Variation within Spanish adnominal possessive constructions

According to Lyons (1986) and Giorgi & Longobardi (1991), languages vary parametrically as regards the categorial status of possessive pronouns. Two main groups of languages can be distinguished: in ‘adjectival-genitive’ (AG) languages, the possessive is an adjective and occurs in combination with a determiner such as an article; in ‘determiner-genitive’ (DG) languages, the possessive has determiner status and occurs in complementary distribution with articles and other determiners (cf. Lyons 1999: 24, 130-134). It has been proposed that these two states are different steps on a grammaticalization cline leading from possessive adjectives within the noun’s inflectional domain towards possessive determiners in D° (Alexiadou 2004). Romance languages do not behave uniformly as regards the categorial status of possessives. While Italian, Portuguese and Catalan fall within the group of AG languages, French and Spanish are generally considered to be representatives of the DG type, cf. (1).

(1)       a.         It.        [_DP la [_AgrP mia_i … [macchina_k [_np t_i [_n t_k [_NP t_k ]]]]]]

1. Fr.        [_DP ma_i [_AgrP t_i voiture_k [_np t_i [_n t_k [_NP t_k ]]]]]

(Alexiadou 2004: 35)

But even within one language, both properties may coexist. Spanish has prenominal possessive determiners (with only number morphology), but postnominal possessive adjectives (with number and gender morphology), as illustrated in (2).

(2)       a.         tu-s libro-s

                        2sg.poss-pl book-pl

1. libro-s tuy-o-s

                        book-pl 2sg.poss-m-pl

In this talk, I will focus on prenominal possessives in Spanish, as in (2a), and show that their morphosyntactic status is more complex than is generally assumed. In particular, the existence of regional varieties of Spanish that show article-possessor combinations in prenominal position challenges the determiner analysis of possessives in (1b), which proves to be oversimplified. Relevant cases may be found in Central America where article-possessor combinations predominantly occur with the indefinite article (cf. 3a) and in Northern Spain where their occurrence is largely confined to the definite article (cf. 3b).

(3)       a.         Un hermano de mi esposa está casado con una mi hermana. (GUA080)

‘A brother of my wife is married to a my sister.’

1. El zorro si entra un día en el tu gallinero no te deja ninguna viva. (COSER-3412_01, Muñeca, Guardo)

‘The fox, if it enters one day in the your henhouse, it does not leave you a single one alive.’

In order to assess the morphosyntactic status of these constructions, a corpus analysis has been performed comparing all variants of attributive possession in Guatemalan Spanish (Preseea Guatemala corpus: PRESEEA 2014-2017, Verdugo de Lima et al. 2007, N=3377 possessive-marked noun phrases) and peninsular Spanish (COSER corpus: Fernández-Ordóñez 2005, N=9643 possessive-marked noun phrases). By means of a variable rule analysis performed with GoldVarb X (Sankoff et al. 2005), the influence of several language-internal environmental factors on the variation of possessive forms has been tested.

The results of the logistic regression analysis show that variants involving article-possessor combinations are favored by factors that relate to discourse-pragmatic functions such as contrast and emphasis. More precisely, article-possessor cooccurrences are particularly likely when the possessed noun has specific reference, when it is modified by a postnominal relative clause and when it is thematically persistent in being established as a discourse topic. In addition, they are favored when the possessee has human reference, when it is modified by diminutive suffixes and when the possessor has first person singular reference.

