Aspectual concord and aspectual relativization nantes

Please download to get full document.

View again

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
 12
 
  1. 1 Aspectual concord and aspectual relativization: telicity in Serbo-Croatian Boban Arsenijević, Nantes, April 15th 2011 2. 2 Part 1: Event-related quantification ã…
Related documents
Share
Transcript
  • 1. 1 Aspectual concord and aspectual relativization: telicity in Serbo-Croatian Boban Arsenijević, Nantes, April 15th 2011
  • 2. 2 Part 1: Event-related quantification • Next to the standard conservative readings of quantifiers (collective, distributive, cumulative), there is a type of interpretation that targets events rather than referents of the quantified nominal expressions (see Krifka 1990). (1) The referee showed seven yellow cards. • Possibly 1 yellow card, but 7 events.
  • 3. 3 Targets non-specific arguments • Only non-specific quantified arguments can receive this type of interpretation. (2) a. 4000 ships passed through the lock last year. b. Certain 4000 ships passed through the lock last year. c. The 4000 ships passed through the lock last year.
  • 4. 4 Some facts from S-C • Targets direct objects. • Goals possible targets, iff there is no direct object with the right semantic properties (non-specific, not undergoing predicate incorporation). • Agentive subjects possible targets, iff there is no direct object and no goal with the right semantic properties. • DO<Goal<SubjectAg
  • 5. 5 Targets direct objects • When there is a non-specific direct object, only the direct object can bear an event- related quantifier (to the extent that it has the right properties). (3)a.Danas je dodao 8 šrafova majstoru. today Aux handed 8 screws craftsman b.Danas je dodao šraf 8rici majstora. today Aux handed screw 8Dat craftsmen
  • 6. 6 Goals possible targets… • … iff there is no object with the right semantic properties (non-specific, not undergoing predicate incorporation). (4) Danas je dodao šraf 8-rici majstora. today Aux handed screw 8.Dat craftsmen • The DO must be either specific/definite or it must semantically incorporate into the predicate for an event-related reading to be possible.
  • 7. 7 Agentive subjects • The same goes for agentive subjects. (5) 5 ljudi je zvalo dok nisam bio tu. 5 people Aux called while I was away
  • 8. 8 Similar to negative concord • Only non-specific non-incorporating participants. • Different in allowing only one instance of a concord marker. WHY? (6)a. Niko nije odveo nikoga nigde. N-who n-Aux lead n-who n-where b. (6) vodiča je odvelo 6 ljudi do (6) izlaza 6 guides Aux lead 6 people to 6 exits
  • 9. 9 Negative concord • Typology of negative concord with respect to the hosts: - all non-specific participants (S-C), - all non-specific participants c-commanded by the verb (Czech), - only one (non-specific) participant (German). • The German pattern for quantity? Why?
  • 10. 10 Neg vs. # • Semantically, and pragmatically, under negation, non-specific expression have an additional domain-broadening effect (Chierchia 2005); also, no introduction of referents to the discourse. • In other words, the interpretation of non- specific expressions is more strongly affected by a negative operator than by a quantifier.
  • 11. 11 Argument/role hierarchy • DO<Goal<SubjectAg - WHY? • A gradation in locality? • Targeting a pair of elements: the verb and an argument, where the argument is the most local one to the verb, carrying the right semantic properties to mark concord. • See Zeijlstra (2004) or Haegeman & Lohndal (2011) on West Flemish for analyses of this type of agreement.
  • 12. 12 • Adopting Haegeman & Lohndal (2011): AspP …VP 5 show card The analysis 5 -s
  • 13. How about universal quantification? 13 • Yes for all, but how about each? S-C: yes? AspP …VP each close window each
  • 14. 14 Why it can trigger concord? • Projections from the middle field known to trigger agreement: Fin, T, Mood, Pol. • AspP has been argued to be the projection at the edge between the lexical (VP) and the inflectional (IP) domain (Smith 1991). • Similar to TP (Demirdache & Uribe Etxebarria 1997). • Arguments that Slavic languages have no tense (e.g. Borik 2011) – Asp pushed to the middle field?
  • 15. 15 Features involved • I assume, with Verkuyl (1972), Tenny (1989), Borer (2005) among others that aspectual features, together with number and quantifier features, belong to the same group of quantity features. • Interpreted in AspP. • Other bearers of this type of features: the undergoer (as an incremental participant), the goal (as the bounder) and the verb (i.e. whether its lexical meaning is dynamic).
  • 16. 16 The hierarchy • This gives us another possible way to account for the hierarchy: the involvement in the quantity of the event. • incremental theme < bounders < other participants. • The verb needs to agree with something in the [asp] feature; interpretability of this feature is favored to just carrying it, which is favored to neither.
