The Pashto genitive Diachronic and synchronic perspectives on a - PowerPoint PPT Presentation

The Pashto genitive Diachronic and synchronic perspectives on a typologically unique “prepositional” prenominal genitive Brianna Wilson blw512@york.ac.uk 1

Overview Pashto: Iranian language; Afghanistan and Pakistan • Word order • SOV • Adj.N • GenN • Circumpositions (+ some prepositions, postpositions) • Genitive marker də precedes the possessor • [[Prep – Gen] N ] • 1. 2

Typological considerations • Prepositional prenominal genitive unexpected Generalized consistency: (Hawkins, 1984; Longobardi & Silvestri, 2013) GenN à postpositions, NGen à prepositions Final-over-Final constraint (FOFC): (Holmberg, 2000) Head-initial phrase dominated only by head-initial • Head-final phrase dominated by head-initial OR head-final • Head-final = FOFC Violation Head-initial 3

Genitive case theory • Free genitive vs. functional genitive (Longobardi & Silvestri, 2013) Free genitive Functional genitive 1) Formally marked 1) Formally marked or zero-realization -Adpositionally or inflectionally -Never marked prepositionally 2) Freely iterable 2) Not iterable 3) Does not satisfy definiteness 3) Interacts with definiteness marking on head nominal Ex: Ex: English of- genitive (Romance de , di , English Saxon-genitive (German –s , German von ) Romance possessives mi, mio ) Pashto genitive: Free or functional? • Is də truly a preposition? • Is də- genitive iterable? 4

Diachronic explanations GREENBERG (1980) Amharic and Pashto • A prepositional prenominal genitive due to a series diachronic syntactic changes • Ordered Changes 1) NA à AN 2) NGen à GenN 3) Prepositional à circumpositional à postpositional • Origin of asymmetric genitive from a relative clause Development of a construction based on a relative clause • HOUSE THAT IS MAN’S [N [Rel Gen]] • Relative pronoun reanalyzed as a preposition and genitive phrase becomes • prenominal OF MAN HOUSE [[Prep Gen] N] • When language shifts to postpositional, the genitive preposition may be dropped • 5

Word order changes in Pashto Comparison of Trumpp (1873) to more recent grammars * 1) NA à AN order Trumpp: NA order was possible though considered archaic • 2) NGen à GenN order Trumpp: GenN is the usual order, but NGen is possible • 3) prepositional à circumpositional à postpositional Middle Iranian - prepositional (Hewson and Bubeník, 2006) • Modern Pashto circumpositions • də … pore ‘across’ • pə ... bānde ‘on top of’ • 1 st element commonly omitted • What about də ? * (Tegey & Robson, 1996; Babrakzai 1999; Roberts, 2000; David, 2014) 6

Relative clause origin? OLD IRANIAN Inflectional genitive, and a relative clause construction • 2. MIDDLE IRANIAN Loss of inflectional genitive & relative pronoun reanalyzed as a linker ( ezafe) • Phonetics forms of the ezafe: é, yé , or ə • MODERN IRANIAN (e.g. Farsi, Kurdish, Gilaki) 3. Problem : Pashto did not develop an ezafe construction… 7

də : a borrowing from Punjabi? • Indo-Aryan languages (e.g. Punjabi, Hindi) spoken to the east of Pashto • Punjabi and Hindi have a genitive enclitic (“postposition”) daa/kaa which agrees with the possessum (Thakur 1997) • Punjabi: daa/de/dii • Hindi: kaa/ke/kii • Proposal • Pashto borrowed the genitive marker from Punjabi • Explains certain dialectal variants of Pashto genitive: də, de, di , and perhaps da • Two other variants, ye and e, from ezafe? 8

Synchronic considerations Optional omission of də • Not discussed in previous literature • də dropping - a FOFC violation repair strategy? • Only in certain syntactic contexts 1. when possessor is modified by a determiner (except the indefinite) 2. in non-initial position in recursive genitives 3. with strong pronouns 4. when the object of a circumposition has a genitive 9

Də dropping: Possessor modified by determiner 4a. 4b. 5. 10

Də dropping: non-initial də in recusion 6. 7. 11

Interim summary • Pashto də- genitive: a FOFC violation and a typological anomaly • Evidence of syntactic changes proposed by Greenberg (1980) • Relative clause in Old Iranian led to ezafe in Middle and (some) Modern Iranian • Pashto is non-ezafe • Certain variants ( e, ye) from relative clause/ ezafe • Certain variants ( də, de, di ) borrowing from Punjabi • Synchronic evidence of də dropping in certain contexts: FOFC repair • What about free (of-genitive, iterable) vs. functional genitive (saxon genitive, non-iterable)? If də = preposition à də = iterable à də = free genitive If də ≠ iterable à də ≠ preposition à də = functional genitive 12

Slide 4 Genitive case theory • Free genitive vs. functional genitive (Longobardi & Silvestri, 2013) Free genitive Functional genitive 1) Formally marked 1) Formally marked or zero-realization -Adpositionally or inflectionally -Never marked prepositionally 2) Freely iterable 2) Not iterable 3) Does not satisfy definiteness 3) Interacts with definiteness marking on head nominal Ex: Ex: English of- genitive (Romance de , di , English Saxon-genitive (German –s , German von ) Romance possessives mi, mio ) Pashto genitive: Free or functional? • Is de truly a preposition? • Is de- genitive iterable? 13

Is də iterable? • Iteration: multiple genitives modifying the same head noun • The picture of Mary of John • The painting of Mona Lisa of Leonardo of the museum • de is not iterable 8. 9. 14

10. 11. 15

Conclusions • Given Longobardi & Silvestri (2013): • If də is not iterable then it must be free gen, not funct gen • Consequence: də cannot be a preposition • Conclusion • də- genitive is a functional genitive • də must be reclassified, an inflectional marker(?) • Advantages: if də is not a preposition, then there is no longer a Generalized Consistency violation (but still a FOFC violation) • What is the syntactic status of də? • Is də non-iterable in all Pashto dialects? • Is də dropable in all Pashto dialects? • What are the constraints on dropability? 16

Select references David, A. (2014). Descriptive grammar of Pashto and its dialects (Vol. 1). Walter de • Gruyter Greenberg, J. H. (1980). Circumfixes and typological change. In Papers from the 4 th • International Conference on Historical Linguistics (Vol. 14, pp. 233-241) Hawkins, J.A. (1984). Modifier-head or function-argument relations in phrase • structure? The evidence of some word order universals. Lingua, 63 , 107-138. Hewson, J., & Bubeník, V. (2006). From case to adposition: The development of • configurational syntax in Indo-European languages (Vol. 280). John Benjamins. Holmberg, A. (2000). Deriving OV order in Finnish. In The Derivation of VO and OV • (pp. 123-152). Amsterdam: John Benjamins. Larson, R.K. (2009). The nature of ‘attributive markers’. ConSOLE XVII , Nova Gorica. • Longobardi, G. & Silvestri, G. (2013). The structure of NPs. In The Bloomsbury • companion to syntax (pp. 88-117). New York: Continuum Publishing. Tegey, H., & Robson, B. (1996). A reference grammar of Pashto . Washington D.C. • Center for Applied Linguistics. Trumpp, E. (1873). Grammar of the Paštō or language of the Afghāns: Compared • with the Irānian and North Indian idioms. Trübner. 17

Strong pronouns With də dropping Standard 18

Object of circumpositions Object of circumpositions with a genitive with a genitive 19

Structure of recursive genitive 20