UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The syntactic recoverability of null arguments Roberge, Yves 1986

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1986_A1 R62.pdf [ 13.71MB ]
Metadata
JSON: 831-1.0097333.json
JSON-LD: 831-1.0097333-ld.json
RDF/XML (Pretty): 831-1.0097333-rdf.xml
RDF/JSON: 831-1.0097333-rdf.json
Turtle: 831-1.0097333-turtle.txt
N-Triples: 831-1.0097333-rdf-ntriples.txt
Original Record: 831-1.0097333-source.json
Full Text
831-1.0097333-fulltext.txt
Citation
831-1.0097333.ris

Full Text

THE SYNTACTIC RECOVERABILITY  OF NULL  ARGUMENTS  By YVES ROBERGE B.A.,  Universite  de S h e r b r o o k e ,  1981  M.A.,  Universite  de S h e r b r o o k e ,  1983  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR  THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE (Department  We  accept to  STUDIES  of L i n g u i s t i c s )  this thesis the required  as conforming standard  THE UNIVERSITY OF BRITISH COLUMBIA October ©  1986  Yves R o b e r g e ,  1986  In p r e s e n t i n g  t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of  requirements f o r an advanced degree a t the  the  University  o f B r i t i s h Columbia, I agree t h a t the L i b r a r y s h a l l make it  f r e e l y a v a i l a b l e f o r reference  and  study.  I  further  agree t h a t p e r m i s s i o n f o r e x t e n s i v e copying o f t h i s t h e s i s f o r s c h o l a r l y purposes may  be  department o r by h i s or her  granted by  the head o f  representatives.  my  It i s  understood t h a t copying or p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l gain  s h a l l not be  allowed w i t h o u t my  permission.  Department o f  UNGUIS.TILS.  The U n i v e r s i t y of B r i t i s h 1956 Main Mall Vancouver, Canada V6T 1Y3 Date  )E-6  (3/81)  S ocTofeeft.  Columbia  written  OBSTRflCT Advisor: In of  Dr.  most  Michael natural  missing  processes.  syntactic It  type of  Grammar  this  parameter designed main  to  goal  that are  allow  of  verbal  agreement  results  i n the  proposal  that  and/or  object  clitics  are  agreement  licensing We NPs  similarities  of  then  in  hypothesis  the  empty  here  clitics  are  surface  and  this  can  that  various  types  null  there  to  as  need  is  doubling subject  realizations  clitic  for  of  the  regarding  the  a  systematic  clitics. subject  as  clitics  clitics  far  as  rich the  and  overt The object  same a b s t r a c t  element  existing  licensing  and  extractions  constructions  at  and  at  ii  This  and  for doubling,  LF.  through  constructions.  possibilities  S-structure  a  concerned.  symmetry the  for  languages with  agreement  for  se.  through  same  the  Universal  recoverabi1ity  the  pro  in  present  with  i s that  made  argument  i s no  per  languages  clitic  account  of  in  r e l a t i o n s h i p between c l i t i c s  so-called  defended  is  between  pronominal  examine the  the  that  object  is  investigates  embedded  recoverabi1ity  and/or  variety  mechanisms r e s p o n s i b l e  deeply  here  a  which  thesis  empty arguments  pursued  and  the  include of  found  suggests  account  subject  the  This  recoverabi1ity  i s supposed  that  s e n t e n c e may  recoverabi1ity  recoverabi1ity  and  The  a  different  type  this  languages, the  by  languages.  Rochemont  elements  possible of  S.  out  of of  between pro,  the  doubling  TABLE OF  CONTENTS  Abstract  i*  flknow 1 e d g e m e n t s  vi i  I nt r od u c t i on  1  Chapter  1: S y n t a c t i c T h e o r y and N u l l  Arguments  1.0.  I nt r o d uct i on  13  1.1.  Variation  14  1.1.1.  l.£. N u l l  and L e a m a b i 1i t y  The Government  and B i n d i n g Framework  1. 1. 1. 1. S y s t e m o f R u l e s  18  1.1.1.2.  System o f P r i n c i p l e s  £0  1.1.1.3.  Interactions  £5  Arguments  £9  I . E . 1. P r e v i o u s A n a l y s e s  34  !.£.£.  40  Zero-Topic  Languages  l.£. 3. A T h e o r y o f p r o 1.3.  On E m p t y  1.4.  Summary  Footnotes  44  Categories  47 5£  t o Chapter  Chapter £: Null £. 0.  IS  1  Arguments  54  i n Romance  I nt r o d uct i on  Languages 5  E. 1. A G e n e r a l i z a t i o n  7  58  iii  £. £.  P r o and C l i t i c s  .  &^  £. £. 1. On  pro  &4  £.£.£.  On  Agreement  77  £.£.3.  Subject  £.£.4.  Base G e n e r a t i o n o f C l i t i c s  90  £.£.5.  C a s e and  107  £.3.  On  t h e Non  £.4.  Free  £.5.  Causatives  £.6.  Summary  Footnotes  Chapter  Clitics  and  Object  Clitics  87  ©-Role  Existence of the Null  Subject  Parameter  Inversion  114 l  ^  7  *3b  t o Chapter  3:  109  On  £  138  Doubling  3.0.  I nt r o d u c t i on  151  3.1.  Subject  15£  Doubling  3.1.1.  Northern  Italian  3. l.£.  Previous Analyses  154  3.1.3.  French  159  Dialects  3.1.3.1.  Colloquial  3. 1.3.£. P i e d 3.1.4. 3. £.  3.3.  Object  Dialects  French  Noir French  Analysis Doubling  15£  160 169 170  and S u b j e c t  Doubling  178  3. £. 1. S t r u c t u r e s  180  3. £. £.  183  Case Assignment  Extractions 3.3.1.  The  198  Facts  £00  iv  3.3.1.1, 3.3.  River  Plate  £00  1. £. C o 11 o q u i a 1 F r e n c h  3.3.1.3.  Pied  3.3.1.4.  Trent ino  3.3.  Spanish  Noir  £00  French  £03 £05  1.5. F i o r e n t i n o  3. 3. 1. 6. 3.3.1.7.  £07  Romanian  £09  Standard  French  £15  3. 3. £. S u m m a r y 3.3.3.  Previous  3.3.4.  C-Chain  £15 Accounts  ££4  3.4.  A Note on Complex  3.5.  A Note on R e f l e x i v e  3.6.  Summary  Footnotes  _£17  Inversion  £33  se  £43 £51  t o Chapter  3  £54  C h a p t e r 4: C l i t i c s and Agreement  Markers  4.0.  Introduction  £70  4.1.  Subject  £71  CI i t i c i z a t i o n  4. 1. 1. I n t e r n a l 4. l.£. V e r b 4.1.3.  Structure  Raising  o f I NFL  £74  o r I NFL L o w e r i n g ?  £75  Coordination  4. £. A g r e e m e n t  £80  Markers vs. C l i t i c s  ;  £9£  4. £. 1. C o o r d i n a t i o n  £95  4. £. £. D o u b l i n g  £97  4. £. 3. E x t r a c t i o n s  300  4. £. 4. O b l i g a t o r i n e s s  301  v  4.12-5. M o r p h o l o g y 4. £.6.  Conclusion  4.3.  Historical  4.4.  Summary  Footnotes  Conclusion References  Speculations  t o Chapter  4  ACKNOWLEDGEMENTS  Le  travail  plusieurs qui  presente i c ia beneficie  personnes  j'exprime touts  dont  d e s c o n t r i b u t i o n s de  M i c h a e l Rochernont,  ma r e c o n n a i s s a n c e p o u r  collaboration.  Je tiens  Ingram,  d e rnon c o m i t e d e l e c t u r e  rnembres  examinateurs, leur  interet Pour  Diane  Massarn,  Helen L i s t , Hurtado,  Adams, G r a n t  McGill, Je  commentaires,  Alfredo  Marianne  Kayne,  Robert  annees de  Guy Garden  ainsi  et David  q u e mes  L e v i n e e t Herve  tout  Guy N e p h t a l i ,  Goodall,  Curat  pour  Hunt,  di  Jaeggli, Ian Roberts,  Sciullo.  " m e r c i " a mes c o l l e g u e s e t  Helen L i s t ,  Cathy  Osvaldo  e t Anne-Marie  de main:  Anne Lobeck, l e  Hyon Sook Choe,  simplement  ami(e)s du departements Morrisson,  s u g g e s t i o n s e t coups  Marie-Therese Vinet,  Denis Bouchard  d i s aussi  Christina  a remercier  nost r o i s  a  d a n s mes t r a v a u x .  leurs  regrete  Jeff  Richard  aussi  rnon d i r e c t e u r ,  Diane  Rodgers,  Henry  Davis,  J u m a h Hameed e t B r u c e  Bagemih1. Les  idees  developpees  i c isont  inspirees  linguistes  que j e n ' a i rnalheureusement  brievement  ou pas du t o u t ;  Luiggi  Rizzi,  J'ai la  part  Ken S a f i r  eu l a chance d e ma f a m i l l e  beneficie  de l'appui  r e n c o n t r e sque  c e s o n t s Hag i t  e t Dominique  de p r o f i t e r durant  vii  B o r e r , Noam  Chomsky,  Sportiche.  d'un s o u t i e n  mes a n n e e s  financier  d e s t r a v a u x de  a UBC.  de 1 ' U n i v e r s i t y  i n e s t i m a b l e de J ' a i aussi of British  Columbia, Canada  du c o n s e i l  de r e c h e r c h e s en s c i e n c e s  e t du f o n d s FCAC de  1' e n s e i g n e r n e n t s u p e r i e u r En d e r n i e r  lieu,  particulierement  du  l a direction  g e n e r a l e de  Quebec.  j e tiens  Diane  humaines  Massam  a mentionner pour  equivoque...  viii  tout  son soutien  sans  du  1  INTRODUCTION  Linguistic highly  o f fundamental  language-learner  value  as i t i sunderstood  structured theory  composed the  theory  of thevarious  here  o f u n i v e r s a l grammar principles  includes a (henceforth,  and parameters.  is, i na simplified  sense,  UG)  The t a s k o f  to set the  parameters on t h e b a s i s o f a v a i l a b l e  evidence. Recent that  developments  i ti snecessary  syntactic abstract account these  rules  t o avoid  i n favor  principles  f o r t h e formal  necessary  constructions had  t h e general  form  for  some e l e m e n t a r y  and fundamental  by t h e  i s to  l a n g u a g e a n d how  concerns t h e transformational works  i nthe generative language-particular  f o r thevarious syntactic  i n a given given  theory  language-learner.  t h e number o f h i g h l y  t o account found  language-particular  properties of natural  shift  have p o i n t e d o u t  of linguistic  o f t h e grammar. E a r l i e r  framework m u l t i p l i e d  theory  o f more g e n e r a l  properties a r eacquired  component  very  i f t h e goal  One e x a m p l e o f t h i s  rules  i n linguistic  language.  These  transformations  i n ( 1 ) , where t h e v a r i a b l e s  constituents of a string  stand  (e.g. s y n t a c t i c  categories).  (1)  X - Y - Z Aside  =£>  from  general  transformations, order  is  t o reduce  anywhere as  violated.  I t may be  detailed  discussion  introduction  condition trace the  forced  on  on  that not  I wish  a m b i g u o u s and  element  the other  i s always  t h e movement I return  on  a  i s often  element.  recoverable  not  a more  i n 1.1.  In  w o r k s and referred  to  to  as  a syntactic category  Chomsky  element  He  are  of deletion".  deletes  follows  of  to a  (1964)  i s deleted  from  the  assumes t h a t  because,  points is  out  generally  "recoverable" a  a  any  particular  condition  this  where  p r i n c i p l e s o f UG  in earlier  an  M o v e a,  t o move  assumed  environment. which  In  subsequent  c o m p o n e n t s o f UG  "recoverabi1ity  that  of the deleted  grammars,  to concentrate  a transformation  from  the  character.  rule called  the case that  This  structural  a sentence  status  as  of  language-particular  of the various  the  form  specific  possible  the  transformations  Suppose that given  a very  these principles.  i t s development.  condition  a  be  by  the  It i s thus permitted  long  also  category  this  on  t o t h e more g e n e r a l  a s y n t a c t i c category.  category  in  conditions  the class of  attempted  transformations  Z  t h e r u l e s had  to restrict  studies  Y - X -  deleted  (£) ft t r a n s f o r m a t i o n c a n d e l e t e a n e l e m e n t o n l y i f t h i s element i s t h e designated representative of a category, o r i f the s t r u c t u r a l condition that defines this transformation states that the deleted element i s s t r u c t u r a l l y identical to another element o f t h e transformed string. CChomsky ( 1 9 6 4 s 4 1 ) 1 In t h e sense j u s t recoverabi1ity grammatical formulated  (3)  expressed,  of deletion  i s a c o n d i t i o n on a t h e o r y  transformations.  Such  a s i n ( 3 ) ; s e e Chomsky  Only recoverable t h e grammar.  ft d e l e t i o n Let  t h e c o n d i t i o n on t h e  illustrate  be  (1965s£££).  deletions are permitted  i s recoverable  us c o n s i d e r  a condition could  of  in  under t h e c o n d i t i o n s  the following English  i n (£).  sentences t o  how t h e c o n d i t i o n o n r e c o v e r a b i 1 i t y o f d e l e t i o n  funct ions. (4)a.  Want s o m e  coffee?  b.  Want t o s e e a m o v i e  tonight?  c.  Getting  d.  She f i n i s h  her thesis  e.  Looks  an  f.  Turns out you can't  pretty excited,  like  aren't  yet?  accident. do t h a t  These s e n t e n c e s a l l have one t h i n g clause  are missing,  (4a,b,c),  you?  namely  an a u x i l i a r y  i n Texas. i n commons  an a u x i l i a r y  some p a r t s  and a s u b j e c t  i n ( 4 d ) , and a s u b j e c t  in  i n (4e,f).  of the  Nevertheless Schmerling are  Akmajian,  D e m e r s &• H a r n i s h  (1973) f o r (4e,f)  q u i t e common i n i n f o r m a l  concerning  them  i s that  without  interpreted  as equivalent  i s expressed  (5) a. Do y o u w a n t  speech.  the missing  "recovered"  what  note that  ambiguity.  sentences  The i n t e r e s t i n g e l e m e n t s c a n be  in  You a r e g e t t i n g p r e t t y e x c i t e d , a r e n ' t  d.  Did she f i n i s h  e.  It looks  like  f.  It turns  out you can't  Does he want  c.  We  t o see a movie  d.  Will  e.  He  f.  She t u r n s  her thesis an  looks  In a t h e o r y  like  do t h a t  i n Texas.  tonight?  her thesis accident. you can't  do t h a t  includes condition  o f r e c o v e r a b i 1 i t y i s a t work  sentences  i n ( 4 ) . Now,  operation  i t must  because these  as expressed  i n Texas.  ( 3 ) , we c a n a s s u m e  that  i n interpretingthe  sentences a r e not  f u r t h e r be assumed  i s involved,  you?  yet?  some t y p e  ambiguous,  you?  coffee?  out that that  mean  yet?  pretty excited, aren't  an  i n (4) a r e  tonight?  t o see a movie  she f i n i s h  easily  accident.  some  are getting  fact  coffee?  c.  b.  these  (6).  Do y o u w a n t  want  as  t o t h e o n e s i n (5) and c a n n o t  b.  (6) a. Do t h e y  such  Thus t h e s e n t e n c e s  by t h e s e n t e n c e s  some  ( 1 9 8 4 ) f o r <4a-d) a n d  that  some  systematic  i n (£'). I n e f f e c t ,  i t  i s  not  the  case  that  any  element  can  be  freely  missing  unambiguous r e c o v e r a b i 1 i t y i s p o s s i b l e . T h i s (7)  below  (7) a . *  Last  b. * Do c. # The  from  Akmajian,  night's  want  party  some  Is f a i l i n g  Demers & H a r n i s h not  went  too  h i s courses, of  sentences  indication  that  be  invoked  ( 7 ) . What h a p p e n s t h e r e  missing  the  specifications  the  the  isn't  an  the  when  is illustrated  in  (1984).  well.  coffee?  ungrammaticality  in  even  element  systematic  cannot  be  relative  to  he? in  (7)  operation  be  taken  alluded to  i s not  recovered. the  must  It  that  the  i s more  systematic  as  cannot nature  likely  operation  of  that  are  not  met. In  a theory  sentences the ones (8)  such  application in  (8)  i n c l u d i n g (£) as of  the  ones  very  R u l e 1: To v e r b s (do, contracted  Question  the  non-ambiguity  (4a,b,d) would  be  Demers & H a r n i s h  of  attributed as  to  the  (1984s304).  Rules  form an a b b r e v i a t e d q u e s t i o n , t h e a u x i l i a r y h a v e a n d be) c a n be d e l e t e d , o r e l s e t h e y c a n onto the subject you.  R u l e £s I n f o r m i n g p e r s o n s u b j e c t you verb i s contracted Besides  (3), the  l a n g u a g e - s p e c i f i c r u l e s such  from Akamajian,  Abbreviated  in  and  blocking the  an a b b r e v i a t e d c a n be d e l e t e d onto i t .  sentences  f u r t h e r consequence of  in  question, the as l o n g as an  (7a,b),  allowing the  the  be  second auxiliary  rules in  (8)  r e c o v e r a b i 1 i t y of  have the  6 missing  element(s).  In other  words,  ( 4 a , b , d ) c a n be i n t e r p r e t e d o n l y  given  <£), ( 3 ) , and  (S) ,  a s i n (5a,b,d) and not a s i n  ( 6 a , b, d ) . The  important  possible, rules  point  here  i s that  i n the theory  just  o u t l i n e d , through  applying  Let ( 9 ) a.  only  i n a given  u s now c o n s i d e r  r e c o v e r a b i 1 i t y i s made very  specific  language and i n a g i v e n  the sentences  in  context.  (9).  Italian Ho trovato i l libro. have-Is found t h e book ' (I) have found t h e book.'  b.  Swahili a - l i -mi -ona S U B J ( 3 s ) PAST O B J ( I s ) s e e '(He) saw (me).'  The of  general  t h e ones  comment  i n (9),that  missing  but t h e c o n t e n t  without  ambiguity.  considered  concerning  i n (4) h o l d s  i s , some e l e m e n t s a r e ( i n a of these  The s e n t e n c e s  as equivalent  the sentences  elements  sense)  i s recoverable  i n (9) c a n t h e r e f o r e n o t be  i n meaning  t o the sentences  in  (10a,b). ( 1 0 ) a. b. It  He h a s f o u n d They saw y o u .  seems o b v i o u s  that  t h e book,  permits  that  missing  something elements,  systematic on one hand,  i s happening and  their  i n (9)  7 recoverabi1ity, To (9)  on t h e o t h e r  illustrate  hand.  the position  i t i s possible to establish  sentences  i n (9) and  t o the sentences  missing  but i t i s , i n a sense,  the tag question  the  sentence.  affix  a certain  with  respect  to  between t h e view,  (4c) i s very  i n (9): the syntactic subject i s still  (4c), the subject  In (9a), the subject  representing  The  a parallel  ( 4 c ) . From  similar  In  we a r e a d o p t i n g  subject-verb  same c a n be s a i d  present  i n the  sentence.  you a p p e a r s a t t h e end o f  i s represented  by t h e v e r b a l  agreement.  o f an E n g l i s h sentence such  as t h e one  in (11). (11)  John wanted  Part  before  Mary d i d .  of the predicate associated  "missing" is  t o leave  still The  from  i t susual  there,  present  same h o l d s  represented  with  t h e argument  syntactic position i n t h e same  o f (9b) where  by t h e s u b j e c t  Mary i s  but t h e m i s s i n g  VP  sentence.  the subject  and t h e o b j e c t a r e  marker and t h e o b j e c t  marker on t h e  verb. We  thus  syntactic (9)), is  consider  argument  where  their  a property  property In  here that  p o s i t i o n s (such content  as the "null  languages.  argument"  of having  empty  a s s u b j e c t s and o b j e c t s i n  i s recoverable  o f some n a t u r a l  a theory  the possibility  i n a very I will  local  refer  sense,  to this  property.  i n c l u d i n g (£) a n d  (3),a p l a u s i b l e account  of  8 the  property  existence Lasnik also  of a subject  (1977).  But  r u l e s have  especially component  and  foundations  of  t h e pure  Italian  i s reduced  and must  i s now  pronominal  that  work  i n the consequences  as t h e one presented  t o Move  above  a, t h e n ,  clearly,  of transformations, by more  i s  fundamental  or condition(s).  often  i n ( 9 ) , namely t h e n u l l  attributed to the availability  empty c a t e g o r y  "pro" i n languages  the phonologically  of a phonologically realised  proposed  language-particular  h a s some  be r e p l a c e d  and S w a h i l i . P r o i s s i m p l y  counterpart widely  clearly  see i n l.S, the property  property,  the deleted  t o (£). I f t h e t r a n s f o r m a t i o n a l  principle(s)  we w i l l  argument  shift  i n recent  i s a c o n d i t i o n on t h e form  general As  respect  o f t h e grammar  which  without  This  explicit  o f r e c o v e r a b i 1 i t y such  with  &  d e l e t i o n would  rule(s) renders  earlier,  been a v o i d e d  framework.  a theory  o f such  i n Chomsky  recoverable.  a s we h a v e s t a t e d  generative  be t o p o s t u l a t e t h e  d e l e t i o n r u l e as proposed  The form  and o b j e c t s  syntactic  (3),  i n (9) w o u l d  I n (9b) a f u r t h e r r u l e o f o b j e c t  be r e q u i r e d .  subjects  for  illustrated  the verbal  agreement  pronoun. somehow  like  null  It i s also permits  the  presence of pro. Thus, deletion It not  i t appears that i s involved  follows that  no  language-part  i n the generation  icular  of the sentences  the recoverabi1ity of the missing  a t t r i b u t a b l e t o (2) a n d  (3) a b o v e .  rule of  Something  in  (9).  elements i s more  general  9 is  involved. Before  examining  recoverabi1ity whether the  of  type  more c l o s e l y what  the of  missing  elements  recoverabi1ity  d i f f e r e n t from  the  words,  (9)  analysed  believe  that  h a v e no  direct  However,  a  should In used in  to  (9) The  type  i s probably  evidence  to  put  i n the  the  rest  refer but  of  to  not  forward  to  the  deletion  be  possible,  case of  of  UG  thesis  the  term  in  light  (£)  and  of  the  of the  (3).  fundamental  is  In  other  same way?  I  this  shift  from  rule to  the  is  missing "null  objects  I  claim.  a  language-  assumption  involved  elements  in  argument" such  as  (4).  will  the  be  ones  of  natural  languages  p r i n c i p l e s must  deletion.  It should  above d i s c u s s i o n ,  I propose to  Recoverabi1ity  also  include  apply  therefore  to in  replace the  Principle in  (1£).  Principle  fin e m p t y c a t e g o r y to  and  fundamental  recoverabi1ity  the  (4)  (9).  support  specific properties  general  Recoverabi1ity  According  the  subjects  ask  (4).  i n the  statements  for  missing  that  to  account  this  in  proposal  the  an  the  can  in  in  also  provide  we  in  a f f i r m a t i v e even though  p l a u s i b l e that  (or object)  found  pronominal  in  (1£)  be  (9)  u n i v e r s a l l y a v a i l a b l e empty  attributed  theory  answer  subject  be  the  (4)  i t is highly  particular that  the  and  in  illustrated  intrinsically should  i s involved  must  be  licensed.  Recoverabi1ity  P r i n c i p l e , the  presence  of  an  empty c a t e g o r y  sense, In  in a syntactic structure  be  permitted.  the  present  period  generative  framework,  categories  can  of  UG»  A  of  M o v e a,  trace,  Chomsky  the  we  by  The  the  d i f f e r e n t p r i n c i p l e s or  the  left  Empty C a t e g o r y  only  that  condition  by  application  P r i n c i p l e of  a trace  on  theories  the  must  be  ungoverned.  up  to  now,  must  be  goals  of  this  thesis  Pro,  licensed  be  presence of  in a syntactic structure  i t o c c u p i e s must  have seen  in  certain  d i f f e r e n t empty  empty c a t e g o r y  a n a p h o r PRO  in a  research  the  principle states  governed".  pronominal  by  i s the  is licensed  position  what  licensed  This  linguistic  i t appears that  which  (1981).  "properly empty  be  of  must,  is  the  that  according  by  to  verbal  agreement. One  of  the  characterization constructions. Principle  apply  licensing  of  role  i n the A  in  (9)  this  pro  pro? and  recent  It w i l l PRO,  to  are  here  rather  (1.1).  be  in  a  formal  various  does the  Recoverabi1ity  suggested  that,  government  i n the the  to  well  I present  how  propose  theory  as  plays  with  a  the  central  pro.  work  i t is fair  is licensed  words,  l i c e n s i n g of  thesis,  relevant  to  i s devoted  and  involved  how  In o t h e r  trace  l o t of  framework  of  i s to  study  say  that  of  the  the  understood.  the  This  Government  features  and  property  general In  of  presentation  Binding  the this  illustrated  mechanisms  first  chapter  framework that  i s closed  in  1.£  by  of are  brief to  r e v i e w s o f some o f t h e t h e o r i e s  account f o r the missing  development that the  follows  proposal  that  recoverabi1ity. of  some  issues  Chomsky missing  a n NP a s s o c i a t e d  (p.£78)s  "a c l o s e  recoverabi1ity extent Chapter  the property  verbal  this  statement  i n which clitic-argument  there  symmetry with  pairs are similar.  i n the literature. existing  respect  holding In  the third  NP g e n e r a l l y  between  clitics, We  also  subject  these  Most  referred  focus  c l i t i c -  and B i n d i n g  theory".  (£.1, £.£, £ . 3 ) , a n d I will  examine  verbal  of the discussion  which have  received  less  c l o s e l y examine t h e  clitics  and o b j e c t  clitics  and t h e r e l a t i o n s h i p  two t y p e s o f c l i t i c s  chapter,  and  considerable  p a i r s and  to the syntactic structure  between  between  (£.4, £ . 5 ) . I n p a r t i c u l a r ,  i s devoted t o subject  attention  t o be e m p t y  i s to a  t h e Government  the  agreement-argument  c l i t i c  can also  and t h e c o n d i t i o n o f  isssues  way  discussion  allowing the  agreement,  i s established  some r e s i d u a l exact  of syntactic  i n 1.3 b y a  f o r pronouns  £ i s concerned with  governed t o  categories.  agreement,  within  historical  from t h e b e l i e f  example  an o b j e c t  relation  of deletion  accommodated  a good  out that  with  NP p a i r s a n d s u b j e c t - v e r b  been proposed  i n (9) was r u l e  t o empty  points  leading  i s followed  elements i n ( 9 ) , namely  permit says  section  relating  (1981)  path  involved  i t i s i n fact  This  have  e l e m e n t s i n ( 9 ) . The  the natural  the recoverabi1ity  that  and empty  i s put on t h e p a i r  t o as " c l i t i c doubling".  We  arguments. clitic-overt  explore  the  consequences o f t h e theory  developed  i n t h e second  problems r e l a t e d t o c l i t i c  doubling.  The m a i n t o p i c s  concern (3.1,  subject/object  (a)symmetries with  3 . 2 ) , and t h e p a r a m e t e r i z a t i o n  possibilities  out o f subject  constructions  (3.3).  some a p p a r e n t l y Inversion The  of  subject  explore  part  constructions,  of the last  onto the verb  the possibility  se  chapter  the syntactic operation clitics  and o b j e c t  ends with  reflexive  the  i s that  same r o l e  they  are different  i n t h e grammar,  markers are s y n t a c t i c devices recoverabi1ity  of null  arguments.  namely,  of  Complex  some  proposals  i n the cliticization  clitics  part,  I  and agreement  The c o n c l u s i o n even though  is, clitics  necessary  doubling  In t h e second  entities  that  doubling  an e x a m i n a t i o n  provides  (4.1).  markers as a s i n g l e s y n t a c t i c notion. there  addressed  (3.5).  involved  of analyzing  to  on  of the extraction  doubling  chapter  problematic  (3.4) and F r e n c h  first  regarding  This  respect  chapter  reached they  have  and agreement  f o rthe syntactic  CHAPTER  1.0.  I nt r o d u c t i on 1 present  different put of  1: SYNTACTIC THEORY AND NULL ARGUMENTS  here  thetheoretical  analyses  on t h e system t h e Government  a r e based. of rules  framework on which t h e  In the first  and p r i n c i p l e s  and B i n d i n g  framework  section  constituting  as developed  (1981).  Some o f t h e m o d i f i c a t i o n s t o t h i s s y s t e m  Chomsky  (1982),  introduction  into  corresponding assumed the  property. government  We w i l l  on  see that  occupying  thefirst  part  therelationship  variation  analyses  of thesubject  empty c a t e g o r y  pronominal,  The a i m o f t h e second  to different  these  presented i n  and i n Chomsky section  analyses  have  (1986) a r e  i st o introduce null  focused  subject ont h e  by INFL and on t h e t y p e o f  thesubject  o f t h i s chapter,  as i t i sconceived  i n Chomsky  category  of theso-called  position  between  t h e core  t h e one concerning t h e  t h e framework o f t h e empty  t o an overt  as well.  reader  In  i nparticular  emphasis i s  position. I present  a discussion  l e a r n a b i l i t y and c r o s s - l i n g u i s t i c o f i n t h e framework  adopted i n  this  1.1.  thesis.  Variation T h e way  and  Learnability  i n which  a specific  u p o n how t h e l a n g u a g e must  be h a n d l e d s  individual  faculty  language  <learnabi1ity)  languages and d i a l e c t s . It  i s generally  t h e o r y i s used  c a n be d e s c r i b e d .  development  a s we w i l l  f o r one b e a r s consequences  example,  i f one w i s h e s t o account  providing the  grammar,  account  can explicate  on t h e b a s i s  formulation  setting.  evidence  a "trigger  availability range  parameter trigger  learner  of positive  difference  i t i s fixed i n  be a b l e  into  1  analyses.  That  i s ,the  account t h e  that  must  ( 1 9 8 £ ) , we c a l l this  i s no way  f o r which  this  a  type of  of  on t h e limiting  proposed  no  s h o u l d be r e j e c t e d  i n which  be u s e d f o r  constraint  I n o t h e r words, i.e.  also  to f i x the  experience has t h e e f f e c t  i s unleamable,  there  take  For  t h e n o n e must  evidence.  constructions  experience i s available,  grammar s i n c e  will  related.  a n d d o e s s o by  o n how  experience". In turn  of possible that  must  Light foot  of a trigger  among  the explanation  fora certain language  tasks  i nthe  f o r the other.  the difference,  of particular Following  see, that  depending  o f t h e parameter  accessibility  the  which,  f o r how t h e l a n g u a g e  parameter  its  o f t h e same  a parameter  o r growth  major  t h e two a r e i n t i m a t e l y  provided  between two d i a l e c t s  Two  and g r a m m a t i c a l v a r i a t i o n  In fact  t h e case,  i s dependent  plausible from t h e  i t c a n be f i x e d  i n one  way  or  the  variation  other. and  following  Many e x a m p l e s o f  language  growth  i t i s not  obvious that  handled  straightforwardly  assumed  here  for  A.  one  we  while  being  cf.  exposed (198£),  Chomsky  attains  a  (1984).  In  other  represents  goes  far  beyond  by  knowledge what  h a v e some s p e c i f i c  enough  to  grammars o f  further  include  ( o r more) evidence  by  the  can  grammar  & L igh t foot  argue that  a  structures can  child  and  be  deduced some  knowledge, (henceforth,  properties  in order  to  I t has  like  set t i n g s to  language  to  only  (possible)  some n o t i o n  p o s s i b1e  how  Grammar  f o r a l l and  the  so  Universal  program.  allow  accounted  assume t h a t  i s available. This  is called  be  (1966),  what  thus  theory  Chomsky,  adult  Kat2  i t s language that  innate  to  Hornst e in  They  i s the  words,  the  among o t h e r s ,  environment.  be  language with  deficient stimuli.  plausible research  positive  the  Noam  problem  complex  must  t wo  central  then to  and  particular  has  How  rich  represented  must  theory.  the  of  abstract  in  r e l a t i o n s h i p can  introduction to  knowledge of  kind  a  presented  much a b o u t  that  (197551980b),  i t s linguistic  in  state  the  from  UG),  the  know s o  to  this  any  Chomsky  steady  (1981),  of  be  t h i s ? According  others,  come t o  the  L ig h t foot  rules  and  evidence;  reach  by  accomplishing  Fodor,  i s how  little  will  r e l a t i o n s h i p between  chapters.  Now,  Jerry  this  be the  be  general  included and  possible  natural  languages.  that  parameter  be  learner.  of  f i x e d on  the  UG that  bas i s  Alternatively, a  of  16 parameter can particular  already  be  set  in a  given  s e t t i n g c o n s t i t u t e s the  evidence  i s needed  to  Concrete  examples of  modify this  this  type  way  i n UG.  "unmarked" setting to  are  This  case. the  presented  Positive  "marked"  i n the  case.  following  chapters. Pi h u m a n b e i n g constitutes goes from types of the  one  as  Fodor the  to  the  to  of  a w e a k e r one  "mediated  by  eventually Now,  variation  of  (1960c).  set  of  unless  the  of  the  among t h e such  The  state  accept  that  to  the  the  idea  of  any  of  a conception  with of  and  as  genetic is  the  In  fact,  learning  is  surrounding  on  the  evolution that  then  1980s 152).  i n the  i n d i v i d u a l and  world,  i t will  example.  a  grammar,  basis  can  (Fodor  an  it  is  r e l a t i o n s h i p between  the  component  this  beliefs"  language  languages of  the  inference,  concept  organism's  UG,  next.  a r i c h e r system  assumes t h a t  of  that  the  UG  reached!  alone,  to  acquire  that  different  is  theory  environment  make c l e a r t h e  transformational  state  assumption  one  the  inductive  one  growth  of  b e l i e f s by  i n the to  basis  b e l i e f s about  availability up  the  from  i f we  the  state  p r i n c i p l e s of  a steady  linguistic  impossible  turn  illustrate The  the  some s e t  i n order  conception  Chomsky  evolution  another  logically  to  to  from  on  settings until  argues that  is  other  makes e x p l i c i t  i n the  passage  world  state  opposed  (I960)  initial  g e n e t i c a l l y encoded  here,  main t o o l  an  of  grammar; c f .  presented  s t a r t s o f f at  set  parameter  adult  code,  the  then  be  which  this the  useful  has  the  17 universal learner  form  Move a  i n the  initial  to  "learn"  or  c a t e g o r i e s can  could a  be  prase can a  do  so.  of  two  deduced  initial the  fact  becomes more s p e c i f i c ,  when, will  and  how  M o v e Wh  provide the  trigger of  state.  moving  that  the  Wh  elements  those  etc.)  and  those  sense  i n which  cross-linguistic  one  does not  o n l y which  experience  and  language  in this  a minimal  a sentence  From  these  two  n a m e l y M o v e Wh.  case  pair,  say,  with a  Wh-  sentences  other  answers.  i s not  general  i t  arise,  principles  then  the  even considered.  (Japanese,  principles  and  I t i s now  etc.).  parameters  cf. Light foot  type  clear of  French,  That  o f UG  (1982)  UG of  types  (English,  Chinese,  of  possibility  between two  i n syntax  Move  where,  Suppose t h a t the  above d i s t i n g u i s h e s  variation?  Pis t o why,  have  category  t h a t W h ~ e l e m e n t s u n d e r g o movement.  applies,  without  but  forming  position.  w i t h Wh-movement  the  moved  in situ  above does not  discussion  languages:  be  sentences  necessary  experience  t o any  In o t h e r words,  Pi t r i g g e r  s  w i t h a Wh~phrase  in clause  be  is available  that c a t e g o r i e s can  composed  sentence  i n UG,  can  i s the reflect  for other  examples. Many in  the  instances of t h i s  following  chapters,  languages  differ  syntactic  arguments.  main modules of  type  of  relationship  especially  with respect to the  UG  But  first,  assumed  with regard  rest  see  presented  to  recoverabi1ity  l e t us  i n the  are  how of  i n more d e t a i l  of the  thesis.  the  18 1.1.1.  The  1.1.1.1.  Government and  System o f  The m a i n g o a l of  Binding  Framework  Rules o f Chomsky  the distribution  o f NPs  and show  grammar  i s a c t u a l i z e d by t h e f o l l o w i n g r u l e  (1)i. ii. i i i .  lex icon syntax a. c a t e g o r i a l c o m p o n e n t b. t r a n s f o r m a t i o n a l c o m p o n e n t p h o n o l o g i c a l form component  iv.  logical  system  (£)  Move a  analysis  particular  character  to  of the  system.  -  component  i s organized  D-structure  This  an  a modular approach  can achieve  form  task.  how  grammar  The  this  (1981) i s t o p r o v i d e  i n UG  i n the following  way:  (i,iia) ( i i b)  S—structure PF It  (iii)  f o l l o w s from  independent (PF). i.e.  LF ( i v ) (£) t h a t  S-structure  c o m p o n e n t s : L o g i c a l Form  T h e r e i s no t r a n s p a r e n c y an o p e r a t i o n  taking  the  operations  taking  the  properties of these  and  4.  place,  place  representations  two  (LF) and P h o n o l o g i c a l  at t h i s  level  of the  s a y , i n PF h a s no  i n LF.  feed  I will  Form  grammar,  influence  on  h a v e more t o s a y on  two components below and  i n chapters  3  19 Pit t h e D - s t r u c t u r e includes the  specifications  lexical  These cf.  items  lexical  we a s s u m e t h e f o l l o w i n g  COMP'' » COMP' —> I N F L ' ' —~> INFL' —>  The  S p e c COMP p o s i t i o n  i n t e r m s o f t h e X' s y s t e m ;  complementisers,  f o r example  o f INFL''  often referred  items  major  D-structures are  i nthe syntactic  expansions:  COMP o f COMP''  ( o r IP) o c c u p i e s  n o d e c o n t a i n s t h e PlGR ( e e r n e n t )  a n d TNS  structure  o f (3) i s g i v e n  INFL  argument this  The  position; i t  of the verb.  3  s u b j e c t . The INFL  (tense)  n o d e s ; s e e 4.1  o f INFL The t r e e  i n(4).  Wh  ( o r CP) c o n t a i n s t h e  t h e Spec  t o as the external  more o n t h e i n t e r n a l  f o r moved  t h a t and f o ri n E n g l i s h .  NP d o m i n a t e d b y I ' ' t o r e p r e s e n t  representation  (1977).  i sthe landing site  t h e head  for  features.  Spec C COMP' COMP INFL'' Spec I I NFL' INFL VP  elements whereas  use  structures of  p r o d u c e d b y t h e X' s y s t e m . F o l l o w i n g C h o m s k y  (3)  is  which  and c a t e g o r i a l  (197£;1986) a n d J a c k e n d o f f  representations  subject  morphological  syntactic  a r e expressed  p r o d u c e d by i n s e r t i n g  (1986),  i s found t h e l e x i c o n  concerning  and t h e i r  specifications  Chomsky  thus  level  I  (4)  C  '  / Spec  \  C  COMF" /  \  C  ji i NP  /  \  /  I' \  I  VP  /  \  INS  AGR  The  highly  earlier Lasnik  rule  versions  moved  the on  the  the  of  element Move a  1.3.  of  (1977) t o  application(s)  s p e c i f i c transformational  always  LF,  semantic  Move a  elements. bearings  also  the  of  Contrary on  for to  the  Chomsky  (1981).  1.1.1.2.  System o f system  of  M o v e a.  D-structure coindexed  the  and  LF.  of  i s the  the  PF  the to  i n Chomsky  result  of  and  raise  of  &  the  itself;  h a v e no  the  constituents;  structure.  LF  May  and  and  In Wh~  movements  structure cf.  affect  effects  quantifiers  i n PF,  A  cf.  movements o n l y  syntactic  case  of  the  i s S-structure.  trace  constituents  example,  of  The  of  not  have on  (1977),  Principles principles  of  of  number o f  certain  a  been reduced  interpretation  organization a  a  i n PF  what  s u p e r f i c i a l order  The  to  interpretation  i s used,  only  rule  leaves  s u p e r f i c i a l order the  has  unique r u l e  this  applies  model  component  grammar.  constitutes  The  these  GB  the  framework  principles  and  internal is  in  their  fact  composed  interactions  with  t h e d i f f e r e n t l e v e l s and components  above. (5)a. b. c. d. e. f. A.  These  i n <£)  principles are the following:  Bounding Theory Government Theory 8-Theory Binding Theory Case Theory Control Theory  The B o u n d i n g  Subjacency  theory  i s one such  applications  defined  in  Chomsky  accounts f o r l o c a l i t y condition,  which  conditions.  constrains the  o f Move a; i t i s n o t a c o n d i t i o n o n  representations;  (6)  illustrated  c f . Chomsky  <198£) f o r a d i s c u s s i o n .  Iti s  (6). (1973)  In t h e s t r u c t ure s no r u l e c a n r e l a t e X and Y i f a and b are bounding nodes. Sportiche Standard on  (1981)  French a r e C',  contrasts  examples (7) a.  demonstrates that  between  are given  t h e Bounding  nodes i n  NP a n d P P . H i s a r g u m e n t a t i o n i s b a s e d  s e n t e n c e s i n v o l v i n g m o v e m e n t . Some  i n (7)~(9)  with  Wh-movement.  T u c o n n a i s l'homme q u i a v u x_. ' Y o u k n o w t h e man who s a w x . '  b. * Q u i t 'Who  connais-tu [ l'homme C « • q u i a v u t_± 3 3 ? d o y o u k n o w t h e man who s a w ? ' N  P  c  (8) a.  I I c r o i t l e f a i t q u e t u a s v u x_. 'He b e l i e v e s t h e f a c t t h a t y o u s a w x . '  b. * Q u i t c r o i t - i l Cwi=. l e f a i t C » • q u e t u a s v u t _ i 3 ] ? 'Who d o e s h e b e l i e v e t h e f a c t t h a t y o u s a w ? ' c  (9) a . * V o i l a q u e l q u ' u n a q u i i C x » » j e s a i s [ >> l e q u e l j E x . . .je c r o i s [ - »ti q u e E x . . j ' o f f r i r a i s t _ j t _ 3II3 ' H e r e i s s o m e o n e t o whom I k n o w w h i c h o n e I t h i n k I w i l l offer.' c  c  b.  In  A  V o i l a q u e l q u ' u n a q u i i Ex«« j e c r o i s [ < < t i q u e E x . . j e s a i s C c » l e q u e l j E x . . j ' o f f r i r a i s t_j t _ i 3 3 3 3 ' H e r e i s s o m e o n e t o whom I t h i n k t h a t I k n o w w h i c h o n e I will offer.' c  (7b) and (8b) t h e t r a c e  since  they a r e s e p a r a t e d from  namely I'',  cannot  C'  i n t h e ungrarnmat i c a l  and  two I ' ' nodes.  ungrarnmat  node  node.  sentence  but o n l y  i n French then  This  node.  (9b) would  and n o t  i s so because  (9a) c r o s s e s  one C '  C'  nodes,  two C '  a  nodes  i n (9b), aq u i  I f I ' ' were a  be e x p e c t e d t o be  ica1.  B. T h e n o t i o n  Government  plays  role  a major  i scentral  i n t h e GB f r a m e w o r k  i nthe discussions  chapters.  Since the f i r s t  (Rouveret  &• V e r g n a u d  have  proposed.  been  o t h e r by t w o B o u n d i n g  In t h e grammatical sentence  two I ' ' nodes  bounding  t o i t s antecedent  i n (9) shows t h a t  i s a Bounding  qui  crosses  each  a n d NP. T h e c o n t r a s t  as i n English,  be r e l a t e d  I adopt  Sportiche  (1981)  reviewing  thedifferent  of the following  attempts at defining  1980 and Chomsky  as modified  this  1 9 8 0 a ) many  here t h e d e f i n i t i o n i n Chomsky  alternative  and  modifications  given  (1981:165)  proposals.  concept  i n ftoun &  without  <10)  Government a g o v e r n s 6 i n a s t r u c t u r e C ...6...a...6...1 where, i. a= X ° ii. 0 i s a maximal p r o j e c t i o n , i f0 dominates 6 then 0 dominates a iii. a c-commands 6 ra  In  general  holding that  t e r m s , Government  between  "depend"  t h e head  o n i t . We  i s defined  as the r e l a t i o n s h i p  o f a p r o j e c t i o n and t h e p r o j e c t i o n s can simply  say that  a head  governs  i t s  com p 1 e m e n t s . It  c a n be d e d u c e d  absolute nodes  from  -  d o m i n a t e d b y Y''  i f Y''  projections are  In (11), X cannot govern t h e  i s a maximal  projection.  X'' /  \  \  /  X  Y' / \ Z  Government (henceforth,  C.  maximal  b a r r i e r s f o r Government. *  (11)  trace  (10) t h a t  must  ©-theory  holding  Y  theory  also  ECP) o f Chomsky be p r o p e r l y  between  According  governed; c f . also  that  an argument  Principle  t o t h e ECP, a  (Kayne 1979; 1981b).  the thematic  arguments and p r e d i c a t e s .  relations  Informally  i n a ©-position  a ©-role d e t e r m i n e d by i t s g r a m m a t i c a l  ©-role o f a n a r g u m e n t occupies  t h e Empty C a t e g o r y  (1981).  aims at d e s c r i b i n g  ©-Criterion r e q u i r e s assigned  includes  stated the be  function.  The  i s d e t e r m i n e d by t h e ©-position i t  at D-structure.  I n (1£) t h e t r a c e  o f John but not t h e  NP  John occupies  (1£) The  John*  9-position assigned  C t_± t o  seemed  ©-Criterion  (13)  a  like  i s defined  as  the in  a  9  role.  cheesecake  (13)  3.  i n Chomsky  (1981:335).  9-Cr i t e r i on G i v e n t h e s t r u c t u r e S, t h e r e i s a s e t K o f c h a i n s , K= - C C i > , w h e r e C = (a * , . . . , ), s u c h t h a t : (i) i f a i s a n a r g u m e n t o f S, t h e n t h e r e i s a C± € K s u c h t h a t a= aj a n d a 9 - r o l e i s a s s i g n e d t o C± b y e x a c t l y o n e p o s i t i o n P. ( i i ) i f P i s a p o s i t i o n o f S m a r k e d w i t h t h e 9 - r o l e R, then there i s a C € K t o w h i c h P a s s i g n s R, and e x a c t l y o n e o i j i n C» i s a n a r g u m e n t . 1  t  1  A  1  D.  Binding  sense that defined  the  reference  domain where a B i n d i n g  i s another govemee,  There are (14)  the  makes c r u c i a l  under Government.  governed and  theory  three  The  category  namely the  Binding  Government  relation  domain which  to  i n which  contains  governing  in  the  applies is a category  both  the  is  governor  category.  conditions:  Binding Conditions Pi. Pi C + a n a p h o r D e l e m e n t m u s t be fl-bound i n i t s governing category. B. Pi C + p r o n o m i n a l 3 e l e m e n t m u s t be P i ~ f r e e i n i t s g o v e r n i n g category. C. Pin R - e x p r e s s i o n ( v a r i a b l e s , n a m e s ) m u s t be f r e e . 2 5  Reference to cannot would  be say,  is  not  to  binding  Pi—binding i n d i c a t e s that  reduced  to  a mere c o n d i t i o n on  f o r example,  coindexed, by  an  and  the  that  bound  element  an  anaphoric  otherwise.  i n argument  Binding  conditions  coindexation element  Rather,  position.  that  i s free  i f i t  Pl-binding r e f e r s It follows  that  an element that E.  i s "bound"  Vergnaud  elements parts:  Chomsky  and  developed  (1980a).  as a r e a l i z a t i o n  an  fl-position  on NPs,  i s assigned very  important  simple role  Case F i l t e r *NP E - C a s e H i f NP filter  movement  applies  and  containing  and  £)  a  under Government  be  two  important  well-formedness assumed  b y +TNS, P,  t h e Case  that  and  Filter  V. plays  grammar.  has a p h o n o l o g i c a l  at S—structure  o r PF  e s s e n t i a l l y expresses that  a non  can  It i s generally  i n i t sformulation,  i n the  theory  It contains  t h e Case F i l t e r . t o a n NP  Case  i n Rouveret &  o f t h e r e l a t i o n s h i p s among  1) C a s e - a s s i g n m e n t c o n t e x t s  Although  The  introduced  in a syntactic structure.  condition  (15)  was  (1980) and  interpreted  an  with  c-commands i t .  Case theory  Case  i f i t i s coindexed  Case-marked  matrix. but a f t e r s y n t a c t i c a syntactic  phonologically  structure  r e a l i z e d NP  i s  u n g r a m r n a t i c a 1. B.  Control  of  the  required  i t ssyntactic  discussions  1. 1. 1.3. It  provides  t h e means t o i n t e r p r e t t h e  C+pronorninal, +anaphor]  mechanisms to  theory  empty  category  PRO.  f o r t h e i n t e r p r e t a t i o n o f PRO,  l i c e n s i n g , a r e not d i r e c t l y  included  in this  reference  The as  relevant  opposed f o r the  thesis.  Interactions  i s now  possible  t o assume t h a t  the interactions  between  the  various elements o f a sentence  principles enables and  and t h e model  us t o r e d u c e  complexity  eliminate  conceived in  t h e system  approach  by  these  t o grammar  o f r u l e s t o a minimum  since the principles  i n number  act as regulators  which  o f ovet—generation.  f o r example t h e t r a n s f o r m a t i o n a l component.  o f as t h e option  t h e sentence.  with  i n (£). T h i s  thepossibility  Consider  c a n be d e s c r i b e d  the feature  o f moving  some  element(s)  i s  anywhere  I n E n g l i s h , t h e e l e m e n t c a n b e a n NP CWh3 o r n o t . E x a m p l e s o f t h i s  It  marked  a r e given i n  (16) . (16)a. 0 seems C P e t e r t o have l e f t b. P e t e r s e e m s t o h a v e l e f t . c. M a r y made N P 3 . d. What d i d M a r y m a k e ? C  In  (16b) t h e l e x i c a l  (16a)) In  w  i s moved  position  NP  (subject Peter  CWh3  position  i s moved  of Peter  i n (16a,b)?  p o s i t i o n ? o r , why c a n ' t  thedirect  clause.  object  following  a n s w e r t o Pi b y t h e a p p e a l  o f grammar a d o p t e d  i n c l u d e s , a s we h a v e s e e n ,  (15). I n essence, NP t h a t  Pi) What  thefilter  forces the  moved  to a  i t move s o m e w h e r e e l s e  The concept  lexical  from  a n d B) Why i s P e t e r  sentence?  in  clause i n  of the higher  we c a n a s k t h e f o l l o w i n g q u e s t i o n s :  specific  theory  o f t h e lower  i n (16c) t o a p r e - s e n t e n t i a l p o s i t i o n .  Now, raising  h  t o t h e subject  ( 1 6 d ) a n NP m a r k e d  3.  here  to certain  provides the principles.  t h e Case F i l t e r  blocks  i snot Case-marked.  i n the  as  any occurrence  I f we a s s u m e t h a t  Case  formulated of a i nan  infinitival Case,  then  subject to  clause  the subject  i t would  o f t o have  be i m p o s s i b l e left.  move s o m e w h e r e e l s e ,  a n s w e r t o B, t h e o n l y position  of the higher  does not s u b c a t e g o r i z e C'. The  There  possibility  through  raising  defined  0-role  i s forced  That  object  one 9 - r o l e .  and t h e Case F i l t e r  sentence  like  available  ©-theory  t o Peter. i s realized  then  that  the subject  each  argument along  In other  of to  must  with i t s i snot  words,  the processing  principles  (17)a. 0 t o seem J o h n t o h a v e l e f t . b. •* J o h n t o s e e m t o h a v e l e f t . c. 0 asks John t o have l e f t . d. * J o h n a s k s t o h a v e l e f t .  pleonastic  the 9-Criterion  through  now s  can have  i f Pet er  i s satisfied.  theory.  of so-called  hand,  Therefore,  i n t h e grammar. The a c t i v e  fora  i n c l u d e s t h e ©—Criterion  specifies  (16b) i s g e n e r a t e d  seem  a ©-role t o t h e s u b j e c t  t o seem t h a t  i s moved t o t h e 0 p o s i t i o n  like  but o n l y  to the 0 position  On t h e o t h e r  Now,  In  i s the subject  i s because a verb  the exception  like  i n (13) above w h i c h  Cons i der  site  NP p o s i t i o n  with  i_t o r t h e r e .  ©-theory a n d C a s e  t o be C a s e - m a r k e d .  landing  VPs a s s i g n  violated  (16a)  why P e t e r  Pet e r  like  one and o n l y  i s the reason  f o r a n NP d i r e c t  i.e. verbs  i s ©-marked.  as t h e  clause.  clause  nominative  t o remain  possible  verbs,  left  does not get  f o r Peter  i . e .i n order  o f moving  of their  subjects have  That  no o b j e c t  t h e ©-theory.  position  get  i s thus  position  a r e , among  a  of others,  In  (17b) t h e Case F i l t e r  position marked  it  as t h e landing  p o s i t i o n s . The s e n t e n c e  without which  as well  i s violated  Case.  therefore  have two 0 — r o l e s s  by a s k . T h i s  and  the sentence  i s ruled out.  syntactic  i s blocked  now t o a c r u c i a l  It s t i p u l a t e s that level  subcategorization  the  same t h r o u g h o u t  syntactic  level,  this  object  make h a s a n o b j e c t  must  an object  NP m a r k e d  overt  object  Projection  The  is  position.  serves  o f empty  t h e empty c a t e g o r y  must  be  f o r example, i s at each  I n <16c), f o r  the feature  CWhJ.  make h a s no  i s not v i o l a t e d  see therefore  i n the determination  dealt  More s p e c i f i c a l l y ,  t h e pronominal  We  words,  since  that the  of the  categories.  empty c a t e g o r i e s  NP p o s i t i o n . involve  in this  a t each  items  i n (ISd),  but t h e P r o j e c t i o n P r i n c i p l e  Principle  distribution  with  and  the Projection  NP, t h e n ,  be p r e s e n t .  a p p l i c a t i o n o f Move B r e s u l t i n g  i s left  left  the lexicon. In other  After  trace  position  o f t h e ©-Criterion  t h e d e r i v a t i o n . I f a verb,  i n the lexicon as taking  a  by t o h a v e  properties of the lexical  specified  example,  i s moved t o t h i s  of the theory,  from  NP i s  position of  the representations  are projected  the  a lexical  the subject  i s a violation  part  a r e not Case-  since  t h e one a s s i g n e d  one a s s i g n e d  Principle.  the original  f o r John  i f John  the  I turn  site  I n (17d) a s k i s a v e r b  i s 0-marked,  will  since  with  in this  t h e problems  empty c a t e g o r i e s found  i n subject  PRO  thesis are a l l i n under study and p r o . The  position of  here former  infinitival  clauses.  Its reference  i s assigned  by  the  whereas  its distribution  i s determined  Binding  theory.  that  of  a  pronominal  positions. subject  Pro  reference; Contrary  has  to  cf.  tensed  Chomsky  PRO,  pro  chapter  Null  will  (1982)  b.  in  specification  with  i n languages to  In  the  where  determine i t s  f o r a d i s c u s s i o n and  positions.  the  ungovemed  associated  enough  be  pro,  fact,  we  i t s properties,  developed  in greater  1.2  which  will  below.  allows  see  that  it it  reference,  detail  in  and  1.2.3  and  Arguments here  i s the  f o r example,  following.  i t i s possible  following; ( 1 8 ) a.  by  feature  appear  sentences  is rich  the  Control  2.  problem  Italian,  i t to  indirectly  i s C+pronomina1,-anaphor3  empty c a t e g o r y  The  has  of  governed.  distribution  1.2.  of  forces  PRO  empty c a t e g o r y  i n governed  be  The  in  i s the  agreement  to  appear  fact  anaphor  position  subject-verb  to  The  theory  Cornprero will+buy~ls ' I w i l l buy  una m a c c h i n a . a car a car. '  Verra. wi1l+come-3s ' He w i l l come. '  to  In  some  find  languages,  sentences  like  the  c.  Lo conosce. h im+know-3s He k n o w s h i m . ' 1  d.  In  Ho telefonato. have-Is telephoned ' I telephoned.'  these sentences,  realisation the  from The  to  o f the sentence.  non pro-drop  seen  the subject  languages,  t h e morpheme  first  say that  reaction  i s not present  Obviously, this i n English  by morpheme  glosses  fact  One c o u l d  s i m p l y assume t h a t  generated  i n D-structure i n Italian  precisely, will  t h e base  optionally  (19)  a p r o p e r t y c a n be  constitute  rules  a  problem.  position  i n t h e c a s e s o f ( 1 8 ) .More produced  b y t h e X'  NP a s s i s t e r  of  system  INFL'.  INFL'' I INFL' / \ VP  t h e r e a r e major all,  that  (18).  t h e r e i s no s u b j e c t  generate a subject  INFL  of  in  a s c a n be  C ' > Spec C C C •> COMP INFL' ' INFL' ' > (NP) INFL' INFL' —•> I N F L VP  (20) =(18)  But  structure  i s impossible i n  f o r example,  when e n c o u n t e r i n g s u c h  i t does not i n a c t u a l  i n the surface  t h e Extended  the subject  within  problems  Projection  position  t h e framework  with  this  type of analysis.  Principle,  i s universally  adopted  here.  which  stipulates  available,  Secondly,  First  i n one  i s violated possible  analysis  (see Chapter  a movement base  for  o f t h e I N F L n o d e t o VP  p o s i t i o n , INFL  subject:  INFL  i s obtained  This  as i n t h e French  by  4). Ini t s  i s i n a c e r t a i n r e l a t i o n s h i p with  t h e NP  r e l a t i o n s h i p accounts  bears a c e r t a i n i n f l e c t i o n  i t ssubject  (£l)a.  inflection  (but c f . chapter  governs the subject.  why t h e v e r b  with  4), the verbal  examples  that  i s  compatible  i n (£1).  Nous v o y o n s beaucoup de f i l m s . we s e e - l p many o f movies 'We s e e many m o v i e s . '  b. # N o u s v o i s ... see-ls c.  Je vois I  ...  d. * J e v o y o n s In  Italian  .. .  t h e same p r o c e s s  c a n b e o b s e r v e d , fl t e n s e d  always e x h i b i t s subject-verb (overt)  NP s u b j e c t ,  representation (£0).  It allows  agreement such  like  agreement  as i n ( I S ) . This (££) s h o u l d  i s another  be p r e f e r r e d  us t o g e n e r a l i z e  i n sentences with  e v e n when t h e r e  overt  the process subject  reason  over of  one  verb i s no why like  subject-verb  t o t h e ones  with  subject.  (££) / NP i 0 Pi t h i r d  INFL'' \ INFL' / \ INFL VP  piece  of evidence  supporting  a  t h e presence  of a  no  subject in  p o s i t i o n can  object  p o s i t i o n as  expression  and,  conditions  which  (£3)  [j.•  If  one  is  free  be  a  and  this  has  two  and  inversion;  INFL  structure  will  the  see  has  second  in  a  Giovanni  (£0)  then  N P  will  to  lexical  an  C of  Rthe  must  be  object  NP  sentence  reading  in  will  structure  in  (££)  free.  in  (£3)  would  John*. (£3) In  i s not the  vedere  meaning  a v a i l a b l e . The  first,  Giovanni  sentence  is  the  a f t e r a p p l i c a t i o n of  in  of  this  type  of  free inversion.  <£5).  see.  in refemce  see  Binding  (£4)  with  the  i s the  subject.  object In  that  of  Giovannij. allows  us  to  properly  r u l e out  vedere  case  means:  ±  NP  3 3113  the  the  ft  coreferent.  for a discussion the  be  R~expression  interpretation, Giovanni  is disjoint  He  C  an  corresponding  readings.  £.4  to  interpretation for this  intransitive see  theory,  under c o n d i t i o n  vedra  V  Binding  i s considered  falls  C p  possible  Giovanni  sent ence  The  such,  John are  sentence  (£5)  (£3)  i n t e r p r e t a t i o n of  of  from  s t i p u l a t e s that  possible  subject  (£6)  a  -a- He*,  But  and  d.  drawn  in  r e f l e x i v e reading  (£4)  In  as  assumes the  w h e r e he  The  be  the  the  reflexive sentence  (£7)  In  i n t e r p r e t a t i o n i n (£4). has t h e f o l l o w i n g  *  Ci». [  N  violates hand, that  reading,  same t y p e  reading as  0 i 3  H i .  In  E  Ex.  (£9), Giovanni  coreferent ial different  this  PiGRi  i s bound  reading  referential  since,  under  i n (£8).  CtMpa  Giovannij  be r e p e a t e d  3 3 33  f o r conditions  does not accept  a referential  element  i s g i ven  XIMF-I-  On t h e o t h e r  f o rthe ungrarnmaticality  that  be t o a s s i g n  s t r u c t ure  Ex..  could  therefore  theory.  i n (£4) i n a theory  resu11 ing  NP a n d  conditions.  vedr a  E VP  move t o a c c o u n t  INFL and assume t h a t  *  INFL  of argumentation  i n <£7) would  (£9)  by t h e s u b j e c t  (£3) has t h e s t r u c t u r e  obvious  EMP. G i o v a n n i i 3 3 3 3  i n (£6) f o l l o w s a u t o m a t i c a l l y  Pi a n d B o f t h e B i n d i n g •ne  [^P vedra  cond it i on C o f t h e B i n d i n g  IMP-  The  i s bound  the reading  i s so because t h e  representation  P 0 i 3 C i i INFL  (£7), Giovanni  This  occupies  empty  index an  of the  subjects  t o PiGR i n  ft—position.  The  i n (£9).  TNS 3  vedra  C V P  in violation  EMF.  Giovanni  I  333 3  o f c o n d i t i o n C and t h e  i s excluded.  I f PiGR a n d G i o v a n n i  i n d i c e s then  the right  bear  i n t e r p r e t a t i o ni s  obtained. One p r o b l e m like  Chinese  with  such  an account  i s that  and Japanese e x h i b i t n u l l  certain  subjects  languages  without  any  overt for  subject-verb  more  on t h i s  assumed, index are  a g r e e m e n t ; c f . Huang  type  t h e s o l u t i o n i n (£9)  with  ftGR  bearing  a  l.£.£  RGRs a r e  referential  conditions violations  unaccounted f o r .  while  would  then  want  r e j e c t i n g empty  phenomenon.  I thus  t h e one  i n (££)  with  empty s u b j e c t s  l.£. 1. P r e v i o u s Now  that  a s k what  sentences explores  t o a s k why NPs  assume  like  t o account  the theory  we  have a b a s i c  makes  i t possible  i n some  here  falls  cross-linguistic should  parameter(s)  syntactic  for a  allows  ftGRs  given  for structures  f o r tensed  clauses  t o have n u l l  under t h e type  therefore  The  in  tensed  section  i n 1.1,  of allowing  and  empty  learnable.  t o ask whether any  from  want  of analysis involving  Once  other  the account.  of the set of properties  languages that  This  we  h a v e b e e n made a t  possibility  want  can f o l l o w  that  subjects  i n others?  be p a r a m e t e r i z e d  a r e s e t , we  languages,  o f t h e problem,  Recalling to the discussion  variation.  properties  The s t u d y  empty  Italian.  statement  l a n g u a g e s and not  the  subjects  that  like  the d i f f e r e n t attempts that  issue  allows  Analyses  question.  these  i n order  i n languages  this  subject  a theory  as the representation  answering  the  argument. U n l e s s empty  i s not a v a i l a b l e and t h e B i n d i n g  We  to  o f empty  (198£;19S4) a n d  associated  with  null  i s ,the syntactic c h a r a c t e r i s t i c s of  c a n be s a i d  to follow  from  the fact  that  35 the  subject  His  generalization,  subject  b.  In  to  to  Italian  that  came?  this  r u l e of  & Lasnik  whereas the  a  containing  aimed  rich  has  leave?'  (30b)  well  i t was  and  as  subject  on  agreement  made u n a v a i l a b l e  making by  f o r more on the  relating  first  empty s u b j e c t s ,  delete a  stated  the  lexical  trace  an left  pronoun?  languages were t h u s  latter  the  the  inflection.  a very one  research the  at  the  Taraldsen  comparison,  Chomsky  (1981)  inflectional  element  cf. named  agreement.  