UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Underspecification, parameters, and the acquisition of vowels Fee, Elizabeth Jane 1991

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
831-UBC_1991_A1 F43.pdf [ 19.14MB ]
Metadata
JSON: 831-1.0100444.json
JSON-LD: 831-1.0100444-ld.json
RDF/XML (Pretty): 831-1.0100444-rdf.xml
RDF/JSON: 831-1.0100444-rdf.json
Turtle: 831-1.0100444-turtle.txt
N-Triples: 831-1.0100444-rdf-ntriples.txt
Original Record: 831-1.0100444-source.json
Full Text
831-1.0100444-fulltext.txt
Citation
831-1.0100444.ris

Full Text

UNDERSPECIFICATION, PARAMETERS, AND THE ACQUISITION OF VOWELS By ELIZABETH JANE FEE B.A., The U n i v e r s i t y of B r i t i s h Columbia, 1978 M.A., The U n i v e r s i t y of B r i t i s h Columbia, 1980  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of L i n g u i s t i c s )  We accept t h i s t h e s i s as conforming to the required  standard  THE UNIVERSITY OF BRITISH COLUMBIA October 1991 ( c ) E l i z a b e t h Jane Fee, 1991  In  presenting  this thesis  in partial fulfilment of  the  requirements  for  an  advanced  degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. copying  of this thesis for scholarly  department  or  by  his  or  her  I further agree that permission for  purposes  may be granted  representatives.  It  is  by the head of my  understood  that  publication of this thesis for financial gain shall not be allowed without permission.  Department of.  ilrJ^U-t^Ti  The University of British Columbia Vancouver, Canada Date  DE-6 (2/88)  C C  extensive  copying  or  my written  ABSTRACT  The  goal of t h i s t h e s i s i s to develop a p a r a m e t r i c model  of a c q u i s i t i o n which i n c o r p o r a t e s  the idea t h a t  phonological  systems are u n d e r l y i n g l y u n s p e c i f i e d f o r c e r t a i n f e a t u r e values.  I examine two  the theory  of R a d i c a l U n d e r s p e c i f i c a t i o n  Pulleyblank  1986), and  Underspecification and  v a r i a n t s of t h i s model: one  one  (Archangeli  based on the theory  ( S t e r i a d e 1987).  based  of  and  Contrastive  I assume the p r i n c i p l e s  parameters framework, where the i n i t i a l  phonological  system of the c h i l d i s assumed t o be c h a r a c t e r i z e d by unmarked parameter s e t t i n g s of UG.  on  The  two  the  types of  parameters t h a t are examined i n d e t a i l are f e a t u r a l parameters and  r u l e parameters.  The  unmarked s e t t i n g s of f e a t u r a l  parameters are s u p p l i e d by u n i v e r s a l redundancy r u l e s . most cases,  the unmarked s e t t i n g s of r u l e parameters  assumed t o be OFF, I provide and  or  analyses  In  are  non-application. of the v o c a l i c systems of Hungarian  Spanish, based on the parametric t h e o r i e s of R a d i c a l  and  C o n t r a s t i v e U n d e r s p e c i f i c a t i o n , which demonstrate t h a t c e r t a i n phonological  parameters i n these languages must be r e s e t t o  the marked o p t i o n . on spreading  The  processes,  Hungarian analyses  focus  particularly  w h i l e those i n Spanish focus  a l t e r n a t i o n s t h a t take p l a c e w i t h i n verb c o n j u g a t i o n Given the d i f f e r e n c e s between the i n i t i a l adult phonological  on classes.  c h i l d s t a t e and  systems of Hungarian and  Spanish,  the  the  u n d e r s p e c i f i c a t i o n a c q u i s i t i o n models make c e r t a i n p r e d i c t i o n s  iii  regarding a c q u i s i t i o n i n these languages.  These predictions  are then tested using data from children acquiring both Hungarian and Spanish. The early phonological systems of children acquiring Hungarian and Spanish are found to i n i t i a l l y be smaller than predicted by either a c q u i s i t i o n model.  To account for these  r e s u l t s , and s t i l l maintain a parametric model, I propose a theory of f e a t u r e a v a i l a b i l i t y , which s p e c i f i e s the order i n which features may system.  become part of a c h i l d ' s phonological  In conjuction with t h i s theory of feature  a v a i l a b i l i t y , the RU model i s able to explain the development of children's early phonological c e r t a i n s u b s t i t u t i o n patterns.  inventories, as well as The contrastive s p e c i f i c a t i o n s  required by the theory of CU cannot account for these aspects of the data.  iv  TABLE OF CONTENTS  ABSTRACT  i i  ACKNOWLEDGEMENT  viii  Chapter 1 INTRODUCTION 1.1 Overview 1.1.1 An A c q u i s i t i o n Model 1.1.2 Components o f the A c q u i s i t i o n Model 1.1.2.1 U n d e r s p e c i f i c a t i o n Theory 1.1.2.2 O r g a n i z a t i o n o f UG 1.1.2.3 C o n s t r a i n t s 1.1.2.4 The Nature o f P h o n o l o g i c a l Development 1.1.3 P r e d i c t i o n s 1.1.4 Some General F i n d i n g s 1.2 O r g a n i z a t i o n  1 4 4 7 7 9 12 13 16 17 19  Chapter 2 PARAMETRIC THEORIES OF UNDERSPECIFICATION 2.1 Aspects o f Phonology 2.1.1 L e x i c a l Phonology 2.1.1.1 The L e x i c o n 2.1.1.2 The C y c l e 2.1.1.3 P o s t - l e x i c a l Component 2.1.1.4 S t r u c t u r e P r e s e r v a t i o n 2.1.2 Representations 2.1.2.1 N o n - l i n e a r Phonology 2.1.2.2 Feature Geometry 2.1.2.3 The OCP 2.1.2.4 M e t r i c a l Theory 2.1.3 Rules 2.2 U n d e r s p e c i f i c a t i o n Theory 2.2.1 R a d i c a l U n d e r s p e c i f i c a t i o n 2.2.1.1 E l i m i n a t i o n o f Redundancy 2.2.1.2 Rule Types 2.2.2 C o n t r a s t i v e U n d e r s p e c i f i c a t i o n 2.2.2.1 S p e c i f i c a t i o n s 2.2.2.2 R-Rules 2.2.2.3 P h o n o l o g i c a l Rules 2.2.3 A l t e r n a t e T h e o r i e s of Feature Specification 2.2.3.1 Clements (1987) 2.2.3.2 P r i v a t i v e Features 2.3 P r i n c i p l e s and Parameters 2.3.1 B i n a r i t y 2.3.2 M u l t i p l e Parameters 2.3.3 Non-parametric A c q u i s i t i o n 2.4 L e a r n a b i l i t y 2.4.1 L e a r n a b i l i t y C o n d i t i o n 2.4.2 C o n t i n u i t y Assumption 2.4.3 No-negative Evidence Hypothesis /  22 25 25 26 27 30 31 33 33 36 37 39 47 50 55 55 58 62 62 64 67 68 69 70 72 74 75 77 78 79 80 81  V  2.5  Parametric Approaches t o U n d e r s p e c i f i c a t i o n 2.5.1 Parameters and R a d i c a l U n d e r s p e c i f i c a t i o n 2.5.1.1 UG i n RU 2.5.1.2 Input i n RU 2.5.2 Parameters and C o n t r a s t i v e Underspecification 2.5.2.1 UG i n CU 2.5.2.2 Input i n CU Notes t o Chapter 2 Chapter 3 THE VOCALIC SYSTEMS OF HUNGARIAN AND SPANISH 3.1 Harmony Systems i n Hungarian 3.1.1 Hungarian Vowels 3.1.2 Parametric RU A n a l y s i s of Hungarian Harmony 3.1.2.1 Back Harmony 3.1.2.2 Round Harmony 3.1.2.3 Low Front Vowel Formation 3.1.3 Parametric CU A n a l y s i s of Hungarian 3.1.3.1 Back Harmony 3.1.3.2 Round Harmony 3.1.3.3 P e r i p h e r a l Rules 3.1.4 Summary and Comparisons 3.2 Spanish V o c a l i c A l t e r n a t i o n s 3.2.1 Spanish Vowels and S t r e s s 3.2.2 Parametric RU A n a l y s i s o f Spanish Alternations 3.2.2.1 Spanish V e r b a l C l a s s e s 3.2.2.2 A l t e r n a t i n g Vowel/Diphthongs 3.2.2.3 High/Mid A l t e r n a t i o n s 3.2.2.4 P e r i p h e r a l Rules 3.2.3 Parametric CU A n a l y s i s o f Spanish Alternations 3.2.3.1 High/Mid A l t e r n a t i o n s 3.2.3.2 Vowel/Diphthong A l t e r n a t i o n s 3.2.3.3 P e r i p h e r a l Rules 3.2.4 Summary and Comparisons Notes t o Chapter 3  83 85 85 101 103 103 114 116 120 123 123 126 131 143 155 156 161 169 175 177 184 185 188 190 200 215 226 229 231 234 241 243 249  vi  Chapter 4 A PARAMETRIC ACQUISITION THEORY 4.1 Aspects o f P h o n o l o g i c a l Development 4.1.1 Speech P e r c e p t i o n 4.1.2 Speech P r o d u c t i o n 4.1.2.1 Stages of P h o n o l o g i c a l Development 4.1.2.2 Consonantal I n v e n t o r i e s 4.1.2.3 N a t u r a l Processes 4.1.2.4 E a r l y S y l l a b l e S t r u c t u r e s 4.1.3 Some General Issues 4.1.3.1 V a r i a b i l i t y 4.1.3.2 L e v e l s o f O r g a n i z a t i o n 4.1.3.3 U n i t s o f O r g a n i z a t i o n 4.1.3.4 P h o n o l o g i c a l Rules 4.1.4 I m p l i c a t i o n s f o r the A c q u i s i t i o n o f V o c a l i c Systems 4.2 Assumptions f o r a Model o f P h o n o l o g i c a l Acquisition 4.2.1 P e r c e p t i o n 4.2.2 P h o n o l o g i c a l Development 4.2.2.1 F e a t u r a l Parameters 4.2.2.2 Rule Parameters 4.2.2.3 Other Parameters 4.2.3 S u b s t i t u t i o n Processes 4.2.3.1 Paradigmatic S u b s t i t u t i o n s 4.2.3.2 Syntagmatic S u b s t i t u t i o n s 4.2.4 Summary 4.3 P r e d i c t i o n s f o r the A c q u i s i t i o n o f Hungarian and Spanish 4.3.1 RU P r e d i c t i o n s 4.3.1.1 Hungarian 4.3.1.2 Spanish 4.3.2 CU P r e d i c t i o n s 4.3.2.1 Hungarian 4.3.2.2 Spanish 4.3.3. Comparison of RU and CU P r e d i c t i o n s 4.3.3.1 Hungarian 4.3.3.2 Spanish 4.3.3.3 Summary Notes t o Chapter 4 Chapter 5 VOCALIC ACQUISITION IN HUNGARIAN AND SPANISH 5.1 Phonetic I n v e n t o r i e s and P h o n o l o g i c a l C o n t r a s t s Methodology 5.1.1 Phonetic I n v e n t o r i e s 5.1.2 S u b s t i t u t i o n s 5.1.3 P h o n o l o g i c a l I n v e n t o r i e s 5.2 V o c a l i c A c q u i s i t i o n i n Hungarian 5.2.1 Samples 5.2.2 Hungarian Vowels 5.2.3 P h o n o l o g i c a l I n v e n t o r i e s 5.2.4 Mismatches 5.2.5 S u b s t i t u t i o n P a t t e r n s 5.2.5.1 Syntagmatic S u b s t i t u t i o n s 5.2.5.2 Paradigmatic S u b s t i t u t i o n s  253 256 256 258 258 262 265 267 271 271 273 277 279 281 282 283 284 284 285 286 288 290 292 300 302 304 306 313 317 319 324 328 328 332 336 341 347 347 347 352 353 355 356 357 360 364 372 372 375  vii  5.3  V o c a l i c A c q u i s i t i o n i n Spanish 5.3.1 Samples 5.3.2 Spanish Vowels 5.3.3 P h o n o l o g i c a l I n v e n t o r i e s 5.3.4 Mismatches 5.3.5 S u b s t i t u t i o n P a t t e r n s 5.3.5.1 Syntagmatic S u b s t i t u t i o n s 5.3.5.2 Paradigmatic S u b s t i t u t i o n s 5.4 Summary Notes t o Chapter 5  377 377 379 380 383 389 389 392 392 396  Chapter 6 TESTING THE PREDICTIONS 6.1 P r e d i c t e d v s . A t t e s t e d P h o n o l o g i c a l I n v e n t o r i e s 6.1.1 The I n v e n t o r i e s 6.1.2 Gradual A c q u i s i t i o n o f Complex Vowels 6.1.3 Gradual A c q u i s i t i o n o f the Inventory 6.1.3.1 Jakobson and H a l l e (1956) 6.1.3.2 C a l a b r e s e (1988) 6.1.3.3 A Theory o f Feature A v a i l a b i l i t y 6.2 Inventories 6.2.1 Hungarian 6.2.1.1 RU Account 6.2.1.2 CU Account 6.2.2 Spanish . 6.2.2.1 RU Account 6.2.2.2 CU Account 6.3 S u b s t i t u t i o n P a t t e r n s 6.3.1 Syntagmatic S u b s t i t u t i o n s 6.3.1.1 C r o s s - l i n g u i s t i c S i m i l a r i t i e s 6.3.1.2 C r o s s - l i n g u i s t i c D i f f e r e n c e s 6.3.2 Paradigmatic S u b s t i t u t i o n s 6.3.2.1 Hungarian 6.3.2.2 Spanish 6.4 Rules 6.4.1 Hungarian 6.4.2 Spanish 6.5 Summary 6.6 Implications 6.6.1 I m p l i c a t i o n s f o r P h o n o l o g i c a l Theory 6.6.2 I m p l i c a t i o n s f o r a Theory o f P h o n o l o g i c a l Acquisition Notes t o Chapter 6  398 399 400 403 405 407 411 413 416 416 419 425 428 428 431 433 434 434 437 439 439 441 442 443 448 455 462 464  REFERENCES  475  466 473  viii ACKNOWLEDGEMENT  I would l i k e t o thank the members of my t h e s i s committee, David Ingram, P a t r i c i a Shaw and Ewa-Czaykowska-Higgins, f o r t h e i r comments, c r i t i c i s m s and encouragement throughout t h e writing of this thesis. Writing a long-distance d i s s e r t a t i o n i s never easy, but t h e f a c t t h a t t h e members o f my committee were always a v a i l a b l e has made i t l e s s d i f f i c u l t than i t c o u l d have been. David Ingram has been a f r i e n d , c o n f i d a n t and source o f never-ending support s i n c e I f i r s t began my s t u d i e s a t U.B.C. David sparked my o r i g i n a l i n t e r e s t i n language a c q u i s i t i o n , and has kept t h e f i r e s going over t h e many years I have known him. He has helped t o shape many o f the ideas p r e s e n t e d here and has read every word t h a t has gone i n t o t h i s t h e s i s , o f t e n t h r e e o r f o u r times. My i n t e r e s t i n n o n - l i n e a r phonology was i n i t i a t e d by Pat Shaw, and I have continued t o p r o f i t from h e r i n s i g h t s over the y e a r s . Pat's comments and c r i t i c i s m s o f t h e v a r i o u s d r a f t s o f t h i s t h e s i s have always been e x a c t i n g and constructive. She has been a v a l u a b l e r e s o u r c e f o r someone who w r i t e s and t h i n k s i n a r a t h e r i m p u l s i v e manner. T h i s t h e s i s c o u l d not have been completed without t h e h e l p o f Ewa Czaykowska-Higgins. Ewa, perhaps because o f a f a i r l y r e c e n t l y completed d i s s e r t a t i o n , always seemed t o know when I needed a push and when I needed a k i n d word. She has been p a r t i c u l a r l y h e l p f u l i n p o i n t i n g out where my assumptions were i n v a l i d o r unsupported, and where the a c q u i s i t i o n s e c t i o n s needed t o be expanded o r r e v i s e d so they c o u l d be understood by p h o n o l o g i s t s . I have p r o f i t e d g r e a t l y from t h e l i n g u i s t i c s community a t U.B.C. T o n i Borowsky was h e l p f u l i n much o f t h e groundwork f o r t h i s t h e s i s . M.D. Kinkade has always been a v a i l a b l e f o r a d m i n i s t r a t i v e h e l p , and i n t h e e a r l y stages o f my degree was a v a l u a b l e source o f i n f o r m a t i o n on anything t o do w i t h fieldwork. Bernard St-Jacques and Mike Rochemont p r o v i d e d u s e f u l a d v i c e on some o f t h e more b u r e a u c r a t i c a s p e c t s o f f i n i s h i n g a degree. The members of the student p o p u l a t i o n a t U.B.C. have p r o v i d e d h e l p and encouragement over t h e y e a r s . Susan Blake, Kathy Hunt, Bruce Bagemihl, David M i l l a r d , Henry Davis and N i c o l a B e s s e l l have p r o v i d e d u s e f u l d i s c u s s i o n o f many o f t h e ideas presented here, as w e l l as moral support when t h e going got rough. The p h o n o l o g i s t s a t M c G i l l , i n c l u d i n g Glyne P i g g o t t , Jose T o u r v i l l e and Dominique Rodier, were s u p p o r t i v e and f r i e n d l y t o a s t r a n g e r i n a strange l a n d . Glyne, e s p e c i a l l y , helped me t o i n t e g r a t e i n t o t h e Montreal academic community, and p r o v i d e d me w i t h a forum f o r some of work presented here. Perhaps t h e most f r u i t f u l p e r i o d of my student c a r e e r o c c u r r e d when I attended the 1987 L.S.A. Summer I n s t i t u t e a t S t a n f o r d U n i v e r s i t y . The c o n t a c t s and f r i e n d s I made t h e r e , as w e l l as the ambiance o f S t a n f o r d i t s e l f , made f o r a  ix p r o d u c t i v e and i n t e r e s t i n g summer. Courses I took from John McCarthy, John Goldsmith and Nick Clements helped me t o d e c i d e the path of my d i s s e r t a t i o n r e s e a r c h . My f a m i l y (and more r e c e n t l y my extended f a m i l y ) have been of i n v a l u a b l e support over the y e a r s . My parents i n s t i l l e d i n me the b e l i e f t h a t i t was p o s s i b l e t o do a n y t h i n g , so long as one wanted i t badly enough. The encouragement and support of my p a r e n t s , s i s t e r s , b r o t h e r s , and in-laws i s both acknowledged and a p p r e c i a t e d . My sons, P a t r i c k and C h r i s t o p h e r Lougheed, have watched t h i s t h e s i s develop and e v o l v e . They have grown up w i t h the p h y s i c a l and mental d e b r i s of academia and have had t o put up w i t h a mother who was not always t h e r e (be i t m e n t a l l y or physically). While t h e r e may have been times t h a t they d i d not understand, t h e i r presence and l o v e has always helped me t o keep a t l e a s t one f o o t i n the r e a l world. L a s t , but by no means l e a s t , I would l i k e t o thank Tom. He has been i n c r e d i b l y s u p p o r t i v e and p a t i e n t and has always been t h e r e when I needed someone t o l i s t e n , someone t o argue w i t h , or simply a shoulder t o c r y on. This thesis i s d e d i c a t e d , w i t h l o v e , t o him.  1  CHAPTER 1 Introduction  There i s an abundance of evidence i n c r o s s - l i n g u i s t i c phonological  research  t h a t not a l l segments are  s p e c i f i e d f o r a l l f e a t u r e values  underlyingly.  fully There i s  l i t t l e agreement, however, as t o e x a c t l y which f e a t u r e s may underspecified occur.  The  Archangeli  be  or under what c o n d i t i o n s u n d e r s p e c i f i c a t i o n  may  t h e o r y of R a d i c a l U n d e r s p e c i f i c a t i o n , o u t l i n e d i n and  Pulleyblank  p r e d i c t a b l e f e a t u r e values representations.  (1986), maintains t h a t o n l y are i n c l u d e d i n  Archangeli  and  Yoruba, w h i l e Abaglo and A r c h a n g e l i r a d i c a l l y underspecified  underlying  Pulleyblank  r a d i c a l u n d e r s p e c i f i c a t i o n of the f e a t u r e  non-  (1989) argue t h a t  [ATR]  i s required  in  (1989) argue t h a t a  system i s necessary i n o r d e r  to  account f o r the v o c a l i c f a c t s of Gengbe. An a l t e r n a t e theory  of u n d e r s p e c i f i c a t i o n , t h a t w i l l  be  r e f e r r e d t o here as C o n t r a s t i v e U n d e r s p e c i f i c a t i o n , i s outlined i n Steriade  (1987).  assumes t h a t u n d e r l y i n g f e a t u r e values  Contrastive  representations  Underspecification  contain only  those  t h a t are necessary to d i s t i n g u i s h p a i r s of  segments i n a g i v e n  language.  Mester and  I t o (1989) argue  t h a t c o n t r a s t i v e u n d e r s p e c i f i c a t i o n must be assumed i n t o account f o r the mimetics.  order  f a c t s of p a l a t a l prosody i n Japanese  In a r e c e n t d i s s e r t a t i o n , C a l a b r e s e (1988) examines  both types of u n d e r s p e c i f i c a t i o n and  suggests t h a t t h e r e  may  2  be a parameter which allows  languages to choose a  system t h a t i s e i t h e r r a d i c a l l y u n d e r s p e c i f i e d contrastively  phonological  or  underspecified.  Given a theory  of U n i v e r s a l Grammar (UG),  children's  l i n g u i s t i c systems are p o s s i b l e a d u l t grammars/ and therefore constrained grammars.  are  by the same s e t s of p r i n c i p l e s as  Data from f i r s t  language a c q u i s i t i o n should  adult then  a r e v e a l i n g a l t e r n a t e source of evidence f o r the v a l i d i t y l i n g u i s t i c theories. from p h o n o l o g i c a l of R a d i c a l and determine how  In t h i s t h e s i s I propose t o use  a c q u i s i t i o n to t e s t the u n d e r l y i n g  Contrastive Underspecification.  of  data premises  I attempt t o  these competing t h e o r i e s must be o r g a n i z e d  order t o r e p r e s e n t  be  coherent models of a c q u i s i t i o n , and  in  I will  examine the p r e d i c t i o n s these t h e o r i e s make f o r the a c q u i s i t i o n of the v o c a l i c systems of Hungarian and  Spanish.  These p r e d i c t i o n s are then t e s t e d u s i n g r e a l - t i m e a c q u i s i t i o n data. The  task the c h i l d faces i n a c q u i r i n g the grammar of a  language can be shown as i n  (1.1).  (1.1) Input  ;->  Grammar  The  c h i l d must determine, on the b a s i s of the i n p u t  the  form of the grammar.  the s c h e m a t i z a t i o n incomplete.  