noun Vanessa
with a pronoun, as in (1b), in the same context is decidedly odd.

(1) a. I ran into Vanessa the other day.

b. I ran into her the other day.

But if Vanessa has already been mentioned in the discourse, then (1b) is perfectly felicitous, as
illustrated in the mini-discourse in (2).

(2) A: Have you seen Vanessa recently?

B: I ran into her the other day.

Pronouns, then, in contrast to ordinary noun phrases, are referentially dependent on

some antecedent in the discourse.
Chapter 15: Binding theory 391

The term ‘antecedent’ is potentially misleading. Since it is derived etymologically from Latin
ante-cedens ‘one who walks before’, an antecedent might reasonably be expected to be required to
precede a referentially dependent expression. However, contrary to this expectation, precedence
is neither a necessary nor a sufficient condition for antecedenthood. In a sentence like (3), for
instance, Zelda does not precede she, yet can still serve as the pronoun’s antecedent.

(3) If she calls, tell Zelda that the package arrived.

Conversely, Zelda precedes her in both examples in (4), but is unable to serve as the antecedent
of her in (4b) (in other words, (4b) cannot mean ‘Zelda likes herself’).

(4) a. Zelda believes that nobody likes her.

b. Zelda likes her.

A less misleading term for the notion of antecedent might be ‘referential anchor’, but we will
continue to use the standard term ‘antecedent’.

1.2 Coreference
A discourse will often contain more than one possible antecedent for a pronoun. For instance, in
(5), he can refer to either Tim or Tom.

(5) Tim told Tom that he needed some time off.

In any particular discourse context, of course, one interpretation may well be favored over the
other, given what is known about Tim, Tom, their respective work loads, and so on. In case the
antecedent for he must be explicitly specified in an unambiguous way, this can be achieved as in
(6).

(6) a. Tim told Tom that he, Tim, needed some time off.
b. Tim told Tom that he, Tom, needed some time off.

When (5) is given the interpretation in (6a), he and Tim are said to corefer. On the alternative
interpretation in (6b), it is he and Tom that corefer.

1.3 Coindexing
It is convenient to introduce a general notational device to represent coreference relations. Let us
begin by asssociating any noun phrase with a referential index. In the literature, it is standard
to use the letters of the alphabet as indices, beginning with i (for ‘index’). But as we already
use i, j, k, … as indices with another function1 , we adopt the convention in this chapter of using
the natural numbers as referential indices. Specifically, in order to indicate an interpretation in
which two expressions refer to the same discourse entity (that is, in which they corefer), we
assign the same index to both expressions. The two expressions are then said to be coindexed.
On the other hand, in order to indicate an interpretation in which two expressions refer to distinct
discourse entities (that is, in which they do not corefer), we assign distinct indices to each of the
Chapter 15: Binding theory 392

two expressions. Such expressions are said to be contraindexed. In neither case are the specific
indices important - only whether the indices are the same or not. That is, both indexings in
(7) represent the interpretation in (6a), and both indexings in (8) represent the interpretation in
(6b).

(7) a. Tim1 told Tom2 that he1 needed some time off.
b. Tim1097 told Tom18 that he1097 needed some time off.
(8) a. Tim1 told Tom2 that he2 needed some time off.
b. Tim380 told Tom7 that he7 needed some time off.

(9) gives a further possible indexing for the sentence in (5). Of course, in any particular
discourse, this indexing is felicitous only if a discourse entity with the index 3 (say, Tim’s brother
Mike) has already been mentioned.

(9) Tim1 told Tom2 that he3 needed some time off.

1.4 Grammatical and ungrammatical index assignments

Take good care to distinguish between reference and indexing. Reference relations are actual
linguistic relations that we have intuitions about. For instance, we have the intuition that Tim
and him can corefer in (10a), but not in (10b).

(10) a. Tim thinks that everyone admires him.

b. Tim admires him.

The assignment of indices, on the other hand, is a notational device that is intended to
represent arbitrary reference relations; the indices themselves have no independent linguistic
or psychological status. As a result, it is perfectly possible to assign referential indices to noun
phrases so as to represent interpretations of a sentence that are possible in principle, but impos-
sible in fact. Two such ungrammatical indexings are illustrated in (11).

(11) a. * Tim1 admires him1 .

b. * He1 admires Tim1 .

Notice that the proposition that both (11a) and (11b) are trying to express is not inherently
semantically anomalous. It can be expressed perfectly grammatically as in (12).

(12) Tim1 admires himself1 .

Notice furthermore that what makes these sequences ungrammatical is the index assignment.
The same sequences of words as in (11) are grammatical sentences when associated with the
indices in (13).

(13) a. ✔ Tim1 admires him2 .

b. ✔ He1 admires Tim2 .
Chapter 15: Binding theory 393

In other words—as mentioned in Chapter 1—the grammaticality of a sequence is always deter-

mined with reference to a particular interpretation.
It is often convenient to combine the information in (11) and (13) as in (14). The descending
numerical order of the indices on the object noun phrases is intended to make clear the scope of
the asterisk.

(14) a. ✔ Tim1 admires him2, *1 .

b. ✔ He1 admires Tim2, *1 .

Why can’t sentences like (11a) or (11b) express the proposition that is expressed gram-
matically in (12)? Such questions are what binding theory seeks to answer, the topic of Chap-
ter 15.

Hellan 1988 2

2.1 The co-argument condition

English makes a distinction between reflexive pronouns ending in -self (myself, yourself, and so
on) and the corresponding ordinary pronouns (I, you, and so on). Norwegian, too, distinguishes
between pronouns with forms containing selv ‘self’ and ones without. As (15) and (16) show, selv
is in complementary distribution with zero (that is, where a form with selv can occur, a form
without selv cannot, and vice versa).

(15) a. ✔ Jon1 beundrer seg selv1 .

admires seg self
b. * Jon1 beundrer seg1 .
‘Jon1 admires himself1 .’
(16) a. * Jon1 bad oss beundre seg selv1 .
asked us admire seg self
b. ✔ Jon1 bad oss beundre seg1 .
‘Jon1 asked us to admire him1 .’

In an effort to understand what determines the complementary distribution, let us consider

the derivation of the sentences. Both sentences require the elementary tree in (17) for the verb
beundre(r) ‘admire(s)’.
Chapter 15: Binding theory 394

(17) VP

»DP« V′

V◦ »DP«
beundre(r)

Observe now that in the derivation of (15a), seg selv and Jon both substitute into the elementary
tree for beundre(r). In the derivation of (16a), on the other hand, the subject position of the
elementary tree for beundre(r) is filled by oss ‘us’, whereas Jon substitutes into the elementary
tree for bad ‘asked’. (We treat bad ‘asked’ as taking a VP small clause complement, like its English
cognate bid.) The structures resulting from substitution in the two cases are shown in (18) and
(19).

(18) VP

DP V′

Jon V◦ DP
beundre(r)
seg selv

(19) a. VP b. VP

DP V′ DP V′

Jon V◦ »VP« oss V◦ VP

bad beundre
seg selv

Given this derivational difference, we can define a notion of co-argument as in (20a) and
propose a condition governing the distribution of selv as in (20b).

(20) a. Definition of co-argument (to be revised):

A and B are co-arguments if and only if they substitute into the argument positions
(specifier or complement) of the same elementary tree.
b. Co-argument condition (to be revised):
Selv and its antecedent (more precisely, the minimal DP nodes dominating them) must
be co-arguments.
Chapter 15: Binding theory 395

One type of syntax/semantics mismatch

While adequate for (15) and (16), the definition and condition in (20) does not extend to examples
like (21) and (22).

(21) a. ✔ Jon1 snakket om seg selv1 .

talked about seg self
b. * Jon1 snakket om seg1 .
‘Jon1 talked about himself1 .’
(22) a. * Jon1 bad oss snakke om seg selv1 .
asked us talk about seg self
b. ✔ Jon1 bad oss snakke om seg1 .
‘Jon1 asked us to talk about him1 .’