I will argue that these patterns can best be accounted for against the background of a split-DP analysis which presumes a tripartition of nominal expressions into a lower lexical domain, a central inflectional domain and an upper left periphery (in analogy to the clausal structure). The latter encodes discourse-pragmatic functions (cf. Ihsane & Puskás 2001). However, contrary to proponents of a split-DP analysis who postulate the presence of topic and focus phrases within the nominal structure (cf. Aboh 2004), I rather consider these notions to be of an inherently clausal nature and reserved for the sentential organization of discourse (cf. Giusti 2006). Instead, nominal expressions are sensitive to notions such as emphasis and contrast. This explains why in the Guatemalan data, the indefinite article preceding the prenominal possessive may vary with demonstratives, e.g. este/esta ‘this’, and the indefinite adjective otro/otra‘other’. The relevant structure is displayed in (4) where the DP splits into an upper DP (in analogy to the clausal ForceP) and a lower dP (similar to FinP, cf. Giusti 2006) which enclose projections with discourse-pragmatic functions, such as contrast (ContrP).

(4)       [_DP [_ContrP una/esta/otra [_dP [_d’ mi [_AgrP hermana [_nP mi [_n’ hermana [_NP hermana ]]]]]]]]

Apart from accounting for the observed discourse-pragmatic effects, the analysis in (4) has the additional advantage of considering the fact that the prenominal possessive has all signs of a fully grammaticalized D-element (as evidenced, among others, by its defective agreement morphology). In other words, article-possessor combinability does not necessarily imply that the possessor is an adjective within the noun’s inflectional domain, as suggested in (1).

Finally, in the absence of any discourse-pragmatic effects, the Spanish prenominal possessive acts as a definiteness marker and the nominal left periphery reduces to an unarticulated DP:

(5)       [_DP [_D’ tus [_AgrP libros [_nP tus [_n’ libros [_NP libros ]]]]]]

The analysis proposed here sheds a more differentiated light on the interrelation between the grammaticalization of the possessive into a determiner and its compatibility or incompatibility with other D-elements.

References:

Aboh, Enoch O. (2004): Topic and Focus within D. In: Leonie Cornips & Jenny Doetjes (eds.): Linguistics in the Netherlands 2004. Amsterdam & Philadelphia: John Benjamins, 1–12.

Alexiadou, Artemis (2004): On the development of possessive determiners: Consequences for DP structure. In: Eric Fuß & Carola Trips (eds.): Diachronic Clues to Synchronic Grammar. Amsterdam & Philadelphia: John Benjamins, 31–58.

Fernández-Ordóñez, Inés (dir.) (2005-2017): Corpus Oral y Sonoro del Español Rural. <www.corpusrural.es> [consulted: June 14^th, 2017] ISBN 978-84-616-4937-2.

Giorgi, Alessandra; Longobardi, Giuseppe (1991): The Syntax of Noun Phrases: Configuration, Parameters and Empty Categories. Cambridge: Cambridge University Press.

Giusti, Giuliana (2006): Parallels in clausal and nominal periphery. In: Mara Frascarelli (ed.): Phases of Interpretation. Berlin: Mouton de Gruyter (Studies in Generative Grammar, 91), 163–184.

Ihsane, Tabea; Puskás, Genoveva (2001): Specific is not definite. In: Generative Grammar in Geneva 2, 39–54.

Lyons, Christopher (1986): The syntax of English genitive constructions. In: Journal of Linguistics 22 (1), 123–143.

Lyons, Christopher (1999): Definiteness. Cambridge: Cambridge University Press.

PRESEEA (2014-2017): Corpus del Proyecto para el estudio sociolingüístico del español de España y de América. Alcalá de Henares: Universidad de Alcalá. <http://preseea.linguas.net> [consulted: June 14^th, 2017].

Sankoff, David; Tagliamonte, Sali; Smith, Eric (2005): Goldvarb X. A Variable Rule Application for Macintosh and Windows. Department of Linguistics: University of Toronto.

Verdugo de Lima, Lucía; Palma Chacón, Ana María; Fong de Rivera, Mayra; Acevedo-Halvick, Ana (2007): PRESEEA Guatemala – Proyecto para el estudio sociolingüístico del español de las ciudades de Guatemala. Proyecto PRESEEA-Guatemala. Guatemala City: Intituto de Lingüística y Educación, Universidad Rafael Landívar.