Krzysztof Migdalski, University of Wrocaw On the Decline of Tense and the Emergence of Second Position Pronominal Clitics This

talk analyzes diachronic changes in the position of pronominal clitics in Slavic and argues that they are related to the impoverishment of tense marking. In Old Church Slavic (OCS), pronominal clitics were verb-adjacent, while operator clitics that express Illocutionary Force (such as bo because, the ethical dative, e and li (focus/interrogation markers)) invariably occurred in second position (2P, see Radanovi-Koci 1988; Pancheva 2005; cf. 1). In modern Slavic, pronominal clitics are still verb-adjacent in Bg and Mac, yet in some other languages (Czech, S-C) they shifted to 2P. OCS had a complex system of tenses, with two simple past tenses, aorist and imperfect. In addition, aspectual distinctions (imperfective vs. perfective) were marked on all tenses and conflicting combinations of tense and aspect values (e.g. perf. aspect + imp. tense, cf. 2) were possible, which suggests the independence of tense/aspect systems. Compound tenses consisted of the non-tensed l-participle and the auxiliary BE that was marked for imperfective/perfective aspect in the present perfect/future tenses, respectively. Currently, aorist and imperfect are productive in Bg and Mac; in other Slavic languages compound tenses constructed with various aspectual forms of the auxiliary BE and the l-participle or the infinitive are adopted to describe past and future events. This means that temporal interpretations in these languages are derived from aspectual markings or from modality (as in the case of the future auxiliary will in South Slavic). Strikingly, the decline of the past tenses coincided with the emergence of 2P pronominal cliticization. In Czech and Slovene the simple past tenses disappeared by the 14 th-15th c.; in S-C aorist became restricted to certain dialects, losing its temporal semantics (Stieber 1973, Lindstedt 1994). During the same period verb-adjacent pronominal clitics switched to 2P in these languages (see Radanovi-Koci 1988 for S-C). I propose that the decline of tense distinctions was reflected by the loss of TP, as a result of which pronominal clitics could not adjoin to T 0 any more. The idea that certain languages lack TP is in line with Bokovis (2010) claim that languages without the DP layer also do not project TP (this explains the lack of expletives, the Sequence of Tense, and certain subject/object asymmetries in DP/TP-less languages; see also Lin 2010/Osawa 1999/Van Gelderen 1993 for arguments against the presence of TP in Chinese/Ancient Greek/Old English). Clitics are standardly assumed to be ambiguous categories sharing XP and X 0 properties which move from argument XP positions within VP and adjoin to T0 (Chomsky 1995: 249). Given Diesings (1992) Mapping Hypothesis, they must move out of VP, as they are inherently referential (cf. Uriagereka 1995). Moreover, Nash and Rouveret (2002) postulate that clitics need to be licensed by adjoining to a category endowed with active -features, i.e. T0. Yet, once T0 is lost, there is no suitable head for pronominal clitics to adjoin to and they end up in 2P, in separate maximal projections. The proposal developed here receives support from a number of contrasts in the syntax of 2P-clitic languages versus Bg/Mac, such as the distribution of the Person Case Constraint (PCC). The PCC restricts the occurrence of the dative clitic with an accusative clitic other than the 3 rd person. Assuming with Anagnostopoulou (2003) that the PCC results from the incompatibility of person and number feature checking on a single head, this analysis correctly predicts that the PCC strongly holds in Bg and Mac, but not in languages with 2P clitics (S-C, Slovene) that lack T0 (cf. 3 vs. 4). Furthermore, Stjepanovi (1998) notes that a higher part of pronominal clitics may be deleted in VP-ellipsis in S-C (cf. 5), while Bokovi (2002) shows that a similar operation in Bg or Mac is impossible (cf. 6). This is to be expected under the current analysis: if pronominal clitics in Bg/Mac all adjoin to T 0, deletion of a part of the clitic cluster involves deletion of a non-constituent. Next, Progovac (1993) observes that in S-C pronominal clitics may climb from an embedded subjunctive clause to the main clause (cf. 7a). Example (8b) shows that this never happens in Bg, which is not surprising, since X 0 elements may not raise from an embedded clause. Lastly, examples (9) and (10) indicate that negation attracts pronominal clitics in Bg, but not in S-C, where it attracts the verb instead. Since negation incorporates into other elements in Slavic (see Rivero 1991 for arguments based on the interaction of 2P and negation in S-C and Baszczak 2001 for arguments based on stress patterns), the finite verb in S-C may incorporate into ne located in Neg0, but pronominal clitics, which occupy XP positions, may not, given the Chain Uniformity Condition (Chomsky 1995: 253). Thus, the present account shows that the impoverishment of tense has led to radical changes in cliticization patterns in Slavic, which in turn gave rise to additional syntactic repercussions. Interestingly, what happened in Slavic seems to be the reversal of the process