As is evident, the distribution of selv in (21) and (22) is parallel to that in (15) and (16).
However, the grammatical (21a) violates the condition in (20b) because the DPs dominating seg
selv and its antecedent do not substitute into the same elementary tree. Rather, seg selv substi-
tutes into the elementary tree for om ‘about’, whereas its antecedent Jon substitutes into that for
snakket ‘talked’.
One way of addressing this problem is to introduce a semantically-based notion of argu-
ment, based on the idea that situations can be conceptually decomposed into events (activities,
states, qualities, and so on), on the one hand, and different types of participants in such events
(agents, experiencers, beneficiaries, goals, etc.), on the other. In general, the participants in an
event denoted by a lexical item correspond to the substitution nodes in the elementary tree asso-
ciated with that lexical item. Indeed, this correspondence is so general that substitution nodes are
often equated with participants in a situation, with both being referred to as arguments. How-
ever, locutions like snakker om DP ‘talk about DP’ provide evidence that syntactic arguments
(constituents of the elementary tree projected by a lexical item) must be distinguished from se-
mantic arguments (participants in the event denoted by the lexical item). In such cases, there is
a mismatch between the syntax and the semantics. Specifically, the DP, being a complement of
the preposition, stands in no local syntactic relation with the verb. Nevertheless, it is a semantic
participant in the ‘talking about’ activity. For the purposes of binding, the prepositional com-
plement’s status as a semantic argument is sufficient to license selv. This motivates a revision of
(20b) to read as in (23b).

(23) a. Semantic co-argument:

A and B are semantic co-arguments if and only if they denote co-participants in the
same situation.
b. Co-argument condition (to be revised):
Selv and its antecedent must be semantic co-arguments.

It should be noted that the status of a particular phrase as a semantic argument is not always
easy to determine, and that there is some individual variation regarding the relevant judgments.
This is reminiscent of the situation that we encountered in connection with the distinction be-
tween syntactic arguments and adjuncts discussed in Chapter 4, § 3.1.
Chapter 15: Binding theory 396

Another type of syntax/semantics mismatch

As we have just seen, a noun phrase can function as a semantic argument in an expression denot-
ing an event without being a syntactic argument of the verb itself. Also possible is the converse
state of affairs, where a noun phrase counts as a syntactic argument of a head without being its
semantic argument. We illustrate this second type of syntax/semantics mismatch with reference
to perception verb complements. Consider (24).

(24) Jon1 så seg selv1 sitte i stolen.

saw seg self sit in chair.def
‘Jon1 saw himself1 sit in the chair.’

As the tree for (24) in (25) shows, the matrix verb så ‘saw’ (like bad ‘asked’) takes a VP small
clause complement, so Jon and seg selv are not syntactic co-arguments in the sense of (20a).

(25) IP

DP1 I′

Jon I◦ VP
[pst]
V′ VP

V◦ DP1 V′
så
seg selv sitte i stolen

Seg selv is also not a semantic argument of så, as required by (23a), since the complement
subject position can be occupied by expletive det, the Norwegian counterpart of expletive there.
(The structure for (26) is beyond the scope of our discussion.)

(26) Vi så det sitte ett spøkelse i stolen.

we saw there sit a ghost in chair.def
‘We saw a ghost sit in the chair.’

The facts discussed so far motivate extending the concept of syntactic co-argument as in
(27).

(27) Syntactic argument:

A and B are syntactic co-arguments if and only if they are locally head-licensed by
the same head C.

The notion of local head-licensing was introduced (though not by that name) in Chapter 7, §3,
where we distinguished three case assignment configurations: head-comp, spec-head, and head-
spec. These structural notions are also relevant for binding theory. In (18), Jon and seg selv are
syntactic co-arguments because they stand in the spec-head and head-comp relation, respectively,
with beundrer ‘admires’. In (26), (the trace of) Jon and seg selv are co-arguments because they
Chapter 15: Binding theory 397

stand in the spec-head and head-spec relation, respectively, with så. Wait—what about (16a),
though? Fortunately, in that sentence, Jon and seg selv are too far apart to be co-arguments,
even under the extended definition in (27), and so (16a) continues to be ruled out—correctly—as
ungrammatical.
Our final revision of the co-argument condition on selv is given in (28).

(28) Co-argument condition (final version):

Selv and its antecedent must be semantic or syntactic co-arguments.

The statement of the condition is not yet ideal. The fact that (28) contains a disjunction (an
or statement) suggests that we do not have a complete understanding of the notion of co-
argument.

2.2 The predication condition

In addition to the distinction between elements with and without selv, Norwegian distinguishes
between the two third-person singular forms seg and ham. Like seg, ham occurs on its own as
well as before selv. As (29) and (30) show, seg and ham are in complementary distribution.

Word of Caution
Note that the noun phrases in (29) and (30) are all semantic arguments. Therefore, since
the co-argument condition in (28) is satisfied, the contrast in (29) and (30) must be due to
independent conditions governing the distribution of seg and ham.

(29) Jon1 snakket om { seg / *ham } selv1 .

talked about seg ham self
‘Jon1 talked about himself1 .’
(30) a. Vi snakket med Jon1 om { ham / *seg } selv1 .
we talked with about ham seg self
‘We talked with Jon1 about himself1 .’
b. Vi fortalte Jon1 om { ham / *seg } selv1 .
we told about ham seg self
‘We told Jon1 about himself1 .’

The difference between the two cases is that the (potential) antecedent is a subject in (29),
but not in (30). Further evidence that the antecedent of seg must be a subject comes from (31),
which shows that what is crucial is the antecedent’s status as a subject of predication.

(31) Vi gjorde Jon1 glad i { seg / *ham } selv1 .

we made fond of seg ham self
‘We made Jon1 fond of himself1 .’

The sentences in (29)–(31) obey the condition in (32).

(32) Predication condition:

Chapter 15: Binding theory 398

The antecedent of seg must be a subject of predication, whereas the antecedent of ham
must not be.

The co-argument condition in (28) and the predication condition in (32) are independent
of one another. The domain in which the referentially dependent elements under discussion can
appear can therefore be partitioned without overlap as in Table 15.1.2

Antecedent co-argument?
Antecedent subject of predication? Yes No
Yes seg selv seg
No ham selv ham

Table 15.1: Domains of Norwegian referentially-dependent forms seg, ham, and their selv variants

2.3 The finite IP condition

The contrast in (33) shows that seg is subject to a final condition, given in (34).

(33) a. ✔ [IP Jon1 bad oss forsøke å få deg til å snakke pent om seg1 ].
asked us try to get you towards to talk nicely about seg
‘Jon1 asked us to try to get you to talk nicely about him1 .’
b. * Jon1 var ikke klar over at [IP vi hadde snakket om seg1 ].
was not aware over that we had talked about seg
‘Jon1 was not aware that we had talked about him1 .’

(34) Finite IP Condition:

The minimal finite IP dominating seg must also dominate its antecedent.

2.4 Strict vs. non-strict co-arguments

As we saw earlier, selv is subject to the co-argument condition in (28), repeated in (35).

(35) Co-argument condition (final version):

Selv and its antecedent must be semantic or syntactic co-arguments.

The contrast in (36) might at first glance be taken to indicate that the reciprocal pronoun
hverandre ‘each other’ is not subject to such a condition.

(36) a. * Jon1 traff noen venner av seg selv1 .

met some friends of seg self
‘Jon1 met some friends of his1 .’
b. [ Jon og Marit ]1 traff noen venner av hverandre1 .
and met some friends of each.other
‘[ Jon and Marit ]1 met some friends of each other’s1 .’
Chapter 15: Binding theory 399

Nevertheless, as (37) shows, the distribution of hverandre is not completely unconstrained.

(37) ?* [ Jon og Marit ]1 leste mine bøker om hverandre1 .

and read my books about each.other
Intended meaning: ‘Jon read my book about Marit, and Marit read my book about Jon.’

The contrast between (36b) and (37) can be accounted for by introducing a distinction be-
tween strict and non-strict co-arguments. Strict arguments are defined as in (20) or (23), but
non-strict co-arguments are more loosely related. Consider the two configurations in (38).