received,  When the input i s examined c l o s e l y ,  of a c q u i s i t i o n i n (1.1)  can be shown t o  White (1989), summarizing r e s e a r c h  c a r r i e d out over the past 20 y e a r s ,  t h a t has  i d e n t i f i e s three  be  been  problems  3  t h a t r e l a t e t o i n p u t : u n d e r d e t e r m i n a t i o n , degeneracy, and the No-negative Evidence Hypothesis. The problem of u n d e r d e t e r m i n a t i o n i s t h a t many a s p e c t s of grammar a r e not o v e r t l y v i s i b l e i n the i n p u t . example,  Traces, f o r  which a r e assumed t o be ah i n t e g r a l p a r t o f the  s y n t a c t i c r e p r e s e n t a t i o n of sentences, are a b s t r a c t  entities  which are not p r e s e n t i n the spoken language a c h i l d h e a r s . The degeneracy problem i s one o r i g i n a l l y addressed i n Chomsky (1965), which p o i n t s out t h a t the language c h i l d r e n hear c o n t a i n ungrammatical o r incomplete u t t e r a n c e s .  How  may  isi t  p o s s i b l e f o r a c h i l d t o determine which sentences a r e grammatical, and t h e r e f o r e generated by the grammar, and which are the r e s u l t of performance e r r o r s ?  The No-negative  Evidence problem p o i n t s out t h a t our grammatical competence a l l o w s us t o make judgments  about the u n g r a m m a t i c a l i t y of  sentences as w e l l as t h e i r g r a m m a t i c a l i t y .  I f we assume (as  i s g e n e r a l l y done i n t h i s type of r e s e a r c h ) t h a t l e a r n o n l y through p o s i t i v e evidence, how  children  can they p o s s i b l y  l e a r n t h a t c e r t a i n c o n s t r u c t i o n types are not generated by the grammar? These problems demonstrate t h a t c h i l d r e n must a t t a i n the complex  a d u l t grammar u s i n g i n p u t t h a t does not p r o v i d e a l l  the c l u e s n e c e s s a r y f o r the language b e i n g a c q u i r e d . been r e f e r r e d t o as the p r o j e c t i o n problem problem o f language a c q u i s i t i o n .  T h i s has  o r the l o g i c a l  In an attempt t o s o l v e t h e  p r o j e c t i o n problem, r e s e a r c h e r s such as Chomsky (1981a,b) have p o s i t e d a m e d i a t i n g component t o (1.1) c a l l e d  Universal  4  Grammar or  UG.  (1.2) Input  UG  >  UG  >  Grammar  i s our s p e c i e s - s p e c i f i c endowment f o r language, which  c o n s t r a i n s the conception for  form of p o s s i b l e human grammars.  of a c q u i s i t i o n i n (1.2)  has  The  l e d t o two  consequences  l i n g u i s t s interested i n a c q u i s i t i o n research.  First  has  been the c r e a t i o n of the f i e l d of l e a r n a b i l i t y , which i n v e s t i g a t e s the l o g i c a l problems r e l a t i n g t o language acquisition.  T h i s f i e l d attempts t o e x p l a i n how  g i v e n an incomplete and  imperfect  the  child,  s e t of i n p u t s t r i n g s , can  a c h i e v e the a d u l t grammar, i . e . t o s p e c i f y what form the and UG  components of  be a t t a i n a b l e .  (1.2)  input  must take i n order t h a t the grammar  Secondly, the schema i n (1.2)  has  led to  acceptance of a c q u i s i t i o n data as an important and  the  useful  source of evidence f o r the t e s t i n g of a l t e r n a t i v e l i n g u i s t i c theories.  1.1 1.1.1  Overview An A c q u i s i t i o n Model In t h i s t h e s i s I develop a model of a c q u i s i t i o n t h a t i s  based on the assumption t h a t p h o n o l o g i c a l  systems  are  underlyingly unspecified for c e r t a i n feature values. examine two  v a r i a n t s of t h i s model: one  assumption t h a t u n d e r l y i n g p r e d i c t a b l e information,  t h a t i s based on  representations  and  one  I  lack a l l  t h a t i s based on  the  the  5  assumption t h a t u n d e r l y i n g contrastive  information.  representations  lack only  non-  I assume t h a t t h i s a c q u i s i t i o n model  i s o r g a n i z e d a c c o r d i n g to the p r i n c i p l e s and  parameters model  of grammar o r i g i n a l l y proposed i n Chomsky (19813,13). variants  of the a c q u i s i t i o n model make c e r t a i n  regarding  the a c q u i s i t i o n of p h o n o l o g i c a l  t e s t these p r e d i c t i o n s u s i n g a F i n n o - U g r i c language, and o r d e r t o r e s t r i c t my  systems, and  I will  a c q u i s i t i o n data from Hungarian, Spanish, a Romance language.  In  t a s k somewhat, I have chosen t o focus  has  taken p l a c e .  g i v e n an RU  on  of work i n  I also r e s t r i c t  i n v e s t i g a t i o n s t o the types of f e a t u r a l and t h a t are r e q u i r e d  two  predictions  v o c a l i c systems, because t h i s i s where the m a j o r i t y underspecification  The  my  r u l e parameters  or CU a c q u i s i t i o n model.  While  parameters have been proposed f o r many o t h e r a s p e c t s of phonological  organization,  these are the two  sets  of  parameters t h a t are the most c r u c i a l to a t h e o r y of underspecification. The  underlying  theme of t h i s t h e s i s i s v e r y s i m i l a r t o  t h a t of C a l a b r e s e (1988). universal  filters  a c q u i s i t i o n and t h a t UG  C a l a b r e s e develops a h i e r a r c h y  t o e x p l a i n c e r t a i n f a c t s of  f o r e i g n language t r a n s f e r .  language, c h i l d r e n may  C a l a b r e s e argues  hierarchy  have a l r e a d y  objectives  constraints,  In the a c q u i s i t i o n of a  l e a r n a segment which v i o l a t e s a  c e r t a i n f i l t e r i f segments v i o l a t i n g f i l t e r s  My  phonological  c o n s i s t s of a s e t of f i l t e r s , or n e g a t i v e  which are h i e r a r c h i c a l l y ordered.  of  been  f u r t h e r down the  acquired.  d i f f e r from C a l a b r e s e ' s i n a number of  6  c r u c i a l ways.  First,  I have chosen t o use r e a l - t i m e  a c q u i s i t i o n data t o t e s t the p r e d i c t i o n s o f t h e p a r a m e t r i c t h e o r i e s o f RU and CU. linguistic  T h i s has been done so t h a t  cross-  d i f f e r e n c e s and r e l a t i o n s h i p s between a d u l t  phonological demonstrated.  systems and a c q u i s i t i o n can be c l e a r l y C a l a b r e s e attempts o n l y t o account f o r some  very g e n e r a l  aspects of t h e Jakobsonian p i c t u r e o f  phonological  development, which sometimes s u f f e r from having  been m i s i n t e r p r e t e d  over t h e years (see Ingram 1989a).  Secondly, w h i l e C a l a b r e s e attempts o n l y t o account f o r t h e development o f p h o n o l o g i c a l for  i n v e n t o r i e s , I attempt t o account  t h e types o f s u b s t i t u t i o n processes and p h o n o l o g i c a l  rules  t h a t young c h i l d r e n use, as w e l l as t h e development o f t h e i r initial  inventories.  The  a c q u i s i t i o n data t h a t a r e analyzed  come from p r e v i o u s l y p u b l i s h e d  i n this thesis  s t u d i e s o f young c h i l d r e n  a c q u i r i n g Hungarian and Spanish.  The c h o i c e o f languages  s t u d i e d was made by f o r c e r a t h e r than by c h o i c e . there  i s a large inventory  Although  of c r o s s - l i n g u i s t i c a c q u i s i t i o n  samples t o choose from, I wanted t o use o n l y languages where data from s e v e r a l d i f f e r e n t c h i l d r e n was a v a i l a b l e , i n o r d e r t o be c e r t a i n t h a t t h e samples were t r u l y r e p r e s e n t a t i v e . a l s o wanted t o choose languages f o r which u n d e r s p e c i f i c a t i o n analyses  I  previous  o f the v o c a l i c systems had been  proposed.  These two c r i t e r i a narrowed down t h e c h o i c e o f  languages  considerably.  7  1.1.2  Components o f the A c q u i s i t i o n Model I have d u a l o b j e c t i v e s i n w r i t i n g t h i s t h e s i s .  First, I  wish t o examine t h e l e a r n a b i l i t y aspects of t h e t h e o r i e s o f R a d i c a l and C o n t r a s t i v e U n d e r s p e c i f i c a t i o n , l o o k i n g a t t h e i n n a t e mechanisms t h a t must be a t t r i b u t e d t o t h e c h i l d and t h e l e a r n i n g procedures t h a t a r e r e q u i r e d t o a t t a i n t h e a d u l t language.  Secondly, I wish t o examine p h o n o l o g i c a l  cross-  l i n g u i s t i c a c q u i s i t i o n data i n order t o determine whether t h e u n d e r s p e c i f i c a t i o n analyses patterns  o f development.  which allows  can c o r r e c t l y account f o r t h e  In d e v e l o p i n g  an a c q u i s i t i o n model  me t o i n v e s t i g a t e these i s s u e s I have made  c e r t a i n assumptions r e g a r d i n g representations  1) how u n d e r l y i n g  a r e s t r u c t u r e d , 2) t h e o r g a n i z a t i o n o f UG, 3)  the c o n s t r a i n t s t h a t h o l d i n an a c q u i s i t i o n model such as t h i s , and 4) t h e nature o f p h o n o l o g i c a l  development.  Some o f  these i s s u e s a r e addressed i n the f o l l o w i n g s e c t i o n s .  1.1.2.1  U n d e r s p e c i f i c a t i o n Theory  The  a c q u i s i t i o n model developed i n t h i s t h e s i s adopts t h e  b a s i c premise t h a t a l l f e a t u r e s have b i n a r y v a l u e s , underlying  representations  lack c e r t a i n feature values.  v a r i a n t s of t h i s model a r e i n v e s t i g a t e d : theory  but that Two  one based on t h e  o f R a d i c a l U n d e r s p e c i f i c a t i o n , and t h e o t h e r based on  the t h e o r y  of Contrastive Underspecification.  Underspecification  Radical  (RU) i s based on the n o t i o n of minimal  redundancy developed i n K i p a r s k y i n d e t a i l i n Archangeli  (1982, 1985) and i s o u t l i n e d  and P u l l e y b l a n k  (1986).  In t h i s  8  theory  i t i s assumed t h a t o n l y non-redundant f e a t u r e  e x i s t i n underlying representations. f e a t u r e values  values  Redundant o r p r e d i c t a b l e  a r e i n s e r t e d by redundancy r u l e s a t some p o i n t  d u r i n g t h e l e x i c a l o r p o s t - l e x i c a l phonology. C o n t r a s t i v e U n d e r s p e c i f i c a t i o n (CU), sometimes r e f e r r e d t o as R e s t r i c t e d U n d e r s p e c i f i c a t i o n (Mester and I t o 1989), i s developed from t h e theory o f f e a t u r e s p e c i f i c a t i o n s used i n Halle  (1959) and i s o u t l i n e d i n S t e r i a d e  discussed,  (1987).  CU i s a l s o  i n a somewhat r e v i s e d form, i n t h e works o f  Clements (1988), C h r i s t d a s  (1988) and Calabrese  (1988).  S t e r i a d e ' s theory assumes t h a t o n l y c o n t r a s t i v e f e a t u r e are marked u n d e r l y i n g l y , and t h a t n o n - c o n t r a s t i v e values  values  feature  a r e i n s e r t e d l a t e i n t h e d e r i v a t i o n by redundancy  rules.  The t h e o r i e s o f u n d e r s p e c i f i c a t i o n t h a t a r e examined  here make d i f f e r e n t claims about how c h i l d r e n s p e c i f y t h e i r phonological  i n v e n t o r i e s , and about t h e r o l e t h a t redundancy  rules play i n phonological  systems.  An a l t e r n a t i v e conception underlying  o f the s p e c i f i c a t i o n o f  f e a t u r e values assumes t h a t f e a t u r e s a r e p r i v a t i v e ,  or have o n l y a s i n g l e f u n c t i o n a l v a l u e . der H u l s t Steriade  Den Dikken and van  (1990) argue t h a t a l l f e a t u r e s a r e p r i v a t i v e , w h i l e (1987), Mester and I t o (1989) and P i g g o t t  (1990, t o  appear) argue t h a t c e r t a i n f e a t u r e s a r e best viewed as b e i n g privative.  A p r i v a t i v e o r unary f e a t u r e can have o n l y one  p o s s i b l e marked value  i n any language, and t h e unmarked  value  i s never s p e c i f i e d , even a t some l a t e p o i n t i n t h e phonology. The  data presented  i n Chapter 3 f o r Hungarian p r e s e n t a  9  challenge  t o a theory which assumes t h a t a l l f e a t u r e s a r e  p r i v a t i v e , because t h e r e  i t i s argued t h a t  the l e x i c a l l y s p e c i f i e d value o f Hungarian.  [-round] must be I believe  cases l i k e t h i s , where a u n i v e r s a l l y redundant v a l u e  that  can be  shown t o be marked u n d e r l y i n g l y on a l a n g u a g e - s p e c i f i c  basis,  w i l l h e l p t o demonstrate t h a t some o r a l l d i s t i n c t i v e  features  have b i n a r y values  1.1.2.2  available for  Organization  manipulation.  of UG  I assume t h a t UG c o n s i s t s of a s e t of u n i v e r s a l p r i n c i p l e s , and a s e t o f parameters. a s p e c t s o f UG t h a t a r e h e l d constant  P r i n c i p l e s a r e those across  a l l languages,  w h i l e parameters a r e p r i n c i p l e s which have s e v e r a l s p e c i f i e d options.  S y n t a c t i c parameters have been proposed t o account  f o r d i f f e r e n c e s between languages i n the use o f "empty" subjects  (the Pro-drop parameter, see Hyams 1983 and Wexler  and M a n z i n i 1987), subject-aux i n v e r s i o n (Davis theory  (Solan  1987) and subjacency ( R i z z i 1982).  1987),  binding  Phonological  parameters have more r e c e n t l y been proposed t o account f o r d i r e c t i o n a l i t y and maximal/minimal e f f e c t s o f the f e a t u r e hierarchy  (Archangeli  and P u l l e y b l a n k  1986, P i g g o t t , t o  appear), f e a t u r e h i e r a r c h i t e c t u r e ( P i g g o t t , t o appear), branching p o s s i b i l i t i e s of s y l l a b i c constituents and  (Kaye 1987)  s t r e s s placement (Hayes 1981, Dresher and Kaye 1988). While suggestions have been made f o r parameters w i t h  m u l t i p l e s e t t i n g s (e.g. Wexler and Manzini 1987), I assume t h a t t h e parameters t h a t a r e r e q u i r e d i n an a c q u i s i t i o n model  10  based on u n d e r s p e c i f i c a t i o n theory have o n l y b i n a r y  options.  I show t h a t u n i v e r s a l redundancy r u l e s can be viewed as unmarked s e t t i n g s of f e a t u r a l parameters, although  the  languages  may  choose t o r e s e t f e a t u r a l parameters t o the marked o p t i o n .  The  unmarked s e t t i n g of f e a t u r a l parameters  initially  c o n s t r a i n the s p e c i f i c a t i o n of a l l c h i l d r e n ' s p h o n o l o g i c a l i n v e n t o r i e s , and consequently,  i n languages which r e q u i r e the  marked s e t t i n g of a f e a t u r a l parameter, the i n i t i a l hypothesized system.  system  by the c h i l d w i l l be d i f f e r e n t from the a d u l t  T h i s parametric  model can h e l p t o e x p l a i n why  a c q u i s i t i o n r e s e a r c h e r s have noted t h a t  many  cross-linguistically  c h i l d r e n ' s e a r l y p h o n o l o g i c a l systems are very s i m i l a r , o n l y l a t e r take on l a n g u a g e - p a r t i c u l a r  and  qualities.  In the p r i n c i p l e s and parameters model of grammar i t i s assumed t h a t c h i l d r e n ' s grammars may  d i f f e r from a d u l t s o n l y  i n the parameter s e t t i n g s t h a t c h a r a c t e r i z e them. then p l a c e s very severe r e s t r i c t i o n s on the types  T h i s model of  p h o n o l o g i c a l systems t h a t can be a t t r i b u t e d t o c h i l d r e n , and on the developments t h a t can occur i n the attainment a d u l t grammar. necessary  I b e l i e v e t h a t t h i s i s an important  of  the  and  c o n s t r a i n t on a c q u i s i t i o n models, because f o r f a r  too long a c q u i s i t i o n r e s e a r c h has  ignored advances i n  l i n g u i s t i c t h e o r y and has attempted to e x i s t as an  independent  field.  acquisition  I b e l i e v e t h a t t h e o r e t i c a l l i n g u i s t i c s and  r e s e a r c h can both b e n e f i t from c l o s e r t i e s between fields.  these  A c q u i s i t i o n r e s e a r c h can help t o determine which a r e  the unmarked parameter s e t t i n g s , w h i l e t h e o r e t i c a l  linguistics  11  can  p r o v i d e a model a g a i n s t which we.can t e s t  acquisition  data. In the  course of t h i s t h e s i s  I examine parameters  that  r e l a t e t o 1) redundancy r u l e s , 2) p h o n o l o g i c a l r u l e s , and complex vowels.  The  f i r s t two  types of parameters  are  d i s c u s s e d i n d e t a i l i n Chapter 2, as s p e c i f i c t o t h e o r i e s underspecification. i s not  The  i s required  l o n g vowels i n Hungarian and The  t o account f o r the  model of  acquisition  and  CU.  of  diphthongs i n Spanish.  parameters t h a t r e l a t e t o p h o n o l o g i c a l r u l e s  i d e n t i c a l i n RU  of  parameter r e l a t i n g t o complex vowels  a p a r t i c u l a r f e a t u r e of an u n d e r s p e c i f i c a t i o n  a c q u i s i t i o n , but  3)  Based on work i n c u r r e n t  phonology I assume t h a t r u l e parameters a l l o w f o r  are  non-linear the  s p r e a d i n g , d e l e t i o n or i n s e r t i o n of p h o n o l o g i c a l elements. The  unmarked s e t t i n g of a l l r u l e parameters i s OFF  application.  In t h i s t h e s i s  I d i s c u s s how  be used t o d e s c r i b e what have been c a l l e d child  or  non-  spreading r u l e s 'reduplications*  can in  language. As  stated  above, the unmarked s e t t i n g s  r u l e s are  assumed t o be p r o v i d e d by UG.  t h e r e are  two  r u l e s and  context-sensitive  possible  context-free rules  of  In the  redundancy t h e o r y of  types of redundancy r u l e s : rules.  RU  context-free  At the unmarked s e t t i n g ,  i n s e r t u n i v e r s a l l y unmarked f e a t u r e v a l u e s  r e d u n d a n t l y , w h i l e at the marked s e t t i n g they p r o v i d e u n i v e r s a l l y marked v a l u e s redundantly. setting context-sensitive  At the  the  unmarked  rules provide universal  feature  o c c u r r e n c e r e s t r i c t i o n s , w h i l e at the marked s e t t i n g a  co-  12  c o n t e x t - s e n s i t i v e r u l e i s suppressed or e l i m i n a t e d language-specific feature value  grammar, and  i n the  the u n i v e r s a l l y redundant  i s marked u n d e r l y i n g l y .  In the theory  of  CU  o n l y c o n t e x t - s e n s i t i v e redundancy r u l e s are p e r m i t t e d . unmarked s e t t i n g these r u l e s p r o v i d e occurrence r e s t r i c t i o n s  (as i n RU),  universal feature  1.1.2.3 The  co-  the  contrastive  underlyingly.  Constraints a c q u i s i t i o n model I develop i s c o n s t r a i n e d  L e a r n a b i l i t y Condition Condition  (Atkinson  (Pinker 1979,  1982,  Pinker  Evidence Hypothesis (Williams  1984), the  1984), and  1976,  the  Baker 1979,  learnability  by  the  Continuity No-negative Berwick 1985),  which are the most commonly h e l d assumptions i n the  field  of  research.  Both the L e a r n a b i l i t y and C o n t i n u i t y C o n d i t i o n s the o r g a n i z a t i o n of UG.  The  L e a r n a b i l i t y Condition  constrain says t h a t  any developmental stage p o s i t e d by an a c q u i s i t i o n t h e o r y  must  be a t t a i n a b l e v i a an a c q u i s i t o n mechanism t h a t begins w i t h and  the  w h i l e at the marked  s e t t i n g these redundancies do not apply and f e a t u r e values must be marked  At  ends a t the a d u l t grammar.  t h a t the p r i n c i p l e s and  The  Continuity Condition  a c q u i s i t i o n mechanisms t h a t  UG  says  are  a v a i l a b l e t o the c h i l d must be the same throughout the whole course of development.  These two  d i r e c t l y i n t o a p r i n c i p l e s and  c o n d i t i o n s are  built  parameters model of phonology.  Given the parameter s e t t i n g account of a c q u i s i t i o n c h i l d r e n ' s grammars can d i f f e r from a d u l t s ' o n l y i n the s e t t i n g of  13  parameters, and the o n l y type o f a c q u i s i t i o n t h a t takes  place  i s the s w i t c h i n g o f parameters. The  No-negative Evidence Hypothesis,  p o s i t i v e evidence  which says t h a t o n l y  may be used i n a c q u i s i t i o n , i s a l s o be  assumed i n t h i s t h e s i s .  