that  the  English  exhibits this  to  subject (1978)  quite  would  property  to  INFL  would  recover  system  developed.  proposes  subject  pointed  agreement  element  PlGR w h i c h  i t possible fact  missing  poor s u b j e c t - v e r b  i s , by  observation,  property  not.  as  Null  verbal  Italian  this  concentrate  abstract  subject  "that-trace"  (30).  deletion could  (1977).  step  English  Following  both  null  1anguages.  next  that  subject the  in  (1971).  says that  usual  characteristic,  languages allowing  out  does  say  explain  to  that  e x h i b i t the  c r e d i che p a r t i r a ? do y o u t h i n k t h a t w i l l  property  in  not  in Perlmutter  stated,  Chi 'Who  "pro-drop"  £.£.£  approximately  you  e x t r a c t i o n of  The  initiated  did  in  Chomsky  was  Who  optional by  empty,  Compare E n g l i s h  order  that,  be  l a n g u a g e s do  effects. (30)a.*  can  have the  some information  i s missing;  differ  from  whereas the  see  Italian former  Chomsky d i s c u s s e s subject  parameter.  relevant  (31)  a set of properties associated  examples  Chomsky  They a r e g i v e n from  i n (31) a l o n g  with  the null  with the  Italian.  (1981:£40)  a.  Missing subject: Ho t r o v a t o i l l i b r o . MI) f o u n d t h e book.'  b.  Free inversion i n simple Ha mang i a t o G i o v a n n i . 'Giovanni a t e . '  c.  L o n g Wh-movement o f s u b j e c t s L' uorno C c h e m i d o m a n d o C c h i a b b i a v i s t o 11. 'The man ( x ) s u c h t h a t I w o n d e r who x s a w . '  J sentences:  d.  Empty r e s u m p t i v e p r o n o u n s i n embedded c l a u s e s E c c o l a r a g a z z a C c h e m i d o m a n d o C c h i c r e d i [ c h e p o s s a VP. ' T h i s i s t h e g i r l who I w o n d e r who t h i n k s t h a t s h e may V P . '  e.  Apparent v i o l a t i o n s o f t h e t h a t - t f i l t e r s C h i c r e d i C c h e p a r t i r a 3? 'Who d o y o u t h i n k ( t h a t ) w i l l leave?'  Ch oms k y argued from  pro poses a s o 1 u t i on based that  the cluster  languages subject involved subject  postverbal  t h e ECP a t L F . T h i s  reducing  being  extraction of the subject  a governed  violate  on Ri z z i' s  solution consists,  of p r o p e r t i e s proper i n (a) and  inversion. Since  i n these  in Italian  constructions,  anaphor,  Ri z zi  place  does not  i n fact,  to null  in  subject  ( b ) , namely  missing  no e x t r a c t i o n o p e r a t i o n i s Chomsky assumed  p o s i t i o n w a s PRO c o s u p e r s c r i p t e d  a pronominal  takes  p o s i t i o n and t h e r e f o r e  i n (31) t o t h e o n e s  and f r e e  (198£) w o r k.  i s excluded  with  that  AGR.  t h e empty  B u t PRO,  by t h e B i n d i n g  conditions the  from having  theorem  Binding  that  PRO  conditions  a  governing  must  be  apply,  i n a l l languages with  INFL  i n s i d e VP  for  the  subject  propose the  <3£) R In  a  null  In PRO  PRO  parameter  the  h a v e an  and  level  INFL  clauses,  where  the  Assuming  agreement,  that  therefore  a rule  is a  moves  governor  i t is possible  to  (3£) .  inverted  English,  subject  for  in subject at  S-structure  i s blocked  inverted  NP  inversion,  by the  his  Chomsky  from  INFL.  Borer  —  pronoun,  to VP  i n the  (3£)  nominative instances  i s fixed  clause  R applies  Case cannot  the  i n the  and  will  it Case  of  negatively, always  i n PF  —  assigned  instances  French  typology  the  be  in subject  there  i.e. a  assign  making  of  the  free  position is  pleonastic  i1. empty  empty c a t e g o r y  pronominal  is  free to  of  be  and  p o s i t i o n . Furthermore,  Note that  of  by  S-structure  i f  (1984).  since  since  English  discussion  and  tensed  this  (198£) i n t r o d u c e s  overt  a  i s a v a i l a b l e and,  INFL at  i n which  empty c a t e g o r y  analogous to In  also  p o s i t i o n of  inversion  by  dominated  instance,  excluded  option  empty s u b j e c t  inversion; cf.  an  in  ungoverned  therefore  to  R),  language t h i s  appears to  governed  PRO  it rule  the  S-structure.  subject-verb  p o s i t i o n i n tensed  subject  possible  free  i s , at  at  deriving  Chomsky (1981) may a p p l y i n s y n t a x .  taken,  to  (call  ungoverned  that  that  category,  pro  categories, corresponding  non—anaphor,  and  argues  that in  i t i s best  null  s u i t e d as  subject  assumption normally accounted  languages.  i s that  be  a  i f PRO  missing  One  missing  arbitrary  for  the  of  the  subject  in reference. i s used  Furthermore,  obligatory  verb-fronting operation  a  position (properly)  Torrego to  use  (1984) PRO  The  in subject of  pro  unnecessary  to  invoke  pro,  as  opposed  must  be  added  identified enough  governed  for the  use  to  or  the a  can  be  can  act  as  an  pro  appear  in subject  One  of  the  of  and  i n v e r s i o n are  dialects subject  the  goals  to  Jaeggli  (198£). of  case  is  not PRO  may  which  places  fronted This  the  the  at  of  English.  pro In  in  impossible above.  making i t in  that  PiGR e l e m e n t  identifier  i n , say,  i t  (3£)  S~structure.  requirement The  one  an  see  reasons stated  as  is  subject  verb;  makes  be  since What  pro  be  being in  rich  CNP,  IP]  conclusion,  p o s i t i o n i n languages with  rich  agreement.  approach free  property  clauses  interrogatives there  governed  determined.  the  subject—verb  cannot  position since  the  i s the  i s not  only  languages  parameter such  whereas t h i s can  this  f u r t h e r consequence of  theory  locally  in Italian  underlying  p o s i t i o n f o r the  PRO,  the  by  reasons  This  argumentation.  has  to  i n Spanish  i n tensed  i n these  in this  arbitrary.  subject  missing  His  (1985)  subject closely  but  Trentino which  i s to  problem  and  have  based  challenge  i n which  r e l a t e d as  arguments are  Italian,  languages  Safir  i n Chomsky  on  Modenese, free  this  the  any property (1981)  existence  that  are  inversion,  and  not  of  and two  null  Portuguese, any  possibility  possible by  which  Bouchard  parameterized  INFL  that  a  to  at  assigned  "invisible" On  (198£),  the  assigning  feature  before  at  therefore  in  some  parameter  property  delaying until  only  percolation  In t h i s  i f i t has  gender,  and  i s the  case  in null  subject  of  analysis,  and  LF).  nominal and  certain  Case),  i f a  it will  to  subject  the  VP  nominal  be languages.  languages r e q u i r e  percolate  The  is  S-structure,  i n VP,  null  the  LF.  INFL  section  here,  concentrate  A)  cause  subject  by  INFL t o  this  subject-verb  neglected;  will  Case  the  Case  projection  in subject  position is  i t follows that  it will  be  PF.  presented  Japanese, rich  of  null  nominative  L e x i c a 1 i z a t i o n which s t i p u l a t e s  i f and  LF  S-structure.  conclude  analyses  at  non  of  INFL  of  always Case-marked  lexicalized To  of  number,  That  hand,  of  This  subject  (D-structure,  (person,  PF.  null  option  feature  is lexical  at  other  or  the  the  Case o n l y  NP.  the  without  inversion.  his Principle  PF  of  property  i n v e r s i o n i s made  assignment  postverbal  i n syntax  nominal  features  a  treatment  through  i s i n VP  According  to  subject  inversion. Free  free  Case a s s i g n i n g  missing  a n a l y s i s by  f o r any  r e l a t e s i t to For  is  free  clitic  difficulties  the  of  in Safir's  a subject  that  e x h i b i t s the  on  except Romance  agreement.  those  with  i t should  no  Two  be  noted  that  the  f o r Bouchard's d i s c u s s i o n l a n g u a g e s w h e r e we types of  agreement  at  languages a l l ; and  B)  often  of find  are those  with  object-verb  agreement.  (198£;1984) t r e a t m e n t latter  type  l.£.£.  Zero T o p i c  Let topic  language  (34) .  The  Ni you(s)  c.  Ta he/she  discuss the  Languages  (Mandarin) using  Chinese  Huang's  a s an example o f a  (198£;1984) t e r m i n o l o g y  a t a l l a s c a n be s e e n  zero and data.  i n (33) and  s h i xuesheng. s h i xuesheng.  Women 1a i we come  b.  Nimen l a i . you(p)  c.  Tarnen l a i . they  form  then  of t h e verb that  external  there  finite  infinitives  should  I briefly  Wo s h i x u e s h e n g . I be student ' I am a s t u d e n t . '  implying true  type.  Huang's  t h e tones.  b.  ( 3 4 ) a.  section reviews  4.  h a s no agreement  I omit  ( 3 3 ) a.  o f t h e former  i n chapter  us t a k e  Chinese  The f o l l o w i n g  i s invariable i s n o AGR  clauses  argument  i n INFL  i n Chinese  i n English with position  be p o s s i b l e t o f i n d  and t h i s  respect  c a n be a n a l y z e d  i n Chinese.  should  I f this  behave  t o the content  (thes u b j e c t ) .  a  an empty s u b j e c t  In other  as  i s  like of the words, i t  in finite  clauses  41 in  Chinese.  Huang  This  i s verified  i n t h e f o l l o w i n g examples  from  (198£s35£). Z h a n g s a n s h u o C E0II x i a w u h u i l a i 1. say a f t e r n o o n w i l l come ' Z h a n g s a n s a i d t h a t he w i l l come t h i s a f t e r n o o n . '  ( 3 5 ) a.  b.  Zhangsan shuo  CC  C03 y o u - r n e i - y o u qianH meiyou guanxiH. h a v e - n o t - h a v e money no matter i t d o e s n ' t m a t t e r w h e t h e r he h a s money  'Zhangsan s a i d or not. '  Huang  reports that  t h e empty s u b j e c t  (35)  i s i n t e r p r e t e d as r e f e r r i n g  that  i t i s p r o , t h e empty  must  be c o i n d e x e d  with  (1982s365-377) . T h i s assigns sense,  a reference  t o Zhangsan.  pronominal  i s reminiscent t o PRO.  subject  i n a very appears  of Control  manner.  in infinitival  clauses  which  certain  the null  In effect, in this  s e e Huang  theory  in a  as exhibiting  assumes  and t h a t i t  element;  Note here t h a t ,  restricted  clause i n  Huang  non-anaphor,  the c l o s e s t nominal  E n g l i s h c a n be c o n s i d e r e d  property  o f t h e lower  subject  an empty  language as i n  (36) . (36) We  John wants  c a n now  understand  arguments appear is  very  C C03 t o g o ] .  rich  why H u a n g  p o i n t s out that  i n two s i t u a t i o n s  agreement;  i n languages:  a n d B) w h e n t h e r e  null  subject  Pi) w h e n  there  i s no agreement a t  all. In  t h e second  languages with  chapter  subject  of this  clitics  t h e s i s we w i l l  must  be a n a l y s e d  see that as n u l l  subject  languages.  Whether they  agreement  type  will  The d i f f e r e n c e sentence subject by  between  respect  <1982:355)  C03  l a il e . c o m e PiSP ( e c t ) came.'  Rather, empty  i t i s assumed  category  of the rich  to identification  Chinese  o f t h e empty  category  i s not i d e n t i f i e d  operator  which  Contro1.  Huang  'He  as being  (35) a n d (36) a n d t h e o t h e r  i n (37), t h e e m p t y  t h e p r i nc i p1es o f  (37)  be c o n s i d e r e d  be c o n s i d e r e d .  i n (37) w i t h i s that,  must  that  i s present  an empty  i n a position  adjoined  ft'-binds  the  t o IP as i n  (36) . (38)  This  Cop*]  assumption  category as  ...  (39)  position  by t h e f a c t  that  can appear i n object (37). T h u s ,  t h e empty position  the following  i n Chinese:  Huang a.  . . . II  i s supported  discussed here  i n subject  possible  [ j p C0t  (198£:355)  Q: L i s i j . ,  s h e i k a n j i a n - l e t_± ? who see PiSP 'Pis f o r L i s i , who s a w ( h i m ) ? '  b. P i : Z h a n g s a n s h u o C n i k a n j i a n - l e C03 3 . say you see PiSP ' Z h a n g s a n s a i d t h a t y o u saw ( h i m ) . '  as well  discourse i s  What  (39b) shows  topic  (Lisi)  i s that  introduced  t h e empty c a t e g o r y i n a previous  can r e f e r  discourse  though  this  topic  i s not p h o n o l o g i c a l l y present  (39b).  This  topic  i s called  represented the  zero  the  reason  type  structurally  topic  found  discussed  clauses  i n non z e r o  the sentence  counterparts pronominal is  for  (38). This  i s relevant here;  (39b) i n t h e s e t o note  Huang  a different  topic  i s no  "closest"  t h e r e f o r e proposed  i n tensed  Italian.  Consider  to the  Italian  i n (37) c a n n o t  nominal  that  element  be (there  i t i s a variable  A'-  (1986)  approach.  t o Huang's a n a l y s i s ,  the null  i s not incompatible  The two a r e d i f f e r e n t Italian,  and  null  subject  null  subject.  Chinese  would  be z e r o  German,  in this  view,  could  null  of  o f empty  a s i n ( 3 8 ) ; but c f . L i e j i o n g  i n languages.  non  like  Contrary  ways  (35).  the type  from t h e one  languages  i n (37) above.  in Italian  possibility.  i t i s part  of  languages.  that  i s different  by a d i s c o u r s e t o p i c  i s found  sentence  i s the characteristic  i n ( 1 8 ) , t h e empty c a t e g o r y  According it  here  since there  no AGR).  bound  as  i t i s important  subject  again  i n the  i t i s p o s s i b l e t o have empty arguments o f t h e  i n (37) and  Finally  (39a) even  " d i s c o u r s e - t o p i c " and i s  languages that  why  back t o a  with  interact  f o r instance,  language whereas E n g l i s h  subject  the zero  i s a non z e r o topic  (37) and n u l l  be a n a l y z e d  s u b j e c t ; s e e t h e e n d o f s e c t i o n £.2.1  as  topic  in particular  i s non z e r o  topic  property  as zero  topic non subject  topic  f o r more  on  but  44 German. The order  conclusion to this  to achieve  the  null  and  Japanese,  the  present  type  argument  larger  because  property  research  of null  discussed  the  and if  that  appearance  fashion. is  Pis we  reasons,  pronominal,  The  assignment why  on  problems  i n languages  So,  the  and  languages  in  arguments of  i n the Introduction.  f o r the development Chomsky  t h e empty c a t e g o r y i s , pro.  (1982)  i s that  of pro are not  defined  in a  questions  arises  f o r i t s l i c e n s i n g ? ? and of content  t o p r o two  pro d i f f e r respect?  Projection  Principle  a l l o w s PRO  position  must  B)  be  pure  the conditions  kind  Plre t h e  on  element  licensing  of  of and  empty  Extended  t o appear  ungoverned  of  operations?  types  instance, the  an  to a  p r e c i s e enough  distinct  ( a n NP)  of  introduces, for  Pi) What  from the o t h e r For  of the rest  corresponding  problem  in this  This  though  to  Italian  topic  The  does  Chinese  even  related  like  in  a n a l y s i s of  like  languages a l l o w i n g n u l l  have seen,  categories  position.  languages.  that,  pro  following  sufficient  not,  topic  include the zero  section i s crucial  thesis.  various  we  of the  1 . 2 . 3 . fl T h e o r y o f  this  include languages  are a l l a p p l i c a t i o n s of the p r i n c i p l e  recoverabiIty  This  must  found  clitics,  typology  o f e x p l a n a t i o n , any  concentrates  arguments  they  level  some o f t h e z e r o  languages with a  a higher  s e c t i o n i s the r e a l i z a t i o n  one  i n argument according  to  the  Binding  theory  of  Control.  structure by  and  the  by  second first  ( 1 9 8 6 ) who  rich  i n chapter  first  inflection.  when  i t has  a  leads  Gianni  According internal  to  e  under  in  as  problem  by  i t s presence are  the  in  the  made p o s s i b l e  i n 1.2.1, the  being  The  i t s identification from  to  i n the  position  the  to  cases can  properly  by  X . v  of  rich)  pro  through INFL  can  assign  pointed  in  out  contexts  i s not  met.  Consider  sequente cone1 usione.  following conclusion.'  accontentare please  be  f l a gente").  i  0  \  (people).  where the  only  object  filled  identifying  following licensing  Rizzi  Rizzi. alia  ad  return  on  OGR  problem  appearance of  (40)  I  i s done  element  (that of  government.  pronto  Rizzi (1986) pro i s Case-marked  section?  B.  shown  i s always ready  a r i s e s of  in this  speaking,  (people)  Rizzi,  both  d i s c u s s i o n i s based  ( l a gente/  sernpre  argument  proposes the  (41)  The  property  Questo conduce  'Gianni  the  Formally  examples  'This b.  £.  pro,  c o n d i t i o n on  Italian  ( 4 0 ) a.  i s treated  (1986) c o n c e r n s t h e  where t h i s the  of  pro  pro,  feature-content  raised question  certain  features to Rizzi  case of  i t s feature content  inflection.  question  rich  in  i s assigned  In the  Identification  its  PRO  i t s particular  government The  to  theory.  principles  by  this  i s empty, pro.  7  But  element.  the then  Rizzi  where  X i s a governing  languages X  v  i s accounted  i n a given  consequently Italian, pro  X  never c a n be  y  be  of the type  y.-Variation  f o r by t h e d i f f e r e n t  language.  i n object The  head  In E n g l i s h ,  licensed  INFL  X  y  among  v a l u e s assumed  h a s no  v a l u e and  i n any s t r u c t u r a l  by  pro can  position.  f o r pro i n subject p o s i t i o n  and  In  V for  position.  content  of pro i s determined  through  the  following  convent ion s (4£)  To  Rizzi (1986) L e t X be t h e l i c e n s i n g h e a d o f a n o c c u r r e n c e o f p r o ; then pro has t h e grammatical f e a t u r e s p e c i f i c a t i o n o f t h e f e a t u r e s on X c o i n d e x e d w i t h i t .  s e e how  sentences ( 4 3 ) a. b. ( 4 4 ) a. b. ( 4 5 ) a. b.  In and is  this  convention operates,  l e t us e x a m i n e  the  Italian  i n (43)~-(45).  Verra. will+come—3s ' He w i l l c o m e . ' [ p pro± I N F L t v e r r a Z  1  S i dorrne troppo. c l s l e e p - 3 s too+much ' P e o p l e s l e e p t o o much. ' [ p pro. b C i INIK--L_ s i „ , t , 1 d o r r n e t r o p p o 3 P  X  v  La buona rnusica r i c o n c i l i a 0 con se s t e s s i . ' Good rnusic r e c o r t c i l e s 0 w i t h o n e s e I f . ' l a buona rnusica [ C r i c o n c i l i a 1 p r o * ^ con V P  (43) p r o i s l i c e n s e d i t scontent licensed  by  v  by t h e n o m i n a t i v e  i s determined INFL  but  b  assigning  1  ...  head  INHL  by t h e same e l e m e n t .  In  (44)  i t s features are specified  by  the  pro  47 arbitrary by  the  subject  clitic  s i _ under  Case a s s i g n i n g head  coindexing  with  a slot  coindexing  with  PiGR u n d e r  corresponding  to  in  and,  the  index for  syntax must  the  In c h a p t e r  necessary  and  Empty  that  a  certain seen as  to the gap to  elements  licensed  (analogous  through  to  9-role  i s assigned  the  index  to the  Projection Principle,  the  objects  NP  theory  will to  see  of  pro  see  Rizzi  developed  whether t h i s  answer q u e s t i o n  i n a sentence be  We  arguments  like  grammars,  in certain  arb this  (1986)  0  in  theory  Rizzi  is  above.  allowed  to  Besides  pro and  and  appear and  PRO  PRO,  Binding  recoverable  i n order  i s therefore plausible independent  contexts  to  be  these  structural  reasons,  missing.  gaps are  We  in Italian  to allow  represented  circumstances and  for  have  representation of  u n d e r what  in structural  I n t r o d u c t i o n , i t seems  be  for  sections that  have a l s o seen pro  must  successful. It  empty c a t e g o r i e s i n t h e  Government  V  is  i s defined  internal  d i s c u s s i o n i n the  previous  sentence.  are  The  position  the  £ we  particular  i n the  i t s content  pro  Categories  communication assume t h a t  (45)  argumentation.  sufficient  Returning clear  the  is essentially  (1986).  On  object  In  ©-grid o f  INFL).  according  of  and  i n the  transferred to  detail  This  1.3.  be  the  V  INFL.  Chinese,  the  null  for  example,  positions.  i t i s g e n e r a l l y assumed framework t h a t  NPs  can  be  in  the  represented  by  two o t h e r this the  empty c a t e g o r i e s , namely v a r i a b l e s and NP~-traces.  section, field  order  I will  concerning  t o make c l e a r  The  first  questions defined  How  enter  an area  the determination my  problem  o f empty  c a n be e x p r e s s e d  categories i n  them.  as t h e following  types  o f empty  simple  categories  i n t h e grammar?  categories produced  adopted  a r e base  generated  by movement  o r p o s s i b l y PF  here  i s as follows. ( p r o a n d PRO)  structural  (NP—traces  come f r o m  context Binding  a n d some  agreement.  from  empty  a n d some a r e  variables).  i t sf e a t u r e s from t h e  and t h e v a r i o u s  features that  are available  theory.  One o f t h e c o n s e q u e n c e s o f t h i s movement  Some  o f NPs i n t h e t r a n s f o r m a t i o n a l component,  Pi g a p i n a s t r u c t u r e r e c e i v e s  of  of controversy i n  assumptions concerning  are the different  The a p p r o a c h  LF  briefly  In  a n NP p o s i t i o n  The D — s t r u c t u r e  i s found  that  o f such  i n the application  i s governed  a sentence  by  rich  has t h e form i n  (46) . (46) The (47) The  C C P  C I P  C,Mp.Wh3  S-structure CC P  I N F L VP  111  i s a s t h e o n e i n (47). IMP- e i  problem  occupies  C,.  3 Cx. I N F L VP  consists i n determining  t h e MP  position  111  what  i n (47) f r o m  type  where  o f empty  movement  category  49 originates. all  the  pro?  necessary  I t can  agreement. the  Recall  Wh  On  i f we the  i f we  of  by  stipulate  other  that  and  context  allows  be  two  i n the  produced  by  a  base generated  as  a  variable only  allows  be  In  pro.  way  this  movement  and  the  that  category  or  other  the  and  neither  In  such  that  in  (47)  e  i t cannot  be  view, as  such;  Jaeggli  a  of  is a which  can  act  cf.  (1984)  the  case,  pro,  pro  is  since  other.  In t h i s  and  only  empty  gap  in  not  important  i f i t i s base-generated this,  is  i s an  the  study  empty c a t e g o r y .  the  a  different  Pro  s  by  is  Thus,  conditions  This  i n one  i s Pi'-bound  f o r two  (47)  (47)  rich  Pi'-bound.  Binding  in  empty c a t e g o r y  variable.  better" than of  by  with  for  3.3.3 a  view.  Note that i s to  resumptive  and  the  resolved  rest  an  allows  A--free.  d i r e c t a p p l i c a t i o n of  different  which  since be  such  PiGR  is e  empty c a t e g o r y  context pro  So,  i s governed  i n t e r p r e t a t i o n s of  is  a  the  pro.  v a r i a b l e s are  therefore  is "intrinsically  variable  for  the  i t must  must  assume  pro  i n S p e c COMP a n d  case,  could  that  I will  hand,  INFL c o n t a i n s  licensing  assume t h a t  definition  problem  them  for  empty c a t e g o r i e s :  excluded  A-free  assume t h a t  assume t h a t  particular  types  we  features  antecedent  variable this  be  that  there  allow  us  i s at  t o ma i n t a i n t h e  p r o n o u n s and  Wh—movement.  Assume f i n a l l y  that  l e a s t one  consequence to distinction  v a r i a b l e s . Suppose that  Suppose  further that  i t has  we  the  view,  took  opposite  this  move,  b e t w e e n em p t y a  language  rich namely  agreement. that  an  empty  category  movement empty pro  took  governed place  category  by r i c h  from  pronoun. subject  position in this  with  Consequently,  an o p e r a t o r ,  there  will  between  pro coindexed  operator I  thus  syntactic  will assume  discussions  Two  with  respect  views  parameter, formulation rich  rules,  t o p r o and,  rich  with  consequence. traces  (1984)  i n the  f o r extensive categories.  the reference  The q u e s t i o n  to  empty  i s the following:  filters,  o f UG  make  categories? According  to the  first,  i n the generative  categories.  can therefore  The  or a given  framework,  o b j e c t i o n t o any null  subject  make r e f e r e n c e  informally, state that  inflection  coindexed  leave  o r rnetatheoret i c a l  t o empty  Subjacency  i s r e f e r r e d t o Chomsky  parameters,  studies  f o r example,  inflection  exhibiting  t o UG.  in this  to  and t r a c e  and B r o d y  c a n be o p p o s e d .  reference  respect  o f d e f i n i n g t h e empty  t o empty  i n current  possible  operations  i s s u e h a s t o do w i t h  i s no t h e o r e t i c a l  explicit  by  (1982),  i s always  resumptive  i s not a d e s i r a b l e  reader  o f the problem  reference  widespread  This  an  be a v a r i a b l e i n  with  an o p e r a t o r  movement  the principles,  explicit  there  that  Bouchard  second  categories Should  lost.  s t r u c t u r e . The  (1981;1982),  The  be  with  Then  i t i s a  never  pro even i f  language  p o s i t i o n i n any o f t h e d e r i v a t i o n s  l a n g u a g e and t h e u s e f u l d i s t i n c t i o n  an  i s always  the p o s i t i o n i t occupies.  i n subject  a n d , when c o i n d e x e d  agreement  category.  c a n be a n u l l  Hence  in i t s  pro i s l i c e n s e d a  subject  language language,  otherwise  i t cannot;  especially there the  i n Bouchard  i s only  other  independent  NPs —  follows  for  between  the  rule,  by t h e i n t e r a c t i o n o f  —  which  t h e same i n t e r a c t i o n s .  parameter  or f i l t e r  t o empty c a t e g o r i e s  since  It  should they are  t h e c o m p o n e n t s o f UG d o n o t  empty and o v e r t o f such  type, and  Therefore, the  NPs; c f .  an approach  adopt  here t h e l a t t e r  more r e s t r i c t i v e visible  model.  view  Bouchard  (1982)  and an e x a m i n a t i o n o f  because  throughout the rest  of this  t o b e made w h e n c o n f r o n t e d  plausible  analysis.  my a t t e m p t , existing  formalization argument only  of this  clitics  and r e l a t e d  must  matter  of this  o r n o t . The p r o b l e m  move  and w i l l  type  will  guide  i s illustrated  by  the relationship positions.  t o whether  be e m p t y o r p h o n o l o g i c a l l y  i s whether  t o provide  more t h a n one  argument  s h o u l d make no r e f e r e n c e  position  relevant  bound  One e x a m p l e  thesis  with  i n c h a p t e r 3, t o f o r m a l i z e  between  i t appears  The c o n s e q u e n c e s  decisions  be  partitions  o f empty c a t e g o r i e s  from  that  consequences. I  be  exclusively  t h e development  its  derive  NPs. I n o t h e r words,  discriminate  exhaustively  NPs. The d i s t r i b u t i o n ,  c o m p o n e n t s o f UG.  no p r i n c i p l e ,  defended  on t h e a s s u m p t i o n  NPs a r e d e t e r m i n e d  should also  that  i s based  t y p e , and r e f e r e n c e  make r e f e r e n c e regular  of lexical  of lexical  distribution, are  <198£),  one empty c a t e g o r y t h a t  distribution  reference  c f . i.£. T h e o t h e r v i e w ,  This  o r not t h e realized.  o r n o t t h e NP o c c u p y i n g  The  i t can  o f d e t e r m i n i n g t h e empty o r o v e r t  status Th e  o f t h e NP  p a r t i c u 1 a r s t a t u s rn u s t  occupied To  by  t h e NP  to the status  empty o r o v e r t this  thesis.  1.4.  Summary First,  faculty  we  the  here,  parameters,  this  The  i n the  approach on  individual  syntactic  recoverabi1ity  trigger  I will with  to the  language  with  and  of  i s assumed  lexical  and  items w i t h i n  proposed  to the  and  for variation  t o be  among  growth  and  crucial  problems  in  language  principles  variation  arguments  being  related  language and to  are approached  a way  that  on  the here.  plausible  i n c o r e grammar  be  learner.  the principles  parameters,  explicit  f o r language  between  the various  o f UG  the  pos i t i on  to their  approach  account  hand,  of null  avoid  regard  t h e use  can  the  definitions  formulated i n such  c o n c e i v e d of as  different the  be  representation  together thus  i n which  whether  experiences f o r i t s proper s e t t i n g s  available The  one  by  relationship  t h e way  must  how,  in chapter £ .  not.  the particular  restricts  Pi p a r a m e t e r  possible,  and  dialects  o t h e r ' hand.  development  or  o f t h e NPs  have seen  addressed  from  i n t h e p a r a m e t e r s and  adopted  l a n g u a g e s and  f o 11 ow  i s Case-marked  c o n c l u d e , whenever  reference  open  i s a separate issue  assumed  h e r e was  composing  interaction  then presented  i t . Pi c o r e g r a m m a r i s  between  those  principles,  the subcategorisat ion properties the representation  in  (£).  of  I  then  discussed  turned i n the  recoverabi1ity subject  of  a  the  null  of  tested In  i n the  to  particular  to  of  final  I t was found  those  the  previous pointed  In  interaction found  as  in Rizzi  (1986) and  ones  of  empty c a t e g o r y avoiding  i n the  principles,  etc.  and  of  this  the  specific  followed  by  between t h i s  property  and  which  was  was  was  i n , say,  Italian.  The  be  further  summarized  in  l.£.3.  here,  I have d i s c u s s e d p o i n t i n g out  i s contextually defined  of  theory  will  chapter,  reference  f o r m u 1 a t i on  null  l.£.£, H u a n g ' s a n a l y s i s  introduced  property  an  that  in  t o p i c languages  analyses.  this  analyses out  are  the s y n t a c t i c  instantiated  i n zero  s e c t i o n of  problems that  namely  empty c a t e g o r i e s a d o p t e d how  parameters,  the  following chapters  possibility overt  arguments as  t o p i c l a n g u a g e s was  developed  approach  the  argument  subject  the  more s p e c i f i c  Some o f  discussion of  null pro  null  i n most  zero  brief  the  of  were reviewed.  neglected of  the  following chapters,  languages.  property type  to  t o e m p t y NPs  as  the  in and  opposed  w e l l --formed n e s s c o n d i t i cms,  FOOTNOTES TO CHAPTER 1  1.  Cf. Baker  evidence  £.  X  3.  (1986s4)  (maximal  m a > > <  1.1.1  concerning  The term  (1980).  this  Projections relatively governor. presented  5.  projection)  Chomsky  only  X  only  (minimal  0  are"visible"  "external  argument"  positive  argument  an e x p l i c i t  n o t i on  into  i sbuilt  the further argument.  account  of Barriersi n  t h e d e f i n i t i on  o f Government.  a r enot b a r r i e r s i nthemselves t o the p o s i t i o n o f the govemee shift  i n Williams  and " i n t e r n a l "  (1986) d e v e l o p s  This  projection)  f o r M o v e m; s e e  i s introduced  (1981) c o n c e r n i n g  h a s no d i r e c t  effect  anymore with  but only  respect  to the  on t h e a n a l y s e s  here.  (198£) p o i n t s o u t t h a t  definition  o f empty c a t e g o r i e s  particular  definition  theory  arguments that  t h e X' s y s t e m .  between e x t e r n a l  Chomsky  wh i c h  assumes t h a t  See a l s o W i l l i a m s  distinction  4.  who p r e s e n t s  i sa v a i l a b l e t o t h e c h i l d .  Chomsky  and  (1979),  c a n be d i s p e n s e d  given  t h ef u n c t i o n a l  he adopts t o g e t h e r  with  a  of variables, condition C of Binding with.  I will  have nothing  t o say here  55 about  this  though still  p o s s i b i l i t y . Nevertheless,  c o n d i t i o n C i s not b u i l t exist  and t h a t  an  explicit  6.  This  part  i s true  nominative  into  i s the reason  i t remains that the theory,  i t s  even effects  I opt f o r maintaining  i t as  of the theory.  only  under t h e assumption  Case a s s i g n e r ;  c f . £.4  that  AGR  i sthe  f o r more on n o m i n a t i v e  Case  a s s i gnment.  7. E n g l i s h the  and I t a l i a n ,  presence  position are  Rizzi  (1986) a r g u e s ,  i n English  can serve  controller,  8. B r o d y the  In  on t h e b a s i s  in Italian  there  the verbs  For instance,  in  ( i ) , the pronominal  f o r the  t h e empty o b j e c t i n  f o r an anaphor, of a small  a similar  evidence,  syntactically  i s no e m p i r i c a l e v i d e n c e  discusses  sentence  of empirical  i s "active"  i t c a n be t h e s u b j e c t  Who« d i d PR0«  antecedent  i n the object  i n (40). In English,  as antecedent  (1984s363)  English (i)  by  intransitive.  p r e s e n c e o f an empty o b j e c t . Italian  are differentiated  i n t h e l a t t e r o f an empty c a t e g o r y  t h e empty o b j e c t  whereas,  respect,  of V i n the sentences  simply  that  in this  i t c a n be a  clause.  problem  with  respect  to  (i).  losing  his*  anaphor  w a y a n n o y e><?  PRO  (who) o f e. A c c o r d i n g  i s locally  fl'-bound  t o the contextual  by t h e  definition  of  empty c a t e g o r i e s g i v e n  bound  empty c a t e g o r y  Pi' - b o u n d other (i) ECP.  must  i s a variable.  empty c a t e g o r y  hand, be  Brody a  i n Chomsky  In other  i s a non-pronominal  p o i n t s out  pronominal  (1981:330),  that  because  the f i r s t i t would  a  words,  l o c a l l y Pi'— a  locally  non-anaphor.  On  empty c a t e g o r y  otherwise  violate  the in the  CHAPTER £ : NULL ARGUMENTS IN ROMANCE  £. O.  LANGUAGES  Int r o d u c t i on The  main  of  subject  of  INFL.  clitics  linguistic  observations  filler  language,  i n which  After motivating  1  possible  purpose o f t h i s  chapter they  this  approach  null  subject  chapter,  p o s i t i o n c a n be o c c u p i e d  proposal  that  which notion  clitics  "subject  as well  i n which  by pro.  as t h eother  c a n be c o n s i d e r e d  p r o i n CNP,IP3  argue that  pro.  ont h e  A s we h a v e  as null  i snot a spurious  that  the external  This  leads  languages subject  i s l i c e n s e d by a c l i t i c  clitic"  the only  i t i s now g e n e r a l l y a c c e p t e d  argument  subject  on t h e b a s i s o f c r o s s  NP a n d / o r  languages a r e t h e ones  French  u n d e r AGR  p o s i t i o n i s , depending  a phonetically realized  i nthe first  an a n a l y s i s  a r e base-generated  i n £.1, I w i l l  for t h esubject  seen  i st o present  to the with  languages i n  and that t h e  o n e i n t h e grammar.  It  i s argued  referred  pro  I wish  subject  clitics  by a c l i t i c  i n INFL. operates  i s no p r o - d r o p  or null  other  subtheories  this  property  puts  this  o f UG s u c h  i s needed  result of this  property  continue  with  problems  f o r t h e general  free  dialects,  to the conclusion  parameter  a look  that  no e x p l i c i t grammar.  (1981;198£).  and c a u s a t i v e s  i n Standard  reference Section  We  seem  a n a l y s i s i n the preceding Italian  follows to  £.3  i tto the  a t some c o n s t r u c t i o n s t h a t  i n v e r s i o n i n Standard  that  p e r s e . The  arguments  by c o m p a r i n g  i n Chomsky  of  i n £.£, e s p e c i a l l y w i t h  i n any p a r t i c u l a r  i n perspective  here a r e  o f how t h e l i c e n s i n g  lead  subject  and assume  under study  f o r a language t o e x h i b i t n u l l  from  namely  chapter  i n different  f o r some  conclusion  a r e presented  in this  treatment  this  The d e t a i l s  The d i s c u s s i o n s  generally  s e e £.£.4  i n the languages  ©-role.  possibility  £.1.  languages;  t o Case and  there  the elements  c a n be g e n e r a t e d  t o argue against  generated  respect  clitics"  positions across  details.  always  (1984) t h a t  t o as "subject  syntactic  that  in Rizzi  then t o cause  sections,  a n d some o f i t s  French.  A Generalization Rizzi  (1984) a p p e a l s  concerning  subject  t o some d i s t r i b u t i o n a l  clitics  dialects  and S t a n d a r d  "subject  c l i t i c " defines  i n c e r t a i n Northern  French  (St.Fr.)  a spurious  facts Italian  t o argue that  syntactic class.  (N.It.)  the notion e  This  59 conclusion represent (St.It.)  i s based an  a  the  intermediate  and  presence of  on  St.Fr. a  such  preverbal  INFL.  clitics  between S t . F r . St. I t . ,  they  contrary subject differ pro  to  and  former;  see  properties dialects  simply  be  clitic"  based  the  French  exists  from  that  of  assume t h a t must  be  a  existence  a  the  and  the  of  N.It.  Pied  Noir  possibility the  N.It.  subject  generated  of  clitics  i n the  of  empty  the  flGR  and  so  in chapter  real.  that  This  apparently  (PN  Fr. ) .  doubling  and  pro  license the The  i n the  3.  For  I  notion is case  by C o l l o q u i a l  In C.Fr. which  there  differs  i t is optional. N. I t .  same s y n t a c t i c p o s i t i o n .  It.  now  intermediate  i n the  in  would  hypothesis  3  to  and  in  the  represented  in that  i n C. F r .  contrary  latter  do  in  states  (1984) a n a l y s i s .  hypothesis  subject  node  pronominal  phenomenon f o u n d  French  is also  suggests  dialects  cannot  dialects  dialects  Rissi  i n the  extensively  another  is obligatory  i n St. I t .  N.It.  Rissi's  working  dialects  intermediate  as  clitics  doubling  It.  Italian  in St.Fr.  but,  the  i s grammatically  (C. F r . ) a n d  also  and  subject  discussed  assume as  between S t . F r .  as  also allow  subject  NP.  represent  clitics  f o r more on  the  "subject on  that  N.  e v e n when t h e r e  realisations  would  St.Fr.  the  there  subject  p o s i t i o n whereas they  £.£.4  will  they  So,  fact  of  clitic  St. I t . i n that,  St.Fr.,  the  dialects  preverbal  dialects  position. by  N.It.  phonetic  that  between Standard  subject  have s u b j e c t  in subject  will  are  These N.It.  state  In the  phonetically realised  that  assumption  This  I  thus  dialects analysis  is  then  extended  Thus,  lexical  pronoun.  These  the  St.Fr.  i n some l a n g u a g e s ,  where a  affixed  to  NP  subject  as  <1)  V .. .  SC1  clitic  i n saying  certain ask  circumstances  lexical  Pis  i s therefore  what  (2)  [ we  that,  occupies material  N P  0 :  SCI  as  t o be e x a m i n e d  that  I will  when  clitic.  assume  pronouns,  position.  lexical  i n 3.1.  on  NP  This  under  But then,  we  can  i t i s not f i l l e d  b  i n (2).  V .. . gap seems  conditions  necessary  f o r the presence  pronominal  pro. This  i s basically  f o r the N.It. dialects. with  a  position  see, t h i s  concerned  f o r subject  i n subject  will  (1984)  are  represented  to the English  to receive  the subject  c l i t i c  because they  c a n be  construction  contrary  free  sentences  i n t h e same s e n t e n c e a t l e a s t  pronouns a r e not generated  position  "clitics"  construction  of this  to find  by a c o r e f e r e n t  i n ( 1 ) , where SCI s t a n d s  Pi p l a u s i b l e a c c o u n t consists  verb  The d o u b l i n g  schematically  NP  i s doubled  pronouns are c a l l e d  to the inflected  surface.  i t i s possible  t o meet  discussion  of  (1) t o t h e n e x t  argue that  the representation  of the pure  the approach  In t h i s  the representation  a l l and o n l y  chapter,  empty  assumed  be v a l i d  be  delay  In p a r t i c u l a r ,  i n (2) must  i n Rizz  I will  i n (2) and w i l l  chapter.  the  I want  fora l l  t  61 languages To  with  subject  clitics.  do  so,  l e t us  languages  and  dialectss  without. the  In the  external  includes,  group  argument  Trentino,  with  (Italian,  no  with  there  i s no  pronoun above,  group  (clitic this  are  pronouns The  C.Fr.,  clitics  there  subject  or n o t ) .  observation  a r e not  that  no  is a  dialects Pashto  as  be  stated  and  Italian,  languages  (English,  of  full  as t h e  the  languages German).  known member o f language lexical  Given the representation  can  in  others.  pro drop i t a  found  group  are s o - c a l l e d pro drop  a t a l l be e  such  of  those  are  Piraha,  some o f t h e  s e e m s t o be  clitics  and  first  Modern Hebrew and that  groups  the N.It.  languages  whereas others  subject overt  and  o f two  clitics  F i o r e n t i n o , Modenese),  subject  Interestingly, group  the subject  St.Fr.  known f a c t  Spanish)  subject  p o s i t i o n CNP, I P 3 .  E n g l i s h , German,  i s a well  the existence  those with group  In the second  A  Spanish, It  latter  f o r example,  (including others.  postulate  in  B  the i n which  NP  or  a  (£)  generalisation in  (3) . (3) F i r s t v e r s i on Pi l a n g u a g e w i t h s u b j e c t c l i t i c s c a n n e v e r h a v e b o t h CNP,IP3 and t h e SCI p o s i t i o n s l e f t empty in simple tensed clauses. In  fact  (3) s t a t e s t h a t  languages  with  (4)  3  Chap. 0  C  subject  Q  C  1  0  3  the representation clitics.  V  in  (4)  i s barred  in  The  first  reference  version  to  "simple  constructions after  where  e x t r a c t i o n of  both  same  i s true  to  here.  be  generally  of  INFL,  have seen verbal  to  a simple  agreement,  The  clitics  second as  the 1,  morphology  different  have r i c h  empty  in  cannot  group we  are  not  the  in directly  clause  form  carries  and  in  1.2,  to  be  agreement  in  in  a  (4).  This  that  person  Italian,  accounted  for.  say, which  that  a  to  by be  English, does have  Pis  "rich"  that  see  is  pro.  morphology  number?  it  property  agreement,  features  this  and  clitics),  occupied  i s said  agreement  a way  f o r each  agreement,  subject  subject-verb  subject  the  i n such  no  have seen rich  NP  (with  d i f f e r e n c e between,  i s thus  material  tensed  have the  where  for.  i n Chapter  generally  overt  i s that  allow  realized  no  point  i s probably  lexically  agreement.  instance,  Inversion  constructions  assumed,  which  sufficient  the  Stylistic  derived  subject  to  is  the  The  f o r French  Such  accounted  Turning  for  p o s i t i o n s are  postverbally with  position.  language with has  (5),  various  venir?  is realized  relevant  In  are  (4) .  i l va  preverbal  subject  explicit  because there  the  Qui va v e n i r ? 'who w i l l come?'  subject  clauses" hold.  (5)a.  The  g e n e r a l i s a t i o n makes  i t does not  as  Qui  the  tensed  French  b.*  in  of  rich  when  are is  £. £„ £  which  we  on  does  not  63 Let is  us  just  pro  another  in  If this  languages includes  I return i s true,  i s much the  i n the  manifestation  CNP,IP].  done.  rich  assume t h a t ,  subject-verb  of  a  then  the  the  or  the  number o f  clitics no  of  subject INFL  details  previously  subject  agreement  group,  property  shortly to  larger than  ones with  first  as  of  null  agreement  allowing how  this  is  subject  thought well  clitic  since  as  it  those  whatsoever  with (see  1. c!. c') . We formal  can  now  terms  reformulate i n order  "simple  tensed  of  The  pro.  subject  as  to  clauses"  (5a)  generalisation  eliminate  possible  restriction  empty c a t e g o r y in  the  in  in  in  more  ambiguities.  The  i s replaced  (3)  by  the  mention  CNP,IP] a f t e r e x t r a c t i o n o f  is a trace  which  does not  fall  under  the  the  generali zation. (6)  Second v e r s i o n I f t h e grammar o f L i n c l u d e s s u b j e c t clitic p r o u n o u n s t h e n , w h e n CNP, I P ] i s p h o n e t i c a l l y n u l l a n d f i l l e d by p r o , a s u b j e c t clitic m u s t be present. 7  One  of  discussion want  to  the  numerous q u e s t i o n s  concerns the  argue that  status  i t can  be  of  raised this  derived  by  the  preceding  generalization  from  other  i n UG.  I  p r i n c i p l e s of  grammar.  £.£.  Pro In  and  this  Clitics section  I present  a way  i n which  the  generalization  in  (6)  here  can  i s to  appear The  be  implemented.  account  i n tensed  analysis to  similarities  subject  doubling  an  other  i n which  simply  different  notion  "clitic"  sake as  of  On  Pis  we  object i n the  be  clitics  and  and  syntactic subject  over  It  of  i s my  This  derive  belief  clitics  the  one  between  shown t o  object  realizations  of  be  same  both  that  are unique  where they  i s so  must  clitics.  asymmetries  will  NP  are  for  s y n t a c t i c r e c o v e r a b i 1 i t y as  the well  reasons.  have seen,  identified  of  Chomsky  where pro,  the  strictly  by  the  element  is rich  rest  of  this  t h e s i s with  must  be  nominal  element  e l e m e n t s and (gender)3.  a  to  PiGR o f  typology  the  basic  following points  set  of  features  agreement Contrary  usually  markers. to  of  the  of  null .  I N F L and  I assume t h i s  the a  presents  C+pronominal,-anaphorl,  enough.  l i c e n s e d by  number,  (1982)  corresponding  features  this  [person,  the  issue  Pro  realization  with  clitics  preferred  theory  subject  chapter.  s y n t a c t i c elements.  empty c a t e g o r i e s  Pro  next  the  lexical  by  The  doubling  subject  leamability  £. £. 1.  i s supported  cases.  structura1  a coherent  for  and/or a  between o b j e c t  i s to  different  i t differently,  i n languages with  non-doubling and  put  a clitic  presented  principles  account  quite  clauses  existing  i n the  from  f o r why  be  clitics  To  is only  idea  in  when the  elaboration. associated  These f e a t u r e s proposal  are  i n Chomsky  (198£) a n d  Rizzi  the  features  set  of  assignment present  (1986),  approach  as  will  particular  £.£.5.  that,  in English  pro  they  condition be  on  are  the  not  for  language-learner  be  the  type  of  the  surface  able  for this  realized  sense  coindexed  be  Finally,  licensing imposing  f o r pro  restrictions  the  with  on  with idea  the  of  the that  on  verb  the  also  argument  there  i s no  where pro  d i s c u s s i o n on being  a  in be,  two suppose  of  are  a  finite  not  cannot  license this  licensing  features  would  The  the involving  i f nothing set  govern  such  i n a very  thing  as  local  in  as  one  D-structure.  empty c a t e g o r i e s pronominal  by  a  i s understood  appear  on  of  p o s i t i o n occupied  notion can  this  Secondly,  How  recoverabi1ity. must  the  must  to  i t s language as  where t h i s  pro,  they  under d i s c u s s i o n here  the  for  features  realization  even though  analyze  features  I assume t h a t  domain  In c o n f o r m i t y maintain  to  reasons  in  subject-verb  l e a m a b i 1i t y reasons.  phonetically and  rich  of  Case  f o r pro  There are  lexically realized.  recoverabi1ity allows  these  each  lexical realization  adopted  member  pro.  i t i s natural  i s no  These f e a t u r e s  a  following sections,  First,  features  of  The  lexically realized.  required  be  domain  (1986).  f o r example,  to  licensing  licensing  i n the  assumption.  realized.  pro.  the  CCase3  I assume f u r t h e r t h a t  lexically since  take  f o r the  i n languages where t h e r e  bears the  must  not  i t is in Rizzi  sense,  for this  agreement, verb  not  become c l e a r  some o b v i o u s  reasons  required  i s therefore  assumption  in  I do  element  i n 1.3, with  I no  66 phonological  matrix,  that  restrictions  imposed  on  In  other  words,  w h e r e an  overt  positions and  pro  i s , an  i t as  are:  f a r as  i s allowed  pronoun can 6-marked  empty  pronoun,  i t s position  i n a l l and  appear at  A-position,  should  only  is  the  have  concerned.  positions  D-structure. non  not  These  6-rnarked  A-position,  A'-position. To  said:  summarize the  assumptions with  respect  to  pro,  I  have  s  P r o can a p p e a r a n y w h e r e an o v e r t p r o n o u n i s a l l o w e d and i t i s " l i c e n s e d " i n t h e s e p o s i t i o n s i f i t i s c o i n d e x e d w i t h a n d g o v e r n e d by s o m e ( p h o n e t i c a l l y ) r e a l i z e d s e t o f f e a t u r e s i n c l u d i n g p e r s o n , number and, i n s o m e c a s e s , gender . -1  It  follows  license  that  pro;  fulfillment  any  v a r i a t i o n among or  recoverabi1ity  to  of  the  of  of  pro.  empty  features in  the  lies  potential  therefore  to  in  condition  on  the  I return  to  the  the  £.3.  review that  the  some o f  are  the  attested  functioning  of  putative  i n the  the  syntactic  literature  in  order  above c o n d i t i o n  on  this  category.  One verb.  the  this  pro  illustrate  languages of  briefly  positions  grammar has  non-fulfillment  consequences of I now  particular  of Most  various (Jaeggli  the of  positions the  object 19S5).  cases  clitics French  available studied  in  for  i n the  Italian  pro  object  literature  (Rizzi  (Zubizarreta  i s as  198£),  (198£b),  of  involve  Spanish  Arabic  (floun  a  67 1979), (7)  Romanian  (Farkas  1978).  Some e x a m p l e s a r e p r o v i d e d  here.  11 a 1 i a n Gianni  l a presentera pro a Francesco. h e r int r o d uce~3s to 'Gianni w i l l introduce her t o Francesco. t  t  5  (8)  Spanish Juan  l o tv i s i t o pro*. him visited—3s 'Juan v i s i t e d him.'  (9)  French Marie  l a mange pro*. it eat-3s 'Marie eats i t . '  (10)  A  Arabic &i f - t - a i pro saw I s him ' I saw h i m . '  (11)  t.  Romanian L i -am va" z u t p r o . him h a v e - I s seen 'I have seen him.' 4  In  a l l t h e preceding examples,  appearance object  of pro are s a t i s f i e d :  position  such  the object c l i t i c features ruled if  out i n these object  that  affixed  f o r proper  the null  ( 1 2 ) a.  t h e two c o n d i t i o n s t h e verb  9-marks  i t c o u l d be o c c u p i e d t o the verb  "licensing". languages  bears  on t h e i t s  (direct)  by a p r o n o u n a n d  the necessary  Note that  these examples a r e  i fthe c l i t i c  i s not p r e s e n t and  i s not t h e r e s u l t  * Gianni presenter! pro a  of  extraction.  Francesco.  68  b.  *  Juan  c.  * Marie  visito  pro.  mange  pro.  d. * 2 i f - i t p r o . e.  Thus,  *  Am  even  positions other  vJlzut  though  pro  i t i s not  words,  The  pro.  the  as  a  pronoun  l i c e n s e d by  content  of  the  pro  context  where  preposition.  Stowell  (1981:144),  Hag i t  proposes that  marked which and to  but  that  i t i s the  receives a  the  pro  licensed.  In  Hebrew,  clitic  a  necessary  the  Modern  on  f o r the  of  suggestion  of  i s true  in this  the  of  the  or  not,  position  from  as  Arabic  p r e p o s i t i o n and  licensing  of  pro;  are  not  cf.  the  i t i s natural long  and  as  the  carries Borer  the (1984).  t  above examples  are  barred  i f the  i t is  Modern  Arabic  again  9-  verb  proi  pro  a  preposition  c o m p o s i t i o n a l l y by  this  In  object  complements themselves  assigned  appears  features.  i s as  following a  complement  appear  these  Hebrew  wiya-hai with her  Here  can  in  recoverable.  is attested  f o l l o w i n g examples  features  ' i t —o i with him b.  NP  p r e p o s i t i o n . Whether  assume t h a t  (13)a.  9-role  PP  appear  appropriate  i s not  next  Borer,  pro  could  clitic  is  69  absent. I  13  turn  Modern  now  Hebrew.  between  a  in  and  (12)  the an  overt  (13),  to  examples  below  Borer  (14)  state  complement  the  As  head  clitic  examples position in  available for  the  noun.  i n the  pro  This  is affixed  i s the to  the  pro  should  necessary  and  relations  complement  Nevertheless,  grammar a l l o w s  on  the  potentially  i s allowed.  genitival  noun.  assume t h a t  i s one  since  appear  lexicalized case head  in  the  noun;  cf.  Hebrew  a.  be i t — o ± pro ± house h i s  b.  beit - a i pro± house her Arabic  case  ruled  the out.  The  fact  does not  proi  clitic  i s omitted,  the  NPs  in  (14)  and  (15)  are  1 0  subject  languages  is  can  where a  ^ urnrn - a i mother h i s  the  a  in Arabic  (1964).  Modern  (15)  In  we  appear  state  construction expresses  particular  features  so-called construct  n o u n and  pronoun  i f a  the  This  head  construct  only  to  position in different  p o s e s some that  the  receive  possible only  interesting  external a  9-role.  problems.  argument In  this  i f i t i s moved  sentences  to  from  different  This  follows  from  p o s i t i o n of  certain  verbs  case  a referential  subject  this  p o s i t i o n from  a  9-  70 marked  position;  ©-marked explore  c f . i.1.1.3.  external what  argument  happens  I will  positions  i n the case  first  l o o k a t some  f o r p r o and w i l l  of expletive  licit  then  subject  posit ions. Pis we h a v e allow  empty  (16)  Spanish  (17)  i n l.£,  Spanish  (Jaeggli  b.  pro estamos cansadfsimos. be-lp very+tired 'We a r e v e r y t i r e d . '  Italian  (Rizzi  p r o comprero una macchina. buy-Is one c a r ' I w i l l buy a c a r . '  b.  pro verra. come—3s 'He w i l l c o m e . ' that  i n these  examples  and c a n r e c e i v e  the external  a full  This  pronoun  i s where  of  the verbal  paradigms i n I t a l i a n  It  i s usually  assumed  (18)  PiGR o f t h e h e a d  d p  [  N  others  19812)  a.  pro i s licensed.  with  among  198£)  pro b a i l a bien. dance-3s w e l l 'He/she dances w e l l . '  ©-marked  and I t a l i a n  subjects i n tensed clauses.  a.  Recall  how  seen  P  pro  4  1  that  the rich  and S p a n i s h  PiGRi TNS  property  i s appealed t o .  as i n  1  obvious  inflection  p r o i s g o v e r n e d by and  CiNP-u.  position i s  but i t i s not  INFL o f t h e s e n t e n c e  [ I N F L -  argument  coindexed  (18).  ...  Ill  71 The  A6R e l e m e n t  thus  licensed.  clitic  the  Let  to the possibility  c a s e s where t h e e x t e r n a l  directly  from  following  (19)  I return  the necessary  as l i c e n s e r of pro i n external  The role  o f INFL c o n t a i n s  Italian  surfaces  with  overt  pro i s subject  position in  does not r e c e i v e  a  9-  Consider  sentences.  s e e how t h i s  languages with  a  t h e VP a r e l e s s s t r a i g h t f o r w a r d .  Non e c h i a r o come r a c c o n t a r e '(It) i s u n c l e a r how t o t e l l us f i r s t  o f having  argument  argument  features;  type  of construction  pronominals.  the overt  certe storie. certain stories.'  pleonastic  In English  works i n  the string  pronoun there  o r i_t_ a s i n  (£0) .  (£0)a.  I t seems t h a t  b. These  There  i s a man  pleonastic  allowed  since  is  not projected  of  t h e verb,  Projection  left.  i n t h e garden.  pronouns a r e not r e f e r e n t i a l  i n non 8-marked  obligatory  the  John  subject  the external from  Principle.  no  i n t e r p r e t a t i o n with  presence i s  p o s i t i o n , even though i t  why t h e s e  required  properties  by t h e E x t e n d e d  A l t e r n a t i v e l y , i t c a n be a s s u m e d  pronouns a r e l e x i c a l the reason  argument  Their  t h e l e x i c o n b y t h e ©--marking  i s nevertheless  is  positions.  and a r e o n l y  r e a l i s a t i o n s of nominative pronouns a r e e x p l e t i v e  respect  t o thematic  since  that  Case.  That  they  have  r e l a t i o n s . Now,  they  cannot  be p h o n o l o g i c a l l y  English  given  Introduction verbal rich the  the poverty concerning  inflections.  inflection external  Once t h i s  expletive  pronoun w i l l  the  subject  assign  appear  languages,  be o c c u p i e d  and t h e c o n d i t i o n  predicted  that  expletive  external  argument  features  c a n be r e c o v e r e d .  involved Notice  subcategorized marked  object  but t h e i r  Principle from  this  since  the lexicon.  t h e Extended  In other  It follows  an  that  9-role  to  (non-thematic) Projection  licensing  words,  that  p o s i t i o n . Now, i n  i n languages where  i s referential  of pro, i t  i s  ©-marked  i t s nominal  t h e Extended sufficient  when t h e m a t i c  f o r the  relations  otherwise.  discussion positions.  occurrence they  i n the subject  on proper  though  i s  i t must be a s s u m e d  i s , i n a c e r t a i n sense,  and e x p l e t i v e  that  i t  p r o e x i s t s and o c c u p i e s non  positions  presence o f pro, which  Even  with  Projection  by a n o n - r e f e r e n t i a l  Principle  are  c a n be n u l l .  i s n o t ©-marked,  given  on  t o languages  does not have an e x t e r n a l  pro. Therefore,  Principle  applies  p r o c a n be l i c e n s e d .  that  but c f .  c a s e s a n d £.£.£  due t o t h e Extended  expletive  Projection  English  i s established,  of a verb  will  inflections;  the subject  position  obligatory  subject  pro i s not licensed i n  of verbal  particular  i n which  Principle.  null  since  T h e same a n a l y s i s  argument  nevertheless  null  cannot These  be e x t e n d e d  p o s i t i o n s a r e n o t ©-  v i o l a t e s t h e Extended  are lexically  t o non  governed  Projection  but not  projected  These inversion  remarks facts  have  implications f o r the analysis  i n languages  examine  them  in  The  last  construction  is  the subject  French  and  since  from  t h e TNS  Italian  that  p o s i t i o n of such  i t i s not  a subject  of  receive  infinitives.  governed  element  will  J'espere boire 'I  realized  languages only  to  Case-marked;  Case-marking. makes What  But  i t possible  and,  a r g urnent p o s i t i o n is  ©-marked  which the  and  pro could  therefore,  particular  agreement  like,  say,  phonologically  does  not  get  Case  cafe. coffee  coffee.  coffee.  subject  will  i f there  f o r PRO,  in English  "exceptional"  (1985)  being generally  constitutes, Two  that  c 1 a uses with  therefore,  exceptional ungovemed,  arise.  language does  the subject  non  to f i l l i t .  external raising  verbs  a potential position in  possibilities  grammar o f t h e  the  and  means f o r i t  f o r more on  i s the fact  tensed  p r o c e s s between  allowed  t h e pronominal anaphor,  here  o f non  be  i s an  the position,  appear.  In languages  here  coffee.'  c f . Massarn  i s more r e l e v a n t  attention  cafe.  to drink  other be  du  I hope t o d r i n k  ft l e x i c a l l y  our  INFL.  hope t o d r i n k  (££)a.# I h o p e J o h n b.  Spanish. I  i s generally  <£l)a.# J ' e s p e r e Jean boire du I hope-Is to+drink of+the 'I hope Jean t o d r i n k c o f f e e . ' b.  and  £.4.  English,  null  like  of  and  not  In t h e include  the verb.  Pro,  first, any even  74 though is  i t can  the case  external  appear,  in Italian,  argument  infinitives. sentences  i s not  f o r example,  position  PRO  structures.  The  in  the d i s t r i b u t i o n  1.3  with NP  that  that  of  cannot  null  lexical  appear  appear  infinitival could  be  pro  be  taken  infinitives. would  not  <£3)  Rizzi  as  But  be  identical  phonologically  realized  t h e o t h e r hand  view,  evidence  of  of nominative since  the  that  pro,  surface we  Case  following pro  is unlikely  might  since  i f a  which  NP,  case  of  overt subjects Case.  This  would  relevant  contrast  this  lexical  i s the  present  i s not  be  did  pattern  pronominal  i n the  ungrammatically lack  that  in  since  i t i s t o assume as  supposes  This  in  o f empty c a t e g o r i e s s h o u l d  appearing  this  have  not  to verify  then  the  According to t h i s  could  to verify  This then  to the  from  will  barred.  is available  c l a u s e s but  is difficult pro  i s thus  given position  On  sentences  prevent  pro.  a  either.  attributed  way  NPs.  in a  counterpart of  cannot  not  obvious  or  and  where pro  of tensed  This prediction  with either  licensed  present  in in  occurrence  for  Italian Italian of  pro  licensed.  (1982)  a. * M a r i o a f f e r m a v a C q u e s t a d o n n a n o n v o l e r l o s p o s a r e II. ' M a r i o s t a t e d C t h i s woman n o t t o w a n t t o m a r r y h i m 3.' b. * ?  c.  Possiamo r i t e n e r e L queste persone avere sempre f a t t o i l l o r o d o v e r e 3. 'We c a n b e l i e v e C t h e s e p e r s o n s t o h a v e a l w a y s d o n e t h e i r d u t i e s 3.'  O f f e r mo C d i PRO a v e r e s e m p r e f a t t o i l rnio d o v e r e ' I s t a t e C PRO t o h a v e a l w a y s d o n e my d u t y 3.'  