(38) a. AP b. AP
XP A′
XP A ′

A◦ BP A◦ BP

B′ YP B′
B◦ YP B◦ ZP

In (38a), we will say that the absence of a specifier in BP allows B to extend its argument-taking
domain up to the first maximal projection that is ‘closed of’ by a specifier (in this case, AP). In
such a domain, which we will call a projection chain of B, the arguments of the heads in the
projection chain count as co-arguments in a non-strict sense. By contrast, in (38b), the presence
of Spec(BP) prevents the formation of such a projection chain.
The similarities and differences in the distribution of selv and hverandre can now be suc-
cinctly stated as in (39).

(39) a. Selv and its antecedent must be strict co-arguments.

b. Hverandre and its antecedent must be co-arguments (possibly non-strict ones).

The application of the condition in (39b) to (36b) and (37) is illustrated in the structures in
(40).
Chapter 15: Binding theory 400

(40) a. IP

DPi I′

Jon og Marit I◦ VP
[pst]
ti V′

V◦ DP
traff
D′

D◦ NP
noen
N′

N′ PP

N◦ P′
venner
P◦ DP
av
hverandre
b. IP

DPi I′

Jon og Marit I◦ VP
[pst]
ti V′

V◦ DP
leste
DP D′

mine D◦ NP
[poss]
N′

N′ PP

N◦ P′
bøker
P◦ DP
om
hverandre
Chapter 15: Binding theory 401

(40a) is a (recursive) instantiation of the projection chain configuration in (38a). None of

the maximal projections in the object DP is closed off by a specifier, and the projection chain of
the preposition av (the head of the elementary tree into which hverandre substitutes) therefore
extends to VP. As a result, (the trace of) Jon og Marit is a non-strict co-argument of hverandre and
can serve as its antecedent. (40b) is a recursive projection chain configuration as well, but here,
the projection chain of the preposition om extends only up to the object DP, which is closed off
by mine. The ungrammaticality of the sentence is then due to the fact that Jon og Marit is too far
away from hverandre to serve as an antecedent. Meanwhile, the potential antecedent mine (1st
person singular) clashes with hverandre (3rd person plural).

Extending Hellan’s binding theory to English 3

As we have just seen, Norwegian distinguishes between two third-person pronouns, seg and ham,
both of which can stand alone or be combined with selv. This yields four referentially dependent
forms—seg, seg selv, ham, and ham selv. Lacking the distinction between seg and ham, English
has only two referentially dependent forms. In order to maximize the parallel with Hellan’s
Norwegian examples, we illustrate these in what follows by him and himself. The question that
guides the discussion in this section is how these two forms divide up the syntactic territory
covered by the four forms of Norwegian.

3.1 The co-argument condition

Since English -self is cognate with Norwegian selv, it is reasonable to assume that English re-
flexive pronouns are subject to a condition analogous to the co-argument condition in (35). Such
a condition is given in (41a). The companion condition on ordinary pronouns in (41b) makes
explicit the complementary distribution between the two types of pronouns.

(41) a. Reflexive pronouns and their antecedents must be (semantic or syntactic) co-
arguments.
b. Ordinary pronouns and their antecedents must not be co-arguments.

The conditions in (41) are borne out by the data in (42) and (43).

(42) a. John1 admires { himself1 / *him1 }.

b. John1 talked about { himself1 / *him1 }.
c. We made John1 fond of { himself1 / *him1 }.
(43) a. John1 asked us to admire { him1 / *himself1 }.
b. John1 asked us to talk about { him1 / *himself1 }.
c. John1 made us fond of { him1 / *himself1 }.
Chapter 15: Binding theory 402

In (39a), we proposed that the antecedents of Norwegian selv must be strict co-arguments,
and so we are led to wonder whether this is true of English reflexive pronouns as well. Evidence
bearing on this question comes from examples like (44), which has the structure in (45).

(44) John1 bought some books about himself1 .

(45) IP

DPi I′

John I◦ VP
[pst]
ti V′

V◦ DP
bought
D′

D◦ NP
some
N′

N′ PP

N◦ P′
books
P◦ DP
about
himself

Himself has no strict co-arguments (the PP dominating it lacks a specifier), and (45) is therefore an
instance of a projection chain. In particular, given that the entire object DP contains no specifiers,
the projection chain of about extends up to VP, making himself and (the trace of) John non-strict
co-arguments. Since (44) is grammatical, we conclude that the antecedent of himself need only
be a non-strict co-argument. In this respect, then, himself differs from Norweian seg selv and
resembles hverandre ‘each other’.
As expected, if the projection chain of about is closed off by a specifier of the object DP, then
it is that DP that must contain an antecedent that agrees with himself in the relevant grammatical
features (gender, number, and person). This is illustrated by the contrast in (46).

(46) a. ✔ { She, they, I } bought John1 ’s book about himself1 .

b. * John1 bought { her, their, my } book about himself1 .

Finally, the contrast between (44) and (47) on the one hand and (46) and (48) on the other
confirms the complementary distribution between reflexive and ordinary pronouns.
Chapter 15: Binding theory 403

(47) * John1 bought some books about him1 .

(48) a. * { She, they, I } bought John1 ’s book about him1 .
b. ✔ John1 bought { her, their, my } book about him1 .

3.2 The predication condition

The predication condition in (32) governs the distribution of seg versus ham, and so we expect it to
be irrelevant in English. This is in fact so. (43) and (48b) show that the antecedent of him, unlike
that of its Norwegian counterpart ham, can be a subject of predication, and (46a) shows that the
antecedent of himself needn’t be. The latter fact is confirmed by (49), where the antecedent of
himself is not a subject of predication.

(49) a. We talked with John1 about { himself1 / *him1 }.

b. We told John1 about { himself1 / *him1 }.

Another way of putting this is that himself corresponds to both of the Norwegian selv
forms.

3.3 The finite IP condition

Like the predication condition, the finite IP condition governs the distribution of seg versus ham,
and so again, we expect it to be irrelevant in English. This expectation is confirmed by the paral-
lelism in (50), as opposed to the contrast in (33).

(50) a. ✔ [IP John1 asked us to try to get you to talk nicely about { him1 / *himself1 } ].
b. ✔ [IP John1 was not aware that we had talked about { him1 / *himself1 } ].

In summary, then, English -self corresponds almost perfect to Norwegian selv. The only
difference that we have seen is that the antecedent of -self, unlike that of selv, needn’t be a strict
co-argument.

Chomsky (1981) 4
So far in this chapter, we have presented an approach to binding theory that was developed in
order to account for the particularly rich data from Norwegian. In this section, we present the
standard binding theory of Chomsky (1981), which was developed based on the simpler data of
English. Chomsky’s binding theory contains three conditions (or principles, as they are more
commonly known), which govern the distribution of reflexive and reciprocal pronouns, ordinary
pronouns, and full noun phrases, respectively. We present each of these principles in turn.
Chapter 15: Binding theory 404

4.1 Principle A
Consider the English binding facts in (51).

(51) a. [ Zelda ]1 helped [ herself ]1 .

b. [ Zelda’s sister ]1 helped [ herself ]1 .
c. * [ Zelda ]1 ’s sister helped [ herself ]1 .

In Hellan’s binding theory, (51a) is grammatical because herself and its antecedent Zelda are syn-
tactic co-arguments. The same is true of herself and Zelda’s sister in (51b), and so it is grammatical
as well. In (51c), on the other hand, the intended antecedent Zelda is not a co-argument of herself,
and so the sentence is ungrammatical under the intended interpretation.
Chomsky (1981) derives the grammaticality pattern in (51) on the basis not of co-
argumenthood, but on the basis of a purely structural relation called c-command, defined as
in (52).

(52) A c-commands B if and only if

a. neither A nor B dominates the other, and
b. the first branching node that dominates A also dominates B.

The structures for the sentences in (51) are shown in (53). The branching node dominating
the intended antecedent is highlighted by a box. Notice now that the red boxed nodes dominates
that anaphor in (53a) and (53b), but not in (54). In other words, the anaphor is c-commanded by
the intended antecedent in (53a) and (53b), but not in (54).