In a parameter s e t t i n g model, t h i s  c o n s t r a i n t says t h a t parameter r e s e t t i n g can o n l y be t r i g g e r e d by p o s i t i v e evidence  from the i n p u t .  Given the No-negative  Evidence Hypothesis,  the u n g r a m m a t i c a l l y o f c e r t a i n  r e p r e s e n t a t i o n s o r c o n s t r u c t i o n s can never be l e a r n e d , but r a t h e r w i l l have t o be p r o v i d e d  f o r the c h i l d as a p r i n c i p l e  of UG.  1.1.2.4  The Nature o f P h o n o l o g i c a l  Development  In o r d e r t o a t t a i n the p h o n o l o g i c a l system o f a language the c h i l d must c o r r e c t l y p e r c e i v e the a d u l t  phonetic  r e p e r t o i r e , determine the a p p r o p r i a t e cues t o a t t e n d t o , and l e a r n which sounds i n the language being a c q u i r e d a r e used distinctively.  I make c e r t a i n assumptions about each o f these  s t e p s , some o f which r e s u l t from p r e v i o u s  research into  p h o n o l o g i c a l development, and some of which f o l l o w d i r e c t l y from t h e a c q u i s i t i o n theory assumed. I assume, f o l l o w i n g work i n p e r c e p t i o n r e s e a r c h , c h i d r e n a c q u i r e the phonetic  that  i n v e n t o r y of t h e i r language i n  advance o f p h o n o l o g i c a l o r g a n i z a t i o n .  In a d d i t i o n , f o l l o w i n g  previous  a c q u i s i t i o n evidence,  I assume t h a t c h i l d r e n b e g i n t o  organize  speech sounds i n t o a p h o n o l o g i c a l system sometime i n  the f i r s t h a l f of the second year.  Given the p a r a m e t r i c  model  14  of a c q u i s i t i o n , UG w i l l provide initial  phonological  system i n the form o f a s e t o f u n i v e r s a l  p r i n c i p l e s and parameters. phonological  the s t r u c t u r e f o r the c h i l d ' s  The c h i l d ' s f i r s t  attempts a t  o r g a n i z a t i o n w i l l show the e f f e c t s o f t h e  unmarked f e a t u r a l and r u l e parameters p r o v i d e d  by UG, and  p o s i t i v e evidence w i l l be r e q u i r e d t o t r i g g e r t h e r e s e t t i n g o f any  one o f these parameters. According  t o these assumptions,  children's e a r l i e s t phonological t h a t they a r e a l l c o n s t r a i n e d  cross-linguistically  systems w i l l be s i m i l a r i n  by the same s e t o f u n i v e r s a l  p r i n c i p l e s and parameters, but may d i f f e r i n t h a t these c o n s t r a i n t s may be a p p l i e d t o d i s t i n c t phonetic The  inventories.  p a r a m e t r i c a c q u i s i t i o n t h e o r i e s based on the t h e o r i e s o f  RU and CU make v e r y s i m i l a r claims s u b s t i t u t i o n patterns  about the types o f  t h a t w i l l be found i n c h i l d r e n ' s e a r l y  speech. One type o f p a t t e r n , t h a t I c a l l  paradigmatic  s u b s t i t u t i o n s , a r i s e when the unmarked parameter s e t t i n g provided  by UG f o r c e s the c h i l d t o r e p r e s e n t  sounds i n a n o n - d i s t i n c t i v e f a s h i o n .  distinctive  T h i s type o f  s u b s t i t u t i o n i s p r e d i c t e d t o occur i n languages where t h e marked s e t t i n g o f c o n t e x t - s e n s i t i v e parameters i s c a l l e d f o r . The  c h i l d ' s i n i t i a l phonological  according  t o the f e a t u r a l parameters (redundancy r u l e s ) o f UG,  w h i l e the language being  acquired  pattern of s p e c i f i c a t i o n . two  i n v e n t o r y w i l l be s p e c i f i e d  calls fora different  In t h i s case the r e p r e s e n t a t i o n s o f  sounds may be c o l l a p s e d and the unmarked sound w i l l be  15  produced i n p l a c e of a more marked one. allow  f o r c o n t e x t - s e n s i t i v e r u l e s , and  Since both RU  and  s i n c e I assume the same  b a s i c s e t of c o n t e x t - s e n s i t i v e r u l e s i n both t h e o r i e s , types of paradigmatic s u b s t i t u t i o n s they p r e d i c t identical.  D i f f e r e n t s u b s t i t u t i o n patterns  Hungarian and  CU  the  are  are p r e d i c t e d  Spanish, however, because t h e i r  for  phonetic  i n v e n t o r i e s are q u i t e d i f f e r e n t . I a l s o assume, again  f o l l o w i n g both p r e v i o u s  research  the c o n s t r a i n t s of the parameter s e t t i n g model, t h a t  and  the  s t r u c t u r e of c h i l d r e n ' s word forms i s i n i t i a l l y p r o v i d e d  by  s e t of templates.  CVCV  These templates p r o v i d e  simple CV and  s y l l a b l e s t r u c t u r e s t o which f e a t u r a l i n f o r m a t i o n mapped.  can  a  be  In c e r t a i n cases the a d u l t t a r g e t w i l l c o n t a i n more  information  than the c h i l d can r e p r e s e n t  i n the template,  and  c e r t a i n elements i n the t a r g e t w i l l be omitted i n the c h i l d ' s representation information fill  of t h a t form.  In other cases the f e a t u r a l  t h a t the c h i l d can r e p r e s e n t  does not  exhaustively  the template, and p o r t i o n s of the template w i l l  underlyingly unspecified. S a t i s f a c t i o n Condition c h i l d to provide  In t h i s case I assume t h a t  (McCarthy and  Prince  f e a t u r a l information  1986)  f o r the  the  forces  the  unspecified  s l o t s , e i t h e r by a paradigmatic s u b s t i t u t i o n p r o v i d e d through the spread of f e a t u r a l i n f o r m a t i o n  be  by UG  from another  segment or s y l l a b l e i n the word form. S u b s t i t u t i o n s t h a t r e s u l t from a p h o n o l o g i c a l filling  i n the  f e a t u r a l information  rule  of an u n d e r l y i n g  empty  s k e l e t a l s l o t are r e f e r r e d to as syntagmatic s u b s t i t u t i o n s .  or  16  Syntagmatic s u b s t i t u t i o n s then occur a c r o s s a word form, as opposed t o paradigmatic s u b s t i t u t i o n s , which a r e segment specific.  I t i s predicted  t h a t both syntagmatic and  p a r a d i g m a t i c s u b s t i t u t i o n s w i l l occur when a sound i n t h e a d u l t t a r g e t i s not p r e s e n t i n the c h i l d ' s i n v e n t o r y , t h e r e i s some complexity i n t h e a d u l t  form t h a t  i n t e r f e r e s w i t h how much f e a t u r a l i n f o r m a t i o n  o r when  somehow  t h e c h i l d can  represent.  1.1.3  Predictions In Chapter 4 i t i s shown t h a t the p a r a m e t r i c a c q u i s i t i o n  models based on t h e t h e o r i e s s i m i l a r claims regarding system.  Both t h e o r i e s ,  of RU and CU make remarkably  the c h i l d ' s i n i t i a l  f o r example, p r e d i c t t h a t  a c q u i r i n g Hungarian and Spanish w i l l d i s t i n c t vowels, which w i l l s u r f a c e [a],  phonological children  i n i t i a l l y represent 5 as [ i ] , [ u ] , [ e ] , [o] and  and t h a t they w i l l r e p r e s e n t these i n v e n t o r i e s  features  [high],  [back] and [low].  Both t h e o r i e s  t h a t c h i l d r e n a c q u i r i n g Hungarian w i l l i n i t i a l l y  using the  predict substitute  f r o n t unrounded vowels f o r f r o n t rounded vowels, a low back vowel f o r a low f r o n t vowel, and s h o r t o r simple vowels f o r complex vowels. phonological  Both t h e o r i e s  also predict s i m i l a r sets of  r u l e parameters f o r both Hungarian and Spanish,  w i t h t h e major d i f f e r e n c e being i n how these r u l e s a r e assumed t o operate. The  two v a r i a n t s of the parametric a c q u i s i t i o n t h e o r y  d i f f e r i n their predictions  regarding  the number and types o f  17  f e a t u r a l parameters t h a t must be r e s e t i n o r d e r t o a c h i e v e a d u l t systems of Hungarian and  Spanish.  RU p r e d i c t s t h a t 4  f e a t u r a l parameters must be r e s e t i n the a c q u i s t i o n Hungarian, w h i l e CU p r e d i c t s only 2. that only a s i n g l e context-free CU p r e d i c t s t h a t no  f e a t u r a l parameters w i l l be r e s e t a t a l l . phonological feature  t h e r e f o r e p r e d i c t s t h a t a g r e a t e r number of  r e s t r u c t u r i n g stages w i l l take p l a c e  i n both Hungarian  Spanish than i s p r e d i c t e d by the t h e o r y of  1.1.4  predicts  parameter w i l l be r e s e t , w h i l e  system w i l l be r e s t r u c t u r e d t o accommodate the new RU  of  In Spanish, RU  When a f e a t u r a l parameter i s r e s e t the c h i l d ' s  markings.  the  Some General  and  CU.  Findings  In both languages i t i s found t h a t a l a r g e number of mismatches t h a t occur between a c h i l d ' s form and  the  the  adult  t a r g e t are the s u b s t i t u t i o n of a s h o r t or simple vowel f o r a complex one.  The  o n l y p r e d i c t e d paradigmatic s u b s t i t u t i o n  p a t t e r n t h a t i n f a c t occurs i s the s u b s t i t u t i o n of i n Hungarian.  feature values,  in  underlyingly  so t h a t both vowels i n a  m u l t i s y l l a b i c c h i l d form are i d e n t i c a l on the s u r f a c e . f a c t t h a t Spread i s found i n the speech of c h i l d r e n Hungarian and  and  The  acquiring  Spanish, even when Spanish does not appear t o  have a p r o d u c t i v e One  /e/  In both languages a l a r g e number of mismatches  are the r e s u l t of a r u l e of Spread f i l l i n g unspecified  [a] f o r  Spreading r u l e i s an i n t e r e s t i n g f i n d i n g .  of the main d i f f e r e n c e s between the  Spanish data r e l a t e s t o the c o n d i t i o n s  Hungarian  under which  18  s u b s t i t u t i o n s occur.  In Hungarian, i t i s found t h a t  s u b s t i t u t i o n s g e n e r a l l y occur when a sound i n the a d u l t i s not present  i n the c h i l d ' s p h o n o l o g i c a l  inventory,  target  although  i n a few cases s u b s t i t u t i o n s a l s o occur when the a d u l t form i s complex i n some way, such as c o n t a i n i n g a consonant c l u s t e r o r being  trisyllabic.  In Spanish, however, s u b s t i t u t i o n s seem t o  be a f u n c t i o n o f the s t r e s s p a t t e r n s  o f the language.  Vowels  whose f e a t u r e s spread t o another element i n t h e word form a r e almost e x c l u s i v e l y i n the s t r e s s e d s y l l a b l e i n t h e a d u l t target.  I t i s hypothesized t h a t because o f the complex s t r e s s  system i n Spanish c h i l d r e n mark s t r e s s i n every form.  This  adds t o the complexity o f the c h i l d ' s form, and as a r e s u l t only the f e a t u r a l information  of the s t r e s s e d vowel o r  s y l l a b l e i s mapped t o the template. The  e a r l y phonological  systems o f c h i l d r e n a c q u i r i n g  Hungarian and Spanish a r e found t o i n i t i a l l y  be s m a l l e r  than  p r e d i c t e d by the parametric t h e o r i e s of u n d e r s p e c i f i c a t i o n . While both t h e o r i e s p r e d i c t t h a t c h i l d r e n ' s e a r l i e s t  systems  w i l l c o n t a i n 5 d i s t i n c t vowels, the data suggest t h a t c h i l d r e n o n l y g r a d u a l l y a c h i e v e a 5 vowel system.  The f i r s t vowel i s  /a/, f o l l o w e d by the mid vowels /e/ and /o/, then by / i / f i n a l l y by /u/.  and  In Hungarian the l a s t vowels added a r e t h e  f r o n t rounded vowels /u7 and /o7.  To account f o r these  r e s u l t s , and s t i l l m a i n t a i n the p a r a m e t r i c model o f acquisition, provides children.  I propose a theory  a b a s i c order T h i s theory  of f e a t u r e a v a i l a b i l i t y , which  i n which f e a t u r e s become a v a i l a b l e t o i s i n p a r t based on the t h e o r y o f  19  d i s t i n c t i v e f e a t u r e s developed i n Jakobson and H a l l e I m a i n t a i n a UG p e r s p e c t i v e  of f e a t u r e s and assume t h a t  c h i l d r e n have access t o a l l d i s t i n c t i v e f e a t u r e s the theory  (1956).  i n n a t e l y , but  of feature a v a i l a b i l t y s p e c i f i e s the order  i n which  f e a t u r e s can be used when the c h i l d begins t o o r g a n i z e phonological  a  system.  I t i s found t h a t g i v e n t h i s theory  of feature  a v a i l a b i l i t y , t h e RU model i s a b l e t o account f o r t h e development o f c h i l d r e n ' s e a r l y p h o n o l o g i c a l w h i l e t h e CU model i s not.  inventories,  The types o f s p e c i f i c a t i o n s  r e q u i r e d i n t h e CU model made i t impossible  t o p r e d i c t which  segments w i l l be added t o the c h i l d ' s i n v e n t o r y time.  a t a given  I t i s a l s o found t h a t the type o f u n d e r s p e c i f i c a t i o n  r e q u i r e d by t h e theory  o f RU i s a b l e t o e x p l a i n t h e  i d i o s y n c r a t i c behaviour o f [e] i n the Spanish data, w h i l e t h i s i s not p o s s i b l e g i v e n a c o n t r a s t i v e u n d e r s p e c i f i c a t i o n system. I t h e r e f o r e conclude t h a t an RU a c q u i s i t i o n model i s a b e t t e r representation  1.2  o f UG than the CU model.  Organization T h i s t h e s i s i s organized  i n the following fashion.  In  Chapter 2 I develop the parametric t h e o r i e s o f R a d i c a l and C o n t r a s t i v e U n d e r s p e c i f i c a t i o n , based on t h e t h e o r i e s o u t l i n e d i n Archangeli  and P u l l e y b l a n k  (1986) and S t e r i a d e  f o c u s s i n g on t h e type of i n f o r m a t i o n are p r e s e n t  i n underlying  and f e a t u r e v a l u e s  representations  redundancy and p h o n o l o g i c a l  rules.  (1987), that  and t h e form o f both  In doing so I f i r s t  20  outline some current aspects of non-linear phonology that are assumed i n the course of t h i s thesis, with a p a r t i c u l a r focus on those aspects of phonology that are assumed to be universal principles.  I then discuss some of the basic assumptions of  the p r i n c i p l e s and parameters model of grammar, and of the theory of l e a r n a b i l i t y .  F i n a l l y I examine the theories of RU  and CU as parameter setting models.  I describe how UG w i l l be  organized given each theory, and the types of input that w i l l be required to t r i g g e r the r e s e t t i n g of parameters. In Chapter 3 I present certain facts regarding the v o c a l i c systems of Hungarian and Spanish, which help to argue for a s p e c i f i c underspecification system. argumentation  This type of  i s more necessary i n the parametric theory of  RU, since t h i s theory allows for a greater number of languages p e c i f i c choices than the parametric theory of CU.  Each  analysis concludes with a summary of how the grammar would be organized, given the proposed feature s p e c i f i c a t i o n s and phonological r u l e s . Chapter 4 presents a parametric theory of phonological acquisition.  I f i r s t present some of the major findings of  previous research on phonological a c q u i s i t i o n , and then discuss the assumptions that I make regarding these findings. F i n a l l y the CU and RU predictions for the a c q u i s i t i o n of the v o c a l i c systems of Hungarian and Spanish are outlined, given the analyses developed  i n Chapter 3.  Chapter 5 presents the a c q u i s i t i o n data from Hungarian and Spanish.  I present analyses of the phonological  21  i n v e n t o r i e s used by these c h i l d r e n a t two time p e r i o d s , and the types o f s u b s t i t u t i o n p a t t e r n s t h a t occur i n t h e d a t a . In Chapter 6 I look more c l o s e l y a t the a c q u i s i t i o n  data,  w i t h an eye t o s e e i n g how s u c c e s s f u l the p a r a m e t r i c a c q u i s i t i o n t h e o r i e s o f RU and CU a r e a t c a p t u r i n g the acquisition facts.  F i r s t I examine the d i s c r e p a n c i e s t h a t  e x i s t between the i n v e n t o r i e s used by the Spanish and Hungarian c h i l d r e n and the i n v e n t o r i e s p r e d i c t e d by RU and CU, and  show t h a t a theory o f f e a t u r e a v a i l a b i l i t y can h e l p  e x p l a i n t h e e a r l y p a t t e r n s o f development.  I then attempt t o  account f o r t h e development o f these e a r l y p h o n o l o g i c a l i n v e n t o r i e s , the s u b s t i t u t i o n p a t t e r n s , and the p h o n o l o g i c a l r u l e s used by these two groups o f c h i l d r e n , g i v e n RU o r CU. L a s t l y , I d i s c u s s the i m p l i c a t i o n s these analyses  have both  f o r p h o n o l o g i c a l theory and f o r a theory o f p h o n o l o g i c a l acquisition.  22  CHAPTER 2 Parametric Theories  of U n d e r s p e c i f i c a t i o n  In t h i s chapter I examine the t h e o r i e s o f R a d i c a l and C o n t r a s t i v e U n d e r s p e c i f i c a t i o n as r e p r e s e n t a t i v e and  parameters models o f phonology.  of p r i n c i p l e s  T h i s w i l l be a w h o l l y  l o g i c a l e n t e r p r i s e ; the development o f a c q u i s i t i o n models capable o f h a n d l i n g u n t i l Chapter 4.  r e a l - t i m e a c q u i s i t i o n data w i l l be l e f t  I f i r s t o u t l i n e the two t h e o r i e s o f  u n d e r s p e c i f i c a t i o n , showing the types o f f e a t u r e s and f e a t u r e values  t h a t a r e l e f t u n s p e c i f i e d and how redundancy and  phonological  r u l e s are dealt with.  I w i l l then p r e s e n t  some o f  the b a s i c assumptions o f the p r i n c i p l e s and parameters model of phonology and of l e a r n a b i l i t y theory,  and demonstrate how  R a d i c a l and C o n t r a s t i v e U n d e r s p e c i f i c a t i o n may be r e v i s e d t o fit  i n w i t h these assumptions.  Many of these i s s u e s have  a l r e a d y been faced i n R a d i c a l U n d e r s p e c i f i c a t i o n it  (RU), s i n c e  i s p r e s e n t e d as a p r i n c i p l e s and parameters model o f  phonology i n A r c h a n g e l i Underspecification  and P u l l e y b l a n k  (1986).  Contrastive  (CU), on the other hand, has not been  developed as e x t e n s i v e l y as R a d i c a l U n d e r s p e c i f i c a t i o n , and has  not been d i s c u s s e d w i t h i n the p r i n c i p l e s and parameters  framework. In t h e 1960s and 1970s i t was assumed t h a t  language  a c q u i s i t i o n took p l a c e as the c h i l d c r e a t e d r u l e s compatible w i t h the l i n g u i s t i c data.  An " e v a l u a t i o n m e t r i c " would l e a d  23  the c h i l d t o the a p p r o p r i a t e a d u l t grammar by a l l o w i n g t h e c h i l d t o choose between competing r u l e s o r r u l e systems. t h e o r y o f UG (and i n p a r t i c u l a r t h e parametric  The  model)  g r a d u a l l y r e p l a c e d t h i s r u l e - w r i t i n g model, i n p a r t because i t was found t o be extremely d i f f i c u l t t o understand o r even c h a r a c t e r i z e the evaluation m e t r i c . 1  In t h e t h e o r y o f UG, t h e  s e t o f p o s s i b l e hypotheses t h a t a r e necessary i s assumed t o be i n n a t e .  for acquisition  T h i s type o f t h e o r y r e q u i r e s more  i n n a t e machinery than the r u l e - w r i t i n g approach, but on t h e o t h e r hand i t s e v e r e l y l i m i t s the types o f hypotheses t h a t c h i l d r e n a r e assumed t o make about language. The  parametric  model i s a s p e c i f i c theory o f UG which  assumes t h a t t h e c h i l d comes equipped t o t h e language l e a r n i n g process w i t h a s e t o f u n i v e r s a l r u l e s o r p r i n c i p l e s and a s e t of parameters, which t o g e t h e r make up UG.  Parameters  present  the c h i l d w i t h s p e c i f i c options o f a p a r t i a l l y u n s p e c i f i e d principle . 2  settings —  Parameters a r e g e n e r a l l y assumed t o have b i n a r y one achieves  marked o p t i o n .  