This book is based in large part on fieldwork that I conducted in Brittany and Wales in 1983 and 1985. I am thankful for a Fulbright Award for Research in Western Europe and a Faculty Development Award from the University of North Carolina that funded that fieldwork. lowe a less tangible, but no less real, debt to Steve Anderson, G. M. Awbery, Steve Harlow and Jim McCloskey whose work initially sparked my interest, and led me to undertake this project. I want to thank Joe Emonds and Alec Marantz who read portions of Chapter 3 and 5. I am particularly grateful though to Kathleen Flanagan, Frank Heny and two anonymous referees who read a dyslexic and schizophrenic manuscript, providing me with criticisms that improved this final version considerably. The Welsh nationalist community in Aberstwyth and its Breton coun­ terpart in Quimper helped make the time I spent in Wales and Brittany productive. I am indebted to Thomas Davies, Partick Favreau, Lukian Kergoat, Sue Rhys, John Williams and Beatrice among others for sharing their knowledge of their languages with me. Catrin Davies and Martial Menard were especially patient and helpful. Without their assistance this work would have been infinitely poorer. I am hopeful that this book will help stimulate more interest in the Celtic languages and culture, and assist, even in a small way, those in Wales and Brittany who struggle to keep their language and culture strong.

Inhalt

1 Introduction.- 1.1. The Descriptive and Theoretical Goals.- 1.2. An Overview of Government Binding Theory.- 1.3. An Overview of the Major Results of This Study.- 2 Celtic Agreement, the Avoid Pronoun Principle, and Binding Theory.- 2.1. Introduction.- 2.2. Breton Agreement Markers Determined by the Avoid Pronoun Principle.- 2.3. Generalizing the Analysis of Breton Agreement to Welsh.- 2.3.1. The Welsh Facts.- 2.3.2. The Null Subject and Prepositional Object Contain Instances of pro.- 2.3.3. The Avoid Pronoun Principle and the Distribution of Welsh NP Agreement.- 2.3.4. Welsh Pronouns with AGR and the Avoid Pronoun Analysis.- 2.3.5. Pronouns that Co-occur with Agreement in Breton.- 2.4. Evidence from the Binding Theory: Breton and Welsh Have a Null AGR.- 2.5. AGR as a SUBJECT for the Binding Theory.- 3 Raising and Passivization in Breton: An Argument for Anaphoric Traces.- 3.1. The Theoretical Status of Anaphoric Traces.- 3.2. The Breton Raising to Subject Construction.- 3.2.1. Agreement and Identifying the Breton Subject.- 3.2.2. Breton Raising to Subject Involves Anaphoric Traces.- 3.3. Raising Structures Parallel Passive Structures.- 3.4. Breton Raising and Pseudopassive: Further Implications.- 3.4.1. A Problem for Theories Lacking Anaphoric Traces.- 3.4.2. Breton Raising and Passive: Problems for Other Theories of Agreement.- 3.4.2.1. The Weakness of Case Absorption Accounts.- 3.4.2.2. Breton Agreement Is Not an Incorporated Pronoun.- 3.4.3. Reconciling Breton Passives with GB Theory: A Revision of Burzio's Generalization.- 3.5. Conclusion.- 4 PRO-INFL and Reduced Structures.- 4.1. Reduced Structures Have Missing INFLs.- 4.2. Some INFLs Missing in Welsh and English Are PRO-INFL.- 4.3. Corroborating Evidence for the PRO-INFL Analysis.- 4.3.1. PRO-INFL Only Appears When INFL Is Ungoverned.- 4.3.2. Reduced Wh-Questions Do Not Contain PRO-INFL.- 4.4. Contraction and Reduced Structures.- 4.5. A Competing Analysis.- 4.6. Breton is Consistent with the PRO-INFL Analysis.- 5 Government and the Connection Between Relative Pronouns, Complementizers and Subjacency.- 5.1. Introduction.- 5.2. Relative Pronouns in English.- 5.2.1. Relative Pronouns Are Pronominal Anaphors.- 5.2.2. Relative Pronouns Are Interpreted by Control Theory.- 5.3. Relative Pronouns Are Pronominal Anaphors.- 5.3.1. The Distribution of French Relative Pronouns.- 5.3.2. An Explanation for When S Is a Bounding Node.- 5.3.3. Infinitival Complementizers and the Head of $$ \bar{S} $$.- 5.4. Welsh and Breton Lack Relative Pronouns.- 5.4.1. The Particle A Does Not Have the Distribution of a Pronoun.- 5.4.2. Welsh and Breton Infinitival Complementizers Are Not the Head of $$ \bar{S} $$.- 5.4.3. S Is Not a Bounding Node for Subjacency in Welsh and Breton.- 5.4.3.1. Movement Structures in Welsh and Breton.- 5.4.3.2. Welsh and Breton Bounding Nodes.- 5.5. Competing Analyses and Other Arguments.- 5.6. Conclusion.- 6 The Interaction of Government Theory with Synthetic Agreement.- 6.1. Introduction.- 6.2. The ECP Gives a Unified Treatment of Complementizers and Agreement in Welsh Movement Structures.- 6.2.1. Welsh Structures Involving Short Wh-movement.- 6.2.2. Welsh Structures Involving Long Distance Wh-movement.- 6.2.3. Extraction over Uninfected Verbs in Welsh.- 6.3. Two Asymmetries in Breton and Welsh Extraction.- 6.3.1. A Difference Between Welsh and Breton Extraction over Non-Finite Verbs.- 6.3.2. Welsh and Breton Extraction Over Finite Verbs.- 6.4. Welsh and Breton Extraction from Negatives.- 6.5. Competing Analyses and Arguments.- 6.5.1. An Alternative Way of Reducing the ECP.- 6.5.2. Another Analysis of Breton Agreement and the ECP.- 6.5.3. Another Analysis of Welsh Relativization.- 6.6. Subject-Object Asymmetries at LF and the ECP.- 6.6.1. Subject-Object Asymmetries in Multiple Wh-questions.- 6.6.2. Quantification and Subject-Object Asymmetries at LF.- 6.7. Conclusion.- References.- Index of Languages.- Index of Names.