Complex predicates (CPs) are very common in many Iranian languages. However, the source of transitivity of CPs in these languages is, according to previous studies, somewhat uncertain. In some cases the transitivity of the light verb (LV) is found to determine the transitivity of the CP, whereas in other cases it depends on the semantic transitivity of the whole CP. In Southern Balochi (SB), as well as in several other Iranian languages, where ergative alignment is found with transitive (tr.) predicates formed with the past stem, and accusative alignment with intransitive (itr.) predicates formed with the past stem, as well as with all predicates formed with the non-past (present–future) stem, the transitivity of a CP is evident from the alignment.
Haig (2008: 11–12) discusses the transitivity of CPs and notes that not only in Balochi, but also in Northern Kurdish and Vafsi, semantically itr. CPs can trigger ergative case marking. He also makes two additional remarks for Iranian languages, namely that ―there are examples of etymologically tr. verbs shifting class under semantic pressure‖ and that ―there are interesting interactions between main verbs and auxiliaries‖.
The purpose of this presentation is to investigate the impact of the transitivity of the LV in determining the transitivity of a CP. The investigation is based on a corpus consisting of six texts and on interviews with three speakers of SB. The study shows that the transitivity of CPs is generally determined by the syntactic transitivity of the LV rather than the semantic transitivity of the whole CP. There is, however, one exception to this rule, namely the LV girag ‘to get‘, which occurs with accusative alignment in semantically itr. CPs.
Farrell (2003: 199) proposes that the unexpected accusative alignment in the past temporal domain for a CP with a tr. LV in SB could reflect the alignment of this CP in Urdu (i.e. when Urdu forms the corresponding CP with an itr. LV). However, there is only one verb, at least in the present corpus, that breaks the alignment rule, and it does so not only in dialects of SB spoken in Pakistan, e.g. Karachi Balochi, the dialect described by Farrell, but also in dialects spoken in Iran, which are hardly influenced by Urdu syntax at all. Two of the persons I interviewed are from Iran, and they both rule out ergative alignment for semantically itr. CPs involving the LV girag ‘to get‘.
A more reasonable explanation of the fact that girag ‘to get‘ appears at first sight to break the alignment rule, is that it is an ambitransitive verb. Assuming it is, we can conclude that syntactic transitivity always takes supremacy over semantic transitivity in SB, a rule that applies both to CPs and to periphrastic verb constructions.
At the end, I compare alignment of CPs a) in SB and Kurdish (a closely related Iranian language) b) in SB and Urdu (a language that SB is in close contact with, particularly in Pakistan).
Farrell, Tim 1995. Fading Ergativity? A Study of Ergativity in Balochi‖. In: Bennett, David C., Bynon, Theodora, and Hewitt, B. George (eds), Subject, Voice and Ergativity. Selected Essays. London: SOAS, 218–243.
Farrell, Tim 2003. Linguistic Influences on the Balochi Spoken in Karachi‖. In: Jahani, Carina, and Korn, Agnes (eds), The Baloch and Their Neighbours. Ethnic and Linguistic Contact in Balochistan in Historical and Modern Times. Wiesbaden: Reichert, 169–210.
Haig, Geoffrey L. J. 2008. Alignment Change in Iranian Languages. A Construction Grammar Approach [Empirical Approaches to Language Typology, 37]. Berlin and New York: Mouton de Gruyter.
Korn, Agnes 2009. The Ergative System in Balochi from a Typological Perspective‖. Iranian Journal of Applied Language Studies 1:1, 43–79.