t h e unmarked o p t i o n , t h e o t h e r t h e  Thus i n a d d i t i o n t o p r o v i d i n g t h e c h i l d w i t h a  v e r y l i m i t e d s e t o f p o s s i b l e hypotheses, t h e p a r a m e t r i c framework p r o v i d e s a theory of markedness t h a t w i l l  lead to  p r e d i c t i o n s about order o f a c q u i s i t i o n and complexity o f phonological  systems.  F a i r l y r e c e n t l y , the p r i n c i p l e s and parameters model has become w e l l - i n t e g r a t e d i n t o p h o n o l o g i c a l theory.  Phonological  parameters have been proposed t o account f o r d i r e c t i o n a l i t y o f mapping o r a s s o c i a t i o n ( A r c h a n g e l i and P u l l e y b l a n k 1989,  24  P i g g o t t , t o appear),  f e a t u r e h i e r a r c h i t e c t u r e (Davis 1990;  P i g g o t t , t o appear),  branching  constituents  possibilities  of s y l l a b i c  (Kaye 1987), s t r e s s assignment (Hayes 1981, H a l l e  and Vergnaud 1987, Dresher and Kaye 1988) and t h e u n d e r l y i n g s e l e c t i o n o f f e a t u r e values and  Pulleyblank  t h i s chapter  (e.g. A r c h a n g e l i  1988; A r c h a n g e l i  1986, 1989; Abaglo and A r c h a n g e l i  I argue t h a t i f a parametric  1989).  In  theory of  u n d e r s p e c i f i c a t i o n i s adopted, UG w i l l i n c l u d e a s e t o f f e a t u r a l parameters, which i n s e r t redundant f e a t u r e v a l u e s , and a s e t o f r u l e parameters, which p r o v i d e t h e p o s s i b l e form of p h o n o l o g i c a l r u l e s .  RU assumes t h a t UG s u p p l i e s both  c o n t e x t - f r e e and c o n t e x t - s e n s i t i v e f e a t u r a l parameters.  When  a c o n t e x t - f r e e r u l e i s r e s e t t o t h e marked o p t i o n a new r u l e ( c a l l e d a complement r u l e ) w i l l be c r e a t e d t o i n s e r t t h e opposite  f e a t u r e value p r e d i c t e d by UG.  When a c o n t e x t -  s e n s i t i v e r u l e i s r e s e t , a marked f e a t u r e s p e c i f i c a t i o n  will  be added t o t h e u n d e r l y i n g r e p r e s e n t a t i o n o f a segment, b u t the r u l e i t s e l f w i l l remain i n t h e grammar. for  CU a l l o w s  c o n t e x t - s e n s i t i v e r u l e s , and parameter s w i t c h i n g  only  will  occur and have t h e same e f f e c t s as i n RU. T h i s chapter w i l l be organized  as f o l l o w s .  In 2.1 I  d i s c u s s some g e n e r a l  issues i n current phonological  looking s p e c i f i c a l l y  a t the theory of L e x i c a l Phonology,  aspects  theory,  of p h o n o l o g i c a l and m e t r i c a l r e p r e s e n t a t i o n s , and  phonological rules.  Many o f t h e i s s u e s d i s c u s s e d i n 2.1 a r e  both c o n t r o v e r s i a l and complex, but r a t h e r than o u t l i n i n g a l l approaches and c o n t r o v e r s i e s , I w i l l d i s c u s s o n l y those  issues  25  that are c r u c i a l f o r t h i s d i s s e r t a t i o n . u n d e r s p e c i f i c a t i o n theory,  In 2.2 I look a t  f o c u s s i n g p a r t i c u l a r y on the  t h e o r i e s o f R a d i c a l and C o n t r a s t i v e U n d e r s p e c i f i c a t i o n . 2.3 I p r e s e n t  some assumptions concerning  In  the p r i n c i p l e s and  parameters model of grammar, and i n 2.4 I d i s c u s s some of the b a s i c assumptions of l e a r n a b i l i t y theory. parametric  In 2.5 the  t h e o r i e s of R a d i c a l and C o n t r a s t i v e  U n d e r s p e c i f i c a t i o n a r e o u t l i n e d by superimposing the b a s i c premises o f these two t h e o r i e s upon the p r i n c i p l e s and parameters model of phonology.  2.1 2.1.1  Aspects of P h o n o l o g i c a l  Theory  L e x i c a l Phonology In The Sound P a t t e r n of E n g l i s h ( h e n c e f o r t h SPE, Chomsky  and H a l l e 1968) the s y n t a c t i c component was assumed t o feed d i r e c t l y i n t o the p h o n o l o g i c a l component of grammar. formation processes  operated  j u n c t u r e s were r e p r e s e n t e d  i n the syntax.  Word  Morphological  by boundary symbols, and these  remained v i s i b l e t o the phonology.  L e x i c a l Phonology  developed out of the SPE framework, u s i n g i n s i g h t s from the work of Chomsky (1970) and A r o n o f f  (1976) on word  formation.  In L e x i c a l Phonology boundaries a r e encoded through s e t s o f b r a c k e t i n g s and through l e v e l s , r a t h e r than through boundary symbols.  There a r e two components t o the l e x i c o n —  l e x i c a l component and the p o s t - l e x i c a l component.  the In the  l e x i c a l component r u l e s apply o n l y t o words and/or morphemes, while  i n the p o s t - l e x i c a l component r u l e s apply t o the output  26  of the syntax, i . e . t o words, or t o l a r g e r s e t s of s t r i n g s such as phrases o r sentences.  2.1.1.1  The L e x i c o n  The work of authors such as S i e g e l and A l l e n  (1974), A r o n o f f  (1976)  (1978) demonstrated t h a t t h e r e i s o f t e n a  s i g n i f i c a n t i n t e r a c t i o n between p h o n o l o g i c a l r u l e s and morphological processes.  C e r t a i n types of m o r p h o l o g i c a l  o p e r a t i o n s appear t o t r i g g e r c e r t a i n types of p h o n o l o g i c a l r u l e s , w h i l e m o r p h o l o g i c a l o p e r a t i o n s o f t e n seem t o c l u s t e r together.  Pesetsky  (1979) t r i e d t o capture these f a c t s by  p r o p o s i n g t h a t p h o n o l o g i c a l r u l e s can apply i n s i d e the l e x i c o n , t o the output of s p e c i f i c kinds of m o r p h o l o g i c a l operations.  Groupings  of m o r p h o l o g i c a l o p e r a t i o n s a r e c a l l e d  l e v e l s o r s t r a t a , and the output of each l e v e l i s s u b j e c t t o the r u l e s o f the phonology.  Mohanan (1982) and P u l l e y b l a n k  (1986) assume t h a t t h e r e i s a s i n g l e s e t of p h o n o l o g i c a l r u l e s i n the grammar of a language,  but a s p e c i f i c r u l e may be  c o n s t r a i n e d t o a p p l y o n l y w i t h i n the l e x i c a l component, o n l y w i t h i n the p o s t - l e x i c a l component, o r w i t h i n b o t h . 3  In t h i s  model r u l e s w i l l d i s p l a y d i f f e r e n t p r o p e r t i e s depending upon where they apply i n the phonology, because of the d i f f e r e n t c o n s t r a i n t s t h a t h o l d of these two components.  27  LEXICON  (2.1)  PHONOLOGY  MORPHOLOGY  Underived  lexical  items  rule A  L e v e l 1:  *  p -rule i  (domain L)  p -rule i  (domain P)  > p- r u l e k.  .(domain L&P)  r u l e B... •r  L e v e l n:  /  rule X r u l e Y...  SYNTAX  POSTLEXICAL PHONOLOGY  2.1.1.2  The C y c l e  Underived l e x i c a l items a r e f e d i n t o t h e f i r s t  l e v e l of  the morphology where they undergo m o r p h o l o g i c a l o p e r a t i o n s . A f t e r each m o r p h o l o g i c a l o p e r a t i o n takes p l a c e these forms a r e f e d back i n t o t h e phonology, and then back i n t o t h e morphology t o undergo another m o r p h o l o g i c a l o p e r a t i o n , and so on.  Each  feed through a m o r p h o l o g i c a l process and t h e s e t o f phonological rules  i s called a cycle.  K i p a r s k y (1982) notes  t h a t many p h o n o l o g i c a l p r o c e s s e s , such as T r i s y l l a b i c  28  Shortening  (TSS), do not operate  on underived  a p p l i e s t o d e r i v e d words such as s e r e n i t y and while and  i t does not apply t o underived  ivory.  f o r these  The  C y c l e C o n d i t i o n given i n K i p a r s k y  The  TSS  invitation,  forms such as n i g h t i n g a l e  S t r i c t C y c l e C o n d i t i o n was  f a c t s (Mascar6 1976).  forms.  proposed t o account  statement of the  1982:  41)  Strict  i s reproduced i n  (2.2). (2.2)  The  S t r i c t Cycle Condition  a.  (SCC)  C y c l i c r u l e s apply o n l y t o d e r i v e d representations.  b.  A r e p r e s e n t a t i o n 4> i s d e r i v e d w i t h r e s p e c t t o r u l e R i n c y c l e j i f f 4> meets the s t r u c t u r a l a n a l y s i s of R by v i r t u e of a combination of morphemes i n t r o d u c e d i n c y c l e j or the a p p l i c a t i o n of a phonological r u l e i n cycle j .  T h i s c o n d i t i o n w i l l p r o h i b i t forms such as i v o r y from undergoing c y c l i c p h o n o l o g i c a l r u l e s such as TSS, are underived Kiparsky  forms i n the sense of  of r u l e s t h a t has been adopted as a standard  1974).  can be  (EC), a c o n d i t i o n on the  r u l e a p p l i c a t i o n (Kiparsky 1973,  they  (2.2).  (1982) then argues t h a t the SCC  from the Elsewhere C o n d i t i o n  since  derived ordering  c o n s t r a i n t on  Koutsoudas, Sanders and  Noll  29  (2.3)  The  Elsewhere C o n d i t i o n  Rules A and  (Kiparsky  1982:  136-7)  B i n the same component a p p l y  d i s j u n c t i v e l y to a form $ i f f : a.  The  s t r u c t u r a l d e s c r i p t i o n of a  rule) properly  includes  the s t r u c t u r a l  d e s c r i p t i o n of R (the g e n e r a l b.  The  r e s u l t of a p p l y i n g  rule)  a to $ i s d i s t i n c t  the r e s u l t of a p p l y i n g fi t o $. i s applied i s not  first,  and  (the s p e c i f i c  from  In t h a t case, a  i f i t takes e f f e c t , then 15  applied.  K i p a r s k y argues t h a t every underived l e x i c a l item i s i n f a c t an  i d e n t i t y r u l e , which by the EC w i l l b l o c k the  of any  other phonological  always be more s p e c i f i c ) . formulation  of the EC  r u l e ( s i n c e the  many a s p e c t s of the phonological  identity rule  With t h i s assumption, o n l y  i n (2.3)  s p e c i f i c statement of the SCC While the SCC  i s needed, o b v i a t i n g in  (or the EC)  successful  in  i n t e r a c t i o n of m o r p h o l o g i c a l  SCC.  s t r e s s assignment, and may  ( t h i s argument i s l a t e r r e f u t e d  the  and certain  K i p a r s k y (1982) argued  t h a t s t r u c t u r e - b u i l d i n g r u l e s , such as r u l e s  i n feature values,  the  capturing  p r o c e s s e s , t h e r e i s some evidence t h a t  s y l l a b i f i c a t i o n and  will  (2.2).  was  types of r u l e s do not obey the  which f i l l  application  of  redundancy r u l e s ,  apply to non-derived forms  i n K i p a r s k y 1985).  Halle  and  Mohanan (1985) argue t h a t the r u l e of f i n a l - n d e l e t i o n i n E n g l i s h , which i s hot a s t r u c t u r e - b u i l d i n g r u l e , must a l s o  be  30  non-cyclic.  This  r u l e deletes  the n i n forms l i k e columns,  hymn-book and column, y e t does not apply i n forms l i k e hymnal. The f a c t s show t h a t t h i s r u l e i s a l e x i c a l r u l e and must  apply  b e f o r e compounding and i n f l e c t i o n s , y e t i t must not a p p l y on the  same l e v e l as the - a l s u f f i x .  applies  In a d d i t i o n ,  this rule  t o underived forms such as column and hymn.  Mohanan suggest t h a t the morphology t h a t  H a l l e and  f i n a l - n d e l e t i o n i s p a r t of a l e v e l of i s non-cyclic.  K i p a r s k y (1985) i n t e r p r e t s  these f a c t s by s a y i n g t h a t e i t h e r t h e r e i s o n l y a s i n g l e l e x i c a l l e v e l i n English, the  end of L e v e l  2.1.1.3  o r t h a t b r a c k e t s are not e r a s e d a t  1.  P o s t - l e x i c a l Component  The most r e l e v a n t  c r i t e r i o n f o r d e t e r m i n i n g whether a  r u l e i s l e x i c a l o r p o s t - l e x i c a l i s the domain of a p p l i c a t i o n . S i n c e p o s t - l e x i c a l r u l e s may be f e d by s y n t a c t i c  information,  they o f t e n a p p l y t o segments a c r o s s word boundaries as w e l l as w i t h i n words.  L e x i c a l r u l e s , on the o t h e r hand, can a p p l y  o n l y w i t h i n words.  The r u l e of F l a p p i n g i n E n g l i s h  typical post-lexical rule.  is a  F l a p s occur i n s i d e words, as i n  l a d d e r ( [ l a D a r ] , but a l s o occur a c r o s s word boundaries, as i n h i t i t ! [hIDIt].  P o s t - l e x i c a l r u l e s always a p p l y i n an  across-the-board fashion,  s i n c e they cannot be s e n s i t i v e t o  l e x i c a l l y marked e x c e p t i o n s .  I t has a l s o been suggested  p o s t - l e x i c a l r u l e s may be o p t i o n a l r a t e o r s t y l e of speech (Kaisse  that  and may be s e n s i t i v e t o  and Shaw 1985).  31  2.1.1.4  Structure  Kiparsky  Preservation  (1982,1985) argues t h a t l e x i c a l r u l e s a r e  s u b j e c t t o S t r u c t u r e P r e s e r v a t i o n , while p o s t - l e x i c a l r u l e s are n o t .  S t r u c t u r e P r e s e r v a t i o n i s the c o n s t r a i n t t h a t  f e a t u r e s , f e a t u r e combinations, s y l l a b l e types o r any o t h e r s t r u c t u r e s t h a t a r e not present may not be r e f e r e n c e d  u n d e r l y i n g l y i n the language  or d e r i v e d .  In K i p a r s k y  (1985) i t i s  shown t h a t S t r u c t u r e P r e s e r v a t i o n w i l l r e s t r i c t the r u l e of V o i c i n g A s s i m i l a t i o n i n Russian t o apply t o o b s t r u e n t s ,  since  v o i c i n g i s not s p e c i f i e d on sonorants anywhere i n the l e x i c o n . Kiparsky  assumes t h a t the formal work of S t r u c t u r e  P r e s e r v a t i o n i s c a r r i e d out by marking c o n d i t i o n s o p e r a t i n g i n the l e x i c o n .  In the case of Russian V o i c i n g A s s i m i l a t i o n the  r e l e v a n t c o n s t r a i n t i s g i v e n i n (2.4) ( K i p a r s k y (2.4)  *  1985: 108).  [avoiced] [+son]  T h i s c o n s t r a i n t says t h a t n e i t h e r [+voice] nor [-voice] i s marked u n d e r l y i n g l y on sonorants i n Russian,  and t h e r e f o r e by  S t r u c t u r e P r e s e r v a t i o n n e i t h e r f e a t u r e v a l u e may be added t o the r e p r e s e n t a t i o n of a sonorant anywhere i n the l e x i c a l phonology.  T h i s prevents sonorants from being a f f e c t e d by the  r u l e of V o i c i n g A s s i m i l a t i o n , and allows V o i c i n g A s s i m i l a t i o n t o be s t a t e d without a t a r g e t c o n d i t i o n e x c l u d i n g the sonorant class. Phonetic  implementation r u l e s , such as those t h a t d e r i v e  a s p i r a t e d stops  i n E n g l i s h , are p o s t - l e x i c a l r u l e s t h a t a r e  32  non-structure  p r e s e r v i n g s i n c e they d e r i v e segments t h a t  not u n d e r l y i n g l y d i s t i n c t i v e . (2.4)  must be turned  are  In Russian the c o n s t r a i n t i n  o f f p o s t - l e x i c a l l y i n order  sonorants be p h o n e t i c a l l y r e a l i z e d as v o i c e d  that  segments.  There have been a number of arguments g i v e n i n the l i t e r a t u r e t h a t S t r u c t u r e P r e s e r v a t i o n as s t a t e d by i s too s t r o n g .  Borowsky (1986) claims t h a t  P r e s e r v a t i o n must i n f a c t be turned  Kiparsky  Structure  o f f b e f o r e the end  of  the  l e x i c o n i n E n g l i s h i n order t o account f o r the a s s i m i l a t i o n of velar nasals.  Mohanan and Mohanan (1984) argue t h a t c e r t a i n  p l a c e s of a r t i c u l a t i o n t h a t are not u n d e r l y i n g l y  distinctive  i n Malayalam must be d e r i v e d i n the course of the phonology, and  lexical  t h e r e f o r e t h a t S t r u c t u r e P r e s e r v a t i o n does not  p l a y a r o l e a t a l l i n the l e x i c o n of t h i s language.  Sproat  (1985) a l s o argues t h a t C a t a l a n v i o l a t e s S t r u c t u r e Preservation.  Archangeli  marking c o n d i t i o n s such as  and P u l l e y b l a n k  (1986) employ  (2.4), but argue t h a t i t must be  s t i p u l a t e d f o r each c o n s t r a i n t whether i t holds  i n the  component, the p o s t - l e x i c a l component or both.  In  f o l l o w i n g chapters Preservation  the  I assume the o r i g i n a l v e r s i o n of  (given by K i p a r s k y  1982,  lexical  Structure  1985), as t h e r e i s no  evidence e i t h e r i n Hungarian or Spanish t h a t an a l t e r n a t e version i s required.  I t w i l l be shown i n 3.3.1  that  Structure  P r e s e r v a t i o n can e x p l a i n n e u t r a l vowel behaviour i n the Hungarian Back Harmony system.  33  2.1.2  Representations  2.1.2.1 The  Non-linear  Phonology  focus of g e n e r a t i v e phonology, as presented  such as SPE,  was  on r u l e w r i t i n g , r u l e o r d e r i n g and  d e r i v a t i o n of l e x i c a l items.  i n works the  In the l a t e 1970s, due  t o such  works as Goldsmith's Autoseqmental Phonology (1976), t h i s focus changed t o show a g r e a t e r i n t e r e s t i n r e p r e s e n t a t i o n s . Goldsmith proposed t h a t i n d i v i d u a l f e a t u r e s belong  to  separate  t i e r s , where they can be l i n k e d by a s s o c i a t i o n l i n e s t o one more than one  segment.  or  Autosegmental r e p r e s e n t a t i o n s were  o r i g i n a l l y proposed t o account f o r tone p a t t e r n s , which t r a d i t i o n a l l y posed problems f o r standard approaches, s i n c e tones o f t e n spread a segment, o r s h i f t from one  over a domain l a r g e r than  segment t o another by  p h o n o l o g i c a l or morphological I t has  generative  operations.  s i n c e been r e c o g n i z e d t h a t many melodic f e a t u r e s ,  i n c l u d i n g n a s a l i t y , v o i c i n g , g l o t t a l i z a t i o n and p l a c e s a r t i c u l a t i o n can be autosegmentalized, independently  of other f e a t u r e s .  may  be a m o r p h o l o g i c a l  may  ' f l o a t ' or be u n a s s o c i a t e d  or g e n e r a l l y a c t  An example i s a f e a t u r e t h a t  p r o p e r t y of a r o o t  and  underlyingly  to any p a r t i c u l a r segment i n  t h a t r o o t (e.g. A r c h a n g e l i and P u l l e y b l a n k ' s of [-ATR] i n Yoruba).  of  1986  discussion  A l t e r n a t i v e l y , a segment may  be  l e x i c a l l y l i n k e d to a s i n g l e segment but a p h o n o l o g i c a l r u l e may  operate  t o add a s s o c i a t i o n l i n e s between t h i s f e a t u r e  and  o t h e r segments i n a r o o t (e.g. the Type A languages g i v e n i n Piggott  (to a p p e a r ) ) .  Both these cases are examples of  34  autosegmentalized f e a t u r e s ; f e a t u r e s which behave i n d e p e n d e n t l y o f the remainder of the melodic r e p r e s e n t a t i o n . Goldsmith  (1976), i n h i s treatment o f tone, proposed  that  tones a r e always u n d e r l y i n g l y u n a s s o c i a t e d , and t h a t they become anchored  segmentally by means o f the WeiIformedness  Condition. (2.5)  Well-formedness  C o n d i t i o n (Goldsmith 1976: 27)  a.  A l l vowels a r e a s s o c i a t e d w i t h a t l e a s t one tone.  b.  A l l tones a r e a s s o c i a t e d w i t h a t l e a s t one vowel.  c.  A s s o c i a t i o n l i n e s do not c r o s s .  These conventions allowed more than one tone t o be l i n k e d t o a s i n g l e vowel t o c r e a t e contour tones. Ford  Work by Clements and  (1979) and H a l l e and Vergnaud (1982) demonstrated  that  (2.5) i s t o o weak as i t p r e d i c t s t h a t when t h e r e i s a mismatch between the number of tones and the number o f vowels, options are possible.  In f a c t , i n a g i v e n language,  several there i s  g e n e r a l l y o n l y one method of r e s o l v i n g the mismatch. and Ford argue t h a t  Clements  (2.5) i s a l s o too s t r o n g because i t  p r e d i c t s t h a t tones may be m u l t i p l y l i n k e d , which i s not t h e case i n a l l languages.  H a l l e and Vergnaud (1982) then  propose  t h a t a s e t o f a s s o c i a t i o n conventions a p p l y o n l y t o f r e e o r f l o a t i n g tones.  