(53) a. IP b. IP

DPi,1 I′ DPi,1 I′

Zelda I◦ VP DP D′ I◦ VP
[pst] [pst]
ti V′ Zelda D◦ NP ti V′
’s
V◦ DP1 sister V◦ DP1
helped helped
herself herself
Chapter 15: Binding theory 405

(54) IP

DPi I′

DP1 D′ I◦ VP
[pst]
Zelda D◦ NP ti V′
’s
sister V◦ DP1
helped
herself

These configurational relations suggest the condition in (55).

(55) Principle A (to be revised):

An anaphor must be c-commanded by a coindexed antecedent.

Word of Caution
The term anaphor in Chomsky’s usage refers to reflexive and reciprocal pronouns. This
usage is potentially confusing because in general linguistic usage, anaphora refers to ref-
erential dependence regardless of morphological form. In other words, ordinary pronouns
and even full noun phrases can count as anaphors in this wide sense. Here, we will use
the restricted sense of the term, in keeping with Chomsky’s usage.

(55) is generally expressed more succinctly as in (56a), where the notion of binding is defined as
in (56b).

(56) a. An anaphor must be bound.

b. A binds B if and only if
i. A c-commands B, and
ii. A and B are coindexed.

As it turns out, (56a) is a necessary but not sufficient condition on anaphors in English,
since it fails to account for complex sentences like (57).

(57) John1 thinks that Mary2 will help { herself2 / *himself1 }.

The grammaticality of the variant with herself is unproblematic (herself is bound by Mary). But
given (56a), the ungrammaticality of the variant with himself is surprising, because himself is
bound by John.
In Hellan’s version of the binding theory, John is ruled out as an antecedent of himself
by the co-argument condition in (41a). In Chomsky’s version, a similar effect is achieved by
introducing the notion of governing category, a locality domain within which an anaphor must
Chapter 15: Binding theory 406

be bound. The definition of governing category is fairly complex, so we will work our way up to
it in several steps, motivating each revision of the definition in turn.

(58) Principle A (final version):

An anaphor must be bound within its governing category.
(59) Governing category (to be revised):
The governing category for an expression is the lowest IP that contains that expression.

(58) correctly describes the contrast in (57), but is unable to account for contrasts as in (46),
repeated as (60).

(60) a. ✔ { She, they, I } bought John1 ’s book about himself1 .

b. * John1 bought { her, their, my } book about himself1 .

In both sentences, the lowest IP that contains himself (the only IP in the sentence) also contains
John, in accordance with the definition of governing category in (59). The ungrammaticality of
(60b) is therefore unexpected.
A first step in accounting for such contrasts lies in making reference to DPs in addition to
IPs in the definition of governing category, as in (61).

(61) Governing category (to be revised):

The governing category for an expression is the lowest IP or DP that contains that expres-
sion.

But now (61) overshoots the mark by incorrectly ruling out sentences like (44), repeated as
(62).

(62) John1 bought some books about himself1 .

A further refinement of the definition of governing category yields (63).

(63) Governing category (to be revised):

The governing category for an expression is the lowest IP or DP that contains
a. that expression
b. and a specifier.

The reference to a specifier in (63) is the formal counterpart to the distinction between strict
and non-strict co-arguments in Hellan’s approach to the binding theory. There is a difference,
however, in what the two approaches regard as the unmarked domain within which an anaphor
must be bound. In Hellan’s approach, the availability of non-strict co-argument antecedents
extends a binding domain that would otherwise be smaller. By contrast, in Chomsky’s approach,
the addition of (63b) restricts a binding domain that would otherwise be larger.
There remains one final revision to make to the definition of governing category. The
revision is motivated by sentences like (64a), which have the structure in (64b).
Chapter 15: Binding theory 407

(64) a. They1 believe themselves1 to be indispensable.

b. IP

DPi,1 I′

they I◦ VP
[pRs]
ti V′

V◦ IP
believe
DPk,1 I′

themselves I◦ VP
to
tk V′

V◦ AdjP
be
indispensable

Let’s consider in detail why (64) poses a difficulty for the definition in (63). As is evident
from (64b), the lowest IP or DP that contains the anaphor and a specifier is the complement IP
(indicated by the box). (The anaphor and the specifier turn out to be the same node, but nothing
in (63) rules this out.) But this complement IP contains no antecedent for the anaphor, so (64a) is
incorrectly predicted to be ungrammatical.
In order to accommodate sentences like (64a), the governing category must be extended in
just the right way to allow matrix subjects to act as antecedents in sentences like (64a), but not
in ones like (57). This is achieved by the formulation in (65).

(65) Governing category (final version):

The governing category for an expression is the lowest IP or DP that contains
a. that expression
b. a specifier,
c. and the expression’s case-assigning head.

The addition of (65c) has the same effect as Hellan’s extension of the notion of syntactic argument
in (27) (recall the discussion of the second type of syntax/semantics mismatch discussed earlier
in connection with the distribution of Norwegian selv).
Chapter 15: Binding theory 408

4.2 Principle B
We turn now to the distribution of pronouns.

Word of Caution
As noted earlier, reflexive and reciprocal pronouns are combined in Chomsky’s usage un-
der the rubric of anaphors. This leaves the unqualified term ‘pronoun’ to refer to ordinary
personal pronouns (I, you, he, and so on). Unless otherwise noted, we follow this usage
below.

Replacing the anaphors in (51) by pronouns yields the sentences in (66).

(66) a. * [ Zelda ]1 helped [ her ]1 .

b. * [ Zelda’s sister ]1 helped [ her ]1 .
c. ✔ [ Zelda ]1 ’s sister helped [ her ]1 .

The resulting grammaticality pattern, which is the converse of that in (51), suggests the condition
in (67).

(67) Principle B (to be revised):

A pronoun must be free, that is, not bound by an antecedent.

As in the case of our first formulation of Principle A, the formulation of Principle B in (67)
is not yet quite adequate. This time, however, the condition errs on the side of caution, incorrectly
ruling out grammatical sentences like (68).

(68) John1 thinks Mary will help him1 .

Since John c-commands him (along with everything else in the sentence), (67) incorrectly leads
us to expect (68) to be ungrammatical.
The fact that the antecedent and the pronoun belong to different clauses in (68) suggests
reformulating (67) to incorporate the concept of governing category, as in (69).

(69) Principle B (to be revised):

A pronoun must be free in its governing category.

As is evident, both Hellan’s and Chomsky’s approaches to the binding theory agree that
anaphors and pronouns are in complementary distribution.
Chapter 15: Binding theory 409

4.3 Principle C
Consider the sentences in (70).

(70) a. * She1 treats Mary1 well.

b. * She1 claims that they treat Mary1 well.
c. * She1 claims that we know that they treat Mary1 well.

In Hellan’s approach to the binding theory, (70a) is ruled out by the condition in (41b),
according to which an ordinary pronoun and its antecedent cannot be co-arguments. It is not,
however, ruled out by Principle B, the counterpart to (41b) in Chomsky’s binding theory, since
the pronoun is free. In order to rule out sentences like (70a), Chomsky’s binding theory there-
fore contains a third principle that governs the distribution of full noun phrases (referred to in
Chomsky’s usage as R(eferential)-expressions).

(71) Principle C:
R-expressions must be free.

Principle C is reminiscent of Principle B, but differs from it in that the anti-binding require-
ment is absolute (= independent of a governing category), as required by the ungrammaticality
of (70b) and (70c).
The ungrammaticality of the sentences in (70b) and (70c) means that an anti-binding condi-
tion on full noun phrases also needs to be incorporated into Hellan’s approach to binding theory.
This is because the co-argument condition in (41b), which ruled out (70a), fails to apply to (70b)
and (70c). They must therefore be ruled out by separate means. The necessity for an independent
anti-binding condition on full noun phrases is underscored by sentences like (72).

(72) * We made herself1 treat Mary1 well.