  • 17. 17 [asp]: hosting and interpretability [asp] interpretable carried verb + + incremental theme + + goal - + other participants - -
  • 18. Evidence for quantification over events 18 • Chinese event classifiers (Zhang 2002). a. Ta da-le Baoyu liang bazhang. he hit-PRF B two CLpalm ‘He hit Baoyu twice with his hand.’ (not necessarily with two hands) b. Ta da-lewo liangzuiba. he hit-PRF I two CLmouth ‘He slapped my mouth twice.’ Mandarin (Zhang 2002:
  • 19. 19 Part 2: Verbal prefixes • Almost all morphologically bare verbs in S-C are atelic and perfective. • Prefixation makes a verb perfective; it can be imperfectivized again, and/or receive another prefix. • No telic (culminative) meaning in S-C is expressed by a verb without a prefix (even kill, die, join, divide: u-biti, u-mreti, s-pojiti, raz-dvojiti).
  • 20. 20 Internal and external prefixes • Internal prefixes contribute a resultative component to the VP and correspond to the preposition heading the goal phrase. • The contribution of external predicates relates to the quantity of the eventuality (and of its incremental theme). (7)Iz-na-vlačio si reze na vrata. from-on-pull Aux.2Sg bars.Acc on doors ‘You placed all the bars on the doors.’
  • 21. 21 Internal vs. external • External prefixes can stack, unlike the internal (i.e. at most one prefix on a verb can be internal, and then there can be more than one external prefix). (8) Iz-na-po-razbijao sam tanjire. from-on-over-broke Aux1Sg plates ~‘I broke plates and thereby I exhausted all the (breaking of) plates, resulting in a lot of (breaking of) plates and I covered all the available (breaking of) plates.’
  • 22. 22 Internal vs. external • Internal prefixes take the position closest to the lexical verb. • In a series of prefixes, one of which is internal – this one must be the last one in the series, adjacent to the verb. (9) (Iz-)pod-(*iz-)vlačio si olovke pod from-on-from-pull Aux.2Sg pens.Acc under orman. cupboard ‘You placed all the pens under the cupboard.’
  • 23. 23 Internal vs. external • Internal prefixes may add an argument to the argument structure of a bare lexical verb, while the external prefixes cannot have this effect. (10) *(Od-)plesali su na terasu. of-danced Aux3Pl on terrace ‘They danced away onto the terrace.’
  • 24. 24 Standard analyses AspextP … vP … RP/AspintP • Two different functional projections for the two types of aspect. (Svenonius 2004, DiSciullo&Sla bakova 2005).
  • 25. 25 Grammatical aspect • Usually the external aspectual projection relates to the grammatical aspect, and the internal to the lexical aspect. • While the lack of resultative contribution in external prefixes indicates a difference at the lexical aspectual level, both types of prefixes have the same type of grammatical aspect effects (inducing perfectivity).
  • 26. 26 Resultativity – no difference • External prefixes are also resultative; the difference is in the complement of the preposition, DP vs. VP (Arsenijević 2006). (11) Jovan je na-pekao palačinke. J Aux on-baked pancakes ‘Jovan baked a lot of pancakes.’ ~Jovan acted and the result was ‘on pancake-baking’, i.e. event-accumulation.
  • 27. 27 Resultative interpretation • The quantity effects of the typical external prefixes can be derived from the semantics of the corresponding preposition: • na ‘on’: acummulation; • po ‘over’: full or partial coverage; • iz ‘from’: exhaustion; • za ‘behind’: cross a phase transition (zapevati, zalomiti, zaleteti se, zabrazditi)
  • 28. 28 Added argument – no difference • External prefixes also add non-selected arguments Arsenijević (2006), Žaucer (2010). (12)a. Pio sam (*se)vode. drink.ptcAux1Sg Refl water.Gen ‘I drank water.’ b. Na-pio sam *(se)vode. drink.ptcAux1Sg Refl water.Gen ‘I got my fill of drinking water.’
  • 29. 29 One projection for both • Conclusion: an analysis with both types of prefixes related to the same structural position is not only parsimonious, but also empirically better. • Both types of prefixes relate to the specification of result, but take different arguments.
  • 30. 30 Žaucer (2010) vP Jovan v’ V2P v palačinke V2’ [DO] PP pekao na [DO] V1P pekao na pekao na palačinke pekao
  • 31. 31 Remaining general problems • General problem 1: how is it syntactically encoded that the result predicate describes a subevent of the (referent of) macropredicate? • General problem 2: why the preposition still receives phonological realization in internal prefixation – together with the corresponding prefix?
  • 32. 32 A problem for Žaucer • Žaucer (2010) notes himself that his structure gives the wrong ordering: while the internal prefix ends up correctly on the left of the verb, the external prefix ends up on the right. PP … VP … PP pod vukao iz pod