3.  d.  R i t e n g o C d i PRO ' I b e l i e v e C PRO  The  second  exhibit  that  agreement.  Raposo  (1985) r e p o r t s  of  verifying  infinitives  infinitives other  with  words,  i f the  be  coreferent  object  or  to  be  not  the  with  pro  would  sentences  granted  see  on  £. £. £  be  Galician  1 1  availability showing  are  not  w e r e PRO,  This  of  pro  that  as  we  would the  The  i s the  case  subject  of  inflected  Control  in reference.  ambiguous.  3.  would  P o r t u g u e s e and  ( c o n t r o l l e d by)  arbitrary  be  language,  tensed  that  c o n s i s t s of  subject  a  agreement;  empty s u b j e c t s  either  therefore  non  subject-verb  way  inflected  p o s i t i o n of  this description.  language that  In such  is rich  with  One  involves a  infinitives.  in subject  there  accord  possibility  inflected  licensed  a v e r e s e m p r e f a t t o i l rnio d o v e r e t o h a v e a l w a y s d o n e my d u t y 3.'  structures. expect  matrix  sentence  In  i t to subject  or  would  in English  in  (£4) . (£4)a.  I expect  b.  I hope  c.  I tried  C PRO  In P o r t u g u e s e , construction  C PRO  the  (£4)  only  leave  arrive  to  win  3. soon  the  other  i s ambiguous,  between a c o r e f e r e n t The  to  C PRO on  to  reading  and  the  infinitive  verb  3.  a very  argued  d i f f e r e n c e between the  i s that  prise  hand, as  3.  similar  i n Rouveret  a disjoint sentences  in  type  (1980),  reference (£5)  in Portuguese  is  of  and  reading. those  inflected.  in  The  following  agreement  examples are  markers are  taken  from  Rouveret  (1980);  cf.  footnote  1£.  underlined?  (£5)a. A c r e d i t a m C t e r e m g a s t a d o e s s e d i n h e i r o p a r a n a d a 3. ' T h e y t h i n k C t o h a v e s p e n t t h i s money f o r n o t h i n g 3.' b.  C o n f e s s a m C d e v e r e m l h e a v i d a 3. ' T h e y a d m i t C t o owe h i m l i f e 3.'  c.  L a m e n t a m C n a o s a b e r e m l e r 3. ' T h e y r e g r e t n o t t o k n o w (how)  d.  (£'5a) i s a s  ambiguous as  sentential  complement  complement  in  content.  g.  I  thus  then  verb  (£5e-g) b e l o w  f.  has  and  from  (£5d). no  subject  can  be  the  should  be  no  be  possible to  on  the  sernos t o be  Afirrna C pro terem 'He a f f i r m s ( t h e y )  the as  of  inflected on  This  argument  that  position  of  pro  i t s marker  i s shown  on  in  (1980).  pobres poor.'  3.  l h e r o u b a d o e s t e l i v r o 3. t o have s t o l e n t h i s book f r o m  D i r - s e - i a C p r o n a o p o d e r e m c a r l a r - s e 3. 'One w o u l d s a y ( t h e y ) n o t t o be a b l e t o s h u t conclude  tensed  (£5a).  restriction  verb.  a  nada3.  sentential  in  subject  with  have a d i f f e r e n t  embedded  Rouveret  E l e d i z I pro 'He s a y s (we)  i t s counterpart  Note that  overt  pro  there  It should  matrix  (£5)e.  (£5d)  in  i t i s true that  infinitives,  the  read.'  A c r e d i t am Cque p r o tern g a s t a d o e s s e d i n h e i r o p a r a ' T h e y t h i n k C t h a t t h e y h a v e s p e n t t h i s money f o r noth ing3.'  Sentence  If  to  can  in fact  inflected  occupy  infinitives  the at  up.'  external  least  in  him.'  77 P o r t uguese. It must  should  be o t h e r  licensed. ones  be c l e a r  particular  The ones  instance,  the preceding  constructions  I reviewed  i n the generative  guarantee that  from  here  literature,  only  Yiddish  offers the  one p r e s e n t e d  of  here even though,  governed,  which  empty e x p l e t i v e s  analysis  cf. Safir  here;  incompatible  £.£.£.  On  with  demonstrate  appealed  that  problem  i s compatible  f o r h e r , p r o must  with  be  Travis'  account  feature  pro. I w i l l  fora different  agreement This  on t h e n o t i o n t o up t o now.  the intrinsic  pronouns and t h e i r  property.  (1984:££6~£67)  not review approach  this  that  assumptions.  now c o n c e n t r a t e  been r e p e a t e d l y  like  (1985)  Malagasy  Agreement  I will  much  my  Travis  on a p a r t i c u l a r  by INFL o f e x p l e t i v e  we11-documented  Irish,  i s not t h e case here.  relies  p r o c a n be  (or pleonastic)  phenomenon t h a t  identification  is  i  of this  there  a r e languages, f o r  German,  sort.  that  does not c o n s t i t u t e a  There  empty.  a r e languages of t h i s  an a n a l y s i s  properly  which  (some) e x p l e t i v e  p r o n o u n s c a n be p h o n o l o g i c a l l y and  i n which  a r e t h e most  the set i s exhaustive.  i n which  discussion  nature  I will  discussion  of determining  will  revive  the differences  to  look  to the null  t h e much  has  clitic  make t h e m  respect  which  attempt  of subject  syntactic properties markers with  "agreement"  very subject  discussed  and s i m i l a r i t i e s  between  78 agreement greater In  detail the  generally carries a  way  m a r k e r s and  of  agreement this  we  The  information  by  CNP,IP3  i s absent.  concept  of  rich  that  are  i s generally  verb  stem  question  inflection  can  relevant  for  leamability  words,  can  how  will  that and  i t s grammar an  of  pro  question, Let  inflection  must  must  be  looked  Approached  but  also  only  That  such  an  become t r a n s p a r e n t  and  the  concept  even  is  In  other  to  tense of  trivial.  verb  the to  answer  paradigms of  across  the  establish  for  some  show t h a t paradigm  agreement  across  how  between the  i n terms of  —  in  problem  In order  Italian  —  recover  to  learner  verbal  vertically  angle  This  "rich"  subject  regard  process  i s , the  horizontally  to  lexical  such  each  agreement  the  considerations.  and  only  for  in  that  position?  the  morphology  realized  is "sufficient"  at  not  verbal  it is  called  with  agreement  that  look  defined  not  in  is often  language  argument  syntagm.  at  from  must  the  compare French  be  i n terms of  number —  in this  us  also  an  argument  one  languages.  should  includes  (external)  licensing this  asks  Spanish,  different  defined.  especially one  the  i f a  arises be  and  is sufficient  i s unavailable  One  i s addressed  lexically  i s therefore  the  that  Italian  characteristic  intuition  carried  like  that  morphology This  problem  have seen,  features  number.  inflection. morphology  as  This  4.  languages  assumed,  p e r s o n and  the  in Chapter  case  that  clitics.  rich but  morphology person and  rich  and  mood. agreement  It in  i s possible  such  a way  subject-verb  (£6)  that  to study c e r t a i n verbal  p a r a d i gms  one  French  would  agreement.  infinitive present  conclude that  Consider the examples  a.  aller  ind. Is £s 3s lp £p 3p  'to  vais vas va aliens allez vont b.  etre  't o  avoir  f o r £s  ending  f o r each  accepted  and  3s,  person  definition  conclusion  does not  syntagrnat i c a l l y as  each and  of rich hold (£7).  rich  (£6).  go'  be' Es ^ i ] C€] C€] Esom] E€t] [so]  'to  a i as a avons avez ont Except  has  Ev€] Eva] Eva] Ealo] Eale] Evo]  suis es est sommes etes sont c.  in  i n French  have  Ee] Ea] Ea] Eavo] Cave] Co]  of the verbs  in  number.  i s the  That  inflection.  i f one  (£6)  has  a different  generally  Nevertheless,  compares more  this  paradigms  :7>  infinitive s  parler ' to  a.present is £s 3s Ip £p 3p  b. f u t u r e  par l e paries par 1 e par1cms paries parlent  Cpar13 C p a r 13 Cpar13 Cpar153 Cparle3 C par13  The  situation  the  same e n d i n g  combinations. same v e r b  (£8)  becomes c o n f u s e d  next  infinitive: present  conjugation  -are  in Italian  not  rich  evident in  (£9).  the  proves  enough, w h e n we  i t i s often the  useful  Cpar1€3 Cparl€] Cparl€] C p a r 1.jo] Cparlje] Cpari€3  case  that  person/number  t o compare  conjugation of  i s the  paradigms of  the  Italian  verb  the  that  in  (£8).  ending  with  speak' parli parli parli parliamo parliate parliano  same f o r a l l r e g u l a r v e r b s  i n present  which  par1a i s par1a i s parlait p a r 1 i oris pariiez parlaient  tenses.  parlare 'to  to conclude  as  CparIre] Cparlra3 CparIra] [parlro] CparIre] Cparlro3  four different  subjunctive Is £s 3s lp £p 3p  This  natural  represents  It thus  c. p a s t  par1erai par1eras par1era parlerons par1erez parleront  in different  Consider  speak'  s u b j u n c t i v e . It would  s u b j e c t - v e r b agreement  i s the  wrong  compare d i f f e r e n t  then  in Italian  conclusion. This  paradigms of  seem  the  is  becomes  same v e r b  as  (£9)  infinitives  pariare  present Is £'s  Each  speak'  b. f n t u r e  par l o par 1 i par 1 a pariiamo parlate parlano  lp £p  the  'to  parlero par1erai par1era parleremo parlerete parleramo  person/number c o m b i n a t i o n three  syst  c.imperfect  tenses.  This  in  parlavo parlavi parlava parlavamo parlavate parlavano  (£9)  constitutes,  has  i t s own  indeed,  ending  a rich  for  agreement  em. To  summarize,  tenses  should  be  establish  that  developed  enough  exposure sufficient  the to  provide  involved  i n the  conclude  that  problem  The  the  i s a spurious  language-learner  language has  rich  conjugations  for regular  are  presumably  The  P r i n c i p l e of  procedure since  a  subject  can  it  be  must  an  of  pro  learner  to  system  that  i n the  argument  process with  defining  probably  a concept  the  is  verb.  such  as  determines whether  at  the  time  verbs  are  established.  i n the  i n determining  the  making  motivates the  sentence with  constitute evidence licensed.  agreement  in different  I "rich"  one.  involved  tensed  language  l i c e n s i n g of  Recoverabi1ity  involved  simple  f o r the  agreement  agreement  not  d i f f e r e n t paradigms  language has  position  inflection  to  that  of  various Irregular  such  the  richness  no  the  a  "decision".  discovery of  the  phonologically  pro  verbs  i s involved  agreement realized and  that  Having developed that  determined  of  licensing  —  will  the  verb.  We  can  and  ask  pro  drop  why  the  licensing  the  that  grammar o f  i t should  language —  i f there  done through  us  subject  the are  and  agreement  in  Trent i n o *parli, *parla, *parla,  position  of  as  subject  1 e  te+parli el/la+parla i/Ie+parla  pro  by  This  morphology  on  never clitics the  (6),  This  above  in  verb  these of  view  be  i t must  they  the  in  present.  two the  o b l i g a t o r y i f the  prediction  i n the  be  take  makes  g e n e r a l i z a t i o n (6).  is  on  responsible for  available  always  a l l barred clitic  such  in subject  solely  clitics  i s empty.  i s expressed  l a n g u a g e s when no  subject  f o r some r e a s o n ,  subject  (30)-(3£) a r e  are  generalization in  languages should  argument and  the  is  so-called  s u b j e c t - v e r b agreement. First,  obvious  clitics  licensing  that,  be  (6)  realizations  analyzed  sufficiently  generalization in  assume t h a t  f o r the  have a  system,  morphological  pro  these  verified  £s 3s 3p  Let  given  of  does not  r e t u r n to the  clitics  over  predictions.  (30)  be  now  Again,  case  paradigms  not  necessary  the  be  in this  languages.  position.  external  pro  languages with  languages are  priority  French  s u b j e c t - v e r b agreement  licensing  features  that  seems The  different  to  (31)  h i o r e n t i no Is 2s 3B lp 2p 3p  (32)  *vengo, *vieni, #viene, #viene, #venite, *vengono, French  1B 2B 3s lp 2p 3p This  e'+vengo tu+vieni e'/la+viene si+viene vu+venite e'/le+vengono  #parle, *parles, *parle, *parlons, *parlez, *parlent,  i s so  je+parle tu+paries i1/elle+parle nous+parlons vous+parlez ils/elles+parlent  because  identification  licensing  of  i s done through  pro the  subject subject  by  clitic.  subject  position  i s p h o n e t i c a l l y empty,  then  present  in order  to  then  pro The  system is  i s not second  of  the  verb  agreement,  I  not  in  the  for  clitic, the  some o f  to  be  the i n the  that  i s ruled  and  are as  the  clitics.  clitics  the  should  comparing  available  i s given  in  (30)  systems  and  first  always  a  as  of  pro  subjectpresent.  paradigms & Cord i n  above, and  shown can  p r e d i c t i o n s the  rich  be  in Brandi  in Trentino  Nevertheless,  of  is rich  (31)  always o b l i g a t o r y . This the  out.  different  data  be  present  If licensing there  the must  i s not  coexistence  even though  to a study  agreement  s a m e way  string  subject  q u i t e developed. clitics  the  then,  which the  clitic  available  have access  indicates  (31),  with  I f the  concerns  If  a clitic  pro.  and  prediction  N. I t . d i a l e c t s  (1981),  seem  licensed  agreement  d o n e by  do  license  feature  Florentine  in be  (30)  and  accounted  clitics  must  be  present  in order  Subject surface the  clitics  and  this  (£7).  has  the  I t can  follows  thus  be  Taraldsen's  the  sense that  of  the  missing  of  by  the  the  that  in  verbs  the  intuition  clitics  to  i n the  at  least  on  (3£)  this the  respect recover  s a m e way  the  (+)  affixation paradigm  to  feature  agreement  of in  here  agreement  the  in  found  a n a l y s i s presented  with  help  position.  notational device  disambiguating  said that  arguments  to  i n argument  markers content  markers  do;  1. 1. 1. Let  us  asking  element clitic  now the  with  not the  a regular  subject  empty  suggestion  develop verb  the  in  the  agrees with element  verb,  the  in  in external  detail  of  i n the this  CNP,IP].  i t s external external  In  argument  argument  as  the  subject  which  is in  languages  to  fact  with  that  the  position.  This  position  clitic. only  I will  here;  the  propose  maintaining  grammar.  is  either  where  angle  i t seems r e a s o n a b l e  a subject  process  be  pronoun,  argument  advantage of  different  (30)-<3£),  In E n g l i s h ,  to  a  It could  i t seems a l s o n a t u r a l t o  the  an  what,  agrees?  e m p t y NP.  agreement  i s not  verb  cliticized  NP  agrees with  clitic  same f a c t s f r o m  construction with  has  subject-verb  the  nominal  the  in a  the  verb  clitics,  agrees with  the  following questions  pronoun or  assume t h a t just  approach  which  pronouns are  is  Notice  effect  pro  affixed  i s expressed  in  cf.  identify  are  paradigms above.  clitics  by  to  I simply  and  that  proposed  This one  not  verb  process  attempt  to  assume t h a t  the  of  a  subject  in Safir  (1985).  We  have a l r e a d y  argument clitic  position  licenses  seen  pro.  The  the  licensee  p r o . ft s u b j e c t  in  CNP,IP3  licenser  rule  inside  of Verb  subject  with  licensing  t o govern  position.  (30)-<3£) m a k e s  generated  pro can  i n languages  requires  the  that  clitic  The  Raisings  the  clitics  position  i f such  4  be  itself  t o assume t h a t  CI i t i c i z a t i o n  in  able to  of the  f o r the d e t a i l s  the govern  pronouns  they are  operates through  a  £.£.1  o c c u p i e d by  therefore  nature  cf. Chapter  external  process developed  must  i t possible  INFL.  subject  the  clitic  occupy  basea  late  concerning  cliticization.  The  structure  in  (33)  results  from  the preceding  discussion. (33)  IP / NP  \  ±  /  INFL' \  INFL / flGR I Cl i Many a u t h o r s h a v e (1982), for  proposed  Brandi & Cordin  the N.It. dialects,  (1982)  the  VP \ TNS  and  Sportiche  possibilities  one  i n ( 3 3 ) : t h e NP  material  or  i t i s not.  structure;  (1981), J a e g g l i Everett  (1983)  Two  a similar  arise  (1985)  (1984)  cf. and  for Piraha,  Bouchard Rizzi  (1984)  Jaeggli  for St.Fr. i n the case of a s t r u c t u r e  position In the  is filled  latter  case,  with  some  such lexical  pro o c c u p i e s the  as  86 subject  position  The this  clitic  and i s l i c e n s e d  accounts f o r the functional pronouns  g r a m m a r a n d may b e m o v e d languages having  fill of  with  S e c o n d l y , t h e TNS e l e m e n t  therefore  free  n o t a good  this  Since  and  this  clitics simple them  VP p r i o r  to S-structure  i n the  node i s i n INFL. f o ra  i s a primary d i v i s i o n  with  Kayne  theory  among  o f pro, t h e c h i l d  t o recognize  respect  the c l i t i c  subject  status  i n the behavior to inversion  (34a) and (34b),  has t o  pronouns a r e  position  as l e x i c a l  subjects.  there are  o f pronouns.  of lexical  One o f  NPs and  i n F r e n c h . The c o n t r a s t s  and between  pronouns  NP  whether  ( 1 9 7 2 ; 1 9 7 5 ) , we know t h a t  evidence that  (35) a.  i n the  f o r Cl_ t o o c c u p y  positive  b.  of  another purpose  number o f i m p l i c a t i o n s  there  agreement  to the licensing  c f . £.4. T h i s  c a n be done by o b s e r v i n g  i sthe difference  (34) a.  between  o r n o t t h e l a n g u a g e h a s t h e means t o l i c e n s e p r o  tests  between  serves  the availability  o r not. Since  pronouns  similarity respect  inversion; position  languages concerning whether  into  has a c e r t a i n  l e a m a b i 1i t y .  fix  i n INFL.  i s g e n e r a t e d u n d e r PlGR f o r t w o r e a s o n s . F i r s t ,  markers and c l i t i c pro.  by t h e c l i t i c  (35a) and (36b) c o n s t i t u t e  a r e n o t i n t h e same  Viendra —t—il? ' W i l l he come?' Quand v i e n d r a ( * - t ~ ) Y v e s ? 'When w i l l Y v e s c o m e ? ' Quand a - t — i l d i t q u ' i l irait? 'When d i d h e s a y t h a t h e w o u l d g o ? '  syntactic  b.*  Quand  (36)a.* b.  a Jean  'When d i d J e a n  say that  Quand  a dit-il  qu'il  Quand  a d i t Jean q u ' i l  In o t h e r  words,  clitics,  then  Fiorentino). clitics,  d i t qu' i l  irait? he would go?'  irait? irait?  i ft h e pronouns a r e never missing  p r o c a n be l i c e n s e d  (French,  Trentino,  I f t h e pronouns a r e never missing  then  and a r e  and a r e not  pro i s not a v a i l a b l e ( E n g l i s h ) . I f t h e pronouns  are  missing,  and  t h e p r e s e n c e o f p r o i s made p o s s i b l e b y l i c e n s i n g  rich  verbal  fourth is  then  they  inflection  possibility  excluded  —  o f which  similarities  of  and B i n d i n g  a r e presented  i n the behavior  to the principles  structural  arguments) differently  Portuguese,  t o (6)  —  through  Spanish).  The  and a r e c l i t i c s  —  and o b j e c t s with  framework,  of elements  i n the sentence  o f UG c a n b e c h a r a c t e r i z e d subjects  arguments)  from  their  with  i n terms  (external  c a n behave  to extraction possibilities  section,  symmetry  the relevant  i n 1.1, c e r t a i n d i f f e r e n c e s a n d  (internal  respect  In this  subject/object  Clitics  p r o p e r t i e s . F o r example,  f o 1 1 o w s o n some a c c o u n t s predicate.  according  pronouns a r e missing  C l i t i c s and O b j e c t  In t h e Government  respect  (Italian,  —  by ( 6 ) .  £.£'.3. S u b j e c t  aspects  a r e not c l i t i c s  and  this  re1at ionship with the  I present  one major  general  i n t h e r e l a t i o n s h i p between c l i t i c s  and  the  argument  section,  position  they  i t was a r g u e d  are related  that  the structure  subject  clitics  account  f o r the cross-1inguistic  specifically, argument clitic must  position  licenses  appear  position This empty  clitics. Arabic,  which  account  c a n now  argument  be f i l l e d  of pro. This subject  i n D-structure  Consider  the following  and F r e n c h  (Standard,  Conosco Gianni know—Is' ' I know G i a n n i  Colloquial  c.  L o i c o n o s c o e± r n o l t o b e n e , him ' I know hirn v e r y w e l l . '  Spanish a.  (Jaeggli  1982)  Vimos l a casa de M a f a l d a . saw-1p t h e ho use o f 'We s a w M a f a l d a ' s h o u s e . '  b. * V i m o s e. c.  bene.  La* vimos her/it 'We s a w i t . '  e . t  from  why a  the  c l i t i c  when t h e s u b j e c t clause. the cases of  i n languages with  rnolto bene, very well very well.'  b. * C o n o s c o e r n o l t o  explains  t o cover  to  external  by p r o . I n t u r n ,  clitics  examples  suited  i n ( 6 ) . More  the  i n a tensed  be e x t e n d e d position  i s best  generalisation  Italian a.  (38)  i n flGR o f I N F L ,  can only  the presence  i s empty  i n ( 3 3 ) , where  i n (33) c a n g o v e r n  i n languages with  internal  (37)  are generated  the c l i t i c  to. In the preceding  object  Italian,  and P i e d  Spanish,  Noir).  1  3  (39)  Arabic a-  sif-it saw-is ' I saw  Xalid Xalid.'  b. * s i f - i t c.  sif-t Is ' I saw  (40)  e.  -a* e±. him him. '  French a.  Pierre a vide la bouteille. has e m p t i e d t h e b o t t l e ' P i e r r e emptied the b o t t l e . '  b. * P i e r r e a c.  e  Pierre l ' a videe her/it ' P i e r r e emptied i t . '  e±.  t  In t h e s e cannot  vide  languages with  h a v e an  object  internal  clitics,  argument  generated  empty c a t e g o r y  (c).  generalization patterns  This  concerning  external  presence of forced  by  the  the  (b)  empty  category  Projection Principle. presence of  the  pronominal  pro  f a r as  between an must  be  the  object  similar  to  by  object the  one  a  in object The  a clitic  i n argument  verb  (a)  base  clitic- is in  present  (6)  same e x p l a n a t i o n  (pro)  forces the  As  an  with  The  Principle empty  represented  unless  arguments.  a transitive  holdss  position is  Recoverabi1ity in order  to  license  position.  s t r u c t u r e i s concerned,  the r e l a t i o n s h i p  clitic  associated  the  and  one  in  the  argument  (7),  the  i . e . the  clitic  with being  i t  affixed  to  the  illustrated  verb  in  can  govern the  object  position. This  is  (41).  (41)  IP /  \ NP* INFL' I / \ pro INFL VP / \ / \ PiGR TNS V NP.j I I I C1i C 1 j +V pro  Note that the  this  structure  generative  (198£b),  1  Sportiche  doubling,  important clitics  to  can  (1983),  same e l e m e n t , in  a r g urnent  1,  £.£.4. In are  of  to  be  two  object a noun  of in  a  precedes  they  i n Chapter  distinct  are  that  and  others.  be  subject  i s to  pre  t o ot her  presented and  the  license  e x t end ed  (1£)  tc  is  r e a l i z a t i o n s of  have been in  respect  What  and  which  can  preposition  in  (13),  (15).  Clitics  I have assumed base  belong  of  in  Zubizarreta  among  3.  clitics  d i s c u s s i on  (14)  (198£),  accepted  e s p e c i a l l y with  syntactic function This  generally  (1984;1985)  object  involving c l i t i c s  elements that  construction  Jaeggli  i s that  be  Chomsky  discussed  Base G e n e r a t i o n of what  cf.  asymmetries,  be  assumed  pos i t i on.  namely,  complement  will  the  constructions £.£.  and  note here be  seems t o  literature; -*  More s y m m e t r i e s clitic  now  generated  to.  The  that  clitics  on  some h e a d  problem  of  in in  general the  determining  whether  91 clitics  are  base  surface  or are  related  to  generated  moved  i s one  from  i s developed  Rouveret  & Vergnaud  be  found  (19823.1985), here  different Jaeggli mostly  i n Kayne  (1984) and arguments  approaches.  (1982:15-20) concerned  construct ions:  argument  they  i n the  Quicoli  The  (1976), Safir  put  reader  f o r such  I return  Rivas  (1985).  forward  a review.  to the  concerning  generation across  theories  differ  Rizzi  (1984),  the  subject  clitics  clitic"  i s not  referred  t o as  second,  view  to the  "cliticization"  In  refers  due  and  which  clitic"  to a very  even w i t h i n I want  i s not  a single  precise syntactic should  be  these  two to  on  different  two  two  already  (1984)  and  notion "subject i s often  i n v a r i o u s ways particular  notion.  The  exhibiting  what  i n the  base  present  I have  f o r here,  a s p u r i o u s one  that subject c l i t i c s  and  operate  t o argue  attempt  In the  i n Kayne  that the  i n UG  not  languages.  languages  fact  can  and  t o which  mainly  among  a homogeneous one  languages  "subject  defended  differences  are  across the  first,  respect  Jaeggli  in particular,  i n 2.5.  theories  major  former  causative  subject c l i t i c s  i n one  latter  These s t u d i e s are  I c o n c e n t r a t e on their  the  and  the  to defend  and  latter  The  (1980)  I will  on  are  (1977),  i s referred,  with object c l i t i c s  In the  they  literature.  section,  referred.  occupy  position  (1975),  in Strozer  Borer the  position  (1978) whereas d i s c u s s i o n s o f  mainly  to review  the  much d i s c u s s e d  approach  can  i n the  the  grammar. concept  grammar  It follows  generated  and from  in a  this  unified  way  across  follow  languages.  from  subject Let (1984),  i n the  us f i r s t  review  besides  counting  sent ence  1 i ke s  (4£)  a  On  has  subject  which  hand,  takes  First,  i n syntax  leaves  Rizzi  data.  a trace  of  Clitic-  i n argument  i n a St.Fr.  of object  i t i s p o s s i b l e t o account  Qui ca,  les?  of component  clitics,  i n t h e mapping  i f i 1 s i n (43) i s " l e s s  (44) a.? Q u i 9a, iIs? who t h a t t h e y ? ' 'Who i s t h a t , t h e y ? '  by a r u l e  i n the phonological  p o i n t s out v a r i o u s  l i s 1es vo i ent. t h e y them s e e 'They s e e them.'  b.*?  the relevant  proposes that  (PF).C l i t i c i z a t i o n  operates  l e s , then  place  (44) .  (43)  which  as an anaphor,  S - s t r u c t u r e . Kayne  than  (1984) and  rung.'  t h e grammar  approach.  i n Kayne  pronoun c l i t i c i z e s t o t h e verb  cliticization  to  properties of the  the existence of a rule  clitics,  then  Sonne.  'One  other  would  1 3  the analyses  assuming  f o r object  position  of  languages.  i n t r o d u c i n g a t t h e same t i m e  Placement  variation  differences i n the intrinsic  clitics  Kayne,  the  Cross-linguistic  facts  from  on t h e  D-structure  supporting  thoroughly"  a  this  clitic  f o r the contrast i n  Secondly, strings  there  well-known  i n French 5 c f . Kayne  requirement clitics  are  as  that the  subject  first  ordering (1975).  clitics  One  appear  member o f t h e  c o n s t r a i n t s on o f them to the  clitic  i s the  left  string  clitic  of  as  object  shown  in  (45)  ( 4 5 ) a.  II le l u i a dit. he i t t o - h i m / h e r h a s t o l d 'He h a s t o l d i t t o h i m / h e r . '  b. * L e  i l l u i a dit.  c. * L e  l u i i1 a d i t .  If  i_l_above c l i t i c i z e s  do  so  prior  accounted  to S-structure  inflected  ( 4 6 ) a.  then  i n PF  the r e l a t i v e  the negative  verb  and  the  m a r k e r ne  subject,  be  always i t a  clitic. Pierre  n' est neg i s ' P i e r r e i s not  pas a Paris. not at in Paris.'  b.  Tout n'est pas a P a r i s . ' E v e r y t h i n g i s not i n P a r i s . '  c.  I I n ' e s t pas a P a r i s . 'He i s n o t i n P a r i s . '  (47) a . *  and  the object ordering  clitics  can  be  for.  Finally,  subject  to the verb  Ne  Pierre  b. * Ne  tout  c. * Ne  i lest  est  est  pas  pas  pas  a  a  a  Paris.  Paris.  Paris.  appears  full  between  lexical  NP  or  the a  9 4  Nevertheless, subject at  pronouns  which  the  i nSt.Fr.  i t applies.  The  To i l l u s t r a t e ,  process o f  as f a ras the level  consider  Inversion.  a sentence  like  i e i  Cela e s t - i l faux? that i s i t false 'Is that false?' derivation ofthis  follows; (=IP), then  t h esubject  sub j e c t -  sentence  NP c e l a  presented  i n Kayne  i s moved t o t h e l e f t  inserted  i nsubject  at S—structure  pos i t i on.  t o properly  D-structure: [ Q Civie> c e l a  II I N F L  [^P etre  faux  - NP m o v e m e n t o u t o f S s [ N P cela* I I Cs C I M P e_i II C i N F L .  -  i_l_ i n s e r t i on s L'iMp.ce 1 a i II C i iMi=ri_est j 3  Frenchs  t o INFL,  govern  t h e empty  IIII  CQ  IIII  CiiMFi_e_j3  CtMP>ili3  Ci  N F L S J  1  two operations  one applying  h4  o f subject  t otheright  faux]]  Cvpe  ,  3  C^pe  c l i t i c i z a t i o n (= S ~ s t r u c t u r e ) s C i N i = - i _ e s t j i 1 3 Ca II (=.e 3 II xiMF-i_e_j 3  CiMi=.celai 3  There a r e thus  pronoun  e s t II L'vpa e f a u x IIH  I N F L movement o u t o f Ss CNi=.ce 1 a i II C i iMFL-Sst j 3 Let C ^ p e i l l  -• L e f t w a r d  ofS  %  - V e r b movement t o INFLs Co L ' N P c e l a 3 II i N F L . e s t II [ V P e f a u x  -  outside  p o s i t i o n and c l i t i c i z e s  i n order  i s as  ( 1 9 8 4 )  t h e I N F L n o d e a l s o m o v e s o u t o f S, t h e e x p l e t i v e  crucially,  (49)  thec l i t i c i z a t i o n  i snot u n i f i e d  one i n ( 4 8 ) i n v o l v i n g Complex  (48)  is  i n Kayne's approach,  i  faux33  C p.e f a u x 3 3 V  cliticization i n  i n PF, t h e o t h e r  t othe left  at  S-structure. Kayne  ( 1 9 8 4 5 £'£1, £30)  conceivable clitics  that  (198£) a n d that  that  the  The  & Cord i n  to  i s no  the  want  in subject  a l l subject  have the  in  St.Fr.,  in  Kayne's a n a l y s i s , can  INFL" (33)  type above.  Recall  accounted  of  f o r by be  that  we  to  the  NP I pro  /  AGR I Cl ±  /  V I Clj+V  a n a l y s i s not  process  of  account  of  (subject)  be  This subject  appears  properties of  to  clitics. subject  clitics  handled  "clitic  easily  the  a  lines  of  \ NPj I pro  the  Inversion  D-structure  VP \ TNS  an  empty  as  INFL' \  INFL /  bound  in  is  pro,  cliticisat  \  t  an  is  phonological  IP /  this  rightward  (50) . (50)  such  the  Such  Complex  adopted  behind  improperly  can  theory  a n a l y s i s along  (I r e t u r n  i n Chomsky  cross-linguistically.  single rule.  argue here that  subject  suggested  reasoning  an  reducing  more r e s t r i c t i v e to  of  clitics  i t is  position in Italian  i s why  advantage of a  The  that  exhibit  i n INFL as  (1981).  here  fact  dialects  occurrence  question  c l i t i c i s a t ion to  I  generated  empty c a t e g o r y  category.  provide  base  Brandi  there  would  certain Italian  are  therefore  extended  acknowledges the  by one  a  presented  ion  in in  £.4.)  representation  as  in  96 What  i s important  constitutes  a separate  Consider the  t o notice here  subject  morpheme a t l e a s t  PF.  I t i s thus  of  (44a) a s opposed  the  natural  to this  until  + V3  v  structure.  ( 4 4 ) .Under ( 5 0 ) ,  cliticization  t o expect  clitic  C Clj  c o n s t i t u t e s a somewhat  t o (44b) s i n c e  i s an object  the string  i n <43) a n d  i n INFL s t i l l  in  derivation  according  now t h e s e n t e n c e s clitic  independent  entity  i s that  the relative  a t no p o i n t  morphologically  takes  place  acceptability  i nthe independent  of  verb. Consider  next  the contrasts  the subject  to  t h e s t r u c t u r e i n (50) and t h e a n a l y s i s o f c l i t i c i z a t i o n ,  and  clitic  which  plausible be  simply  an i n f l e c t e d  (45b),  after I  for "the  cannot  verb  clearly  respect  C Clj v  now t u r n  + V3.  t o object  a subject  clitic  I return to this  possibility  i n (46) and  i s true  the  relative  and  lexical  clitic  cluster, then  and t h e s u b j e c t both  there  ordering NP s u b j e c t  i s absent  being  clitics  from  clitic  of subject  would  i n £.4.  ( 4 7 ) . L e t us assume Rizzi  (1984.6).  a r e members o f t h e  t o expect  clitics  other  appear  c o n s t i t u e n t s o f INFL".*'  i s no r e a s o n  3  If  a difference in  and ne, on one hand,  and ne, on t h e o t h e r  a s i n (46a,b)  could  a  clitic  i s a d e s i r a b l e p r e d i c t i o n . The o n l y  t o the sentences  clitic  According  be e i t h e r b e t w e e n a n o b j e c t  t h e moment t h e f o l l o w i n g s t a t e m e n t negative  clitics.  (45c) o r between two o b j e c t  p o s i t i o n i n which  same c l i t i c this  with  the position  of  subject  clitic  i n (45) r e g a r d i n g  the negative  hand, marker  since would  when a still  follow  the subject.  negative  marker  languages with and  negation,  various I  Rizzi  would  points out that  lead  respect  one t o e x p e c t  to the particular  As he i l l u s t r a t e s ,  N.It.  that,  given  (50) , t h e p r o p e r t i e s o f F r e n c h  believe in  ordering  among of  clitics  prediction i s verified  i  clitics  subject  clitics  i n French  t h e one i n  presented i s then  a b o v e c a n be  no r e a s o n t o  a r e not base  generated  INFL. The  second  theory  t o be p r e s e n t e d  (1984) where he p r o p o s e s t h a t refer  t o two very  favor  of this  and  variation  a structure like  f o r i n a n a t u r a l way. T h e r e  that  analysis of the  dialects.  thus conclude  accounted  this  this  cases  crucial  ( 5 1 ) a.  b.  i s suggested  "subject  s y n t a c t i c elements. a comparison  i nR i z  clitic"  can  Arguments i n  between  St.Fr.  dialects. data  of subject  dialects.  the notion  p o s i t i o n come f r o m  some N . I t . The  different  here  involved  doubling  A Trentino  in Rizzi's  (cf.  example  (1) a b o v e )  i s given  in  presentation i n these  concern  Italian  (51).  E l G i a n n i e l magna. DET he e a t - 3 s 'Gianni eats.' E l magna.  c. * M a g n a . According  to Rizzi,  intermediate  case  these  dialects  represent  an apparent  between c l e a r examples o f non n u l l  subject  98 languages  like  English  languages  like  Standard  convincing and  against  evidence  and c l e a r c a s e s o f n u l l Italian  and S p a n i s h .  f o rthe structural  t h e ones  i n (53) and  (54).  subject  He  provides  representation 1 9  I will  i n (5£)  not review h i s  argumentat i o n here.  <5£)  IP / El  NP I Gianni  (53)  \ / INFL I el  VP I magna  TOP' / El  \  TOP I Gianni  IP / \ NP INFL' I / \ e l INFL VP I magna  (54)  IP / El  In  INFL' \  NP I Gianni  \ /  INFL' \  INFL  VP I e l magna  (5£) w h e n t h e s u b j e c t  this  position,  here  —  ways.  where  by t h e r i c h  p o s i t i o n i s empty,  i t i s licensed — PiGR w h i c h  1) b y t h e s u b j e c t  pro i s found i n  i n t h e sense  adopted  i s phonetically realised  clitic;  i n two  a n d , £) b y t h e v e r b a l  morphology. What  i s more i m p o r t a n t f o r o u r d i s c u s s i o n  i s the assumption  that  t h e two s e n t e n c e s  i n (55) have d i f f e r e n t  representations  i n Rizzi's  ( 5 5 ) a.  French  Standard  syntactic  approach.  I I mange. 'He e a t s . ' b.  T r e n t i no E l magna. ' He e a t s . '  Let  us r e v i e w  subject (56)  the reasoning  doubling  Standard  i n Standard  i l mange.  b.  Jean  mange.  c.  I I mange. there  this.  French,  First,  compare  i s a pause  between Jean  ( 5 6 a ) i s ungrarnmat i c a l .  speakers  (56a) a c c e p t a b l e w i t h o u t  of  find  c o n s t r u c t i o n b e c o m e s ungranimat i c a l  subject. ( 5 7 ) a.  This  i s no  (56a) t o (51a).  and i 1 . i . e . i f i t i s a  l e f t-d i s l o c a t ion, who  there  French  a. * J e a n  Unless  behind  i s not t h e case  Standard  French  * P e r s o n n e i .1 n e nobody h e NEG 'Nobody e a t s . '  mange, eat-3s  Furthermore, a pause,  this  w i t h a quant i f i e r  i n Trentino.  even  f  type as  100 b.  Trentino Q u a l c h e p u t e l 1' some boys they 'Some b o y s c a m e . '  In  Rizzi's  one  approach, t h i s  i n (53) w o u l d  Brandi  & Cord in  Secondly, in  their  in  St.Fr.  (58) Is £s 3srn 3sf IP £p 3pm 3pf  best  (1381)  0 te el la 0 0 i le  Florentine  paradigm  the  one  St.Fr. c. Is £s 3sm 3sf IP £p 3 pm 3pf  in  (1984)  clitics  (58b).  v e n go v i en i v i ene v i ene v i ene venite v e n gone v e n gone  cf.  the  also  for similar  but t h i s  like  views.  i s not  have  gaps  the case  i n (58).  Standard French  not have  i n (58c) w i t h  F i e r e n t i no e' tu e' la si vu' e' le  for St.Fr.;  je tu i1 el le nous vous i 1s el les do  a structure  number o f t h e N . I t . d i a l e c t s  b.  vegno vegni ven ven v e g n i rn vegni ven ven  mean t h a t  Kayne  Compare t h e p a r a d i g m s Trentino  vegnu. come  suited  and  a certain  N.It. dialects  (58)  would  paradigm of subject  a.  Other  be  e have  viens v i ens v ient vient venons venez v i ennent viennent such  gaps.  Compare  the Trentino  one  the i n (58a)  and  101 Rizzi  (1984s9)  proposes t o account  between F i o r e n t i n o and involving  the p o s s i b i l i t y  "obligatory,  optional or  specifications and  always  does not  since  him.  two  subject  Therefore,  hold  this  consequence Third, (59)  t o be  number  are  consider  Italian". still  i n IIMFL i s  [Italian] But  i n (58b,c)  this to  account  in St.Fr.  Rizzi. cannot  according  are explained  t h e y seem  distinction  dialects,  according  Rizzi's  i n CNP,IP]  though  arbitrary  of Rizzi's  parameter  different  i n the  e x p l a i n why  the paradigms  an  for  for St.Fr.,  ways even  a  s p e l l i n g - o u t PiGR  i n Standard  clitics  quite different  believe  by a s s u m i n g  impossible  s t r u c t u r e i n (53) w o u l d  apply to  that  o f p e r s o n and  impossible  explanation The  Trentino  f o r the difference  identical.  and  an  in I  undesirable  approach.  the contrasts  i n (59) and  (60).  Standard French a.  Elle  chante et  danse.  'She  sings  dances.'  b. ? ? E l l e (60)  and  chante et e l l e  danse.  Trentino a. * L a c a n t a e b a l l a . 'She s i n g s and d a n c e s . ' b. La canta e l a b a l l a .  Rizzi  suggests that  a  zero  pro-form  i s not  allowed  coordination.  Therefore,  i f t h e s t r u c t u r e i n (5£)  for  then  i s excluded.  Trentino,  (60a)  On  the other  i n PiGR  under  i s assumed hand, i f  (53)  i s the correct  be g r a m m a t i c a l second  Fr.  since  conjunct.  Finally,  representation there  cf.  Safir  (61)a.  b.  (1985)  (6£)  pro—form  (59b) r e m a i n s  1) t h e v e r b  a n d , £) t h e i n v e r t e d for a  must  subject  (59a)  i n AGR  will  of the  unaccounted f o r . i n (61) f o r S t .  be u n a c c u s a t i v e  must  be  or  indefinite;  discussion.  II est arrive une/*la there i s a r r i v e d a/the 'There a r r i v e d a g i r l . '  fille. girl  II a ete arrete un/*l' etudiant. t h e r e has been a r r e s t e d a/the student ' T h e r e was a r r e s t e d a s t u d e n t . '  A s i m i l a r type involve  of  then  of inversion exemplified  obeys two c o n s t r a i n t s :  passivized;  i s no z e r o  The o d d i t y  the type  f o r French,  these  of construction  i n t h e N . I t . d i a l e c t s does not  constraints.  F i o r e n t i no Gl' e venuto l e ragazze. t h e r e h a s come the girl ' T h e r e came t h e g i r l . ' A c c o r d i ng t o R i z z i ,  fact  that  inversion  French i s more  postverbal is for  NP  restricted receive  clitic  or passivized  i s t o be r e 1 a t ed t o t h e  subject  in this  l a n g u a g e and language.  nominative  i_l_which  t h e NP t o b e C a s e - m a r k e d  unaccusative  contrast  i s not a n u l l  cannot  the expletive  this  gets  Case  In short,  from  which  i t . The o n l y  assigns  the  INFL s i n c e i t  i s i f i t i s the object  verb  that  other  way  o f an  partitive  Case.  103 Partitive  Case?,  Belletti is  (1985).  empty making  nominative the  verb  In French  nature,  In the  Italian  i s thus  the type  o f NP  no  in Rissi's  are  generated  base  are  they  appear  i n INFL  null  subject languages  follow  these  assumptions,  from want  t o argue  given the and  here  some d a t a below.  assume t h e  structure  languages  coordination  ( 6 3 ) a.  from  in  i n an  in  in  French  the nature be  involved  of in  —  i n St.  whereas  St.Fr.  i n the  The  reasons.  cone1 u s i o n s shouId  (6), the  discussion  dialects  equivalent to I will  that  seem  four  dialects  structural  will  be in be  preferable  (33)  —  4 and  the  £.4. and  optional  Pied  way.  dialects.  J e a n / 1 ' hornme v i e n t . ' J e a n / t h e man comes.'  Noir French  The  sentences  exhibit in  (63)  to  fora l l  discuss the  i n Chapter  N.  N.It.  i s not.  Italian  i t would  in greater detail  French  i n both  the  Ri ss i ' s  two  (5£)  but  i . e . from  In o t h e r words,  facts  facts  grammatical  that  position  and  with subject c l i t i c s .  Colloquial doubling  can  position receive  i n D - s t r u c t u r e . The  generalisation  introduced  inversion  on  to  subject c l i t i c s  i n CNP,IP3  between S t . F r .  the  constraint  approach,  above  £.£.3,  subject  p o s t v e r b a l NP  subject that  differences  revised  the  cf.  c o n s t r u c t i ons.  dialects  I  indefiniteness;  dialects,  f o r the  There  summary,  dialects  requires  i t possible  Case.  o r on  i n v e r s i on  It.  by  subject are  104 b.  II vient. ' He c o m e s . '  c.  J e a n / 1' hornrne i l v i e n t .  The o n l y subject  d i f f e r e n c e between t h e s e two d i a l e c t s doubling  i s the existence  a c o n s t r a i n t on t h e n a t u r e o f t h e d o u b l e d  NP  cannot  indefinite  hold  f o r PN F r .  (64)  C o 1 1 o q u i a1  regard  to  i n C . F r . b u t n o t i n PN F r .  of  be  with  NP.  In C.Fr.,  n o r c a n i t be q u a n t i f i e d . T h i s  the  does not  French  a. * Un hornrne i l v i e n t . a man he comes b. * C h a q u e each (65)  Pied  Noir  woman s h e  comes  French  Un hornrne i l v i e n t .  b.  Chaque  doubling  i s indeed  like  null  subject  languages then  To  illustrate  types of dialects  subject that  an i n t e r m e d i a t e  (St. Italian)  3.  t h e N.  case  and  i t i s r e a s o n a b l e t o assume  a n d PN F r . a r e i n t e r m e d i a t e  different  f o r evidence that  i n some s e n s e ,  St. French  N.It. dialects.  c o n s t r u c t i o n t o Chapter  a v a i l a b l e i n C.Fr. I f i t i s t r u e  represent,  fully  languages C.Fr.  vient.  i s r e f e r r e d t o 3.1.3.1  dialects  between  fernrne e l l e  detailed analysis of this  The r e a d e r  the  vient.  a.  I defer  It.  fernrne e l l e  cases  between S t . French  this,  l e t us c l a s s i f y  or languages  that  that and  the  are referred to  105 here.  The t e r m  constraint (66)a.  b.  c.  d.  What of  on t h e doubled  Standard + subject - subject  subject  doubling"  NP s h o w n  in  refers to the  (64).  French clitics doubling  C o l l o q u i a l French + subject clitics + optional subject doubling  (restricted)  Pied Noir French + subject clitics + optional subject  (non-restricted)  doubling  Fiorentino + subject clitics + obligatory subject  I want  t o suggest  structural  clitic in  "restricted  assume t h a t  here  i s that  differences with  i s generated,  (66) would  doubling  look  (non-restricted) under an account  respect  t o where t h e s u b j e c t  the cross-1inguistic variation a c c i d e n t a l and a r b i t r a r y .  C.Fr. and S t . F r .  i n terms  have t h e i r  expressed  One w o u l d  subject  have t o  clitics  in  C N P , I P ] w h e r e a s PN F r . a n d F i o r e n t i n o h a v e t h e s a m e c l i t i c s INFL. (57)  This  could  above,  structure But  n o t be o t h e r w i s e  and he i s f o r c e d  Rizzi's  t o say that  i n (53) f o r s u b j e c t  doubling  constructions.  problems  because one would  explain  why s u b j e c t  doubling  i s possible  St.Fr.,  on t h e one hand,  a n d why s u b j e c t  F i o r e n t i n o and o p t i o n a l If  one d e c i d e s  i n PN F r . ,  to classify  C.Fr.,  discussion of  C.Fr. has t h e  causes  in  this  given  in  f u r t h e r have t o  i n C.Fr. but not i n doubling  on t h e o t h e r  i s obligatory hand.  PN F r . a n d F i o r e n t i n o a s  106 languages subject  with  clitics  difficulty and  but S t . F r .  one would  as one where  run into the of subject  doubling  i n C.Fr.  the subject  t o l i c e n s e p r o i n CNP,IP]  clitics  i fthis  i n St.Fr.  i s possible  n o t be  i n some o f  dialects?  dialects  hand,  an a n a l y s i s i n which  i n (66) a r e a l l a n a l y z e d  representation, variation  clitics  analysis  i.e. c l i t i c  more n a t u r a l  motivated  the  character  why w o u l d  On t h e o t h e r  the  i n INFL  a r e i n CNP,IP],  i t s restricted  able  clitics  of explaining the optionality  Finally,  its  subject  involved  principles considers  same s p e c i f i c  with  as having  i n INFL,  given  t h e languages and t h e same  structural  makes t h e d i a l e c t a l  differences i n the properties of  respect  t o other  independently  o f UG.  In other  words,  t h e term  "subject  clitic"  grammatical  element  this  type of  as r e f e r r i n g  to  i n a l l t h e languages  exhibiting i t . Now,  with  discussion to  relate  clitics  various  clitics  allow  of this  object  fact  that  doubling.  property  of obligatory object  some  be  able  subject  properties of constructions with languages  object with  Interestingly, the  i s equivalent  one i n ( 6 6 ) ,as i l l u s t r a t e d  examples  o u t l i n e d would  properties of constructions with  I t i s a well-known  distribution the  t o t h e g e n e r a l i z a t i o n i n (6) a n d t h e  i n £.£.3, t h e a n a l y s i s j u s t  to similar  clitics. object  respect  i n many r e s p e c t s  i n ( 6 7 ) . I do n o t have any doubling.  to  107  (67)a. + -  It  Standard/Colloquial object clitics object doubling  R i v e r P l a t e S p a n i s h Pi, P i e d N o i r French + object clitics + optional object doubling (restricted)  c.  River P l a t e Spanish B + object clitics + optional object doubling thus and  seem t o  subject  different  having  the  does not  different  that  I role  now  assignment  ideal  to  to  detail  my  pro  i n argument subject  and  (58c)  be  they  position. proposed  analyzed  concerning  in the  of s p e l l i n g - o u t  F i o r e n t i n o paradigm  French  are  element  clitics  what i t  paradigm.  assumptions concerning  in c l i t i c  maintain  the  return to  relationship  assume t h a t  impossibility  St.  object  0—Role  subject/object  useful  or  relate  same g r a m m a t i c a l  Whatever makes t h e  C a s e and will  the  (1984) s u g g e s t i o n  optionality  holds.  in fact,  (58b)  a l s o makes p o s s i b l e t h e  £.£»5.  the  of  t r y to  a n a l y s i s of  Rizzi's  obligatoriness,  and,  licensing  unified  require  ways.  PIGR s t i l l  clitics  realisations  the  (non-restricted)  make s e n s e t o  f u n c t i o n of  Finally, here  8 0  a i  clitics  is  Italian  b»  would  two  French,  constructions.  same l i n e  symmetry our  between a  of  reasoning  developed  basic clitic  To  do  C a s e and so,  with  i n £.£.3.  i t would respect  It  is  assumptions concerning and  the  argument  to  @~ be  to  thus the  which  it is  108 related. We  have seen  and  across  can  appear.  role in  i n £.2.1  languages It  the  i n the  i s thus  i s assigned  to  Government  and  on  simplified  definition  licensing  Binding The  status  that  (Pi o r  position  in  an  fi-position,  (SCI  and  NP)  excluded clitics  by  asking  verb  assign the  for clitics  here,  this  whether or  leaves  the  what  are  the  i s one  of  them;  a  coindexed  by  the  select  two  the  holds  see  9case  (1984) has  no  I f i t were  in  argument p o s i t i o n s to  both.  di Sciullo  This  of  s i n c e the marker  is  (198£) f o r whom  in Pl-position.  problem  agreement  tomorrow.  clitic  9~cr i t e r ion.  but  a  for  Borer's  to the  i s a v a c u o u s one.  remark  not  Now,  are  same 9 - r o l e  9~criterion;  assume t h a t  i s u s u a l l y the  I adopt  occupied  would  (1983) a r g u e s t h a t  John  reason,  respect  pronoun  o n l y one 9 - r o l e , t o one and o n l y  Pi'-binders f o r a v a r i a b l e  Pi'-status  outlined  and the  are  Sport iche  the  this  general  lexical  to  as  in  (68).  argument  position  Pi' ) w i t h  licensed  framework.  i s given  For  the  be  9-criterion  i t s related  purposes.  suggestion  (69)  argument  9-theory?  and  can  p o s i t i o n s where a  Chomsky (1981.36) E a c h a r g u m e n t b e a r s one and and e a c h 9 - r o l e i s a s s i g n e d one argument. Pi c l i t i c  or  pro  natural to continue  an  conditions  (68)  that  In  determining In the problem ^s_ i n s  fact, an Pl-  approach amounts  to  109 is  i n an  type  A-  of  o r A ' - p o s i t ion.  question that  remains  considering  the examples  the  following  it  moment,  absorbs  must  that  assignment,  as  assumed  dp  proi  CIMP.  subject  pro  1  position  assumption  that  in  3 w h e n we  Chapter  Finally, for  both  like  subject  plausible  terms  On It  in  in  the  same  IIih4i=i_  i s not  this  clitics  and  in  (66)  i n Chapter  and  (67)  If this  vient  a s s i g n e d Cases by  i s so  for a  then  Case  will  be  The qualified  doubling.  analysis object  o f Case and clitics  this  above.  IIIIII  c f . £.£.1.  a clitic  clitic  3 together with  holding  (1985), that  i n the domain o f  i l l II II v p  i s absorbed discuss  Jaeggli  o f ) Case.  Before  (1986), s i n c e i n :  (50). Moreover,  of a chain  a r g ument  £.3.  adopts  s i m p l y assume,  in a  analysis  candidate for parametric variation  illustrated explored  Case  in Riszi  I will  and  i s licensed  d.  note that  t h e one  (1984)  (or i s a s p e l l - o u t  be  the  also  concerns Case assignment.  of doubling,  Borer  not  (70)  (1985)  approach.  The  clitic  Jaeggli  9-role  configuration  seems  like  of the  type  This possibility the exact  between a c l i t i c  and  applies  a  is  relationship a  in  related  pos i t i on.  the  Non—Existence of  i s now  t i m e t o put  the  Null  the results  a more g e n e r a l p e r s p e c t i v e .  Put  Subject  Parameter  of the preceding  sections  i n very general terms,  the  110 goal  of linguistic  theory,  as conceived  o f here,  i s t o account  for  variation  and l e a m a b i 1 i t y .  try  to attain  a maximum o f g e n e r a l i t y i n t h e p o s t u l a t i o n o f  principles in  a n d p a r a m e t e r s , ft g o o d  1„£.1 w h e r e I r e v i e w e d  null  subject  phenomenon.  We h a v e s e e n t h a t  on p a r t i c u l a r ,  of  languages,  different  merged  e t c . I n Chomsky  a  provokes a chain  sense,  parameter. this But  (1981),  parameter  these  isolated, property,  these  presented  studies  that-t  effects,  characteristics  the result  then,  area  they  argued  that  Portuguese,  linguists  t h e new p e r s p e c t i v e  started to realise the null  believed given  be d i r e c t l y  that  subject  learner, i n  o f which i s  subject  seemed t o be  at f i r s t  looked  achieved  a t more and  developed  the cluster  Safir  i n Chomsky of properties  was n o t a s  (1985),  immutable  i n particular,  o f t h e N . I t . d i a l e c t s and  property  r e l a t e d t o each  inversion without,  follow  way o f a s i n g l e  parameter  or desired.  the existence  the null  The  of research.  a s more and more  with  theory  were  parameter.  by t h e l a n g u a g e  i n a given  of linguistic  have  characteristics  o f a unique  o f events,  the setting  The g o a l  was f i r s t  free  from  particular  associated  not  of this  languages with  (1981),  as  was  a c e r t a i n number o f t h e p r o p e r t i e s o f t h e l a n g u a g e  automatically  more  subject  and a s s u m e d t o be t h e r e s u l t setting  in  seemingly  null  positive  that  example o f t h i s  desirable to  t h e e v o l u t i o n o f t h e s t u d i e s on t h e  concentrated  inversion,  I t seems t h e r e f o r e  other.  and f r e e Some  sight, null  inversion  could  languages e x h i b i t subjects  and v i c e  versa.  It thus  analysed  seemed a s i f t h e s e  as t h e r e s u l t  In Chomsky  (1982),  empty c a t e g o r y the  null  cf.  1(£7)  subjects verbal in  —  w a s now  longer  would  t o an o v e r t  indicating  tied  UG  pronominal  as formulated necessary.  o f an element,  i n Chomsky  The e x i s t e n c e o f  subject-verb  agreement,  of this  e v o l u t i o n process  of properties associated with  the null  T h e r e seems t o be  explicitly  formulated  "null  parameter"  The q u e s t i o n  Most  theory  property less  general  much m o r e d e e p l y  t o be l e a r n e d  of having  a u t o m a t i c a l l y from  null  of having  f o r example.  this  theory  no language  i s a desirable  f o r the  following  of pro renders  rooted  subjects  the theory  empty s u b j e c t s In t h i s  for the  the  i n UG5 t h e r e i s  i n terms of parameter i n a given  settings.  language  o f empty c a t e g o r i e s , i . e .  of recoverabi1ity. I t i s very  possibility English,  i s whether  importantly, the a v a i l a b i l i t y  The p o s s i b i l i t y  a  t o be a s k e d  The  way.  The a n s w e r s e e m s t o be p o s i t i v e  consequently  follows  subject  to set i n a particular  subject  a s i t was  subject  t o a minimum.  null  null  i s the following.  i s reduced  reasons.  —  to the richness of the  phenomenon  result.  (1981)  (1978).  The r e s u l t  learner  be b e t t e r  parameters.  the introduction into  more c l o s e l y  inflection  cluster  with  parameter,  was no  Taraldsen  o f two d i s t i n c t  corresponding  subject  two phenomena  similar  to the  i n non-tensed  particular  of recoverabi1ity nothing  sentences  instance of a n e e d s t o be  more  in  stipulateds Binding  parameter  can  be  same  The  done t h r o u g h  i n the  (71)a.  of  hand,  pros  follows  posit ion that  of  from i s not  Projection  infinitival  clauses.  only  B i n d i n g t h e o r y makes  i n governed learner  in particular positive  language  a child  simple sentences,  learner's  positions  No  as  stuctural  and  On  one  i s exposed (71)  only  hand,  or  this  verbal  directly  t o and,  are,  i f i t  whether  c o n t e x t s and  i n £. £. £ a r e  in  i t generally  i s to detrmine  evidence.  the ones presented  on  the  presumably  also  disposal.  Italian Ho t r o v a t o MI) found  b.  that  assumes t h e Extended  task of the  i s licensed  available  the  i f one  i t appears  like  at  t h e on 1 y argurnent  i s true  paradigms  other  and  n e e d s t o be s e t .  licensed. pro  ungovemed  i s the subject position  available,  not  be  i n E-Inglish,  Principle,  is  must  t h e o r y and  governed  The  PRO  i l libro. t h e book. '  Spanish Estamos cansadisimos. '(We) are very t i r e d . '  c.  Portuguese D i s s e que t e n h a l i d o e s s e l i v r o . '(She) s a y s t h a t (she) has r e a d t h a t  Furthermore, Brandi  & Cord i n  (1984;1985),  according to studies (1981),  Rizzi  Chomsky  (1984),  book.'  i n Bouchard  (198£),  Everett  S p o r t i c h e (1983),  (198£), (1985),  Zubizarreta  Jaeggli  (198£b), and  object)  that to  and  this  clitic  null  makes  verb  languages as Hale  It  "null  proper  out  that  (subject  l i c e n s e r s f o r pro  l a n g u a g e s and null  a  clitics  now  and  and  languages",  be i.e.  clitics  between  language with  seems p r e f e r a b l e t o  argument  or  be  establish a  subject  agreement a  can  to  i s recovered,  possible to  language with  regard  considered  variation  argument  i s t h u s now  subject-verb  I t now  turns  languages are  the  say,  rich  object  agreement.  more  i n which  between,  be  i t now  predictions, e s p e c i a l l y with  object  is licensed.  language with  also  So,  null  way  language with  thesis,  interesting  or  relationship a  can  doubling.  i n the  pro  present  clitics  subject  found how  the  rich  r e f e r to  and  a object—  these  f o l l o w i n g McCIoskey  &  (1984). The  consequences of  this  shift  and  more  undoubtedly  be  numerous,  raised.  of  them c o n c e r n s  One  the  in linguistic important  questions  distinction  between  m a r k e r s and  clitic  pronouns;  part  of  this  Notice  also that  such  a  issue.  shift  i n the  property  conception  under  empty arguments English  for  English  making  study but  here. why  instance. no  answer  i s probably  device  f o r pro.  use  an  4  i s devoted  perspective  provokes  One  not  must  ask,  property words,  i s not why  English  generally,  does not the  why  to a  the  Italian  attested  i s the  empty c a t e g o r y  be  agreement  i t applies to  In o t h e r of  Chapter  will  will  markedness as  this  that  More  of  theory  has  in  grammar  of  a v a i l a b l e i n UG.  h a v e any  Recoverabi1ity  The  licensing Principle  is  114 violated  i n the  In the issues  case  of  this  particular  remainder  of  this  chapter,  concerning  free  The  clitic  doubling  £.4.  Free  Inversion  There are different either  languages  because of  subject  clitics  questions will  was  null  subject  another  appearing Jaeggli argument  of  NP  them, be  i l l u s t r a t ed  i n l.£  languages  crucially, (7£).  that,  some d i s c u s s i o n analysis  raise  of  interesting section,  inversion.  French  because of  at  that  to  the  onset  generative  of  By  government  nominative  the  I  i t s  doubling.  (1981).  prior  3.  In t h i s  i n 3.4  was  lexical  quite  i n s i d e VP.  be  NP  subjects cf.  external a rule  Chomsky  assigned  This  studies the  f r e e manners  INFL t h r o u g h  S-structure,  Case c o u l d  the  literature,  f r e e i n g the by  of  pararnetr i c a l l y r e l a t e d  having  position in a  Chomsky  constructions.  f o r the  variation.  i n the  namely  need  free  related  i n Chapter  because they  empty s u b j e c t s  applying  in  or  that  s a  on  inversion in  pose  discussed  out  p o s i t i o n from  generated,  here  namely  clitic  in postverbal  that  here  to  having  (198E),  subject  parametric of  property,  INF'L-lowering assumed  one  pointed  of  problems they  proposed  similarity  possibility to  the  inversion will  apparent It  types  causatives  i s explored  under study  concerning  treat only  complex  on  various  I concentrate  i n v e r s i o n and  phenomenon  empty c a t e g o r y .  to  operation  a  of  (1981) postverbal  is  <7£>  Case  IP / NPj  INFL'  I  /  PRO  assignment  \  \  INFLi  VP  I e  /  \  VP  NPj  I INFLi+V  Thus,  the  presence of  automatically that  (73)  empty s u b j e c t s  NP  (PRO)  subjects  and  free  allowed  as  in  (73)  and  i n v e r s i o n were  S 3  a.  Ho t r o v a t o ' (I) found  b.  Ha mangiato Giovanni, has e a t e n 'Giovanni ate.'  as  pointed  problems  free  Safir  i n Chao  type  of  (1981),  approach.  e x a c t l y as  (74b,c)  the  are  (74) from  to  Portuguese  This  in Italian,  i n v e r s i o n . Compare  (1985) and  Safir  i l libro. t h e book.'  out  for this  empty s u b j e c t s ,  (74)  lexical  subjects  Italian  But,  for  preverbal  for postverbal  i s p r e c i s e l y how  related.  empty  language a l l o w s  without  (73).  any  Example  Perlmutter  Port u g uese Disse que tenha l i d o esse l i v r o . s a y ~ 3 s t h a t h a v e - 3 s r e a d t h a t book 'She s a y s t h a t s h e h a s r e a d t h a t b o o k . '  causes for  possibility (74a)  (1976),  (1985).  a.  data  is  cited  from in  b.  Deus e x i s t s . 'Bod ex i s t s . '  c. * E x i s t e  The  cases of  the  French  of  inversion  definiteness  ( 7 5 ) a. b.  effect  in  (75)  two this  Nao 'No  i n the sense  from  they  of Safir  Perlmutter  line  to notice  but  properties  no  free  should  to null  be  closer  show some  (1985)  to kind  as  (1976).  