In Chomsky’s binding theory, (72) is ruled out by both Principles A and C. But in Hellan’s version
of the binding theory, (72) satisfies the co-argument condition in (41a), the equivalent of Principle
A, because herself and Mary are both syntactic co-arguments (in the spec-head and head-spec
configuration, respectively). It is therefore only with reference to an anti-binding condition on
full noun phrases that (72) can be ruled out as required.
In traditional grammar, the examples in (70) and (72) would all be called instances of cat-
aphora, instances of referential dependence where the antecedent follows the referentially de-
pendent element. It is worth pointing out explicitly that not all instances of cataphora are un-
grammatical. Examples like (73), where the referentially dependent element precedes the an-
tecedent, but does not c-command it, are grammatical, as expected under Principle C.

(73) If he1 calls, tell John1 I’ll be back in an hour.

Chapter 15: Binding theory 410

Word of Caution
The term ‘anaphora’ is generally used to refer to referential dependence, subsuming both
cataphora and what might be called strict anaphora, where the antecedent precedes the
referentially dependent element. According to the metaphor underlying the terms, dis-
course flows along like a stream, with temporally earlier elements located upstream from
later ones. The antecedent is then upstream of the referentially dependent element in
strict anaphora (< Greek ἀνα ana ‘up(stream)’), but downstream in cataphora (< Greek
κατά kata ‘down(stream)’).

A final point needs to be made about Principle C. We pointed out earlier that Principle C is
an absolute requirement in the sense that it makes no reference to governing categories (= binding
domains). It is absolute in a further sense as well: it makes no reference to whether the binder is
a referentially dependent element. Thus, Principle C rules out (74) on a par with (70a).

(74) *? Mary1 treats Mary1 well.

It has been argued, however, that given the right discourse conditions, sentences like (75),
which are structurally parallel to (74), are in fact acceptable. Two naturally-occurring examples
of the same type are given in (76).

(75) Nobody likes Oscar. Even Oscar1 doesn’t like Oscar1 .

(76) a. Phil1 said that if he2 came by, Phil1 would show him2 around.
([email protected] to [email protected])
b. Luke1 thinks that everyone has as much integrity as Luke1 has.
(overheard at the White Dog Cafe, Philadelphia, PA, 17 Feb 2001)

But equivalent sentences in which the full noun phrase is bound by a referentially depen-
dent element are ungrammatical.

(77) * Nobody likes Oscar. Even he1 doesn’t like Oscar1 .

(78) a. * He1 said that if { he2 / Sean2 } came by, Phil1 would show him2 around.
b. * He1 thinks that everyone has as much integrity as Luke1 has.

Although judgments regarding contrasts of the type illustrated by (75)–(78) can be delicate,
it seems reasonable to weaken Principle C to include reference to the status of the binder, as is
done in (79).

(79) Weakened version of Principle C:

Full noun phrases must not be bound by a referentially dependent element.
 Chapter 15: Binding theory 411

Notes 5
1. Actually, we will usually use i, k, m, … for better legibility at small font sizes.

2. Certain difficult cases are set aside here; for further discussion, see Hellan (1988, chapter 3).
Glossary

asterisk (*) indicates a sentence’s syntactic ill-formedness (ungrammaticality)

(1) * The had pipsqueak the nerve confront me to

Enclosing material that is preceded by an asterisk with parentheses indicates that including
the material in parentheses is ungrammatical. Thus, (2) abbreviates the examples in (3).

(2) a. The (*those) cats like treats.

b. My cats like(*s) treats.

(3) a. i. The cats like treats.

ii. * The those cats like treats.
b. i. My cats like treats.
ii. * My cats likes treats.

On the other hand, prefixing an asterisk to material that is enclosed in parentheses indicates
that the parenthesized material is obligatory. Thus, (4) abbreviates the examples in (5).

(4) a. They consumed *(dinner).

b. My cat like*(s) treats.

(5) a. i. * They consumed.

ii. They consumed dinner.
b. i. * My cat like treats.
ii. My cat likes treats.

72, 162
% percent A percent sign indicates that an example is accepted as grammatical by some speak-
ers, but rejected by others.

(6) % We use a gas stove anymore; anymore we use a gas stove.

(chiefly Midwest, but found throughout the United States except New England;
earliest recorded examples from Northern Ireland)
Chapter 15: Glossary 413

181
adposition Any P, regardless of whether it is head-initial or head-final. See Chapter 5 for ex-
amples. see also postposition
argument From the point of view of formal logic, an argument is an input to a predicate (in the
formal logic sense). From the point of view of syntax, specifically X′ theory, an argument is
a linguistic expression occupying the specifier or complement position of a head. Because a
predicate can have more semantic arguments than the X′ schema provides, semantic argu-
ments are not necessarily expressed as syntactic arguments. For discussion, see Chapter 3,
§ 1 and Chapter 4, § 3. Conversely, not all syntactic arguments are semantic arguments;
see Chapter 3, §§ 3.1, 3 for examples. 414, see also expletive
bound morpheme A morpheme that cannot stand alone, but must form part of a larger word,
like plural -s, un-, or -ness. In contrast, free morphemes, like cat or happy, can stand alone.
A trailing hyphen indicates that a bound morpheme is a prefix; a leading hyphen indicates
a suffix. 116, 414
clause A constituent that contains a subject, possibly silent (in boldface), and a predicate (in
italics). Clauses can be subdivided into ordinary clauses and small clauses. Ordinary
clauses can be further subdivided into finite clauses, which can stand alone, and nonfinite
clauses, which can’t. All ordinary clauses contain an Infl element (underlined)—a modal,
auxiliary, silent tense morpheme, or the nonfinite marker to. Small clauses differ from
ordinary clauses in lacking an Infl element.

(7) a. Our friends must be in Cancun by now. Finite clause

b. Bill has never seen a racoon.
c. They are our friends.
d. Bill [past] arrived.

(8) a. (John seems) to be having problems. Nonfinite clause

b. (I expect) to know tomorrow.
c. (I expect) them to know tomorrow.
d. (We consider) them to be our friends.

(9) a. (They made) us do it. Small clause

b. (We consider) them our friends.

121
compound tense A tense that is expressed analytically, for instance, the English present pro-
gressive she is singing or the French passé composé elle a chanté ‘she sang, she has sung’.
Compound tenses contrast with simple tenses, which are formed synthetically, like the
English simple past she sang or the French imparfait elle chantait ‘she used to sing, she
was singing’.
Compound tenses in one language can correspond functionally to simple tenses in another
language, and vice versa. For instance, the English compound present progressive corre-
Chapter 15: Glossary 414

sponds to a French simple present, whereas the English simple past often corresponds to a
French passé composé. 418
demotion see grammatical relation
determiner A syntactic category that includes the definite article the, the indefinite article a
and its variant an, the demonstratives this and that, and ordinary and reflexive pronouns.
English also has a silent determiner, marked by in (10) below, which resembles some
in that it can be used with both mass nouns and plural count nouns.

(10) a. I see rice on the table.

b. I heard lions out in the bush last night.

137, 140
evidence Here and throughout the book, we will use the term ‘evidence’ to refer to empirical
arguments for a conclusion, rather than in a more liberal sense in which it includes concep-
tual arguments based on theoretical virtues such as simplicity, economy, theory-internal
consistency, and so on. In other words, for us, the collocation ‘empirical evidence’ is re-
dundant.
Evidence consists of a linguistic expression (a sentence or phrase), along with a grammat-
icality judgment (✔ or *).
For instance, we might ask whether the underlined noun phrase in (11a) is a complement
or an adjunct of the italicized verb. The grammaticality of (11b) is evidence that it is an
adjunct (given assumptions spelled out in Chapter 4).

(11) a. They partied the entire weekend.

b. ✔ They did so the entire weekend.

What about the status of the same string in (12a)? In this case, it is the ungrammaticality
of (12b) that is evidence for the expression being a complement.

(12) a. They enjoyed the entire weekend.

b. * They did so the entire weekend.