These conventions as adopted by P u l l e y b l a n k  (1986: 11) a r e g i v e n i n (2.6).  35  (2.6)  A s s o c i a t i o n Conventions Map  a sequence of tones onto a sequence of tone-  bearing units, a. from l e f t t o r i g h t b. i n a one-to-one r e l a t i o n Well-formedness  Condition:  A s s o c i a t i o n l i n e s do not c r o s s .  These conventions do not a l l o w f o r the mapping o f more than one tone t o a g i v e n segment.  M u l t i p l e l i n k i n g s of a s i n g l e  tone t o more than one t o n e - b e a r i n g u n i t o r of one t o n e - b e a r i n g u n i t t o more than a s i n g l e tone are accomplished by  language-  s p e c i f i c r u l e , a c c o r d i n g t o P u l l e y b l a n k (1986). Autosegmentalized f e a t u r e s of any s o r t are assumed t o be l i n k e d t o r e p r e s e n t a t i o n s by a s e t o f A s s o c i a t i o n such as those i n (2.6)  Conventions  (e.g. see P i g g o t t , t o a p p e a r ) , a l t h o u g h  i n A r c h a n g e l i and P u l l e y b l a n k (1989) and L i e b e r (1987) i t i s c l a i m e d t h a t the d i r e c t i o n of i n i t i a l a s s o c i a t i o n i s parameterizable.  In the unmarked case a s s o c i a t i o n  proceeds  from l e f t t o r i g h t , but i n marked cases, such as the a s s o c i a t i o n of Yoruba [ATR], a s s o c i a t i o n operates from r i g h t to  left.  In Chapters 3 and 5 I w i l l  adopt the v e r s i o n o f the  A s s o c i a t i o n Conventions g i v e n i n (2.6), w i t h the added assumption t h a t the i n i t i a l d i r e c t i o n of mapping i s parameterizable.  36  2.1.2.2  Feature Geometry  I t was r e a l i z e d as e a r l y as Jakobson and H a l l e  (1956)  t h a t segments do not c o n s i s t o f bundles o f t o t a l l y unorganized f e a t u r e s , but i t was not u n t i l Mohanan (1983) and Clements (1985) t h a t the f i r s t models o f f e a t u r e geometry were proposed.  F e a t u r e geometries attempt t o account f o r  r e s t r i c t i o n s on how f e a t u r e s i n t e r a c t i n human languages.  The  geometry i n (2.7) i s one p o s s i b l e model o f f e a t u r e h i e r a r c h i t e c t u r e t h a t i s c o n s i s t e n t w i t h r e c e n t work i n t h i s area (2.7)  (cf.  Clements 1985, Sagey 1 9 8 6 ) . 4  F e a t u r e Geometry x Skeleton Root Node [nasal] Laryngeal Node [voice; [spr.gl [CG]  v  Supralaryngeal  Node  [cont] [cons]  [lateral]  Dorsal Node  L a b i a l Node [round]  Tback] [high]  [distr]  37  There i s a d i s t i n c t i o n i n (2.7)  between nodes and  features.  Nodes (given as " o " ) dominate f e a t u r e s or nodes, w h i l e features  (given i n [ ] brackets)  the t r e e .  are the t e r m i n a l elements i n  Three a r t i c u l a t o r nodes -- L a b i a l , Coronal  Dorsal —  and  are immediately dominated by the P l a c e node.  vowel f e a t u r e s of [back], [high] and  [low] are dominated by  the D o r s a l node, which i s the node a l s o used t o v e l a r and u v u l a r consonants. [ a n t e r i o r ] and  The  The Coronal  represent  node dominates  [ d i s t r i b u t e d ] w h i l e the L a b i a l node dominates  the f e a t u r e [round], a f o u r t h f e a t u r e used i n the d e s c r i p t i o n of v o c a l i c  systems.  In A r c h a n g e l i and P u l l e y b l a n k  (1986) a f e a t u r e geometry  i s proposed i n which a separate node, the Secondary P l a c e Node, dominates a l l vowel f e a t u r e s . capture  T h i s node was  meant t o  the f a c t t h a t harmony r u l e s are almost e x c l u s i v e l y  t r i g g e r e d by vowels and have vowels as t h e i r t a r g e t s .  I t has  s i n c e been argued t h a t other aspects of r e p r e s e n t a t i o n s r u l e types can account f o r these  and  facts, while incorporating  the model i n (2 . 7 ) . 3  2.1.2.3 The  The  OCP  O b l i g a t o r y Contour P r i n c i p l e  (OCP)  was  originally  proposed by Leben (1973) to account f o r the d i s t r i b u t i o n of s u r f a c e tones i n Mende, a Niger-Congo language. t r i s y l l a b i c nouns i n Mende the t o n a l p a t t e r n s LLH not occur.  Leben used the OCP  In and HHL  to p r o h i b i t i d e n t i c a l  do  adjacent  38  tones,  so t h a t i f a t r i s y l l a b i c  autosegments, o n l y HLH  and LHL  Other p a t t e r n s , such as HLL  noun has t h r e e t o n a l p a t t e r n s w i l l be  and LHH  permitted.  w i l l be d e r i v e d from  u n d e r l y i n g sequences of HL or LH tones, w i t h s p r e a d i n g second u n d e r l y i n g tone to the f i n a l s y l l a b l e . OCP  and p r i n c i p l e s of spreading  systematic  gap  In t h i s way  was  a p p l i e d t o segmental l a c k of  McCarthy's g e n e r a l i z e d statement  of the OCP  i s g i v e n i n (2.8).  (2.8)  (McCarthy 1986:  OCP  the  language.  p a t t e r n s i n A r a b i c , t o account f o r the s y s t e m a t i c r o o t s of the form /CAC^C-,/.  the  are a b l e to account f o r the  i n the t o n a l p a t t e r n s of t h i s  In McCarthy (1979) the OCP  of  208)  At the melodic l e v e l , adjacent  i d e n t i c a l elements are  prohibited.  McCarthy (1986) s y s t e m a t i c a l l y examines evidence counterexamples a g a i n s t the OCP  i n many d i f f e r e n t  f o r and languages.  In A f a r , f o r example, a r u l e of syncope a p p l i e s t o forms i n (2.9a) but  f a i l s t o apply i n the forms i n (2.9b) where the  a p p l i c a t i o n of syncope would r e s u l t i n two consonants.  identical  adjacent  39  (2.9)  Afar  (McCarthy 1986: 220) digib-t-e  digb-e  'she/I  married'  me*i e r - t a  meSr-a  •you/he k i l l s a c a l f  wager-n-e  wagr-e  'we r e c o n c i l e d / h e r e c o n c i l e d  adad-e  *add-e  'I/he  danan-e  *dann-e  'I/he was h u r t '  xarar-e  *xarr-e  'he burned'  McCarthy concludes  trembled'  t h a t i n the v a s t m a j o r i t y o f languages, t h e  OCP appears t o h o l d of segments a t the l e x i c a l l e v e l , but does not n e c e s s a r i l y h o l d a t the l e v e l o f phonetic  implementation.  He suggests t h a t i f c o n v i n c i n g examples of the OCP not h o l d i n g a t t h e l e x i c a l l e v e l can be found, then the OCP c o u l d be regarded  as a p a r a m e t e r i z a b l e  p r i n c i p l e o f UG, but one t h a t i n  the unmarked case holds of a l l non-prosodic i n f o r m a t i o n i n t h e lexicon  (but see P a r a d i s and Prunet 1990 f o r a c o n v i n c i n g  argument t h a t perhaps the OCP does not have options).  Recently  parametric  i t has been suggested t h a t t h e OCP  holds  not j u s t o f e n t i r e segments, but of i n d i v i d u a l f e a t u r e s ( e . g . A r c h a n g e l i and P u l l e y b l a n k 1986, Mester t o appear).  2.1.2.4  M e t r i c a l Theory  In g e n e r a t i v e phonology s y l l a b l e s were not c o n s i d e r e d t o be p r i m i t i v e s o f the theory, and c o u l d o n l y be r e f e r e n c e d as some sequence of consonants and vowels.  In Kahn (1976) i t i s  argued t h a t s y l l a b l e s a r e c o n s t i t u e n t s of p h o n o l o g i c a l r e p r e s e n t a t i o n s , represented  as a separate  t i e r dominating  40  skeletal  slots.  There have been many subsequent e l a b o r a t i o n s  of the n o n - l i n e a r approach to s y l l a b l e s t r u c t u r e , the most common of which are the r u l e - b a s e d S t e r i a d e 1982, 1978,  approach (Kahn  L e v i n 1985), the templatic-approach  H a l l e and Vergnaud 1978)  (Kaye and Lowenstamm 1984).  1976, (Selkirk  and the government approach  Here I o u t l i n e the  rule-based  approach of L e v i n (1985), and t h i s framework w i l l be i n Chapter 3 i n the a n a l y s i s of Spanish Levin's and  employed  syllable structure.  framework focusses p a r t i c u l a r l y on the u n i v e r a l  l a n g u a g e - s p e c i f i c aspects  of a theory of  syllabicity.  S y l l a b i c c o n s t i t u e n t s are generated by a v e r s i o n of X-bar theory, where each s y l l a b l e c o n t a i n s one The  and  o n l y one  head.  u n i v e r s a l components of t h i s theory are g i v e n i n (2.10)  ( L e v i n 1985: (2.10)  A.  12). X-bar theory i.  ii.  iii.  iv.  C a t e g o r i a l Component a.  N-Placement  b.  Complex-N  Projection a.  Project  N"  b.  P r o j e c t N'  Incorporation a.  Incorporate  i n t o N"  b.  Incorporate  i n t o N'  Adjunction  (to  N")  B.  C o n d i t i o n on Structure-Dependent Rules  C.  Sonority  Hierarchy  41  N-Placement i s the process which determines the s y l l a b l e head. N-Placement may be marked i n the l e x i c o n o r may be determined by a redundancy o r p h o n o l o g i c a l r u l e .  L e v i n assumes t h a t  [ s y l l a b i c ] i s not an o p e r a t i v e f e a t u r e , but r a t h e r t h a t i n many languages the category N i s e r e c t e d by r u l e over segments s p e c i f i e d as [-consonantal] and/or [-high].  Complex Ns may  e x i s t , c o n t r o l l e d by a parameter a s s o c i a t e d w i t h the Complex-N process  o f the c a t e g o r i a l component.  Although L e v i n does not  s p e c i f i c a l l y address the i s s u e o f markedness i n complex Ns, I assume t h a t U n i v e r s a l Grammar i n i t i a l l y t e l l s the c h i l d Ns may not branch, but t h a t based on p o s i t i v e evidence  that from  the i n p u t the c h i l d may switch t h i s parameter t o a l l o w a s i n g l e N t o dominate two s k e l e t a l s l o t s  ( t h i s aspect o f  L e v i n ' s t h e o r y w i l l be d i s c u s s e d i n Chapter 5 ) . (2.10) w i l l apply c y c l i c a l l y ,  The r u l e s i n  so t h a t i f an e p e n t h e t i c vowel  i s i n s e r t e d a f t e r the i n i t i a l a p p l i c a t i o n o f N-Placement, i t w i l l r e c e i v e a d e s i g n a t i o n as a nucleus  on a subsequent c y c l e .  A f t e r N i s found N" i s p r o j e c t e d by p i c k i n g up segments immediately t o the l e f t of N. and  N" i s the maximal p r o j e c t i o n ,  t h e r e f o r e the s y l l a b l e node i t s e l f .  The f i n a l p r o j e c t i o n  i s N', and t h i s p r o j e c t i o n i s p a r t i c u l a r l y i n f l u e n c e d by language-specific information. remaining  post-nuclear  segments.  P r o j e c t N' w i l l p i c k up any I n c o r p o r a t i o n i s a process  which then allows a d d i t i o n a l s k e l e t a l s l o t s t o be i n c o r p o r a t e d under N" o r N*. onsets  T h i s process w i l l a l l o w f o r both complex  and codas on a l a n g u a g e - p a r t i c u l a r b a s i s .  The  42  o p e r a t i o n of N-placement, P r o j e c t N" and N' and N" I n c o r p o r a t i o n are shown i n (2.11) f o r the E n g l i s h word (2.11) a.  N-Placement N  I X  X  X  Redundancy  |  rule inserts  [-cons]  [-cons]  b.  X  t  r  i  p  P r o j e c t N"  X  X  X  X  I [-cons] t  r  i  p  trip.  43  c.  d.  P r o j e c t N'  Incorporation  [-cons] t  Each o f the o p e r a t i o n s  r  i  p  o f N-Placement/  P r o j e c t N" and  P r o j e c t N' must operate w i t h i n the c o n s t r a i n t s of the S o n o r i t y Hierarchy,  as shown i n (2.IOC).  Levin posits a universal  h i e r a r c h y of f e a t u r e s which determine s o n o r i t y r a n k i n g , serves as the base from which l a n g u a g e - p a r t i c u l a r  which  sonority  44  s c a l e s are developed.  I f a given s t r i n g cannot  e x h a u s t i v e l y s y l l a b i f i e d i n adherence w i t h the  be language-  s p e c i f i c s o n o r i t y s c a l e , then s e v e r a l p o s s i b l e o p t i o n s a v a i l a b l e t o r e s o l v e t h i s problem.  The  are  most common o p t i o n , i s  t h a t a s k e l e t a l s l o t , which can then undergo the r u l e of Placement, i s i n s e r t e d o r epenthesized representation. of a new  i n t o the  segment w i l l then serve as the head  s y l l a b l e to which the p r e v i o u s l y u n s y l l a b i f i a b l e  segment may Zee  T h i s new  be  associated.  (1988) d i s c u s s e s such a case of Epenthesis  Bulgarian.  In B u l g a r i a n l i q u i d s are not a b l e t o a c t  n u c l e i , and  t h e r e f o r e a form such as /grk/  undergo epenthesis (2.12)  N-  as shown i n  Underlying Epenthesis  in as  'Greek' w i l l  (2.12). /grk/ x  x  g  r  Syllabification  x  x k  N"  X  X  X  X  [-cons] g  r  k  [grak]  'Greek'  45  Within  t h i s view Epenthesis i s seen as an automatic r e s u l t of  the s y l l a b i f i c a t i o n algorithms  i n a language, and  r e q u i r e t h a t such r u l e s be l a n g u a g e - s p e c i f i c r u l e s such as r u l e s of spreading  does not  phonological  or d e l i n k i n g (to be  discussed  i n 2.1.3). The  past decade has  of m e t r i c a l s t r e s s . m e t r i c a l theory  a l s o seen much r e s e a r c h  Liberman and  Prince  into a  (1975) propose a  i n which s t r e s s i s determined by the  prominence of s y l l a b l e s , determined through b i n a r y t r e e s t r u c t u r e s e r e c t e d over s y l l a b l e s . i s l a b e l l e d as e i t h e r w  t o words of more than two  Hayes (1981) r e v i s e s and arguing  t h a t a theory  branching  expands on t h i s theory  tree  each b i n a r y  Both t e r m i n a l and  t e r m i n a l c o n s t i t u e n t s are l a b e l l e d f o r prominence, s t r e s s t o be assigned  relative  Each node i n the  (weak) or s ( s t r o n g ) , and  branching tree i s c a l l e d a foot.  theory  nonallowing  syllables. of s t r e s s ,  of m e t r i c a l t r e e s t r u c t u r e and  a  small  number of parameters can account f o r s t r e s s assignment i n the world's languages. given  Hayes' p r i n c i p l e s of t r e e c o n s t r u c t i o n  i n (2.13) (Hayes 1981:  48).  are  46  (2.13) a.  Tree  Construction  P r o j e c t rimes.  O p t i o n a l l y form a s u b p r o j e c t i o n  [ + s y l l a b i c ] segments w i t h i n the  of  rime.  b.  S e l e c t e i t h e r r i g h t or l e f t nodes as dominant.  c.  Form the l a r g e s t p o s s i b l e b i n a r y b r a n c h i n g t r e e , such t h a t r e c e s s i v e nodes do not branch.  Optionally, i t  may  be s p e c i f i e d t h a t  The  i.  A l l t e r m i n a l nodes are counted as  ii.  Dominant nodes must be  iii.  Dominant nodes must branch.  non-branching.  terminal.  f i r s t p r i n c i p l e i n (2.13a) says t h a t rhymes form  i n i t i a l p r o j e c t i o n on which t r e e s are e r e c t e d , t h a t t h i s p r o j e c t i o n i s the nucleus. predates L e v i n  the  or o p t i o n a l l y  Since Hayes (1981)  (1985) I assume t h a t i n a L e v i n - s t y l e a n a l y s i s  (2.13a) would say  ' p r o j e c t N',  o p t i o n a l l y p r o j e c t N.  (2.13b) p r o v i d e s  the c h o i c e of whether the  right-dominant.  By  '  foot i s l e f t  (2.13c) f o o t c o n s t r u c t i o n  or  continues,  e r e c t i n g f e e t over rhymes or s y l l a b l e s i n accordance w i t h (2.13b), where nodes marked w never branch.  The  parameters i n  (2.13c) i n the unmarked case are OFF,  and  are i n e f f e c t .  i s chosen f o o t  I f Parameter (2.13ci)  i n the marked case  c o n s t r u c t i o n w i l l be q u a n t i t y s e n s i t i v e , i g n o r i n g b r a n c h i n g w i t h i n the rhyme or nucleus.  I f ( 2 . 1 3 c i i ) i s chosen, then  f o o t c o n s t r u c t i o n i s maximally b i n a r y r a t h e r than unbounded, and  i f ( 2 . 1 3 c i i i ) i s chosen then nodes are l a b e l l e d s i f and  o n l y i f they branch.  Following  (2.13) s t r e s s w i l l be  assigned  47  to  the form merepet from Maranungku u s i n g the parameters g i v e n  in  (2.14a and b)(Hayes 1981:  (2.14)  a.  51).  Going from l e f t t o r i g h t , c o n s t r u c t b i n a r y , q u a n t i t y i n s e n s i t i v e , l e f t dominant  b.  feet.  Group the f e e t i n t o a l e f t dominant word t r e e .  c.  merepet  The most p r e v a l e n t competing theory of s t r e s s assignment i s g r i d t h e o r y , where the branching s t r u c t u r e s of t r e e t h e o r y are not r e c o g n i z e d of  ( P r i n c e 1983).  In the a n a l y s i s  s t r e s s assignment i n Spanish g i v e n i n Chapter  adopt the Hayesian  Hayesian  3 I will  framework, although I b e l i e v e n o t h i n g  hinges on t h i s p a r t i c u l a r c h o i c e of t h e o r y .  2.1.3  Rules N o n - l i n e a r phonology and t h e o r i e s of f e a t u r e geometry  have l e d t o a much c l e a r e r view of how r e p r e s e n t a t i o n s are s t r u c t u r e d .  phonological  T h i s , along w i t h the move i n  syntax t o l i m i t r u l e types, has l e d p h o n o l o g i s t s t o a c o n s t r a i n e d view of the number and types of p h o n o l o g i c a l r u l e s t h a t may  occur i n any language.  P i g g o t t (to appear),  example, a l l o w s f o r the r u l e s shown i n (2.15).  for  48  (2.15)  Phonological a.  Spread a  Rules A  B  a  b.  Delink a  A  B  II  a  d.  Insert a  A  A  I  A  "  > I  a.,  B  a  ->  B  I  1 / : )  A  B  I  I  a*  Each o f these r u l e s i s a simple o p e r a t i o n which adds o r d e l e t e s some p a r t of the p h o n o l o g i c a l r e p r e s e n t a t i o n .  Spread  a i s t h e r u l e i n v o l v e d i n most a s s i m i l a t i o n p r o c e s s e s ,  adding  an a s s o c i a t i o n l i n e between a f e a t u r e a l r e a d y anchored and a new segment.  D e l i n k a i s the d i s s i m i l a t i o n r u l e which d e l e t e s  some element due t o the presence of a second element. i s a process  Fusion  t h a t i s not unique t o P i g g o t t ' s work, although i t  i s f i r s t d i s c u s s e d as a r u l e o f phonology i n P i g g o t t ( t o appear).  F u s i o n c r e a t e s a double l i n k i n g between an element  ( f e a t u r e , s k e l e t a l s l o t , e t c . ) on one t i e r and two elements on  49  another t i e r .  I n s e r t a i s the process which accounts f o r the  i n s e r t i o n of a f e a t u r e  (or f e a t u r e s )  i n t o the melodic  component. Although P i g g o t t views F u s i o n as a s e p a r a t e and r u l e from those i n (2.15a/ b and Fusion  d ) , i t i s p o s s i b l e t o view  as the i n s e r t i o n of an a s s o c i a t i o n l i n e .  a n a l y s i s of Spanish and  distinct  In  the  Hungarian presented i n Chapter 3 I  f i n d no evidence f o r a separate r u l e of F u s i o n  and  c o n s e q u e n t l y I assume t h a t t h i s r u l e i s simply  one  possible  i n s t a n t i a t i o n of the r u l e I n s e r t a. Piggott  (to appear) assumes t h a t the o p e r a t i o n  of any  one  of the r u l e s i n (2.15) i s r e g u l a t e d by a D i r e c t i o n a l i t y Parameter which says t h a t r u l e s may or i n a b i d i r e c t i o n a l fashion.  spread l e f t w a r d ,  rightward  I assume t h a t the unmarked  s e t t i n g of a D i r e c t i o n a l i t y parameter i s always the same d i r e c t i o n as the i n i t i a l mapping r e g u l a t e d by the  Association  Conventions. A more p r e c i s e statement of how depend upon the theory 2.2.1  - 2.2.3).  of f e a t u r e s p e c i f i c a t i o n assumed  of C o n t r a s t i v e U n d e r s p e c i f i c a t i o n  in Radical Underspecification. may  be u n d e r l y i n g l y s p e c i f i e d , i n which case i f  r a d i c a l model, o n l y a s i n g l e f e a t u r e value w i l l be  for  a.  than  In the c o n t r a s t i v e model, both  a spreads t o ft i t w i l l a l s o have to d e l i n k a from  underlyingly,  (see  For example, Spread a w i l l behave v e r y  d i f f e r e n t l y i n a theory  f e a t u r e values  these r u l e s operate w i l l  ft.  In  the  present  so a w i l l o n l y spread to ft i f ft i s not s p e c i f i e d  A theory which assumes t h a t f e a t u r e s are p r i v a t i v e  50  ( i . e . have o n l y a s i n g l e f u n c t i o n a l value) w i l l assume t h a t Spread a f u n c t i o n s as i n R a d i c a l  2.2  Underspecification.  