that occurred in Ancient Greek, where according to Kiparsky and Condoravdi (2002), pronominal clitics shifted from 2P to verb-adjacency with the development of tenses and the emergence of IP. (11) (12) Elisaveti e isplni s vrm roditi ei. Elizabeth CL.FOC fulfilledREFL time give-birth herCL.DAT When it was time for Elizabeth to have her baby. (OCS, Pancheva et al 2007) TENSE/ASPECT IMPERFECTIVE PERFECTIVE 3SG PRESENT neset poneset 3SG AORIST nese ponese 3SG IMPERFECT nesae ponesae 3SG PERFECT nesl jest ponesl jest 3SG FUTURE II bdet nesl bdet ponesl (the verb nesti to carry in different tenses in OCS, cf. Van Schooneveld 1951: 97) *Az im te preporvam / Az te preporvam na tjax I themCL.DATyouCL.ACC recommend1SG I youCL.ACC recommend1SG to themACC I am recommending her to them Az im ja preporvam I themCL.DATherCL.ACC recommend1SG (Bg, Hauge 1999) Ja im te preporuujem I themCL.DATyouCL.ACC recommendPRES.1SG I am recommending you to them (S-C) Mi smo mu ga dali, a i vi ste mu ga dali, (takodje) we are himCL.DAT itCL.ACC gave and also you are himCL itCL gave too We gave it to him, and you did, too Mi smo mu ga dali, a i vi ste mu ga dali, (takodje) (S-C, Stjepanovi 1998) *Nie sme mu go dali, i vie ste mu go dali (sto) we are himCL.DAT itCL.ACC gave and you are himDAT himACC gave too We gave it to him, and you did too (Bg, Bokovi 2002: 331) Milan eli da ga vidi /?Milan ga eli da vidi Milan wishes that himCL.ACC sees /Milan himCL.ACC wishes that see3SG Milan wishes to see him (S-C, Progovac 1993) Manol iska da go vidi /*Manol go iska da vidi Manol wishes that himCL.ACC sees / Manol himCL.ACC wishes that sees Manol wishes to see him (Bg) Ne boli me /*Ne me boli NEG hurts meCL.ACC It doesnt hurt me Ne cini mi se da... /*Ne mi se cini da... NEG seem3SG meCL.DAT REFL that It doesnt seem to me that (S-C) Ne me boli /*Ne boli me NEG meCL.ACC hurts It doesnt hurt me Ne mi se struva, e ... /*Ne struva mi se, e... NEG meCL.DAT REFL seem3SG that It doesnt seem to me that (Bg)

(13)

a. b.

(14)

(15)

a. b.

(16)

(17)

a. b.

(18)

a. b.

(19)

a. b.

References: Anagnostopolou, E. 2003. The Syntax of Ditransitives. Mouton. Baszczak, J. 2001. Investigation into the Interaction between the Indefinites and Negation . Akademie Verlag. Bokovi, . 2002. Clitics as Non-branching Elements. LI 33:329-340. Bokovi, . 2010 On NPs and Clauses. Ms, UConn. Franks, S. & T.-H. King . 2000. A Handbook of Slavic Clitics. Oxford University Press. Gelderen, E. 1993. The Rise of Functional Categories. Benjamins. Hauge, K.R. 1999. The Word Order of Predicate Clitics in Bulgarian. JSL 7: 91-139. Kiparsky, P. & C. Condoravdi. 2002. Clitics and Clause Structure. Journal of Greek Linguistics 2: 1-39. Lin, J-W. 2010. A Tenseless Analysis of Mandarin Chinese Revisited. LI 41: 305-29. Lindstedt, J. 1994. On the Development of the South Slavonic Perfect. EUROTYP Working Papers. Nash, L. & A. Rouveret. 2002. Cliticization as Unselective Attract. Catalan Journal of Linguistics 1:157-199. Osawa, F. 1999. The Relation between Tense and Aspect: The Emergence of the T system. UCL Working Papers in Linguistics 11:521-543. Pancheva, R. 2005. The Rise and Fall of Second-Position Clitics. NLLT 23:103-167. Pancheva R. et al. (2007). Codex Marianus. In USC Parsed Corpus of Old South Slavic. Progovac, Lj. 1993. Locality and Subjunctive-like Complements in S-C. JSL 1:116-144. Radanovi-Koci, V. 1988. The Grammar of Serbo-Croatian Clitics. Ph.D. diss., Urbana-Champaign.

Rivero, M-L. 1991. Long Head Movement and Negation. The Linguistic Review 8:319-351. Schooneveld, C.H. van 1951. The Aspectual System of the Old Church Slavonic and Old Russian verbum finitum byti. Word 7:93-103. Stieber, Z. 1973. Zarys gramatyki porwnawczej jzykw sowiaskich. . PWN. Stjepanovi, S. 1998. On the Placement of S-C Clitics. LI 29:527-537. Uriagereka, J. 1995. Aspects of the Syntax of Clitic Placement in Western Romance. LI 26:79-123.