29, 75, 106, 109

expletive (from Latin explere ‘to fill out’) A syntactic argument that isn’t a semantic argument.
See Chapter 3, § 3.1 for more discussion. 4
finite Finite verbs are verbs that are inflected for tense and person-number agreement. Finite
clauses are clauses that can stand alone. See Chapter 2, § 8 for more discussion. 61, 284
free morpheme A morpheme that can stand alone, like cat or happy. In contrast, bound mor-
phemes, like plural -s, un-, or -ness) are part of a larger word. 280, 413
grammatical relation The grammatical relations in a sentence are listed in (13); see Chapter 3,
§ 5 for more discussion.
Chapter 15: Glossary 415

(13) subject direct object indirect object object of preposition

promotion demotion

Various syntactic operations can change the grammatical relation of a noun phrase. De-
pending on whether a noun phrase moves “up” or “down” the hierarchy in (13), the noun
phrase is said to be promoted or demoted. For instance, the passive in English demotes
the subject of an active sentence to the object of the preposition by. In addition, it promotes
the object to subject. 237
homograph One of two or more linguistic forms that are spelled alike, but different in function
or meaning or in pronunciation.

Example:
bank ‘river bank’ and bank ‘financial institution’ (different meaning)
read (infinitive) and read (participle) (different pronunciation)

419, see also homophone, homonym & word

homophone One of two or more linguistic forms that are pronounced alike, but distinct in
function or meaning or in spelling.

Examples:
bank ‘river bank’ and bank ‘financial institution’ (same spelling, different meaning)
reed and read (infinitive); red and read (participle) (related meaning, different spelling)

213, 419, see also homograph, homonym & word

hypercorrection The psychological process of constructing a non-prescribed linguistic form by
analogy to a prescribed form. Also the resulting form itself. Example: They feel badly, by
analogy to They drive badly. The prescribed form is They feel bad. In They drive badly, badly
modifies drive, which is an ordinary verb. In They feel bad, on the other hand, feel is a
linking verb, and bad is a predicative adjective (it predicates a particular property of the
subject of the sentence; in other words, it attributes the property to the subject).
They feel badly is acceptable in prescriptions for “standard” English only if it is the verb feel
that is being modified; that is, if the sentence is intended to mean They have a bad sense of
touch.
As always, note that the prescribed “standard” form is not a cognitive or linguistic status,
but a social move to privilege one form of speaking over others. 26
intransitive Traditionally used of verbs that take no object. We use the term in a more gen-
eral sense to refer to any syntactic category that takes no complement. For instance, the
italicized heads are transitive in the (a) examples, but intransitive in their (b) counterparts.

(14) a. We have eaten the pizza. verb

b. We have eaten.
Chapter 15: Glossary 416

(15) a. They crawled underneath the table. preposition

b. They crawled underneath.
(16) a. I like that radio. determiner
b. I like that.

see also transitive

lexeme 13, 416, see word
lexical ambiguity The association of a single word with more than one lexeme. 13, 419
linking verb Also known as copular verb. One of a class of verbs including the copula be, as
well as (among others) appear, become, feel, get, grow, look, prove, seem, smell, sound, taste,
and turn, when used as in (17).

(17) a. They { are, appear, became, grew, look, proved, seem } competent.
b. It { feels, looks, smells, sounds, tastes } fine.
c. He { became, got, grew } old.
d. It turned rancid.

415, see also hypercorrection

matrix A structure that contains another structure. Typically used in the collocation ‘matrix
clause’ of a CP or IP that contains another CP or IP. 121
morpheme A minimal meaning-bearing element.
Words are not necessarily the smallest meaning-bearing elements in a language. For in-
stance, cats is a single word, but consists of two morphemes, cat and the plural morpheme
-s. Unhappiness consists of the three morphemes un-, happy, and -ness.
Morphemes can be free or bound. Free morphemes (like cat and happy) can stand alone,
whereas bound morphemes (like -s, un-, and -ness) cannot. A trailing hyphen indicates that
a bound morpheme is a prefix; a leading hyphen indicates a suffix. 413, 414, see also bound
morpheme & free morpheme
nominal Of or relating to a noun or its projections (N, N′ , NP); more generally, of or relating to
a noun phrase (DP). 138
ordinary pronoun Synonymous with “personal pronoun.” The following table lists the English
ordinary pronouns.
Chapter 15: Glossary 417

Possessive
Person Number Nominative Oblique
Prenominal Absolute
(That’s book.) (That book is )
1 I me my mine
2 sg you you your yours
3 he, she, it him, her, it his, her, its his, hers, its
1 we us our ours
2 pl you you your yours
3 they them their theirs

Table 15.2: English ordinary pronouns

see also full noun phrase

postposition A head-final P. Coined in order to avoid the expression ’head-final preposition’,
which offends the etymologically aware as a contradiction in terms. See Chapter 5 for
examples. see also adposition
predicate The term ‘predicate’ in linguistics has two distinct (though related) senses, what we
will call the subject-predicate sense and the predicate-argument sense.
The subject-predicate sense derives from traditional logic, where propositions are divided
into two parts, the subject and the predicate, and the predicate is what is affirmed (or de-
nied) of the subject.
In the history of formal logic, this original sense was generalized to include relations miss-
ing more than a single argument. In the resulting predicate-argument sense, in linguistics,
the term ‘predicate’ refers to a head that expresses a logical relation. Typically, predicates
in this sense are verbs, but other types of heads can function as predicates in this sense as
well.
The two senses are illustrated in (18) and (19). The predicate is underlined; notice that the
two senses can pick out the same expression, as in the examples (18b) and (19b).

(18) a. Bill gives money to charity. Subject-predicate sense

b. Bill swims.
(19) a. Bill gives money to charity. Predicate-argument sense
b. Bill swims.
c. Sheila’s criticism of the plan.

413, see also argument

promotion see grammatical relation
raising The movement of a subject from an embedded clause to a matrix clause. see also control
verb & raising verb
reflexive pronoun The English reflexive pronouns are easy to identify because they all contain
a form of the morpheme self, as laid out in the following table.
Chapter 15: Glossary 418

1 2 3
himself
sg myself yourself herself
itself
pl ourselves yourselves themselves

Table 15.3: English reflexive pronouns

216, see also full noun phrase

simple tense A tense that is expressed synthetically, for instance, the English simple past she
sang or the French imparfait elle chantait ‘she used to sing, she was singing’. Simple tenses
contrast with compound tenses, which are formed analytically, like the English present
progressive she is singing or the French passé composé elle a chanté ‘she sang, she has sung’.
Simple tenses in one language can correspond functionally to compound tenses in another
language, and vice versa. For instance, the English simple past corresponds to a French
passé composé, whereas the English compound present progressive often corresponds to a
French simple present. 288, 307, 413
subject See Chapter 3, § 5.1 9, 413, 417
subject raising see raising
subject-aux inversion The process that forms a yes-no question from the corresponding declar-
ative sentence. In declarative sentences containing a modal, an auxiliary, or main verb be,
subject-aux inversion consists in switching the order of the subject and that element (high-
lighted by italics below).

(20) a. He should apply to both schools. Modal

b. Should he apply to both schools?
(21) a. The guy she met last night is coming along. Auxiliary be
b. Is the guy she met last night coming along?
(22) a. They do have to clean their room. Auxiliary do
b. Do they have to clean their room?
(23) a. The mail has come. Auxiliary have
b. Has the mail come?
(24) a. They are superbly qualified. Main verb be
b. Are they superbly qualified?

