Differential Object Marking with Inanimate Direct Objects
Marco García García (Universität zu Köln)
It is generally accepted that Differential Object Marking (DOM), i.e. the language-internal variation of the realisation of direct objects (DOs), is controlled by different prominence features of the DO-referent such as animacy, definiteness, specificity or topicality (cf. Aissen 2003 or Bossong 1985 among others). With respect to DOM in Spanish, animacy seems to be the most important factor among these conditioning features. Definite and/or specific DO-referents are not marked with the Spanish differential object marker, the preposition a, unless the referent is human or at least animate (e.g., Conozco (*a) esta pelicula vs Conozco a este actor). Despite this general dependency on animacy we do find DOM with inanimate DOs. This is the case with some relational verbs such as preceder ('precede') or seguir ('follow') as well as with some examples of the type el entusiasmo vence (a) la dificultad ('The enthusiasm conquers the difficulty') in which DOM is favoured although not required obligatorily.
Many theories assume that DOM is motivated by the need for disambiguation between subjects and DOs that can hardly be distinguished semantically (cf., Comrie 1979 among others). Usually, the disambiguation principle is associated with the concept of markedness and used as an explanation for cases in which the direct object resembles the subject, for instance, when the DO- as well as the subject-referent are [+animate] and [+definite], which are unmarked values for subjects, but marked values for DOs.
According to de Swart (2003) the need for disambiguation also triggers DOM when it is the subject that resembles the DO, for instance, when the subject- as well as the DO-referent are both [-animate] and [-definite] being unmarked values for objects, but marked values for subjects. Such an extended version of the disambiguation principle seems to provide an account of many cases of DOM with inanimate DO-referents like the one cited above where not only the DO (la dificultad) but also the subject (el entusiasmo) is [-animate]. I will check this extended disambiguation principle by means of a test corpus research in order to show that the (non-)marking of inanimate DOs cannot be predicted by this principle. Therefore, my discussion will address the following questions: What are the factors that trigger and (dis)favour DOM with inanimate DOs? Why does the combination of semantically marked DOs and unmarked subjects trigger DOM more frequently and rather obligatorily, whereas in the mirror-inverted case, i.e., in the combination of semantically marked subjects and unmarked DOs, DOM is less frequently and rather optionally? What consequences does this asymmetry have for the disambiguation principle?