U n d e r s p e c i f i c a t i o n Theory U n d e r s p e c i f i c a t i o n has  been a p a r t of p h o n o l o g i c a l  s i n c e the days of the Prague School l i n g u i s t s . (1969) archiphonemes are elements represented  theory  In Trubetzkoy by the s e t of  f e a t u r e s t h a t are common t o a group of phonemes i f those phonemes are n e u t r a l i z e d i n some context. those f e a t u r e values  Archiphonemes l a c k  t h a t are i n v o l v e d i n the n e u t r a l i z a t i o n  process.  Trubetzkoy d i s c u s s e s  the case of  devoicing  i n German (Trubetzkoy 1958/69:  word-final  79):  In German the b i l a t e r a l o p p o s i t i o n d - t i s n e u t r a l i z e d i n f i n a l p o s i t i o n . The o p p o s i t i o n member, which occurs i n the p o s i t i o n of n e u t r a l i z a t i o n , from a p h o n o l o g i c a l p o i n t of view i s n e i t h e r a v o i c e d stop nor a v o i c e l e s s s t o p but "the nonnasal d e n t a l o c c l u s i v e i n g e n e r a l " . Archiphonemes are o n l y p o s i t i o n a l elements -- i n  contexts  where segment c o n t r a s t s are not n e u t r a l i z e d segments are  fully  specified. Halle  (1959), adopted the Praguian n o t i o n  of  archiphoneme, but a l s o assumed t h a t f e a t u r e s and values  may  be u n s p e c i f i e d due  combinatorial  filled  to d i s t r i b u t i o n a l c o n s t r a i n t s ,  p o s s i b i l i t i e s , grammatical c o n t e x t ,  nature of the segment i t s e l f .  feature  Unspecified  or  the  f e a t u r e values  are  i n d u r i n g the course of the d e r i v a t i o n by morpheme  s t r u c t u r e c o n s t r a i n t s or by p h o n o l o g i c a l f e a t u r e s have one  rules.  of three p o s s i b l e v a l u e s :  In the t r a d i t i o n of H a l l e  "+",  (1959) u n d e r l y i n g  Specified or  "0".  51  r e p r e s e n t a t i o n s i n SPE were assumed to be redundancy f r e e (Chomsky and H a l l e 1968:  381):  Languages d i f f e r with r e s p e c t to the sounds they use and the sound sequences they permit i n words. Thus each language p l a c e s c e r t a i n c o n d i t i o n s on the form of p h o n e t i c matrices and hence of the c o n f i g u r a t i o n s of p l u s e s and minuses ( i n d i c a t i n g membership i n one of a p a i r of complementary c a t e g o r i e s ) t h a t may appear as e n t r i e s i n the c l a s s i f i c a t o r y matrices of the l e x i c o n . These c o n s t r a i n t s make i t p o s s i b l e t o p r e d i c t , i n a g i v e n language, the s p e c i f i c a t i o n of f e a t u r e s i n p a r t i c u l a r segments. Such p r e d i c t a b i l i t y a p p l i e s t o segments i n i s o l a t i o n (e.g. i n F i n n i s h , a l l obstruents are v o i c e l e s s ) as w e l l as to segments i n p a r t i c u l a r c o n t e x t s (e.g. i n E n g l i s h , / s / i s the o n l y t r u e consonant a d m i s s i b l e b e f o r e a t r u e consonant i n w o r d - i n i t i a l p o s i t i o n ) . Rules d e s c r i b i n g these c o n s t r a i n t s can r e a d i l y be formulated w i t h i n our framework, and can be i n t e r p r e t e d as s p e c i f y i n g the c o e f f i c i e n t s of p a r t i c u l a r f e a t u r e s i n p a r t i c u l a r environments. I t i s t h e r e f o r e n a t u r a l t o propose t h a t such r u l e s be i n c o r p o r a t e d i n the grammar and t h a t the f e a t u r e s t h a t are p r e d i c t a b l e be l e f t unspecified i n lexical entries. The  d e s i r a b i l i t y of u n d e r s p e c i f i c a t i o n had been q u e s t i o n e d  Lightner  (1963) and S t a n l e y  the u n d e r l y i n g omission  (1967), however, who  argued t h a t  of f e a t u r e values c o u l d l e a d t o a  system of unwanted t e r n a r y c o n t r a s t s . accepted  by  Chomsky and  Halle  the L i g h t n e r and S t a n l e y arguments and d e a l t w i t h  the  problem by o r d e r i n g a l l redundancy r u l e s i n a b l o c k t o f i l l  in  l e x i c a l l y u n s p e c i f i e d f e a t u r e values b e f o r e the o p e r a t i o n of phonological rules.  In t h i s way  the u n d e r l y i n g  specification  of f e a t u r e s d i d not i n t e r a c t with the p h o n o l o g i c a l r u l e s of the language.  U n s p e c i f i e d f e a t u r e values were t o  be  i n t e r p r e t e d as the unmarked values of f e a t u r e s , a l l o w i n g markedness v a l u e s t o be computed f o r each l e x i c a l Kiparsky  item.  (1982) a l s o adopted the premise t h a t u n d e r l y i n g  r e p r e s e n t a t i o n s are f r e e from a l l redundancy, but d e a l t w i t h  52  the L i g h t n e r - S t a n l e y  objections  Chomsky and H a l l e (1968).  i n a d i f f e r e n t fashion  Kiparsky  proposed t h a t w i t h i n a  g i v e n environment, o n l y a s i n g l e f e a t u r e value may l e x i c a l l y s p e c i f i e d , so t h a t i f a p h o n o l o g i c a l fill  i n the a l t e r n a t e value  from  be  rule applies to  a three-way c o n t r a s t w i l l never be  p o s s i b l e w i t h i n t h a t p a r t i c u l a r environment. Chomsky and H a l l e  (1968) b e l i e v e d t h a t  redundancy r u l e s before  a l l phonological  ordering  r u l e s avoided a  t e r n a r y system and a t the same time allowed s i g n i f i c a n t l e x i c a l g e n e r a l i z a t i o n s t o be c a p t u r e d .  They f e l t the e a r l y  a p p l i c a t i o n of redundancy r u l e s was warranted because "no examples  have been d i s c o v e r e d  good  of e m p i r i c a l l y s i g n i f i c a n t  g e n e r a l i z a t i o n s t h a t r e s u l t from o r d e r i n g these r u l e s " (Chomsky and H a l l e 1968:  386).  Since t h a t time a number o f e m p i r i c a l arguments  f o r the  u n d e r s p e c i f i c a t i o n of segmental f e a t u r e s a t the time t h a t phonological  r u l e s apply have been found.  v o i c i n g of obstruents Steriade  i n Japanese  ( I t o and Mester  1987, Mester and I t o 1989).  i n v o l v e d a r e Rendaku and Lyman's Law. voices (2.16)  i n i t i a l obstruents  One example i s the 1986,  The two phenomena Rendaku i s a r u l e which  i n the second element of a compound:  Rendaku a.  o r i + kami  b.  yo  c.  yama + t e r a  d.  kake + futon -->  + sakura  —>  origami —>  -->  yozakura  'paper f o l d i n g ' 'blossoms a t n i g h t '  yamadera  'mountain  kakebuton  'top  temple'  futon'  6  53  Rendaku i s a p r o d u c t i v e  phonological  rule applying  to  compounds made up of n a t i v e Japanese morphemes whose components stand i n a modifier-head formulated  relationship.  It is  i n I t o and Mester (1986) as a r u l e a s s o c i a t i n g the  f e a t u r e [+voice] b e f o r e the second member of a Rendaku compound.  T h i s f e a t u r e w i l l be a s s o c i a t e d t o the  first  e l i g i b l e segment i n the second member of the compound. Lyman's Law  i s a c o n s t r a i n t i n Japanese which says t h a t  Rendaku i s blocked  i f any  segment i n the second compound  member i s s p e c i f i e d f o r v o i c i n g .  In (2.17) the  initial  consonants of the second member of a compound do not undergo Rendaku because they are f o l l o w e d by a v o i c e d segment. (2.17)  Lyman's  Law  a. s i r o + t a b i -->  sirotabi  b. maru + hadaka — >  'white t a b i '  maruhadaka  c. t a i k u t s u + s i n o g i -->  'completely  taikutsusinogi  'time-killing'  d. doku + tokage — >  dokutokage  'poisonous  e. onna + kotoba — >  onnakotoba  'feminine  Ito  and Mester (1986) f o r m a l i z e Lyman's Law  d e l e t e s the  naked'  lizard' speech'  as a r u l e which  [+voice] Rendaku f e a t u r e when a [+voice] f e a t u r e  i s a l r e a d y p a r t of the r e p r e s e n t a t i o n of the second compound member.  (2.17d and  e) show t h a t Lyman's Law  a v o i c e l e s s obstruent obstruent obstruent. as  a p p l i e s even when  s u r f a c e s between the i n i t i a l  of the second compound member and I f v o i c e l e s s obstruents  voiceless  a voiced  i n Japanese are  specified  [-voice] they would be expected to b l o c k the o p e r a t i o n  of  54  Lyman's Law,  s i n c e they would make the v o i c i n g of the  v o i c e d obstruent  i n v i s i b l e to the r u l e .  I t o and  suggest t h a t t h i s evidence shows t h a t v o i c e l e s s  Mester obstruents  must be u n d e r l y i n g l y u n s p e c i f i e d f o r the f e a t u r e the p o i n t i n the d e r i v a t i o n where Lyman's Law g i v e s the r e p r e s e n t a t i o n lizard'  final  [voice] at  applies.  i n (2.18) f o r the form  This  'poisonous  i n (2.17e).  (2.18)  [+voice]  [+voice]  £ [o  n  n  a]  I  RENDAKU  [k  o  t  o  b  a]  'feminine speech'  T h i s a n a l y s i s demonstrates t h a t i n Japanese o b s t r u e n t s  are  u n s p e c i f i e d f o r v o i c i n g at the p o i n t i n the phonology when c e r t a i n phonological Pulleyblank  rules  apply . -7  (1985) presents  u n d e r l i n g l y u n s p e c i f i e d and  arguments t h a t i n T i v tone i s  c e r t a i n tone b e a r i n g  u n s p e c i f i e d u n t i l the p o s t l e x i c a l component. provides  u n i t s remain  I t o (1984)  evidence t h a t /a/ i n Ainu must be u n s p e c i f i e d  for  backness a t the p o i n t i n the d e r i v a t i o n where a r u l e of dissimilation applies. phonologists  Evidence of t h i s s o r t has  t o attempt to develop t h e o r i e s  led  of  u n d e r s p e c i f i c a t i o n t h a t w i l l p r e d i c t when and where underspecified  representations  C u r r e n t l y , there are two underspecification: Archangeli  and  exist. major t h e o r i e s  of  Radical Underspecification, outlined i n  Pulleyblank  (1986) and  Contrastive  55  U n d e r s p e c i f i c a t i o n , o u t l i n e d i n S t e r i a d e (1987). t h e o r i e s w i l l be d i s c u s s e d i n 2.2.1 w i l l present  2.2.1  and  These  2.2.2, and  some a l t e r n a t i v e views of f e a t u r e  i n 2.2.3  I  specification.  Radical Underspecification  2.2.1.1  E l i m i n a t i o n of Redundancy  The  t h e o r y of RU  Pulleyblank  1986,  1989;  ( A r c h a n g e l i 1984, Pulleyblank  e x p l o r a t i o n of the nature  1986,  redundant f e a t u r e values are omitted  lexicon.  Archangeli  1988)  of p h o n o l o g i c a l r u l e s  r e p r e s e n t a t i o n s which adopts K i p a r s k y ' s  r e p r e s e n t a t i o n s and may  1988;  i s an and  (1982) view t h a t a l l  from u n d e r l y i n g  remain u n s p e c i f i e d throughout  Only n o n - p r e d i c t a b l e  and  the  f e a t u r e v a l u e s are p r e s e n t  in  the u n d e r l y i n g system of a language, and o n l y a s i n g l e v a l u e of any  feature i s permitted.  are d i s t i n g u i s h e d .  Two  types of redundancy r u l e s  Default rules insert u n i v e r s a l l y  p r e d i c t a b l e f e a t u r e v a l u e s , w h i l e complement r u l e s i n s e r t f e a t u r e v a l u e s t h a t are language s p e c i f i c . r u l e s may  Both types  be c o n t e x t - f r e e , i n s e r t i n g f e a t u r e v a l u e s  of  in  g e n e r a l , or c o n t e x t - s e n s i t i v e , i n s e r t i n g f e a t u r e v a l u e s 8  are based on co-occurrence  restrictions.  Redundancy r u l e s are ordered  among the p h o n o l o g i c a l r u l e s  of a language by the s e t of o r d e r i n g c o n s t r a i n t s g i v e n i n (2.19).  that  56  (2.19)  Ordering Constraints  a.  ( A r c h a n g e l i and P u l l e y b l a n k  1986)  Redundancy r u l e s apply as l a t e as p o s s i b l e i n the grammar.  b.  A redundancy r u l e must apply at the l e v e l a t which r e f e r e n c e i s made t o the f e a t u r e v a l u e b e i n g i n s e r t e d (The Redundancy Rule Ordering C o n s t r a i n t or RROC).  c.  A redundancy r u l e a p p l i e s as e a r l y as p o s s i b l e a t  the  l e v e l d i c t a t e d by the RROC.  The  i n t e r a c t i o n of these c o n s t r a i n t s p r o h i b i t s the  use of a t e r n a r y system of f e a t u r e v a l u e s .  At some p o i n t i n  the d e r i v a t i o n , b e f o r e a redundancy r u l e f i l l s [F],  practical  i n a value  of  t h e r e w i l l be a c o n t r a s t between the u n d e r l y i n g v a l u e  [F] ([ctF]) and  [OF],  At a l a t e r stage  p h o n o l o g i c a l r u l e a p p l i e s to [-aF], v a l u e , then by  (2.19b) and  of  i n the d e r i v a t i o n , i f a  the unmarked f e a t u r e  (2.19c) the redundancy r u l e w i l l  apply immediately b e f o r e the p h o n o l o g i c a l r u l e and t h e r e be a c o n t r a s t between [a] and  [-aF].  I f t h e r e are  will  no  p h o n o l o g i c a l r u l e s i n the language t h a t manipulate the unmarked f e a t u r e v a l u e , then the redundancy r u l e i n s e r t i n g [-aF]  does not apply u n t i l  the p o s t l e x i c a l component, by  (2.19a). RU a l s o assumes t h a t redundancy r u l e s are s u b j e c t t o EC  ( K i p a r s k y 1973,  g i v e n i n (2.3).  1982;  The  Koutsoudas, Sanders and N o l l  EC w i l l  the  1974)  force a context-sensitive  redundancy r u l e t o apply before a c o n t e x t - f r e e r u l e i f they r e f e r e n c e the same f e a t u r e v a l u e , s i n c e the environment f o r  57  the a p p l i c a t i o n o f a c o n t e x t - s e n s i t i v e r u l e i s more s p e c i f i c . The  a p p l i c a t i o n of the c o n t e x t - s e n s i t i v e r u l e w i l l b l o c k t h e  a p p l i c a t i o n o f the c o n t e x t - f r e e r u l e i n t h e p a r t i c u l a r environment s p e c i f i e d by the c o n t e x t - s e n s i t i v e r u l e .  The EC  a l s o c o n s t r a i n s the o r d e r i n g of p h o n o l o g i c a l r u l e s w i t h r e s p e c t t o redundancy r u l e s , as i n the case o f Yokuts Echo Vowel Formation, which i s ordered d e f a u l t r u l e o f [high] i n s e r t i o n  b e f o r e the more g e n e r a l ( A r c h a n g e l i and P u l l e y b l a n k  1986). In a h y p o t h e t i c a l language which has the 9 consonants /p,t,k,b,d,g,m,n,1/ i n i t s p h o n o l o g i c a l i n v e n t o r y , RU p r e d i c t s the s p e c i f i c a t i o n of the f e a t u r e s [ s o n o r a n t ] , [lateral],  and [ n a s a l ] i n (2.20),  [voice],  given the accompanying s e t  of d e f a u l t r u l e s . (2.20)  p  t  k  b  d  g  +  +  +  m  n  1  [sonorant] [voice] [lateral]  +  [nasal]  Default  +  +  rules:  [  ] —>  [ [  [-son]  —>  [+son]  ] --> [ - l a t e r a l ]  [+lateral] — >  [+son]  ] —>  [+son]  —>  [+voice]  [-son]  -->  [-voice]  [-nasal]  [+nasal]  These consonants w i l l be f u r t h e r s p e c i f i e d f o r p l a c e o f  58  a r t i c u l a t i o n , which w i l l then d i s t i n g u i s h the l a b i a l s , alveolars  and v e l a r s  from one another.  The f e a t u r e  i s completely redundant i n t h i s h y p o t h e t i c a l  [sonorant]  system, s i n c e t h e  sonorancy v a l u e s o f a l l segments i n (2.20) a r e g i v e n by t h e context-free  r u l e i n s e r t i n g [-son] and the two c o n t e x t -  sensitive rules predicting  [+son] on segments marked  [ + l a t e r a l ] o r [+nasal]. 2.2.1.2  Rule Types  A r c h a n g e l i and P u l l e y b l a n k (1986) p r o v i d e a t y p o l o g y o f possible  r u l e types w i t h accompanying markedness statements.  Each and every p h o n o l o g i c a l r u l e has a f u n c t i o n argument, and may a l s o have a t a r g e t (2.21) g i v e s the p o s s i b l e designates functions trigger/target (2.21) I .  and an  and/or t r i g g e r  condition.  s p e c i f i c a t i o n s o f r u l e s , where " I "  o r o p e r a t i o n s , " I I " arguments, and " I I I "  conditions.  a. ( i n s e r t )  d < 2 > t Q U l t  /delete  b.  (raaximal)  c.  (content) «  d.  (same d i r e c t i o n )  d  d e f a u l t  £au  /minimal  x /structure t  d e f a u  i t / o p p o s i t e direction/  bidirectional II. III.  node o r  feature(s)  trigger  condition  target  condition  There a r e four f u n c t i o n parameters, each o f which has a d e f a u l t o r unmarked s e t t i n g . feature  or association  "Insert/delete"  indicates that a  l i n e i s added t o o r taken away from t h e  59  representation,  w i t h " i n s e r t " being the unmarked o p t i o n .  "maximal/minimal" f u n c t i o n d e s c r i b e s targets.  The  the adjacency o f r u l e  A t the unmarked "maximal" s e t t i n g a r u l e which  t a r g e t s a node o r f e a t u r e scans the h i g h e s t  l e v e l of s y l l a b i c  s t r u c t u r e p r o v i d i n g access t o t h a t t a r g e t , w h i l e a t t h e "minimal" s e t t i n g a r u l e scans the t i e r c o n t a i n i n g  the t a r g e t .  Rules t a r g e t i n g vowels can t h e r e f o r e a p p l y t o segments t h a t are not immediately adjacent a t the s k e l e t a l l e v e l , but must be a d j a c e n t o r l o c a l a t the l e v e l of the n u c l e u s .  This  c o n s t r a i n t on adjacency i s termed the L o c a l i t y C o n d i t i o n i n Archangeli  and P u l l e y b l a n k  (1987).  "Content/structure"  i n d i c a t e s t h a t i t i s a f e a t u r e o r an a s s o c i a t i o n l i n e which i s added t o the r e p r e s e n t a t i o n  by the r u l e , w i t h "content" b e i n g  the d e f a u l t o r unmarked o p t i o n . describes  The d i r e c t i o n a l i t y parameter  the d i r e c t i o n o f spreading, w i t h "same d i r e c t i o n "  i n d i c a t i n g t h a t the unmarked s e t t i n g f o r t h i s f u n c t i o n i s i n the same d i r e c t i o n as the i n i t i a l a s s o c i a t i o n o f s k e l e t a l s l o t s t o melodic s t r u c t u r e  (as given by a s e t o f A s s o c i a t i o n  Conventions such as those i n ( 2 . 6 ) ) . T h i s t y p o l o g y of r u l e f u n c t i o n s o r o p e r a t i o n s that there  a r e two unmarked types o f p h o n o l o g i c a l  predicts rules  a v a i l a b l e t o c h i l d r e n -- d e f a u l t r u l e s and r u l e s o f epenthesis.  Default  r u l e s use only d e f a u l t s e t t i n g s o f  f u n c t i o n parameters and i n s e r t arguments t h a t a r e p r o v i d e d UG.  A context-free  d e f a u l t r u l e i n s e r t i n g the f e a t u r e  by  [-low]  w i l l be s t a t e d as i n (2.22), where the parentheses i n d i c a t e the d e f a u l t s e t t i n g s of parameters.  60  (2.22)  Default I.  [-low]  a.  (insert)  c.  (content)  d.  (same d i r e c t i o n )  [-low]  II.  (2.22) says t h a t Configuration  [-low] w i l l be i n s e r t e d anywhere (a s e t o f  Constraints  w i l l d i c t a t e where i t can a t t a c h and  where i t c a n n o t ) , and i n s e r t i o n w i l l operate i n l e f t - t o - r i g h t f a s h i o n , assuming t h a t t h i s i s the unmarked d i r e c t i o n o f i n i t i a l association.  The maximal/minimal parameter i s  i r r e l e v a n t t o t h i s r u l e , since there A context-sensitive  i s no s p e c i f i e d t a r g e t . 9  d e f a u l t r u l e w i l l use the same f u n c t i o n  parameters as i n (2.22) but w i l l r e q u i r e the a d d i t i o n o f a target  condition.  The second type of maximally unmarked r u l e i s an e p e n t h e s i s r u l e , where the f u n c t i o n parameters a r e a g a i n s e t t o t h e d e f a u l t s , w i t h the argument s p e c i f i e d as a s k e l e t a l slot  1 0  .  (2.23)  Epenthesis I.  II.  a.  (insert)  c.  (content)  d.  (same d i r e c t i o n )  skeletal  slot  Such a r u l e w i l l only be p o s i t e d when s y l l a b i f i c a t i o n o r o t h e r l i c e n s i n g requirements make i t necessary.  