58
tense A linguistic category associated with temporal reference (what is the relation of the time of
the event under discussion to the time of speaking?) as well as with aspect (is the speaker’s
focus on the event’s inception, completion, duration, repetition, general truth, and so on?).
116, 414, see also finite
Chapter 15: Glossary 419

transitive Traditionally used of verbs that take a single object. We use the term in a more
general sense to refer to any syntactic category that takes a single complement. see also
intransitive
word The term ‘word’ has at least three distinct meanings.
1. An orthographic word is a particular sequence of written characters. Bank in the
sense of ‘river bank’ and ‘financial institution’ are the same orthographic word, as are
read as an infinitive (to read a book), as a past tense form (I read the book), and as a
past participle (I have read the book).
See also homograph, homophone.
2. A word form is a linguistic form of a particular grammatical type. For instance,
cut can be an infinitive (to cut), a present tense form (Whenever I stuff envelopes, I
cut myself ), a past tense form (Yesterday, I cut myself ), or a past participle (I have
cut myself again). In this case, four distinct word forms are associated with a single
orthographic word.
3. A lexeme is an abstract meaning unit that can subsume several different word forms.
For instance, the lexeme be subsumes the eight word forms am, are, be, been, being, is,
was, and were. The lexeme cut subsumes the three word forms cut, cuts, and cutting.
In lexical ambiguity, the same orthographic word is associated with more than one
lexeme, as in the case of bank ‘river bank’ and bank ‘financial institution’.
416
yes-no question See Chapter 6, § 1.1 418
References

Allen, Joseph Henry and J. B. Greenough (1903). Allen and Greenough’s new Latin grammar for
schools and colleges. Ginn and Company. Perseus acquisition number: 1999.04.0001.
uRl: https : / / www . perseus . tufts . edu / hopper / text ? doc = Perseus :
text:1999.04.0001.
Bacovcin, Hezekiah Akiva (2017). “Parameterising Germanic ditransitive variation: A historical-
comparative study.” Doctoral dissertation. University of Pennsylvania.
Bayer, Josef (1983). “Comp in Bavarian syntax.” In: The Linguistic Review 3, pp. 209–274.
Berko, Jean (1958). “The child’s learning of English morphology.” In: Word 14, pp. 150–177.
Bernstein, Judy (1993). “Topics in the syntax of nominal structure across Romance.” Doctoral
dissertation. City University of New York.
Besten, Hans den (1989). “Studies in West Germanic syntax.” Ph.D. thesis, Catholic University of
Brabant.
Bever, Thomas G. and D. Terrence Langendoen (1971). “A dynamic model of the evolution of
language.” In: Linguistic Inquiry 2, pp. 433–463.
Blake, Barry J. (2004). Case. Cambridge: Cambridge University Press.
Bobaljik, Jonathan (2002). “Why morphology does not drive syntax.” In: Journal of Comparative
Germanic Linguistics 6, pp. 129–167.
Brown, Roger (1973). A first language: The early stages. Cambridge, MA: Harvard University Press.
Chomsky, Noam (1955). The logical structure of linguistic theory. Manuscript. Published in 1975 by
Plenum Press, New York.
Chomsky, Noam (1965). Aspects of the theory of syntax. Cambridge, MA: MIT Press.
Chomsky, Noam (1970). “Remarks on nominalization.” In: Readings in English transformational
grammar. Ed. by Roderick Jacobs and Peter Rosenbaum. Waltham, MA: Blaisdell, pp. 184–
221.
Chomsky, Noam (1971). Problems of knowledge and freedom. New York: Pantheon.
Chomsky, Noam (1973). “Conditions on transformations.” In: A Festschrift for Morris Halle. Ed. by
Stephen Anderson and Paul Kiparsky. New York: Holt, Rinehart, and Winston, pp. 232–286.
Chomsky, Noam (1977). “On wh-movement.” In: Formal syntax. Ed. by Peter Culicover, Thomas
Wasow, and Adrian Akmajian. New York: Academic Press, pp. 71–132.
Chomsky, Noam (1981). Lectures in Government and Binding. Studies in generative grammar 9.
Dordrecht: Foris.
Chomsky, Noam (1986). Knowledge of language. New York: Praeger.
Chapter 15: References 421

Chomsky, Noam and George A. Miller (1963). “Introduction to the formal analysis of natural
languages.” In: Handbook of mathematical psychology. Ed. by R. Duncan Luce, Robert R.
Bush, and Eugene Galanter. Vol. 2. New York: Wiley, pp. 269–321.
Christian, Donna and Walt Wolfram (1976). Appalachian speech. Arlington, VA: Center for Applied
Linguistics.
Crain, Stephen and Mineharu Nakayama (1987). “Structure dependence in children’s language.”
In: Language 62, pp. 522–543.
Crain, Stephen and Rosalind Thornton (1998). “Investigations in Universal Grammar: a guide to
experiments on the acquisition of syntax and semantics.” In: Cambridge, MA: MIT Press.
Doherty, Cathol (1993). “The syntax of subject contact relatives.” MA thesis. Santa Cruz: Univer-
sity of California at Santa Cruz.
Dugast, Idelette (1971). Grammaire du Tunen. Vol. 8. Éditions Klincksieck.
Ecay, Aaron (2015). “A multi-step analysis of the evolution of English do-support.” Doctoral dis-
sertation. University of Pennsylvania.
Ellegård, Alvar (1953). The auxiliary do: The establishment and regulation of its use in English.
Gothenburg Studies in English. Stockholm: Almqvist and Wiksell.
Emonds, Joseph (1978). “The verbal complex V′-V in French.” In: Linguistic inquiry 9, pp. 151–175.
Engdahl, Elisabet (2006). “Information packaging in questions.” In: Empirical issues in syntax and
semantics 6 (1). Ed. by O. Bonami and P. Cabredo Hofherr. available online from Colloque
de Syntaxe et Sémantique à Paris. http://www.cssp.cnrs.fr., pp. 93–111.
Falk, Cecelia (1993). “Non-referential subjects in the history of Swedish.” Doctoral dissertation.
University of Lund.
Frisch, Stefan (1997). “The change in negation in Middle English: A NEGP licensing account.” In:
Lingua 101, pp. 21–64.
Green, Georgia (1974). Semantics and syntactic regularity. Bloomington, IN: Indiana University
Press.
Gumperz, John J. and Robert Wilson (1971). “Convergence and creolization: A case from the Indo-
Aryan/Dravidian border in India.” In: Pidginization and creolization of languages. Ed. by Dell
Hymes. Cambridge: Cambridge University Press, pp. 151–167.
Halle, Morris and Alec Marantz (1993). “Distributed morphology and the pieces of inflection.” In:
The View from Building 20. Ed. by Kenneth Hale and S.Jay Keyser. Cambridge: MIT Press,
pp. 111–176.
Harley, Heidi (2002). “Possession and the double object construction.” In: Linguistic Variation Year-
book 2, pp. 31–70.
Hellan, Lars (1988). Anaphora in Norwegian and the theory of grammar. Studies in generative
grammar 32. Dordrecht: Foris.
Henry, Alison (1995). Belfast English and standard English: Dialect variation and parameter setting.
Oxford: Oxford University Press.
Heycock, Caroline, Antonella Sorace, and Zakaris Svabo Hansen (2010). “V- to-I and V2 in subor-
dinate clauses: An investigation of Faroese in relation to Icelandic and Danish.” In: Journal
of Comparative Germanic Linguistics 13, pp. 61–97.
Heycock, Caroline, Antonella Sorace, Zakaris Svabo Hansen, et al. (2011). “Residual V-to-I in
Faroese and its lack in Danish: Detecting the final stages of a syntactic change.” In: Working
Papers in Scandinavian Syntax 87, pp. 137–165.
Chapter 15: References 422

