West Coast Conference on Formal Linguistics (WCCFL) (38th : 2020)

Agreement with deficient pronouns in Laki : A syntactic repair to a clitic cluster restriction Taghipour, Sahar; Kahnemuyipour, Arsalan 2020-03-06

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


73804-Taghipour_S_et_al_Laki_clitic_WCCFL38_2020.pdf [ 223.78kB ]
JSON: 73804-1.0389902.json
JSON-LD: 73804-1.0389902-ld.json
RDF/XML (Pretty): 73804-1.0389902-rdf.xml
RDF/JSON: 73804-1.0389902-rdf.json
Turtle: 73804-1.0389902-turtle.txt
N-Triples: 73804-1.0389902-rdf-ntriples.txt
Original Record: 73804-1.0389902-source.json
Full Text

Full Text

 1 Agreement with deficient pronouns in Laki: A syntactic repair to a clitic cluster restriction1  Sahar Taghipour   Arsalan Kahnemuyipour University of Toronto          University of Toronto Mississauga  WCCFL 38, University of British Columbia, March 6, 2020  Goal: We aim to account for the complex agreement pattern in Laki (Southern Kurdish, Northwestern Iranian) using otherwise motivated assumptions. We will argue that the alignment follows a split-ergative system, with T (in the past tense) showing a striking contrast in the realization of agreement with full DPs or strong pronouns, as opposed to deficient pronouns. We will provide an account for this contrast.    Main claims:  - Laki split ergative alignment arises due to the presence of a single locus of Agreement on T in past intransitive and present clauses, versus two loci of Agreement on T and v in past transitive clauses.  - We show that in past intransitive or present clauses, T agrees with the subject (realized as verbal suffixes) and there is no further Agreement.   - In past transitive clauses, v Agrees with the subject (realized as clitics), leaving T available for further Agreement with an accessible argument. When there is no accessible argument (due to locality), the Agreement on T is realized as default Ø.   - We suggest that the accessibility of an argument for T agreement in past transitive clauses depends on a PF constraint banning the adjacency of two clitics. When in-situ realization of the clitic violates this PF constraint, this clitic raises to the edge of the phase becoming accessible to T and realized as T Agreement on the verb. Otherwise, it remains in-situ and does not become accessible for T Agreement.  1. Subject agreement pattern for present and past intransitive verbs  • A present transitive/intransitive verb (1-2) and a past intransitive verb (3) express subject agreement with suffixes in Table I. We use Roman numeral superscripts to mark Table I affixes. Note: The [3sg] subject agreement in the past is Æ (4).                                               Table I. Agreement suffixes                      (1)   ali yo    maryam   to-na2   ma-s̆nās-enI.                      (2) ali-a      ma-s̆in-iI                     Ali and Maryam  you-SP   DUR-know.PRS-3PL                      Ali-SP    DUR-sit.PRS-3SG                              ‘Ali and Maryam know you.’                                            ‘Ali is sitting.’                                                 1 We are grateful to the participants of the annual meeting of the Canadian Linguistics Association (CLA 2019) in Vancouver, BC, the Syntax Project group at the University of Toronto and the audience at the TU+5 workshop at the University of Delaware for their feedback and helpful discussion.  2 We have used SP to mark specificity. This characterization requires further investigation.  1SG 2SG 3SG 1PL 2PL 3PL -(e)m -(i)n -i/Æ -(i)men -(i)nān -(e)n  2 (3) (to) zu       č-inI.                                                              (4) maryam  nis̆t-ÆI.           you soon   go.PST-2SG                                                                          Maryam  sit.PST-3SG          ‘You left/went soon.’                                                            ‘Maryam sat.’                                                                                                                                                                                            2. Subject agreement pattern for past transitive verbs  • A past transitive verb (5-6) expresses subject agreement with enclitics in Table II. These enclitics appear on the first constituent in the verb phrase, second position in the vP.3  Second position clitics in the verb phrase have been attested in other languages (see Kahnemuyipour and Megerdoomian 2011 for Armenian auxiliary clitics and Karimi 2010, 2012, Bonami and Samvelian 2008, among others for subject agreement clitics in Kurdish varieties).                                             Table II. Agreement enclitics          (5) sif-ela=tII                   wārd5.                   (6) ali  yo      maryam   to=nānII      s̆enās-i.                              apple-PL.DEF=2SG       eat.PST                        Ali and    Maryam  you=3PL      know-PST            ‘You ate the apples.’                                    ‘Ali and Maryam knew you.’  • Table II enclitics can also be used as pronominal clitics for direct objects or objects of prepositions in present tense. In (7), we see an example of a strong pronoun used as a direct object and (8) is its cliticized counterpart. A similar contrast is illustrated for prepositional objects in (9-10). (Note: Clitic doubling is not allowed in Laki.)  (7) (me) to-na      ma-s̆nās-emI.                                      I       you-SP    DUR-know.PRS-1SG            ‘I know you.’  (8) (me) ma-s̆nās-emI=etII.                                                            I       DUR-know.PRS-1SG=2SG                        ‘I know you (sg).’                                                (9) az̆      owen-a    ma-pers-emI.      from  them-SP   DUR-ask.PRS-1SG     ‘I (will) ask them.’  (10) az̆en=nanII-a   ma-pers-emI.        from=3PL-SP     DUR-ask.PRS-1SG       ‘I will ask them.’                                                    3 In this talk, we do not attempt to provide an account for the positional distribution of the agreement enclitics or suffixes. This is an interesting area of research left for future work.  4 The subject agreement for [3sg] is an exception as it appears on the verb as a Table I suffix. We are abstracting away from this issue here.  (i)  ali   maryam s̆enās-iI   Ali Maryam  know.PST-3SG    ‘Ali knew Maryam.’ 5 Later in the talk, these verbal forms will be analyzed as containing a default Æ 3sg agreement.   1SG 2SG 3SG 1PL 2PL 3PL =(e)m =(e)t =i4 =mān =tān =(ā)n  3 Side question: Is the split really about past vs. present?    • Following much of the literature on Kurdish ergative alignment, we use the present vs. past distinction as the source of split alignment for illustration purposes (Karimi 2010, cf. Haig 2008). Meanwhile, we recognize that the actual source of the split cannot be tense, as the ergative pattern is also found in the present perfect (11). The difference can very likely be attributed to an aspectual difference (as suggested for other languages, see Coon 2013), but it cannot be a straightforward perfective-imperfective split, as the ergative pattern is also found in the past progressive and simple past. We have left a closer examination of this issue for future research.   (11) ali  yo    maryam   to=nānII    s̆enāsi-ya.                                Ali and  Maryam  you=3PL   know-PERF                                 ‘Ali and Maryam have known you.’  3. Agreement alignment in the past tense: direct and prepositional objects   • Direct objects: Recall from (5-6) that in past transitive clauses, agreement with the subject is realized as a second vP enclitic (Table II). With full DP objects (12) and a full pronominal object (13) there is no further agreement on the verb.  (12) sif-ela=tII                  wārd.                                     apple-PL.DEF=2SG      eat.PST                            ‘You ate the apples.’                                        (13) ali  yo    maryam   to=(n)ānII    s̆enās-i.          Ali and   Maryam  you=3PL      know-PST             ‘Ali and Maryam knew you.’  Meanwhile, when the object has no overt realization, the phi-features are realized as Table I suffixes on the verb.         (14) a. di-(e)nI=etII.                                            b. di-nI=ānII.                                       see.PST-3PL=2SG                                                   see.PST-2SG=3PL                  ‘You (sg) saw them.’                                  ‘They saw you (sg).’  • Prepositional objects: With a full DP (15) or a strong pronoun (16) there is no further agreement (other than the 2P in vP subject agreement enclitic).        (15) vet=emII          aben   maryam.            tell.PST=1SG    to       Maryam               ‘I told Maryam.’                                           (16) az̆      owen=emII    pers-i.                    from   them=1SG     ask-PST         ‘I asked them.’  When the prepositional object is not a full DP or a strong pronoun, a contrast arises between post-verbal and pre-verbal PPs:   4 - When the PP is pre-verbal (17-18), the phi-features of the prepositional object are realized as Table I suffixes on the verb. Here, the subject agreement enclitic appears on the preposition (as the first element in the vP).  (17) a.  aben=emII   vet-inI.                       (18) a. az̆en=emII   persi-(i)nI.           to=1SG        tell.PST-2SG                                     from=1SG  ask.PST-2SG               ‘I told you (sg).’                                        ‘I asked you (sg).’          b.  *aben=etII=emII   vet.                                 b.  *az̆en=etII=emII   pers-i.                         to=2SG=1SG      tell.PST                                           from=2SG=1SG  ask-PST                           intended: ‘I told you (sg).’                              intended: ‘I asked you.’           c. *aben=etII   vet=emII.                                 c. *az̆en=etII   pers-i=m.                       to=2SG      tell.PST=1SG                                            from=2SG  ask-PST=1SG                    intended: ‘I told you (sg).’                               intended: ‘I asked you.’  - When the PP is pre-verbal but there is an additional direct object, the 2P subject agreement appears on the direct object and the prepositional object is realized as a Table II enclitic on the preposition (19-20).   (19) rāz-a=mII              aben=etII  vet.                              secret-DEF=1SG     to=2SG      tell.PST                                  ‘I told you (sg) the secret.’                                                    (20)  soāl-a=mII              az̆en=etII     pers-i.                      question-DEF=1SG  from=2SG   ask-PST              ‘You (pl) asked me.’    - When the PP is post-verbal (21-22), the phi-features of the object are realized as Table II enclitics on the preposition, with subject agreement realized as a Table II enclitic on the verb.       (21) a. vet=mII            aben=etII.              (22) a. persi=tānII       az̆en=emII.      tell.PST=1SG    to=2SG                                   ask.PST=2PL     from=1SG     ‘I told you (sg).’                                              ‘You (pl) asked me.’             b. *vet=etII            aben=emII.                       b. *persi=mII       az̆en=tānII.        tell.PST=2SG     to=1SG                                  ask.PST=1SG  from=2PL      intended: ‘I told you (sg).’                              intended: ‘You (pl) asked me.’             c. *vet-inI            aben=emII.                      c. *persi-mI         az̆en=tānII.        tell.PST-2SG   to=1SG                                   ask.PST-1SG   from=2PL                      intended: ‘I told you (sg).’                           intended: ‘You (pl) asked me.’   4. Agreement alignment in the context of possessives  • The enclitics in Table II can also function as possessive pronouns, encoding the phi-features of the possessor.   5 (23)   Possessor marking on ketew ‘book’       • With a full DP possessor (24) or a full pronominal possessor (25), there is no further verbal inflection, regardless of the tense.  (24) to     sag maryam-a     m-own-inI.              you   dog Maryam-SP  DUR-see.PRS-2SG             ‘You (will) see Maryam’s dog.’   (25) to    sag   men=etII     di.              you  dog  my=2SG     see.PST             ‘You saw my dog.’              • In the present tense, when the possessor is not a full DP or a strong pronoun, the possessor’s phi-features are invariably realized as Table II enclitics on the possessed NP. We see an example of a possessive construction in object and subject positions in (26) and (27), respectively. Agreement is with the subject, realized on the verb as a Table I suffix.        (26) homa      sag-a=mII-a                     m-own-inānI.              You(pl)   dog-DEF=POSS.1SG-SP    DUR-see.PRS-2PL              ‘You (pl) see my dog.’                          (27) rafix-ela=(a)nII                  men-a       ma-s̆nās-enI.              friend-PL.DEF=POSS.3PL     me-SP       DUR-know.PRS-3PL             ‘Their friends know me.’  • In the past tense, when the possessor is a deficient pronoun, there is variation depending on the grammatical function of the possessive noun phrase:  • If the possessive noun phrase is the subject of the clause, the possessor pronoun is invariably realized as a Table II enclitic, and subject agreement is expressed on the first element in the vP as a Table II enclitic.       (28) rafix-ela=(e)tII                     men=nānII    s̆enāsi.             friend-PL.DEF=POSS.2SG      me=3PL        know.PST             ‘Your (sg) friends knew me.’  • If the possessive noun phrase is the object of a transitive clause, the possessor is realized as a Table I suffix on the verb. Meanwhile, the subject agreement enclitic appears as 2P in vP, here on the possessum.  (29) a. to    no   sārā   sag-a=tānII     di-mI.            you and  Sara  dog-DEF=2PL   see.PST-POSS.1SG           ‘You and Sara saw my dog.’ 1sg ketew=em 1pl ketew=mān 2sg ketew=et 2pl ketew=tān 3sg ketew=i 3pl ketew=ān  6     b. *to    no   sārā   sag-a=mII=tānII      di.                  you  and  Sara  dog-DEF=POSS.1SG=2PL        see.PST             intended: ‘You and Sara saw my dog.  c. *to    no   sārā   sag-a=mII      di=tānII.             you and  Sara  dog-DEF=POSS.1SG     see.PST=2PL                  intended: ‘You and Sara saw my dog.       (30) a. keyk-a=mānII     wārd-enI.             cake-DEF=1PL     eat.PST-POSS.3PL            ‘We ate their cake.’   b. *keyk-a=n=mānII               wārd.                   cake-DEF=POSS.3PL=1PL  eat.PST-                  intended: ‘We ate their cake.’   c. *keyk-a=nII                wārd=mānII.              cake-DEF=POSS.3PL    eat.PST=1PL                 intended: ‘We ate their cake.’  5. Schematic summary of the above agreement patterns   We take strong pronouns to have a larger structure than deficient pronouns (Cardinaletti and Starke 1994, and subsequent authors). We will use DP for the former and φP/φ (which is just a bundle of features) for the latter. (underlining tracks subject agreement)  (31) Past Intransitive and Present    a. Full DP / Strong Pronoun DPsubj  DPobj  V-Tab. I b. Deficient Pronoun  DPsubj  φP/φobj V-Tab. I=Tab. II  (32) Past Transitive     Direct object: a. Full DP / Strong Pronoun DPsubj  DPobj=Tab. II V b. Deficient Pronoun  DPsubj  φP/φobj  V-Tab. I=Tab. II  Prepositional object (post-verbal):   c. Full DP / Strong Pronoun DPsubj  V=Tab. II  [P DP] d. Deficient Pronoun  DPsubj  V=Tab. II  [P=Tab. II     φP/φ]   Prepositional object (pre-verbal): e. Full DP / Strong Pronoun DPsubj  [P DP]=Tab. II V   f. Deficient Pronoun  DPsubj  [P=Tab. II     φP/φ] V-Tab. I   Prepositional object (pre-verbal, additional object)  g. Full DP / Strong Pronoun DPsubj    DPobj=Tab. II  [P DP]  V  7 h. Deficient Pronoun  DPsubj    DPobj=Tab. II    [P=Tab. II     φP/φ] V  (33) Possessive DPs (Past tense)   Subject position: a. Full DP / Strong Pronoun [Possessum   DPPoss’r]subj      DPobj=Tab. II      V b. Deficient Pronoun  [Possessum=Tab. II   φP/φ Poss’r]subj DPobj=Tab. II       V   Object position: c. Full DP / Strong Pronoun DPsubj     [Possessum   DPPoss’r]obj=Tab. II        V d. Deficient Pronoun  DPsubj     [Possessum=Tab. II   φP/φ Poss’r]obj        V-Tab. I  6. Analysis  • We propose that Laki agreement follows a split-ergative alignment system.  • In past intransitive and present clauses6 (31), subject agreement can be explained starightforwardly via Agree between T and the subject, realized as the suffixes in Table I. This is an accusative alignment with only T being a locus of Agreement.                      • We posit that in past transitive clauses, there are two loci of agreement, one on T and the other on v (see Legate 2002, 2008, Aldridge 2008 and references therein).  • The v head Agrees with the external argument, realized morphologically as the Table II enclitics. That leaves T available for further Agreement with another argument.                                                 6 As discussed previously, the past-present split may be revisited as an aspectual split, if we take the so-called present/past stems in Iranian languages as realizing an aspectual contrast (see Kahnemuyipour and Megerdoomian 2002 for Persian).           TP                                                                                                                                                                                                                           T’                                     T             vP                                                                                   Subj                    v’                 v                      VP                              V’                          V           Obj T AGREEMENT (44)  8 • In a past transitive clause with a full DP or strong pronominal object (32a), we see v Agreement with the subject realized as a Table II enclitic on the object but no Agreement between T and the object. We take the absence of an Agree relation with a full DP object or strong pronoun to be a locality issue, with T and the object being in different phases (Chomsky’s 2001 PIC), leading to a default realization of the phi-features on T as Æ (3sg).                     • The patterns found with full DPs or strong pronouns in other contexts can be explained in a similar fashion straightforwardly. In all these cases, we only find Agreement of the subject with v appearing as a Table II enclitic (2P in vP). This can be seen in (32a, c, e, g) and (33a, c). No agreement between T and any other argument is established due to locality as above.  • Turning to the context of the past transitive clause with a deficient pronominal object (32b), we see the expected Table II enclitic realizing agreement with the subject, but in addition we find a Table I suffix on the verb. We propose that this is the result of an Agree relation between T and φP/φ. This raises the question of why Agreement with T becomes available in this context? In other words, how are the phi-features accessible to T here?   • This pattern is reminiscent of the complementarity between strong pronouns (or full DPs) and subject agreement in Celtic languages (see, for example, Jouitteau and Rezac 2006 for Breton and McCloskey and Hale 1984 for Irish).  • There is a debate in the literature about the preferred account for this complementarity. Some view it as a case of incorporation of the deficient pronoun into the verb (Anderson 1982, Doron 1988, Ackema and Neeleman 2003) and others take it to be the result of Agreement (McCloskey and Hale 1984, Stump 1984, Legate 1999) between T and the deficient pronoun and seek an explanation for why this Agree relation is absent with full DPs or strong pronouns.    • It is important to note that in Laki, the phi-features of the deficient pronoun are clearly realized as agreement suffixes, both in terms of form and position. These are the same forms     TP                                                                                                                                                                                                                                T’                                    T                      vP                                                                                     Subj                  v’                 v                       VP                              V’                          V            DPObj v AGREEMENT T AGREEMENT (45)  9 found in run-of-the-mill nominative subject agreement in the present tense realized on the verb. This provides support for an Agree account of these markers.  • Furthermore, importantly, in Laki, this complementarity does not hold fully. While Table I agreement is always absent with full DPs or strong pronouns, deficient pronouns do not always lead to Table 1 agreement, as we can see in (32d, h) and (33b). In these cases, the deficient pronouns are realized as Table II enclitics.  • The generalization that seems to arise can be formulated as a clitic cluster restriction (see Arregi & Nevins 2012 for Basque and Tyler 2019 for Choctaw): Deficient pronouns cannot be realized on an element which already hosts a Table II enclitic: * X-Table II-Table II.   • We propose that in contexts where a deficient pronoun is competing for the same host with the Table II subject agreement enclitic, the phi-bundle cannot be realized on that host and instead moves to the edge of vP, becoming accessible to T for Agreement. As a result of this Agree relation between T and φP/φ, we see a Table I suffix being realized on the verb, similarly to the straightforward subject agreement between T and the subject in past intransitive and present tense sentences.                          • Accordingly, if a different host becomes available for the realization of the deficient pronoun in the vP domain, the competition does not arise, and the deficient pronoun is realized on this other host as a Table II enclitic, as expected. Such cases arose above in the context of a post-verbal PP (32d), preverbal PP with an additional object that hosts the subject agreement enclitic (32h) and a possessive used as the subject of the clause, where the object hosts the subject agreement enclitic and the deficient possessive pronoun is realized on the possessor (33b).    • A question may arise as to what happens to the deficient pronoun φP/φ when the Agree relation with T is established. Why don’t we see it realized as a Table II enclitic on some other element? We posit that whenever an Agree relation is established with a deficient pronoun, the deficient pronoun is deleted. Crucially, this behavior of deficient pronouns                  (46)                  TP                                                                                                                                                                                                                                         T’                                 T            =φP/φ  vP                                                                                     Subj                    v’                   v                     VP                                  V’                                        V                      v AGREEMENT T AGREEMENT  10 stands in sharp contrast with strong pronouns. The latter will not get deleted after Agreement, as we see in the case of agreement with a strong pronominal subject. We attribute this difference to the φP status of deficient pronouns, as opposed to the DP status of strong pronouns.   • This analysis raises an interesting interface question with respect to how we can have a syntactic resolution to an apparently PF problem. One undesirable option would be to allow for a global assessment (or filtering) of different derivations, which would lead to the illicit derivation crashing and the licit one winning out. Meanwhile, a strictly local assessment can also give us the right result with close attention to the timing of various operations. Here is a sketch of one such solution:   - Second position clisis has to take place in syntax or as an early PF phenomenon. In other words, when we reach the phasal head v, second position clisis is established.  - Once the second position clitic finds itself next to another deficient pronoun, the PF constraint is identified and the offending deficient pronoun moves to the edge of the phase and escapes spell-out.  - Spell-out takes place, and in the next phase, the Agree relation between T and the deficient pronoun is established, at which point the deficient pronoun is deleted and as a result not realized at PF.  7. Conclusion  • We have investigated agreement alignment in Laki and have proposed the following to account for the complex distribution of suffixes and enclitics realizing phi-features:   - Laki shows a split ergative alignment. In the past intransitive and present tense sentences, there is only a single locus of Agreement, T, which exhibits straightforward nominative agreement with the subject realized as a Table I suffix.   - In the past transitive clauses, there are two loci of Agreement: v and T. v establishes an Agree relation with the subject, realized as a Table II enclitic on the first element in vP (2P in vP). T is available for further agreement.   - In the context of full DPs and strong pronoun, T cannot establish an Agree relation due to locality and is instead realized as a default 3sg Æ.   - With deficient pronouns, a division arises. When the deficient pronoun has a host that is different from the one which hosts the subject agreement enclitic, each of them is realized as a Table II enclitic on its distinct host.   - When the deficient pronoun competes for the same host with the subject agreement enclitic, it moves to the edge of the phase, becoming accessible for Agreement with T, leading to the realization of its phi-features on the verb and a Table I suffix.  - This proposal raises interesting questions about the nature of PF constraints and their interaction with syntax. We have sketched one possible analysis for how the interface  11 mechanisms may play out. Meanwhile, these issues need to be further investigated in future research with an eye to variations observed in Kurdish dialects.   References  Ackema, P., Neeleman, A., 2003. Context-sensitive spell-out. Natural Language and Linguistic Theory 21, 681–735. Aldridge, E., 2008. Generative approaches to ergativity. Language and Linguistics Compass 2.5: 966-95. Anderson, S.R., 1982. Where’s morphology? Linguistic Inquiry 13, 571–612. Arregi, K. and A. Nevins. 2012. Morphotactics: Basque auxiliaries and the structure of spellout. Springer. Bonami, B. and Samvelian, P., 2008. Sorani Kurdish person markers and the typology of agreement. In Thirteenth International Morphology Meeting, Vienna. Cardinaletti, A. and Starke, M., 1994. The typology of structural deficiency: A case study of the three classes of pronouns. Clitics in the languages of Europe, 8. Chomsky, N., 2001. Derivation by phase. In Ken Hale: A life in language, ed. M. Kenstowicz, 1–52. Cambridge, MA: MIT Press. Coon, J. 2013. Aspects of split ergativity. Oxford University Press. Doron, E., 1988. On the complementarity of subject-verb agreement. In: Barlow, M., Ferguson, C.A. (Eds.), Agreement in Natural Language: Approaches, Tendencies and Descriptions. CSLI, Stanford, CA, pp. 201–218. Haig, G. 2008. Alignment change in Iranian languages: A construction grammar approach. Vol. 37. Walter de Gruyter. Jouitteau, M., and Rezac, M., 2006. Deriving the complementarity effect: Relativized minimality in Breton agreement. Lingua, 116(11), pp.1915-1945. Kahnemuyipour, A. and Megerdoomian, K. 2002. The Derivation/Inflection Distinction and Post-Syntactic Merge. Talk presented at the Annual Conference of the Canadian Linguistic Association, Toronto. Kahnemuyipour, A. and Megerdoomian, K., 2011. Second-position clitics in the vP phase: The case of the Armenian auxiliary. Linguistic Inquiry, 42(1), pp.152-162. Karimi, Y., 2010. Unaccusative transitives and the Person-Case Constraint effects in Kurdish. Lingua, 120(3), pp.693-716. Karimi, Y., 2012. The evolution of ergativity in Iranian languages. Acta Linguistica Asiatica, 2(1), pp.23-44. Legate, J., 1999. The morphosyntax of Irish agreement. In: Arregi, K. (Ed.), Papers on Morphology and Syntax, Cycle one. MIT Working Papers in Linguistics 33. MITWPL, Cambridge, MA, pp. 219-40. Legate, J., 2002. Warlpiri: theoretical implications, dissertation. Cambridge, MA: MIT. Legate, J., 2008. Morphological and abstract case. Linguistic Inquiry 39.55–101. McCloskey, J., Hale, K., 1984. On the syntax of person-number inflection in Modern Irish. Natural Language and Linguistic Theory 1, 487–534. Stump, G., 1984. Agreement vs. incorporation in Breton. Natural Language and Linguistic Theory 2, 289–426. Tyler, M. 2019. Absolutive promotion and the condition on clitic hosts in Choctaw. Natural Language & Linguistic Theory 37.3, 1145-1203. 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items