The i n s e r t e d  61  s k e l e t a l s l o t w i l l have no f e a t u r a l content;  however, s i n c e  (2.22) i s not a redundancy r u l e , i t w i l l not be s u b j e c t t o t h e O r d e r i n g C o n s t r a i n t s i n (2.18) and w i l l apply i n the l e x i c a l phonology b e f o r e the redundancy r u l e s  1 1  .  In t h i s way t h e  f e a t u r e s p e c i f i c a t i o n s of the e p e n t h e t i c segment w i l l be p r o v i d e d by the redundancy r u l e s o f the language. I f we f o l l o w c u r r e n t assumptions about Epenthesis (as d i s c u s s e d i n 2.1.2.4) then the unmarked s t a t u s o f (2.23) i s e x p l a i n e d by the f a c t t h a t i t i s simply an automatic r e s u l t o f the s y l l a b i f i c a t i o n algorithms  i n a p a r t i c u l a r language.  The  r u l e i n (2.23) w i l l not have the same s t a t u s as o t h e r p h o n o l o g i c a l r u l e s (such as those g e n e r a l have a b e a r i n g on p r o s o d i c  i n (2.15), which do not i n representations.  Rules t h a t w i l l i n v o l v e s e t t i n g a t l e a s t one o f t h e f u n c t i o n parameters i n (2.22) t o the marked v a l u e w i l l be those t h a t d e l e t e f e a t u r e s o r s t r u c t u r e ; s p r e a d i n g o r a s s i m i l a t i o n r u l e s which use the marked " s t r u c t u r e " parameter; r u l e s whose t a r g e t s a r e s p e c i f i e d as "minimal"; and r u l e s t h a t operate  i n some d i r e c t i o n other than the c a n o n i c a l  of i n i t i a l  association.  direction  Thus RU r e c o g n i z e s t h r e e o f t h e r u l e  types g i v e n i n (2.15): D e l e t i o n , Spread and I n s e r t i o n , without allowing f o r Fusion. Both the type o f u n d e r s p e c i f i c a t i o n assumed by RU and the r u l e typology  i n (2.21) have i m p l i c a t i o n s f o r how p h o n o l o g i c a l  r u l e s operate  i n t h i s theory.  v a l u e i s present  Because o n l y a s i n g l e f e a t u r e  i n any environment, Spreading  a b l e t o manipulate o n l y a s i n g l e f e a t u r e v a l u e .  r u l e s w i l l be The t y p o l o g y  62  in  (2.21) f o r m a l i z e s Kiparsky's  claim  (Kiparsky  1985)  that i n  the unmarked case f e a t u r e s spread o n l y t o segments which are • f r e e ' or u n s p e c i f i e d f o r t h a t f e a t u r e .  The  fact that a  t a r g e t i s f r e e i s s t a t e d i n (2.21) by the l a c k of a t a r g e t condition  (2.21III).  harmony g i v e n  In the a n a l y s i s of Khalka Mongolian  i n Archangeli  and  Pulleyblank  (1986), i t i s  assumed t h a t Round Harmony, which spreads [+round], l a c k s a t a r g e t c o n d i t i o n , e x p l a i n i n g why It  i s r a r e t h a t two  once (although present  values  of any  i n the i n v e n t o r y  shown i n (2.20) t h i s l a t e r a l would [ - v o i c e ] ) , but  presumably a l s o be  blocked).  2.2.2  Underspecification  Contrastive  The and  has  i f a rule  [-F]  spreading  spreading  should  Specifications theory  of CU was  first  outlined i n Steriade  been taken up i n works such as Clements  C a l a b r e s e (1988) and Mester and was  I t o (1989).  (1987),  (1987),  S t e r i a d e ' s work  not meant as a t h e o r e t i c a l t r e a t i s e , but r a t h e r as  a l t e r n a t i v e t o the theory the t h e o r y  are not  i n Steriade Steriade  trivial  and  of RU.  an  As a r e s u l t , many a s p e c t s of  f u l l y a r t i c u l a t e d , although most c r u c i a l  a s p e c t s of the theory may given  /li/.  f e a t u r e are s p e c i f i e d a t  encounters a segment s p e c i f i e d as  2.2.2.1  by  i t c o u l d happen: i f a v o i c e l e s s l a t e r a l were  presumably be s p e c i f i e d as [+F]  the r u l e i s b l o c k e d  be drawn from s p e c i f i c  analyses  (1987).  (1987) makes an i n i t i a l  d i s t i n c t i o n between  non-trivial underspecification.  Trivial  63  u n d e r s p e c i f i c a t i o n occurs  when a segment l a c k s  f o r a p a r t i c u l a r f e a t u r e at a l l stages  specification  in a derivation.  a r i s e s i n the case of monovalent f e a t u r e s or  It  articulator  nodes/ where o n l y a s i n g l e value of the f e a t u r e or node i s o p e r a t i v e , and value  i n the case of f e a t u r e s which never a c q u i r e  a  f o r a second f e a t u r e because of r e s t r i c t i o n s i n the  feature hierarchy.  An example of the l a t t e r type of  u n d e r s p e c i f i c a t i o n i s a D o r s a l consonant which w i l l  trivial never  receive a s p e c i f i c a t i o n for [anterior], since [anterior] i s a f e a t u r e t h a t i s dependent on the Coronal (2.7)).  a r t i c u l a t o r node  (see  N o n - t r i v i a l u n d e r s p e c i f i c a t i o n occurs when a segment  l a c k s a s p e c i f i c a t i o n f o r a f e a t u r e at some stage i n the d e r i v a t i o n , but e v e n t u a l l y becomes s p e c i f i e d f o r t h a t  feature  value. The  important d i s t i n c t i o n w i t h i n  non-trivial  u n d e r s p e c i f i c a t i o n i s made between R-values of f e a t u r e s , which are redundant f e a t u r e v a l u e s , and D-values, which are those v a l u e s of a feature, t h a t c o n t r a s t w i t h i n a s p e c i f i c c l a s s of segments.  R-values are absent from  underlying  r e p r e s e n t a t i o n s , w h i l e D-values must both be p r e s e n t . c o n t r a s t i v e s p e c i f i c a t i o n of the h y p o t h e t i c a l system shown i n (2.20) i s given i n  (2.24).  A  consonantal  64  (2.24)  p  t  k  b  d  g  m  n  l  [sonorant] [voice]  - - - + + + a  [lateral]  -  [nasal]  -  +  -  +  +  Redundancy r u l e s : [+nasal]  —>  [+son]  [+lateral] —>  [+son]  [+son]  —>  [+voice]  [+cons]  —>  [-son]  [+cons]  -->  [-lateral]  [+cons]  —>  [-nasal]  The v o i c e d and v o i c e l e s s obstruents the f e a t u r e  [ v o i c e ] i n t h i s system, so v o i c e l e s s  are u n d e r l y i n g marked [+voice].  contrast with regard to obstruents  [-voice] and the v o i c e d ones as  / I / contrasts with  /d/ i n terms of l a t e r a l i t y ,  both a r e s p e c i f i e d f o r a value of the f e a t u r e two n a s a l s c o n t r a s t w i t h the corresponding  [lateral].  voiced  The  obstruents,  and t h e r e f o r e /b/ and /d/ are s p e c i f i e d as [-nasal] w h i l e and /n/ a r e s p e c i f i e d as [+nasal]. i s completely  2.2.2.2  so  In t h i s system  /m/  [sonorant]  redundant.  R-Rules  There i s no formal  c o n t r a s t i n CU between u n i v e r s a l and  l a n g u a g e - s p e c i f i c redundancy (R-^rules).  R-rules  insert  R-  65  values.  They are always c o n t e x t - s e n s i t i v e , s i n c e they  insert  redundant values based on co-occurrence r e s t r i c t i o n s , and they may have one of two f u n c t i o n s (2.25)  a.  ( S t e r i a d e 1987: 359):  the enhancement of p e r c e p t u a l e.g.  b.  [+back] — >  salience,  [+round], o r  the demonstration o f the r e s t r i c t e d of a content  distribution  f e a t u r e w i t h i n the realm o f a  s t r i c t u r e feature ( s t r i c t u r e features are [sonorant], e.g.  [cons], [continuant],  [+son] — >  [ h i g h ] , [low])  [+voice].  S t e r i a d e p o i n t s out t h a t assuming t h a t R-rules may have o n l y the f u n c t i o n s i n (2.24) r u l e s out the p o s s i b i l i t y of an such as [+round] — > Pulleyblank  [+high],  R-rule  which i s used by A r c h a n g e l i  (1986) i n the a n a l y s i s of Nyangumarta.  This  and rule  i s not excluded on the grounds t h a t i t i s l a n g u a g e - s p e c i f i c , but on the grounds t h a t i t does not have one of the f u n c t i o n s in  (2.25). The R-rules  are g i v e n i n  f o r v o c a l i c segments used i n S t e r i a d e  (2.26) . 1 2  (1987)  66  R- r u l e s 1.  [+low]  - --> [-high]  2.  [+low]  — > [+back]  3.  [+back] - -->  [+round]  [-low] 4.  [-low]  - -->  [-back]  [-round] 5.  [+high] - -->  [+round]  [+ATR]  In S t e r i a d e ' s a n a l y s i s  o f Hungarian Back Harmony an R - r u l e i s  e x t r i n s i c a l l y ordered b e f o r e a p h o n o l o g i c a l r u l e .  Two R - r u l e s  and a r u l e o f Back Harmony a r e p o s i t e d f o r Hungarian ( t h e CU analysis  o f Hungarian BH w i l l be d i s c u s s e d i n d e t a i l i n  3.1.5.2): (2.27)  R - r u l e 1:  [+low] — >  [+back]  R - r u l e 2:  [-low, -round] — >  Back Harmony ( i t e r a t i v e , [aback]  V . . .  [-back]  feature-changing)  [Aback]  V  The r u l e o f Back Harmony i s ordered between R - r u l e s 1 and 2 i n o r d e r t h a t the low long vowel /a:/ can a c t as a t r i g g e r o f Harmony. it  fills  R - r u l e 2 cannot be ordered b e f o r e Harmony, because i n the backness s p e c i f i c a t i o n s of / i / ,  /i:/  and /e:/  and these vowels a r e t r a n s p a r e n t t o the Harmony p r o c e s s .  The  67  EC,  which orders  s p e c i f i c r u l e s before more g e n e r a l  cannot f o r c e the o r d e r i n g of R-rule  1 before  s i n c e the s t r u c t u r a l d e s c r i p t i o n of R-rule  ones,  Back Harmony,  1 i s not p a r t of  the s t r u c t u r a l d e s c r i p t i o n of Back Harmony. Steriade  (1987) a l s o p o s i t s an e x t r i n s i c  ordering  r e l a t i o n s h i p between the r u l e of Round Harmony i n Mongolian and  the redundancy r u l e i n s e r t i n g the p r e d i c t a b l e v a l u e  [round].  She  argues, however, t h a t l a n g u a g e - s p e c i f i c  between p h o n o l o g i c a l and  r u l e s and  of  ordering  redundancy r u l e s i s unwanted,  i n s t e a d assumes t h a t i n Mongolian [round] a c t s as a  monovalent f e a t u r e . inserting  In t h i s language then the redundancy r u l e  [-round] i s not r e q u i r e d .  Unfortunately,  a l t e r n a t e a n a l y s i s of the Hungarian f a c t s i s not  2.2.2.3  Phonological  an  provided.  Rules  Because CU assumes t h a t both values  of a c o n t r a s t i v e  f e a t u r e are u n d e r l y i n g l y s p e c i f i e d , the theory must a l s o assume t h a t a s s i m i l a t i o n r u l e s may or feature-changing.  Steriade  be e i t h e r f e a t u r e - f i l l i n g  (1987: 339)  states:  n o t h i n g , i n my view, prevents a r u l e of vowel harmony from o p e r a t i n g s e q u e n t i a l l y , by s u c c e s s i v e s p r e a d i n g and d e l i n k i n g s t e p s , as shown i n ( 1 ) : (1) F  -F  -F  -F  V C V C V C V --> F  F  blocked  -F  V C V C V C V  -F  V C V C V C V --> A r u l e p r o p o g a t i n g the  -F  feature  -->  F V C V C V C V [F] w i l l t h e r e f o r e not  be  when a segment i s encountered t h a t i s s p e c i f i e d f o r  68  [ F ] , as i s assumed i n the unmarked case by R a d i c a l Underspecification.  S t e r i a d e p r o v i d e s examples of f e a t u r e -  changing processes i n the Height Harmony system of  Pasiego,  f o l l o w i n g McCarthy (1984) and i n the Back Harmony system o f Hungarian,  f o l l o w i n g the a n a l y s i s of Farkas and Beddor  shown i n (2.27).  (1987),  In Pasiego e i t h e r [+high] or [-high]  i n i t i a t e s Height Harmony, spreading t o a vowel s p e c i f i e d either  [+high].  In Hungarian,  as  S t e r i a d e assumes t h a t e i t h e r  [+back] o r [-back] i n i t i a t e s the harmony process and t h a t t a r g e t s may may  be s p e c i f i e d as e i t h e r  [+back].  If rule  be s p e c i f i e d f o r e i t h e r v a l u e of a f e a t u r e , then  triggers these  r u l e s w i l l have t o be s t a t e d u s i n g alpha n o t a t i o n , as shown i n (2.27) .  2.2.3  A l t e r n a t e T h e o r i e s of Feature  Specification  The m a j o r i t y of c u r r e n t analyses d e a l i n g w i t h u n d e r s p e c i f i e d v a l u e s adopt e i t h e r the R a d i c a l or C o n t r a s t i v e framework.  Calabrese  (1988), Clements (1987) and C h r i s t d a s  (1988) argue f o r s l i g h t l y m o d i f i e d v e r s i o n s of  CU.  C a l a b r e s e ' s r e v i s i o n s i n v o l v e d e r i v i n g the p a t t e r n of u n d e r s p e c i f i c a t i o n i n a g i v e n language from the s e t of u n i v e r s a l f i l t e r s t h a t are v i o l a t e d  language-specifically  ( t h i s i s e s s e n t i a l s i n c e Calabrese uses f i l t e r s or n e g a t i v e c o n s t r a i n t s , whereas S t e r i a d e uses u n i v e r s a l redundancy o r rules).  R-  C a l a b r e s e ' s claims r e g a r d i n g the h i e r a c h y of  u n i v e r s a l f i l t e r s w i l l be d i s c u s s e d i n 6.1.2.2. w i l l b r i e f l y o u t l i n e Clements' theory.  In 2.2.3.1 I  A number of o t h e r  69  works> such as Den Dikken and van d e r H u l s t  (1990) and P i g g o t t  (1990, t o appear) assume t h a t f e a t u r e s a r e p r i v a t i v e r a t h e r than b i n a r y , and t h e r e f o r e do not adopt a theory o f underspecification at a l l .  2.2.3.1  Clements (1987)  Clements' (1987) p o s i t i o n on u n d e r s p e c i f i c a t i o n i s t h a t both v a l u e s  o f a c o n t r a s t i v e d i s t i n c t i v e f e a t u r e a r e always  f u l l y s p e c i f i e d i n URs, as a r e the s o - c a l l e d 'primary' o r major c l a s s f e a t u r e s such as [sonorant]  and [ c o n s o n a n t a l ] .  Incomplete u n d e r l y i n g s p e c i f i c a t i o n o f values  occurs  f e a t u r e s a r e not used c o n t r a s t i v e l y , o r when they a r t i c u l a t o r nodes ( i . e . L a b i a l , Coronal  represent  and D o r s a l  There a r e no u n i v e r a l d e f a u l t r u l e s t o f i l l  when  i n (2.6)).  i n missing  r u l e s may add m i s s i n g  but  language-particular  The  h y p o t h e t i c a l 9 consonant language s p e c i f i e d i n (2.20) and  (2.24) i s s p e c i f i e d a c c o r d i n g (2.28)  p  t  k  b  d  g  m  n  l  [sonorant]  -  -  -  -  -  -  +  +  +  [voice]  - - - + + + -  [nasal]  feature  [sonorant]  values.  t o Clements' system i n (2.28).  [lateral]  The  feature  values,  -  -  + +  +  i s f u l l y s p e c i f i e d s i n c e i t i s a major  c l a s s f e a t u r e , w h i l e the f e a t u r e s [nasal] are c o n t r a s t i v e .  [ v o i c e ] , [ l a t e r a l ] and  A r t i c u l a t o r nodes would behave i n a  monovalent f a s h i o n ( i . e . a r e e i t h e r present  or absent),  with  70  L a b i a l present  f o r /p,b, and m/ and absent f o r a l l other  features, etc. In t h i s t h e s i s I w i l l t r e a t the Clements Christdas  (1987) (and  1988) view of u n d e r s p e c i f i c a t i o n as another form of  CU, s i n c e vowel f e a t u r e s , which are n e i t h e r major c l a s s f e a t u r e s nor a r t i c u l a t o r nodes, w i l l be c o n t r a s t i v e l y specified. equivalent  2.2.3.2  Clements* treatment of vowels w i l l t h e r e f o r e  be  t o a c o n t r a s t i v e one.  P r i v a t i v e Features  Each of the t h e o r i e s of u n d e r s p e c i f i c a t i o n assume t h a t f e a t u r e s are b i n a r y , as was Jakobson and H a l l e (1956).  discussed  f i r s t proposed i n  Trubetzkoy (1939/69), on the  hand, proposes t h a t f e a t u r a l o p p o s i t i o n s where f e a t u r e s are e i t h e r present  may  other  be p r i v a t i v e ,  o r absent; g r a d u a l ,  where  f e a t u r e s have degrees; o r e q u i p o l l e n t , where f e a t u r e s  have  p o l a r d i s t i n c t i o n s t h a t are each l o g i c a l l y p o s s i b l e .  In the  f e a t u r e geometry privative  i n (2.7) nodes a r e assumed t o behave i n a  (or monovalent)  underspecification  f a s h i o n , even by proponents of  theory.  More r e l e v a n t t o a d i s c u s s i o n of u n d e r s p e c i f i c a t i o n i s the i s s u e of whether t e r m i n a l privative.  Piggott  f e a t u r e s are b i n a r y o r  (1990) proposes t h a t the f e a t u r e  [nasal]  i s u n i v e r s a l l y a p r i v a t i v e f e a t u r e , w h i l e Mester and I t o (1989) and S t e r i a d e  (1987) propose t h a t  are p r i v a t i v e f e a t u r e s  [ v o i c e ] and  [round]  i n Japanese and Mongolian r e s p e c t i v e l y .  Den Dikken and van der H u l s t  (1990) assume t h a t a l l f e a t u r e s  71  are p r i v a t i v e , and Goldsmith (1987) proposes t h a t a l l f e a t u r e s are e i t h e r p r i v a t i v e o r e q u i p o l l e n t .  In t h e m a j o r i t y o f cases  i t w i l l be i m p o s s i b l e t o determine whether a f e a t u r e i s p r i v a t i v e o r whether i t i s b i n a r y but has o n l y a s i n g l e f e a t u r e v a l u e present u n d e r l y i n g l y .  T h i s i s because  u n d e r s p e c i f i c a t i o n theory assumes t h a t u n d e r l y i n g l y u n s p e c i f i e d f e a t u r e values a r e f i l l e d  i n by redundancy r u l e s  l a t e i n t h e d e r i v a t i o n , and t h e r e f o r e redundant v a l u e s g e n e r a l l y have no p h o n o l o g i c a l e f f e c t s . 283)  will  Mester and I t o (1989:  warn t h a t t h i s may make comparisons of RU and a t h e o r y o f  p r i v a t i v e features d i f f i c u l t : I f not a l l f e a t u r e s a r e p r i v a t i v e (see den Dikken & van der H u l s t 1988 and r e f e r e n c e s c i t e d there f o r a s t r i c t l y s i n g l e - v a l u e d approach), a p r i n c i p l e d t y p o l o g y o f f e a t u r e s must be developed. Otherwise p r i v a t e f e a t u r e s c o u l d be invoked ad l i b i t u m whenever i t seems t h a t RU i s needed. I t i s o n l y i n t h e case o f a b i n a r y f e a t u r e whose p r e d i c t a b l e v a l u e has p h o n o l o g i c a l e f f e c t s t h a t we w i l l be a b l e t o d i s t i n g u i s h i t from a p r i v a t i v e f e a t u r e .  In RU, t h e  p h o n o l o g i c a l use o f p r e d i c t a b l e f e a t u r e s i s accomplished by the RROC, g i v e n i n (2.19b). RROC a r e presented for  Some examples o f t h e use o f t h e  i n A r c h a n g e l i and P u l l e y b l a n k  Yoruba and i n A r c h a n g e l i and P u l l e y b l a n k  Yokuts.  I b e l i e v e that i t i s contingent  (1986, 1989)  (1986)  for  upon those who assume  t h a t f e a t u r e s may be p r i v a t i v e t o account f o r these c a s e s .  In  t h i s t h e s i s I assume t h a t t e r m i n a l f e a t u r e s a r e b i n a r y , and always have two p o s s i b l e v a l u e s .  Based on t h e h i s t o r i c a l  development of d i s t i n c t i v e f e a t u r e s as b i n a r y u n i t s , I b e l i e v e t h a t t h e b i n a r y approach i s more c o n s e r v a t i v e than t h e  72  privative  2.3  one.  P r i n c i p l e s and  Parameters  The p r i n c i p l e s and parameters  t h e o r y of grammar was  first  o u t l i n e d i n Chomsky (1981a,b) and has been w i d e l y used t o account f o r a v a r i e t y of f a c t s i n both syntax and Parameters  are used t o e x p l a i n c r o s s - l i n g u i s t i c  phonology.  differences,  and a t the same time make p r e d i c t i o n s about the course of language a c q u i s i t i o n .  For these reasons, i n t h i s t h e s i s I  w i l l develop a theory of p h o n o l o g i c a l a c q u i s i t i o n t h a t assumes t h a t UG c o n t a i n s both p r i n c i p l e s and parameters. f o l l o w i n g s e c t i o n s I o u t l i n e how  the p r i n c i p l e s and  model of grammar i s o r g a n i z e d , and show how proposed parameters  In the parameters  some p r e v i o u s l y  are assumed t o o p e r a t e .  