Heycock, Caroline and John Sundquist (2017). “Don’t rush to rehabilitate: A remark on Koeneman
& Zeijlstra 2014.” In: Linguistic Inquiry 48, pp. 173–179.
Heycock, Caroline and Joel Wallenberg (2013). “How variational acquisition drives syntactic
change: The loss of verb movement in Scandinavian.” In: Journal of Comparative Germanic
Linguistics 16, pp. 127–157.
Hiaasen, Carl (1995). Stormy weather. New York: Warner.
Holmberg, Anders and Christer Platzack (1995). The role of inflection in Scandinavian syntax. New
York: Oxford University Press.
Hutchinson, Amelia P., Janet Lloyd, and Cristina Sousa (2018). Portuguese: An essential grammar.
3rd ed. London: Routledge. doi: https://doi- org.ccl.idm.oclc.org/10.
4324/9781315307190.
Jackendoff, Ray (1990). “On Larson’s treatment of the double object construction.” In: Linguistic
Inquiry 21, pp. 427–456.
Jerome, Jerome K. (1889). Three men in a boat. Reprinted 2001. New York: Tom Doherty Associates.
Joshi, Aravind, L. Levy, and M. Takahashi (1975). “Tree-adjunct grammars.” In: Journal of the
computer and system sciences 10, pp. 136–163.
Kayne, Richard (1984). Connectedness and binary branching. Dordrecht: Foris.
Kayne, Richard (1989). “Notes on English agreement.” In: CIEFL Bulletin 1, pp. 40–67.
Kendall, Tyler and Charlie Farrington (2021). The Corpus of Regional African American Language.
Eugene, OR: The Online Resources for African American Language Project. uRl: http:
//oraal.uoregon.edu/coraal.
Kerr, Elizabeth (2023). “Appendix to: On OV and VO at the Bantu/Bantoid borderlands.” In: Pro-
ceedings of ACAL 53. Ed. by Sharon Rose. Language Science Press. to appear.
Kiefer, F. and K.É. Kiss (1994). The syntactic structure of Hungarian. Vol. 27. Syntax and semantics.
Academic Press.
Kiss, K.É. (2002). The syntax of Hungarian. Cambridge Syntax Guides. Cambridge University Press.
Koopman, Hilda (1984). The Syntax of Verbs : from verb movement rules in the Kru languages to
Universal Grammar. Dordrecht, The Netherlands: Foris Publications.
Kroch, Anthony (1989). “Reflexes of grammar in patterns of language change.” In: Language vari-
ation and change 1, pp. 199–244. uRl: https://www.ling.upenn.edu/~kroch/
online.html.
Kroch, Anthony (1994). “Morphosyntactic variation.” In: Proceedings of the 30th Annual Meeting of
the Chicago Linguistics Society. Ed. by K. Beals. Vol. 2. Chicago: Chicago Linguistic Society,
pp. 180–201.
Kroch, Anthony (1998). “Amount quantification, referentiality, and long wh-movement.” In:
University of Pennsylvania Working Papers in Linguistics 5.2, p. 3. uRl: https : / /
repository.upenn.edu/pwpl/vol5/iss2/3/.
Kroch, Anthony and Cathy Small (1978). “Grammatical ideology and its effect on speech.” In:
Linguistic variation. Models and methods. Ed. by David Sankoff. New York: Academic Press,
pp. 45–55.
Kroch, Anthony and Ann Taylor (2000a). The Penn-Helsinki Parsed Corpus of Middle English. uRl:
https://www.ling.upenn.edu/ppche/ppche-release-2016/PPCME2-
RELEASE-4.
Kroch, Anthony and Ann Taylor (2000b). “Verb-object order in early Middle English.” In: Di-
achronic syntax: Models and mechanisms. Ed. by Susan Pintzuk, George Tsoulas, and An-
Chapter 15: References 423

thony Warner. Oxford: Oxford University Press, pp. 132–163. uRl: https://www.ling.
upenn.edu/~kroch/online.html.
Kroch, Anthony, Ann Taylor, and Donald Ringe (2000). “The Middle English verb-second con-
straint: A case study in language contact and language change.” In: Textual parameters in
older languages. Ed. by Susan Herring. Philadelphia: Benjamins, pp. 353–391. uRl: https:
//www.ling.upenn.edu/~kroch/online.html.
Labov, William (1972a). Language in the inner city: Studies in the Black English vernacular. 3. Uni-
versity of Pennsylvania Press.
Labov, William (1972b). “Negative attraction and negative concord in English grammar.” In: Lan-
guage, pp. 773–818.
Larson, Richard (1988). “On the double object construction.” In: Linguistic Inquiry 19, pp. 335–391.
Larson, Richard (1990). “Double objects revisited: reply to Jackendoff.” In: Linguistic Inquiry 21,
pp. 589–632.
Lees, Robert B. (1960). “The grammar of English nominalizations.” In: International Journal of
American Linguistics 26.3, p. 2.
Lippi-Green, R. (2012). English with an Accent: Language, Ideology and Discrimination in the United
States. Taylor & Francis.
Liu, Yingtong et al. (2022). “Structural, functional, and processing perspectives on linguistic island
effects.” In: Annual Review of Linguistics 8.1, pp. 495–525. doi: 10 . 1146 / annurev -
linguistics-011619-030319.
Lutz, Uli, Gereon Müller, and Arnim Stechow, eds. (2000). Wh- scope marking. Amsterdam: Ben-
jamins.
Marantz, Alec (1984). On the nature of grammatical relations. Cambridge, MA: MIT Press.
Marcus, Gary et al. (1992). “Overregularization in language acquisition.” In: Monographs of the
Society for Research in Child Development 57 (4).
McDaniel, Dana (1989). “Partial and multiple wh- movement.” In: Natural language and linguistic
theory 7, pp. 565–604.
Mossé, Fernand (1968). A handbook of Middle English. Trans. by James A. Walker. Baltimore, MD:
Johns Hopkins University Press.
Mous, Maarten (1997). “The position of the object in Tunen.” In: Object Positions in Benue-Kwa.
Ed. by Rose-Marie Déchaine and Victor Manfredi. Holland Academic Graphics, pp. 123–37.
Mous, Maarten (2003). “Nen (A44).” In: The Bantu Languages, pp. 283–306.
Mous, Maarten (2005). “The innovative character of object-verb word order in Nen (Bantu A44,
Cameroon).” In: Studies in African Comparative Linguistics with Special Focus on Bantu and
Mande. Tervuren: MRAC, pp. 411–24.
Mous, Maarten (2014). “TAM-Full Object-Verb Order in the Mbam languages of Cameroon.” In:
Preverbal Domain (s), p. 72.
Nagelhout, E. et al. (2018). Analyzing Grammar in Context. Retrieved from
https://nagelhout.faculty.unlv.edu/AGiC/. University of Nevada, Las Vegas.
Oehrle, Richard (1976). “The grammatical status of the English dative alternation.” Doctoral dis-
sertation. Massachusetts Institute of Technology.
Platzack, Christer (1988). “The emergence of a word order difference in Scandinavian subordinate
clauses.” In: McGill Working Papers in Linguistics: Special Issue on Comparative Germanic
Syntax, pp. 215–238.
Chapter 15: References 424

Radford, Andrew (1988). Transformational syntax: A first course. Cambridge textbooks in linguis-
tics. Recommended as an introduction to syntactic argumentation. The style can be grating.
Cambridge: Cambridge University Press.
Ringe, Donald and Ann Taylor (2014). The development of Old English. Oxford: Oxford University
Press.
Roberts, Ian (1993). Verbs and diachronic syntax. Dordrecht: Kluwer.
Ross, John Robert (1967). “Constraints on variables in syntax.” Doctoral dissertation. Mas-
sachusetts Institute of Technology.
Sande, Hannah, Nico Baier, and Peter Jenks (2019). “The syntactic diversity of SAuxOV in West
Africa.” In: Theory and description in African Linguistics: Selected papers from the 47th An-
nual Conference on African Linguistics. Ed. by Emily Clem, Peter Jenks, and Hannah Sande.
Berlin: Language Science Press, pp. 667–701.
Sigurðsson, Halldór (1991). “Icelandic Case-marked PRO and the licensing of lexical arguments.”
In: Natural language and linguistic theory 9, pp. 327–363.
Sobin, Nicholas (1987). “The variable status of Comp-trace phenomena.” In: Natural language and
linguistic theory 5, pp. 33–60.
Sundquist, John (2003). “The rich agreement hypothesis and Early Modern Danish embedded-
clause word order.” In: Nordic Journal of Linguistics 26, pp. 233–258.
Sutherland, Flora (2000). “Do we have or have got? Have/have got alternation.” MA thesis. Uni-
versity of Edinburgh.
Thornton, Rosalind (1995). “Referentiality and WH-movement in child English: Juvenile D-
linkuency.” In: Language acquisition 4, pp. 139–175.
Wagner, Jane (1986). The search for signs of intelligent life in the universe. New York: Harper &
Row.
Zaenen, Annie, Joan Maling, and Höskuldur Thraínsson (1985). “Case and grammatical functions:
The Icelandic passive.” In: Natural language and linguistic theory 3, pp. 441–483.
Zimmermann, Richard (2017). “Formal and quantitative approaches to the study of syntactic
change: Three case studies from the history of English.” Doctoral dissertation. University
of Geneva. uRl: https://archive-ouverte.unige.ch/unige:96500.