of having  t h e o r y and  pointed  out  inversion  i n £.3,  and  accounted  null  null  then  since  independent  parameter  that that,  has  consequently,  i t i s consistent so  with  far. Since  should  be  f i x e d by  reason  the r e s u l t of  inversion should therefore the  be  adopt  the  the  Projection Principle, obvious  as  why a  free  single  the r e s u l t of  language  the  the  i s a l l o w e d f o r by  t h e r e i s no  t o be  Portuguese  for independently.^ I  subjects  subjects  Free  learner  an  on  the a v a i l a b l e evidence.  t h e r e a r e many w a y s  implemented.  fact  i n v e r s i o n and  the Extended  parameter.  of  the  arguments developed  Binding  basis  be  here  of reasoning here  possibility  Now,  i n that  would  existe t a l ilha. such i s l a n d e x i s t s . '  subjects  approach  the  i n Portuguese  E x i s t e n t hornens c a p a z e s d e r n a t a r a t e a s a v e s c a n o r a s . ' T h e r e e x i s t men c a p a b l e o f k i l l i n g e v e n s o n g b i r d s . '  is sufficient  null  found  impersonal construction  illustrated  It  Deus.  i n which  Some o f t h e m h a v e  been  this  idea  proposed  can  i n the  be literature.  My  i n t e n t i o n here  different The  first  proposed to  options  but merely  (1981).  free  According  to this  The Picallo  second (1985),  postverbal position Case.  option,  null  else  being  also  allow  presented  in Rizzi  9-role  (198£) a n d a d o p t e d  r u l e moving  the subject  by  to a  NP a n d p r o i n s u b j e c t  and form  a chain  i s transmitted  which  i s assigned  by p r o t o t h e NP5 p r o  expletive. IP /  NP I pro* *  >  Case  assignment  \ /  INFL' \  INFL  VP  1  /  \  VP I V  NP*  a movement a n a l y s i s m a k e s some p r e d i c t i o n s w i t h  languages with an  would  null  i s a wrong p r e d i c t i o n .  are cosuperscripted  (76)  Such  i t fails  i n v e r s i o n and  subjects  p o s i t i o n . The p o s t v e r b a l  therefore  free  seen that  analysis, everything  a s s u m e s a movement  The e x t e r n a l  of the  different alternatives.  have a l r e a d y  between  any language a l l o w i n g  i n v e r s i o n . But t h i s  review  a b o v e a n d i n l.£. 1 a n d i s  We  make a c l e a r d i s t i n c t i o n  equal,  t o explore  one i s d e s c r i b e d  i n Chomsky  subjects.  is  i s n o t t o make a n e x t e n s i v e  subject  clitics  and f r e e  analysis of subject  clitics  that  then  we  predict  that  with  the postverbal  a preverbal subject,  base  subject  inversion.  respect  I f we  g e n e r a t e s them clitic  and, f u r t h e r , t h a t  would  to  adopt i n INFL  agree  the verb  would  118 show an  ending  subject.  This  this  can  c f . £. 5. and  be  fit  (78),  (77)  the  person we  and  will  number o f  see  later  the  in  least from  two  X'  to  assume t h a t  theory  the  r i g h t of  (1984s190).  an  external  (or a l t e r n a t i v e l y  alternative structures  Travis  VP  dominated  a r i s e , as  by  in  s s  V I V  in a  pronominal  an  so  an  associated  i t should  since  clitic  be  then  lost.  should  in  and INFL  made p r e c i s e  position. Given be  i f clitics  argument  Portuguese  clitics  c l e a r that  between c l i t i c s  appearance are  that  a subject sense to  association  excluded  subject  i n argument  far,  be  their  requirements)  prob1ems s i n e e  developed such  through  languages with  associated,  empty  simply  I' / I \ I V N' I V  Here again cause  i s to  generated  I' / I\ N' I  (78)  is  to  p r e d i c t i o n as  possibility  Case adjacency  (77)  be  i s a wrong  third  argument  IPs  corresponds  sect ion. The  by  that  can most  The  and  appear  the  free in  the  with  not  an  approach of  p o s i t i o n must  f r e e l y on  free  wou1d  possibility  argument  problem  either exhibit  (78)  i n 3.3,  generalizations  second  inversion  heads  without  concerning  with  this  inversion,  option which  is  not  the  which  case,  i s not  or  the  case  (1984.£07) c l a i m s , occur  with  being  studied  should  either.  "the  are  S-I-V-0 do  position.  such  Without  unidentified  gap.  however,  identified  gap  to  by  the  gap  Travis without she  free  i.e.  to  have  subjects, O-V  reason  that  pro—drop  seems t o  i s that  most  i s the  pro-drop  co-  languages  governed  pro-drop  would  l a n g u a g e has possibility  p o s i t i o n and  Travis  s-I-V-Q l a n g u a g e s  properly  position,  there  inversion i t to  free  be  Irish but,  inversion to  there as  an  of  with  subject leave  an  I-V-O-S  having  features  can  an be  supplied  to  MeCloskey  to  have an  language  & Hale  t h u s does not  Italian  pointed  Therefore, exist  Irish.  subject  need  (1984), to  identified  gap,  subjects.  contrary  should  Irish  in order  as  is a null  following  have n u l l  in  which  I-S-O-V.  inversion).  variant  as  S-I-V-0  I-V-O-S v a r i a n t  free  language,  because,  l a n g u a g e s and  in subject  subject  i s so  have a  a  a null  This  I f an  Portuguese,  h a v e an  be  INFL".  in order Now  not  does examine  assumes  appeal  only  subject-inversion  subject-inversion  variant,  not  out  means,  would  Spanish,  before  in order  other  This  or  need  does  ( i . e . i t does  for  i t to  allow  f o r example, to  not  be  argued  not  null  an  I-S-  for  independently. The  last  insight, right  one.  of  option which This  a  i s that  of  variant  i s adopted  insight  Safir  i s that  the  (1985). here,  free  Safir s 5  basic  i s , I believe,  inversion  the  possibility  ISO is  the result  there like  exist  o f an independent  languages with  Portuguese.  approach The  propose here,  subjects  F o r arguments against  of the free  following  communication),  +TNS e l e m e n t  Chomsky  an idea  assume t h a t  o f INFL,  Suppose  a &  inversion  i sa synthesis  us f i r s t  (1981).  null  (1981),  that  free  to the possibility  i s , presumably  that  i n syntax  Chomsky's r u l e  by a r u l e  possible  R. T h i s  whereas  structure,  free  u n d e r VP. T h i s According  inversion  position  2 2 - 7  there  i sassigned  i stied,  the nominative i sassigned or  similar to  inversion i s  m o v e p r i o r t o S-  since  nominative base  say, i n Portuguese  Case  generated  NP  and French.  of the definiteness i n an  as a t r i g g e r experience  i sfree  as i n  possibility: i n  free  i tcannot  analysis  by t h e  i n Chomsky  a d e f i n i t e NP c a n a p p e a r  c a n be c o n s i d e r e d  deduce that  language.  that  i n syntax,  above.  T h e +TNS n o d e i s  t o a postverbal  (1985)  (personal  o r lowering  i sexcluded  i st h e case,  the fact  where Case  S-structure.  where  directly  to Safir's  o f having  i sa parameterized  i n those  be a s s i g n e d  Case  inversion  o f hopping  l a n g u a g e s w h e r e TNS may move  to  inversion,  I want t o  theories  n o t b y PiBR a s a s s u m e d  further  that  Rochemont  nominative  checked,  effect,  that  thedetails of Safir's  o f t h e four  i n VP a t t h e l e v e l  cannot  but no f r e e  property  b y M.  Case a s s i g n e r  moved  and p r e d i c t s  s e e 3. I . E . account  Let  parameter  inversion  inverted  sufficient  i na particular  The  S-structure r e p r e s e n t a t i o n o f a sentence  inversion  i s thus  (79)  •>  Case  assignment  \  NP  INFL' \  /  e  free  a s i n (79) „  IP /  with  INFL  VP /  \  VP  PIGR  NP  TNS+V  Three questions a r i s e  with regard  to the representationi n  (79) s Pi.  HOW i s the external does t h e verb  C.  i s t h e s t a t u s o f t h e empty c a t e g o r y  What  t o answer these  compare two s p e c i f i c already  been  which  subject.  namely t h e type difference to  free  agree  (1981). the  preverbal  of free  found  position  e  type  to  One h a s  of free  (translated ( 8 0 ) .  inversion  dialects,  major  the verb  with  respect  does not  i n B r a n d i &• C o r d i n  when f r e e  as English M  The  and S t . I t .  i n t h e former,  f o r example,  as i n  inversion.  subject c l i t i c s .  t h e post v e r b a l N P a s shown  clitic  be n e c e s s a r y  i n t h e N. I t .  between t h e N . I t . d i a l e c t s i s that,  i n CNP, IP3?  apparently, with the postverbal  i s the type  In Fiorentino,  neuter  i twill  i t i s the St.It.  involving  inversion  with  questions,  agrees,  The o t h e r  with?  realizations  presented;  t h e verb  agree  assigned?  B. What  In o r d e r  in  ©-role  inversion  i_t_) a p p e a r s  occurs in a  (SO)  E' v i e n e le ragazze. i t corne-35 t h e g i r l s 'The g i r l s come.' t  Compare t h e i n f l e c t i o n where t h e s u b j e c t personal (81)  clitic  on t h e v e r b  i s i n preverbal  position  i n (SI)  and accompanied  by a  pronoun.  Le r a g a z z e l_e vengano. the g i r l s t h e y come—3p 'The g i r l s c o m e . "  ft f r e e  inversion construction with  clitic  pronoun  subject  therefore,  postverbal ( 8 3 ) a.  there  corresponding  inversion  agrees with  coreferential  the postverbal  a s s h o w n i n (8£).  i s a gap i n t h e p a r a d i g m t o the neuter  find  any p r e v e r b a l  c o n s t r u c t i o n but t h e verb subject  as  (83)  E vegnu  'na  'na  putela.  putela.  clitic; clitic  of  subject  c f . £.£'.4. We  i n the Trentino  does not agree  illustrates.  ' Na p u t e l a 1_' e i vegnuda. a girl s h e h a s come ' fl g i r 1 c a m e . '  b. * L' e i v e g n u d a c.  the verb  a preverbal  vengono l e ragazze. come—3p t h e g i r l s g i r l s come.'  In T r e n t i n o clitics  i n which  i s ungrarnmat i c a l  (8£) * L e they 'The  not,  i n (SO) t o t h a t  do free  with the  The  same  i s true,  extraposition  in Trentino,  f o r impersonal  c o n s t r u c t i o n s (85),  (84)  (*E1_) p a r c h e e l M a r i o '(It) seems t h a t M a r i o  (85)  (*E1_) s a r a m e i o ' n a r v i a . '(It) w i l l be b e t t e r t o go  ( 8 6 ) a.  E sta t r o v a 'na has been found a ' Pi b a g was f o u n d . '  b. * L ' e i s t a d a These  facts  previously. (79)  found  suggest  Let in  trovada  us  (87)  el_ s i a left.'  impersonal  passives  (86)  parti.  away.'  ' na  borsa.  some a n s w e r s t o t h e  f o r the  (84),  borsa. bag  assume t h e  (87)  and  verbs  more d e t a i l e d  questions  raised  r e p r e s e n t a t i o n of  F i o r e n t i n o sentence  in  (80).  IP / NP I proi  \ /  INFL' \  INFL. VP I / \ PiGR VP NP I I I e TNS+V le ragazze 4  In  answer t o  (C) , a  external  argument  clitic  in  To saying  e  position  candidate  i s pro  licensed  f o r the by  the  empty  subject  INFL.  answer q u e s t i o n that  plausible  the  verb  which  i s 3s  since  INFL.  This c l i t i c  ( E ) , we  agrees  can  with  i t i s licensed i s 3s.  maintain  the by  our  assumption  e x t e r n a l argument the  expletive  clitic  by  positio in  The The  process  only  of external  unquestionable  bear  a 9-role  from  the Trentino  instance, argument  since  position,  agreement  (138S)  argument  proposal  variant.  necessitates  introduced  pros  postverbal  NP.  review  respect  assigned  i s equivalent  f o r St. Italians from  NP i n to the  to  Rizzi's  the  i n a chain  with  & Rizzi  (1981s136)  for a  expletive. This chain  c a n be f o r m e d  problems  solution  containing either  p r o and  by  o r by c o s u b s c r i p t i n g t h e t w o members. potential  to  being  of a "concrete"  extensively i n Safir  second  consists  possibility,  i n saying  directly  assignment the  with  the postverbal  normally  and i s t h u s  chain  to the external  inversion construction.  The f i r s t  i t s 9-role  comes  This  f o r E-iinding t h e o r y  (1985) and w h i c h  that  I do n o t  here.  The  9-role  This  c a u s e s some  discussed  9-role  above  forthis  o f the verb  cf. also Belletti  formation  cosuperscripting  are  position.  clear.  i n <84), f o r  a 9-role  two ways i n which  P r o h a s no 9 - r o l e  solution  where  and t h e b e h a v i o r  NP r e c e i v e s  preverbal  i s less  p r o i n CNP, IPII d o e s n o t  i s t h e same a s i n t h e f r e e  postverbal  the  i n (84)-(86)  can receive the external  external  the  i s that  does not a s s i g n  There a r e at least (87)  fact  assignment  i t i s e x p l e t i v e . Evidence  data  t h e verb  9-role  that  in this  suggested  the postverbal position.  o f an e x t e r n a l  9-role  9-cr i t e r i on i s n o t v i o l a t ed,  in Safir  (1385s188),  NP c a n s o m e h o w  Safir's  solution  t o any s i s t e r  receive a  involves  o f VP a s l o n g  i . e . a s 1on g a s t h e r e  as  i s on1y  one  argument  f o r a given  s i m i l a r " t o t h e one  9-role.  i n (87) b o t h  CNP, S3  positions are sisters  9-role  i f any.  (8S)a.  I n t h e s t r u c t u r e s i n (88) the VP-adjoined  o f VP  and  NP  and  can r e c e i v e  the  the external  e 9  S i \ IMP I N F L VP +0 I V /  b. / NP  To  summarize,  Case-marked position  subject  external  made p o s s i b l e moving  The  3 1  in i t sposition.  clitic  argument  i n INFL.  position.  i n the f i r s t  The  Case  place  The  This  a 9-role  preverbal  verb  type  assigner  and i s  subject an  agrees with  pro i n  of construction i s  by t h e p a r a m e t r i z e d  i s s u e t o be d i s c u s s e d  in free  have a l r e a d y Northern  receives  +TNS i n t o  VP  option  prior  to  of  S~  3 0  last  agreement  NP  by a n e x p l e t i v e p r o l i c e n s e d by  the nominative  struct ure.  NP.  the postverbal  directly  i s filled  impersonal its  S I \ I N F L VP / \ VP NP I +9 V  Brandi  inversion constructions  pointed  dialects  here concerns  out that  i n that  & Cordin  St. I t a l i a n  the verb  i n St. Italian. differs  agrees with  (1981) show t h a t  verbal  the  I  from i t s postverbal  languages vary  with  i£6  respect  to feature  material the  i n preverbal  following  (89)  sharing  Brandi  between a p o s t v e r b a l  p o s i t i o n . They  & Cord i n  11 a 1 i a n Eng1ish French F i o r e n t i no T r e n t i no  c o r r e c t l y remark subject  s a m e way.'Now n o t i c e established  i n (89).  h a v e no s u c h  inversions from  property  H-sharing property  inconsistency  attributed  This  since that  property Italian  there  The t h r e e  property  Data  and E n g l i s h  a systematic  with  license  pro i n external  reason.  I f i t i s true  clitics;  a r e not " f r e e " —  property  that  i n the feature the proposal  free  i n (89) me  language.  clitics  subject  block  property. clitics  position f o rthe following i n v e r s i o n i s a base  —  nevertheless  sharing  that  with the  c a n be  in a  subject  t h e two  prevent  generalization. I can  argument  that  do n o t have  o f i n v e r s i o n found  i t c a n be s u g g e s t e d  i s consistent  behave i n  together  clitics  from  correlation to  subject  limitation  of subject  i s involved  and E n g l i s h  i s another  t h e absence o f a sharing  process  i s independent  languages which  the types  t o the presence  Alternatively, whatever  this  clitics.  i n French  that  -sharing  a r e t h e ones with  regarding  suggesting  speculate  property  that  the  others  on  Vengono l e r a g a z s e *Viene l e ragazze T h e r e come some g i r l s • T h e r e c o m e s some g i r l s #11 a r r i v e n t d e s f i l l e s 11 a r r i v e d e s f i l l e s #Le v e n g a n o l e r a g a z z e E' v i e n e l e r a g a z z e •&Le v e n l e p u t e l e Ven l e p u t e l e  null  the  observation  (1981.15)  the  be  and t h e  comparisons.  H-sharing  They  base t h i s  subject  generated  construction there it.  and t h a t  n e e d s t o be a s u b j e c t  Pro i s expletive  have one such Fiorentino by  this  without  with  i n INFL  i s used  here,  f o rt h i s  purpose  clitic,  t o English.  This  being  3s i n t h e  In languages of pro  o r pro i s not i n Introduction  o p e n s t h e way t o a d i f f e r e n t  p r o c e s s o f agreement  when a p o s t v e r b a l  in  This  the construction.  (French i I ,  the features  (Italian)  ( E n g l i s h ) ; but s e e t h e d i s c u s s i o n  respect  may  o f pro a r e determined  Italian),  agreement  then  to license  clitics  makes p r o 3 s a s w e l l .  (English,  by v e r b a l  i n order  subject  the impersonal  clitics  determined  clitic  0 ) . The f e a t u r e s  that  under study  subject  available  that  e, T r e n t i n o such  t h e CNP,IP] p o s i t i o n ,  and languages w i t h  clitic  clitic  languages  are  pro occupies  would  NP s u b j e c t  account  i s involved  f o rthe contrasts i n  (89) .  2.5.  Causatives Another  for  type  of construction  an a n a l y s i s o f object  generated internal (90b)  on t h e v e r b a l argument  i n French,  ( 9 0 ) a.  Marie  b.  Marie  clitics  head  position usually  that  seems t o c a u s e  i n which  they  a r e base  and l i c e n s e p r o i n a  i s illustrated  referred  t o as  problems  9—marked  by a s e n t e n c e causative.  a f a i t manger l a t a r t e a 1' enfant. h a s made t o + e a t t h e p i e to the child ' M a r i e made t h e c h i l d e a t t h e p i e . '  c. * M a r i e  l_i' a fait a fait  manger  l a _ manger 4  pro  t  pro*  a  1'enfant.  a  1'enfant.  like  1£8  The  problem  first  i s that  sight,  the object  i n t o t h e very  £.£.3.  This  i s thus  departs  from  t h e normal  clitic  local  (90b) does n o t e n t e r ,  relationship described  a construction pattern  i n which  which  would  at  in  cliticisat  ion  be t h e o n e i n  (90c). The  aim o f t h i s  analysis  analyzed  to  first  (1980),  his  assumed  here.  analysis I  i s then  construction  move,  has been  interesting  (1978),  Manzini  Goodall  (1983),  Rouveret  described.  i t i s argued, follows  respect  to  The a n a l y s i s s u g g e s t e d i n I intend  a t minimum c o s t , helps  t o show  that  t o t h e one us p r o v i d e  an  automatically. the faire+infinitive  verb  i n t r a n s i t i ves  J'ai f a i t s o r t i r Jean. I h a v e made t o + g o ' I made J e a n g o o u t . '  that  i s referred i n  facts with  e x c l u s i v e l y with  with  complete  (198£b).  where t h e non—tensed  same c o n s t r u c t i o n (91)a.  briefly  of t h e f a c t s that  am c o n c e r n e d  (1975),  causatives.  c a n be a d a p t e d , This  Ernonds  the relevant  i n French  (1984)  approach  Kayne  Zubizarreta  describe  cliticization Goodall  (1981),  raising  The r e a d e r  (1974 s i 9 7 9 ) ,  a  i s an a r e a  i n the literature,  flissen  Jaeggli  & Vergnaud I  This  f o rs y n t a c t i c theory.  particular (1984),  i s not t o provide  o f Romance c a u s a t i v e s .  extensively problems  section  i stransitive.  i s illustrated  3  S  i n  The (91).  The  b.  Ce film fait p l e u r e r l e s gens, t h i s m o v i e makes t o + c r y t h e people ' T h i s movie makes p e o p l e c r y . '  c.  Marie  a fait entrer le chat. h a s made t o + e n t e r t h e c a t ' M a r i e made t h e c a t c o m e i n . '  subject  with  of the infinitive  respect  J e 1_'ai  fait  b.  Ce f i l m  l e sfait  c.  Marie  ( 9 3 ) a.  to faire in  a fait  i stransitive,  a s an i n d i r e c t  as i n (93), t h e  object  a s shown by  (94).  manger  l atarte  a 1'enfant.  <=90a)  II a fait boire du v i n au chat, h e h a s made t o + d r i n k s o m e w i n e t o + t h e c a t 'He made t h e c a t d r i n k s o m e w i n e . '  c.  II a fait signer l e document a sa patronne. he h a s made t o + s i g n t h e d o c u m e n t t o h i s b o s s 'He made h i s b o s s s i g n t h e d o c u m e n t . ' Elle  1ui a fait  manger  b.  II 1ui a fait  boire  c.  II 1ui a fait  signer  regular  object and  entrer.  b.  ( 9 4 ) a.  In  clitics  Marie  object  pleurer.  of the infinitive  cliticizes  dative  a direct  sortir.  1_' a f a i t  When t h e v e r b  the  like  t o c l i t i c i s a t ion.  (9£)a.  subject  i n (91) a c t s  constructions  o f t h e embedded  not on any o t h e r  du v i n . l e document.  with  verb  verb  l a tarte.  a n embedded  i scliticized  (95c)„  tensed onto  clause, the  this  verb  (95b),  ( 9 5 ) a-  II exige 'He  b.  que  demands  II exige  l a patronne that  que  In  causative  that and  t h e boss  l a patronne  c. * I I 1_'exige q u e  cliticizes  not on t h e n o n - t e n s e d  ( 9 6 ) a.  Elle  1_'a f a i t  b.  Elle  en a f a i t  c.  111.' a f a i t  t h e document.'  l_e s i g n e .  though, onto  l e document.  signs  l a patronne  constructions,  the object  signe  signe. i t i s generally  faire  or i t s auxiliary  manger a boire  signer  1'enfant.  au a  chat.  l a patronne.  a fait  l e manger a  1'enfant.  b. ? * E l l e  a fait  en b o i r e  chat.  II a fait  Goodall  is  with  not the r e s u l t  rules.  le,s i g n e r  (19S4) a r g u e s  construction  representation  Goodall's  analysis  coordinated distinct  constitute  l a patronne. the peculiar  to cliticization  follow, which  i s based  structures" structures,  terminal  a  of the interaction  structure  simultaneous  au  that  respect  The p e c u l i a r i t i e s  verb,  verb.  (97) a . ? * E l l e  c. ?#  the case  and o t h e r  o f a number o f  rather,  I will  from  now  a given  phrase  an example o f a t e r m i n a l  phenomena specific  a t more  of  phrase  closely.  "parallel  In the case of  marker  The c a u s a t i v e  this  a specific  look  on t h e e x i s t e n c e  i n t h e grammar.  strings.  behavior of  can contain  construction  string  with  two  would  more t h a t  one  structural  analysis.  According the  t o Goodall  two s i m u l t a n e o u s  (98)a.  (1984s144),  structures  the sentence  given  in  (98).  i n (90a) has  3 3  IP / NP I Marie  \ /  INFL' \  INFL  VP /  \  Vj. I fait  IP / \ INFL' NP / \ I INFL VP 1 ' e n f a n t / \ V NP I I manger l a tarte  a  a  b.  IP / NP I Marie  \ /  INFL' \  INFL  VP Vo / Vj. I fait  a Zubizarreta subcategorization frame that  \  NP I l a tarte  V I manger a  (198£b) a r g u e s t h a t frames:  NP I 1'enfant  f a i r e must  S ( o r IP) and  V.  i s s a t i s f i e d i n the parallel structures t h e NP  contrary sentences  subject  i n (98a) i s i n a p o s t v e r b a l  to the position as i n (99).  i t normally  occupies  satisfy This  two  particular  i n (98). Notice position, i n tensed  (33)  IP /  \  NP  INFL' \  INFL Goodall (99)  VP  adopts Pesetsky's  i s required  assignment.  (i38£)  by a d j a c e n c y  That  1'enfant i n (98a),  adjacent  since  assign  Case  NP s u b j e c t I  now  (98b),  behavior  stipulation. respect the  t o Case  object, marked  e  results  but i f both  hand,  nominative  INFL  does  the structures i n  a plausible i n causatives  of this  the  (98),  account f o rthe without  0  i s assigned  are ruled  n o t be C a s e - m a r k e d Q  further  one Case. may  t h e NP This  withhold  assigns  optionally  more t h a n  l a tarte  being  Case t o t h e p o s t v e r b a l  Case  subject  once. Case  ends  up  impossible,  i t s Case  accusative  and  o u t by t h e C a s e  i n (38b) c a n a s s i g n  do s o t h e n  f o r more t h a n  dative  that  a s s i g n m e n t s Pi) C a s e  V  f o r t h e t w o t o be  i t i s +TNS. C o n s e q u e n t l y ,  Goodall  i n (38a) and V  i t s own.  i s no need  (1384s150) makes two a s s u m p t i o n s w i t h  B, o n e o f t h e v e r b s  assigns and  V  Case  t o INFL.  of c l i t i c s  B) a n NP may  Thus,  in  (nominative)  and does not a s s i g n  since  can provide  Goodall  undesirable  filter;  to  argue with  the order  governs the external  In ( 9 9 ) , on t h e o t h e r  t o the subject  especially peculiar  there  must be a d j a c e n t  will  INFL  I N F L h e r e i s -TNS  to the subject.  that  r e q u i r e m e n t s on  i s , even though  argument  Case  proposal  to the Caseaccording  and t h e o t h e r t o t h e NP  since  object  neither V i nor  govern  this  NP  subject  belonging  to faire  Case  ACC)  two  ( NPs  Case.  ACC  (DAT)),  s i n c e manger a s s i g n s  i n (98b). The  (  i n (98). V  Otherwise  t h e NP  postverbal subject  0  a s s i g n s t h e Cases  and manger w i t h h o l d s i t s only  one Case and t h e r e  1' e n f a n t  then  would  surfaces with  ar  not r e c e i v e the dative  a.  (100)  Marie  We first  a fait  c a n now  manger  l a tarte  concentrate  a  1'enfant.  o u r a t t e n t i o n on c l i t i c s .  Consider  t h e c o n t r a s t s i n (101).  (101) a . * M a r i e b.  Marie  a fait  l a manger a  l'a fait  1'enfant,  manger a  The s t r u c t u r e c o r r e s p o n d i n g  1'enfant.  t o (98b) f o r (101b)  i s t h e one i n  (102) . (10£)  IP / NP I Marie  \ /  INFL  a According this  INFL' \  t o what  VP lai+Vo NP / \ I Vj. Va proi I I fait manger  we  have s a i d  point, the c l i t i c  (101b).  More  precisely,  NP I 1' e n f a n t  concerning  la_ c a n n o t  object  clitics  up t o  appear anywhere e l s e i n  affixation  to V  0  i s the only  possible  position First, that  for this  since  the c l i t i c  l_a b e a f f i x e d  coindexed  The r e s u l t  t o V . Secondly,  has t o govern  a  condition  a  appears  would  on t h e s u r f a c e l i k e  but i t i s i n f a c t  affixation  n o t be  met.  affixation  t o the verb  Lv  Consider dative  natural  posit ion i n order t o 1icense i t . I f  to V i or V , this of this  reasons.  i t i s only  a clitic  0  the higher verb  corn p i e x  f o r the following  i s a Case absorber  p r o i n argurnent  1a w e r e a f f i x e d  to  object c l i t i c  now  object  t h e same c o n t r a s t  with c l i t i c i z a t i o n  of the  i n (103) and t h e c o r r e s p o n d i n g s t r u c t u r e i n  (104) . ( 1 0 3 ) a.# b.  Marie  a fait  Marie  l u i a fait  (104)  l u i manger  l a tarte.  manger  l a tarte.  IP /  \  NP I  Marie  /  INFL' \  INFL  VP /  luij+Vo /  a The a n a l y s i s  NP  \  Vi  V  I  I  fait  a  above h o l d s o f t h e d a t i v e  one  i n (105) which  Marie  proj  manger  objects are cliticized,  (105)  I  l a tarte  When b o t h given  NP  I  l a i luij  appearing  the predicted  i s the right  a fait  clitic  manger  result.  pro  t  proj.  3  string 3  o n V >. c  i s the  135 This of  i s essentially  object  clitics.  causatives; (1976)  In  c f . Goodall  verb  and w i t h o u t that  t o our analysis  to clitics  a n a l y s i s of Spanish  Grammar f r a m e w o r k  o f t h e lower  adapted  (1984.184-186) and  conclusion, the fact  object  approach  I t c a n be e x t e n d e d  fora similar  Relational  verb  Goodall's  i n Spanish  flissen  &  causatives  Perlrnutter in a  "parallel structures".  a clitic  corresponding  s u r f a c e s as an a f f i x  t o an  on t h e h i g h e r  i n c a u s a t i v e c o n s t r u c t i o n s i s not a counterexample  claims  i n 2. S. 3 i f we a d o p t  construct i o n .  2.6.  3  Goodall's  (1984)  views on  to the  this  S  Summary In  with  this  chapter,  how t o a n a l y z e  where they position clitics  subject  a r e not doubled  they  are related  represent  consequently exhibiting  we h a v e  been  clitics  concerned  mostly  in syntactic structures  by a n o v e r t  NP  i n t h e argument  t o . The h y p o t h e s i s  a grammatically  be a n a l y z e d  principally  real  that  subject  n o t i o n and  i n t h e s a m e way  should  i n a l l the languages  them was d e f e n d e d  on t h e b a s i s o f t h e f o l l o w i n g  i t was n o t e d  languages with  f a c t s. First, never  present  realized overt from  simple  subject.  NP o r b o t h .  that  tensed  There This  the hypothesis  sentences  with  i s always e i t h e r was a r g u e d  that  subject  subject  no p h o n o l o g i c a l l y  a subject  t o f o l l o w without clitics  clitics  clitic  o r an  stipulation  a r e generated  i n INFL  136 and  serve  as  position. pro  and  licensing  I f the  CNP,IPJ  a clitic  occurrence  of  elements  must  such  characterisation  appear  an  of  the  and  object  license  pro  i n argument  then  suggested  as  having  binding,  fts  affixes,  they  Furthermore,  others,  that  clitics  the  s e c t i o n was  null  subject  because  pro,  available  i n UG  Recoverabi1ity  any  due  to  licensed  i n the  causative  license  between to  be  this  two  distinct  subject  their types  capacity of  clitic  realizations  to  clitics  respect  i n which  to  to can  of  the  they  are  or  ft'-  fl-binding  Pi-posit ion nor  tentatively  assumed,  Case n o r m a l l y  by  which  other  an  following  assigned  to  empty c a t e g o r y ,  is  the  the  a more g e n e r a l argued  Binding  Principle,  turned  the  followed  in determining  then  to  only  by  the  to.  consists  I  i t can  principled  n e i t h e r i n an  i t was  parameter  like  two  are  absorb  affixed  This  as  The  status with  ion.  are  to a  respect  approach  ft'-posit  they  i n order  argument  "clitic".  an  no  with  position. of  element  analyzed  head  i n INFL  symmetry h o l d i n g  clitics  conceived  same s y n t a c t i c I  i s empty t h e n  a n a l y s i s leads  clitics  be  position  in external  pro.  Secondly,  therefore  f o r pro  was  theory. task  of  d i s c u s s i o n of be  non-existent  initially  In c o n f o r m i t y the  with  the  language-learner  c o n d i t i o n s under which  pro  is  language. to related  c o n s t r u c t i o n s and  i s s u e s , namely, argued  that  the  free  i n v e r s i o n and  a n a l y s i s of  clitics  137 presented found  i n these  In this  here  Chapter  can  and  the apparently problematic  facts  constructions. 3,  c h a p t e r can  external  handle  we  s t u d y how  account  internal  the  general approach  f o r t h e d o u b l i n g phenomena  arguments.  outlined f o r both  in  i38  Footnotes t o Chapter  1.  The m a t e r i a l  in  Roberge  £.  The main s o u r c e  Cord i n  i n £.1  n o t be  spoken  i n Quebec.  French  dialects  relevant  cf.  also  data  i n North  I t shares  found  A good  i n Bacri  4. C f . E v e r e t t P a s h t o.  The s o u r c e s  Africa,  form  i s Brandi  &  f o rthe individual  c a n be f o u n d  to the dialect  properties with the  (or Patouete)  mainly  from  in  Some  Laurendeau,  (198£) a n d R o b e r g e  come f r o m  (1986a);  i s t h e French  i n Algeria.  dialect  The P i e d N o i r  a single  i n f o r rnant,  t h e G r a n r e g i o n a n d now  of confirmation f o rthese  (1983) and i n L a n l y  (1985)  to refer  some s y n t a c t i c  Carroll  thesis  source  French  dialect  Pied Noir  Ne ph t a 1 i , o r i g i n a 1 1 y Quebec.  dialects  i n some p r o v i n c e s o f F r a n c e .  (198£),  of this  i n these  (1985).  Colloquial  forthis  3. l.£.  the rest  i n a condensed  mentioned.  spoken  & Foumier  spoken in  f o rthe data  (1981) and S a f i r  I use t h e term  Neron  a n d £. £ i s p r e s e n t e d  (198551986a).  examples w i l l  3.  £  data Guy  living in  data  c a n be  (1970).  f o r some P i r a h a d a t a  and Huang  (1984) f o r  5.  But c f . S a f i r  in  which  only  ( 1 9 8 5 ) who  pleonastic  presents  evidence  Cf.  Travis  6.  also  Rouveret  Ungues]  p l e i n nominal  of  ou p r o n o m i n a l s  adrnettent  propositions  finies."  observation  Case.  lexical  This  clitics  i s that  Case  feature  element  i s very  and  i n subject  i n VP, P P a n d g e n i t i v a l  incompatible  with  as  l i c e n s i n g elements  i n order  clitic  posit ion  This  in  nuls  dans  t o account  i s a spell-out obligatory i f this  here  that  This treats  hypothesis clitics  f o rpro.  Raising  French.  Jean semble e t r e malade. ' J e a n seems t o be s i c k . '  construction  ( ii),  pas de  (1981) a n a l y s i s o f  7. ft p o t e n t i a l c o u n t e r e x a m p l e t o ( 6 ) i s t h e c o n s t r u c t i on  d'un  can receive  constructions.  t h e one adopted  Lies  sujets,  1'absence  i s therefore  s i m i l a r t o Borer's  is  (i)  i n German.  phonet iquernent  the subject  (1984)  zero-topic.  c e l l e s q u i n'ont  by R o u v e r e t  language  Huang  but  subjects  des s u j e t s  put f o r w a r d  the nominative  no o t h e r  i s non p r o - d r o p  e s t o b i i g a t o i r e en  sujets  this  c a n be m i s s i n g .  makes a s i m i l a r s t a t e m e n t s " D a n s  du c l i t i q u e  The h y p o t h e s i s for  German  a s Norn-drop  q u i o n t d e v e l o p p e d e s f o r m e s de c l i t i q u e s  clitiques les  subjects  (1984) on p l e o n a s t i c  (1980s99)  1'apparition sujet  that  t r e a t s German  i s generally  represented  i n D-structure  as i n  (ii)  e semble  w h e r e t h e NP forced  subject  t o move  position to  this  be  an  Jean  etre  Jean  i n order  of sernbler  o f t h e embedded  t o get  does not  construction  expletive  malade,  pro  C a s e and  get  i n £.£.1,  an  where t h e  external  where  i n subject  infinitival  is  subject  0-role.  I  i t i s shown t h a t  p o s i t i o n of such  clause  verbs  return there  can  as  sernbler.  8.  This  approach  Zubizarreta's  i s compatible with,  (198£b) a s s u m p t i o n s  though  regarding  different the  from,  properties  of  pro. Cf. a two  Rizzi  way  The his  condition  d e t a i l s of  account  the  assignment. Rizzi's argues  (1986)  that  9.  There  i n examples on  theory  are  i n any  problem  only  Zribi-Hertz  of  given  and  i n l.£.3. of  In  Case  i s less restricted  one  by  a head.  She  than  further  d i r e c t i o n (to the  left  p a r t i c u l a r grammar.  f o r such  (1984)  i s that  government  with  (13)  requiring  pro.  presented  pro  holds i n only  o r agreement  (1£)  f o r approaches  licensing condition  i s a construction  have c l i t i c s  based  Rizzi's  government  seems t o c a u s e  (1986)  the appearance  i t implies  to the right)  as  ftdarns  l i c e n s i n g domain o f  or  not  on  ftdams'  since  and  e m p t y NP an  analysis  markers  i n French  since  affixed to  i n the text. who  i n PPs  presents  The  French  does  prepositions  following  arguments  that  is  against  /  141 interpreting "ellipsis" ( i ) a.  or  sentences  as  the  in  result  ( i ) below  of  c.  Notre-Dame,  v o u s v e n e z de p a s s e r devant you came of to+pass in+front 'Notre-Dame, you j u s t passed it.'  Zribi-Hertz by  pro on  even though,  pro  i s not  assumes t h a t the as  regarding  obvious  variable  bound  provided  i n l.£. £ s i n c e i t c a n  of  topic,  similar  in  also  the are  Note a l s o  Mais 'You  that  in  (i) i s  usual not  met.  construction should in  to the refer  (i> c o u l d Chinese  t o an  be  which  detailed  i s not  analysis  discussions of a ( i i i ) a.  Who  available and  item  mentioned  from  preposition  i n French;  Dame?  e! stranding in  c f . Kayne  (1981a)  (1979) and  Bouchard  parallel  phenomenon  in Colloquial  vote  a  examples  Vinet  d i d you  be  (ii).  v o u s v e n e z de p a s s e r d e v a n t have j u s t passed it!'  i t i s different  e.  e.  Pi: P o u r r i e z - v o u s m ' i n d i q u e r o u s e t r o u v e N o t r e ' C o u l d y o u t e l l me w h e r e N o t r e - D a m e i s ? ' B:  for  the  type  position  notes,  empty c a t e g o r y  p r e v i o u s d i s c o u r s e as (ii)  a  The  this  in this by  way.  that  she  empty  i t s licensing  analyzed  English  movement.  Cette valise, j e v o y a g e t o u j o u r s a v e c e. this suitcase I t r a v e l always with 'This suitcase, I always t r a v e l with i t . '  conditions  a  instances of  Les arbres, P i e r r e se cache t o u j o u r s d e r r i e r e the trees REFL h i d e s a l w a y s behind ' T r e e s , P i e r r e a l w a y s h i d e s b e h i n d them.'  occupied  in  as  b.  Rather,  It  the  (198£:£35-6) French.  for?  b. •«• Q u ' e s t - c e q u e n o u s v e n o n s d e p a s s e r 'What d i d we j u s t p a s s i n f r o n t ? '  for  devant?  Another,  less a t t r a c t i v e option  discussed  here  sentences  in (iv).  (iv)a. b„ The in  only  Can Oh  good!  differ  phenomenon I  with  especially  gap  concerning  Rizzi  10. with  them  case  category  Consider  (films Eating i n ( i v ) and  Raoul)  t h e ones  follows a preposition. s e e m s t o be  limited  why  pending  further  research,  i s limited  "topic-variab1e"  Note f i n a l l y  such  The  i n French.  the construction  pro causes  The  to the  t h e gap a l w a y s r e f e r s t o £ p e r s o n you.  open  the  Texas)  with!  c a s e s t h e gap  in (i) i f a  i s adopted.  empty  brought  the question  as  in English.  (films Paris,  the English  and  i n ( i ) as being  this  phenomenon  i s not so r e s t r i c t e d  preposit ions  be t o r e l a t e t h e f a c t  between t h e s e n t e n c e s  i n both  i n that  leave  analysis  You  similarity  preposition  the  I come w i t h ?  (i) i s that  two  of  to a similar  would  that  problem  as t h e ones  an  type  to  object  of  interpretation of the  f o r other  i n Chomsky  analyses (198£)  of  and  (198S).  Possessive respect  NPs  t o pro except  specifier  (or subject)  by a n a g r e e m e n t (i)  pro* 'my  i n Turkish  seem t o b e h a v e  that,  in this  p o s i t i o n o f NP  marker on t h e head  i s t a k o z —unii lobster Is lobster'  noun;  i n t h e same  language, where  way  pro i s i n  i t i s licensed  cf. Kornfilt.  (1984).  143 ( ii) prcn  tstakoz-urnuzi lobster lp lobster'  'our It  i s riot  clear  specifier. that a  Massarn  to the  (iii)  also  John's s t o r y  o f N'  m a r k e r s on  11.  genitive and  That  (1985).  The  t h e book  no  disturbed  and  expletive  position  I cannot Case  t h e book  finderson  of the examples pro  (1984), and  disturbed  verify  Chapter 3(footnote  (1983).  is £s 3s  do  assign  i s allowed  the value in this  morphology  me.  Note  i n ( i ) and  that  t o appear as clitics  of t h i s  or  the  agreement  prediction,  fts  see Borer  15).  agreement  i s "rich"  o f t h e agreement  i n European paradigm  markers  1p £'p 3p  from  Raposo  i s t h e same f o r  'to eat'  eu comer+0 I tu comer+es you e l e comer+0 he  i f this  ( i i ) , i t i s then  construction  i s s u p p o r t e d by t h e f o l l o w i n g  comer  assumes  story  c o n jugat i o n c 1 a s s e s . (i)  to the  me.  i n languages with  assignment  subject-verb  Portuguese  all  N.  (1976) also  specifier  reading  Chomsky  a 9—role  (1984)  b.  that  (1984)  below  John's reading  predicted  for  following  a.  holds  noun a s s i g n s  specifier,  Chomsky  Gruber  analysis  t h e head  (1985),  i n the examples  9-role  Cf.  here that  n o s comer+rnos we v o s eorner+des you e l e s corner+ern they  144  For  example,  subject  of  agreement (ii)  12.  an  dialects  clitics.  realized.  the  Brandi  I return to  mentioned  rest  of  this  bom  in Iraq  now  between the  object  (characterized example  lexical these  must  by  are  NP  bear the  a  in  in  the  of  as  the  relevant  Rizzi and  3. the  their  2p;  cf.  Renzi  (1984).  Fiorentino  subject  are  is  3.1.1.  data  found  s i n g l e informant,  in  Jumah  the Harneed,  Canada.  clitic  and  as  NP  Arabic  i n assuming  +)  I s , I p and  in Trentino  issues  to  paradigm  (1981) and  (1984)  in Jaeggli  (i)  i n the  preverbal  otherwise,  living  I follow Borer  the  opposed  to  verb  that  the r e l a t i o n s h i p  i s one  of  affixation  adjunction  as  in  (i)  found  (1985).  V / clit ic  Jaeggli  verb  they  clitics  t h e s i s i s due and  gaps  & Cord i n  subject  e v e n when a  Unless  have  In T r e n t i n o ,  (1982),  obligatory  for  the  lexical  S e r a d i f i c i l C os d e p u t a d o s aprovarem a p r o p o s t a l ' I t w i l l be d i f f i c u l t f o r t h e d e p u t i e s t o a p p r o v e proposal.'  Furthermore,  14.  verb,  a  marker.  Vanelli  13.  following sentence with  infinitive  Some N . I t .  subject &  i n the  V  (1985.2)  syntactically.  \ assumes t h a t  However,  as  the  clitic  iss  i t i s dominated  by  "a  separate  the  same  'word' level  node a s t h e word be the  part  15.  of the verb."  clitic  Borer's  i t i s affixed In t h i s  onto,  i t i s considered  view,  i s t h e same a s t h a t  t h e government  o f t h e verb.  also to  domain o f  The same  i s true i n  analysis.  A similar  approach  i s developed  i n a d i f f e r e n t way  in Safir  <1985).  16.  Cf. Kayne  (1975)  f o r the various  properties  of  this  construct ion.  17.  For analyses  McA'Nulty  18.  19.  (1983),  For Rizzi,  Italian  i n v o l v i n g Verb Lobeck  this  Raising  &• R o b e r g e  holds  only  s e e Emonds  (1978),  (1985) and C h a p t e r  of the subject  clitics  4.  i n t h e N.  dialects.  Note that  (54) i s t h e r e p r e s e n t a t i o n  adopted  in Safir  (1985).  £0.  Cf. J a e g g l i  River  £1.  Plate Spanish  Cf. Hurt ado  Plate  (198£;19S5), B o r e r  Spanish  B.  ( 1 9 8 4 ) a n d 3.£  f o r more on  A.  (1985),  Borer  ( 1 9 8 4 ) a n d 3. £ f o r m o r e o n  River-  ££.  But see I n t r o d u c t i o n  constructions  £'3.  This  with  no o v e r t  type of analysis  providing  f o ra discussion  an account  o f some  English  subject.  also  has t h e non t r i v i a l  f o r t h e apparent  advantage  of  violations of that-t; cf.  1. d. 1.  £4.  For a discussion  Zubizarreta they  of the that-t  (1981s198£a)  i n French  Portuguese,  such  not  French  £5.  que s t a n d i n g  q u i . F o r Bouchard,  structure  (1986:400) f o r free  i s a  f o r t h e nominal  though  i t  i s  unique  equivalent  of  strategy.  i n Travis'  f o r the point  Borer  e x i s t even  In  t h e v i o l a t i o n s a r e t o be r e l a t e d t o  pronoun  The d i f f e r e n c e s  For Zubizarreta,  extraction of the subject.  realized, i.e. there  empty r e s u m p t i v e  crucial  allowing  (198£).  s i m i l a r t o t h e que/qui  an a l t e r n a t i o n would  morphologically  complementizer  an  and Bouchard  are the result of a strategy  alternation  v i o l a t i o n s see i n p a r t i c u l a r  I want  use o f I ' , V  t o make  here.  suggests the following  inversion  a n d N' a r e n o t  constructions.  base-generated  147 (i)  S / INFL  \ /  VP' \  VP I V  £6.  Pis p o i n t e d  problem  with  NP  o u t i n Koopman  this  claim  Portuguese  ( c f . £.£.1)  nominative  Case even  alternative  account  INF"L s i m i l a r also  Villiard  nominative  £7.  that  £8.  £9.  fact  i n terms  instead  argues  infinitives  proposes  o f a V-movement  f o r TNS  an to  f o r i n V a t a and G b a d i .  explicitely  in  i s marked f o r  of this  who  apparent  See  as  assigner.  against  & Rochernont  (1983), P o l l o c k  analyses of the definiteness  (1985) f o r  effect  such  as  (1985).  e b e c o m e s q1' when  Gl' ha portato l a torta C l ( n e u t e r ) has brought t h e cake 'Pi g i r l b r o u g h t t h e c a k e . '  It i s possible  suggests  the subject  i s -TIMS. K o o p m a n  This neuter c l i t i c  Safir's  i n which INFL  (1985)  of Safir's  (i)  regard t o inflected  t o t h e one she a r g u e s  Case  t h e r e i s an  though  But see C u l i c o v e r  arguments  with  (1984s£l£),  to translate  proposal into to define of lexical  framework.  i n terms  projections.  In ( i ) ,  by a  vowel.  una r a g a z z a . a girl  the structures  the Barriers  sisterhood  followed  i n (88) and Chomsky  o f maximal which  (1986s14)  projections  corresponds to  (87)  i n the text, ( i)  b o t h NPs a r e s i s t e r s  o f VP.  IP \ NP I' / \ INFL VP / \ VP NP /  Borer  (1986)  text  but without  also  direct  30.  Note  clitic  This  involving (1978) the  might  position.  i n (80) would  I argue  i s true  to the postverbal  that  also  n o t do s o b e c a u s e  (i)a.  i s base  verbs i n t h e sense of  generated i n direct  firriva Giovanni. 'Giovanni arrives.'  b.  IP / NP I pro  \ /  INFL' \  INFL  VP / V I arriva  i s no  t o 3.1.4 a n d Case.  inversion construct ion  or " e r g a t i v e " i n t h e sense of Burzio  subject  there  do n o t "have t o " a b s o r b  f o rthe free  "unaccusative"  since the preverbal  discussion of this  clitics  NP.  f o r t h e a n a l y s i s i n £. £.  absorb Case  I delay  t o (88b) i n t h e  In her a n a l y s i s , there i s  be a p r o b l e m  clitics  t o be a b s o r b e d .  3. £. £ w h e r e  31.  this  t o which  impersonal Case  a n CNP,S3  ©-role a s s i g n m e n t  that  according  adopts a s t r u c t u r e s i m i l a r  \ NP I Giovanni  Perlmutter  (1981)  object  and where  position.  3£»  The  c o n s t r u c tion with f a i r e  "typical" of  laisser  reqarder,  33.  causative construction.  the faire+infinitive  like  The  and  and  Assume h e r e  the  NP;  35.  Cf. Goodall  indices  a r e found  the  properties  with other  such  verbs  as yoir,  entendre,  (1975).  i n (98) a r e used  h a v e no t h e o r e t i c a l  f o r the sake  of  significance.  that a transmits dative  (i)  (1984.174)  marginal, result  requirements  on Case  ? Marie  There  analysis  Case  i s one  for a discussion i n ( i ) based  from  the verb  to  l a * , m a n g e r pro±  proj.  potential  incompatibility  between  presented  in  a very  limited  number o f  accusative or dative,  as  verbs shown  ( i ) .  (i)a.  Marie  a fait  ecrire  b.  Marie  a fait  ecrire  'Marie  Goodall's  here.  s u b j e c t t o be e i t h e r  by t h e e x a m p l e s  possible,  adjacency  l j ' a fait  and t h e one  their  on  of the  assignment.  causative constructions,  Goodall  cases,  t o as the  c f . 3. £. £'.  although  allow  I n most  some p e r c e p t i o n v e r b s  numerical  34.  In  sentence  e c o u t e r ; c f . Kayne  presentation  36.  i s often referred  1'enfant, a  made t h e c h i l d  (1984.156) assumes  1'enfant. write.'  that verbs  such  as e c r i r e  have  an  implicit is  object  assigned  which  to the subject.  accusative  Case.  It  be  object  might  i s pro«  therefore not up  now.  In  the case that r b  have  licensed to  receives  by  the accusative  Case;  <ia), 1'enfant  i s  such  an  implicit  irithe sense of R i z z i  an example a clitic,  (1986)  o f an argument in contrast  pro  with  dative  Case  assigned  accusative i n ( i b ) . We  would  i n French that  what  I have  is  proposed  CHAPTER 3: ON C L I T I C  3.0.  Introduction In t h i s  2(1).  chapter  I wish  doubling  discussion Italian that  we w i l l  t o suggest  be c o n c e r n e d  a unified  c o n s t r u c t i o n and point  differences  with  respect  on subject  dialects  account  representation  o f the subject  out i t s s i m i l a r i t i e s and  t o object  doubling  with  doubling  i s based  a n d some F r e n c h  (3. 1 a n d 3.2).  mostly  dialects.  Safir  doubling (1985).  and f r e e  Rather  I want  generated  i n t h e P.GR e l e m e n t  optionality I  argument  i sthen  then  chapter  constructions,  where s u b j e c t  position.  clitics are  The p o s s i b i l i t y  r e l a t e d t o a parameter  making  with  of subject  reference  to the  by t h e c l i t i c .  i n 3.3 t o a m a j o r namely  assumed i n  o f INFL and a r e a s s o c i a t e d  o f Case a b s o r p t i o n turn  between  i t seems p r e f e r a b l e t o assume t h e t h e o r y  i nt h e previous  doubling  The  t o argue  inversion as i s implicitly  developed  i nexternal  1  on t h e Northern  i ti snot desirable t o e s t a b l i s h a connection  subject  pro  DOUBLING  problem  related t o doubling  e x t r a c t i o n s . Pin e x a m i n a t i o n  of the facts  concerning  both  lead  explicit  t o an  clitics what  and  I wish  call  This  then  c-chains  e x a m i n e two that  the  some N . I t .  dialects  though  the  two  in  with  how  3.1.  are  i t i s best  Subject  3.1.1.  Brandi  the The  doubled  Jaeggli  are  free  or  these  theory  on  reflexive as  between  developed and  not  surface.  clitic  forming  se a  form  Pi- o r P i ' so f a r .  3.5.  First,  in St.Fr.  involve subject doubling  similar  will  elements  from  I n v e r s i o n found  the  and  even  This chapter  and,  more  I  ends  precisely,  c-chain.  (1984),  inversion  N.It.  dialects  Renzi  Rizzi  data  as  presented  (1982).  and  Safir  i s from  in  Different Bouchard  (1985).  Except  T r e n t i n o and  taken  above. determining clitic  what  in  ( 1 ) , but  in (2).  s  kind of  lexical  i s d e f e r r e d t o 3.3.  i n Trentino whether a as  been  cf. especially  (1984),  the  have  &• V a n e l l i  proposed?  mentioned,  a subject  not,  Dialects  (1981) and  question of  obligatory  present  that  Complex  Italian  references  by  relationship  different  the  i n syntax  i s s u e s i n 3.4  have s i n c e been  where o t h e r w i s e from  on  about  & Cordin  (1982),  are  as  Doubling  facts  analyses  that  does not  handled  Northern  The  well  related  very  a note  as  It i s proposed  i s consistent with  to argue  3.5  LF  formulation of the  arguments.  I will  chains.  e x t r a c t i o n s at  are  lexical  excluded  NP  can  Subject  p r e v e r b a l NP i n the  cases  be  clitics is of  (i)  a . *<E1) m a g n a . cl eat-3s ' He e a t s . ' b.  El Mario * ( e l ) the 'Mario eats.'  c.  * ( L e ) ven. cl come—3p ' T h e y come.'  d.  Le the  putele girls  'The  girls  (£)a. b. c.  come.'  (#Le) v e n l e p u t e l e . E 'vegnu 'na p u t e l a . has come a girl 'Pi g i r l c a m e . ' e i vegnuda  i n St.Fr.  una  the subject  cf.  3.4  (3)  ( E l Mario)  putela.  clitic  c a n be i n v e r t e d  in  questions;  f o r the details.  This  inversion  from  embedded  (4)a.  ven.  (*E1) m a g n a e l M a r i o .  d. * L' cl Pis  *(le>  magna,  maqnelo? i s possible  i n matrix  sentences  but e x c l u d e d  contexts.  Ho d o m a n d a s e ( e l M a r i o ) e l m a g n a a c a s a . ' I w o n d e r e d i f ( M a r i o ) h e e a t s a t home.'  b. * Ho d o m a n d a s e Pi s e n t e n c e  like  ( e lMario)  magnelo  t h e o n e i n (b) , w h e r e  a  casa.  the preverbal  clitic  i s  154 interpreted case  (5)  This  i s not t h e  clitic.  # E l magnelo? facts  Trentino  that  suggest  the following interpretations.  and t h e o t h e r  languages  i n the sense  N. I t . d i a l e c t s developed  the p o s i t i o n s occupied  the lexical  tensed  clause  NP at  (6) a . * Ven. come-3s 'He/she  subject  c. *  Magna. eat-3s ' He/she  cannot  given  and t h e one  g e n e r a l l y both  the  fact  occupied  be e m p t y  in a  eats.'  the free  Spanish  inversion property  i s found  i n £.4.  obligatory,  The  by t h e c l i t i c  2,  E l / l a magna. ' He/she eat s.'  Secondly,  3.1.2.  i n chapter  subject  comes.'  E l / l a ven. ' H e / s h e come. '  argued  are null  First,  D-structure.  b.  d.  and  i s ungrammatical.  i f e l i s t h e hornophonous o b j e c t  These  by  as 3s subject,  a n d c a n be a n a l y s e d  Finally,  subject  first  of  i n t h e same way  doubling  i n some c o n s t r u c t i o n s s u c h  Previous  characteristic  Italian as  i s p o s s i b l e , and e v e n  a s i n (1) a b o v e .  Analyses  a n a l y s i s t o be r e v i e w e d  here  i s that  of  Safir  (1985).  The  account  f o r the  that  must  in  INFL t o t h e  as (7)  L>  derives  impersonal  from  Subject  Case  inside  this  that  are  subject  clitics  Therefore,  (lb,d),  seem t o a c c o u n t  in  are  the  i s assigned  VP  and  coindexed  the  by  lexical  c l i t i c .  3  Case assignment Case i n h e r i t a n c e  Case  f o r why  c l a u s e s and  in subject an  overt  c o n s t r u c t i o n s , as  in  (£)  Case assignment  clitic  to the  i s the  subject c l i t i c  i n v e r t e d NP  subject  the  clitic  lexical  subject c l i t i c (cf. also  government as  transmission of and  how  above  through  "incomplete"  the  is  NP  doubling constructions.  absence of  Case assignment the  to  and  (4a)  to  Cli+V  the  from  i n VP  verb.  generated  Case  in non-inverted  Furthermore,  the  in order  VP  nominative  inversion  on  Safir  ( l b , d ) and  s> —i—;>  These assumptions  receives  in  assigned  corresponding  inherit  by  Case d i r e c t l y .  affixes  S / I \ NPi INFL  obligatory  be  subject c l i t i c  s u b j e c t can  (7)  assigned  adopted  phenomenon  C a s e must  be  generated  structure  NP  doubling  nominative  clitics base  basic assumptions  in a  thus  by  in  free  £(83)), the  (8). That  i s because  f e a t u r e which  renders  phonologically null.  156  (8)  S / I \ INFL VP  NP  /  \  V L>  Case  Cl+V  inheritance  since  the  derived there  clitic  the  fact  definiteness  be  assigned  under  Note  finally  in this  the  may  then  formed 5  for a  detailed  see  of  Safir  free  that  resulting the  the from  inverted  So,  as  all  the  that  clitic.  long  as  an  The  inversion  on  sense  and  that,  must  be  is  i.e. the  is a  by  cf.  also  doubling  inverted i n VP  INFL.  clitic  doubling  by  the  free  £.4.  Two  are NP  of  to a  things the  assigns There  Case  is  inherit  also Case  construction.  assigned  and  can  inversion  p o s i t i o n may  clitic  adopted now  inversion.  subject  i s made p o s s i b l e  an  one  i n VP  NP;  free  for  the  is free  subject  subject  of  i t s C a s e and  £)  in subject  the  presence  retains  (6b,d);  result NP  this  effect,  being  inverted  Case d i r e c t l y  in  is a  the  there  result  there  conditions  Free  the  Case  doubling  clitic  i s as  NP,  possibility the  the  sentence  to  subject i n the  impossible  From  account  resembles  nominative  government  is  (1985).  inversion  in that  1)  coindexed.  definiteness  nominative  happens  not  i s no  approach  assigned  are  constructions  there  realized phonetically,  clitic  met.  NP  analysis (1981)  to  and  effect  Chomsky  be  inversion  that  in  tied  in free  9-chain  i s no  This  NP  Case  in  VP,  inversion  are  parameter  in  (9).  157  (9)  Free  Inversion  Parameter  (Safir  1985:£34)  If the (impersonal) subject c l i t i c C a s e by I N F L , t h e n i t c a n , in turn, For  the  languages  instance  in Standard  stipulate  that  structure.  The  does not with  inversion  Safir  NP  i s no  subject  clitics  can  Case then  inverted  subject  subject  It  (1985:£59) m a k e s t h e  by  i s a well  clitics  e n d s up  but  has  following  doubling,  French,  inheritance  It  only  subject  Standard  not  follows  have f r e e  i s made p o s s i b l e  the  i s no  and  Case F i l t e r .  (1985:£66).  (overt)  there  Italian  there  pass the  subject  Case t o  i n which  i s assigned a s s i g n Case.  up  one  in  f o r Case  only  assignment  a subject  that  inversion.  free  nominative  clitic  known f a c t free  of  affixed  to  S t . I t . has Thus  Safir  claims  If a l a n g u a g e L has any argument c l i t i c s (subject, object, indirect object) associated with c l i t i c slots on a v e r b , t h e n , i n t h e u n m a r k e d c a s e , L has t h e full c l i t i c paradigm.  This  claim  clitics.  him  Note that  speculations  this  concerning  construct ions;  cf.  inversion  suggested  something  like  I will  to  assumption  £(89).  (10),  therefore  assume t h a t  verbal  not  which assume  S t . I t . has  agreement  does not I take  silent  i s incompatible  Moreover,  i n £.4  V: no  (10)  allows  and  languages  inversion since  by  must  the  marked that  for  the  in  to  i t i n the  be  with  an  rest  my  inversion  a n a l y s i s of  require  subject  one  to  free claim  unfalsifiable of  the  thesis.  claim.  The Rizzi in  second  analysis  adopts the s t r u c t u r e  in Rizzi  (1984).  i n (11) f o r s e n t e n c e s  Recall like  that  t h e ones  (1b,d).  (11)  IP / NP I Mario  El  Of  i s found  course  N. I t .  / INFL I el  VP I magna i n (1£)  i s also  available i nthe  dialects. IP / NP I pro  Rizzi  \ /  INFL' \  INFL I el  (1984.2)  agreement"  VP I magna  adopts an a n a l y s i s  and c l a i m s  St. I t . i n that  Phonetic also  INFL' \  the representation  (1£)  from  \  that  i n i t s abstract  commitment  t h e N.It.  "the strong  Form n o t o n l y  i n terms o f " r e d u p l i c a t i o n o f d i a l e c t s only  differ  PiGR i s r e a l i z e d i n  i n the concrete  verbal  morphology, but  s y n t a c t i c p o s i t i o n CINFL]".  There  i s no  made a s f a r a s C a s e a n d © - r o l e a r e c o n c e r n e d  although  Rizzi  o f f e r s t w o a l t e r n a t i v e s : 1) t h e c l i t i c  nominal,  i t i s thus  n o t an argument  lexical  o r empty argument  subject  form  (given  i n CNP,IP]);  a syntactic chain  true  the presence o f a  the c l i t i c  and s h a r e  i s a  a n d t h e NP  t h e s a m e C a s e a n d 0-  159 role;  £) t h e c l i t i c  therefore This  irrelevant  unanswered.  represented  than  How  by C o l l o q u i a l  doubling  French  between s u b j e c t  out  be a c c o u n t e d  object  doubling  given  that  would  a simple  influence (cf.  they  f o r ? Why  morphological  on e x t r a c t i o n  French  clitics  clitics  distinct  o f AGR  that  just  pointed  doubling  i n many r e s p e c t s  possibilities  dialects  can the s t r i k i n g  are subject  realization  analysed?  are truly  and o b j e c t  as two c o m p l e t e l y  are so s i m i l a r  French  cases  and  phenomena  ( c f . 3 . 2 ) ? Why have t h e  i t apparently  has  3.3)? I  will  these  argue  questions  specific with  analyzed  Noir  o f PiGR, how  clitics  assignment.  important  i n these  I f subject  i s  intermediate  and P i e d  realization  similarities i n £.2.3  and 9 - r o l e  a r e t h e apparent  dialects.  morphological  i t s presence  number o f  i s more r e s t r i c t e d  i n the N.It.  another  f o r Case assignment  analysis leaves a certain  questions  Subject  i s not a t r u e nominal;  As  French  concerning  t o Case  sections that  by a t h e o r y  that  answers t o makes  the properties of subject  clitics  theory.  Dialects  mentioned  dialects  c a n be p r o v i d e d  proposals  respect  3.1.3.  i n the following  i n £.1,  we w i l l  i n a d d i t i o n t o the N.It.  be c o n c e r n e d dialects.  with  two  French  3.1-3.1. The  Colloquial first  variation with  dialect  very  respect  French is Colloquial  similar  to  the  to  the  French,  property  d i s t r i b u t i o n of  of  which  the  exhibits  N. I t .  l e x i c a l NPs  and  a  dialects subject  clit ics. In t h i s  dialect  a  referential  subject  clitic  (13)a.  Now,  Marie (elle) ' M a r i e (she)  P i e r r e ( i l ) mange. ' P i e r r e (he) e a t s . '  c.  C e t homme ' T h i s man  d.  Les l o y e r s 'The r e n t s  doubling  in  and  St.Fr.  not  PF;  cf.  by  that  the  s t r u c t u r e would (1984)  with £.£.4.  \  TOP  IP /  Mar i e  \  NP  INFL' /  elle  by  e m b r a s s e r M. Caouette k i s s Mr. Caouette.'  above examples are  l e f t - d i s l o c a t i o n s of  (1984) and  /  doubled  (13).*  in  TOP'  (14)  be  ( i l s ) ont augmente. (they) have increased.'  Rizzi  Kayne  optionally  (il) a dit qu'il allait (he) s a i d t h a t he w o u l d  obvious  appropriate  as  can  vient. comes.'  b.  i t i s not  the  l e x i c a l NP  INFL  \  VP vient  be  the  the one  subject, in  cliticization  (14) to  the  instances for  which  argued right  for in  c  161 Carroll  (198£)  dislocations the  from  a r e not d e c i s i v e l y  her tests.  i n order t o distinguish  doubling constructions.  examples she examines,  (13) to  p r o v i d e s some t e s t s  only those  cases  Interestingly,  similar  t o t h e ones i n  of (left-)dislocations  Some e x a m p l e s o f d i s l o c a t i o n s  ofa l l  according  are provided i n  (15) .  ( 1 5 ) a.  I I e s t p a r t i , Jean. 'He i s g o n e , J e a n . '  b.  Pierre est f i n , l u i . ' P i e r r e i s n i c e , him.'  c.  Dodo e s t m a l a d e , l a p a u v r e . 'Dodo i s s i c k , t h e p o o r (girl).'  That  the sentences  basis  above a r e d i s l o c a t i o n s  of various tests.  between t h e d i s l o c a t e d sentence. that  55  Pi d i s l o c a t e d  F o r example, constituent constituent  there i s a strong and t h e r e s t  a grammatical  function.  the sentences  i n (16) would  thus  o f an element  (16) a . * E s t p a r t i ,  to f i l l  obligatory case  pause  a  position  The u n g r a m m a t i c a l i t y o f  be e x p l a i n e d i n d i r e c t l y  by t h e  function.  l a pauvre. i n i n s t a n c e s o f d o u b l i n g t h e r e i s no  between t h e two elements  i n t h e examples  involved  pause  Jean.  b. * E s t m a l a d e , On t h e o t h e r h a n d ,  this  on t h e  of the  does not occupy  receives  1ack  i s determined  i n (13). Furthermore,  i n doubling constructions  involved. the  This  i sthe  elements  are i n "regular"  syntactic  162 positionsfreedom  i n the  subject, clitic  It should  for  choice  of  instance,  which  structure,  therefore these  could  i s i n INFL,  as  argued  i s impossible  the  in  (17),  b.* Only and  Dodo  verb  as  in  type  of  NP  structure  like  intervene.  (17)  an  second  clitic  is less  only  by  a  NP  subject  This  i s in fact  doub1ing  D-  what  constructions  like  for other only are  the  malade.  there  NPs  reason  CNP,IP!.  On  the  receives  external like  sentences to  argument  1 u i and  exclude  other  a natural  in  hand,  any  with  p o s i t i o n making  l_a p a u v r e  in  (17)  to  and  The  contrast  between  (13)  for. in  French  (17) and  have  with  grammatical  i n C o l l o q u i a l French.  i s made b e t w e e n t h e  (18a,b) whereas,  appears together  a  explanation:  generated.  sentences  in  no  subject  where  o b l i g a t o r y pause as  the  NP  s y n t a c t i c p o s i t i o n a v a i l a b l e i s INFL  i n Standard  NP  lexical  be  fact the  between a  s t r u c t u r e of would  in  this  accounted  Note that  the  there  (15).  I f the  occupies  clitics  equivalents  (14)  (11),  The  i s thus  former,  £.  intervene  appearing  NP  impossible  can  (13).  in  from  leftmost  subject  doubled  find  compare t o  clitics  were t h a t  it  be  to  l a pauvre est  (13)  the  that  P i e r r e l u i est f i n .  subject a  case  c o n s t i t u e n t s . Pi d o u b l e d  i n Chapter  It  (17)a. *  the  a syntactic position available in  happens. ones  be  the  i n the  pause as  in  first  latter,  a  (18c,d).  NP  In  the  and  subject These  163 sentences are (16)a.  used  Pierre,  b.  Dodo,  c.  Pierre,  d.  Dodo,  Studies introduce cases,  est  l a pauvre,  elle  reintroduce  referent  can  particular  Keenan  in  C. F r .  sentences  involve  Moreover,  subject.  malade.  have shown t h a t  that  emphasise the  l u i , i l est f i n .  in  clitics  to  l u i , est f i n .  l a pauvre,  or  the  i n order  an  est  malade.  left-dislocations  a referent be  considered  & Schieffelin like  the  a discourse-new object  can  i n the  be  as  discourse.  in  always  In  both  'discourse-new';  (1976).  ones  in English  It  i s not  (13)  with  the  see case  subject  referent. left-dislocated  i n C.Fr.  as  in  (19) . ( 1 9 ) a. b.  When a  Pi W h i s t l e r , o n y_ v a 'To W h i s t l e r , we go  souvent. there often.'  J e a n , j e J J B r e n c o n t r e de t e m p s en ' J e a n , I meet hirn o c c a s i o n a l l y . ' left-dislocated  doubling,  the  object  object  i s found  appears along to  the  left  (£0)a. Pi W h i s t l e r , mes a m i s i l s y v o n t 'To W h i s t l e r , my f r i e n d s ( t h e y ) b.  In  temps.  with  of  the  souvent. go t h e r e  subject subject.  often.'  Jean, Marie e l l e l e rencontre souvent. ' J e a n , M a r i e (she) m e e t s him often.'  order  to  get  an  interpretation  i n which  both  the  object  and  164  the a  subject  strong  elle the  are left-disloeated,  pause  (##) b e t w e e n t h e s u b j e c t  o r t o move t h e s u b j e c t lexical  ( 2 1 ) a.  object  Mariej  'Pierre,  Marie  Pierre,  c.  Marie,  Pi s e n t e n c e without  Marie  (13), left  ## e l l e  l ' a vu.  regular  pause,  cannot  phonological  phrasing, i . e .  be i n t e r p r e t e d a s i n v o l v i n g a  o f the subject  i n C.Fr.  t h e s t r u c t u r e i n (14) f o r t h e s e n t e n c e s i n  one would  expect  of thematerial  found  namely complementizers more d e v e l o p e d  i n front of  U ' a vu.  (21a) w i t h  one a c c e p t s then  ellej  l ' a vu.  left-dislocation If  further to the left,  elle  a strong  and t h e c l i t i c  (she) has seen him.'  Pierre,  like  Marie  a s i n (21b, c ) .  Pierre*,  b.  i t i s p o s s i b l e e i t h e r t o make  the leftmost  NP t o a p p e a r  i n COMP o r t h e S p e c i f i e r  and Wh-phrases.  structure i sthat  This  i n (22),  to the  o f COMP,  i sso since t h e  f o l l o w i n g Chomsky  (1977). (22)  TOP' (=S''> \  TOP  CP /  (=S') \  Spec C  C /  \  COMP  IP /  <=S) \  NP / INFL The  f a c t s do n o t a c c o r d  with  I' \ VP this  p r e d i c t i o n . I n C.Fr.,  when  the  c o n s t r u c t i o n i s used  is  found  to  the left  (£3) a.  b.  The  to the right  i n embedded  contexts, the lexical  o f t h e complementizer  NP  o r Wh-phrase, n o t  a s (££) w o u l d s u p p o s e .  II a fallu q u e rnarnan e l l e a i l l e t r a v a i l l e r s u r i t w a s n e e d e d t h a t mom s h e go to+work on la ferine, the farm ' I t w a s n e c e s s a r y f o r mom t o g o w o r k a t t h e f a r m . ' Je I ell she 'I  same  me s o u v i e n s quand l a p e t i t e f i l l e du voisin REFL r e c a l l when the l i t t l e girl of+the neighbour e s' e s t noyee. REFL i s d r o w n e d remember when t h e n e i g h b o u r ' s d a u g h t e r d r o w n e d . '  i s t r u e when t h e c o n s t r u c t i o n i s u s e d  i n matrix  sentences. (£4)a. Q u i q u e M a r i e e l l e v e n t v o i r ? 'Who d o e s M a r i e w a n t t o s e e ? ' b.  If  Quand q u e P i e r r e 'When w a s P i e r r e  the lexical  i l a e t e malade? sick?'  NP i s t o t h e l e f t  of the material  S p e c Comp, a s i n (£5) c o r r e s p o n d i n g only  then,  (£5) a.  c a n i t be i n t e r p r e t e d  II a fallu,  rnarnan,  b.  J e me s o u v i e n s , s ' e s t noyee.  c.  Marie,  d.  Pierre,  l a petite  qui qu'elle quand  qu'elle  qu'il  veut  as  i n COMP o r  t o (£3) a n d < £ 4 ) , t h e n , a n d left-dislocated.  aille fille  t r a v a i l l e r s u r l a ferme du v o i s i n ,  voir?  a e t e malade?  quand  elle  Or a l t e r n a t i v e l y : (£6) a. b. (£7) a .  b.  The  Maman,  i l a fallu  II a fallu  qu'elle  qu'elle  La p e t i t e f i l l e s'est noyee. J e me s o u v i e n s du v o i s i n .  aille  quand  i n ( £ 5 ) , (£6) a n d  Finally, NP u s e d  there  non-specific), (£8)a.  t r a v a i l l e r s u r l a f e r i n e , rnarnan  elle  s'est  fille  like  quantified  noyee,  l a petite  on t h e n a t u r e  ( 1 3 ) : t h e NP c a n n o t  c.  o r t+whll  a s shown  (he) i s a l w a y s t h e r e . '  (£9)a.  Mes a m i s i l s v i e n n e n t a s o i r . 'My f r i e n d s ( t h e y ) c o m e t o n i g h t . ' b. * B e a u c o u p d ' a m i s i l s v i e n n e n t a s o i r .  c.  l o tof friends  (they)  Beaucoup d'amis viennent  (30)a.  P i e r r e i l aime l a t a r t e . ' P i e r r e (he) l i k e s p i e s . ' b. •* Q u i i l a i m e l a tarte? who h e l i k e s t h e p i e 'Who 1 i k e s p i e ? '  c.  Qui aime  l a tarte?  come a  tonight.'  soir.  of the lexica  i n (£8)-(30).  Un a m i e s t t o u j o u r s l a .  'P.  fille  be i n d e f i n i t e (o  b. * Un a m i i l e s t t o u j o u r s l a . friend  elle  (£7) b u t n o t i n ( £ 3 ) .  Son ami i l e s t t o u j o u r s l a . ' H i s f r i e n d (he) i s a l w a y s t h e r e . '  'Pi  quand  du v o i s i n a r e i n d i s l o c a t e d  i s a constraint  i n sentences  t r a v a i l l e r s u r l a ferine  d u v o i s i n , .je me s o u v i e n s  N P s rnarnan a n d l a p e t i t e  positions  aille  /  167  This  particular  analysis for  constraint  of this  i s often  construction  used  i n terms of subject  one i n terms o f l e f t - d i s l o c a t i o n ;  Rizzi  (1984).  these  f a c t s c a n be i n t e r p r e t e d v e r y  constraint  conclusion  i s often  construct ions; (1980)  This  cf.  associated Borer  among o t h e r s .  with  (33)  illustrate this.  (31)  * Loi v i a un c h i c o i . h i m saw t o a boy ' I saw a boy. '  Jaeggli  and  (1980),  however, and  differently object  an  doubling  c f . Salarnbos  i s problematic,  (1984),  The R i v e r  t o argue against  since  this  doubling (1982),  Plate Spanish  Steriade  examples  in  (31)-  (3£) * L a s * v i a todas l a s chicasi. them saw t o a l l the girls ' I saw a l l t h e g i r l s . ' (33)  * ft q u i e r i i l o v i m o s t_t ? t o whom h i m saw 'Who d i d we s e e ? ' t  But : (34)  (35)  Loj. v i m o s a Guille±. him saw to 'We s a w G u i l l e . ' ft  quieru  v i m o s t_± ?  ft d e t a i l e d  discussion of this  of  c a n be c l i t i c  NP t h a t  important cannot  t o note  be u s e d  particular  doubled  f o r t h e moment  constraint  on t h e t y p e  i s presented  i n 3 . 3 . What i s  i s that  a constraint  t o argue d e f i n i t i v e l y  such  against  a  doubling  analysis support found  of such  in  an  analysis  (£ia)  ,  (£3),  of  dislocation  (13)  i n the  and  the  ,  doubling  of  the  (29a) , and  and  structure  the  that  as  in  in  (30a)  an  in  property  like  (34).  (13),  (18),  are a l l  obligatory  appeal  (15),  fact,  similar  sentences  interpretation necessitates of  i t can,  constructions  C.Fr.  (£8a)  that  light  doubling  that  (£4),  subject  modifications (£7).  in  object  I conclude  instances  or  sentences  "attested"  Thus, (£0) ,  the  to  left-  further  (£lb,c),  (£5),  (£6),  &  The  structure  the  one  in  the  a n a l y s i s of  (11),  (36)  associated repeated  subject  in  with  subject  (36),  clitics  and  doubling  which  introduced  i n C.Fr.  i s consistent  i n Chapter  is with  £.  IP /  \  IMPi  INFL'  I  x  /  \  INFL / \ 06R TNS  VP  I  Cl i With  respect  Vanelli sirnili anche (1973)  to  &• R e n z i varieta  this  (198£:134) w r i t e s CNP-pronoun—verb]  i l francais and  conclusion,  populaire"  Meyer-Lubke  i t i s interesting to " Pinch e (...)5  f o r which  (1889).  i l francese e cosi  they  si  note  that  present a  comporterebbe  refer to  Guiraud  3. 1.3. £.  Pied  Noir  The s e c o n d often  French  dialect  r e f e r r e d t o as Pied  The s i t u a t i o n of  French  in this  the facts already  subject  doubling.  t o be c o n s i d e r e d  Noir  French;  dialect  introduced  This  i s t h e one  c f . £ ( f o o t n o t e 3)."''  i s rather  f o r C.  i s illustrated  here  simple  French.  i n the  PN F r e n c h  light has  i n (37).  ( 3 7 ) a. L' hornrne ( i 1) v i e n t . ' T h e man ( h e ) c o m e s . ' b.  This to  doubling  i s o p t i o n a l as i t i s i n C.Fr. but i t i s c l o s e  the N.It. dialects  This  i s studied  (38)a.  Pis  L e s o l e i l ( i l ) b r i l l e p o u r t o u t l e monde. 'The s u n ( i t ) s h i n e s f o r e v e r y b o d y . '  i n that  extensively  any t y p e  o f NP  c a n be  i n 3.3.  Un hornrne i l v i e n t . ' Pi man ( h e ) c o m e s . '  b.  C h a q u e fernrne e l l e p a r l e . ' E a c h woman ( s h e ) s p e a k s . '  c.  Brua (1938s11) P e r s o n n e i l s a i t q u i c ' e s t l e u r mere. ' N o b o d y ( h e ) k n o w s who t h e i r m o t h e r i s . '  i n RP S p a n i s h ,  (39)a.  b.  doubled.  there  i s also object  Marie  I t ' aime a Jeani. h i rn 1 i k e s t o 'Marie l i k e s Jean.'  La fille the g i r l 'The g i r l  l i 'a vu a Jean*, him has seen t o has seen Jean.'  doubling  as i n (39).  170 There  i s also c l i t i c  (40) a.  J e l u i t a i donne 'I  b.  The  with  indirect  l a pomme a u  gave t h e a p p l e  more d e t a i l subject  3.1.4.  doubling i n 3.£.  objects  a s i n (40).  professeur*..  t o the teacher.'  Tu 1 u i i a s donne d e s f l e u r s a 'You gave f l o w e r s t o M a r i e . '  object  property  Marie . t  o f PN F r . w i l l  I now t u r n  be d i s c u s s e d  to the specific  in  analysis of  doubling.  Analysis  The  crucial  doubling shown  doubling  fact  that  will  i s the impossibility  guide  my a n a l y s i s o f s u b j e c t  of free  i n v e r s i o n i n C.Fr. as  i n (41>.  (41) a . * E s t p a r t i J e a n . is gone ' J e a n i s gone.' b. * A r r i v e 'Marie Recall  that  property  Marie. arrives.'  Safir's  (1985)  i n Trentino  influenced  account  and t h e o t h e r  by t h e f r e e  the subject  to  i t seems n a t u r a l  subject adjoined  clitic  can, i n turn,  t o and d i r e c t l y  particular  definition  clitic  dominated  o f c-command,  exhibit.  i n s i d e VP a n d a s s i g n i n g  to Safir  assign  doubling  N. I t . d i a l e c t s i s  i n v e r s i o n c o n s t r u c t i o n they  When g e n e r a t i n g i tdirectly,  of the subject  t o assume  that  Case t o an i n v e r t e d b y VP s i n c e , the subject  Case  this NP  under a clitic  governs  171 the  i n v e r t e d NP. •n  the  requires and  the  hand,  preverbal  possible  i s no to  to  type  of NP  of  account  a chain  i n h e r i t a n c e of  an  analysis that  in order  Case t o  a n a l y s i s leads  a  the  i n languages with  subject  clitics  such  inversion  that the  directly  w i t h i n VP.  general  as  St.It.  proposed  does not  clitics;  postverbal  absorb  Case;  a  Case.  The  It should  NP.  free  and  thus  require NP  be  an  to  empty  inherit  Recall also  that  (silent)  i n v e r s i o n and  no  overt  Spanish.  particular to  subject  get  can  clitic  3.I.E.  p o s t u l a t i o n of  make a p p e a l  I n £.£.5,  cf.  preverbal  postverbal to  doubling  between the  does not  f o r the  clitics  I have  subject  (coindexation)  subject  I n £.4,  of  i n v e r s i o n i n C.Fr..  i n VP  assign  this  and  free  suggest  clitic  C a s e and this  other  formation  There  subject  53  i t was  a n a l y s i s of  empty  preverbal  nominative  suggested  f o l l o w i n g account  free  that  i s based  subject  Case clitics on  these  prem i s e s . Let subject  us  assume,  clitic  as  we  i s base  have done so  generated  (4£) . (4£)  IP \  INFL' /  INFL / \ AGR TNS C1 i  \  VP  far, that  i n C.Fr.  u n d e r PiGR i n s i d e I N F L a s  the in  in  Given  t h e Case  assignment that  filter,  t o t h e NP w h e n  n o m i n a t i v e Case  assume t h e f o l l o w i n g  (43) If  Pi s u b j e c t  the c l i t i c  possibility be  the question that  This  parameter  clitic  need  f o r t h e NP t o h a v e  phonetically  c. * J e a n d. *  Case,  Case.  t h e n t h e r e i s no  lexical  filter.  content since  i t would  In order f o r a lexical  t h e r e must  be no s u b j e c t  i n Standard  French,  clitic  w h e r e we  NP t o b e i n PiGR.  gets  comes.' i l vient.  Vient.  If  the c l i t i c  be  a s s i g n e d t o t h e NP w h i c h  so  even  under  process  does not absorb  an approach  i s optional.  no l e x i c a l  full  I  II vient. ' He  a  realized. Considering  Jean vient. ' J e a n comes. '  b.  then  Case  t o be m o d i f i e d l a t e r on.  not absorb  i n (4£) a b s o r b s  i s the situation  ( 4 4 ) a.  concerns  i s a s s i g n e d by t h e (+)TNS e l e m e n t ,  e x c l u d e d by t h e Case  realized  i t i s lexically  remains  element  appearing  would  then t h i s  therefore  t o Case assignment  Suppose that  NP o r a s u b j e c t  category  Case,  Case  can appear clitic.  i n CNP,IP3?  What  C a s e c a n and must  be l e x i c a l . i n which  This i s this  i s not assigned at a l l ,  i n preverbal then would  I t cannot  position,  be i t  be t h e e m p t y  be a t r a c e  since the  position  i s not p r o p e r l y  is  governed.  is  not  Therefore,  lexical  NP  realised.  clitic We  subject  (45)a.  II vient.  c.  Jean  d. *  Vient.  since NP  always absorb  clitic  in  i n the case  a  where  t h e NP i s  manner f o r t h e o p t i o n a l where  pro  e i t h e r pro or  only  we  status  get.  of a subject  patterns  i n ( 4 3 ) . The o f NP  that  With  respect  Case)  position.  (43) with  i s fixed  i s t h e one  an  will  fact  associated hand,  a preverbal  lexical  positively.  i n the setting them  of the with  be c a p t u r e d  i t must  where  negatively  on t h e o t h e r  and  d i f f e r e n c e s between  t o l e a m a b i 1i t y ,  situation  clitic  c a n be d o u b l e d  i s  on t h e b a s i s o f t h e  I n C. F r .  C.Fr.  St.Fr.  (43) i s f i x e d  always appears without  i s attested,  type  unmarked  then  i n C o l l o q u i a l French  i n preverbal  French  parameter  flGR,  i n (43). In St.Fr. ,  co-occurrence  subject PN  the  as  clitics  NP  Of c o u r s e ,  t h e d i f f e r e n c e between C.Fr. and  a subject  lexical  by a c l i t i c  of  i l vient.  parametrised (subject  only  since i t  occurrence  i s generated,  in  in this  this  b e PRO  vient.  b.  I n summary,  i s done  i s generated  doubling  i t cannot  p r o but t h e n  i n CNP,IP3.  can account  Jean  and  when a c l i t i c  can appear  no s u b j e c t  that  be  l i c e n s e d because t h i s  French.  of  I t can only  governed  be assumed  the c l i t i c  respect i n 3.3. that  absorbs  the  Case.  to  174 Contact the  with  I  the a n a l y s i s presented  now  argue that  accounts  f o r the optionality  a n d PN F r e n c h  extended  and t h e r e  following.  In Trentino  i s free  clitics  i n flSR. A c c o r d i n g l y ,  inversion occurs,  so that Case.  (46)  i n S t . French,  have s u b j e c t  a subject  the preverbal This  subject  must  Case  i n C. can  doubling  holds  clitics  the neuter  be r e a l i z e d  also  i sthe  phonetically,  i s always  the parameter lexical  possible to  NP,  generated i n  i n (43) i s s e t  when p r e s e n t ,  can  i n Fiorentino. are excluded  e appears  i n the free  free  i n preverbal  inversion  p o s i t i o n as i n  £.4.  E' v i e n e le ragazze. i t come—3s t h e g i r l s 'The g i r l s come.'  Crucially,  subject-verb  agreement  i s  postverbally, the subject  clitic  i n Trentino  clitic  doubling  i s assigned  c o n s t r u c t i o n b u t i n F i o r e n t i n o , when  (46) 5 c f . a l s o  doubling  c f . (£). I t i s t h e r e f o r e  D-structure  subject  which  i n v e r s i o n . The s i t u a t i o n  i n Trentino  Now,  that  i t i s assigned  do n o t s u r f a c e ,  assigned  above,  In Trentino,  assume t h a t  be  (43).  of subject  when n o m i n a t i v e  the subject  ( 1 ) , w h e r e a s when  positively  i s sufficient for  and F i o r e n t i n o .  obligatory  clitics  dialects  differences are clear.  preverbally,  in  and i t s i m p o s s i b i l i t y  t o the N.It.  as Trentino The  cf.  constructions  o f t h e parameter  French  such  doubling  positive setting  readily  be  subject  holds  with  p r o i n CNP,IP]  licensed is  by  the  consistent  structure  neuter  with  in  (47)  (47)  clitic,  the  not  approach  with  the  developed  corresponding  to  i n v e r t e d NP.  here.  This  Consider  the  S~  (46).  IP /  \ NP± INFL' I / \ pro INFL VP I / \ AGR VP I I CU TNS+V In  (47),  clitic  TNS  assigns  does not  licenses  pro  not  ©-marked,  or  impersonal)  so  pro  (1981)  clitic  paradigm  and  others).  This  does not  agree  The  preverbal  with  NP an  postverbal (43)  subject  and  NP,  the  it  position is  expletive  (or  neuter,  can  simply (1984),  assumption the by  assume, that  to  the  there  the  contrast  is a  neuter  i s supported i n v e r t e d NP  following Brandi  by  the  gap  clitic fact  as  i t does  in  (48).  in  & the  (among that  the  verb  in Florentine.  E' v e g n u 'na p u t e l a . h a s come a girl 'A g i r 1 c a m e . '  b. * L ' e i (48a)  i n compliance  corresponding  with  the  appears.  Rissi  is illustrated  ( 4 8 ) a.  In  we  Case t o  i s e x p l e t i v e and  clitic  Cord i n  Case  CNP,IP3.  In T r e n t i n o ,  This  nominative  absorb  in  NP  vegnuda  then,  as  'na  putela.  in a l l other  instances  of  free  inversion in  Trentino, none  no  would  be  this  the  Standard  an  empty  last  point  of  presence  believe  there  that  assumption the  in  concept  i t must a  preverbal  since  The  verb  features can  thus  position that i f there  in Trentino.  features  the  were  Pro  in  as  i t always  be  considered  is as  d e l i c a t e i s s u e r a i s e d by clitics.  Safir's  Italian  approach  to  of  silent  clitics  here.  some S t . I t . subject  "learn" the  resort to  This  such  case  clitics.  existence  which,  as  the  i s the  one  was I  Of  presence  a  of  the  course,  (10) of  of  order subject  Safir,  above.  object  none  in  silent  to  to  language  Therefore,  according in  this  d i f f e r e n c e has  sentences.  UG  concerning  d i f f e r e n c e between  here.  the  i n 3. 1. £  However,  l e a r n a b i 1 i t y . Suppose t h a t  contains  in this  Recall that  (1985) a s s u m p t i o n  developed  presented  of  statement  available  the  that  one  learner to  experience  The  of  is a significant  the  sentences  includes  the  and  up  (or s i l e n t )  the  i s exposed  clitic,  generated  dialect.  features  of  be  in Fiorentino)  set  set  brings  i n Standard with  the  (as  this  this  i n the  those  This  I have observed  learner  clitic to  empty  incompatible  for  a  Italian.  postulation  with  features  can  clitic.  This  these  as  clitic  grammar o f  i s l i c e n s e d by  in  do  of  corresponding  case  above  set  realized  clitic  the  subject  i s a v a i l a b l e i n the  agrees with  a  preverbal  The  trigger  clitics  in  data.  situation paradigm  of  is quite different clitics  as  i n the  i n the case of  instances the  of  subject  gaps clitic  177 paradigm  i n Trentino.  In t h i s  particular  learner  i s exposed  clitics  i n t h e grammar and, a l s o ,  features  i n INFL,  corresponding statement In  like  this  Rizzi  is  the strong  that  there  AGR  trivial  my  parameter  o f AGR  *2)  n  1  d  sets of  here  t o t h e one  for a  INFL  for different  I disagree  subject  Italian  content  different  o f AGR;  also  in i t s this  values f o r  i s obligatory, or optional,  and a l w a y s Rizzi  impossible  phonetic or  and number i n  i n Standard  in h i s interpretation  w i t h i n t h e above d e s c r i p t i o n  and  only i n  i.e., the  o f person  clitic  language,  It i s natural to think of  specifications  with  proposed  the subject  Standard  receives phonetic  under  o f where  of the facts; cf.  .  To s u m m a r i z e , is  from  specifications  EN. I t . 1 d i a l e c t s ,  n  no need  i s closer  as p o s s i b l y assuming  spell-out  fits  are subject  realisations  pro i n a null  position.  grammatical  French  there  for specific  i s thus  systems d i f f e r  syntactic  Italian".  that,  " I n t h e EN. I t . ] d i a l e c t s  different  the  position  specification  impossible  There  licensing  grammatical  abstract  that  t h e t h e one i n( 1 0 ) .  (1984:9):  AGR  evidence  a r e no p h o n e t i c  to a clitic.  respect  in  these  t o concrete  instance, t h e language  always  positively  generated —  i n my  account  i n D-structure  as i t i s i n C.Fr. —  between C.Fr. and T r e n t i n o inversion  of Trentino,  i n the latter.  i n AGR  clitic  (43) i s s e t  but a major d i f f e r e n c e  f o l l o w s from  Base  and  a subject  generation  the possibility  of free  o f t h e NP s u b j e c t i n  178 postverbal  position accounts  agreement  between  inversion  constructions  subject  clitics  the verb and  in this  f o r the fact  and  this  this  pro  in external  argument  (or  impersonal)  clitic  postverbal  follows  dialect.  that  The  from  verb  p o s i t i o n which  in Fiorentino  there  i s no  NP  free  in  the presence  agrees with  i s licensed  of  expletive  by a  neuter  or a set of features  in  Trent ino. The  differences  between  St.Fr.,  on t h e o t h e r  setting  of the parameter  distinguished of  free  inversion  clitics. with  C.Fr.  respect  explored  3.2.  developed  adopt  and  differs  from  F r . , on one  a difference  from  PN  and  Fiorentino  obligatory  to extraction  Doubling  £.£.3,  clitics  follow  PN  i n (43). C.Fr.  Trentino and  and  PN  by t h e  Trentino,  possibilities,  and  i n the  Fr. are  base generation  French,  hand,  a topic  possibility of  subject  and to  Fiorentino be  i n 3.3.  Object In  from  hand,  C.Fr.  here  i t was also  in a variety  the structure  and  Subject  argued covers  that  Doubling the theory  the constructions  of languages. i n £(41),  This  repeated  of subject with  object  i s made c l e a r i n (49).  clitics  i f we  <49)  IP /  \ NPj. INFL' I / \ pro INFL VP / \ / \ HGRi  TNS  I  V  NPj  1 Clj+V  Cl  I pro  M o r e p r e c i s e l y , i t c a n be d e r i v e d Principle  that  when t h e r e position. the  an o b j e c t  i s a base  generated  The o b j e c t  clitic  —  clitic  from  must  the  Recoverabi1ity  be a f f i x e d  empty c a t e g o r y  p o s i t i o n i s coindexed  i n fact,  the verb  —  i n object  with  which,  t o t h e verb  and governed  i n turn,  by  licenses the  em p t y p r onorn i n a 1. There the this  i s t h u s one major  licensing process The  goal  o f an empty argument operates of this  section  with  a lexical  the  type  of discussion that  argument.  constructions i n 3.1  constructions.  In other  i t t o object  The r a t i o n a l e b e h i n d  two  realizations  o f which  that  namely  subject  this  clitics  that  condition.  I want  i n £.£.3  t h e account  as f a r as  subject/object  words,  c a n be f o u n d  and a p p l y i n g  argued  function  i s t o explore  by u s i n g  have a l r e a d y distinct  i s concerned,  i n the r e l a t i o n s h i p that  has  introduced  symmetry  u n d e r t h e same s t r u c t u r a l  s y m m e t r i e s and a s y m m e t r i e s  doubling  subject/object  a  clitic  to  to  of subject  apply  clitic doubling  doubling  should  be c l e a r .  and o b j e c t  I  clitics  are  o f t h e same s y n t a c t i c e l e m e n t , t h e  i s t o l i c e n s e p r o i n argument  position.  Consequently, the in  argument  position  the relevant  internal is  respects  the relationship  doubled In  clitic  lexical  already  respect I  and B i n d i n g  with  lexical and  position  i s  explored  argument  external  one  look  framework,  qualification.  d i f f e r e n c e s should  t o government  here  and i t s clitic  I f there  are  be e x p l a i n a b l e  i n terms  asymmetries, e s p e c i a l l y  o f t h e argument  with  positions involved.  at the s t r u c t u r e s i n v o l v e d  and t h e n  the predicted  i n the  two  t u r n t o t h e problem o f Case assignment.  Structures  This (49)  i s a l w a y s t h e same  to the question  a doubled  and  argument?  these  constructions  3. £. 1.  leads  known s u b j e c t / o b j e c t  first  with  t h e same f o r i n t e r n a l  i s positive  differences,  between a c l i t i c  whether t h e argument  between  t h e Government  answer  relationship  i t i s associated  or external. This  associated  of  the licensing  i s a relatively  i s a p l a u s i b l e one  doubling  i s a s shown  (50)  simple and  matter.  i f the structure f o r subject  i n ( 5 0 ) , where  IP /  \ NPi INFL' I / \ x INFL / \ flGRi TNS I Cl  VP  I f the structure i n  x_ i s  lexical,  181 then  the structure  f o r object  doubling  s h o u l d be  t h e one  in  (51).  (51)  IP /  \  NP /  INFL' \  INFL  VP /  \  V I Clj+V It  i s an  (51)  is a valid  major all  affixed  (53)  1 0  The  t o determine whether  of the c l i t i c  does  In the examples  i n (52)~(55),  the c l i t i c s  i t s h o u l d not  Plate  the structure  position  t o t h e v e r b on  that  River  issue  one.  problems.  believe (52)  empirical  NPj I x  surface.  There  be  D—structure.  Spanish  a.  Vimos a Juan. saw-1p ' We s a w J u a n . '  b.  Lo vimos. c l saw-lp 'We saw h i m . '  c.  Lo  vimos  'We  saw  a Juan. to Juan.'  Roman i a n a.  ftm vazi.tt Popescu. have-Is seen ' I have seen Popescu. '  b.  L -am vazut. c l have-Is seen ' I have seen him.'  so at  i s t h u s no  not  in  raise  reason  are to  18S  c.  (54)  (55)  L-am  va21.it p e Popescu. Ob.j. M a r k e r ' I have seen Popescu. '  Pied  Noir  a.  Marie  b.  Marie  c.  Marie  aime Jean. 1i ke-3s 'Marie l i k e s Jean.'  1' a i m e . cl like-3s 'Marie l i k e s him.' 1'aime a Jean. to 'Marie 1 ikes Jean.'  Iraqi  Arabic  a.  Sif-it Xalid. saw 1 s ' I saw X a l i d . '  b.  §if~t - a . saw I s c l ' I saw him. '  c.  §if~t-a ' I saw  More the  French  1 -Xalid t o Xalid.'  problematic i s the position doubling construction.  object  i s , f o r obvious  of the lexical  I n t h e (a) s e n t e n c e s  reasons,  i n internal  CNP, VP3 b u t i n t h e ( c ) e x a m p l e s t h i s (19845 1985) Spanish  sentence  A'-position Strozer  defends  i n (5£c) , t h e c l i t i c  usually  (1978),  occupied  Rivas  (1977),  evident.  i n the River  doubled  NP J u a n  by r i g h t - d i s l o c a t e d Jaeggli  above, t h e  argument  i s less  the hypothesis that  NP o b j e c t i n  NPs  (198S31985) and  position Hurtado Plate i s i nthe whereas Borer  (1984) a r g u e position. accept  f o r the hypothesis  I do n o t r e v i e w  t h e second  CNP,VPD.  convincing  as  object  predicted,  3.1.4  Case  parallel  a clitic  construction  (1985)  NP  i si n  f o r some  o n i n t o n a t i o n , t h e u s e o f dummy and bounding  effects.  i n t h e examples  i n every  one t o  (5£)-(55) a n d i t  i s ,  respect  t o t h e one s u g g e s t e d i n  to a clitic  doubled  NP  i s at f i r s t  o f d i f f e r e n c e between s u b j e c t  though,  Jaeggli like  Case normally  (198£),  Chomsky  doubling  sight and  assigned  the c l i t i c  may  by t h e v e r b o r may  Case  t o be w e 1 1 - f o r m e d  hypothesis. that  position  t h e NP o b j e c t .  The i n t u i t i o n  a "9-role  P and i s t h e n  containing  This  chain  according  Crucially  Case and i t  must  to the  manifest Visibility  the V i s i b i l i t y  t o an element  inherited  a, b u t o n l y  this  C...1  a in a  hypothesis 0-marked  by a n a r g u m e n t  i fthe chain  in a  absorbs the  t o i t s complement.  underlying  i s assigned  the c l i t i c  not manifest  with  i n order  (198£) a r g u e s t h a t  t h e one i n (51) above,  forms a chain  chain  object  doubling.  Following  is  doubled  but I  Assignment  main source  object  arguments here  doubling.  Case assignment the  i s i n argument  t h e s t r u c t u r e i n (51) i s a v a l i d  doubling  f o r subject  3.£.£.  based  ( c f . 3. £.£'),  In c o n c l u s i o n , describe  NP  i s referred to Jaeggli  arguments  Case-markers  this  the various  hypothesis:  The r e a d e r  that  has Case"  i nthe  (Chomsky  184 1981:334). to  Thus,  the chain  receive a 9-role.  (cl,NP)  Therefore,  some o t h e r  order  f o r t h e NP t o r e c e i v e C a s e ,  since  t h e verb  Case  i t could  This  i s Chomsky's  Jaeggli  states  preposition  French,  has already  ( a dummy  grammars o f R i v e r and I r a q i  Arabic,  t o Case-mark  (56)  River  Spanish  Loj vimos a Juanj. ' We s a w J u a n . ' Pied  Noir  French  M a r i e l ' j airne a J e a n j . 'Marie l i k e s Jean.' (58)  Romanian L j - a r n v a z u t p_e P o p e s c u j . ' I have seen Popescu.'  (59)  Iraqi  do s o . I n  means must task  what  be u s e d  because t h e  by t h e c l i t i c . i sreferred t o generalization a  Case-marker).  preposition i sunderlined  (57)  must  NP a l w a y s a p p e a r s w i t h  This  Plate  i n order  does not manifest  been a b s o r b e d  t o t h e use o f a p r e p o s i t i o n i n order  this  Case  gneralization. This  a clitic-doubled  Romanian,  some o t h e r  (1982) way o f c a p t u r i n g  particular  recourse marker)  assign  (cl,NP)  accomplish  (1982) a s K a y n e ' s  that  The  of the chain  c a n no l o n g e r  manifest  i fthe c l i t i c  Case,  in  member  must  Arabic  !~>if-t-aj 1_-Xalidj. ' I saw X a l i d . '  P l a t e Spanish, f o ri n s t a n c e ,  Pied  a l l have  ( t h e s o - c a l l e d dummy  thec l i t i c  doubled  i n (56)—(59).  Noir  case-  o b j e c t NPs.  185  This  process  literature.  seems t o be w i d e s p r e a d a n d i s w e l l S i m i l a r processes  constructions  such as t h e construct  Iraqi  Arabic  Tigre  and Lebanese  (60)  (61)  (6£)  (63)  Hebrew  c a n be f o u n d  ( 6 1 ) , and o b j e c t  (Borer  Arabic,  b e i t - a j * (^el) ha —rnoraj house h e r of the teacher 'the teacher's house'  ?urnm - a mother c l 'his mother'  b.  ?umrn -aj *(l_)~Xalidj mother c l to ' X a l i d ' s mother'  i n Iraqi £.£.1.  Arabic  a.  wiya—ha with c l 'with her'  b.  w i y a - h a j * ( U -mara withcl t o woman ' w i t h t h e woman'  Lilat  (60) and  Arabic  a.  Tigre  i n Hebrew  1984)  b.  Iraqi  doubling  (6£)-(64); c f . a l s o  beit —a house c l 'her house'  (Jake  1980)  * (' i q a l ) ? a l i j to A1i ' L i l a t sent a l e t t e r  warakat nad'at 1et te r sent to A l i . '  i n the  i n other  of a preposition  a.  Iraqi  state  attested  'it-tuj. t o c1  1 1  Arabic,  186  <64) L e b a n e s e A r a b i c Hk i t talked-Is 'I ta1ked Now in  —aj * (l_a) - K a r i m j .  ma">  with with  c l to Karim.'  l e t us c o n s i d e r C. F r e n c h ,  (65)  (floun 1981)  t h e f o l l o w i n g examples o f subject  PN F r e n c h ,  C o l l o q u i a l French;  Trentino,  Pied  Noir  doubling  F i o r e n t i n o , and P i r a h a .  French  Jearu  i l l part. cl leave-3s 'Jean leaves. '  (66)  Trentino El Marid e l * magna, the c l eat-3s 'Mario eats.'  (67)  F i o r e n t ino Le r a g a z z e i l e * vengano. the g i r l s c l come—3p 'The g i r l s come.'  (68)  Piraha Koxoii 'Koxoi  It  ( E v e r e t t 1985) h i i gi xibaoba. cl you h i t h i t you.'  i s c l e a r from  generalization  t h e examples  does not appear  Case-markers on t h e doubled The no  question  the  we t h u s  dummy C a s e - m a r k e r In order  i n (65)-(68)  want  a r e n o dummy  i n these  languages.  i e  t o a s k i s t h e f o l l o w i n g ! why i s t h e r e  such  syntactic structures  Kayne's  as there  NP s u b j e c t s  i n subject  t o answer  t o hold  that  doubling  a question,  involved  constructions? I will  i n the four  first  /  look a t  doubling  187 construct ions  involved  generalization. some r e c e n t The object  I will  work  then  propose  by J a e g g l i  o f a noun, They  object  (69)  o f which  obey  a solution  (1985) and L e v i n with  ares  Kayne's  based  i n part  & Massam object  of a  three.  verb,  of a  i n (69) t o <7£). I a d o p t  of the first  on  (1985).  o f a p r e p o s i t i o n , and s u b j e c t  are represented  (1984) r e p r e s e n t a t i o n s  Borer's  1 3  VP /  \  v I Clj+V  NPj  (70)  N'' I N' / \ N +C l j Nj''  (71)  P" I P' / \ P +C l j N j "  <7£>  IP / NPj  \ I' /  INFL / \ AGR TNS I Cl ,  here  three  s t r u c t u r e s we a r e c o n c e r n e d  sentence.  The  here,  first —  markers  major  i s that  \  VP  difference — the three  i n ( 6 9 ) , (70) and  a n d maybe t h e o n l y  constructions accepting (71) i n v o l v e a d o u b l e d  relevant dummy NP  one  Case-  governed  by  isa a  lexical  head.  Before necessary  This  i snot t h e case  b u i l d i n g an account t o examine t h e exact  i n (7£) «  on t h i s  observation,  i t  i s  f u n c t i o n a n d s t a t u s o f dummy  Case-markers. It  seems o b v i o u s  devices", assigned out  by t h e Case  marker  functions.  help  This  somehow a l l o w  NP w h i c h  would  a s t o t h e way i n w h i c h  In fact,  this  consequences  otherwise  element  t h e dummy  be r u l e d that  Case-  c a n be e i t h e r a C a s e o f one o r t h e o t h e r  f o rthe status of c l i t i c s  t o Case a b s o r p t i o n .  i n opting  C a s e t o be  i sa descriptive? statement  o r a Case t r a n s m i t t e r . The c h o i c e  important  respect  they  doubled  filter.  no a c c o u n t  assigner  t h e dummy C a s e - M a r k e r s a r e " s a v i n g  means t h a t  to a c l i t i c  provides  has  which  that  with  These consequences should,  f o r one o r t h e other  hypothesis.  i n turn,  L e t us look a t  them. Recall and  t w o o f my b a s i c  subject  clitics  arejust  unique s y n t a c t i c concept. concerning argument  their  given  suggested  that  but  that  this  i s based  object  clitics  realizations  o n my  of a  observations  elements  f o r pro i n the  a r e r e l a t e d t o ; c f . £. £ a n d £ . 6 .  clitics  i sa parametrized  allowing  subject  and  Case c a n f r e e l y  this  This  t h e f a c t s about subject  First,  two d i s t i n c t  function as licensing  p o s i t i o n they  Secondly,  assumptions.  doubling,  subject absorb  nominative  option  subject  doubling,  (3.1.4).  clitics  be a s s i g n e d  need  I have Case  (£.£.5)  In languages not absorb  to a lexical  NP i n  Case  189 external If the  argument  position.  we p u t t h e s e  optionality  property  o f Case a b s o r p t i o n  of object  Now,  two assumptions together,  clitics  i f we a s s u m e ,  i s a Case a s s i g n e r  clitic  not t o absorb  be  no need  suspect case  clitic  a dummy  Case s i n c e ,  clitics  since  i twould these  absorb  option.  e x h i b i t s both  which  object  For  by a n o b j e c t  i f this  clitic  object  makes a d i s t i n c t i o n a s t o how t h e y exist  reasons,  even  happened,  Furthermore,  there  level. doubling  Finally,  i n a given  Case  look  between by a  recall  c l i t i c s and  rather  particular that  object  Case o p t i o n a l l y and that  this  i sa  i n (43) above c a n t h u s  be  and subject  between s u b j e c t  absorb  would  i t would  and Case a b s o r p t i o n  I assume here  The parameter  Case-  i t i s not p o s s i b l e f o r an object  at a cross-linguistic  PN F r e n c h  doubling,  in  then  that  i n t e r m s o f l e a m a b i 1 i t y t o make a d i s t i n c t i o n  subject  also  parametrised  (1984),  f o r a dummy C a s e - m a r k e r .  absorption  that  expect  also.  as i n Borer  marker  t o be a  we t h e n  strange  grammar. c l i t i c s can parametrised  be g e n e r a l i s e d  as  (73).  (73)a. b.  Unmarked Pi c l i t i c  Marked s e t t i n g : Pi c l i t i c n e e d n o t a b s o r b  With learner  setting: a b s o r b s Case.  respect will  Case.  t o l e a r n a b i 1 i t y , i tremains that  have t o e s t a b l i s h t h e v a l i d i t y  t h e language-  o f (73) f o r  each  190 clitic-paradigm other  words,  clitics sets  (subject,  o f P, e t c . ) .  In  f o r one s e t o f  the a v a i l a b i l i t y of doubling  for a l l  of clitics. move c a u s e s a p r o b l e m  suppose,  functions as subject  Case-markers need i s t h e case  not  o f V, o b j e c t  the a v a i l a b i l i t y of doubling  does not e n t a i l  This  as  object  n o t be u s e d  f o r object  doubling  doubling.  We p r e d i c t t h a t  i n object  doubling  doubling  constructions.  the r e a l i t y of object  doubling  which  a s shown  i n (56)-(64)  we now  dummy  constructions  i n subject  Case-markers  which,  But t h i s i s  seems t o r e q u i r e  and expressed  dummy  by K a y n e ' s  g e n e r a 1 i z at ion. One  o b v i o us s o 1 ut i on t o t h i s  p r o b 1 em i s t o c o n s i d e r d urn my  Case-markers a s Case t r a n s m i t t e r s . J a e g g l i view of  of the preposition a i n the River  object  "Case  doubling,  bridge"  assigned  t h e Case  to a c l i t i c  Such  for this  person  neuter  dative  Case,  doubling  comes  full  from  pronominal  Pis p r e d i c t e d ,  constructions  i saccusative  feature  direct  assigned this  view,  examples  a i sa kind of  o f a verb  t o be  object  i s accusative  by t h e p r e p o s i t i o n a, n a m e l y  p r e d i c t i o n i s borne outand  t h e use o f e l l o . form  which  e11a cannot  E11o i s a  can only  appear  (74a) o r i n s u b j e c t  i nthe first  a  an a n a l y s i s p r e d i c t s that t h e  doubled  J a e g g l i argues that  evidence  it  allows  not t h e Case normally  dative.  P l a t e Spanish  as i n (56). In t h i s  t o t h e NP o b j e c t .  Case assigned and  which  (1985) a d o p t s s u c h  and nominative  occur  third i n the  i n object  p o s i t i o n (74b) s i n c e i n t h e second. But  it  appears  (74c).  i n regular  The s e n t e n c e s  dative  assignment  a r e from  Jaeggli  environments  asi n  (1985s£5).  ( 7 4 ) a . # Y o lo± e n t i e n d o a e l l o ± . ' I understand i t . ' b. * E l l o e s e v i d e n t e / d i v e r t i d o . 'It i s obvious/fun.' c.  If  H a y q u e r e s t a r l e a e l l o l o q u e me a c a b a s d e d a r . ' I t i s n e c e s s a r y t o s u b s t r a c t f r o m i t what y o u h a v e j u s t g i v e n me.'  t h e p r e p o s i t i o n a i n (74a)  clitic  doubled  NP, t h e n  there  assigned would  ungrarnmat i c a l i t y o f t h e s e n t e n c e . argument markers worth  from  Jaeggli  (at least  noting  here  that  i t really  of  features  undesirable such  optional  a view,  morphologically saying not,  that  which  i tgives  a  convincing  of analyzing  like  to this  This  Making  sometimes  i sthus  dummy  Case-  It  i s  t h e one i n (73)  on seeing  clitics  as spell-outs  t h e Case  because,  itself  i to p t i o n a l  would  under  realized amount t o  i s t h e Case and sometimes i s  i srather counterintuitive. with  t h e problem  o f a dummy C a s e - m a r k e r  raised  by t h e o b l i g a t o r y  i n object  doubling  construct ions. The  real  problem  consists  i s  s p e l l i n g - o u t an  i sundesirable  i s i n fact  on a head.  theclitic  We a r e l e f t presence  i f a parameter  status.  thec l i t i c  This  for the  a s Case t r a n s m i t t e r s .  puts a s t r a i n  since  Case t o t h e  be no r e a s o n  i n favor  i n RP S p a n i s h )  adopted Case  (1985)  dative  i n fact  i n accounting  for the  difference respect  t o dummy  It that  i s proposed  The element  assigning  us  define  (73)  assigns  & Massam  both  This  f o r an element  a Case f e a t u r e  i st h e "normal"  Case t o an o b j e c t "absorption"  t h e C+CP)] f e a t u r e  thereby  doubling  clitic.  Now,  b e a C+CfiH  as stated  t o assign  pattern  and t h e a s s o c i a t e d that  i n (73).  when, Now,  l e t  parameter i n  a clitic  head,  (1985)  always  but only In languages  both  E+CD a n d E+CflU  a r e absorbed  by t h e  making  i timpossible  f o r t h e verb  t o assign  Case t o an o b j e c t object  with  C+CII a n d a C a s e  NP i t g o v e r n s .  of the lexical  t h e C+CII f e a t u r e  no d o u b l i n g  clitic,  doubling  (1985) a n d Massam  be f u l f i l l e d  have  C+CPlII.  t h e term  optionally  to  and o b j e c t  i n a more p r e c i s e way a n d s u g g e s t  absorbs  with  i n Levin  must  feature  a verb  doubling  Case-markers.  t w o c o n d i t i o n s must  case.  say,  between s u b j e c t  even though  only  i n order  C+CD c a n b e f r e e d for this  i t . In languages  from  absorption  C a s e t o be a s s i g n e d  with  by t h e  there  needs  i n t h e c o n s t r u c t i o n and that  i s what t h e  role  o f t h e dummy C a s e - m a r k e r c o n s i s t s i n . Pi dummy  case-marker  thus  provides  assigned  in  1 3  t h e C+CPlII f e a t u r e  t o a doubled  construct (71).  feature  i tgoverns  state  NP. *•* T h i s  necessary  c a n be e x t e n d e d  i n (70) and t o o b j e c t s  The s t r u c t u r e s i n (69)-(71)  the light  Case-Marker".  o f t h e preceding  f o r C+CII t o b e to the  of prepositions i n  a r e repeated  i n  d i s c u s s i o n . DCM s t a n d s  (75)-(77) f o r "dummy  (75)  VP / V / Clj+V C+CII  (76)  \  NPj / I DCM N P j C+CPiII  N'' I N' / \ N+Clj Nj'' C+CII / \ DCM Nj'' C+CM  (77)  P'' I pi  /  \  P+Clj C+CII  Nj'» / \ DCM Nj'' E+Cft]  In  t h e s e n t e n c e s where no c l i t i c  the This  lexical  head t h e n  i sillustrated  (78)  VP /  \  V C+CD C+CPlII (79)  NP  N" I N' / \ N C+CII C+CPI 1  N' '  absorbs  t h e C+Cft] f e a t u r e o f  Case assignment f o l l o w s t h e normal  i n (78)-(80).  path.  194 (80)  pi  i  p» /  \  P C+C3 C+CA3  Now,  with  N' '  respect  to the non-availability  markers  i n subject  suggest  two p l a u s i b l e and r a t h e r  one  consists  assignment that  carries  with  by a s s u m i n g  AGR provides nominative  following  (81)  Case assigner.  that  i s assigned  In a standard  t h e C+CA3  In other  i n £.4 we  feature  words,  b y TNS. T h i s  clitic,  first  Case  assumed  construction  and no s u b j e c t  The  Suppose that  TNS p r o v i d e s feature.  Case-  like to  accounts.  way. I h a v e a l r e a d y  case which  NP s u b j e c t  i s assigned  different  I would  the process o f nominative  t h e C+CH  consequences.  a lexical  AGR has t h e  a s i n (81)  t h e Case o f  t o t h e NP b y TNS.  IP /  \  NPi /  I' \  INFL / \ AGR* c+c:  In  a construction  as  i n (82),  since  constructions,  i n the following  this  that  AGR  i n modifying  TNS i s t h e n o m i n a t i v e  qualify and  doubling  o f dummy  VP TNS  IH-CAU  involving a subject  the c l i t i c  the c l i t i c  absorbs  i s not part  C+C3  clitic  and no l e x i c a l  but cannot  o f t h e TNS n o d e .  absorb  C+CA]  NP  .195  <ae>  IP /  \  NP i I' i / \ p r o INFL VP / \ fiGRj. TNS I C+CPi] Cl In  (8£),  familiar In  the  position  r e a s o n s and  because  languages with  parameter does  CNP.IP]  i n (73)  t o some  illustrated  subject  C+C]  be  filled  p r o does not need doubling,  i s operative  not absorb t h e  assigned  can only  f o r subject  material  pro f o r the  Case.  i . e . where  f e a t u r e o f PiGR a n d  lexical  by  the  clitics, Case  i n CMP,IP].  the  can  clitic  be  This i s  i n (83).  (83)  IP /  \  NPi /  I' \  INFL VP / \ fiGRi TNS I C+Cfl] Cl C+C] In o t h e r  words,  a subject  clitic  (optionally)  PiGR b u t c a n n o t a b s o r b C+Cft] i n TNS assigned  to the subject  available, clitic opt i o n .  does  namely  position  i f (73)  not absorb t h e  as  i s fixed C+C]  and  Case  long  a b s o r b s C+C]  can p o t e n t i a l l y  as a  C+C]  positively;  feature.  1 6  in  This  be  feature i s  i.e. i f the i s the  first  196  The Let  second  option  us assume t h a t  assign head.  dummy C a s e - m a r k e r s ,  or transmit This  (69)-(71)  Case  i s t h e case,  inserted  i n front  position  then  structural  stipulative.  prepositions, can  a r e governed  not assign  by a  under  transmitters, note that  only  prepositions  a preposition  not appear  i n a governed  Case  t o t h e NP. I n o t h e r  doubling.  i s d i c t a t e d by  be a c c o u n t e d f o r i n a  there  o f dummy c a s e - m a r k e r s a s  a r e no " p o s t p o s i t i o n s " b u t  i n t h e languages under  subject  doubling  Since  a conception  Case  the  assignment t o t h e subject i n  c o n s t r u c t i o n s must  way. E v e n  lexical  i nthe structures  i n subject  t o a dummy C a s e - m a r k e r  c o n d i t i o n s and Case  exhibiting  Mow,  i squite different.  i t would  doubling  different  for  like  a s we h a v e s e e n ,  o f CNP, IPII w o u l d  t h e appeal  subject  i f they  a r e generated.  thesituation  words,  only  b u t more  i n w h i c h V,N, a n d P g o v e r n t h e p o s i t i o n w h e r e  dummy C a s e - m a r k e r s (7S),  i s less complicated  There  study  i stherefore  a p r e p o s i t i o n i n (84) t o t r a n s m i t  t h e Case  here and no  possibility  released  by t h e  clitic.  (84)  IP / NP /  (*DCM)  The  only  \  I' \ / \ NP I N F L VP / \ PiGR TNS I C+Cfl] Cl C+CD  other  m e a n s f o r a d o u b l e d NP t o g e t C a s e  would  be  through  the  regular  languages where Case  i n the  CNP,IP3  (73)  (73)  when a  absorbs the  i s set  clitic  nominative  positively,  whether  C+C3 f e a t u r e  (85)  of  i s set  s a m e way  language where to  process  a  clitic  of  a subject  i s present  negatively,  i s present  Case assignment.  as  NP or  receives not.  no  case  can  in  (85)  since  In  be  In  a  assigned the  clitic  OCR.  IP / NP I pro  \  i  /  I' \  INFL VP / \ PiGRi TNS I C+CHII Cl  This  hypothesis  initial  question  necessary  subject  in,  of  u n p r o b l e m a t i c and  determining  in object  doubling  words,  i s not  doubling  constructions. clitics,  why  say,  can (69)?  why the  i s i t not  Such  account so  process  a question  us  back t o  dummy C a s e - m a r k e r s  constructions  I f the  regular  why  brings  but  above  for object of  not  in  are subject  is available for clitics?  In  Case assignment  f i n d s no  the  answer  other  not  i n the  '  apply  second  account. I  thus  leave  the  further  research  that  the  options.  For  two  to  find  a  in  (85)  but  question would  with  allow  example,  language with  open  f o r the one  i n the  a sentence  to  moment  make a  second  pending  choice  account,  structure similar  postpositions instead  of  between  i f we to  the  p r e p o s i t i o n s , and  were one i f  198 this the  language? h a d o b j e c t p r e d i c t i o n i sthat  subject  doubling  doubling  and s u b j e c t  dummy C a s e - m a r k e r s  but not i n object  doubling,  would  doubling  then  be used i n  as illustrated i n  (86) .  (86)  IP /  \  MP* /  \ / DCM I N F L / \  NP  fiBRi  and  and o b j e c t  that  V  NPj  C+Cfl: I / C l j + V NP  conclude this  clitics  VP / \  TNS  I Cl C+CD To  INFL' \  section,  clitics  thedistinction  \ (*DCM)  i t i s assumed  do n o t have  between  that  intrinsic  subject differences  t h e s t r u c t u r e s i n (69)-(71)  w h e r e dummy c a s e - m a r k e r s  a r e o b l i g a t o r y a n d t h e o n e i n (7£)  w h e r e dummy c a s e - m a r k e r s  a r e never  independently To  do so,  the  3.3.  two o p t i o n s  unique  object  required  parameter  clitics. " x  structural  i n (73) v a l i d  must  follow  subject/object  a r e presented  which  allow  from  asymmetries. us t o m a i n t a i n  f o r both s u b j e c t  c l i t i c s and  7  Extractions We now h a v e a b a s i c  phenomena w h i c h phenomenon left  used,  with  dialects  account  encompasses t h e d i f f e r e n t  illustrated one major  studied  parametric  i n (75),  (76),  namely  instances  of the  ( 7 7 ) , a n d ( 8 3 ) . We a r e  source o f variation  here,  of doubling  among t h e l a n g u a g e s a n d  extraction possibilities  outo f  199  doubling  constructions. This  attention Borer  i n the literatures  (1984),  fit f i r s t , doubling that  i s an a r e a  Roberge  cf. Jaeggli  (1986b)  when o n l y  among  i n River P l a t e Spanish  Later  languages, assumed,  for certain very  aim o f t h i s  doubling  i n previous  unified  account  will  In  the f i r s t  Romance  argue  of c l i t i c These data  accounts.  such  from  previously doubling  ones. data  from  NPs w i t h  a r e compared  subject respect  with  to  t h e ones  i s t o provide  Pi c l o s e e x a m i n a t i o n  of  can provide  of lexical  more  NPs t h a t  these  a  c a n be  clitic  languages.  part  a review  of this  section, the data  from  o f some o f t h e p r e v i o u s  i n 3.3.2. accounts  o f e x t r a c t i n g out o f property.  i n terms of c-chain  seven  and t h e d i f f e r e n t  f o r a r e summarized  must be a p a r a m e t r i z e d  an account  than  The u l t i m a t e g o a l  the (im)possibi1ity  constructions  assumed  e x t r a c t i o n from  doubled  an account  o f the type  t o be a c c o u n t e d  with  that  section,  intricate  languages/dialects are presented  properties proceed  more  t o t h e RP S p a n i s h  of the facts.  i n various  i t was  c o n s t r u c t i o n s w e r e t o be  section i s to introduce  reveal that  characterization doubled  similar  possibilities.  studied  examples o f object  were examined,  languages allowed  and t h e b e h a v i o r  extraction  facts  proved  much  (1982;198451985),  t h e i n t r o d u c t i o n o f more d a t a  the situation  constructions The  on, w i t h  has received  others.  the particular  e x t r a c t i o n out o f doubling  barred.  that  then  and  doubling  In the  i s proposed.  I  final  a  £00  3.3.1.  The  3.3.1.1.  Facts  River Plate  Jaeggli cannot shown  (198£) p o i n t s  be e x t r a c t e d in  ( 8 7 ) a.  Spanish out that  a clitic  doubled  b y Wh-movement  i n River  Plate  object  Spanish  NP as  (87).  L d vimos a Bui 1l e i . cl saw-lp t o ' We s a w G u i l i e . '  b. * Pi q u i e Y i t l o i v i m o s t _ i ? t o whom c l saw-lp 'Whom d i d we s e e ? ' c.  Pi q u i e n i v i rnos t _ i ?  Furthermore,  extractions  also  The q u a n t i f i e d  barred.  sentences (88)a.*  taking  place  i n Logical  NP o b j e c t s  i n (88) undergo Q u a n t i f i e r  i n the  Raising  Form  ungrammatical (QR) a t L F .  Las* v i a todas l a s c h i c a s i . cl saw-Is t o a l l the girls ' I Sciw a l l t h e g i r l s . '  b. * L a s * e n c o n t r e a algunas cl met—Is t o some ' I met s o m e women.'  mujeres*.. women  c. * L d v i a un c h i c o . cl saw-Is t o a boy ' I saw a boy. ' t  3.3. :L.£„ C o l l o q u i a l Subject  doubling  (LF) a r e  French i n C . F r . o b e y s t h e same t y p e o f  l < 3  restriction from  as object  subject  subject  doubling  position  clitic  (89) a . * Q u i i who 'Who  i.e. extraction  i s i m p o s s i b l e when t h e r e  i n t h e same c l a u s e .  This  i s a coreferent  i s illustrated  in  (89).  i i iest alle? c l i s gone went?'  b. *  C C P ' Wh i  Cipa t i c l i +V  c.  Qui e s t  alle?  d.  E c p Whi  llipa t i V ...  There  i n RP S p a n i s h ,  i s also,  definiteness  parallel  clitic.  ( 9 0 ) a.  La f i l l e elle the g i r l c l 'The g i r l w i l l  33  t o t h e RP S p a n i s h  requirement  subject  ... IIII  on t h e l e x i c a l  example i n (88c),  NP a s s o c i a t e d w i t h  a a  va chanter. go-3s t o + s i n g sing.'  b. * U n e f i l l e e l l e v a c h a n t e r . a 'Pi g i r l w i l l sing.' c. The  string  ( 9 1 ) a. b.*  If and,  Une f i l l e v a  i n (91b) must  dp. [  chanter.  I P  t h e r e f o r e be  NPi  cU+V  ...  3  =(90a)  +def NPi -def  cli+V  ...  3  =(90b)  i t i s true that as such,  must  excluded.  indefinite  NPs a r e q u a n t i f i e d  u n d e r g o QR  at LF then,  expressions  i n (91b),  t h e NP i s  extracted  f o r scope  i n t e r p r e t a t i o n requirements.  The s t r i n g i s  therefore s  <9£) As  * NPt  [  expected,  I  t± c l i + V  P  regular  ... 1  q u a n t i f i e d NPs a r e a l s o  constructions.  barred  subject  doubling  (93)a.*  Quelques g a r s i i l s i ont trop bu a+few guys c l have-3p too+much d r u n k 'Pi f e w g u y s d r a n k t o o much yesterday.'  from  1 9  hier. yesterday  b. # P e r s o n n e l i l l a i m e travailler. nobody c l l i k e — 3 s to+work 'Nobody l i k e s w o r k i n g . ' c. * T o n s l e s h o m m e s i i l s i a i m e r i t all t h e men c l like—3p ' P i l l men l i k e e a t i n g . ' d. * N ' i m p o r t e q u i i i l l p e u t anybody c l can 'Anybody c a n do t h a t . ' The  syntactic representation  the  one i n  (89b).  What  section (89a)  ( 9 £ ) , wh i c h  i s that  the  that  QR  as well  constraint  structure River  (90b) and  to extract  Plate  f o r the sentences  here  similar  i n (93) i s a l s o t o t h e one i n  and f o r t h e r e s t  (93) a r e e x c l u d e d (1982),  a clitic-doubled a s Wh—movement  NP.  of  this  f o r t h e same  reasor  i t i s generally Furthermore, the  i s excluded  f o r blocking  as LF, at l e a s t  Spanish.  very  out i n J a e g g l i  responsible  as well  f a i r e ca. make this  i s it se1f  t o suggest  i s . As p o i n t e d  impossible fact  I want  manger. to+eat  suggests  movement h o l d s  i n C o l l o q u i a l French  that  a t Sand  I will (88c)  and  continue (90b)  3.3.1.3,, P i e d  subject they  both  quantified C.Fr., found  Wh  be  words or  in  very  But  PN  clitic  Fr.  Noir  phrases are  French the  differs  doubled  exhibits  C.Fr.  from  in subject  excluded.  in  The  one  C.Fr.  in  in  a that  that  p o s i t i o n . Pis i n  relevant  data  is  (94).  Un hommet i i i v i e n t . a ' Pi man comes. '  c.  P l u s i e u r s femrnesi e l l e S i many women cl ' M a n y women c a m e . ' Personne i l s a i t qui nobody h e k n o w — 3 s who ' N o b o d y k n o w s who t h e i r  e. * Q u i i  iii  preliminary  constraint  movement  sont have-3p  venu. come  c' e s t leur i t be~3s t h e i r mother i s . '  mere, mother  vient? conclusion  against  construction  Now,  Pied  s i m i l a r to  b.  d.  effect  L' homrnei il± vient. t h e man cl come—3s 'The man c o m e s . '  ( 9 4 ) a.  The  property  can  definiteness  "no-quantifier effect".  ( 3 . 1.3.12),  optional.  NPs  a  the  French  earlier  doubling  are  assume t h a t  i s in fact  Noir  Pis s t a t e d  to  holds  in  light  e x t r a c t i o n out  of  of  subject  only  at  S-structure  not  QR  (94b,c,d),  (94e),  but  a very  i n t e r e s t i n g aspect  of  these  i n PN is PN  facts  i s that  the  doubling Fr.  since  only  Wh-  blocked. Fr.  syntax  referred  to  in  footnote? 8 becomes e x t r e m e l y  subject  doubling  dialect. that  We t h u s  subject  and o b j e c t  realizations  doubling  have a d i r e c t  doubling  r e l e v a n t here.  doubling  t h e s a m e way. R e c o r d i n g  to this  on  extraction  doubling  from  subject  t h e same way i t d o e s  our hypothesis  a r e two d i f f e r e n t  hypothesis,  t h ec o n s t r a i n t  i n PN F r . s h o u l d  i n object  i tshould  object  c o n s t r u c t i o n s w h e r e Wh w o r d s w o u l d  doubling  (96).  (95)a.  That  this  be p o s s i b l e t o f i n d  i strue  apply i n  doubling constructions.  More p r e c i s e l y ,  disallowed.  i sillustrated  quantified  NPs i n  be  i n (95)  and  a o  Jean  1i'a vu a plusieurs c l h a v e - 3 s s e e n t o many ' J e a n s a w many p e o p l e . '  personnes*. people  b.  Leurs complices leSi ont accuses a their accomplices c l have~-3p a c c u s e d t o t o u s 1 e s voleurs±. all the thieves ' T h e i r a c c o m p l i c e s have accused a l l t h e t h i e v e s . '  c.  II l e s enterre a tous au cent metres, he t h e m b u r y — 3 s t o a l l a t + t h e h u n d r e d m e t r e s 'He b e a t s t h e m a l l a t t h e o n e h u n d r e d m e t r e s . '  d. * A q u i i que t u l i ' a s vu t_i? t o whom t h a t y o u c l h a v e ~ £ s s e e n 'Who d i d y o u s e e ? ' e. * A q u i i q u e t u le± f r a p p e s t _ i ? t o whom t h a t y o u c l b e a t ~ £ s 'Who a r e y o u b e a t i n g ? ' Naturally,  both  o f t h e same phenomenon a n d a r e t o be a c c o u n t e d f o r  in  exactly  that  are attested i n this  way o f t e s t i n g  and o b j e c t  Recall  n o n - q u a n t i f i e d NPs a r e a l s o a l l o w e d  t o be  clitic  doubled. ( 9 6 ) a.  