The p r i n c i p l e s and parameters  t h e o r y emphasizes the  u n i v e r s a l i t y of the t h e o r e t i c a l c o n s t r u c t s used i n a l l languages, w h i l e p r e d i c t i n g t h a t languages d i f f e r i n c e r t a i n c o n s t r a i n e d ways: What we expect t o f i n d , then, i s a h i g h l y s t r u c t u r e d t h e o r y of UG based on a number of fundamental p r i n c i p l e s t h a t s h a r p l y r e s t r i c t the c l a s s of a t t a i n a b l e grammars and narrowly c o n s t r a i n t h e i r form, but w i t h parameters t h a t have t o be f i x e d by e x p e r i e n c e . I f these parameters are embedded i n a t h e o r y of UG t h a t i s s u f f i c i e n t l y r i c h i n s t r u c t u r e then the languages t h a t are determined by f i x i n g t h e i r v a l u e s one way or another w i l l appear t o be q u i t e d i v e r s e , s i n c e the consequences of one s e t of c h o i c e s may be v e r y d i f f e r e n t from the consequences of another s e t ; y e t a t the same time, l i m i t e d e v i d e n c e , j u s t s u f f i c i e n t t o f i x the parameters of UG, w i l l determine a grammar t h a t may be v e r y i n t r i c a t e and w i l l i n g e n e r a l l a c k grounding i n experience i n the sense of an i n d u c t i v e b a s i s . (Chomsky 1981a: 3-4) A c c o r d i n g t o t h i s view a c q u i s i t i o n becomes a r e l a t i v e l y  simple  73  endeavour. evidence  A p a r t i c u l a r hypothesis i s s u p p l i e d by UG, and  from t h e language being a c q u i r e d w i l l c o n f i r m t h a t  t h i s i n i t i a l hypothesis i s c o r r e c t , o r suggest t h a t an a l t e r n a t e h y p o t h e s i s should be chosen.  In t h e l a t t e r  case,  the c h i l d need not c r e a t e a new h y p o t h e s i s , but simply " s w i t c h " a parameter, and a new hypothesis w i l l be g i v e n . The p r i n c i p l e s and parameters approach p r o v i d e s a b u i l t i n t h e o r y o f markedness.  The parameter s e t t i n g s p r o v i d e d by  UG a r e t h e d e f a u l t o r unmarked s e t t i n g s , w h i l e t h e o t h e r s e t t i n g ( s ) , which can o n l y be achieved v i a evidence from t h e i n p u t , a r e marked.  A language t h a t makes use o f t h e d e f a u l t  s e t t i n g o f a parameter w i l l be l e s s complex o r marked than a language t h a t uses t h e marked s e t t i n g .  In a c q u i s i t i o n ,  t h i s model p r e d i c t s t h a t the c h i l d w i l l  f i r s t assume t h e  then,  unmarked parameter s e t t i n g , and o n l y s w i t c h t o a marked parameter s e t t i n g i f t h e r e i s evidence t h a t t h i s i s n e c e s s a r y . Developmentally,  the c h i l d ' s speech  should f i r s t be  c h a r a c t e r i z e d by t h e unmarked parameter s e t t i n g , and o n l y l a t e r be c h a r a c t e r i z e d by t h e marked s e t t i n g . (1983),  In Hyams  f o r example, t h e a c q u i s i t i o n o f t h e "pro-drop" o r  AGR/PRO parameter (Chomsky 1981a, R i z z i 1982) i s d i s c u s s e d . T h i s parameter was proposed languages clauses . 1 3  t o e x p l a i n d i f f e r e n c e s between  i n t h e a b i l i t y t o have n u l l - s u b j e c t s i n tensed Hyams argues  t h a t the unmarked parameter s e t t i n g  a l l o w s n u l l - s u b j e c t s , as i n I t a l i a n , w h i l e t h e marked s e t t i n g r e q u i r e s l e x i c a l s u b j e c t s , as i n E n g l i s h . of  The low i n c i d e n c e  l e x i c a l i z e d s u b j e c t s i n t h e speech o f young c h i l d r e n  74  a c q u i r i n g E n g l i s h i s due t o the i n i t i a l AGR/PRO ( t h i s a n a l y s i s has sparked  unmarked parameter f o r  much c o n t r o v e r s y  G u i l f o y l e 1984 and Lebeaux 1987 f o r two r e a n a l y s e s  -- see o f these  facts).  2.3.1  Binarity The  m a j o r i t y of parameters t h a t have been d i s c u s s e d i n  the l i t e r a t u r e , i n both syntax and phonology, have b i n a r y settings.  F o r example, Bach (1965) claims t h a t languages have  the c h o i c e o f having a r u l e of Wh-movement o r not ( W i l l i a m s 1987  c a l l s t h e Wh-movement parameter an " e x i s t e n c e "  parameter). Pulleyblank  A p h o n o l o g i c a l example i s g i v e n i n A r c h a n g e l i and (1989) where t h e i n i t i a l  autosegmentalized  association of  f e a t u r e s i s assumed t o operate  i n left-to-  r i g h t f a s h i o n i n the unmarked case and i n r i g h t - t o - l e f t f a s h i o n i n t h e marked case (see 2.1.2.1).  Bach's Wh-Movement  parameter i s a b i n a r y parameter t h a t has a c h o i c e between ON and OFF, w h i l e A r c h a n g e l i and P u l l e y b l a n k ' s  initial  a s s o c i a t i o n parameter i s b i n a r y but has two s p e c i f i e d  choices:  LEFT-TO-RIGHT o r RIGHT-TO-LEFT. Suggestions f o r parameters w i t h m u l t i p l e s e t t i n g s have been g i v e n i n Wexler and Manzini category  (1987) f o r t h e governing  o f anaphors, o r f o r the d i r e c t i o n o f s p r e a d i n g i n  A r c h a n g e l i and P u l l e y b l a n k  (1986).  Fodor (1989) argues t h a t  non-binary parameters can and should be analyzed parameters, each with a b i n a r y c h o i c e .  as s e t s o f  Only b i n a r y parameters  w i l l p r o v i d e a c l e a r statement of the markedness o f p a r a m e t r i c  75  systems.  A r c h a n g e l i and P u l l e y b l a n k * s d i r e c t i o n o f s p r e a d i n g  parameter (1986) can be s t a t e d as two separate i s done i n P i g g o t t (to (2.29)  parameters, as  appear):  D i r e c t i o n a l i t y Parameter(s) i.  M u l t i p l e S e t t i n g s ( A r c h a n g e l i and P u l l e y b l a n k 1986) Same d i r e c t i o n / o p p o s i t e / b i d i r e c t i o n a l  ii.  B i n a r y S e t t i n g s ( P i g g o t t ( t o appear)) a.  Spread l e f t w a r d  b.  Spread rightward  (yes/no) (yes/no)  A r c h a n g e l i and P u l l e y b l a n k ' s parameter i s t i e d t o another parameter which r e g u l a t e s t h e d i r e c t i o n of i n i t i a l a s s o c i a t i o n (see, f o r example, ( 2 . 6 ) ) , but P i g g o t t ' s parameter does not make t h i s c o n n e c t i o n . achieved  i f t h e YES o p t i o n i s chosen f o r both parameters i n  (2.29ii). 2.5  The b i d i r e c t i o n a l s e t t i n g i n (2.29i) i s  The f e a t u r a l and r u l e parameters i n v e s t i g a t e d i n  w i l l a l l be b i n a r y , c e r t a i n of them p r o v i d i n g ON-OFF  c h o i c e s , and others two s p e c i f i e d  2.3.2  choices.  M u l t i p l e Parameters Dresher and Kaye (1988) encounter cross-parameter  dependencies i n t h e i r account of parameterized  s t r e s s systems.  An example i s g i v e n i n v o l v i n g the three parameters i n (2.30). (2.30)  PI  The word-tree i s s t r o n g on t h e (LEFT/RIGHT)  P2  Feet a r e (BINARY/UNBOUNDED)  P5  Feet a r e q u a n t i t y s e n s i t i v e (QS) (YES/NO)  76  In o r d e r f o r a l e a r n e r t o determine which s e t t i n g of PI i s r e q u i r e d , a "window" a t the l e f t or r i g h t edge of a word must be sampled.  P2 w i l l determine the s i z e of the window, but the  s e t t i n g of P2 i s dependent upon the s e t t i n g of P5.  I f the  NO  o p t i o n of P5 i s chosen, then the BINARY o p t i o n of P2 must a l s o be chosen, and i f the YES o p t i o n of P5 i s chosen, then P2 must be UNBOUNDED.  Thus the d e f a u l t s e t t i n g of P2 w i l l depend upon  the a p p r o p r i a t e s e t t i n g of P5 i n the language. somehow be t i e d t o g e t h e r , or the l e a r n e r may t h a t i t i s not p o s s i b l e t o r e t r e a t from.  P2 and P5 must  make a c h o i c e  W i l l i a m s (1987: x i )  says of such i n t e r c o n n e c t i o n s : The q u e s t i o n i s , how complicated are such c o n t i n g e n c i e s — i n the worst case, one can imagine the parameters were so p a r a l y z i n g l y i n t e r c o n n e c t e d t h a t they a l l had to be s e t "at one time" and the evidence was the unionof a l l the evidence r e l e v a n t f o r any of them. Dresher and Kaye do not p r o v i d e evidence as t o how dependencies  these  are d e a l t w i t h , but simply note t h a t they e x i s t .  Such i n t e r d e p e n d e n c i e s p r e d i c t t h a t c h i l d r e n w i l l remain w i t h the unmarked parameter  setting until  a l l the evidence i s i n ,  then a l l parameters w i l l be r e s e t a t once.  With r e g a r d t o the  interdependence of d e f a u l t v a l u e s , i t remains q u e s t i o n how  an  empirical  these problems w i l l be s o r t e d out.  Wexler and Manzini (1987) i n t h e i r d i s c u s s i o n of the parameter  a s s o c i a t e d w i t h anaphors,  P a r a m e t e r i z a t i o n Hypothesis (1984).  f i r s t d i s c u s s e d i n Borer  The LPH assumes t h a t parameters  individual lexical whole.  (LPH)  adopt the L e x i c a l  are s e t f o r  items, r a t h e r than f o r a c o n s t r u c t as a  There w i l l be d i f f e r e n t p a r a m e t r i c p o s s i b i l i t i e s  for  77  each l e x i c a l anaphor.  Davis (1987) argues t h a t the a d o p t i o n  of the LPH w i l l l e a d t o " u n d e r g e n e r a l i z a t i o n " problems, i . e . random s c a t t e r i n g of v a l u e s of a parameter throughout the lexicon.  P a r t of the r e a s o n i n g behind the development  p a r a m e t r i c model was  of the  t o c a p t u r e g e n e r a l i z a t i o n s t h a t spread  through the syntax and the l e x i c o n , and the a d o p t i o n of the LPH i g n o r e s these g e n e r a l i z a t i o n s .  I assume t h a t the f e a t u r a l  and r u l e parameters d i s c u s s e d i n t h i s and subsequent c h a p t e r s are parameters which once s e t , a p p l y t o a l l e l i g i b l e and s y n t a c t i c  2.3.3  lexical  categories.  Non-parametric A c q u i s i t i o n While parameters are assumed t o account f o r many a s p e c t s  of a c q u i s i t i o n , they o b v i o u s l y cannot account f o r e v e r y t h i n g . Chomsky (1981a: 7-8) makes a d i s t i n c t i o n between "core grammar" and the " p e r i p h e r y " . E x p e r i e n c e -- i n p a r t , a c o n s t r u c t based on an i n t e r n a l s t a t e g i v e n o r a l r e a d y a t t a i n e d — s e r v e s t o f i x the parameters of UG, p r o v i d i n g a c o r e grammar, guided perhaps by a s t r u c t u r e of p r e f e r e n c e s and i m p l i c a t i o n a l r e l a t i o n s among the parameters of the c o r e t h e o r y . . . But i t i s h a r d l y t o be expected t h a t what a r e c a l l e d "languages" o r " d i a l e c t s " or even " i d e o l e c t s " w i l l conform p r e c i s e l y o r perhaps even v e r y c l o s e l y t o the systems determined by f i x i n g the parameters of UG. This c o u l d o n l y happen under i d e a l i z e d c o n d i t i o n s t h a t a r e never r e a l i z e d i n f a c t i n the r e a l world of heterogeneous speech communities. Furthermore, each a c t u a l "language" w i l l i n c o r p o r a t e a p e r i p h e r y of borrowings, h i s t o r i c a l r e s i d u e s , i n v e n t i o n s , and so on, which we can h a r d l y expect t o — and indeed would not want t o — incorporate w i t h i n a p r i n c i p l e d t h e o r y of UG. For such reasons as these, i t i s reasonable t o suppose t h a t UG determines a s e t of c o r e grammars and t h a t what i s a c t u a l l y ' r e p r e s e n t e d i n the mind of an i n d i v i d u a l even under the i d e a l i z a t i o n t o a homogeneous speech community would be a c o r e grammar w i t h a p e r i p h e r y of marked elements and constructions.  78  L e x i c a l a c q u i s i t i o n , word meaning, s u b c a t e g o r i z a t i o n requirements periphery. of  lexical  and exceptions w i l l a l l be aspects o f t h e C h i l d r e n w i l l have t o a c q u i r e the form and meaning  items and a f f i x e s by r o t e , although UG w i l l  c e r t a i n l y supply s t r u c t u r e and some content t o the l e x i c a l entries. While t h i s s e p a r a t i o n of core and p e r i p h e r y i s l o g i c a l and p r o b a b l y necessary, i t has sparked some concern. (1987) w o r r i e s t h a t t h i s d i v i s i o n may cause a s p e c t s o f a c q u i s i t i o n t o be n e g l e c t e d .  Williams  non-parametric  Fodor (1989) warns  t h a t we must be c a r e f u l i n assuming d i f f e r e n t  acquisition  mechanisms f o r core and p e r i p h e r a l f a c t s s i n c e t h i s may r e s u l t i n p o s i t i n g r u l e - w r i t i n g mechanisms a l l over a g a i n . suggests account  She  t h a t the n u l l h y p o t h e s i s , t h a t the same d e v i c e s t h a t f o r c o r e grammar a l s o h o l d i n the p e r i p h e r y , s h o u l d be  maintained  as long as p o s s i b l e .  In Chapter  4 I w i l l discuss  some a s p e c t s o f p h o n o l o g i c a l a c q u i s i t i o n t h a t I b e l i e v e t o be p e r i p h e r a l t o a core system of p r i n c i p l e s and parameters. I t remains an e m p i r i c a l i s s u e whether o r not UG w i l l be a b l e t o account  f o r a l l remaining aspects of the a c q u i s i t i o n data t h a t  w i l l be i n v e s t i g a t e d i n Chapter 5.  2.4  Learnability L e a r n a b i l i t y theory i s the f i e l d of i n q u i r y t h a t  developed w i t h the c o n c e p t i o n of a c q u i s i t i o n g i v e n i n (1.2) (e.g.  Gold 1967, Wexler and Hamburger 1973, Baker 1979,  Berwick 1985).  L e a r n a b i l i t y attempts  t o account  f o r how the  79  c h i l d constructs, the a d u l t grammar based on l i m i t e d i n p u t . L e a r n a b i l i t y research p r i n c i p l e s of UG  and  has  focussed  on two  areas:  the  the type of input a v a i l a b l e t o the  child.  While t h i s i s not a homogeneous f i e l d , t h e r e are c e r t a i n working hypotheses t h a t are h e l d to by the m a j o r i t y researchers.  In the f o l l o w i n g s e c t i o n s I d i s c u s s t h r e e  hypotheses t h a t I assume i n developing a c q u i s i t i o n models:  The The  the  such  underspecification  the L e a r n a b i l i t y C o n d i t i o n ,  C o n t i n u i t y Assumption and  2.4.1  of  the  the No-negative Evidence H y p o t h e s i s .  L e a r n a b i l i t y Condition  L e a r n a b i l i t y Condition  (Pinker 1979,  1984)  is  i n s p i r e d by the f a c t t h a t a l l c h i l d r e n are u l t i m a t e l y s u c c e s s f u l a t language a c q u i s i t i o n .  I t says t h a t  any  developmental stage must be a t t a i n a b l e v i a an a c q u i s i t i o n mechanism t h a t begins w i t h U n i v e r s a l Grammar and the a d u l t grammar. any  intermediate  Furthermore i t must be p o s s i b l e t o  with  convert  r u l e system i n t o the a d u l t s t a t e of grammar  by means of the a c q u i s i t i o n process. instrumental  ends up  T h i s c o n d i t i o n has  been  i n the development of the p r i n c i p l e s and  parameters model of grammar, s i n c e i f we a c q u i s i t i o n a l stages and  assume t h a t both  language-specific  d i f f e r e n c e s are  r e s u l t of the r e s e t t i n g of parameters, then by d e f i n i t i o n  the any  a c q u i s i t i o n a l stage w i l l be a p o s s i b l e human grammar. While the L e a r n a b i l i t y C o n d i t i o n appears to be an c o n s t r a i n t on a c q u i s i t i o n r e s e a r c h , f i f t e e n y e a r s has  obvious  o n l y i n the l a s t ten  i t been g e n e r a l l y accepted.  The  or  80  L e a r n a b i l i t y C o n d i t i o n would p r o h i b i t many of the grammars w r i t t e n f o r c h i l d r e n ' s speech d u r i n g the  1960s and  1970s,  which were composed of s e t s of r u l e s t h a t bore a b s o l u t e l y resemblance t o any Condition  type of a d u l t grammar . 14  t e l l s us t h a t i n t e r m e d i a t e  cannot be s t u d i e d i n a vacuum — for  how  they are achieved, and  we how  no  The L e a r n a b i l i t y  stages of a c q u i s i t i o n must be a b l e t o account  they develop i n t o the  adult  system.  2.4.2  The The  Continuity  Condition  C o n t i n u i t y Assumption (Atkinson  s t a t e s t h a t the p r i n c i p l e s and  1982,  Pinker  1984)  a c q u i s i t i o n mechanisms t h a t  are  a v a i l a b l e t o a c h i l d must be the same over the whole course of development.  Within  the p r i n c i p l e s and  the C o n t i n u i t y Assumption says t h a t UG  parameters framework contains  a l l the  p r i n c i p l e s t h a t are both necessary f o r a c q u i s i t i o n and  for a  c h a r a c t e r i z a t i o n of the a d u l t grammar. C r o s s - l i n g u i s t i c d i f f e r e n c e s are the r e s u l t of d i f f e r e n t p a r a m e t r i c  choices.  Developmental changes i n the c h i l d ' s grammar are a l s o assumed t o be the r e s u l t of parameter s w i t c h i n g ,  and  not the r e s u l t of  changes i n the c h i l d ' s c o g n i t i v e makeup. An a l t e r n a t i v e t o the C o n t i n u i t y Assumption e x i s t s i n the form of the M a t u r a t i o n a l  Hypothesis (Borer and Wexler 1987).  T h i s h y p o t h e s i s assumes t h a t language i s a b i o l o g i c a l mechanism t h a t i s not a s p e c t s of UG w i l l initiate  f u l l y s p e c i f i e d at b i r t h .  Certain  "mature" at s p e c i f i e d times, and  changes i n the c h i l d ' s grammar.  will  Even i f the  81  a p p r o p r i a t e evidence i s a v a i l a b l e t o a c h i l d a t some e a r l i e r p o i n t , a c o n s t r u c t i o n w i l l not be " a c q u i r e d " u n t i l the p r i n c i p l e r e g u l a t i n g i t appears i n UG.  T h i s view o f  a c q u i s i t i o n i s a much l e s s c o n s t r a i n e d one than the C o n t i n u i t y Assumption, s i n c e i t a l l o w s the p o s t u l a t i o n o f new  just-  matured p r i n c i p l e s t o account f o r more d i f f i c u l t a s p e c t s o f acquisition.  In t h i s t h e s i s I w i l l assume t h a t the  p h o n o l o g i c a l a c q u i s i t i o n i s c o n s t r a i n e d by the C o n t i n u i t y Assumption, and t h a t m a t u r a t i o n does not p l a y a p a r t i n t h e developmental o f a p h o n o l o g i c a l system.  2.4.3  The No-negative Evidence Hypothesis The primary l i n g u i s t i c data (PLD) o r i n p u t a c h i l d  r e c e i v e s i s c r u c i a l t o the a c q u i s i t i o n o f the grammar o f a s p e c i f i c language.  Without evidence about the language b e i n g  a c q u i r e d the c h i l d would never move on from the s e t o f unmarked parameters p r o v i d e d by UG.  The No-negative Evidence  Hypothesis (Williams 1976, Baker 1979, Berwick 1985) says t h a t o n l y p o s i t i v e evidence i s a v a i l a b l e t o language Research by Brown and Hanlon  learners .  (1970) has demonstrated  1 5  that  d i r e c t n e g a t i v e evidence i n the form o f c o r r e c t i o n s a r e r a r e and B r a i n e (1971) has shown t h a t c h i l d r e n do not make use o f c o r r e c t i o n s even i f they a r e p r o v i d e d .  The No-Negative  Evidence h y p o t h e s i s a l s o r u l e s out the a v a i l a b i l i t y o f i n d i r e c t n e g a t i v e evidence —  the d e t e r m i n a t i o n o f  ungrammaticality based on computations of the p r o b a b i l i t y o f o c c u r r e n c e i n a g i v e n amount of i n p u t .  The a c q u i s i t i o n  82  l i t e r a t u r e g e n e r a l l y assumes t h a t n e i t h e r type o f n e g a t i v e evidence i s a v a i l a b l e , although t h e r e have been t h e o r e t i c a l suggestions  (e.g. Chomsky 1981, L a s n i k 1989) t h a t c h i l d r e n may  make use o f i n d i r e c t n e g a t i v e The No-negative  evidence.  Evidence Hypothesis  applied to a  parameter s e t t i n g t h e o r y of grammar makes p r e d i c t i o n s about markedness and l i n g u i s t i c systems, i n the form o f the Subset Principle  (Berwick  1985, Wexler and Manzini  1987).  I f the  language b e i n g a c q u i r e d i s a proper subset o f a