  • 33. 33 Agreement • General problem 2: why the preposition still receives phonological realization in internal prefixation – together with the corresponding prefix? • This is a pattern typical for agreement. • Not surprisingly, established between the verb and the bounder, two elements that carry/contribute to the [asp] feature.
  • 34. 34 The analysis, step 1: scope sto [aspu] pod [aspu:pod] vukao [aspi:dyn] stolice VP PP
  • 35. 35 Analysis, step 1, effects • C-command: checking, valuation. sto [aspu:dyn:pod] pod [aspu:dyn:pod] stolice pod-vukao [aspi:dyn:pod] VP PP AspP
  • 36. 36 Analysis, step 2: relativisation sto [aspu:dyn:pod] pod [aspu:dyn:pod] pod-vukao [aspi:dyn:pod] stolice AspP PP AspP AspP pod-vukao [aspi:dyn:pod]
  • 37. 37 Analysis: step3, recursion PP AspP PP pod-vukao [aspi:dyn:pod] iz [aspu:iz]
  • 38. 38 External prefixes: recursive agreement iz [aspu:dyn:pod:iz] iz-pod-vukao [aspi:dyn:pod:iz] PP Asp2P PP Asp2P
  • 39. 39 Multiple aspectual relativization • The analysis derives structures analogous to Bianchi’s (2000) Kaynean analysis of multiple relative clauses. • Aspectual relativization: enables multiple (recursive) resultative specification. • The rather abstract interpretation of external prefixes comes from the fact that the complement of the respective preposition is a quantity specification (AspP), not a lexical phrase (VP).
  • 40. 40 Aspectual concord, i.e. agreement • Aspectual agreement with the preposition enables the relativization strategy (it encodes that probing has taken place and the original position of the VP/AspP). • Languages that have lexical elements to mark this agreement may also have aspectual relativization – others not.
  • 41. 41 Explanations • Stacking of external prefixes: prefixes stack in general, the fist one is special (agreement involving VP, not AspP). • Internal prefixes closest to the verb: only first round of relativization involves VP. • Quantitative vs. lexical interpretation: quantitative vs. lexical complement of P. • Prefixation of all the elements: agreement via left adjunction.
  • 42. 42 Explanations • Unification of the result- and the macro- event: relativization specifies one eventuality as an aspectual modification of another (~’e1 which is e2-bounded’). • Doubling of the prefix and preposition: agreement.
  • 43. 43 New issues: PP specifiers • How come external prefixes have no specifiers? • They actually do: either the direct object moves up to each of the SpecPPs, resulting in a measuring out interpretation. (13) Jovan je na-pekao palačinke. J Aux on-baked pancakes ‘Jovan baked a lot of pancakes.’ #..., although he baked a couple only.
  • 44. 44 New issues: PP specifiers • … or a new one is (i.e. externally) merged. (14) Jovan se na-pekao palačinki. J Refl on-baked pancakes.Gen ‘Jovan got his fill of baking pancakes.’ ..., although he baked a couple only.
  • 45. 45 New issues: goal-PP above VP • Standard analyses have it below, perhaps assuming that this corresponds to the temporal posteriority. • In line with general (cartographic) practice: all the modifiers specific for a projecting lexical item are in its extended projection. • P goes with a nominal expression which has its case assigned by the verb, not P.
  • 46. 46 New issues: goal-PP above VP • Trčim u sobu/sobi. run.1Sg in room.Acc/Loc ‘I’m running into/within the room.’ • Complements of goal Ps bear the same case as direct objects. • Postpositions in some languages: Ik ren de (in) kamer (in). Dutch I run the (in) room (in) Loc Acc
  • 47. 47 How come the order is P, DP? • i.e. how come not all languages are like Dutch? • Prepositions lexically specified as proclitics? • Agreement?
  • 48. 48 New issues: lexical bottoms • Kayne (2009) argues that lexical nouns cannot have complements because of their lexical nature – they have to bottom t heir structure. • The argument extends to lexical verbs, to the extent they are open class lexical elements. • Indeed, in the present analysis the verb bottoms its structure.
  • 49. New issues: Internal and external aspect • Unification of the internal and the external aspect. • Unboundedness of an event embraces the reference time and boundedness leaves it out of the event time. • This effect does all the syntactic temporal organization: no tense in Slavic languages. 49
  • 50. 50 Wrap up • Asp probes into the c-commanded structure, looking for elements with the [asp] feature. • When found, checking and/or evaluation are triggered, and possibly reflected in agreement. • In some languages, in the right syntactic context, an empty AspP may be projected, prompting aspectual relativization.
  • 51. 51 THANK YOU!
  • Related Search
    We Need Your Support
    Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

    Thanks to everyone for your continued support.

    No, Thanks