Compare  (96)  U'a vu c l have-3s seen ' M a r i e saw J e a n . '  a Jean*.. to  Tu l e i f r a p p e s au petit you c l beat-£s t o + t h e l i t t l e 'You b e a t t h e l i t t l e guy.'  Furthermore,  PN  (95a,b, c) .  Marie  b.  the  to  case Fr.  i n RP  as  i t i s interesting  Spanish,  illustrated  garsi. guy  t o note  a clitic-doubled  by  the  that,  NP  discourse in  can (97).  contrary to be  focused  in  a i  ( 9 7 ) tts S a v a i s ~ t u q u e s o n c o m p l i c e a a c c u s e P i e r r e ? 'Did y o u know t h a t h i s accornp 1 i c e a c c u s e d P i e r r e ? ' B:  If the  a rule  S o n c o m p l i c e l±'a a c c u s e a JEAN*.. H i s a c c o m p l i c e a c c u s e d JEAN.' of  that  quantified  My  LF  i n PN  raises  shown  in  raised  (1983) and  out  t h e NP  in  (97b)  at  LF)  (95a,b,c).  See  Rochemont  r e p r e s e n t a t i o n of  extraction  3.3.1.4. The  (also  at  LF  then  i s expected  are allowed Chomsky  (1986)  for  in  (1976), detailed  focus.  p r e 1 i m i n a r y cone1 u s i o n t h u s  against at  NPs  & Rochemont  analyses of the  interpretation  of the construction  same c o n t e x t , a s  Culicover  not  focus  granirnat i c a l i t y  given the  Non! 'No!  holds, the  of doubling applies  at  constraint S - s t r u c t u r e but  Fr.  Trent ino behavior of subject c l i t i c s  and  lexical  NP  subjects  was  presented  clitics  i n 3.1.1.  are obligatory  preverbal  NP  Recall either  i s present.  that,  in this  in isolation  They a r e e x c l u d e d  dialect,  subject  o r when a i n the  lexical  free  i r i v e r s i on c o n s t r uct i on. The  sentences  i n OS)  quantifier  requirement  quantified  NPs  coreferent  subject  and  illustrate  i n subject  indefinite  NPs  4  It  But  4  i n Trentino, i . e .  can f r e e l y co-occur  with  a  4  p l a u s i b l e t o assume  doubled  NP  subject  that  a s shown  surface  extraction of a  i s possible  I n e f f e c t , Wh-movement  construction cannot  no-  vegnu. come  t h e f a c t s a r e not so simple  here.  doubling  i s no  Una p u t e l a * . l ' e i vegnuda. a g ir1 c 1 b e - 3 s come ' Pi g i r l c a m e . '  i s thus  clitic  there  clitic.  ( 9 8 ) a. Qua 1 c h e p u t e l l ' e some boys c l be~3p 'Some b o y s c a m e . ' b.  that  cannot  take  (10£)  b. * Q u a n t e p u t e l e  4  c. * Q u a n t e p u t e l e  4  e na via? be-3p gone away have l e f t ? ' el_e  4  nade  l_e e nade 4  via? via?  place  dialect. i s in  The  order  in this  where a s u b j e c t  extraction.  under1ined. ( 9 9 ) a. Q u a n t e putele how+many g i r l s 'How many g i r l s  N.It.  and a q u a l i f i c a t i o n  i n (99) t o  after subject  in this  preverbal  clitics  c l i t i c  are  ( 1 0 0 ) a. C h i h a who h a v e ~ 3 s 'Who a t e ? '  magna? eaten  b. * Ch i i h a l _ O i  magna?  c. * C h i i ( 1 0 1 ) a.  e l _ i h a magna?  Quante putele how+many g i r l s 'How many g i r l s  b. * Q u a n t e ( 1 0 2 ) a.  must  thus  penset  to allow  3 . 3 . 1.5. In  (103)  magna? est—3s i s eating?'  that  holds  whatever constraint  at S-structure  f o r (98) where  other  QR  N.It. dialect,  same a s i n T r e n t i n o , in isolation  present  che l _ e i s i a vegnude?  i s violated i n  but not at LF i n  applies.  Fiorentino  this  either  vegnu? come  c h e e l i magna?  be assumed  (99)-(10£) c r u c i a l l y order  penset  C h i penset che who t h i n k - £ s t h a t 'Who d o y o u t h i n k  b. * C h i i It  putelei  penset che s i a think—£s t h a t be-3p do y o u t h i n k came?'  a s shown  in  i . e . subject  the facts  are basically  clitics  are obligatory  o r when a l e x i c a l  NP s u b j e c t  i s  (103).  Qua1che r a g a z z o i g l ' e some boy c l be~3s ' Some b o y c a m e . ' A  One o f t h e m a j o r d i f f e r e n c e s  venuto. come  between  concerns the presence o f a preverbal inversion  preverbal  the  constructions.  Contrast  Fiorentino  and  impersonal  clitic  (104) w i t h  (105),  Trentino i n free  where  no  impersonal (104)  (105)  clitic  appears  i n preverbal  position.  a a  F i o r e n t i no a.  61' cl(imp) 'A g i r l  e venuto b e - 3 s come came. '  una r a g a z z a . a g i ^ l  b.  61'ha t e l e f o n a t o una c l have-3s telephoned a 'A g i r l telephoned.'  c.  61'ha portato l a torta c l have-3s brought t h e cake 'Pi g i r l b r o u g h t t h e c a k e . '  ragazza. girl una a  ragazza. girl  T r e n t i no E' vegnu 'na p u t e l a . h a v e - 3 s come a girl 'Pi g i r l came. '  Pis  f a r a s Wh-rnovernent  surfaces in  i n preverbal  i s concerned,  the impersonal  position  i t applies to a subject  when  clitic  <106).  ( 1 0 6 ) a.  b.  Una r a g a z z a * l _ ' e a girl c l be—3s 'Pi g i r l c a m e . ' t  61' cl(imp)  e v e n u t o una b e — 3 s corne-m a  'Pi  came. '  girl  c. # L ' e v e n u t a d.  venuta. come—f  C h i q1'e  e. * C h i e  una  ragazza. girl  ragazza.  venuto?  venuta?  f. * C h i l ' e v e n u t a ? The  sentences  in  (106a,b,c)  show t h a t  only the  impersonal  as  clitic  can  subject is  be  used  i s extracted  grammatical.  because the without (f).  a  The  subject  clitic  Pis p o i n t e d  subject  with and  or  construction  gramrnat i c a l ,  but  ungrarnmat i c a l The with  The possible  subject  in  preverbal (1984), from  (106d),  (106e,f)  are  blocked  clitic  i s evidence  postverbal  this  position either  referential  this  the  i s present;  preverbal  that  position in  extraction in Trentino  to  take  place  in  i s not  and  LF  out  (103)  allowed,  the  same a s  constraint blocking  i.e.  of  subject and  (106a)  witness  are  the  (106e, f ) .  i s therefore  the  at  in Fiorentino since  sentences  doubling,  3. 3. 1. 6.  place  Wh—movement  situation  respect  subject  a  clitic  cases from  In  i n 3.3.4.  Thus e x t r a c t i o n can doubling  with  subject.  impersonal  in Jaeggli  I return to  Fiorent ino  the  i s extracted  extraction takes  Fiorentino.  postverbal  ungrarnmatical  <e) out  a  i t a p p l i e s at  i t is in  Trentino  extractions  S-structure  from  but  not  at  LF.  Romanian Romanian d a t a to  concentrate  generalizations. (1984) and  restricted  The  Steriade  Dobrovie-Sorin Clitic  i s considerably  i n the  the  relevant  well  Romanian examples are (1980);  (1985)  doubling  on  more c o m p l e x  of  cf.  also Farkas  f o r more on a direct  familiar  way  clitic  object to  but  it is  documented  taken  from  (1978)  doubling  i n Romanian  Borer  and in  Romanian.  is  C + h u m a n / p r o n o m i n a l II,  C+specif ic/def initell  NPs.  In t h e s e n t e n c e s below  "object  marker"  and t h e c l i t i c s  ( 1 0 7 ) a.  Am vSzut Popescu. have-Is seen ' I have seen Popescu. '  b.  L -am vSzut. c l have-ls seen ' I have s e e n him. '  c.  Li-am  d. # L i - a m  ( 1 0 8 ) a.  Am  v&zut  vazut vSzut  Li-am  Popescu.  un b u c a t a r . a cook seen a cook.'  vazut. vazut  pe un bucatar±.  d. * L i - a m  vazut  un  ( 1 0 9 ) a.  Am  v&zut vazut  ' I have b.*  be  pe  Popescut.  c. * L i - a m  e. * Am  The  are underlined.  pe Popescu±. OM seen Popescu.'  ' I have b.  Li-am  pe un  bucatari. bucatar.  pe a l t i c i n e v a . some bod y e l s e seen somebody e l s e . '  vSzut  impossibility attributed  s t a n d sf i  v<5zut  ' I have  e. * Am  OM  pe alticineva±.  of object  doubling  i n (108c)  t o the nature o f the doubled  NP,  f i c , - d e f i n i t e , + h u m a n , - p r o n o m i n a l II i n ( 1 0 8 ) a n d - d e f i n i t e , + p r o n o r n i n a 13  i n (109).  f i c , + h u m a n , + p r o n o m i n a l 3 NP s u c h  On t h e o t h e r  and  (109b) i s  namely  C-spec  [-specific, hand,  a  C+spec  a s t h e o n e i n ( 1 0 7 ) c a n be  £11 clitic  is  doubled  as i n  (107c).  Note that  according  hypothesised  that  [+hurnan] s h o u l d Romanian.  This  indefinite The  t o what  a  i s true.  i n object  Therefore,  NP u n b u c a t a r  possibly  i n (110) i s a l s o  (110)  This  very  doubling (108c)  point, i t i s also  constructions i n  i s grammatical  i f the  i s C+specifici.  fata" de l a n o i d i n s a t ] , a l t h o u g h also  to this  L + s p e e i f i c , - d e f i n i t e ] NP t h a t  be a l l o w e d  sentence  we h a v e s e e n  grammatical  becau se  LIMI=i  o  i t i s indefinite i s  specific.  0 i caut p e L"o f a t a " d e l a n o i d i n s a t ] * , cl be+looking+for-ls 0M a g i r l of the our v i l l a g e ' I am l o o k i n g f o r a g i r l f r o m o u r v i l l a g e . ' situation  specific  indefinite  quantified It  i s predicted  NPs  i s thus  extractions  here  under t h e hypothesis  NP d o e s n o t u n d e r g o QR  a t LF a s do  (109) and n o n - s p e c i f i c  indefinite  natural  the constraint  t o assume t h a t  out o f doubling  constructions  that  NPs  a  regular  (108). against  i n Romanian h o l d s  at  LF. The Consider  data  involving Wh~extractions  the relative  i s more  problematic.  and i n t e r r o g a t i v e c l a u s e s  i n (111) t o  (113). ( 1 1 1 ) a.  b.  Casa pe c a r e * c r e d e a i c<=i am v&zut—o± t h e + h o u s e 0 M w h i c h thought~£s t h a t h a v e - I s s e e n - c l 'The h o u s e w h i c h y o u t h o u g h t t h a t I h a v e s e e n ...' Pe c a r e * c r e d e a i ca 0M w h i c h thought—£s t h a t 'Which one d i d you t h i n k  am vazut-o*? have-Is seen c l t h a t I have seen?'  ...  (ll£)a.  P e c i n e c r e d e a i e a am v a z u t ? OM who 'Whom d i d y o u t h i n k t h a t I h a v e  b» *  Pe  (113)a.  ciriBi  Ce what  credeai  credeai  'What d i d b. * C e fit f i r s t  sight,  doubling  are  happens  in  (113b) NP  specific?  (111).  doubling  cf.  can  NP  am  the  be  is  direct  conclusion are  clitics  since  this  out  in  a b o v e on  the  the  the  in  (111)  of  (ll£:b)  NP  hand,  what  and  nature  extracted  other  object  i s presumably  a clitic  On  would  allowed  of  the  i s non-  object  because the  clear  constructions  are  doubled  corresponding  to  object  have seen  saying  in  (108c)  that  W h - e x t r a e t i o n s out i n Romanian;  argues that allow  moved  Wh—movement o f  of  object see  just  why.  like  doubled  structures  constituent".  and  l e t us  "Romanian,  wh-movement  that  doubling  barred.  words, the  of  of  are  does not  in other  out  allowed  (1985.9)  Spanish,  thus consist  fts we  a 3  extractions  i t i s not  object;  i f Wh-extraetions  possible  LF  Plate  t  (1980).  (109b) above,  River  seen?'  vlzut-o ?  doubled  in Romanian.  Dobrovie-Sorin  I have  constraint  constructions  doubling  that  specific.  Wh-extraetions  However,  vazut?  appearance of  Steriade  fi p r e l i m i n a r y only  The by  vazut-Oi?  i n Romanian s i n c e  is initially  (extracted)  am  think  ea  possible  that  am  i t seems as  i s blocked  object  you  credeai  t  ea  c&  seen?'  I will  exclude not  describe review  here  the detail  t h e assumptions  Dobrovie—Sorin stands  (114)  t  underlying  (116)  Cpe c a r e ] *  l*~ai cl  vazut  e  (117)  * Cpe c i n e ] i  l -ai  vazut  e*  t  These c l a u s e s i l l u s t r a t e  and  clauses.  8 4  that  possibilities they  naturally  the correct  extend  LF  f o rc i n e - s e n t e n c e s and c a r e - s e n t e n c e s a r e  (il8)  (119) r e s p e c t i v e l y .  ci ne-sent ence f o r w h i c h x, y o u s a w x  (119)  care-sentence Cisip. f o r which  cine  i s thus  care  she says  bind  a variable  constructions in  clauses;  She argues  (118)  Pe  e  t  the structural  to interrogative  representation  where  x  a i v & z u t <=?*  relative  i t .  a i v S z u t e_ have-Ss seen  Cpe c i n e ] * OM who  to  simply  category.  (115)  corresponding  but w i l l  examines t h e Romanian c l a u s e s below  f o ra n empty  * Cpe c a r e ] OM w h i c h  of her argumentation  x,  (x=N')  a quantifier  (p.7): i nS  "Care (= I P ) .  t h e scope  that  can f i l l  binding a variable  i sa quantifier, We c l a i m  that  of the quantifier  COMP; t h e q u a n t i f y i n g  elements  ( o r x=PRO) ] * y o u s a w e *  domain  of care  t h e argument  i n I P . O f fie.  but i t does not  i n Romanian i s limited  care t o t h e NP  i s not the c l a s s  position  ofthe  (the one occupied  by  e )  i n S,  t  but  PRO  itself,  elements that  satisfy  antecedent  PRO:  that  this  of  the  i s true.  We  (117).  In  (114)  cannot  be  the? b i n d e r .  occurrence  the  the  can  h y p o t h e s e s on  the  nature  proper  sentence  A'-binder  is  grammatical.  is  thus  (117),  presumably since  variable  but  unacceptable doubling A  pe  be  of  (116) pro  e  the  blocked  out  maintained  of  LF  related to constraint will  extractions  throughout  aware of  the  semantic  and  the  the  fact  that  by  against be  of  as  care  but  to  cine  in  this  our pro in  and  the  a  is  Romanian d a t a  This  conclusion  this  section account  will  i s based  position.  is  are  although  r e l a t i o n s h i p between  i n i n t e r n a l argument  of  i n 3.3.4.  Wh—extract ions  Dobrovie-Sorin's  it  In  e x t r a c t i o n out  the  is  sentence  (114),  is  a  (115)  clitic.  which  introduced  well  the  the  clitic,  concerning  as  Pe  the  (114)-  pe  empty c a t e g o r y a  remainder  i n t e r p r e t a t i o n of  in  pro  v a r i a b l e as  constructions.  the  empty c a t e g o r y  a  this  A  be  according  i s licensed  binds e ,  doubling  could  since  by  Suppose  clauses  Pi'-bound t  etc.".  i s v i o l a t e d and  i s not  4  that  the  of  defined  r e l a t i o n s h i p between  more p l a u s i b l e c o n c l u s i o n that  to  u n g r a r n m a t i c a 1.  which  the  boys,  licensed  nothing  A  i t is also given  not  for e ;  cine  i s not  i s thus  constructions  therefore  am  In  return  class  properties  girls,  Alternatively e  would  a  now  variable  pro  the  referential  c l a s s of  of  clitic;  more p r e c i s e l y t h e  care  be I on  3.3.1.7.  no  French  The  situation  i n Standard  free  inversion  (ISO),  there is  Standard  subject  c a n be no wh-movement  present  property  i n the structure  could  availability  French  i s the following, there i s  doubling  o f a n NP (1££).  i s barred  (1£1) a n d  i fa corresponding  Note that  the  clitic  latter  be i n t e r p r e t e d a s a c o n s e q u e n c e o f t h e n o n -  of doubling  in this  dialect.  8 2 3  (l£0)a.* E s t v e n u P i e r r e , is come ' P i e r r e h a s come.' b.  P i e r r e e s t venu.  (l£l)a.# P i e r r e * b.*  Les t r o i s f i l l e s * e l l e s * ont bu. the three g i r l s c l have-3s drunk 'The t h r e e g i r l s d r a n k . '  (l££)a.* Q u i * who 'Who b.  i l * e s t venu. cl  i l * est venu? c l b e - 3 s come h a s come?'  Qui e s t venu?  c. * C o m b i e n d e f i l l e s * d i s - t u qu' e l l e s * o n t bu du t h e ? how+many o f say you that c l of tea 'How many g i r l s d i d y o u s a y d r a n k t e a ? ' Summary We  c a n now e s t a b l i s h s o m e b a s i c  languages/dialects properties  under study  a r e t h e ones  here  i n (1£3).  ways i n which t h e can d i f f e r .  The  different  (l£3)a.  +/-  subject  b.  +/-  object  c.  + /- W h — e x t r a c t i o n o f a c l i t i c  doubled  NP  d.  +/-  doubled  NP  These  doubling doubling  LF-extraction of a c l i t i c  properties are distributed  languages,  where  means t h a t  the property  property  either  set of c l i t i c s  out of doublings  Subject  exists,  "-"  "0" means t h a t t h e  because t h e language does not  and hence no p o s s i b i l i t y  extracting  <1£4>  the property  does not e x i s t ,  i s not relevant  have t h e r e l e v a n t doubling  "+" m e a n s t h a t  i n t h e f o l l o w i n g w a y among t h e  o r because there  i s no  (or impossibility)  but see f o o t n o t e  of  £5.  Clitics  \ p r o per t y language"""" RP S p a n i s h C. F r e n c h PN French Trent ino F i o r e n t i no Roman i a n St. French  doubling  Wh e x t r a c t i o n  LF e x t r a c t i o n  0  0  0  + + + +  0  -  0 0  —  + + +  0 0  £17  (1£5) O b j e c t  Clitics  """"—-~^propert y l a n g u a g e^~~~~~--~~^__  doubling  positions  extraction  __  -_  0  0  0 0  0 0  -  0  -  0  -  +  +  of the relationship  will  LF  extraction  +  RP S p a n i s h C. F r e n c h PN F r e n c h T r e n t i no Fiorent ino Roman i a n St. French Piny a c c o u n t  Wh  have t o leave  +  -  between  clitics  and  argument  opens  (1£6) Pi. t h e o p t i o n o f h a v i n g d o u b l i n g i n t h e f i r s t place? B. t h e o p t i o n o f e x t r a c t i n g o u t o f d o u b l i n g i n s y n t a x ; C. t h e o p t i o n o f e x t r a c t i n g o u t o f d o u b l i n g i n L F . Such it,  an account I want  facts  i s suggested  t o review  mentioned  3.3.3. To  knowledge none  the three  review  three  impossibility position clitic  options  o f them  <19B£).  of the analyses  o f some o f t h e  here.  In t h i s  already  available can  i n (1£6) a t t h e s a m e t i m e .  analysis,  of extracting  t o V absorbs  I  a s  proposed  a clitic  i n River P l a t e Spanish,  affixed  accounts  introducing  Accounts  handle  Jaeqq1i  some p r e v i o u s  but b e f o r e  above.  Previous my  i n 3.3.4  both  t o account  doubled  NP  i t i s assumed t h e government  for the  i n object that  an o b j e c t  and t h e Case  assigning object  properties  position is  This  In  r u l e out  object  V.  The  e f f e c t , the any  empty c a t e g o r y  i n the  ungoverned  PRO.  a n a l y s i s makes s t r o n g  extraction. will  of  Empty C a t e g o r y  occurrence  position since  this  predictions  of  a trace  trace  would  with  respect  to  Principle in  in a not  clitic  be  (1£7)  doubled  properly  governed.  (1£7)  ECP  (Chomsky  L"o.e3 m u s t Note that well  as  with  i n LF  resulting trace for  at  the  Borer  be  are  includes  a  govemee.  the  argument  the  clitic  and  is,  i n her  account,  called Spanish  that  on  s a m e way Move a  since  i n syntax  a  as  trace  in syntax  is still  impossible  this  definition  Case agreement further  that  i t is related  to  a  spell~out  argument  of  of  proper  a  account  a clitic  Romanian s e n t e n c e s  in  governor  properly  i f , among o t h e r same C a s e .  Case  features  position receives Consider  government  between the  to  p o s i t i o n bear the  dummy C a s e - m a r k e r . and  extractions  i s therefore  the  argues  position  the  It  Borer,  the  follows  i n the  LF,  (l£6b,e).  condition She  at  a p p l i c a t i o n of  in  For  governed.  applying  level.  options  (1984)•  ECP  blocked  an  LF  .£50)^  properly  the  from the  1981  again (87b)  the and  Since on  River  governs things, a  clitic  a head,  Case from  and  the  it so-  Plate  (111b) r e p e a t e d  in  (128). (1 £8) a. * Pi DAT b.  In  quierii whom  Pe c a r e * ACC w h i c h  1o± cl(ACC)  credeai ca thought—£s t h a t  (l£8a) t h e p r e p o s i t i o n  quien. The  After  object  mismatch rules  am va'zut-o* t_i? have seen c l ( A C C ) ACC  a assigns  Wh e x t r a c t i o n ,  clitic  v i r n o s t_±7 s a w - l p DAT  dative  thetrace  l_o i s a c c u s a t i v e .  between t h e governor  out t h e sentence since  Case t o t h e object  i salso  There  marked  dative.  i stherefore  Case  and t h e g o v e r n e e and t h e ECP thetrace  i snot properly  governed. In Borer. thus  (l£8b) t h e s i t u a t i o n The p r e p o s i t i o n  marked  requirement  govern t h e trace, type  Case—markers variation.  clitic can  and a b s t r a c t  itself  accusative  extraction relies  o i salso  extractions  i n subject  accusative.  The  can properly  language  doubling  type o f "rescuing  In t h i s  either  the trace i s  o n t h e u s e o f dummy  Case t o c h a r a c t e r i z e  t o assume t h a t  to  possible.  heavily  particular  according  Case,  met, t h e c l i t i c  do n o t b e a r t h e same  the i l l i c i t  explained?  being  t o ; c f . 3.£.£.  i s no r e a s o n  different  clitic  causes a problem  where t h i s  not appealed  there  making  of analysis  This  constructions is  pe. a s s i g n s  ACC b u t t h e o b j e c t  Case matching  This  i squite  particular  theclitic  i n syntax  construction,  doubled  (nominative)  device"  Case.  NP a n d t h e How  o r a t L F be  then  Moreover, evidence dative. person this  suggesting In fact,  neuter  trace  there  that  here  full  pronoun  (1£5);  Jaeqq1i recent  respect  of Dobrovie-Sorin  then RP  i s therefore RP S p a n i s h  who  (t_i)s  merely  argues that c f . 3.3.1.6.  apparent  The  regard  problematic  i s no  to  as i l l u s t r a t e d  £ 5 a n d 31 f o r s o m e  care  and t h e r e  and Romanian w i t h  constructions  pe  in cases  Hebrew.  inversion.  This  a n a l y s i s c o n s t i t u t e s o n e o f t h e most  at e x p l a i n i n g  Jaeggli  presents  ECP.  H i s approach  from  preverbal  t h e f a c t s about  an account  syntactic extractions  position.  (1985)  a variable  out o f doubling  attempts  Italian.  i s true,  adopted  (1984).  certain  If this  that  i n (1£8).  but s e e f o o t n o t e s  Modern  t h e use o f t h e 3  hirn t o t h e c o n c l u s i o n  c f . 3. £. £.  with  i sactually  t h e i n t e r p r e t a t i o n o f t h e Romanian d a t a  between  extractions  i n (l£8a)  i s no  between  i n (1£8)  difference  bring  there  t o Case-marking  (l£8b) d o e s n o t b i n d  contrast  out that  c e r t a i n f a c t s concerning  and Romanian  i s that  points  the trace  i s accusatives  Finally,  in  (1985)  i s no d i s t i n c t i o n  Spanish  in  Jaeggli  i n which  i s motivated  E x t r a c t i o n must Consider  take  of the impossibility no a p p e a l  by t h e f a c t  p o s i t i o n t o the nearest place  the contrast  e x t r a c t i o n and  that  i s made t o t h e WH-rnovement  COMP i s b l o c k e d  from  i n (1£9).  the  of  postverbal  in  (129)a.  Quante ne sono how+rnany o f + t h e r n be-~3p 'How many o f t h e m h a v e  b. * Q u a n t e s o n o Italian  ne  i n an  adjoined  t o VP.  with  structure  the  wh i c h i s  ( 1 3 0 ) a.  Ne  internal  c.  C|=.  in  arrivano  V  (130c)  [ P molti N  t e l e f o n a n o mo11 EVP  equivalent  of  the  of  not  in  Chomsky-  possible in (131b) t h e  (130b)  structure  of  student i  11  ne  in  mo11 i i n  then  f o r more on point  i.  t e l e f onano  subject  c l i t i c i s a t ion)  The  but  an  i.  here  1  (i£9a)  in postverbal position  (1981)  i . e . not  to  T e l e f o n a n o mo11 i s t u d e n t i . t e l e p h o n e d many students 'Many s t u d e n t s t e l e p h o n e d . '  presence  (l£9b),  t o VP,  i s related  Ne_~cl i t i c i s a t i o n i s t h u s  a r r i v a n o mo11  c.  is  NP  t o V which  A r r i v a n o mo11 i s t u d e n t i . arrived many students 'Many s t u d e n t s a r r i v e d . '  b. * Ne  The  affixed  (131c).  b.  ( 1 3 1 ) a.  cadute?  is a clitic  empty c a t e g o r y  cadute? fallen fallen?'  at  (130)  molti  L"IMI=.  i s an  student i  indication  D — s t r u c t u r e and and  (131)  i s in preverbal extraction  11  that  subject  extraction  i s allowed.  position  i s barred 5  the  of  I f , as  ( t h e r e i s no  see  Belletti  &  the in ne-  Rissi  ne-cliticisation. i s that  i f i t i s assumed  that  postverbal  extraction  i s forced  should  apply  (l£9b)  i s i n (13£).  (13£)  [  C  i n (129).  Whi  P  by t h e ECP, t h e n  d p  The S — s t r u c t u r e  The  trace  the  structure i s rejected.  i s no r e a s o n  why i t  representation f o r  11  t i I N F L VP  i s properly  there  governed  by t h e W h - p h r a s e  This  i s what  i n COMP, y e t  Jaeggli  (1984) a i m s a t  explaining. To  do s o , J a e g g l i a d o p t s a c o n t e x t u a l  categories actually thus  (133) i n s t e a d  C C P Whi  Jaeggli  t o which  pro i n Italian  have  (133)  according  d p  proi  since of  t h e empty c a t e g o r y rich  agreement  I N F L VP  governs  i t . We  f u r t h e r assumes t h e f o l l o w i n g :  Hence  i f e x t r a c t i o n i n (l£9b>  preverbal  position,  sentence  extraction  grammatical Jaeggli  then  i s from  effectively  i n accordance  a postverbal  with  takes  place  from  a  p r o i s i n t e r p r e t e d a s a v a r i a b l e and  i s excluded  not coindexed  certain  i n (13£) i s  11  P r o cannot f u n c t i o n as a v a r i a b l e i f l o c a l l y bound by an o p e r a t o r .  is  o f empty  (13£).  (134)  the  definition  with  position,  the operator  (134).  then  If  the pro subject  and t h e s e n t e n c e i s  a s i n (l£9a). claims  that  this  facts i n Trentino  approach  since,  c a n be e x t e n d e d  a s we h a v e s e e n  to  i n 3.3.1.4,  preverbal  extraction i s impossible  rnovement.  In t h i s  place  from  dialect,  postverbal  The m a j o r  problem  quantifier  requirement  languages.  Pis s h o w n  no s u c h  requirement  indefinite in  fact,  by  accounted  different however,  would  since here  explanatory Another  that Since  apply  preverbal  1.3.  This  position, leaves  b u t no  facts.  free  i n (1£6)  of  a t LF i n  i s not  o f an empty  then  ideal,  category  different  at d i f f e r e n t  languages,  levels  of  i t loses  crucial  domain.  f o r J a e g g l i ' s a n a l y s i s h a s t o do  In t h i s  though  language,  there  i t i s possible to  i s rich  subject-verb  i n v e r s i o n . I t can t h e r e f o r e  i s a null  bound  a r e not  or only  i s to allow  e x t r a c t i o n o r i g i n a t e s from  Portuguese  in  require a parametrization  i n the particularly  even  which,  a pro l o c a l l y  at S-structure  p o t e n t i a l problem  t o be  Since  type of parametrization  in different  adequacy  seems  at LF, t h e n t h e r e i s ,  The o p t i o n s  o f empty c a t e g o r i e s  the subject  subject  only  QR  in certain  there  and F i o r e n t i n o .  i t refers to the definition  the Portuguese  agreement  constructions  undergo  pro. I f the theory  representation  extract  i n doubling  i n Trentino  take  concerns t h e no-  i n 3.3.1.4 and 3.3.1.5,  languages.  definitions  with  claim  this  f o r . To do s o w o u l d  which  itself,  with  cf. also  wh~  s  o f empty c a t e g o r i e s ,  an o p e r a t o r ;  (134)  a  a n d q u a n t i f i e d NP  view  cases of  extractions of the subject  posit i o n .  e x t r a c t i o n from  Jaeggli's  i n most  subject  n o t be  postverbal  language  similar  argued  position. to  Italian  then,  according  Jaeggli's,  o f empty c a t e g o r i e s  the result of subject  something  like  extractions in  to a definition  (133),  which  of a preverbal  extraction  violates subject  in this  (134).  as  language i s  Nevertheless,  are apparently  grammatical  P o r t uguese.  (135)  Portuguese  (Zubizarreta  198£)  Que rapazes acreditas que w h i c h c h i l d r e n believe---£s t h a t 'Which c h i l d r e n do y o u b e l i e v e To  such  summarize,  ECP  —  Jaeggli  we h a v e s e e n t h a t (198£),  Borer  tenham gasto esse have-3p spent that s p e n t t h a t money?'  two a n a l y s e s  (1984) —  fail  dinheiro? money  i n terms of the t o make a  necessary  distinction  between s y n t a c t i c e x t r a c t i o n and LF e x t r a c t i o n out  of  constructions.  doubling  not  rely  o n t h e ECP, s t i l l  particular syntactic  3. 3„ 4.  grammar c a n a l l o w extractions  us c o n t i n u e  argument  Celt C e J ]  LF e x t r a c t i o n s  this  t o assume t h a t  a clitic  while  which  that  does  a  blocking  construction.  further  argument particular  p o s i t i o n form of chain  that some  be r e l a t e d t o  essentially the  "the configuration  pronominal  by a s s u m i n g  type  must  (1984s118) s t a t e s  when s h e m e n t i o n s t h a t  i s a discontinuous  intuition  associated call  handle the fact  i n t h e same d o u b l i n g  position. Borer  same c o n d i t i o n  this  cannot  (1984) a c c o u n t ,  C—Chain  Let an  Jaeggli's  element". a clitic kind  L e t us push and t h e  of a chain.  a "c-chain"  I  and d e f i n e  will i t as  in  (136).  (i36)  C-chain  (definition)  a a n d 13 f o r m a c - c h a i n i f s i. of a n d 13 a r e c o i n d e x e d ; ii. a g o v e r n s (3; i i i . f3 i s l o c a l l y f r e e . Conditions  ( i )and  ones necessary argument  and  f o rthe licensing  position  extensively  i n Chapter  pos i t i on  otherwise  constructions concerning free, that  £. T h e l a s t  governs  some  clitics. which  the definition  i n s t a n c e s o f agreement  There reason  i s no e x p l i c i t t h a t we w o u l d  diagnostic  of c-chain  free  cf.  fordifferentiating  e1ement  3.£.£.  i s c o i d e x ed  This  i n simple  with  i s t h e case i n tensed  sentences.  i n (136) f o r t h e  t o use (136) a s a  clitics  us suppose t h a t one i s f a c e d  are locally  i n (136) a p p l i e s n o t  reference to " c l i t i c s " e v e n t u a l l y want  i n doubling  a r e not ruled o u t ;  element.  markers  The  s e e 3.5  which  r e c e i v e Cases  b u t a 1 s o t o wh a t e v e r locally  Pi-bound;  are free,  out o f t h e  antecedent.  anaphors  Pronominals,  i s new  s i n c e f3 w o u l d  by t h e c o i n d e x e d  are locally  c a n somehow  another  in (iii)  (a)  o u t ( o v e r t o r empty)  inthe  and a r e d i s c u s s e d  condition  to a clitic  0'-bound  and R ~ e x p r e s s i o n s ,  to clitics  pronominal  t o b l o c k any e x t r a c t i o n  related  s i n c e they  reflexive  Note t h a t  Let  rules  i s , i fthey  only  ((3 )  be l o c a l l y  same c o n d i t i o n  o f an empty  associated with the c l i t i c  i s introduced i n order  argurnent  and  ( i i ) on t h e f o r m a t i o n o f a c - c h a i n a r e t h e  and agreement  with data  markers.  that are problematic  as  f a r as  the  is  concerned.  distinction In  tests  i n order  or  agreement  an  d ist inguish. be  that,  way  up  with One  clitics  such  to  to  following  T h e s e two  the  to  one  w h e t h e r an  reasons  a certain  the  and  decide  of  respect of  a situation,  marker.  One  between c l i t i c s  would  affix  on  elements  for this  point,  and  they  agreement want  a head  are  apply is a  often  state  both  to  of  behave  markers  clitic  hard  affairs i n the  to mi g h t  same  (136).  ways  i n which  agreement  (136)  can  be  markers c o n s i s t s  well-formedness  condition  used  to  distinguish  i n adopting  on  the  the  occurrence  of  clitics. (137)  Pi c l i t i c  This  must  condition  Principle principle. presented other  cannot  since On in  be  only  the  part be  hand, and  be  In  remainder  of  properties The role  level i n my  the  definition  presented at  which  this in  the pro  the  principle  will  see  how  can  the  pro,  and that  a  we  account  holds  a  Principle  be  section  c h a r a c t e r i s a t i o n of  such  could e  condition  from  that  i n 3. .3. 1 i n v a r i o u s  this  derives  requires  (137). '  (136)  Projection  Recoverabi1ity  licensed  condit ion  and  of  which  mot i v a t e s  (137)  from  the  uIt imate1y the  c-chain.  presence  Introduction  empty c a t e g o r i e s ,  a  derived  the  other  of  for  condition  the  languages/dialects. will  properties  play in  a  crucial  (i£4)  and  In —  RP S p a n i s h  QR o r Wh-movement  Condition of  as well —  <137) t h e r e f o r e  o r QR v i o l a t e s  involving The  them  with  except  condition  (137) h o l d s  extractions  a r e barred.  condition  Now, movernent position,  o f QR  with  that  t o doubling  by  Wh-  doubling  doubling  i n C.Fr. i s l i m i t e d t h e same,  only  and Romanian.  care  at S-structure  whereas o f i f we where  out as desired.  to Jaeggli's  (1984) c o n c l u s i o n  i n Trent ino takes with  place  from  LF  Wh assume  i t blocks  i s not f i l t e r e d  respect  namely  i sdifferent:  are allowed  c a n be t a k e n  arethe  Pt v a r i a b l e r e s u l t i n g  we a r e l e f t  from t h e  that  Wh-  postverbal  two p l a u s i b l e hypotheses i n  the data:  E x t r a c t i o n from fact  left  b y Wh-rnovement.  of a subject  interpreting  respect  a t L F a s i n RP S p a n i s h  This  (137) h o l d s  variable left  application  t h e appearance  The t r a c e  and F i o r e n t i n o t h e s i t uat i o n  out o f subject  Pi.  a clitic.  The a n a l y s i s i s t h u s  extractions  the  constructions.  ( i i i ) i n (136) and t h e s e n t e n c e s  position.  In Trent i n o  that  doubling  a t LF and b l o c k s  p r o p e r t i e s o f C.Fr. with  the subject  that  holds  i s no e x t r a c t i o n  are ruled out.  s a m e a s i n RP S p a n i s h to  out o f object  a variable associated  movement  a s i n Romanian t h e r e  that  otherwise  postverbal  condition be  position i s forced  (137) h o l d s  violated;  at S-structure  by t h e and would  B.  Extraction  takes  because  (137),  of  structure The  test  case  B  is  questioned.  c l i t i c  (138d),  that,  the subject  us w i t h  Una r a g a z z a  1_' e v ' e n u t a .  b.  G l ' e venuto  una  L >  Wh  e.  Chi e venuta?  L P>  Wh *  right  important  one, which here  CP  C  (138e)  neuter  B,  that  such  an  have t h e form i n on t h e o t h e r  hand,  output.  o u t t o me  reinforces both  C i  t  proj  P  C i  that  i s rejected.  i s that  but a  in  ragazza.  Chi g l ' e venuto?  interrogative;  subject  ragazza.  d.  Wanner p o i n t e d  the  o f ft  constructions  would  i s the grammatical  una  a t S-  i n favor  i n which  Hypothesis  ( 1 3 8 ) a.  the  clause  t o (138a),  that  Dietter  evidence  postverbally  predicts  c. # L ' e v e n u t a  be p r e s e n t  H y p o t h e s i s ft p r e d i c t s  i n (138e).  (138e)  p o s i t i o n but,  the variable.  appears  corresponding  not t h e one  cannot  i n the inverted  preverbally.  interrogative,  postverbal  interrogative  Recall  appears  with  can provide  i s a Fiorentino  Fiorentino,  from  a clitic  together  that  or  place  P  t * V  (138d)  clj+V  t  4  33  33  i s the grammatical  H y p o t h e s i s ft i s t h e r e f o r e  Jaeggli's  hypotheses  conclusion.  are compatible  What i s with  my  suggest ions. I subject but  now  turn  doubling  w h e r e Wh  t o PN F r . i n w h i c h or object  extractions  LF e x t r a c t i o n s  doubling  a r e barred.  constructions We  see here  out o f e i t h e r are allowed that  the  situation with  the  i s the  exception  dialects.  The  S—structure  in  the  then  of  respect  unmarked to  sentences  to  are  case,  i n v e r s i o n found  this  (137)  NP  thus  earlier  i s allowed,  parameter  type  holds  setting  involving c l i t i c  apply  same,  i.e.  of  Fiorentino  i n the <137)  latter  — two  applies only  parametrization  a v a i l a b l e i n each  change t h i s  quantified  The  free  and  at  Fr.  experiences  has  i t i s in Trentino  a n a l y s i s i s the  i n PN  Note with trigger  same a s  and  LF.  The  this  doubling then  i n the  at  of  (137)  case.  i s d o n e by  that  learner  exposure  a q u a n t i f i e d NP.  cannot  here would  Suppose  language  hold  d e r i v a t i o n , i . e . at  involved  that  at  LF  to  Since and  a  will  S-structure.  thus have the  form  in  (139). (139)a.  Unmarked s e t t i n g : (137) h o l d s a t LF.  b.  Marked s e t t i n g s (137) h o l d s a t S - s t r u c t u r e .  Consider, other  finally,  language with  doubling.  Needless  language;  that  doubling  at  to  say,  is  together  since  a  the  case of that (137)  unmarked  a l l i s blocked to  marked  clitics  i s the  assignment present  the  doubled with  clitic  NP a  Standard  French  does not  allow  holds  LF  case.  because of presented clitic,  at  The  a b s o r b s C a s e and  any  type  of  of  having  c o n d i t i o n s on  i n 3.£.£.  then  or  clitic  in this  option  the  —  If a  Case  lexical  this  NP  i s not  the  NP  i s r u l e d out  NP  Caseby  the  Case The  f i l t e r . analysis  3  0  just  that  I now e x p l o r e .  must  be n o t e d ,  such  a way t h a t  clitic  They  first,  doubling  doubling  introduced  that  there  constructions  different  condition  i n (137) w i l l  clitic.  behind  will  also  this  empty  and  i s designed i n grammar  be e x c l u d e d  o f grammar  of a variable  out o f  adopted  illustrated  associated t h e empty  i n the transformational  once t h e s t r u c t u r e  here, the  with  a  category  component  i s f e d t o LF  In other  i nt h e  NPs a t LF. I n  at LF, then  i sa variable.  allowing  MPs a r e i l l i c i t  the quantified  are blocked  b y Wh—movement  possibility (1£4)  block  the presence  category  languages. I t  are allowed  quantified  predictions  levels of syntactic representation,  I f variables  left  account  extractions  t h e model  two  "possible"  be no p a r t i c u l a r  while  including  i t blocks  Wh  Given  concern  t h e present  should  i n which  same c o n s t r u c t i o n .  fact,  both  makes a t l e a s t  since  words, t h e  i n (140) i s e x c l u d e d .  Compare  (140) t o  (1£5).  (140) doubling language  +  L The  second  extractions should  Wh  extraction  LF  extraction  ~~~——  prediction  i s that  and LF e x t r a c t i o n s  exist.  _  +  languages allowing out o f doubling  It i s the representation  in  (141).  both  Wh  constructions  (141) Wh  doubling  extraction  LF  extraction  1 ang u a g e ^ \ ^ +  L  Argentinian set  Spanish  of properties.  studied  extensively  sentences  them a s a s i n g l e (1984a51984b51385).  i n Hurtado  this  dialects  i t  i s  Consider the  (1384b)  A quierii l o acuso su complice t_i? t o whom c l accused h i s accomplice 'Whom d i d h i s a c c o m p l i c e a c c u s e d ? '  b.  Sus complices los* acusaron a todos l o s e s t u d i a n t e s . their c l to a l l the students 'Their accomplices accused a l l t h e students.'  t  t  (14£a) t h e Wh  This  situation. (137)  dialect  At l e a s t  thus  former,  doubled  NP  allowing  a  patterns with the description i n  a s i n t h e unmarked  dialect.  t o handle case,  (iii)  A'-bound  this  condition  The o n l y d i f f e r e n c e  l a n g u a g e s / d i a l e c t s s t u d i e d above  condition  though  i s e x t r a c t e d a t LF".  t h r e e a n a l y s e s c a n be p r o p o s e d  at LF i n t h i s  and t h e o t h e r  i s e x t r a c t e d even  i s t r u e o f (14£b) b u t , i n t h i s  clitic  L e t us s a y t h a t ,  holds  therefore  a quien  T h e same  a quantified  Spanish  (141).  phrase  i s present.  sentence,  the  seem t o e x h i b i t  a.  clitic  it  treat  Spanish  i n (14£).  (14£) H u r t a d o  In  and Uruguayan I will  +  +  between  i s that,  in  o f (136) i s somewhat r e l a x e d empty c a t e g o r i e s i n i t s c - c h a i n s .  Sentences trigger (iii).  similar  experiences  in  markers.  (142)  sentences be  variable  in  account To namely, in  language be  actual case,  can  the  case  clitics  then  (137)  the  empty c a t e g o r y  not  analysis of  a q u i en.  the  constitute  Romanian d a t a  extraction and  LF;  the  i n syntax  possibilities languages/  the  are  aside  in  this  to  The  one  i s the  which  type  (136iii).  3.3.l.S.  i n LF  extraction  in  one  in  i s excluded (143)  with  of doubling  the  a  particular  at  only,  i s not  of Dobrovie-Sorin's in  and  possibility  (142a)  are  last  (143).  in  agreement  apply  another  four p o s s i b i l i t i e s  summarized  dialects.  Yet  case,  discussed  all,  as  verbal  rather,  a counter-example  i s reminiscent  summarize, no  In that  the  does not  to claim that by  considered  that  but,  problematic.  bound  be  learner to relax condition  not  syntax  set  not  (142)  are  extraction  is  in  i t might  i s the  (142)  example would latter  are  If this  ones  f o r the  Alternatively,  affixes  would  to the  (1985) 3 1  accounted only,  (140), i n my  the  The  for, extraction  i.e. account.  The  appropriate  (subject or  object)  (143) ^^~~~~-~-^DT o per  d oub 1 in g  t y  Wh  extraction  LF  extraction  language^— RP S p a n i s h C. F r e n c h Romanian  +  -  -  firgent inian, Uruguayan Spanish  +  +  +  PN F r e n c h Trent ino Fiorentino  +  -  +  no  +  +  -  language  3.4. ft N o t e o n C o m p l e x The well  Complex  studied  referred prep.)  is  of c l i t i c s  clitic  not studied fact.  doubling  Kayne  (1972)  (1984), cited  section  introduced Inversion  i s affixed  here;  What  property  f o r more  i s to explore  I will  with  t o the verb  we a r e i n t e r e s t e d exhibited  (1981;1985),  i n the preceding  the references  by t h i s  i s  but a l s o t o D a v i s ( i n  there,  data.  French i s  The r e a d e r  Safir  the position of the c l i t i c  subject  this  t o Kayne  The a i m o f t h i s  t o t h e Complex  here with The  (1982),  i n Standard  literature.  (1983) and r e f e r e n c e s  theory  applies  construction  i n the generative  Jaeggli  discussions. the  Inversion  in particular  Sportiche  Inversion  above  t h e way  concerned  to the  and i t s s u f f i x provide  verb. status  accounts of  i n i s the apparent construction.  that  sections  n o t be  respect  complete  subject  Consider  (144). ( 1 4 4 ) a.  Jean  viendra -t-il? corne+FUT-3s he ' W i l l J e a n come?'  b.  Marie  a  -t-elle fait she done do t h a t ? '  have-35  'Did In t h i s clitic  Marie  chapter  we  doubling  have  eela? that  been c o n c e r n e d  constructions,  with  the  properties  e s p e c i a l l y i n the  of  configurations  below.  (145)  We  a.  Cxp.  b.  CII=.  [ P  X i  N  NP  x  seen  when t h e  CNP,IP3  material  because the  Now as  in  the  [ I P  Cvp,  that  113  c l i + V  (145a)  x  L. NP  by  p o s i t i o n i s not in  (144)  4  3333  i s excluded  position is filled  sentences  CMI=-  The  only  the  clitic, to  has  X i  3  C . T  V+cli  d i f f e r e n c e between about  ask  instances then  c l i + V  i n Standard  phonologically  French  realized  Case—marked.  could  tentatively  be  represented  i s the  p o s i t i o n of  (146).  (146)  want  Ci»  INFL  L ,  have a l r e a d y  II  the  of to  acceptable  which  (145a)  I will  doubling?  followed  i n Standard  by  (146)  are  I f the  another  French  and  have nothing  following questions  subject be  33  given  the  answer  questions that  to  say.  We  sentences  thus in  (144)  i s positive, why  (145a)  is  it  (146)  i s not?  In  other have  words,  i f we c a n h a v e d o u b l i n g  i t elsewhere? I argue here that  negative  and t h a t  introduced Complex for  t h e answer t o t h e f i r s t  this  follows  i n Chapters  Inversion  from  £: a n d 3. T h i s  i s then  clitics  us assume, a r e base  instance  position  realized  material  mat e r i a 1 ,  Inversion. Note  that  first  that  I n v e r s i on is opt (147)a. b. The with  c l i t i c  t h e presence  iona1.  fait  no  argument  an  t h e Case i nthe  phonologically  when t h i s I f there  therefore  subject  position were  such  filter.  This  occupy  another  i n Complex  of this  approach.  of the lexical  NP  i n Complex  3 S  Viendra-t-i1? fi-t-elle  govern  by t h e C a s e  i n favor  St.Fr.  participate  that  to  construction.  they  i n INFL.  NP) m u s t  respect  absorb  i n CNP, I P ]  of the external  i s evidence  they  already  they  predict  b e made i l l i c i t  (the lexical  There  words,  then  a subject  i t wo u1d  than  We  that  and t h a t  c a n be p r e s e n t  with  material  position  i n INFL,  In other  of a c-chain.  coindexed  overt  generated  i s  consequences  of this  a s we h a v e d o n e s o f a r , t h a t  C+C] o f RBF3.  formation  with  shown t o have d e s i r a b l e  o f p r o i n CNP,IP],  feature  question  the hypotheses  c h a r a c t e r i z i n g some o f t h e p r o p e r t i e s Let  is  i n ( 1 4 6 ) why c a n we n o t  =(144a) cela?  =(144b)  sentences  i n (147) a r e t h e n  t h e same a s t h e o n e s  respect  to the relationship  between t h e s u b j e c t  i n (148) c l i t i c and  the  external  differ  argument p o s i t i o n i l l u s t r a t e d  i n the surface  (148) a.  position of the  i n (149).  They  clitic.  II viendra.  b.  Elle  (149)  a fait  cela.  IP / NP*  \ I'  pro  I  / PiGR  VP \ TNS  cl± In  Standard  us  assume t h a t  accept a  French,  no  this  the fact  i s also  that  posit ion externa1  (15u)  CIMP  T h e NP since  located  3  true  the lexical t o IP as  L xF>  NP  LINIP  outside  o f Complex NP  3  The o n l y  i n a TOP  assumed,  because  Wh-quest i o n s  (151)a.  b.  node  II i *  as  indication  i fthis t h e NP  in  II  3  node  appears  3 -  c li  xNFL  we  *  and  i s probably  3  ...  in  IIVP  i s i t s  can get i s that o f CP a s  to the right  Quand P i e r r e v i e n d r a - t - i 1 ? 'When w i l l P i e r r e c o m e ? ' cela?  333  ...  Case-marking  B u t what  i s outside  a-t-elle fait do t h a t ? '  Let  3  (151).  Pourquoi Marie 'Why d i d M a r i e  Inversion  o f IP does not r e q u i r e  clear  pro i n (149).  i n (144a,b)  i n (150) .  proi  1  can replace  i t i s not i n argument p o s i t i o n .  position? not  ...  lexical  exact  i ti s  i s often  o f t h e wh-word  in  Kayne  (1972),  citing  NP d o e s n o t u s u a l l y This  observation  Gross  (1968),  appear between  i s based  mentions that t h e Wh-word  on t h e c l a i m e d  a  dislocated  and t h e v e r b .  contrast  i n (152); c f .  a 1 s o 3. 1.3. 1. ( 1 5 2 ) a.  P o u r q u o i e s t - c e rnauvais, why i s i t bad 'Why i s m o n e y b a d ? '  b.??  Pourquoi,  I do n o t b e l i e v e analysis (153) way,  1'argent,  that  proposed  (153)a.  est-ce  t h i s argument  here since,  are a l l acceptable. which  1' argent? t h e money rnauvais? i s sufficient to reject  i n my  judgement,  the  the sentences i n  I u s e t h e comma i n t h e t r a d i t i o n a l  i s not relevant  here.  J e me d e m a n d e q u a n d , P i e r r e , i l v a v e n i r . ' I w o n d e r when, P i e r r e , h e w i l l come.'  b.  J e me  demande quand  c.  J e me  demande, P i e r r e ,  d.  Pierre,  j e me  demande quand  e.  Pierre,  quand  va-t-il  f.  Quand v a - t - i l  g.  Quand P i e r r e  venir, va-t-il  i l va venir, quand  according  be i n e x t e r n a 1  remain  unmarked  f o r Case  venir.  venir?  external  cannot  i l va  venir.  Pierre?  why  simply  i l va  venir?  T h e r e s e e m s t o be r e a s o n t o IP. In f a c t ,  Pierre.  t h e NPs  t o our analysis,  argument  i n such  i n (151) cannot  posit ion;  a position.  they  L e t us  be t h e NPs wou1d therefore  assume t h a t  the  them  to the  appear  NPs  in  (144)  right  are  of  adjoined  to  IP.  This  COMP i n W h - q u e s t i o n s a s  makes  in  (151)  above. This  approach  has  three  d e s i r a b l e consequences that  I  now  present. First,  a  lexical  NP  i n Complex  I n v e r s i o n cannot  undergo  Wh-  movement.  ( 1 5 4 ) a.  Jean v i e n d r a - t - i 1 ?  b. * Q u i i v i e n d r a - t - i 1 ± ? 'Wh o w i l l c o m e ?' c.  Ces these 'Will  personnes seront -elles people be+FUT~3p c l t h e s e p e o p l e be p r e s e n t ? '  d. * Q u e l l e s p e r s o n n e s 'Which people w i l l  seront-elles± be p r e s e n t ? '  4  This  is a trivial  subject  p o s i t i o n s they  Alternatively, subject  i t can  extraction,  sentences  are  then  Wh-movernent o f are  (  1  5 5  matter  )  the  undergo s u b j e c t  assumed NPs  the  that  must  be  subject  in  (155)  i s not  and  NPs  in order  in subject  C I P  (156)  j Wh  NP x  *  CIP-  [ j  P  CNI=.  proi  J  below.  C  *»  j HIMP> t i  -case  3  C x»  ...  33 3  ...  to  not  in  undergo  position.  variable  Case-marked.  (156)  are  extraction.  *  (C  Ccsf='  tcp  cannot  assume t h a t  presentes?  ungrarnmat i c a 1 b e c a u s e t h e  the  illustrated  i f we  be  presentes? present  .13.13  The  two  The  left  by  options  Mote the  that  (154b,d>  sentences  a r e excluded,  i n fact,  i i i vient? c l come-~3s comes?'  b.* J e c o n n a i s l a f i l l e qu' i e l l e i I know-Is t h e g i r l that she ' I k n o w t h e g i r l who s i n g s , ' e.  Qui vient?  d.  Je connais  Secondly,  l a fille  a subject  ( 1 5 8 ) a. b.  If  Kayne  NP i n a r g u m e n t  of this  & Po11ock  chante. sing-3s  q u i chante. p o s i t i o n i n St„Fr. c a n  undergo t h e well-known r u l e o f S t y l i s t i c  discussions  reason  i n (157a,b) a r e , a s opposed t o (157c,d).  ( 1 5 7 ) a.# Q u i i who 'Who  (197£;1975),  f o r t h e same  (1978),  Inversion;  Safi r  c f . Kayne  (1985) f o r  construction.  Quand v i e n d r a P i e r r e ? 'When w i l l P i e r r e c o m e ? ' Quand a c h a n t e M a r i e ? 'When d i d M a r i e s i n g ? '  t h e NPs i n (151) and more g e n e r a l l y  constructions that  they  this  i s true  Inversion  a r e not i n subject  cannot  and Complex  b.* Quand  position, then  undergo t h e r u l e o f S t y l i s t i c  i s illustrated  (159) a . * Quand  i n Complex  i n (153) where  Inversion  viendra-t-il a-t-elle  a r e found.  Pierre?  chant!  Marie?  both  Inversion  i t i s predicted Inversion. Stylistic  That  Thirdly, Complex  Inversion.  quantified cf.  i t can  NPs  3. 1.3. 1.  (160)a.  as  We  be The  the  construction  type of  i s not  C o l l o q u i a l French  b.  Tout e s t - i l en all is c l in 'Is everything  c.  Une t a b l e p e u t - e l l e a v o i r d e u x a t a b l e can cl have two 'Can a t a b l e h a v e two legs?' argue that  of  subject  doubling  language  i n which  doubling, to  doubling  as  our  the  only  LF  can  limited  subject  be  to  found  in  non  doubling  i s ;  be  language s i n c e  what  is  d i f f e r e n t enough  is  equivalent  to  be  no  position  French  i s not  But  licit  this  reason i n such a  good first  from  cast  i t to Complex  to a  to  a  of  bring  why  subject  us  St.Fr.  verify  and  out  would  interrogatives in  doubling  instances  corresponds  determining  appears at  St.Fr.  are  interesting to  subject  There would  Colloquial  in  of  indeed  French  extractions Fr.  pattes? legs  above are  Standard  i n PN  only  i t would  i n argument  that  question  i s possible  Inversion.  sentences  case  original  language a l l o w i n g  NP  NP  ordre? order in order?'  and  i s the  Finally,  the  any  Chaque s p e c t a t e u r d o i t ~ i l payer? each spectator must c l pay 'Must e a c h s p e c t a t o r pay?'  could  a  that  haves  One  back  noted  what  happens  in  e x h i b i t i n g Complex  exclude  the  presence  of  language.  example of sight  as  d o u b t s on  Inversion.  3 8  this  type  Complex the  claim  of  Inversion that  it  Nevertheless,  Trentino  exhibits  a property  Inversion.  Compare t h e f o l l o w i n g (a) s e n t e n c e s  (b) s e n t e n c e s  (lGl)a.  similar  i s , according  (1985s££1),  the  very  what  to St.Fr-  to  Safir  Complex i n Trentino  to  i n St.Fr.  Maqnelo?  b.  Manqe~t~i1? 'Does he  (16£)a.  ftl_o  b.  magna?  A-t-i_l  mange?  'Has he ( 1 6 3 ) a. L a b.  eat?'  eaten?'  maqnelo?  La  mange—t—i1?  'Does he e a t  i t ? '  (164) a . * E l * maqnelo*? b.* I I i mange-t—i1±? ( 1 6 5 ) a. Q u a n d o rnaqne 1 o ? b. Q u a n d m a n g e - t ~ i _ l _ ? 'When d o e s h e e a t ? ? ( 1 6 6 ) a. E l M a r i o maqnelo? b. M a r i o m a n g e - t — i 1 ? 'Does M a r i o It  i s hard  argument  NP  cannot  Inversion in  to verify  position  r e a s o n s why  CNP,IPD  we  eat?' whether  the lexical  or not. Recall that  assume t h a t  be moved  either  i n Wh-questions  i s in  crucial  i s because  this  or i n S t y l i s t i c  i s so because  i n Trentino,  i n (166a)  one o f t h e  i t i s n o t i n (166b)  constructions. This position  NP  even  i f t h e NP  i t s e x t r a c t i o n i n syntax  was would  be  prohibited  doubling  as  in  (167)  (167)a.# C h i i b. *  in  the  in  St.Fr.,  hal_Oi  the  of  The  only  whether  the to  plausible  clitic  i s acceptable  conclusion  lexical IP,  NP  with  French  assume t h a t  in  a  our  a  in  the  presence  subject under  clitic.  a  right-  between  the  verb  i s that  i t is  As  and  pro  in  to the  Mario?  stage is in  CNP,IP]. pursue  theory  lexical  Note however that that  only  pause  this  on  the  NP  by  analyses  matter  position are  further.  p o s i t i o n as  the  Complex i n such  probably  proposed  two  in a  unclear  r e l a t i o n s h i p between  involved  IP,  CNP,IP] o r  The  this  argument  i s supported  position external  involves  with  Mario?  (166a)  not  i t s associated  we  a  el  at  in  pro  I will  conclusion,  Standard  i s blocked  coreferential inverted  r n a n g e - t - i _ l #(##)  ones.  and  The  NP.  valid  external  In  a  Inversion  i n t e r p r e t a t i o n with  Quand  dialect.  magna?  Q u a n d o magnel_o *(##)  b.  in this  of  magna?  sentence  postverbal  E x t r a c t i o n out  ungramrnat i c a l .  Stylistic  sentence  dislocation  ha  i n 3.3=1.4.  i s blocked  are  eJL*  Chii  Furthermore,  (16S)a.  have seen  constructions  sentences  the  we  CNP,IP] c o i n d e x e d  Inversion a  adjoined  structure with  i t applies  in and  a in  facts i f  construction  is  to i t . (150)  above  governed  by  the  subject  clitic  i n I N F L must  also  be a s s o c i a t e d  external  t o IP, probably  by c o i n d e x a t i o n .  (136iii)  and s h o u l d  i t impossible  c-chain and  with  p r o i n CNP, IPD.  resumptive assumed result  that  (136iii)  In other  would  t o form  a  case,  be  violated.  t o r u l e o u t empty  words,  r e f e r s t o an element  In that  f o r (136) and  (150) would  i t must  be  Pi'-bound  as a  not cause  (137).  A Note on R e f l e x i v e s e In t h i s  previous  section  however t h a t respects.  The  this this  time  (169)a.  the exercise  t o Complex  on r e f l e x i v e  section  presentation  done  i nthe  Inversion,  clitic  i s , at best,  and r e c i p r o c a l French  se i n French.  incomplete  Note  i n many  of the various  properties  s e i s a v a i l a b l e i n Kayne  3 &  problem  posed  by s e n t e n c e s  analysis of c l i t i c s  category  t o repeat  regard  Pin e x h a u s t i v e  reflexive  (1975).  I wish  section with  concentrating  our  i n general. only  violates  f o rthe c l i t i c  Principle  we d o n o t w a n t  (136iii)  o f movement.  problems  3. 5..  pronouns  This  t h e NP  Pis a c o n s e q u e n c e c o n d i t i o n ( i 3 7 )  ultimately the Recoverabi1ity More g e n e r a l l y ,  of  make  with  h a s t o do w i t h  related to the reflexive  Jean  like  se voit. c l (REFL) s e e - 3 s 'Jean sees himself.'  t h e ones the status  clitic.  i n (169)  for  o f t h e empty  b.  .Jean s e l a v e . 'Jean washes h i m s e l f - '  To s e e t h i s , that  l e t us e x p l o r e one p o s s i b l e a n a l y s i s .  i n order  anaphoric  to capture  the fact  that  t o t h e s u b j e c t , we s a y t h a t  internal  argument  position  argument  position  as i t presumably  translations  ( 1 7 0 ) a. b. The to  [ i p [  i n (169).  N  P  category  the Binding  3  t h e empty  i s coindexed  with  II  INFL i.  C  V  INFL  t h e NP  se +voit  [ v  P  [  V  P  t  i n external  likes  3 [  sei +lave 3  [ v  i n (179a,b) i s c l e a r l y  (by t h e s u b j e c t )  category i n  i s i n the English  conditions. In fact,  i t i s bound  the reflexive i s  would get something  J e a n a. 3 Ci»  C i p IIrsiP> P i e r r e t  empty  since  We  Suppose  N  P  I I M P -  C+anaphorD  i t cannot  333 3  e±  S J  according  be a n y t h i n g  i n i t sgoverning  333 3  else  category  (IP) . With The  respect  reflexive  to the clitic,  clitic  NP o r i t i s n o t . T h a t first  i s where  o p t i o n , t h e empty  conditions then,  i s either  s t i l l  category  must  be l o c a l l y  order  empty  on c - c h a i n s  license  element i n i n t e r n a l  t o be a s s i g n e d  a reflexive  possibilities. t h e empty  lies.  object  Following the  h a s t o be p r o b e c a u s e o f t h e ( c f . 3.3.4) b u t  of c-chain,  (Pi o r Pi') f r e e . rn u s t  with  t h e problem  to the definition  c o n t r ad i c t i o n s t h e c l i t i c anaphor3  a r e two  coindexed  category  established earlier according  there  a  There i s thus  a  II + p r onom i n a 1,  —  argument  reading  t h e empty  position  t h e empty  but i n  category i n  (169)  must  The  be  second  problematic. with  the  desired object  C+anaphor]. option  Suppose that  internal  adoption  such  generalizations We  would  lose  an  and,  approach  made up  to  in particular  solution stipulating  such  an  independently.  clitics.  the  other  This hand,  approach.  coindexed  with  This  an  Let  One  would  (137)  us  then  argument  can  to  for  the  subject  the  and  the  l o s s of  the  the  (object)  (137)  have to  clitics.  above which  propose  does not  seems r a t h e r evidence  allows  coindexed  However,  regard  condition  that  i s not  t r a n s l a t e as  with  the  more  the  coreferent.  could  now  ad  On  even  clitic  object).  thus,  for  reflexive  sight  reflexive  (the  •argued hoc  first  i n t e r p r e t a t i o n i n s o f a r as  coindexed  of  the  argument  reflexive are  seems at  a  was  seemingly  apply  to  undesirable. be  put  assume t h a t  p o s i t i o n and  forward  that  reflexive  se  see  how  this  support is  can  not be  implemented. Let  us  formulate  return  condition  generally, of  to  some  Such  maybe g e n d e r  features.  certain  depending  question.  the  fact that  information  as  in  on  intuition  referring  includes  Pi c l i t i c  features on  In Standard  which the  lead  to  null  person,  this  the  to Very  function  syntactic  number,  Case,  and  instantiates this  phonologically  particular  us  clitics.  fulfills  c a r r i e s or  are  French,  that  appearance of  a clitic  information  information way  basic  (137)  i t i s the  recovering  arguments.  the  realized in  combination  realization  of  of  a  features  the  features ( 1 7 1 ) a.  comprises t h e ones C3, s g , m a s e ,  ta. C3,  s g , masc,  accusative]  = le  nominative] =  i l  (i_l_  voit)  l u i  ( j e 1 u i par1e)  dative]  d.  C3, p i , m a s c / f e r n ,  accusative]  = l e s (je l e s vois)  e.  C£, s g , m a s c / f e r n ,  accusative]  = te  Standard  French  combinations. clitics  to rich vary  agreement  in their  form  clitics  for different with  in  feature the  other  respects.  t h e same p h o n o l o g i c a l  accusative  ( j e t_e v o i s )  in Italian,  R e f l e x i v e se does not p a t t e r n  i n many r e l e v a n t  First,  =  ( j e l_e v o i s )  C3, s g , m a s c / f e r n ,  similar  with  (171).  c.  I n a way  both  in  and d a t i v e  realisation  reflexive  objects.  serves to Compare  license (17i~")  (173).  (17£)a. J e a n  l u i cl(DAT) 'Jean speaks  b.  Jean  (173)a.  Jean  b.  Jean  le voit. c l (ACC) s e e - 3 s 'Jean sees him.'  se parle. DAT 'Jean speaks t o himself.' se voit. ACC 'Jean sees himself.'  Secondly, contrast  parle. speak—3s t o him.'  se does not vary between  (174) and  i n gender as i l l u s t r a t e d (175).  by t h e  ( 1 7 4 ) a.  Jean  le voit. cl(MfiSC) 'Jean s e e s him. '  b.  Jean  ( 1 7 5 ) a.  Jean  la voit. cl(FEM) ' Jean sees her.' se voit. MPiSC 'Jean sees himself.'  b. M a r i e s e voit. FEM 'Marie sees herself.' Thirdly,  se  i s invariable  a s f a r a s number  i s concerned;  other  c1i t ics are not. ( 1 7 6 ) a. J e a n b.  Jean  ( 1 7 7 ) a.  Jean  b.  