The Definite Determiner in Romanian: A Biolinguistic Perspective

Abstract We present a new biolinguistic explanation for the evolution of the Nom/Acc and Gen/Dat forms of the Romanian definite determiner, thus providing further evidence to the hypothesis that the Directional Asymmetry Principle provides an explanation for the variation and evolution of the order of head-dependent constituents. As predicted, the choice between a valued and an unvalued variant of a functional feature, which was available in Old Romanian, is gradually reduced through the development of Modern Romanian. We provide an explanation for this evolution in terms of a more general natural complexity-reduction phenomenon. Keywords: Definite DeterminerHistorical DevelopmentDirectional AsymmetrySymmetry BreakingRomanian Notes *This work is supported in part by funding from the Social Sciences and Humanities Research Council of Canada to the Major Collaborative Research on Interface Asymmetries, grant number 214-2003-1003, and by funding from the FQRSC (programme de soutien aux equipes de recherche) on Dynamic Interfaces, grant number 137253. See www.interfaceasymmetry.uqam.ca; www.biolinguistics.uqam.ca. 1Reiss (Citation2012) argues for a model of the grammar including underspecification on the grounds of simplicity. 2Aldea (Citation2005) equally assigns the increased use of the preposition to the process of disambiguation. 3Without the associated focused interpretation that seems to be needed for the Latin examples, see Nicolae (Citation2009). 4See Nicolae (Citation2009) for more discussion. According to Pisani (1959: 105–106), cited in Rosetti (1986: 236), the enclitic definite determiner must have existed in Thracian. 5The earliest traces of Romanian Nom/Acc enclitic determiners date back to the fourteenth century, by which stage the grammaticalization process seems to have been completed and the paradigm of the definite article is enclitic. 6 lu' is a phonetically shortened form of lui. la is also a possible Dative case marker that can be employed with both masculine and feminine forms. However, while for most speakers lu(i) is best followed by a definite noun, la is mostly followed by an indefinite. 7When the prepositional marker is used, the definite forms seem to be more easily acceptable than the indefinite forms. Still, for some speakers of Romanian the following constructions are acceptable: (i) I-am spus lu' copil să nu întârzie. I have told to boy not to be late. (ii) I-am spus si la fată. I have told it to girl as well. 8More recently, Cornilescu and Nicolae (Citation2011) point out differences between Old Romanian and Modern Romanian with respect to the order of the definite determiner in the N-A/A-N pair, illustrated in (i)–(ii) below. Having argued for the standard derivation of the enclitic definite article from the post-nominal demonstrative ille as a case of reanalysis from [Spec,DP] to D0, they assume that the resulting suffix changes its c-selection features to [+N], thus allowing either a combination with an N or an A, with differences between OR and MR given by the availability of Long Distance Agreement in OR. The authors assume this combination to be the result of a morphological rule, rather than the effect of syntactic movement, thus the suffix combines with N in the lexicon. (i) spre ticăloase cuvintele mele audzul îti pleacă … (Cantemir) (OR) to vicious words.the my hearing your turn 'Lend your ear to my vicious words.' (ii) spre ticăloasele cuvinte ale mele auzul îti pleacă … (MR) to vicious.the words of my hearing your turn 'Lend your ear to my vicious words.' (Cornilescu and Nicolae 2011: 193) 9This analysis would differ from Giusti (Citation1993), where the definite article is a Case element and where the D0 is a Case position where the α-Case feature of the noun is assigned. 10Alexiadou et al. (Citation2007: 245) mention that psycholinguistic evidence supports a split between the Number Phrase and the Gender Phrase, cf. De Vincenzi and Di Domenico (Citation1999). Note that the proposed structure does not imply that the topmost projection of the extended Romanian DP is KP. While our discussion is limited to definite DP, the complete extended projection of the Romanian DP includes a determination area, an area of morpho-syntactic features projections and an agreement area. Our analysis is also compatible with assumptions where Case and Agreement are in the middle field of the DP, such as Ticio's (Citation2003) analysis of the Spanish DP, who follows Grohman's (Citation2000), among others, current assumption that DP parallels the CP structure and includes a thematic, an agreement and a discourse-related domain. The structure of the Spanish DP also includes an AgrP that regroups NumP, GenP, PossP, among other agreement related projections. 11Note that the definite features are equally part of the form of demonstratives (i) and possessive particles (ii): (i) ce-l ce-a ce-i ce-le this-masc.sg this-fem.sg this-masc.pl this-fem.pl (ii) a-l a a-i a-le POSS-masc.sg POSS-fem.sg. POSS-masc.pl POSS-fem.pl Complex determiner forms are also found in the Scandinavian DP (see Roehrs Citation2006). 12See Mardale (Citation2008) for a discussion on the means of expressing case in idiomatic expressions such as that in example (24). 13In Roehrs (Citation2006) the Modern Scandinavian enclitic definite determiner also follows an evolution from phrasal elements (the Early North Germanic demonstratives) to a suffixed article in a head position. Subsequent suffixation to the head noun occurred by movement of the noun, later replaced by movement of AgrP to (Spec,DP). The determiner itself is split and part of it undergoes movement within the extended projection of the DP. The possibility of there being a mapping between the split in the Scandinavian determiner and the Romanian determiner in terms of Case and referentiality features is a matter of further research. 14In Pesetsky and Torrego (Citation2001), structural Case is uninterpretable Tense on D. 15For ease of presentation, we'll illustrate the KP with its fully inflected case marker. 16Demonstrative adjectives are placed in a lower position in the extended DP, such as (Spec,FPagr)/or (Spec,NumP) (Giusti Citation2002; Cornilescu 1993; Brugè Citation2000, 2002). 17In Cornilescu and Nicolae (2011: 213) optionality in the numeration accounts for the analysis of the definiteness checking in OR and MR, which would be linked to the fact that adjectives 'optionally entered the derivation with uninterpretable unvalued definite feature, valued by Agree with the noun'.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.