Finally, usual  voit.  les voit. cl(PL) ' J e a n s e e s them.' se SG  voit.  Les gens se voient. t h e p e o p l e PL s e e — 3 p 'The p e o p l e s e e t h e m s e l v e s . ' reflexive  object  (176)a.  le cl(SG)  se i s e x c l u s i v e l y  clitics  J e me Is  a r e used  vois.  ' I see myself.' b. * J e s e  vois.  a 3 person c l i t i c .  for first  and s e c o n d  The  person.  'You d. * e.  Tu  see se  yourself.'  vois.  Nous nous IP 'We  see  voyons.  ourselves.'  f. * N o u s s e  voyons.  g.  Vous vous voyez. £p 'You s e e yourselves.' h. * V o u s s e v o y e z . We  see,  is  concerned,  to  what  of  evidence  the  therefore,  taken  as  Consider (179)a.  as  f a r as  phonological  happens with  other The  no  that  in favor  regular of  a  the  feature  variation clitics.  principled  composition  i s involved,  This  i s the  distinction  of  contrary  first  piece  between se  and  clitics.  existence evidence the  of  the  so-called "inherent"  for a different  sentences  in  status  se  for this  can  I I s'en va. 'He i s leaving.'  b„  Paul s'evanouit toujours a 'Paul always f a i n t s at the  c.  J e a n s ' e f f o r c e de b i e n f a i r e . ' J e a n s t r i v e s t o do w e l l . '  d.  Le c h a t e a u s ' e s t e c r o u l e . 'The c a s t e l t u m b l e d a w a y . '  e.  Les e n f a n t s se chamai1 l e n t . ' 'The c h i l d r e n s q u a b b l e . '  also  clitic.  (179).  l a vue du s a n g . s i g h t of blood.'  se  be  In  these  sentences  reflexive can  be  to  there  relate  assumed these  simply  entry  of  exist  independently  (180)a.*  I I en  to,  i s no as  that  noticed the  intransitive  va  of  argument  position  i n Kayne  clitic  verbs.  In  se  for  the  (1975s386).^  i s part  fact,  most  bien  faire.  of  the  of  them  It lexical do  not  se.  (lui(-merne)). h i m(se1f)  b. * P a u 1 c. *  Jean e f f o r c e  d. * L e e. *  chateau  Les  Finally, there  Borer  i n the  objects.  She  Spanish.  This  similar  accounted The  ecrou 1 e  (1984s1£3) n o t e s  the  interesting  doubling  otherwise  refers in particular of  very  to  allow  Romance  to  that  reflexive  such  R o m a n i a n and  properties specific  in other  with  fact  doubling  River  Plate  reflexive  l a n g u a g e s must  of  se  be  for. that  se  does not  feature combinations  the  same f u n c t i o n a s  recovering  missing  an  reason  obvious  (eux(—memes). them(selves)  clitic  languages which  cluster  de  (1 u i (-merne) ).  chamaillent  reflexives  fact  different fulfill  (lui(-mime))  a  enfants  (1 u i (-mime) ).  i s complete absence of  clitics  and  evanou i t  vary  in i t s surface  suggests other  that  clitics,  syntactic information. for this  s t a t e of  form  i t does  not  namely t h a t Intuitively,  affairs.  for  of there  If i t i s true  is  that  the it  NP  object  must  be  position  bound  available  i n i t s governing  fi-binder  works as  has t h e p r o p e r t i e s o f an anaphor,  i s the subject-  i f the subject  itself  recoverabi1ity  of the missing  (136)  t o account  designed  clitics no  i s not needed  missing  category  and  In other  was  everything  for the  Therefore,  f o r the recovering  for reflexive  only  words,  sufficient  object.  the  then  condition  function of  se simply  because there i s  f i nd s s u p p o r t  i n the e x i s t ence  i n f o r m a t i on.  Th i s p r e 1 i m i n a r y of  inherent  of  other  s e . The  similar  illustration  cone1 us i on sentences  verbs  of  (179) —  i n Standard  o f se not even  being  and  French * 3  able  there  —  3  i s a  wealth  are the perfect  to enter  i n the r  formation for  of a c-chain  this  process  Following with the  empty  writes: be  and  particular It the  (1984),  NPs  a n NP  will  interpretation  (181)  CVP  C  v  take  be  both  for clitic se  + V  II [  verbs. an  Cof t h e  who  se. N P  e*  doubling  f o r the anaphoric Borer  by v i r t u e  ] J  status of  (1984:122)  R-expression,  expression,  (1984),  necessary  the absence of  which has  of the  sentences]".  p o s s i b l e t o make a c o n c r e t e  a n a l y s i s of Borer  representation  we  position  available.  of reflexive  an a n a p h o r i c  i s now  i s not  se as an argument  object  "Such  free,  t o happen  Borer  reflexive  s i n c e t h e argument  suggests  proposal. the  I assume  following  to  This  representation  coindexed ©-role  with  the  ( i n the  gender,  [VP>  the  a combination  tw  c l i i  + V  the  that  assigned clitics  of  the  fact  that  se  i t i s associated to  that  i s given  features  subject.  that  anaphor  since  in  The  to  of  the  the  NP. in  is  not  with  the  The (182),  [person,  where c l _  number,  3 3  is specified  acts  reflexive (137) se  the of  cannot  i t i s governed  condition  as  features  e m p t y NP  characteristic  absorbs  S i  [IMP  (181)  latter  In c o n c l u s i o n , respect  3  C+anaphor] empty c a t e g o r y  finally,  3.6.  V)  but  the  a  representation  recoverabi1ity  only  NP  for regular  ©  forces  of  illustrates  Case3»  (182)  The  object  9-grid  representation represents  simply  in  be  PRO,  the  se  does not  to  be  This bound  necessary  missing  by  shares with  anaphoric.  CNP,VP3  element  the  f o r the  as  for  object.  the  the  Note,  pronominal  verb. act  as  a clitic  reasons stated other  clitics  above. i s that  with The it  Case.  Summary The  doubling concept  main  assumption  p h e n o m e n o n was "clitic"  which  that  guided  that  there  encompasses  by  our  a n a l y s i s of  exists both  i n the  subject  the  grammar clitics  a and  object  clitics.  Under t h i s  as  distinct  syntactic  two  Chapter a  £,  simple  clitics  i t was  argued  manner f o r t h e and  argument  object  position  Pi n a t u r a l instances  that  are  consequence of  whether they  involved  subject  i n 3.1  and  3.£  Case a b s o r p t i o n  and  developed  involved  i n the  of  way  this be  a  seen In  account  in  subject  view  in a  or  pro  in  a  Case t o  was  the  that  unified  object  particular  of  could  license  particular  I suggested  assignment  are  same e l e m e n t .  between  they  analyzed  a  clitics  to.  clitics  where  of  assumption  related  should  is  this  i n the  doubling  done  types  symmetry e x i s t i n g  of  was  both  realizations  clitics they  view,  clitics.  formal way  way  to  c l i t i c  This  approach describe  doubled  to what  overt  NPs. I n 3.3, doubling  I have s t u d i e d  constructions  concluded  that  parameter  referring  related  these  argument of  this  conditions  on  the  One  effects  of  the  extraction clitic.  of  an  to  the  level  i n the  particular  NP  of  this  that  at  which  type  of  i s coindexed permitted  When t h e y  either  at  LF  or  are both  or  be a  of  chain  clitics  definition  languages/dialects. or  should  formation  appearance of  possibilities  languages/dialects  possibilities  E x t r a c t i o n s are  in syntax  extraction  i n seven  enter  definition  the  with barred  allowed, but  handled c l i t i c  never  by  and  a c-chain. gathers  block and in  a the The  a l l the examined. any  governed  by  certain  they only  of  and  previously  i s to  out  can in  happen syntax.  a  This at  was t a k e n  which  this  c-chains  type  exhibits  that  a parameter  c a n be b r o k e n  both  subject  doubling  and support  Pied  Noir  and o b j e c t  on e x t r a c t i o n from  parallel  involving the  i s involved.  o f a n a l y s i s comes f r o m  restrictions exactly  as evidence  these  level  Confirmation f o r  French  which  doubling.  The  two c o n s t r u c t i o n s a r e  the unified  analysis  proposed  here. In t h e l a s t problematic  c o n s t r u c t i o n s were  were concerned found in  assumption  can  be v i e w e d  CNP,IP3  that  Inversion  involved a lexical  as being  chapter,  briefly  t h e apparent  Complex  no d o u b l i n g  the  subject  some  that  interpreted relation  doubling  This  NP s u b j e c t outside  i n core  of a c-chain.  argument  we  phenomenon  that  i n Complex  there i s ,  Inversion  o f IP, with clitic.  pro i n  I n 3.5, I  se i n French  This  and  i s , i n turn,  grammar a s c h a r a c t e r i s i n g an  between t h e e x t e r n a l  7>  w a s made p o s s i b l e b y  problems r e l a t e d t o r e f l e x i v e  i t i s not part  potentially In  and c o n c l u d e d  there.  generated  two  examined.  l i c e n s e d by t h e i n v e r t e d s u b j e c t  discussed argued  with  i n French  fact,  two s e c t i o n s o f t h i s  and t h e  anaphoric  internal  argument. The important  following chapter issues raised  i s devoted  t o some  by t h e d i s c u s s i o n s  r e s i d u a l but  i n Chapter  £ a n d 3.  Footnotes to  1.  Some o f  3.3,  Chapter  the  3  material  i s presented  in a  in this  chapter,  condensed  and  especially  form  3.1  i n Roberge  and  (1986a;b)  respect ively.  £.  There are  cf.  3.  £(30),  I use  since for  4.  The  out of  Renzi  for  with  i s , l p , and  & Vanelli  t r i p a r t i t e expansion  data  for this  (198£),  the  person  dialect  Carro11  of  one  £p  in  Trentino;  (198£).  S  (=IP)  adopted  come f r o m  (198£) a n d  pronouns have the  used here  in  are  Safir not  crucial  my  Laurendeau, own  Neron  i n t u i t ion.  following  surface  &  Note forms  that in  dialect s 3sm s 3sfs 3pm s 3pfs  5.  clitics  discussion.  third  this  and  differences  Fournier the  subject  3.1.4,  the  the  our  no  But  see  that such  Nernni  i1 elle i l s elles  (1973),  it is difficult facts.  (1983s£6)  to  Ci D Call  >  lil  >  Lil  cited  Nevertheless,  in their  > >  in Carroll  build  an  paper  who  argumentation  Dubuisson,  acquisition  (198£),  on  Ernirkanian dislocations  points  on  the  &  Lemay and  basis  topicalizations comporte en  (i)„  un  SN  (i)a.  This authors  are  pause".  sujet  g e n e r a lenient  du  sujet  t e l qu'illustre n i accent sur l e  i lcompte j u s q u ' a d i x . he c o u n t s u p + t o ten counts to ten.'  points  observe that  what  that  at o t h e r i n t e r e s t i n g they c a l l  i s called  here  i n f r e q u e n c y o f use language.  The  decreases from  but  language  the written  closely  to the standard  6.  Tellier  in  C.Fr.  (1986)  that  a construction exist  use  the  "subject  during  language  s u g g e s t s an  could with  be  used  as  analysis  t o argue  a subject  position  at school  —  is and  in  written  i t increases  Interpretations this  i s so  c o r r e s p o n d s more  dialect.  some f l o a t i n g  final  learned  —  i n the  6 whereas  t h e same p e r i o d .  The  sujet"  language  doubling  £ t o grade  du  doubling"  i n the written of subject  grade  facts.  "reprise  i t seems r e a s o n a b l e t o assume t h a t  because  sentence  clitique  ni  the spoken  There  un  disloque,  study also  spoken  open  et  reprise  Se bouton i lsert a alume le jeu. (sic) t h i s b u t t o n i t s e r v e s t o t u m + o n t h e game ' T h i s b u t t o n i s u s e d t o s t a r t t h e game.'  different  in  sujet  comporte  i n ( i ) and  language  phrase avec  ne  Marc  b.  "Une  elle  'He  the  (NP)  Pi l ' o r a l  const ituant  as  notes  clitic  quantifiers in  ( i ) .  of floating  quantifiers  f o r the presence of pro and  without a  i n C.Fr.  that  lexical appear  in  NP. in  c!5G  ( i ) a.  l i s o n t p a r l e a u d i r e c t e u r t o u t l e rnonde. 'They a l l spoke t o t h e d i r e c t o r . '  b.  Tellier empty must in  I l s ont pas p a r l e au d i r e c t e u r personne. 'None o f them s p o k e t o t h e d i r e c t o r . '  reports  category  that  a quantifier  at S-structure.  apparently  ( i i ) are acceptable  the  ones  in (iii)  ( i i ) a.  b.  are  with  personne  Crucially,  C +pronorninal1.  be  like  must  this  empty  Consequently,  p r o i n argument  bind  an  category  the sentences  p o s i t i o n whereas  barred.  pro± i l s o n t p a s p a r l e a mon p e r e ' N o n e o f t h e m s p o k e t o my d a d . ' 4  J e l e s * a i p a s v u s pro± 'I saw none o f them.'  personne.  personne.  ( i i i ) a . * L e s e n f a n t s o n t p a s p a r l e a mon p e r e p e r s o n n e . ' T h e c h i l d r e n h a v e n ' t s p o k e n t o my d a d . ' b. * J ' a i p a s v u l e s e n f a n t s personne. 'I h a v e n ' t s e e n t h e c h i l d r e n . ' Similarly, are  (iia)  i n ( i v ) , where  personne  binds  PRO  grammati ca1. (iv)  The  the sentences  a.  L e s e n f a n t S i o n t p a s v o u l u PRO± p a r l e t - a mon p e r e personne. ' N o n e o f t h e c h i l d r e n w a n t e d t o s p e a k t o my d a d . '  b.  J ' a i p a s demande aux e n f a n t s d e PRO* me r e n c o n t r e r personne. ' I d i d n ' t a s k a n y o f t h e c h i l d r e n t o m e e t me.' t  interesting  fact  i s excluded  present,  i.e.  here  i s that  a sentence such  when a p h o n o l o g i c a l l y r e a l i z e d  when t h e r e  (v) * L e s e n f a n t S i  ils*  i s subject  a s t h e one i n NP s u b j e c t i s  doubling.  o n t p a s p a r l e a mon  pere  personne.  This  i s accounted  personne les  in  enfants  for  i f we  ( i i a ) and occupies  that the  assume t h a t there  CNP,IPD  pro  simply  can  i s no  position  be  pro  bound in  occupied  by  (v)  by  since  pro  in  ( i i a) .  7.  This  dialect  literature  has  except  (1979sfootnote 8  and  subject  3s/p  below.  and  A  point  exhibits  pronouns  only  property subject  subject  the  South  encore  be  West  sous  of  i s the and  to  object  fact  et  qui  principal  floun  cf.  also  the  £s  than  the  une  is  the  having  French  this  this  literature  on  the  dialects  construction. i l s'est  introduit  habitude  qui  priceder  de  consiste a  faire  repete  emphase".  par  It  i n French,  other  gascon  French  aware of  a l s o have t h i s  du  PN  doubling.  I am  doubling  yl  "Dialectalement,  1'influence  in  forms of  that  object  It appears that France  Ztj  or  language that  writess  tres vivante,  pronorn c o m p l e m e n t  —-> Ct/iD — C i II > C€13 > C i II —•> C€1I1  here  as  dialects  more w i d e s p r e a d  suppose.  (1956s690)  Sud-Ouest,  or  remarks such  the  (198£s56,footnote 1 7 ) 5  in this  doubling  respect  seems t o would  Bourciez au  With  much a t t e n t i o n i n  following surface  tu i1 elle i1s elles  noting  Romance d i a l e c t  property.  from  worth  both  Jaeggli  Note the  £ss 3sm s 3sfs 3 pm s 3pss 8.  received  f o r some s c a t t e r e d  10)  footnote  not  He  y a  est le  provides  the  following (i)  examples:  M o n t a i gne Que je 11' escout e a t oy . that I c l l i s t e n t o you 'I 1 i s t e n t o you. ' t  (ii)  Moliere Comment, t u me± t r a i t e s a moi ± a v e c c e t t e how you c l treat t o me with this ' W h a t ! Y o u t r e a t me w i t h s u c h h a u g h t i n e s s ! '  According Bearnais  t o Bourciez  i s t o be a t t r i b u t e d  Lanly PN  (1970:222)  i s t h e case  many f a c t o r s object  to the influence  proposes  something  i n RP S p a n i s h  interact  i n Gascon and o f Spanish.  similar  with respect t o  NP m u s t  matters  here.  often  9. A c c o r d i n g t o t h i s maximal  projection  definition,  dominating  (1980:1982)),  F o r example, t h e  I do n o t d i s c u s s t h e s e  a c-commands Q i f t h e f i r s t  a also  projection  In the following  X' ' / \ X' ' Z / \ X' W / \ X Y  i n PN F r e n c h .  be C+animateD.  c o n t a i n 13. A m a x i m a l X ° , X".  (cf. Jaeggli  with t h e g r a m m a t i c a l i t y judgements o f  doubling constructions  doubled  of  "habit"  French. As  not  (1959:458),. t h i s  hauteur. haughtiness  dominates  13, a n d a  i s the highest  structure,  does  projection  £59  X c-commands Z as  10.  Only  the  Whether the not  affix  11.  The  Tigre  in  PPs  1£  Sportiche  Kayne's  and  i s discussed  prefix,  i s presumably  or  here.  e v e n an  a very  infix  language  is  specific  out  that  examples are  clitic  doubling  from  Borer  i s not  available?  (1983) s u g g e s t s t h a t in  (i) i s another  French  Complex  Inversion  p o t e n t i a l counterexample  to  Pierret  would a  viendra-t - i i i ? come—FUT—3s c l P i e r r e come?'  The  be  true  i f i t can  construction  obvious.  See  3.4  number o f  relevant  The  a  Lebanese Arabic  points  'Will  14.  this  clitic  generalizat ion.  (i)  13.  the  (1985:6-8).  i n Hebrew.  illustrated  really  of  Safir  clitics.  ( 1 9 8 4 . 9 7 ) who  This  W 5 cf.  Y and  is a suffix,  since  of  as  status  clitic  relevant  property  well  be  shown t h a t  involving subject  for a  Complex  doubling;  Inversion this  is  is  not  discussion.  bars allowed  f o r each  projection is  not  here.  same c a n  causative  be  said to  constructions  hold  i n French  for dative where the  Case assignment preposition a  is  in  used  15.  i n a very  Borer  (1384.46-48)  construct  state  structural property  s i m i l a r way; c f . 2 ( f o o t n o t e 3 2 ) .  argues that  i s assigned  conditions.  o f a l l nouns  g e n i t i v e Case  by t h e head  noun  The g e n i t i v e Case i n Semitic  under  feature  languages,  i n the specific  i sa potential  h e r e Hebrew and  Arabic.  16.  This  have  hypothesis  concerning  nominative  i n t e r e s t i n g consequences f o r i n f l e c t e d  Portuguese. having  Suppose  that  t h e IH-CAD f e a t u r e .  Then,  t o CNP,IPD  infinitival  clauses.  The d i f f e r e n c e  in this  Portuguese  inflected  AGR w i t h  C+C3  a  CNP, I P 3  Case  suggestions C+CA3  assigned  17.  would  This  —  would  —  of  would account f o rCase assignment  theory  o n how t o h a n d l e f r e e  feature  E n g l i s h and  but not i n E n g l i s h  infinitives  this  between  be t h e p r e s e n c e i n I N F L i n  a s s i g n m e n t t o t h e same p o s i t i o n that  infinitives in  i n any language, even i n  infinitives  i n inflected  Note however  the  respect  feature.  could  C a s e w o u l d a l w a y s be  assigned  Portuguese  assignment  TNS i s a l w a y s s p e c i f i e d i n UG a s  potentially  to  Case  i n P o r t u g u e s e and t h e non i n English  infinitives.  c a u s e s a p r o b l e m f o r my inversion  i n 2.4 s i n c e  be i n VP a t S - s t r u c t u r e  where  only  Case i s  o r checked.  Jaeggli's  (1385) h y p o t h e s i s  analyzing  dummy C a s e  markers as  "Case b r i d g e s "  does not  provide  the  d i f f e r e n c e between s u b j e c t  His  account  usual  i s based  requirement  nominal  element-  <i)  Jaeggli  that This  the  doubling  formation  a s i n g l e Case requirement  of Case be  doubling-  p a i r s and  assigned  t o at as  for  the most  one  i n( i ) .  pairs;  (a,6)  13 a n d  a cannot  both  nominal  are considered  markers  ( p r e p o s i t i o n s ) a r e n a t u r a l l y not an  nominal  be  Clitics  in (ii),  object  i s formulated  a Case f e a t u r e . a r e two o r more Case  (ot,G>  Thus,  and  explanation  (1985s£4)  L e t a be If there then  on  a principled  elements  example of object  elements.  w h e r e a s dummy nominal  doubling  Case  elements.  in River  Plate  Spanish, (ii)  three a. b. c. The  Loi cl ' We  Case  vimos a Juan*, saw-lp to saw J u a n . '  pairs are  formed:  (Case (Case (Case  feature feature feature  of the verb, c l i t i c lo) o f t h e v e r b , p r e p o s i t i o n a) o f t h e p r e p o s i t i o n a, o b j e c t  Case  feature  of the verb  nominal  element  preposition object Now French  —  i s associated  the c l i t i c  i s associated  with  —  and only  with  the Case one  NP  Juan) only  feature  nominal  one of  element  the —  the  NP. i n an one  in  example of s u b j e c t (iii),  doubling  like  the C o l l o q u i a l  (iii)  two  Case  a. b.  (Case (Case  This  Jeant  i l l vient. cl come-3s ' J e a n comes.'  p a i r s would feature feature  situation  associated  be  o f INFL, o f INFL,  i s ruled  with  subject  NP.  subject  clitics  assumption  not i n t r i n s i c a l l y  18.  C f . May  (1977),  that  indefinite argument  NPs and r e g u l a r  a t L F . T h e same r e m a r k  of  Rochemont  quantifier  like  This  sentences  clitics  feature i s and t h e  be t o assume  This  would  and o b j e c t  go  that  against  clitics  (1984).  similarity  I simply  assume  that  f u r t h e r comments. i n the behavior  this  Note,  of  q u a n t i f i e d NPs c a n be u s e d  as an  them a s q u a n t i f i e r s u n d e r g o i n g  holds  o f Focus  (see below);  c f . Heirn  Raising (198£)  i n t h e sense for a  approach.  The s e n t e n c e  (88a).  would  elements.  subject  of analyzing  (1986)  the c l i t i c  problem  i n (88) w i t h o u t  QR  19.  this  Homstein  the striking  i n favor  different  elements,  one Case  different.  o f a l l t h e NPs that  Jean)  ( i )s i n c e  a r e not nominal  are  however,  o u t by  One way a r o u n d  basic  true  c l i t i c i1) s u b j e c t NP  two nominal  our  is  formed.  i n (93c) c o n t a i n s t h e one found  i s excluded  containing  a definite  universal  i n t h e RP S p a n i s h  and s h o u l d  a collective  example i n  n o t be c o n f u s e d NP,  which  with  similar  i s not a q u a n t i f i e r  as  i n  ( i ) .  (i)  The  T o u t l e monde± i l e t a i t all t h e people c l was ' E v e r y b o d y was t h e r e . '  NP i n ( i ) b e l o n g s  collective fami l i e ,  NPs which  classe,  ( i i ) a.  However,  la. there  ±  to theclass  o f NP c o n t a i n i n g  a r e undoubtedly  other  not q u a n t i f i e r s  such  as  foule,e t c .  L a f a m m i l e i e l l e i e s t venue au p a r t y . 'The f a m i l y came t o t h e p a r t y . '  b.  La classei 'The c l a s s  e l l e i e s t devenue i n c o n t r o l a b l e . became u n c o n t r o l a b l e . '  c.  L a f o u l e i e l l e * c r i a i t t o n s l e temps. 'The c r o w d was s h o u t i n g a l l t h e t i m e . '  t h e NP i n ( 9 3 c )  i sa q u a n t i f i e r  similar  t o t h e onei n  (iii). (iii)  Subject but  T o u t homme i (*i1±) a i me trava i l l e r . all man c l l i k e s to+work ' A n y man l i k e s working.' doubling  impossible with  collective they  quantifiers,  Neron  F o r an account  fronted Lefebvre  collective  WH w o r d (138£),  & Fournier  Compare  Roberge  Note  s i n g u l a r even  (33c)  that though  t o ( i ) . Cf. also  (138,2).  o f t h e que appearing  i nquestions  NPs, ( i ) and ( i i )  (33) and ( i i i ) .  NPs a r e u s u a l l y g r a m m a t i c a l l y  aresemantically plural.  Laurendeau,  £0.  i spossible with  i nother  (1383),  optionally  French  and Kayne  along  with  dialects, cf. (1384s£31).  £1. a  Jaeggli  (198£:49)  clitic-doubled (i)*  This  provides  NP c a n n o t  t h e example  be f o c u s e d  i n <i) t o a r g u e  i n RP  Spanish.  Yo l o t v i a JUAN*. ' I s a w JUPiN. '  i s expected  doubling  since  constructions  LF e x t r a c t i o n s a r e excluded in this  C f . £.4 a n d £ ( f o o t n o t e £ 7 ) f o r m o r e o n f r e e  the  impersona1  £3.  Borer  (1984)  (111b) c a n n o t  Cornorovski  £4.  (or neuter)  provides  clauses  independent  evidence  as a resumptive  (1986) w h e r e t h e same c o n c l u s i o n  whether there  object  i n v e r s i o n and  clitic.  be a n a l y z e d  Note however that  from  dialect.  ££.  in  that  (pe) c i n e  the c l i t i c  pronouns  cf. also  i s reached.  i s excluded  i s a clitic  that  from  o r not5 c f .  c  relative  Dobrovie-Sorin  (1985s6).  £5.  Note however t h a t  possibility  i t i s not n e c e s s a r i l y true  o f e x t r a c t i n g an element  constructions  entails  without  extracted  allowed  only  out o f  the availability  element.  clitic-doubled  One c o u l d  of  that the  clitic-doubling  clitic-doubling  imagine  a language  that  variables.  Modern Hebrew a p p e a r s t o p r o v i d e  examples o f t h i s  sort.  According only sel  to  i n the  Borer  (1984),  construct  state  as  be  (i)  beit —o s e l ha -more* house h i s of the teacher 'the t e a c h e r ' s house'  to  Hebrew a l l o w s  preposition  the  complement  a p p e a r s on  (i i)a.  ' i t with  b.*  cannot of  a  that  (i) because the  Case-mark the  be  inserted  preposition  noun.  Dan Dan'  out  of  doubling  not  The  well  as  be from  (iii)  in questions.  extracted a  be  it  follows  empty when  constructions  relatives, can  must  and  a  ±  They a r e  word  i n PPs  him  i n Modern Hebrew.  PP  with  from a  the  possible  are  very  only  i n t e r e s t i n g point  construct  state  in  free  i s that  with  a  a  Wh  clitic  as  clitics  kaniti ma± i£e -xasavt b o u g h t - I what t h a t t h o u g h t — y o u 'I bought w h a t e v e r you t h o u g h t  'a l -av about i t about.'  ±  Ce]  t  'anaxou 'ozrirn l e - k o l mi± se - b e i t -o± Cell i n i s r a f we help t o e v e r y who t h a t h o u s e h i s burned 'We h e l p e v e r y o n e whose house burned.'  Although grammatical  A  complement  preposition.  restricted  £6.  preposition  Ce3 ±  —Ox  'it-ot 'with  Extractions  (iv)  clitic-doubling  ±  This  clitic  in order  in  can  that  inserted  Modem  fourth  ( i i b ) i s ungrarnmat i c a l , with  a  the  sentence  in  (iii)  clitic-doubled variable.  possible  analysis  suggests  itself.  It could  be  is  suggested  that  constructions, (i) is  *  due  [  C  t h e i m p o s s i b i l i t y o f e x t r a c t i n g out o f i . e . the ungrarnmaticality  P Whi  variable. occupied  that  However,  a properly  Cf. a l s o  two o p e r a t o r s  binds  (Wh  f o r independent  &  two v a r i a b l e s o r ,  and c l i t i c )  reasons,  c a n be n e i t h e r  i s neither  bind  one  the position  a n P i - n o r a n P' - p o s i t i o n ,  a v a r i a b l e n o r an Pi'-binders c f .  0.  a i s  (1985) f o r an e x t e n s i o n  out o f object  doubling  of this  constructions  analysis in certain  dialects.  Jaeggli a path  differences different  This  further  i f a g o v e r n s (3 a n d  1981s£73).  Jaeggli  extractions  follows  30.  operator  g o v e r n s (3 i f a n d o n l y  (Chomsky  Spanish  £9.  11  ( 1 9 8 4 ) a n d £'.£.5.  lexical  to  ( i ) , f o r instance,  o f t h e B i s e c t i o n P r i n c i p l e (Koopman  by t h e c l i t i c  hence t h e c l i t i c  £8.  t i cli+V  1 9 8 1 ) w h e r e t h e Wh  alternatively,  £7.  Z P  to a violation  Sportiche  Borer  [  of  doubling  pointed  o u t t o me  that  s i m i l a r t o mine w i t h  respect  between t h e two a p p r o a c h e s  conceptions  differs  from  conditions  o f empty  his line  follow  categories;  the analysis  on t h e n a t u r e  to  research  (137).  The  crucially  from  c f . 1.3.  i n Roberge  o f what  of  (1986b) where  can occupy the  a  argument simpler  31.  p o s i t i o n a r e added and c o v e r s  Cf. Hurtado  alternative (142),  variables the  t h e same  i n which  the extracted  i n IP. T h e i r  sentences  Complex  Subject  argues that  33.  Kayne  NPs  (1984) s u g g e s t s  on  9-role  NP  i s because t h e Case  Visibility  clitics  a lexical  also  the inflected  That  i s based  hypothesis assignment  do.  i s also  called  (i97251975)  ft~chains.  i s derived  (1981),  NPs i n P i ' - p o s i t i o n 5 s e e f o o t n o t e  35.  In C o l l o q u i a l French  Inversion  which  clitics; cf.  from t h e i s a  It i s therefore  for  who  i n a position external  and i t s a f f i x e d  filter  that  construction.  i s found verb  bind  effects  an a n a l y s i s o f Complex  i n Chomsky to  NP  in  on t h e f a c t  i n Kayne  a r e o n e and t h e same  to  34.  a n a l y s i s , do n o t  (Subj-Cl-Inv)  the lexical  4.  i n my  without  where not o n l y  C! . c! n  (1985) f o r  the quantified expressions  without  Inversion  they  IP but a l s o  analysis i s  i n (14£) d o n o t e x h i b i t w e a k c r o s s o v e r  Inversion  Clitic  and J a e g g l i  argumentation  w h e r e a s t h e same s e n t e n c e s  32.  The p r e s e n t  data.  (1984a;1984bs1985)  analyses  i.e.  t o (137).  condition  not  relevant  10.  yes/no questions,  the interrogative  particle  tu. a p p e a r s  ( i ) a. b. Contrary French,  can  to  what  this  clitic  and  both  of  course  in  right  of  a  case  with  i s barred  from  together  with  a  lexical excluded  preverbal  NP  Standard  Wh-questions.  referential  Inversions  i n Standard  and  following  to  examples  Rizzi  (1985),  say, from  a subject  clitic,  which  f o r example, Kayne  Manzini  about  (1975s395).  (1983)  among  middle  Cf.  se  Aoun  others.  C e l a se d i t s u r t o u t pour e n n u y e r l e s gens. 'That i s s a i d e s p e c i a l l y t o annoy people.'  b.  L e s n o i s e t t e s s e mangent b i e n en p a r i a n t . 'Hazelnuts eat w e l l w h i l e t a l k i n g . '  c.  Un 'Pi  t e1 toy  j o u e t po u r r a i t s ' a c h e t e r po u r s o i - m i m e . l i k e t h a t c o u l d be b o u g h t f o r o n e s e l f . '  i s tempting clitics  to relate  discussed  is  French.  have nothing  dative  Inversion in  preverbal  c o n t r a r y t o Complex  I will  It  verb.  Pierre va-tu venir? Marie va-tu chanter?  P i e r r e i 1 va-t u ven i r ? Marie e l l e va-tu venir?  (i)a.  inflected  Complex  (i v)a. b.  the  the  II va-tu venir? E l l e va-tu chanter?  (1985s173),  37.  particle Quand Quand  ( i i i ) a. b.  36.  i s the  a l s o appear  subject  the  Pierre va-tu venir? Marie va-tu chanter?  ( i i ) a.* b.* It  to  this  phenomenon t o t h e  in Jaeggli  (1985)  ethical  f o r Spanish.  The  as  examples a r e from  Jaeggli  (1985511);  cf. also  Lapoirite  (19805£57). (i)a.  E s t e c h i c o n o me c o m e ! ' T h i s k i d w o n ' t e a t ( f o r ) me.'  b. No t e me p o n g a s a g r i t a r a s i , p o r f a v o r . 'Don't s t a r t s c r e a m i n g l i k e t h a t ( o n me), c.  38.  Te l o l l a m o a l d o c t o r ? 'Should I c a l l t h e doctor  Some o t h e r  s'achamer club  verbs  with  'to persist',  together',  s'entraider  se d e d i r e  'to help  one  inherent  f o r you?'  ses s ' a c c r o u p i r  se b l o t t i r  ' t o crouch',  'to retract', another',  lament',  mistaken', se  se rnefier  se pavaner  specialiser  'to specialize',  'to  '  'to frolic'  'to rely', 1ament e r  se meprendre  ' t o be  se r e f u g i e r ' t o take  se targuer  squat'  se c o t i s e r  ' to r e v o I t ' , se  ' t o be w a t c h f u l ' , 'to strut',  s'ebattre  se f i e r  s' i n g e n i e r ' t o c o n t r i v e ' , s' i n s u r g e r 'to  please.'  'to pride  refuge  oneself  CHAPTER 4: C L I T I C S AND AGREEMENT  4.0.  Introduction It  i s a well-known  object other  pronouns  fact  i nFrench,  that  clitics  I t was a r g u e d ,  a r egenerated  argument  position.  theidentity  For  holds  an object this  sense,  different structural  that  they  o f t h em i s s i n g  which  we f o u n d subject  verb  of their  £, t h a t  subject  serve t h e  proi n external  serve  t h epurpose o f  information  We h a v e a l s o  f o r t h e i n t e r n a l argument  clitic,  reason,  hypothesis  proved  pronominal  by t h eempty argument.  observation with  In this  P i r a h a , and  onto t h e i n f l e c t e d  i n AGR o f I N F L w h e r e t h e y  o f l i c e n s i n g an empty  syntax  Fiorentino,  e s p e c i a l l y i nChapter  function  recovering  weak f o r m s o f s u b j e c t a n d  Trentino,  languages a r ec l i t i c i z e d  sentence.  in  MARKERS  represented  seen that  position  associated  i s affixed to theinflected  itsuseful  clitics  t o subscribe  and object  this  verb.  to the  clitics  a r e two  r e a l i z a t i o n s o f a same s y n t a c t i c e l e m e n t . T h e differences  t o be a good  between  basis  i n t e r n a l and e x t e r n a l  arguments  f o r characterizing the differences  between t h e i r  associated  There  further  clitics they  is a  and  achieve  throughout generated subject of  object their  the on  distinction  eliites; status  present  the  clitics  clitics. be  made b e t w e e n  subject  to  do  with  in  verbal  head.  their  affixes.  object  We  this  way  have  clitics  Obviously  base  the  are  which  assumed base  i s not  true  p o s i t i o n i s somewhere  of  i n PiGR  INFL. In  the  options  first  that  section  arise  of  i n order  the  subject  the  analysis  adopted  here  subject  clitics  and  with the  ones The  with  type  pronouns  found  in  second  respect  offer  of  property  Subject Let  us  chapter, for the  the  extended  different  clitic  I concentrate be  having  on  to  status  French  other  syntactic structures  i s concerned or  the  the  doubling  with  the  inflectional differences  status  markers.  of  but  languages similar  regard  to  of  to  the  null  clitics  I wish  between the  so-called concatenative  languages with the  account  could  agreement  l a n g u a g e s and  and  to  present  French.  section  to  the  i s explored.  b r i e f comments on  agglutinative  4.1.  as  work t h a t  verbal since  i t has  to  to  Romance  or argument  phenomenon.  Cliticization  return  to  subject  property.  His  element  INFL must  be  Chomsky's  <1981) a c c o u n t  analysis relies attached  to  the  on verb  the at  of fact  the  null  that  some p o i n t  the in  the  derivation. features verb in  on  R  surface. to  of  ungoverned  languages rule  R  ungoverned and  this  gender,  and  to  the  argument allow  Hopping)  subject  an  this  the  languages,  i t to  a p p l i c a t i o n of  barred. was  Another  be  empty s u b j e c t  believed  INFL  i n VP  Case t o  an  Now  hold  is in a NP  chapters,  following associated  the  clauses.  is  best  to  eliminate  null  made.  subjects  government  the  are  i s the  for  Hence, PRO  i f  is  structure  in  the  level.  In  level  for  consequently  r e l a t i o n s h i p that  and  i t can  l.£.1.  need  PRO.  relevant  subjects  different  be  occupy  by  to  as  subject  particular  subjects  choice  cf.  i s the  surface  null  VP  free inversion. assign  nominative  VP.  Following  suited to  can  the  characterized  S-structure,  i s the  in particular  r e m a r k s must with  to  c o n f i g u r a t i o n where  to  choice  to  INFL t o  property?  Form  for this  to  he  occupied  this  between n u l l  adjoined  according  previous  tensed  to  r u l e R and  reason  moving  conditions apply  Phonological  affixed  position in null  languages where r u l e R a p p l i e s at other  are  of  i s what  applies prior  Binding  i t follows that  null  inflectional  tense)  option  This  underlying  external  when t h e  fact.  of  reasons  in order  (Affix  because the  Chomsky c h o s e t h e  express  the  simply  number,  i n h i s account  One an  i s true  (person,  order  rule  This  £.4  of  the  and  First,  i s no  Chomsky this  analyses  presented  Chapter  the  empty  longer  taken  (1982),  position. subject  we  There  3,  in  the  the  category to  be  PRO  in  have seen t h a t i s t h u s no  p o s i t i o n by  the  pro  need  INFL  node. the  In fact,  licensing  this  type  a n d 3.3.4.  verb  t o be i n f l e c t e d  such  It follows  Pgain,  a way g i v e n  of  application,  in  t h e marked  i t seems,  each  the  case.  Furthermore,  between  and I t a l i a n  these  first  prior  pro i n subject  case  i s no  not apply i n  and i t s l e v e l  and S - s t r u c t u r e  difference  respect  to this  rather  position.  pronouns whereas r i c h  rule.  The  i n t h e way  French has  agreement i s  of pro i n Standard  f o r movement  t o S~  i n 3(137) on t h e  structure  two languages l i e s  f o rthe licensing argument  there with  INFL t o t h e  a s i f i t must  namely LF i n t h e unmarked  to clitic  sufficient  affixing  t o apply  i n the syntactic  language l i c e n s e s  recourse  the rule  i n fact,  o r not 5 c f .  a clitic  the well-formedness condition  say, French  difference  that  does n o t need  appearance o f c l i t i c s  between,  s e e m s t o be r e q u i r e d f o r  o f p r o b y PiGR, c o n t a i n i n g  2.2.  structure.  o f government  Italian.  Thus  o f I N F L t o VP d o e s n o t h o l d  anymore. Secondly,  i t was s u g g e s t e d  property  i s made p o s s i b l e  VP  to S-structure,  prior  government argument related  domain  property to free  i n 2.4 t h a t  the free  inversion  b y t h e p r e s e n c e o f t h e TNS e l e m e n t i n where  o f V. T h i s  i tcan assign  w a s made n e c e s s a r y  as i t i s conceived  inversion;  Case  i n the  as t h e null  o f h e r e c a n n o t be  c f . 2.3. The s e c o n d argument f o r  INFL-rnovement  t o VP, w h i c h i s b a s e d o n t h e r e l a t i o n s h i p  null  and f r e e  subjects I  directly  inversion  between  d o e s n o t go t h r o u g h .  propose t o approach t h e problem o f t h e a f f i x a l  nature of  subject  clitics  accomplishes features  the of  this  (Chomsky  t o INFL  Two  possible  1981);  that  we w i l l  which  inflectional  analyses  arises  o r B) t h e v e r b  briefly  the process  affixes  (Ernonds 1 9 7 8 , M c A ' N u l t y  two o p t i o n s ,  t o be  1983),  A)  INFL/PiBR  inflected  Before  exploring  examine t h e i n t e r n a l  structure  INFL,  4. 1. 1.  Internal Structure  Let INFL  us adopt  i n French  constituents. TNS,  subject  surface  must r e f l e c t  clitics,  Object  clitics  a r e base  + ne +  surface  order  o f I N F L a r e , among  particular  f o r INFL  i s t h uss  INFL \  TNS AGR / I \ (cl)(NEG)AGR  [Object  generated  i n VP a t D - s t r u c t u r e ,  /  the relative  marker verb  structure of of i t s  others,  ne. The  AGR,  basic  and i t s o b j e c t  in(la).  Clitic  (l)b.  the internal  including the finite  i s given  INFL a t t h i s  that  and t h e n e g a t i v e  Subject  st r u c t ure  INFL  The c o n s t i t u e n t s  (Da.  being  of  the hypothesis  ordering,  clitics,  of  by a s s u m i n g  i s t h e same t h a t  t o t h e verb-  moves t o V raises  i n French  Clitics  on t h e f i n i t e  must be e x c l u d e d level  + V3 + -CTNS, AGR> verb.  from  The  verb,  the structure  of representation.  A plausible  £75 where t h e s u b j e c t that  this  there  for  structure  i s a great  respect  and ne a r e o p t i o n a l l y g e n e r a t e d .  i s specific  deal  marker  a discussion of various  the  particles  subject  In  Rornagnolo,  as  i n Standard  such  negative  Note a l s o that different position  Verb  Italian  particle  i n INFL;  there  (lb) indicates that The c l i t i c ,  allows  inasmuch  (1984) by  In Trentino, p a r t i c l e (no).  follows the subject there  i s only  as  with  see R i z z i  dialects.  clitics  Note  c a n be o c c u p i e d  In C o l l o q u i a l French  simply  pas a d j o i n e d  clitic i s no t o VP.  a n d PiGR a r e  i n a sense,  i s i n a Spec  f o r the feature-agreement  AGR  showing  up  on t h e verb.  Verb R a i s i n g o r INFL  Finite  stands  marker  and t h i s  McA'Nulty  (1978).  French.  f o r instance;  follows the negative  the negative  elements.  independently  4. l.£.  always  French  variation  positions that  i n various  clitic  t o Standard  of cross-linguistic  t o the negation  negative  of  clitic  (1983)  presents  Raising  The e f f e c t f o r "object  Lowering? some a r g u m e n t s  (henceforth,  o f such  a rule  clitic(s)".  i n favor  FVR) a s p r o p o s e d  i s illustrated  of a rule i n Emonds  i n (£). "0C1"  <£>  FVRs  IP /  \  NP /  INFL' \  INFL I PiGR / \ COCl+VIU PiGR t  The  rule  clitics just  moves t h e to a  position as  PIDV  V  Q  I e i  i  inflected  position  p r e s e n t e d has  V" \  /  verb together with  adjoined  to the  the advantage  of the c l i t i c s  with  left  i t s object  o f PiGR.  of explaining  the  The  relative  r e s p e c t t o the agreement  i n (3), where t h e r e l e v a n t  items are  account  markers  underlined.  (3) P i e r r e  l e l u i donnera. cl c l give-FUT-3s ' P i e r r e w i l l give i t t o him/her.'  Consider (1983 s i 3 9 )  next  and  the sentences  the differnt  i n (4)-(6) from McPi'Nulty  positions  of the  underlined  elements. ( 4 ) a.  Jean  prefererait ne pas l e s a v o i r manges. p r e f e r - C 0 N D - 3 s NEG n o t c l t o + h a v e e a t e n ' J e a n would p r e f e r not t o have e a t e n them.'  b.  Jean 'Jean  c. * J e a n ( 5 ) a.  Jean  ne  les a cl have-3s has not e a t e n  ne  pas  les a  pas manges. not them.'  manges.  dit souvent l e s a v o i r manges. say-3s often c l to+have eaten ' J e a n s a y s t o have o f t e n e a t e n them.'  £77 b.  Jean  c. * J e a n (6)a.  souvent  1es  a  manges.  manges.  Jean  voudrait tous les avoir rencontres. want-C0ND-3s a l l c l t o + h a v e met ' J e a n w o u l d w a n t t o h a v e met a l l o f t h e m . '  b.  Jean  les a tous rencontres. cl have-3s a l l ' J e a n h a s met a l l o f t h e m . '  c. * J e a n If  les a souvent cl have-~3s o f t e n  the  tons  adverbs  les a  and  rencontres.  quantifiers  (pas,  are  generated  directly  under V '  are  accounted  f o r . The  infinitive  follow  these  position,  as  inflectional verb the  in  <£)  verbs  i n the  because the verb  stays  t h e head  of V'.  a  features through the  (b)  adverbs  sentences  and  However, FVR.  On  i s explained  as  The  follows  quantifiers.  The that in  we  (7).  second call  and  p l a u s i b l e approach  here  INFL L o w e r i n g .  (4)-(6)  contrasts examples  verb  operation  gets i t s  then,  contrast  such  verb  (1981) r u l e is  a  between  the  finite  i s Chomsky's The  (a)  w e l l as  (c) e x a m p l e s where a  adverbs  the  the surface  quantifiers.  in  i n i t s base  finite  ungrarnmat i c a l i t y o f t h e the  tons)  then  elements  precedes (a) and  as  souvent,  R  illustrated  (7)  INFL  Lowering.  IP /  \  NP / INFL  INFL' \ V '  I  /  e*  \  PlDV Q  V I  V / \ cl+V  The r e s p e c t i v e illustrated the  position  unless  the contrasts  the structural  is  revised  is  that  of the c l i t i c s  i n (3) i s a c c o u n t e d  o t h e r hand,  fiGRi  r-  and agreement  f o r , as i s t h e case  i n (4)-<6)  position  quantifiers.  f o r FVR. On  are left  unexplained  of the quantifiers  and adverbs  and m o d i f i e d . The p r e d i c t i o n  a verb  markers  (tensed o r not) always  T h i s i s an u n d e s i r a b l e  made b y t h e r u l e  follows  i n (7)  t h e adverbs and  prediction  a s shown  in  (4)-  (6) . The two a n a l y s e s p r e s e n t e d a b o v e do n o t t a k e clitics  into  account.  French  and I t a l i a n  always  appears  clitic  string  (8)  This  The i s s u e  dialects  under  as t h e f i r s t a s shown  in  i s as follows. study here,  subject  In both t h e a subject  member t o t h e l e f t  clitic  of the surfacing  (8).  I_l_ l e l u i d o n n e r a . cl c l c l give-FUT--3s 'He w i l l g i v e i t t o h i m / h e r . ' causes  clitics  problems  a r e generated  i n both i n PiGR.  a n a l y s e s assuming Both  FVR a n d I N F L  that  subject  lowering  place  the  subject  modify  clitic  the  possible  FVR  types  we of  with  the  assume,  in  then  in order to  of  to  the  verb.  It  overcome t h i s  is easier  i s suggested  in  internal  s t r u c t u r e of  INFL adopted  (9), which  (1986.4),  u n d e r Move a a r e  L'OCl+V] c o m p l e x  can  to  p r o b l e m . Pi  (£)  possible the  that  the  is above. two  substitution  only  be  and  adjoined  to  IP /  \  NP  INFL' \  / INFL / PIGR /  \ TNS \ PiGR  Scl  / COCl+VIU  E-ioth t h e  relative  position  of  Furthermore, tense  ordering  a d v e r b s and the  features  FVR  VP / \ PlDV V Q I ei  \ PiGR  t  inflectional  will  i n 3(137)  T h e r e seems t o  be  end  up  no  forces  i s interesting to  has  as  the  clitics  of  features  i t to  the  note  i n the apply  that  this  i n the  string  accounted included  s u f f i x e s on  late  evidence  s y n t a c t i c component It  of  q u a n t i f i e r s are  necessarily applies  condit ion  here  right  (9).  (9)  the  the  f o l l o w i n g Chomsky  movement  adjunction, PiGR a s  rule  modification  consistent If  to  the  and  the  for.  i n PiGR a n d  verb.  derivation.  1  The  in Phonelogical r u l e should  the  Form.  apply  in  i t i s conceived  of  grammar. that  FVR  f u r t h e r d e s i r a b l e consequences  as  i n the  grammar  of  French. that  Lobeck  i t can  French.  (1985s1988)  explain  In t h i s  proposes  a similar  t h e a b s e n c e o f VP  language,  a sentence  rule  anaphora like  and  argues  or e l l i p s i s  (10)  in  is  u n g r a r n m a t i c a 1. (10)  * M a r i e a vu 'Marie has  Lobeck VPs  argues  that  i n the sense  There  i s such  auxiliary as  other  hand,  proper  s a m e way,  as  or  LF,  The allows  that  no  are  t h e e m p t y VP  where,  i n Kayne  proper  whichever  and  governor level  i t is  and  both  Verb  for a variety  Raising  of  governor.  argued,  on  behave  u n d e r g o FVR  applies.  empty  where t h e y  f o r empty VPs  t h e ECP  of F i n i t e  proper  In French,  main v e r b s  (1975),  a  syntax  f o r empty VPs.  verbs  too.'  the occurrences of  requires  in English  governors  t o account  modulates  i n INFL d u r i n g t h e  modified rule us  ECP  governor  argued  i s thus at  the  auxiliary  There at  a  verbs  serve  l e f i l m et Paul a 0 aussi. s e e n t h e f i l m and P a u l h a s 0  can  the i n the  in  PF.  at S - s t r u c t u r e  5 3  in  phenomena,  (9) t h e r e f o r e at  least  in  French.  4.1.3. C o o r d i n a t i o n I n £.2.4  we  have seen  that  Rizzi  differences  b e t w e e n T r e n t i n o and  behavior  subject c l i t i c s  support c l i t i c s .  of  the existence of 3  Recall  the  two  (1984) a p p e a l s  Standard  French  in  to  certain  the  with respect to coordination different  sentences  in  (11)  types of subject and  (1£).  to  (11)  Standard Elle 'She  (IS)  French  chante et s i n g s and  Trentino a.  La c a n t a e l a b a l l a . ' She s i n g s and she d a n c e s . '  ta.*  La  Recording  canta  to  the  obligatory conjunct clitics  see  subject  explanation  e  balla.  Rizzi,  coordination; of  danse. dances.'  a  £.£.4.  clitics  f o r the  repetition  as  in  (l£a).  are  in  CNP,IP3  i s predicted to  form  i n the  section,  coordination  facts  reevaluated.  This  subject  are to  In  the  i n PGR  assuming  of  he  (l£b)  provide  and  the  clitic  i n the  On  the  hand,  i f French  and  be  later  cliticized  grammatical the  I offer  base  t h e m ; £)  which  FVR  an  second subject  the  verb,  i s no  then  zero  pro-  conjunct.  another  generated  to  since there  second  generation  can  a subject other  under  base  of  interpretation  Rizzi's on  of  during  the  the is  premises. INFL  applies late 3)  of  conclusion  three  i n P.GR  subsumes c l i t i c i z a t i o n ;  representations available syntactic  by  a n a l y s i s i s based  languages e x h i b i t i n g and  Therefore,  following  c l i t i c s are  derivation  i s barred  i n PiGR i n T r e n t i n o ,  PiGR n o d e o f  this  pro—form  ungrammaticality  (11)  In  zero  in  in  1) the  the  coordinate structures syntax  and  are  subject  rules.  remainder  of  this  section,  I simply  represent  the  c!6c!  analysis In  proposed  Colloquial  coordinated  (13)  VPs  as  i n Lobeck  & Roberge  French,  lexical  a  NP  may  be  followed  by  two  i n (13).  Marie danse p i s chante. ' M a r i e d a n c e s and s i n g s . '  When t h e s u b j e c t  i s represented  by  of  the sentence i s questionable,  is  repeated  i n both conjuncts,  (14)  ?? E l l e  danse  p i s chante.  (15)  Elle  danse  pis elle  Let  (1985).  us t e n t a t i v e l y  subject  c1it ics  p r o j ect ions  as  cf.  a clitic  the  grammaticality  i n (14), u n l e s s -  chante.  assume t h e s t r u c t u r e  are generated  in their  i n (16) usua1  f o r (15),  where  p o s i t i o n and  INFL'  IP /  \  NP  INFL' I \ INFL' CONJ INFL' / \ / \ INFL VP INFL VP / \ I / \ I PiGR TNS I PiGR TNS I / \ V / \ V e l l e PGR I e l l e PGR I | danse |> chant e /  Consequently  the structure  clitic  (IS). *  a r e c o o r d i nab e d .  (16)  the  of  (13)  i s as  i n (17).  (17)  IP /  \  NP / Marie  INFL' I \  /  INFL' / \ INFL VP / \ I PiGR TNS I V I danse i  Pin (13)  INFL' / \ INFL VP / \ I PiGR TNS I V I chante i  COIMJ  a l t e r n a t i v e analysis could  i s actually a coordination  illustrated  i n (IS) i n s t e a d  projections  as  i n (17)  (18) /  Marie  VP  projections  of a coordination  of  as  INFL'  above.  /  INFL'  \  INFL  PIGR  hypothesis  VP  \  TNS  /  \  VP C O N J VP I 1 V V I I danse chante  causes a problem  f o r sentences like  dominating forms of the a u x i l i a r y  coordinated. auxiliary (19)  o f two  that  \  NP  /  VPs  suggested. Suppose  IP /  This  be  Following  Kayne  (1975),  verbs avoir i t i s assumed  verbs i n French are dominated  by  VP.  Marie a mange p i s e s t a l l e e v o i r un h a s e a t e n and i s gone to+see a ' M a r i e a t e and went t o s e e a m o v i e . '  film. film  (19), where  and  etre  that  are  £84  (£0)  IP /  \  IMP / / M a r i e INFL /  verb  /  \  TNS VP / V I a  C O N J VP \ / \ VP V VP I I / \ V est V CP I I I mange allee voir  Phonological  leftmost  VP  \  PiGR  In  INFL' \  Form,  i n each  FVR w i l l VP  apply  i s raised  (££a) , g e n e r a t i n g t h e u n g r a r n m a t i c a l (££b). (£l)a.  IP /  \  NP / / M a r i e INFL /  PIGR  b.* M a r i e  INFL' \  \  VP /  \  TNS V P C O N J VP I I V V I I danse chants i  chante  danse p i s .  un  film  i n both  (18) and  (£0). The  t o I N F L a s i n (£la) a n d sentences  i n (£lb) a n d  IP  (££)a. /  \  NP / / M a r i e INFL /  INFL' \ VP  \  AGR  /  \  TNS VP C O N J  AA  /  \  V  VP V  I  I  a  V est V  Besides and  Marie  VPs  (££) v i o l a t e s  into  /  \  CP  allee  voir  a e s t mange p i s a l l e e  generating  Constraint  \  VP  I  mange b.*  VP /  ungrarnmatical Ross'  un  voir  strings,  (1967;89)  film  un  film.  the operation  Coordinate  i n (£1)  Structure  i n (£3) s i n c e t w o v e r b s h a v e r a i s e d  out o f conjoined  INFL.  (£3) C o o r d i n a t e  Structure Constraint  I n a c o o r d i n a t e s t r u c t u r e , n o c o n j u n c t may b e m o v e d , nor may any element contained i n a c o n j u n c t be moved o u t o f t h a t c o n j u n c t . On t h e o t h e r INFL'  hand,  FVR a p p l y i n g  does not v i o l a t e  produces t h e r i g h t structures  i n a structure with  the Coordinate  outputs,  i n (£4) a n d ( £ 5 ) .  coordinated  S t r u c t u r e C o n s t r a i n t and  i . e . (13) and (19) w i t h t h e  (£4)  IP /  \  NP  INFL' I \ INFL' CONJ INFL' / \ / \ INFL VP INFL VP / \ I / \ i PiGR TNS I AGR TNS I V I danse chante i  / Marie  /  IP  (£5) NP  INFL'  /  \  CONJ INFL' INFL' / \ / \ INFL VP INFL VP / \ / \ / \ / \ PIGR TNS V VP PIGR TNS V VP I / \ I I est V CP a V — i | | I a l l e e v o i r un mange  film  appears t o coordinated  are i n fact  Mar i e  So what  VPs on t h e s u r f a c e  coordinated  INFL'  p r o j e c t i o n s i n French.  coordinated  INFL'  structures that  violating  the Coordinate  Sentences Colloquial coordinated and  like  It i s only i n  F V R may  apply  Structure Constraint.  t h e o n e i n (£6) a r e a l s o  F r e n c h where one a u x i l i a r y VPs, i n t h i s  without  verb i s followed  c a s e t h e two past  chante.  (£6) M a r i e a danse p i s chante. has danced and sung ' M a r i e danced and sang.'  possible i n  participles  by danse  0 D-structure for  f o r (£6) w h e r e t w o V P s a r e c o o r d i n a t e d  the reasons just  where  the f i n i t e  instead  outlined?  auxiliary  i t would  verb a  require  i s base  a  i s blocked  structure  g e n e r a t e d i n INFL  o f VP.  (£7) *  IP /  \  NP  INFL' / / \ M a r i e INFL VP / I \ / \ a PiGR TNS V P C O N J VP I I V V i ! danse chante The is  structure also  i n (£8), w h e r e  an i m p o s s i b l e  preceding  t h e past  (£8)*  projections  s o u r c e f o r (£6)3 t h e r e  participle  i n t h e second  are coordinated  i s no  auxiliary  conjunct.  IP /  \  NP  INFL'  / Marie  / INFL' /  \ CONJ  \  INFL VP / \ / \ PiGR TNS V VP I I a V I danse We  INFL'  are able  exception  INFL' /  INFL / \ PiGR TNS  \  VP I V I chante  t o generate sentences l i k e  clause  to the Coordinate  (£6)  Structure  by a d o p t i n g  an  Constraint; the  Across-the-Board  Constraint  (1967) and d i s c u s s e d  i n (£9), w h i c h  i s proposed  extensively i n Williams  (£9) T h e P i e r o s s - 1 h e - B o a r d  by  Ross  (1978).  Constraint  ((£3) may n o t b e v i o l a t e d ) u n l e s s t h e s a m e e l e m e n t i s moved o u t o f a l l t h e c o n j u n c t s . The  Picross-the-Board  the  Coordinate  of  movement  string.  structure only  Pis a r e s u l t  rule,  coordinate  t h e two i d e n t i c a l  - o s s - t h e - f 3oard /  /  X  \  XP C O N J  XP  i i  i i  X  X 1  e  e  1  i n (£6) c a n be g e n e r a t e d t o both  phonologically  conjuncts,  out o f each  i n a surface  i n the surface  XP /  I  verbs  out o f  \  /  C O N J XP  across-the-board  of a  i n(30).  1  sentence  results  element  Y \  1  auxiliary  application  illustrated  =».  1  be moved o u t  elements are  e l e m e n t s moved  XP XP  The  such  of  Rule A p p l i c a t i o n  \ /  violations  E l e m e n t s may  as a s i n g l e  i s schematically  Y  when  restricted  of Across—the-Board  s t r u c t u r e s appear  This  allows  Structure Constraint.  a coordinate  identical.  Constraint  string  realised.  VP  into  b y a l l o w i n g FVR t o  raising  identical  INFL a s i n ( 3 1 ) . T h i s  where o n l y  one a u x i l i a r y i s  apply  £89  (31)  IP /  \  NP  INFL' / / \ M a r i e INFL VP / I \ / \ a AGR TNS VP C O N J VP * / \ / \ V VP V VP e danse e  The  interaction  coordinated  o f FVR a n d c o n s t r a i n t s o n r u l e  structures predicts that  where VPs a r e c o o r d i n a t e d apply We in  a r e now  grammatical identical  conclude  able French  verbs.  i n French, only  structures  w h e n FVR  then  elle  that  sentences  with  i s able  subject  i n a s t r a i g h t f o r w a r d way. i n (15) t h e c o n j u n c t s  B e c a u s e FVR  i t must  cliticizes  t o generate  sentence  the-board,  (3£) .  are allowed  application in  to  across-the-board.  Colloquial  clitic  chante  apply INFL',  i s generated t o the verb  i s blocked  individually n o t VP, i n AGR  after  In the  c o n t a i n non-  from  i n each  applying  across-  conjunct.  i s coordinated.  i n INFL  clitics  i n each  The  We subject  conjunct  FVR a p p l i e s a s i l l u s t r a t e d  and in  IP  (vie!)  INFL'  NP /  \  INFL' CONJ INFL' / \ / \ INFL VP INFL VP / \ I / \ I PiGR TNS V PIGR TNS V / I \ I \ PiGR elle I PiGR e elle chante danse t  We (33),  are also  The  clitic  by a c o o r d i n a t e d  Elle she 'She  f o r (33)  the auxiliary  across-the-board single  and  the auxiliary  verb a  are  VP.  a danse p i s chante. h a v e - 3 s danced and sung danced and sang. '  D-structure  containing  the  to generate the grammatical sentence i n  where t h e s u b j e c t  followed (33)  able  to raise  i s given  verb a are coordinated. the identical  INFL node d o m i n a t i n g  (34)  i n ( 3 4 ) , where two  the subject  IP /  \  NP  INFL' /  auxiliary  \ INFL VP / \ / \ PiGR TNS VP C O N J VP / I \ / \ / \ elle a PiGR V VP V VP I I I I e danse e chante i i  FVR  VPs applies  verbs  clitic.  into  The  questionable  clitic  i s followed  hypothesis. but  the  by  subject  INFL / \ PiGR TNS / \ elle PiGR 4*  conclusion,  auxiliary  and  a p p l i c a t i on whether  be  generated  b e c a u s e FVR  This  under in  our INFL,  violates  is illustrated  in  the  (35).  of  VP VP  with  Tentino,  \ CONJ  VP I V I chante  I V I danse i  the  p r o p e r t i e s of  main verb  interact  coord i nat e or the  VP  s t r uct ures  distribution  in  (9).  to  the  (1985s350)  as  to allow We  op p o s e d  are  subject  finite  us  then  clitics  supports  the  p r o p e r t i e s of to  French.  We  to  rule pred i c t  able in  the  ungrarnmat i c a l i t y o f  alludes to  a d i f f e r e n c e i n the dialect,  of  French  c o n s t r a i n t s on  s t r u c t u r e s which  respect  Safir  the  with  i s coordinated.  cliticization  Now,  this  e l l e may  r u l e d out  /  in coordinate  subject  in  in  INFL'  account  be  i s explained  subject  INFL' \  /  may  where one  \  NP  French  be  (14),  IP /  an  clitic  of  VPs,  Structure Constraint.  ??  In  conjoined  derivation will  Coordinate  (35)  The  grammatically  to  provide  Colloquial  a n a l y s i s of  (l£b)  in  p o s s i b i l i t y that conjoined  there  structures  fo11ow t h i s  i n t u i t i on  and  assume t h a t , s i n c e s u b j e c t  generated to  i n AGR  coordinate  Trentino  INFL'  namely  i n Trentino,  equivalent  i t i s only  possible  possible structure i n  t o t h e one i n ( 3 2 ) . The  t h e one i n ( 3 5 ) , i s b l o c k e d  can thus conclude  subject  clitics  that  other  independently  due t o d i ff e r e n t  the  two l a n g u a g e s / d i a l e c t s  constraints  the d i f f e r e n c e i n the behavior  i n coordination  not  s t r uct ura1  on c o o r d i n a t e  c 1 i t ic s a r e ob1i gat ory  as  between French  of  and T r e n t i n o i s  posi t i ons f o r t h e element s i n  but f o l l o w s  instead  from  s t r u c t u r e s and t h e f a c t  various  that  subject  i n T r e n t i no.  Agreement M a r k e r s v s . C l i t i c s I  now  McCloskey  turn  (1984.525)  reflections  i n which  pro occurs  head  specified  cliticization issue  extensively  a n d t h u s we  the  a n d some  To  illustrate  although  related,  make t h e f o l l o w i n g C...D  look  topic.  statements  simply  like  different  of a single syntactic configuration,  i n an a r g u m e n t - p o s i t i o n f o r person  and agreement  i s unfortunately  problem  different,  and C l i t i c i z a t i o n  morphological  lexical  to a rather  & Hale  "Agreement  one  then  above.  We  4.2.  are obligatorily  p r o j e c t i o n s . The o n l y  i s therefore  possibility, argued  i n INFL  clitics  and number f e a t u r e s " .  t h e same  too vast will  t h e problem  i n every  to allow  respect?  by a But a r e  This  us t o d i s c u s s i t  be s a t i s f i e d  suggestions  governed  with  a s t o how  more c l e a r l y ,  a statement  t o approach i t . consider  the  of  sentences  below  SA= s u b j e c t  (36)  w h e r e SC1= s u b j e c t  agreement  marker,  clitics,  0A= o b j e c t  Cl= other  agreement  clitics,  marker.  French II l u i en a offert. SCH-C1 + C H - h a v e - 3 s + o f f e r e d 'He h a s o f f e r e d s o m e t o h i m / h e r . '  (37)  Italian Me l o d i c e . Cl+Cl+tell-3s 'He t e l l s i t t o me.'  (38)  Swahili  (Barrett-Keach  1980)  a 1i mw o n a SA+PAST+OA+see 'He s a w h i m . ' (39)  Chiicot i n  s  be R a y s t^L •A+ASP+SA+k i c k ' He k i e k e d me.' (40)  Piraha  (Everett  1985)  hi h i xib-ao -b -a S C 1 + C 1 + h i t +t e 1 i c+ASP+MOOD 'He h i t h i m s e l f / h i m . ' What t h e s e  few examples  l a n g u a g e s c a n be c o m p o s e d  point  t o i s that  of a single  object(s)  and/or  expressed  by t h e p r e s e n c e o f a f f i x e s  therefore  legitimate to describe  sentences  i n (36) and  to  a subject.  verb  The o b j e c t s  t h e one i n (41) a s f a r a s t h e v e r b  but s t i l l  and s u b j e c t  on t h e verb.  t h e French  (37) a s h a v i n g  sentences  and  i n these  include are  Iti s Italian  a representation i s concerned.  similar  (41) In  v  (41), c l stands  overt of  material  objects Now,  that  i n both  still  also  subject  f o r an empty s l o t  representing French  i n French  i s affixed  suffix  in  1  L" c l - e l - c l - V  the subject  and and  t o t h e verb.  representing,  of the sentence.  i n V that  c a n be f i l l e d  i n French  by  and a v a r i e t y  Italian. Italian, In t h i s  there case,  i n some s e n s e , Representation  i s something there  more  i s often  a  the grammatical (41) must  be r e v i s e d  as  (4£).  (4£)  C  c l ~ c l ~ c l - V ~ S f t II  v  Finally,  f o r some c o n j u g a t i o n s  between t h e verb this  stem  a tense  and t h e s u b j e c t  i s the - r - i n French  marker  agreement.  indicating future  appears One  tense;  example o f  (4£)  becomes  (43) . (43) It  C  c l - c l - c l - V - T N S - S f t II  v  i s i n t e r e s t i n g t o note that  representation Swahili  s  cf.  representation  C  attributed to concatenative  and C h i i c o t i n .  Swahili;  (44)  often  v  also  (43) i s t h e k i n d  Barrett-Keach  Morrisson  (1986)  in Chilcotin.  NEG+Sfl+TNS+Ofl+V+f i n a l .1  of languages  (1980) s u g g e s t s for a  similar  like  (44) f o r  Now  we h a v e s e e n  that  the c l i t i c s  many m o r e l a n g u a g e s s e r v e argument  position.  agreement of  CNP, I P 3 b y  these  fiGR  licensing  both  widely  suggested  that  through  inflectional  and  subject  government  conclusion  i n view o f  agreement  h a v e t h e same f u n c t i o n ,  p r o i n argument  elements s u f f i c i e n t  f o r pro i n  Pi p r e l i m i n a r y  be t o s a y t h a t  (Qfl,SPi) a n d c l i t i c s  elements  h a s t h e same f u n c t i o n  i n INFL.  f a c t s would  Italian,  as l i c e n s i n g  I t i s now a l s o  in Italian  i n French,  namely  p o s i t i o n . More g e n e r a l l y ,  markers that  of  they are  f o r the syntactic recoverabi1ity  of  null  arguments. Let  use t h i s  statement  seems t o be a n a t u r a l , further, clitics  one c a n asks  syntactic  element?  same way?  I attempt  behavior of  4.8.1.  that  one. G o i n g  agreement  in a particular they  since i t a  first  grammar and  a r e r e f l e x e s o f t h e same  wordss  can they  be a n a l y z e d  t o answer t h i s  question  by e x a m i n i n g t h e  agreement  of "attested"  Trentino  i s obligatory  construction  step  markers and  marker's w i t h  respect  i nthe  t o some  clitics.  Coordination  Consider clitic  mean t h a t  In other  hypothesis  unquestionable,  function  of "attested"  the properties  fairly  does t h e f a c t  h a v e t h e same  cross-linguistically  as a working  and F i o r e n t i n o ,  i n t h e second  conjunct  as i n (IS) above repeated  in  where a of a (45).  subject  coordinate  ( 4 5 ) a.  La c a n t a e 'She s i n g s  b.* Brandi  La  e  & Cord i n  contrast and  canta  in  (clitic)  balla.  (1981)  (45)  Florentine  la balla. and she d a n c e s . '  to  the  observation  the  second  b.*  be  well  evidence  behave more  pronouns.  ( 4 6 ) a.  as  This  that  conjunct  like  the  the  clitics  agreement was  reached  markers are  French  examples  nous mangions et  i t was of  argued  clitics  i n the  as  under a c e r t a i n a n a l y s i s of introduced fact  there,  that  this  a verbal  member o f  a conjunct  agreement  marker,  affix  but  certain clitics  this  needs t o  Under the  repeated  i s not  one  to  i n c e r t a i n languages can  There  i s thus a  m a r k e r s and  intersect with  fuzzy  area  respect  to  that  in  this  approach words,  i n the be  second  an  correlation also where their  be  repeated  agreement behavior  c o o r d i n a t i on. In  conclusion,  necessary  r e p e t i t i o n under  of  only  In o t h e r  i t can  a one  basis  is valid  valid.  be  real  shew.  section  i n d i c a t i o n that  under c o o r d i n a t i o n . clitics  i s not  the  (46)  markers  coordination.  conclusion  i s an  previous  agreement  on  in  boire.  que  Trentino  always repeated  II  faut  in  the  markers than  buvions.  interpretation  since  that  (198£) t a k e  I I f a u t que n o u s m a n g i o n s e t 'We must e a t and drink.'  Nevertheless,  the  Bouchard  conclusion  agreement as  as  coordination  in  appears affix  to  to  4. £. £.  be  a necessary  analyzed  as  but  an  not  sufficient  agreement  condition  for  an  marker.  Doubling  The for  be  an  d o u b l i n g phenomenon c a n eventual distinction  clitics.  I t i s assumed  here  be  used  as  another  test  between agreement  markers  that  be  d o u b l i n g can  ground and  d e f i n e d as  in  (47) .  (47)  The  D o u b l i n g i s a p r o c e s s w h e r e b y some g r a m m a t i c a l f e a t u r e s o f a l e x i c a l NP i n a r g u m e n t p o s i t i o n c a n b e r e p r o d u c e d p h o n o l o g i c a l l y by a n a f f i x o n t h e h e a d o f w h i c h t h e NP i s a n a r g u m e n t . two  ( 4 8 ) a.  b.  Pied  an  (49) The (49)  sentences  in  (48)  involve  doubling.  in  (49), where t h e  L' homme i 1 v i e n t . t h e man c l come—3s ' T h e man c o m e s . ' Marie  1 'aime a cl like-3s to 'Marie l i k e s Jean.'  Consider is  Noir French  next  the English  agreement The  man  marker  the verb. instance  sentence  standing for 3 " v  cl  person  affix  -s  singular.  comes.  characterization since  Jean,  of doubling given  s o m e f e a t u r e s o f t h e man O b v i o u s l y we  do  not  of doubling s i m i l a r  want  are  in  (47)  applies  repeated  to describe  by  (49)  t o the ones s t u d i e d  also  a suffix as  to on  an  i n Chapter  3.  One way o f g e t t i n g a r o u n d the  word  "affix"  out  (49) a s d o u b l i n g  original clit  this  problem  i n ( 4 7 ) by t h e w o r d but would  consists in replacing  "clitic".  simply  bring  That  would  rule  us back t o t h e  i s s u e o f d i s t i n g u i s h i n g between agreement  markers and  ics. Pi m o r e p l a u s i b l e move w o u l d  opt i ona1  process.  Colloquial Standard (50)  ( 5 1 ) a.  French,  English  Jean  with  regard  i s no e q u i v a l e n t  (51) i n t o (51) i n  t o ( 4 9 ) , a s shown i n (52)  II vient.  b» *  come.  Comes.  i s also  problematic,  language where t h e a f f i x never  find  (47) i s an  vient.  (5£)a.* T h e man  This  but t h e r e  t o ( 5 0 ) we  that  i l vient.  Jean  b.  So, p a r a 1 1 e 1  be t o s t i p u l a t e  doubled  conclude clitic. though  that This  conclusion be t r u e .  so that  (53) a . * J e a n Marie  argument  i f one s t u d i e s  say, the subject  then  i s an agreement  one would marker  In Standard finds  i_l_ v i e n t . (a) J e a n .  French,  a  i s  have t o  and not a  i s not n e c e s s a r i l y t h e r i g h t  we n e v e r  1'aime  because  representing,  lexical  the affix  i t could  impossible  b.*  by a  however,  doubling  one i s  even  but  only:  ( 5 4 ) a.  Jean  b.  I_l_ v i e n t .  c.  Marie  aime  d.  Marie  1_' a i m e .  Clearly, and  Jean.  i n Standard  (54) a r e c l i t i c s Even  cases: then  vient.  the other  French,  the underlined  not agreement  i fthe affix  i s obligatorily  i t i s an agreement  marker.  i n (49) but i s n o t t r u e  lexical  NP s u b j e c t  ungrammaticality  ( 5 5 ) a.  E l Mario  spurious certain only  where  agreement  7  but o n l y  This  Chapter  3.  that  of a lexical  i fthe verbal  clitic.  holds  of the English  by a s u b j e c t  data  NP  where  clitic.  a  Witness  (55b).  distinguished.' ' This  features  by a l e x i c a l  f o r the Trentino  phenomenon however  an area  notion  doubled  magna.  Thus t h e d o u b l i n g  clearly  of  i n some  e l _ magna,  b.* E l M a r i o  constitute  This  must be d o u b l e d  i n (53)  markers.  p r e d i c t i o n a p p e a r s t o be wrong  example  the  elements  affix  i s the case  i t i s defined  markers and c l i t i c s  d o e s n o t mean  that  the phonological argument  i n the doubling (49) above  are  doubling  i s a  repetition  of  constitutes  can independently  I t a l s o means t h a t  does not  doubling  be d e f i n e d  as a  examples s t u d i e d i n and t h e  Italian  sentence (56)  i n (56) a r e n o t e x a m p l e s o f  Giovanni  arriva.  To c o n c l u d e , (i.e. not  coindexed)  sufficient  agreement  A. 2 . 3 .  for  the possible with  an a f f i x  condition  o f a n NP  associated  a p p e a r s t o be a n e c e s s a r y  f o r analyzing  the affix  but  as an  Extraction test  blocked  suggests  itself.  out o f doubling  C o l l o q u i a l French 5 c f .  ( 5 7 ) a.  presence  marker  Another often  doubling.  Jean  Recall  that  constructions  extraction i s  a s shown  i n (57)  3 . 3 .  i l vient.  b. * Un hornrne i l v i e n t . c. * Q u i i l v i e n t . This  seems t o p r o v i d e  indefinite  NP  clitic.  have seen  more  We  general  quantified this  us w i t h  i s barred  in 3 . 3 . 2  restriction  NPs  from  restriction  appearing that  Pied  Noir  Trentino,  and o t h e r s ,  fact  a restriction  with  French,  such  a  i s an  constructions.  does not apply  like  this  test.  on LF e x t r a c t i o n s  doubling  languages  that  from  a possible  In (57b), coreferent instance  and On  the other  c e r t a i n Spanish  o n t h e t y p e o f NP t h a t  of a  blocking  across-the-board  NPs c a n be d o u b l e d .  an  since  hand,  i n some  dialects,  Therefore the c a n be  301 associated  with  present  in a  clitic,  but  that  the We  that  the  conclude  that  affix  a  now  l a n g u a g e s do  as  a  on  clitics.  is rather  not  affix  is a  does not  imply  of  the  verb  i s blocked  out  we  have examined.  an  NP  then  i s barred this  of  We  can  when  affix  an  is  I t must  limited  in  to  d i s t i n g u i s h between  be  noted,  unfortunately,  i t s applications  a l l exhibit syntactic extractions  since  like  Wh-  s  conclude, be  necessary  apply  Wh-extraction  have a t e n t a t i v e t e s t  test  can  that  i s present  that  that  the  is  clitic.  i f wh-extraction  m a r k e r s and  To  that  does not  languages/dialects  agreement  movement.  suggests  certain features  clitic.  H e n c e we  this  a  i n 3.3  i n a l l the  probably  representing  restriction  i s not  have argued  associated  affix  language s t r o n g l y  affix  doubling thus  a verbal  the  presence  associated  but  not  with  sufficient  of  a  restrictions  given  affix  condition  on  the  seems t o  type  of  NP  constitute  for analyzing  this  affix  clitic.  4.£.4.  Obligatoriness  That  an  affix  also  a strong  But,  here  clitics Trentino  and  obligatorily  i n d i c a t i o n of  again,  can  be  also  there be  i s no  i t s status one  obligatory  F i o r e n t i n o are  present  to  one  of  a  verb  agreement  stem  is  marker.  correlation since  in certain  examples of  on  languages/dialects.  this  sort; cf.  a  3.3.1.4  and  3 . 3 . 1. 5. In  9  conclusion,  necessary  but not s u f f i c i e n t  as an agreement  4.£.5.  obligatory  presence condition  o f an a f f i x  i s a  f o ranalyzing  the affix  marker  Morphology  Zwicky  & P u l l urn  (1983)  note  that  morphological  idiosyncracies  a r e more c h a r a c t e r i s t i c o f a f f i x e d words  clitic  Pis a n i l l u s t r a t i o n ,  groups.  paradigms change  i n (58) where  i t i s c l e a r that  f o r t h e same c o m b i n a t i o n  whereas  the c l i t i c s a.  is £s 3s 1p £p 3p  remain  Present  b.  je parle tu paries il parle nous p a r I o n s vous p a r l e z i l s parlent  Thus, sufficient  t h e French  t h e agreement  of features  markers  in different  tenses  t h e same. Conditional  c.  je p a r i e r a i s tu p a r i e r a i s il parierait no us p a r 1 e r i o n s vous p a r l e r i e z ils parieraient  mor pho1og i c a 1 c h a n g e s condition  consider  than of  je p a r i e r a i tu parieras i 1 parlera no us p a r 1 e r o n s vous p a r i e r e z i l s par1eront  constitute a  f o ranalyzing  Future  an a f f i x  necessary as an  but no  agreement  marker.  4.£.&.  Conclusion  T h e r e seem t o be c e r t a i n d i f f e r e n c e s agreement often  markers even  though  their  s i m i l a r i n certain respects.  between  respective  clitics  and  properties are  The q u e s t i o n o f d e t e r m i n i n g  the  status  features this  of a verbal  affix  o f t h e argument  status  representing  o f a head  i s not c l e a r . This  i s t h e case  this  on  o f t h e arguments with  syntactic and  phenomena l i k e  coordination.  further It  i s most  doubling,  The q u e s t i o n  likely  i n less  studied  t o have  consequences  respect  null  o n e when  to certain  arguments,  thus remains open  extraction,  pending  research. i s nevertheless  distinction agreement  argument  possible  i s an expected  position.  1  0  that  f i l l  this  fuzzy both  t h e same f u n c t i o n  in a  elements  f o r pro i n  the d i f f e r e n c e s observed  section  and t h e ones s t u d i e d  suggest  that,  affixes  are probably  Historical  that  that  of licensing  Pllso,  although  t o remark  o n e w h e n we c o n s i d e r  m a r k e r s and c l i t i c s  number o f l a n g u a g e s ,  4.3.  nominal  i s an important  languages since the behavior  status  certain  very  i n Zwicky similar,  & P u l l urn  these  in this  (1983)  two t y p e s  clearly  of verbal  distinct.  Speculations  T h e r e s e e m s t o be a p o s s i b l e c o r r e l a t i o n b e t w e e n t h e historical unclear  development  distinction  o f c e r t a i n Romance  between agreement  Although  the following discussion  to  out that  point  this  i t i s natural  l a n g u a g e s and t h e  m a r k e r s and  i s highly  clitics.  speculative,  under t h e analyses  I wish  presented  thesis. There  i s a widespread  intuition  underlying  some w o r k s  on  in  clitics  and  according  agreement  to  which,  markers  as  Lapointe  "person/number markers can clitics  which  productive which  have  been  processes  exhibit  in a diachronic perspective  profitably  left  from  (1980s£34) w r i t e s ,  over  as  previous  be  analysed  the  as  fossilized  residue of  s t a t e s of  the  various d i f f e r e n c e s i n behavior  once  languages  across  and  the  languages". This  view  merged  with  the  attribute  this  agreement  markers have the  seem  to  e v o l u t i o n process  provide  a reason  synchronic  slice  languages.  Nevertheless  without with of  the  the  with  of  for their  this  of  hypothesis  i s a very  fact  that  clitics  and  consider  traditional  (1956s684) and supposes that  the  (circa  c h a n g e was  flexional  t o make n e c e s s a r y  i s not  given  and  across  completely  n a t u r a l one  and  has  markers are The  to  just  just  described.  particular  case  of  f o r example, in Harris  change whereby board  had  1300-1600). the  use  of  the of  interactions  in  To  French. Bourciez  (1978s 111), word-final effect  the  The  do  parts  process  distinctions  approximately  any  the  expressed,  the  at  would  language.  (1966) c i t e d  across  f u n c t i o n and  and  a  the  phonological  consonants dropped eliminating  view  Ewert a  alter  would  clitics  language,  agreement  "inventory" of  l e t us  a  here  that  resemblance  development  phenomena can  fact  same s y n t a c t i c  problem  syntactic  presented  the  The  illustrate,  French  the  to  flaws.  other  The  analyses  verbs  effect  subject  of in of  gradually Middle this  pronouns  which  then  became c l i t i c s .  Italian, the  Spanish,  This  and  s i t u a t i o n found  subject On  Portuguese. i n Old  p r o n o u n s were not the  other  the  facts  the  obligatory  syntactic French  in a  word  passed used  hand,  to  this  obey  pronouns seems t o  process  B i nd i ng  4.4.  the  retained  Latin,  where  reinterpreted  be  that  Subject  i s SVX,  and  Old  this  following  chronologically This  to  pronouns  o b l i g a t o r i n e s s of  above.  that  attributed  others,  Harris,  mandatory  pattern Price  subject  the  p a r t i c u l a r process  presence  of  subject  cliticization.  is referred  s i m i l a r to  to  i s to  that  the  alluded  maybe t h e i r  reader  analysis  that  preceded  have a c c e l e r a t e d  The an  have  order,  have  stage.  generalised.  out  thus  (1978.113) s u g g e s t s  following  a verb-second  points  p r o n o u n s and  notes,  through  have s u b s e q u e n t l y  would  studies  pronouns  He  languages  in  obligatory.  subject  order.  have happened  previously,  use  would  phonological  and,  Harris  be  even  French  d i f f e r e n t way. of  not  These  more r e c e n t  could  (1971),  process would  to  Adams  Harris'  but  (1986) i n the  f o r the  defense  Government  of  and  F r a m e w o r k.  Summary Two  raised clitics.  rather by  the In  separate various  4.1,  between s u b j e c t  an  issues  analyses  account  clitics  and  was  were d i s c u s s e d of  the  constructions  proposed  object  here that  for  clitics.  further Whereas  were  involving symmetry object  clitics  a r e base  constituent adjoined verb in  t o AGR  Emonds  The  pro.  clitics  Verb  head  of the  a r e base generated  Raising  as  affixed to the  of the type  us t o an e x a m i n a t i o n  proposed  o f some o f t h e  o n FVR.  addressed  i n 4.£,  markers with  was t h e n  concerned  h e r e h a v e t h e same  raised  respect  the fact  that  syntactic  to the licensing of  of determining  whether  they  d i f f e r e n t r e a l i z a t i o n s of a single syntactic notion.  o f some o f t h e c o n s t r u c t i o n s  markers  leads  to the conclusion  probably  i s raised  agreement given  lead  are analyzed  as agreement  just  issue  This  issue,  as they  involved  i n the first  markers often  t e s t s that  have not been a b l e  conditions agreement  i n which  that  place  behave  would  marker.  We  and agreement  pointed  out that  because c l i t i c s  still  clearly  a given  remains  of  affix  and  respect  to  providing notions.  sufficient  as a c l i t i c  the  and  d i s t i n g u i s h t h e two  necessary  A  a r e both  i n t h e same way w i t h  t o provide  f o r analyzing  they  clitics  have d i f f e r e n t s t a t u s .  phenomena but t h e p r o b l e m  syntactic We  rule of Finite  The q u e s t i o n  study  subject  of coordination  function  on t h e v e r b a l  i n INFL and a r e s u b s e q u e n t l y  (1378).  second  clitics  are  structure,  by a l a t e  effects  generated  o r a s an  Footnotes t o Chapter 4  1.  Mote t h a t  Fiorentino, structure.  i n the t h e TNS  This  verb,  together  in  as  PF  instances element  does with  not TNS  constitute now  i s moved a  attached  i n VP  problem to  in Trentino prior  f o r FVR.  i t sright,  and  to  S-  The  m o v e s t o PiGR  IP / NP I pro  \ INFL' /  \  INFL I OCR / I \  SCI I AGR CV+TNS3  V ' I \ V ' I V I  t  7*  t  2.  Cf.  Zagona  3.  Cf.  also  (1982)  Brandi  similar  views.  4.  i s also  for  INFL  inversion  i n ( i ).  (i)  This  of  of free  the surface  PN  French.  5.  The  Chilcotin  NP  for a different  & Cord in  the case forms  data,  (1981s£),  i n PN  analysis  Bouchard  F r e n c h . Cf.  anaphora.  (1982s407)  3(footnotes  of the 3 person pronouns  an A t h a b a s k a n  o f VP  for  4,  6)  i n C French  language spoken  in  and  British  Columbia  6.  Note  i s due  to Morrisson  further that  structure pronoun  Barrett—Keach  i n ( i i ) and  clitic  and  ( i)  (1986).  (1986)  argues f o r the  against  t h e one  ( i ) , w h e r e SP=  object  pronoun  clitic.  OP-  subject  V  (ii) NP  OUX SP  7.  Note  wanted  that  8.  Cf.  Bantu  Tense  Case  which  where he  (i)  a property  Mchombo's  of  be  clitic  Mchombo  i s present  as  study  used  t o argue  i s the absence  t o agreement  on  are c l i t i c s  object and  not  of  markers.  markers  in  agreement  i n Bantu even  when  the  in( i ) .  s i t i s i r n e n e mwana u y u a n a ; i b a . did this child steal them?'  (1983)  have  argued  i n t e r r o g a t i v e c o n s t r u c t i o n s without even  f o r the  (1984:11)  M c h o m b o &• M t e n j e  criterion,  as a test  effect  specific  (1984)  they  used  s e e m s t o be W h - e x t r a c t i o n  CKodi n d i nkhukull '[Which chickens] But  cannot  i t s ultimate  argues that  There  underlined  V  since  i s not  f o r example  markers.  OP  absoption  distinction  doubling  VP  though  f o r t h e base appeal  i t i s not v i o l a t e d  f o r the c l i t i c  status  by  generation  t o movement. ( i ) , cannot  of zi.i n t h i s  Our  be  sentence.  Cf  also  Bresnan  & Mchombo  (1985)  for a detailed  Lexical-Functional  Grammar  9.  (1983s503ff)  the  Zwicky  &• P u l l urn  set of combinations  words than  clitic  one  correlation  cf.  3.1.4.  10.  See  Jelinek  that  as  the  fulfilling  Principle. verb.  For  that  "arbitrary  clitics  More p r e c i s e l y ,  of  Unfortunately, there  f o r an  i s suggested elements  suggest  between a r b i t r a r y  it  in a  framework.  a r e more c h a r a c t e r i s t i c  groups".  (1985)  analysis  g a p s and  entirely and  the  agreement  agreement  requirements  they  arguments a g a i n s t t h i s  are  the  type  true  of  one  the  be  where  analyzed  Projection  arguments of  of approach  to  markers?  approach  markers  in  affixed  i s no  different  gaps  see  I.E.  the  CONCLUSION  To c o n c l u d e theoretical more  empty of  rests  category  theory  on t h e major  distribution  o f pro both  grammars,  have opted  coindexed  with  across  Principle o f PRO  specifying  The  that  Each  i n a given  In examining  in syntactic  governed  assumption  type  way.  and since  PRO  the  l a n g u a g e s and w i t h i n  f o r a theory  of features necessary  statement  Category  i n the  theory.  languages.  licensed  positions.  i t spresence and  underlying  modulates the occurrences  i n ungovemed  on  t h e main  in linguistic  in natural  by t h e Empty  appears only  conditions  review  i t i n c l u d e s and t r y t o put them  i s syntactically  are licensed  we  simply  perspective of research  work  government  set  I will  categories are available  empty  Traces  thesis,  proposals  general  preceding  this  particular  minimal  s t r u c t u r e s : p r o must  by some e l e m e n t  which  for i t slicensing.  the English constructions discussed  carries  I exempt  from  be  the this  in Introduction  wh i c h  rerna i n  The empty  un i n v e s t i g a t eel.  a b o v e c o n d i t i o n s on  categories  characteristics fundamental  they  and  empty c a t e g o r y  Pis  we  postulation  of  language-part  like  licensing  device  verbal  sufficient rich way  sufficient is  with  the  to  We  then  hypothesis  that  general  was  of  this  familiar  presented  principle path  which  in  in  leading  replace  an  rich  to  the  very  agreement  instance  i n these  languages  of  features  represented  verbal  and,  f o r the  to  was  how  Pro  licensing  in the  showed  Spanish  intuition  turned  the  analyses  set  This  allow  implemented  one  below.  principles  and  the  agreement  i n which  which  least  of  rules.  f o r pro.  f o r the  verbal  (1)  more  at  types  licensed.  followed  Italian  agreement.  the  development  previous  languages  in  be  fundamental  of  coindexed  the  has  icuiar  Pi r e v i e w  from  Principle  must  have seen, theory  particular  Principle  Pin  linguistic  and  a l l derive  i s repeated  Recoverabi1ity  presence of  s y n t a c t i c structure share  Recoverabi1ity  Introduction  (1)  in a  the  agreement of  pro.  Pro  therefore, that  of  a  i s governed by  i s thus  "allowed".  subject-verb  the  licensed This  agreement missing  is  by the  is subjects  grammar.  subject  and  clitics  proposed  and  by  i s , i n a c e r t a i n sense,  r e c o v e r a b i 1 i t y of  clitics  in  object  the  working  clitics  are  two  different same  r e a l i s a t i o n s o f a s i n g l e s y n t a c t i c element  function  object  clitics  position  they  coindexed the  a s PiGR.  same  are related  with  intrinsic  features  position. This  linguistic clitics which  missing. stated  verb  here  generalisation  one never  clitic  explored,  main  namely  that  f o r pro i n external  i s supported  by t h e c r o s s subject  tensed  and a s u b j e c t  sentences i n  clitic  i n languages with  finds a simple  are c a n be  object  s e n t e n c e where both t h e  NP o b j e c t  are missing  (when t h e  f o r an o b j e c t ) .  the intuition  clitics  are sufficient  missing  arguments.  same p a t h ,  The  cases  i s t h e same a s t h e o n e t h a t  and t h e l e x i c a l  subeategorises  i n most  the languages with  NP s u b j e c t  same a s t h e o n e c o n c e r n i n g  behind  this  subject-verb  devices  clitics  generalisation agreement,  of this  are licensing devices  o r i n t e r n a l argument  material  i s generated  p o s i t i o n i s occupied  position  by p r o w h i c h  must  of  intuition  follows  f o r pro i n  p o s i t i o n . When n o  i n argument  i sthe  that i s ,  f o r the recoverabi1ity  The i m p l e m e n t a t i o n  external  this  was t h e n  " e x h i b i t " simple  f o r i n t e r n a l arguments,  Therefore,  the  markers.  l i c e n s i n g elements  never  a lexical  This  clitics, object  as agreement  generalisation that  studied  both  having  hypothesis  that  pro i s , consequently,  by a c l i t i c ,  assumption  are also  others,  the  f o r p r o i n t h e argument  t o and t h a t  and governed  clitics  argument  following  are l i c e n s i n g elements  consequence o f t h i s subject  I t was a r g u e d ,  with  lexical  i n tensed  sentences,  be l i c e n s e d  by a  clitic-  The a b o v e  properly  licensed  The  other  linguistic they to  generalization only  working  when a c l i t i c hypothesis  a r e generated  i n PGR  NPs  order  t o account  a r e i n complementary  languages,  Case absorbers. position blocked that  by t h e C a s e  a n NP w i t h o u t  accordance behavior  with  In  following  a clitic  assumed  respect that  unmarked  others,  be l e x i c a l i z e d  (1S8£:) p r o p o s a l s . t o Case  that  clitics. and  lexical  clitics  are  i n t h e argument  p r o does n o t need  The p a r a m e t e r  i n PF i n As f a r a s t h e  i s concerned,  complementary r e a l i z e d NP  involved  we  Case.  would  given  (£•)  A c l i t i c absorbs a Case f e a t u r e o f t h e l e x i c a l head t o which i t i s a f f i x e d .  language  i s so i na l l  i n a c e r t a i n number o f  form  The  in  this  i s t h u s non Case-marked and  and a p h o n o l o g i c a l l y  case.  that  are i n a position  clitics  r e a l i z e d NP  t e r m s o f l e a m a b i l i t y a n d UG,  between a c l i t i c  and i m p l i e s  A l t e r n a t i v e l y , i t c a n be a s s u m e d  Case cannot  Bouchard's  that  that  distribution  filter.  of pro with  tentatively  the  with  pro i s  the cross-  exhibiting subject  A phonologically  coindexed  where they  f o r the fact  we h a v e a s s u m e d ,  clitics  I t was a r g u e d  Romance l a n g u a g e s / d i a l e c t s In  h a s t o do w i t h  i n INFL,  since  i s present,,  characterization of subject  l i c e n s e p r o i n CNP,IP].  the  i s thus derived  distribution i s  thus  considered have t h e  (S).  language-learner includes  must f i r s t  determine  s o m e c 1 i t i c - p a r a d i grn <s) .  that  his/her  I t i s supposed  that  positive  evidence  pronouns  (theclitics)  each is  i s a v a i l a b l e that  behave d i f f e r e n t l y  clitic-paradigm  set,  namely that  clitics  close  (subject, there  theory,  The a n a l y s e s  here,  l e a m a b i l i t y and  I have suggested  account  f o r v a r i ous  s y n t a c t i c c o n s t r u c t i ons,  account  of leamability.  those  with  clitics, In  various  certain  allows  and t h o s e  without.  not only  doubling  languages there  defended  i n t o two  Further,  must  with  in  certain  of  t h e head  assigned The  here  a clitic  be r e l e a s e d  some  groups:  among  i s no c o m p l e m e n t a r y  t o allow i s that  a r e made p o s s i b l e  coindexed  in a  those  with  combinations o f c l i t i c - p a r a d i g m s a r e found.  languages a r e s a i d  hypothesis  from  t o which  clitic-doubling. subject  doubling  t o be p h o n o l o g i c a l l y  absorption  they  not absorb  The and  main object  F o r a n NP  realised,  by t h e c l i t i c .  need  distribution  r e a l i s e d NPs.  by t h e same m e c h a n i s m s .  languages c l i t i c s  Case  In other  t h e Case  a r e a f f i x e d . Case c a n t h u s  words,  feature be  i n t h e f a m i l i a r w a y t o t h e NP p o s i t i o n . parameter  a  b u t a 1 s o some  b e t w e e n some c l i t i c - p a r a d i g m a n d p h o n o l o g i c a l l y These  establishes  cross-linguistic  us t o p r o p o s e ,  languages a r e divided  clitics  case  between  a r e t h u s embedded  that  respect  NPs. F o r  r e a l i s e d NPs.  perspective  this  regular  distribution  certain  In  certain  e t c . ) , the? u n m a r k e d  as i t i s understood  r e l a t i o n s h i p between  variation.  object,  from  i s complementary  and p h o n o l o g i c a l l y  Linguistic  indicates that  involved  here  i s once again  t h e o n e i n (£).  In t h e c a s e set  o f each  parameter  absorb  (2) n e g a t i v e l y .  t h e Case  clitic-paradigm in  (2) must  feature.  those  major  Doubling  with  doubling  this  o f Case  clitics  and t h o s e  i n doubling  need n o t  i n v o l v i n g each  that  t h e parameter  device  two  i s t h e source  and o b j e c t  i s necessary  position.  i n 3 . 2 . 2 may  into  without.  clitics  i s not so i n subject presented  are divided  constructions  between s u b j e c t  W h e r e a s a dummy C a s e a s s i g n i n g  theory  a clitic  constructions  the languages with  asymmetry  position,  words,  c o n s t i t u t e p o s i t i v e evidence  Case assignment a  In other  may  be s e t n e g a t i v e l y .  Therefore, groups,  c l i t i c - p a r a d i g m , the language-learner  account  of  clitics.  f o r object  The  particular  for this  asymmetry. Even though doubling there of  C a s e c a n be a s s i g n e d  constructions  quantified related  c a n be c l i t i c NPs a r e o f t e n  t o the other  extractions  of c l i t i c  some r e s t r i c t i o n s  doubled. barred  from  restrictions doubled  We  argued clitic  between a c l i t i c  systematically position last  block  involved  point  clitics, on t h e type  the fact  doubling  that  i s t o be  syntactic  NPs.  and an argument  t h e presence  i n a c-chain,  i s the source  that  concerning  Pi c h a r a c t e r i z a t i o n i n t e r m s o f c - c h a i n s holding  ways i n  i n s p i t e of t h e presence of  are, i n c e r t a i n languages,  NPs t h a t  in familiar  of therelationship  p o s i t i o n allowed  o f a bound  be i t Pi-bound  NP  us t o  i n t h e argument  o r Pi'-bound.  o f some c r o s s - l i n g u i s t i c  This  variation,  which,  i t was a r g u e d ,  syntactic disallow and In  level  t h e former  permitting applies. the as for  at which  syntactic  some a l s o  disallow case,  c-chains  the breaking  syntactic  LF e x t r a c t i o n s subject  like  clitics  QR  found  The d i f f e r e n c e s  differences We  then  concerning subject of  between  subject  clitics  F i n i t e Verb The  namely  determining  of c l i t i c s i n natural  well  that language  various  from  and not from  intrinsic  clitics.  some r e l a t e d  cliticization  between t h e  analysis,  asymmetries  issues,  where  especially  i t was p r o p o s e d by a  that  late  rule  Raising. analyses  suggested  m a r k e r s t h e same r o l e  that  as  a r e t h e same  i n INFL a r e a f f i x e d t o t h e v e r b  different  agreement  i n my  the different  discussed  with  as d i f f e r e n t r e a l i z a t i o n of the  follow,  motivated  be b r o k e n  Wh-movement  a theory  same s y n t a c t i c  independently  QR  clitics.  here  must  a t LF where  C-chains  are blocked.  construction  S-structure  like  They a r e a l l c o n s i d e r e d  clitic-paradigms  at  can never  extractions  clitic-paradigms  element.  constructions  of the chain  I have o f f e r e d  the various  languages  out o f t h i s  c-chains  and o b j e c t  More g e n e r a l l y ,  Most  out o f doubling  a r e formed  case,  f o r by t h e p a r t i c u l a r  i s formed.  LF e x t r a c t i o n s  In the latter that  a c-chain  extractions  therefore  effect  unifies  i s t o be a c c o u n t e d  o f l i c e n s i n g p r o . We whether  clitics  attribute to clitics  i n the syntactic thus  examined  and agreement  and  structure,  the question  of  markers a r e j u s t t h e  same e l e m e n t s and the  concluded,  basis of their  constructions  the  differential  and  i t i s worth  presence  of particular  theory  share  another  o f government.  Recoverabi1ity intimately  noting  such  will  probably  that  with  they  were  regard  to  types  the three  o f empty  characteristic:  not  on  various  Principle  and  they  a central role  the theory  the theory  i n the theory  on  alluded  to  a l l involve the that  the  o f government  o f government  p r o v e t o be o f p r i m a r y  accounts of the English  conditions  categories  It appears therefore  related. Since  play  behavior  that  operations.  Finally,  earlier  tentatively,  are  seems t o  of recoverabi1ity, i t  importance  constructions discussed  in  eventual  i n the  Introduction. The  interesting  government answer amounts in  and  linguistic  of universal  Nevertheless,  to provide  i s such  an  since i t  important  concept  intriguing,  and  syntactic recoverabi1ity i s further  of the various PRO  should  i n t i m a t e l y r e l a t e d ? Pin  of the concept  I t would c e r t a i n l y  and  i s : why  It i s nevertheless  of the importance  pro, trace,  theory  government  o f grammar.  theory.  unification  for  why  point  i s difficult  revealing, that  confirmation  a  a question  to asking  especially  at t h i s  r e c o v e r a b i 1 i t y be s o  t o such  the theory  question  licensing  i n order  be  o f government  interesting  operations  to achieve  a more  to  in  attempt  appealed  to  restrictive  grammar.  e v e n more  interestingly,  i t i s certainly  the  case of UG.  that  empty  the importance categories  o f government  i s indicative  for the  recoverabili  of the cognitive  reality  REFERENCES Adams, ms.  M. ( 1 9 8 6 ) . University  Aissen, 366.  New  J.  (1974).  (1979). York.  "From O l d F r e n c h t o t h e T h e o r y o f C a l i f o r n i a at Los Angeles. "Verb  Raising",  The S y n t a x  Linguistic  of Prodrop",  Inquiry  of Causative Constructions.  5s3£5-  Garlands  & D.M. P e r l m u t t e r ( 1 9 7 6 ) . " C l a u s e R e d u c t i o n i n S p a n i s h " , P r o c e e d i n g s o f t h e Second Annual Meeting o f t h e B e r k e l y Linguistics Society. A k m a j i a n , A., R. A. D e m e r s &• R. M. H a m i s h ( 1 9 8 4 ) . L i n g u i s t i c s s An I n t r o d u c t i o n t o L a n g u a g e a n d C o m m u n i c a t i o n . MIT P r e s s : Cambridge. A n d e r s o n , M. ( 1 3 8 3 ) . " P r e n o m i n a l of Connecticut, Storrs. Anderson, Morris  Genitive  N P s " , ms.  University  S. & P K i p a r s k y , e d s . ( 1 3 7 3 ) . A F e s t s c h r i f t f o r H a l l e . H o l t , R i n e h a r t and Winston:New York.  A o u n , J . ( 1 3 7 3 ) . "On G o v e r n m e n t , C a s e - M a r k i n g P l a c e m e n t " , ms. M I T , Cambridge. (1381). Doctoral  The F o r m a l  Nature  Dissertation,  (1385).  A Grammar  MIT,  of Anaphoric  and  Clitic  Relations.  Cambridge.  o f Anaphora.  MIT  Press:Cambridge.  & D. S p o r t i c h e ( 1 9 8 3 ) . "On t h e F o r m a l T h e o r y o f G o v e r n m e n t " , The L i n g u i s t i c R e v i e w £ s £11-£36. B a c r i , R. ( 1 9 8 3 ) . T r e s o r s d e s r a c i n e s p a t o u e t e s . B e 1 i n : P a r i s . B a k e r , C.L. ( 1 9 7 9 ) . " S y n t a c t i c T h e o r y a n d t h e P r o j e c t i o n P r o b 1 e m " , L i n g u i s t i c I n q u i r y 10s 5 3 3 - 5 8 1 . B a r r e t t - K e a c h , C.N. ( 1 9 8 0 ) . T h e S y n t a x a n d I n t e r p r e t a t i o n the Relative Clause Construction i n Swahili. Doctora1 Dissertation, University of Massachusetts, Amherst. (1386). Aux/Infl",  "Word-Internal Evidence from S w a h i l i f o r L i n g u i s t i c I n q u i r y 17:553-564.  of  B e l l e t t i , A. ( 1 9 8 5 ) . MIT, Cambridge.  "Unaccusat i v e s  a s C a s e A s s i g n e r s " , rns.  , L. B r a n d i & L. R i z z i , e d s . ( 1 9 8 1 ) . T h e o r y o f M a r k e d n e s s i n G e n e r a t i v e G r a m m a r . P r o c e e d i n g s o f t h e 1 9 7 9 GLOW C o n f e rence. Scuola Normale S u p e r i o r e d i P i s a . & L. R i z z i ( 1 9 8 1 ) . " T h e S y n t a x o f n e : Some T h e o r e t i c a l I m p l i c a t i o n s " , The L i n g u i s t i c Review 1:117-154. Borer, and  H. ( 1 9 8 4 ) . P a r a m e t r i c S y n t a x : C a s e S t u d i e s i n S e m i t i c Romance L a n g u a g e s . F o r i s : D o r d r e c h t . (1986).  "I-Subjects", Linguistic  , ed. (Forthcoming). Syntax Academic Press:New York.  Inquiry  17:375-416.  and Semantics,  v o l . 19,  B o u c h a r d , D. ( 1 9 B £ ) . On t h e C o n t e n t o f E m p t y C a t e g o r i e s . D o c t o r a l D i s s e r t a t i o n , MIT, C a m b r i d g e . B o u r c i e z , E. ( 1 9 5 6 ) . E 1 e m e n t s d e 1 i n g u i s t i q u e KlincksiecksParis. B r a n d i , L. & P. C o r d i n ( 1 9 8 1 ) . Some I t a l i a n D i a l e c t s " , rns. Pisa.  romane.  "On C l i t i c s a n d I n f l e c t i o n i n Scuola Normale S u p e r i o r e d i  B r e s n a n , J . & S. Pt. M c h o m b o ( 1 9 8 5 ) . " T o p i c , P r o n o u n , a n d A g r e e m e n t i n C h i c h e w a " , rns. S t a n f o r d U n i v e r s i t y . B r o d y , M. ( 1 9 8 4 ) . "On C o n t e x t u a l D e f i n i t i o n s C h a i n s " , L i n g u i s t i c I n q u i r y 15:355-380. E<rua, E.  (1938).  Fables  bonoises.  and t h e Role o f  T y p o - L i t h o : AI g e r .  B u r z i o, L. ( 1 9 8 1 ) . I n t r a n s i t i v e V e r b s a n d I t a l i a n D o c t o r a l D i s s e r t a t i o n , MIT, C a m b r i d g e .  Auxiliaries.  C a r r o l l , S. ( 1 9 8 £ ) . " R e d o u b l e m e n t e t d i s l o c a t i o n s e n p o p u l a i r e " , i n C. L e f e b v r e , e d . , t o m e l:£91~-357. Cassagrande, J". & B. S a c i u k , e d s . ( 1 9 7 £ ) . G e n e r a t i v e Romance L a n g u a g e s . Newbury H o u s e : R o w l e y . C h a o , W. ( 1 9 8 1 ) . C o n t r o l " , rns.  francais Studies i n  "Pro-Drop Languages and N o n - O b l i g a t o r y University of Massachusetts, Amherst.  C h o m s k y , N. ( 1 9 6 4 ) . C u r r e n t M o u t o n s T h e Hague. (1965). Cambridge.  Aspects  Issues  o f t h e Theory  (197£). S t u d i e s o n S e m a n t i c s Mouton:Paris. (1973).  in Linguistic o f Syntax.  Theory.  MIT  i n Generative  " C o n d i t i o n s on T r a n s f o r m a t i o n s " ,  Press;  Grammar.  i n S.  Anderson  P. K i p a r s k y , e d s . (1975).  Reflections  on Language.  Pantheon:New  York.  ( 1 9 7 6 ) . " C o n d i t i o n s o n R u l e s o f Grammar", L i n g u i s t i c A n a l y s i s £;3030-351. ( 1 9 7 7 ) . "On W h - M o v e m e n t " , i n P. C u l i c o v e r , T. Wasow &• A. Ak.rna.jian, e d s . (1980a).  "On B i n d i n g " , L i n g u i s t i c  Inquiry  11:1-46.  ( 1 9 8 0 b ) . "On C o g n i t i v e S t r u c t u r e s a n d T h e i r D e v e l o p m e n t : A R e p l y t o P i a g e t " , i n M. P i a t t e 1 1 i - P a l m a r i n i , e d . ( 1 9 8 0 c ) . "The L i n g u i s t i c P a l m a r i n i , ed.  Approach",  Piattelli-  (1981). Dordrecht.  Lectures  (198£). Government  Some C o n c e p t s a n d C o n s e q u e n c e s o f t h e T h e o r y o f a n d B i n d i n g . MIT Press:Cambridge.  Use,  on Government  i n M.  (1984). Knowledge o f Languages ms. M I T , C a m b r i d g e . (1986).  & H. Inquiry  Barriers.  MIT  L a s n i k (1977). 8:4£5-504.  and B i n d i n g .  i t s Nature,  O r i g i n s and  Press:Cambridge.  " F i 1 1 e r s and C o n t r o 1 " ,  Comorovski, I . ( 1 9 8 6 ) . " M u l t i p l e WH L i n g u i s t i c I n q u i r y 17s171-177.  Movement  C u l i c o v e r , P. &• M.S. R o c h e m o n t ( 1 9 8 3 ) . E n g l i s h " , L a n g u a g e 59s1£3-165. , T. Wasow & A. Academic PresssNew  Foris:  Akmajian, York.  Linguist ic  i n Romanian",  " S t r e s s and F o c u s i n  eds. (1977).  Formal  Syntax.  D a v i s , H. ( i n p r e p . ) . T h e A c q u i s i t i o n o f t h e E n g l i s h A u x i l i a r y S y s t e m and I t s R e l a t i o n s h i p t o L i n g u i s t i c T h e o r y . D o c t o r a l D i s s e r t a t i o n , U n i v e r s i t y of B r i t i s h Columbia, Vancouver. di  S c i u l l o , A.-M. ( 1 9 8 2 ) . " O p e r a t e u r s e t l i a g e , c l , AGR", d e 1' a s s o c i a t i o n q ue b e c o i s e d e l i n q u i s t i q ue £. £ s 4 7 - 5 6 .  D o b r o v i e — S o r i n , C. ( 1 9 8 5 ) . " C l i t i c D o u b l i n g a n d R e l a t i v e F o r m a t i o n i n R o m a n i a n a n d S p a n i s h " , rns. C N R S s P a r i s .  Revue Clause  D u b u i s s o n , C., L. E m i r k a n i a n & L. Lernay ( 1 9 8 3 ) . " L a dislocation, l a t o p i c a l i s a t i o n e t l a n o n - r e p r i s e du s u j e t s o n t - e l l e s des c r i t e r e s d e mat u r a t i o n s y n t a x i q u e ? " , R e v u e d e r a s s o c i a t i o n q ue b e c o i s e d e l i n g u i s t i q u e £.3 s£3-33. E m o n d s , J . ( 1 9 7 8 ) . "The V e r b a l C o m p l e x L i n g u i s t i c I n q u i r y 9s151-175.  V-V  in  French",  E v e r e t t , D.L. (1985). Parametr izat ion of  " P i r a h a C l i t i c - D o u b l i n g and t h e N o m i n a l C l i t i c s " , rns. MIT, Cambridge.  Ewert,  French  A.  (1966).  The  Language.  FabersLondon.  F a r k a s , D. ( 1 9 7 8 ) . " D i r e c t and I n d i r e c t O b j e c t R e d u p l i c a t i o n i n Romanian", Papers from t h e F o u r t e e n t h R e g i o n a l Meeting o f the Chicago L i n g u i s t i c Society. Fodor, J. (1980). " F i x a t i o n of B e l i e f i n M. P i a t t e 1 1 i - P a 1 m a r i n i , • e d .  and  Concept  Acquisition",  G a l a m b o s , S . J . ( 1 9 8 0 ) . "A C l a r i f i c a t i o n o f t h e N o t i o n T o p i c s E v i d e n c e f r o m P o p u l a r S p o k e n F r e n c h " , i n K r e i r n a n & A. E. Ojeda, eds. G o o d a l l , G.T. (1984). P a r a l l e l Dissertation, University of  S t r u c t u r e s i n Syntax. D o c t o r a l C a l i f o r n i a a t San Diego.  G r o s s , M. ( 1 9 6 8 ) . G r a m r n a i r e t r a n s f o r mat i o n n e l l e d u S y n t a x e du v e r b e . L a r o u s s e : P a r i s . Gruber, J. (1976). Amsterdam.  Studies  in Lexical  G u i r a u d , P. ( 1 9 7 3 ) . L e f r a n c a i s t a i r e s de F r a n c e s P a r i s .  Relations.  francaiss  North  populaire. Presses  H a r r i s , M. ( 1 9 7 8 ) . The E v o l u t i o n o f F r e n c h S y n t a x : Comparative Approach. L o n g m a n s New York.  Hollands  Universia  H e i r n , I.R. ( 1 9 8 2 ) . T h e S e m a n t i c s o f D e f i n i t e a n d I n d e f i n i t e Noun P h r a s e s . D o c t o r a l D i s s e r t a t i o n , U n i v e r s i t y o f Massachusetts, Amherst. Homstein, N. & D. L i g h t f o o t , e d s . ( 1 9 8 1 ) . L i n g u i s t i c s . L o n g m a n s New Y o r k .  Explanation i n  Huang, C . - T . J . (198£). L o g i c a l R e l a t i o n s i n C h i n e s e a n d t h e T h e o r y o f Grammar. D o c t o r a l D i s s e r t a t i o n , MIT, C a m b r i d g e . (1984). Pronouns",  "On t h e D i s t r i b u t i o n a n d R e f e r e n c e L i n g u i s t i c I n q u i r y 15s531-574.  H u r t a d o , A. ( 1 9 8 4 a ) . C. M a l e y , e d s .  "The U n a g r e e m e n t  Hypothesis",  ( 1 9 8 4 b ) . "On t h e P r o p e r t i e s o f L F ' " , P a p e r s i n L i n g u i s t i c s 5. (1985). " C l i t i c ReidelsDordrecht.  Chains",  o f Empty  t o appear  C o m e 11  i n L. K i n g Wprking  i n Pi. H u r t a d o ,  J a c k e n d o f f , R. ( 1 9 7 7 ) . X' S y n t a x s Pi S t u d y MIT PresssCambridge.  o f Phrase  ed. ,  Structure.  J a e gg 1 i , 0. Pi. ( 1 9 8 0 ) . On Some P h o n o l o g i c a l l y N u l l E l e m e n t s S y n t a x . D o c t o r a l D i s s e r t a t i o n , MIT, C a m b r i d g e . ( 1 9 8 1 ) . "Pi M o d u l a r Papers i n L i n g u i s t i c s (198£).  Topics  Pipproach t o P a s s i v e s " , C o y o t e £ s 41—59.  i n Romance S y n t a x .  in  Working  ForissDordrecht.  (1984). " S u b j e c t E x t r a c t i o n and t h e N u l l S u b j e c t P a r arnet e r " , P r o c e e d i n g s o f t h e F o u r t e e n t h P i n n u a l M e e t i n g the North E a s t e r n L i n g u i s t i c S o c i e t y , Department o f L i n g u i s t i c s , U n i v e r s i t y o f M a s s a c h u s e t t s , Plmherst.  of  (1985). "Three I s s u e s i n t h e Theory o f C l i t i c s s Case, D o u b l e d N P s , a n d E x t r a c t i o n " , t o a p p e a r i n H. B o r e r , e d . & C. S i 1 v a - C o r v a l a n , e d s . ( F o r t h c o m i n g ) . P r o c e e d i n g s o f t h e X I V L i n g u i s t i c Symposium o n Romance L a n g u a g e s . F o r i s s Dordrecht. Jake, J . (1980). "Object Verb Agreement i n T i g r e " , S t u d i e s i n t h e L i n g u i s t i c S c i e n c e s , Un i v e r s i t y o f 1 1 1 i n o i s , U r b a n a .  &  J e l i n e k , E. ( 1 9 8 5 ) . "The P r o j e c t i o n P r i n c i p l e a n d t h e A r g u m e n t Type Parameter". Paper p r e s e n t e d at t h e W i n t e r Meeting o f the L i n g u i s t i c S o c i e t y of America, S e a t t l e , Washington. Katz, J . J . York.  (1966).  The  P h i l o s o p h y of Language.  K a y n e , R.S. (1972). "Subject t i v e s " , i n J. Cassagrande (1975). French Press.Cambridge. &  Syntax;  ( 1 9 7 9 ) . "Two N o t e s L. R i z z i , e d s .  on  Harper  Inversion i n French & B. S a c i u k , e d s .  Interroga-  the Transformational Cycle. t h e NIC",  i n A.  Belletti,  ( 1 9 8 1 ) . "On C e r t a i n D i f f e r e n c e s B e t w e e n F r e n c h E n g l i s h " , L i n g u i s t i c I n g u i r y l£s349-371. (1984). Dordrecht.  Connectedness  RowsNew  and  Binary Branching.  L.  MIT Brandi  and  Foris:  & J.Y. P o l l o c k ( 1 9 7 8 ) . " S t y l i s t i c I n v e r s i o n , S u c c e s s i v e C y c l i c i t y , a n d M o v e NP i n F r e n c h " , L i n g u i s t i c I n q u i r y 9 s 5 9 5 — 6 c! 1. K e e n a n , E . 0 . & B. S c h i e f f e l i n ( 1 9 7 6 ) . " F o r e g r o u n d i n g R e f e r e n t s . A Reconsideration of L e f t - D i s l o c a t i o n i n Discourse", P r o c e e d i n g s o f t h e Second Annual Meeting o f t h e B e r k e l y U Linguistics Society. K i n g , L. & C. Linguistic Amsterdam. Koopman,  H.  Maley, eds. (1984). S e l e c t e d P a p e r s from t h e X I I I Symposium on Romance L a n g u a g e s . J o h n B e n j am i n s s  (1984).  The  Syntax  of Verbs. F o r i s : D o r d r e c h t .  & D. S p o r t i c h e ( 1 9 8 £ ) . " V a r i a b l e s a n d t h e P r i n c i p l e " , The L i n g u i s t i c R e v i e w £.139-160.  Projection  K o r n f i l t , J . ( 1 9 8 4 ) . C a s e M a r k i n g , A g r e e m e n t , and Empty Categories i n Turkish. Doctoral D i s s e r t a t i o n , Harvard University, Cambridge. K r e i m a n , J . & A.E. O j e d a , e d s . P a r a s e s s i o n on P r o n o u n s and Society.  (1980). Papers from t h e Anaphora. Chicago L i n g u i s t i c  L a n l y , A. ( 1 9 7 0 ) . L e f r a n c a i s d ' A f r i q u e d u 1 i n q u i s t i q u e . B o r d a s : P a r i s.  Nord.  Etude  L a po i n t e, S. G. ( 1 9 8 0 ) . ft T h e o r y o f G r a m m a t i c a l Agreement. Doctoral D i s s e r t a t i o n , U n i v e r s i t y of Massachusetts, Amherst. Laurendeau, P., M. N e r o n & R. F o u m i e r ( 1 9 8 £ ) . " C o n t r a i n t e s l ' e m p l o i du PR0~echo s u j e t en f r a n c a i s du Quebec", Revue 1 ' a s s o c i a t i o n q u e b e c o i s e de l i n g u i s t i q u e 1.i-£:115-1£8. L e f e b v r e , C. ( 1 9 8 £ ) . " Q u i q u i v i e n t ? o u Q u i v i e n t ? : question", i n C. L e f e b v r e , e d . , t o m e 1 : 4 7 - 1 0 1 .  voila  sur de  la  ed. (198£). L a s y n t a x e c o m p a r e e du f r a n c a i s s t a n d a r d e t p o p u l a i r e s a p p r o c h e s forme?! l e e t f o n c t i o n n e l l e . O f f i c e d e l a l a n g u e f r a n c a i s e , E d i t e u r o f f i c i e l du Quebec. L e v i n , J . & D. M a s s a m ( 1 9 8 5 ) . " S u r f a c e E r g a t i v i t y : C a s e / T h e t a R e l a t i o n s Reexamined", P r o c e e d i n g s of the F i f t e e n t h Annual Meeting of the North E a s t e r n L i n g u i s t i c S o c i e t y , Department of L i n g u i s t i c s , U n i v e r s i t y of Massachusetts, Amherst. L i e j i o n g , X. 17s75-93.  (1986).  "Free  Empty C a t e g o r y " ,  L i g h t f o o t , D. (198£). The L a n g u a g e L o t t e r y : G r a m m a r . MIT Press:Cambridge.  Linguistic Toward  Inquiry  a Biology  of  L o b e c k , A. ( 1 9 8 5 ) . "Why T h e r e I s No VP A n a p h o r a i n F r e n c h " . Paper presented at the Winter Meeting of the L i n g u i s t i c S o c i e t y of America, S e a t t l e , Washington. ( 1 9 8 6 ) . S y n t a c t i c C o n s t r a i n t s o n VP E l l i p s i s . D i s s e r t a t i o n , U n i v e r s i t y of Washington, S e a t t l e .  Doctoral  & Y. R o b e r g e ( 1 9 8 5 ) . " C o o r d i n a t i o n , V e r b R a i s i n g and Subject C l i t i c s " . Paper presented at the Winter Meeting the L i n g u i s t i c S o c i e t y of America, S e a t t l e , Washington. M a n z i n i , M.R. (1983). D i s s e r t a t i o n , MIT,  Restructuring Cambridge.  M a s s a m , D. (1985). Case Theory D o c t o r a l D i s s e r t a t i o n , MIT, May,  and  Reanalysis.  and t h e P r o j e c t i o n Cambridge.  of  Doctoral Principle.  R. ( 1 9 7 7 ) . T h e G r a m m a r o f Q u a n t i f i c a t i o n . D i s t r i b u t e d the Indiana U n i v e r s i t y L i n g u i s t i c s Club.  by  M c A ' N u l t y , J . ( 1 9 8 3 ) . " L e s c l i t i q u e s e t l a c-commande: 3 r a i s o n s p o u r a p p l i q u e r l a rnontee du v e r b e f i n i du c o t e g a u c h e de l a g r a m m a i r e " , Revue q u e b e c o i s e de l i n g u i s t i q u e  M e C l o s k e y , J - & K. H a l e I n f l e c t i o n i n Modern T h e o r y Is 487-533.  (1984). Irish",  "On t h e S y n t a x o f P e r s o n - N u m b e r N a t u r a l Language and L i n g u i s t i c  Mchombo, S . f t . ( 1 9 8 4 ) . " T h e N o n e x i s t e n c e o f V e r b - O b j e c t A g r e e m e n t i n B a n t u " , ms. U n i v e r s i t y o f M a l a w i a n d MIT. &• A.D. M t e n j e ( 1 9 8 3 ) . A n a l y s i s 11s £ 1 9 - £ 3 6 .  " N o n c y c l i c Grammar",  Linguist i c  M e y e r - L u b k e , W. ( 1 8 9 9 ) . G r a m m a t i k d e r r o m a n i s c h e n I l l s Romanische Syntax. L e i p z i g . M o r r i s s o n , C. ( 1 9 8 6 ) . Forthcoming Master Vancouver. N e i d l e , C. Studies  Spraehe, v o l  L e x i c a l Phonology o f C h i i c o t in. Thesis, University of British Columbia,  & R. N u n e z - C e d e i r a , e d s . ( F o r t h c o m i n g ) . L i n g u i s t i c i n Romance L a n g u a g e s . F o r i s s D o r d r e c h t .  N e m n i , M. ( 1 9 7 3 ) . V e r s u n e d e f i n i t i o n p h o n e t i q u e e t s y n t a x ! q u e de 1 ' i n c i s e en f r a n c o - c a n a d i e n e t en f r a n c a i s s t a n d a r d . Doctoral D i s s e r t a t i o n , U n i v e r s i t y of Toronto. P e r l m u t t e r , D. (.1971). D e e p a n d S u r f a c e S t r u c t u r e i n Syntax. H o l t , R i n e h a r t & Winston;New York. (1976). "Evidence f o r S u b j e c t Down-Grading", Schrnidt-Radef e l d t , ed.  Constraints i nJ.  (1978). "Impersonal P a s s i v e s and t h e U n a c c u s a t i v e Hypothesis", Proceedings o f t h e Fourth Annual Meeting Berkeley Linguistics Society.  of the  P i a t t e 1 1 i - P a 1 m a r i n i , M., e d . ( 1 9 8 0 ) . L a n g u a g e a n d L e a r n i n g . T h e D e b a t e B e t w e e n J e a n P i a g e t a n d Noam C h o m s k y . H a r v a r d Un i v e r s i t y P r e s s ; C a m b r i d g e . Picallo, CM. (1985). Opaque Domains. C i t y U n i v e r s i t y o f New Y o r k . Pollock,  J.-Y.  (1985).  " I I " ,  P r i c e , G. ( 1 9 7 1 ) . T h e F r e n c h Arno1d s London.  ms.  Doctoral Dissertation,  Universite  Language:  Q u i c o l i , C. ( 1 9 8 0 ) . " C l i t i c M o v e m e n t L i n g u i s t i c A n a l y s i s 6.£s131-186.  de P a r i s  Present  The  1£.  and P a s t .  Edward  i n French Causatives",  R a p o s o , E. ( 1 9 8 5 ) . " C a s e T h e o r y a n d I N F L - t o - C O M P . A S t u d y o f t h e I n f l e c t e d I n f i n i t i v e i n E u r o p e a n P o r t u g u e s e " , rns. U n i v e r s i t y of C a l i f o r n i a at Santa Barbara (to appear i n Linguistic Inquiry).  C.  ( 1 9 8 6 ) . "Romance I n f i n i t i v a l C l a u s e s N e i d l e &• R. N u n e z - C e d e i r a , eds.  and  Case Theory",  in  R e n z i , L. & L. V a n e l l i ( 1 9 8 £ ) . " I p r o n o m i s o g g e t t o i n a l c u n e v a r i e t a r o m a n z e " , i n S t ud i i n o n o r e d i G.B. Pellegrini, Padova. R i v a s , Pi. ( 1 9 7 7 ) . Pi T h e o r y MIT, Cambridge. Rizzi,  L.  (19B£),  Issues  of  Clitics.  in Italian  Doctoral  Syntax.  ForissDordrecht.  ( 1 9 8 4 ) . "On t h e S t a t u s o f S u b j e c t C l i t i c s 0. Pi. J a e g g l i & C. S i 1 v a - C o r v a l a n , e d s . (1986). Linguistic  "Null Objects i n Italian I n q u i r y 17:501-557,  and  Dissertation,  the  i n Romance", Theory  of  in  pro",  R o b e r g e , Y. ( 1 9 8 3 ) . " L e s i n t e r r o g a t i v e s s a n s inversion: g o u v e r n e r n e n t d u n o e u d INFL. e t h y p o t h e s e d'identite s t r u c t u r a l e " , Revue de 1 ' a s s o c i a t i o n q u e b e c o i s e de 1 i n q u i s t i q u e 3.£s119-131. ( 1 9 8 5 ) . "On D o u b l i n g a n d N u l l A r g u m e n t L a n g u a g e s " , t o appear i n Proceedings of the S i x t e e n t h Annual Meeting of the North E a s t e r n L i n g u i s t i c S o c i e t y , Department o f L i n g u i s t i c s , U n i v e r s i t y of Massachusetts, Amherst. (1986a). "Subject Argument Languages", 79.  D o u b l i n g , F r e e I n v e r s i o n , and N u l l Canadian J o u r n a l of L i n g u i s t i c s 31:55—  ( 1 9 8 6 b ) . " C l i t i c - C h a i n s and t h e D e f i n i t e n e s s R e q u i r e m e n t i n D o u b l i n g C o n s t r u c t i o n s " . P a p e r p r e s e n t e d a t t h e XVI L i n g u i s t i c Symposium on Romance L a n g u a g e s , A u s t i n , Texas. R o c h e m o n t , M.S. (1986), Benjamins:Amsterdam. Ross, J . (1967). Dissertation,  Focus  in Generative  Grammar.  C o n s t r a i n t s on V a r i a b l e s i n S y n t a x . MIT, Cambridge.  R o u v e r e t , A. ( 1 9 8 0 ) . " B u r l a n o t i o n d e p r o p o s i t i o n gouvernement et i n v e r s i o n " , Langaqes 60:70-107.  John Doctoral  finie,  & J . R. V e r g n a u d ( 1 9 8 0 ) . " S p e c i f y i n g R e f e r e n c e t o t h e S u b j e c t s F r e n c h C a u s a t i v e s and C o n d i t i o n s on Representations", L i n g u i s t i c Inquiry 11:97—202. S a f i r , K. ( 1 9 8 1 ) . " I n f l e c t i o n - G o v e r n m e n t L i n g u i s t i c Review 1:417—467. (1985). Cambridge.  Syntactic  Chains.  and  Inversion",  Cambridge U n i v e r s i t y  S c h m e r l i n g , S.F. (1973). " S u b j e c t l e s s Sentences of S u r f a c e S t r u c t u r e " , Papers from the Ninth Meeting of the Chicago L i n g i s t i c Society. Schrnidt-Radefeldt, ed. (1976). R e a d i n g s t i c s . North Hoi landsNew York. Sportiche, D. t i c Review  (1981). "Bounding ls219~246.  Nodes  Press:  and t h e N o t i o n Regional  i n Portuguese i n French",  (1983). S t r u c t u r a l I n v a r i a n c e and Symmetry D o c t o r a l D i s s e r t a t i o n , MIT, Cambridge.  Linguis-  The in  LinguisSyntax.  S t e r i a d e , D. (1980). " C l i t i c D o u b l i n g i n t h e Romanian C o n s t r u c t i o n s and t h e A n a l y s i s o f T o p i c a 1 i z a t i o n " , from the S i x t e e n t h Regional Meeting of the Chicago L i n g u i s t i c S o c i e t y. S t o w e 1 1 , T. ( 1 9 8 1 ) . O r i g i n s o f P h r a s e D i s s e r t a t i o n , MIT, Cambridge.  Structure.  S t r o z e r , J. (1976). C l i t i c s i n Spanish. D o c t o r a l U n i v e r s i t y of C a l i f o r n i a at Los Angeles.  The  WH Papers  Doct ora1 Dissertat ion,  T a r a l d s e n , K.T. ( 1 9 7 8 ) . "On t h e N I C , V a c u o u s A p p l i c a t i o n t h a t - t F i l t e r " . D i s t r i b u t e d by t h e I n d i a n a U n i v e r s i t y L i n g u i s t i c s Club.  and  Tellier, C. ( 1 9 8 6 ) . " D e s q u a n t i f i c a t e u r s f l o t t a n t s e n f r a n c a i s s t a n d a r d , e n q u e b e c o i s e t e n r n o o r e " , ms. M c G i l l U n i v e r s i t y , Montreal. T o r r e g o , E. ( 1 9 8 4 ) . "On I n v e r s i o n i n S p a n i s h Effects", L i n g u i s t i c Inguiry 15s103-129.  and  Some o f i t s  T r a v i s , L.D. ( 1 9 8 4 ) . P a r a m e t e r s and E f f e c t s o f Word O r d e r V a r i a t i o n s . D o c t o r a l D i s s e r t a t i o n , MIT, Cambridge. V i l l i a r d , P. ( 1 9 8 5 ) . " I N F L e t n o r n i n a t i f " , R e v u e d e t i o n q u e b e c o i s e de l i n q u i s t i g u e 4 . 4 s 1 6 9 - 1 7 8 .  1* a s s o c i a -  V i n e t , M.-T. ( 1 9 7 9 ) . " D i a l e c t V a r i a t i o n and a R e s t r i c t i v e T h e o r y o f Grammars a S t u d y o f I n t r a n s i t i v e P r e p o s i t i o n i n a V a r i e t y of French", Recherches 1 i n q u i s t i q u e s a Montreal 13s107-125. W i l l i a m s , E. ( 1 9 7 8 ) . " A c r o s s - t h e - B o a r d L i n g u i s t i c I n q u i r y 9 s 31-43. (1980). (1981). Ling u is t ic  "Pred i c a t i on",  Rule  Application",  L i n g u i s t i c Inquiry  "Argument S t r u c t u r e Rev i ew 1s 8 1 - 1 1 4 .  and  11s 2 0 3 - 2 3 8 .  Morphology",  The  Z a g o n a , K. ( 1 9 8 2 ) . G o v e r n m e n t a n d P r o p e r G o v e r n m e n t o f Project ions. Doctoral D i s s e r t a t i o n , U n i v e r s i t y of Washington, S e a t t l e . Z r i b i - H e r t s , A. ( 1 9 8 4 ) . " O r p h a n P r e p o s i t i o n s the Concept of N u l l Pronoun". D i s t r i b u t e d U n i v e r s i t y L i n g u i s t i c s Club. Z u b i z a r r e t a , M.-L. (1981). "Subject and t h e P r o - D r o p P a r a m e t e r " , ms. in  (1982a). " T h e o r e t i c a l P o r t u g u e s e " , ms. MIT,  Verbal  in French and by t h e I n d i a n a  E x t r a c t i o n i n Portuguese MIT, Cambridge.  Implications Cambridge.  of Subject  Extraction  ( 1 9 8 2 b ) . On t h e R e l a t i o n s h i p o f t h e L e x i c o n t o D o c t o r a l D i s s e r t a t i o n , MIT, Cambridge. Z w i c k y , A.M. & G.K. P u l l urn ( 1 9 8 3 ) . " C l i t i c i s a t i o n v s . I n f l e c t i o n s E n g l i s h n't", Language 59s502-513.  Syntax.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share