In several of the world’s languages, it has been observed that there is one ‘canonical’ and another ‘non-canonical’ marking of subjects. The common ‘canonical’ markings of the subject are nominative for a non-ergative language and agentive (or ergative) for an ergative language. ‘Non-canonical’ markings could be, e.g., genitive, dative, or locative.
Van Valin (2006: 684) makes a distinction between ‘experiencers’ and ‘purposeful instigators’ or ‘agents’ and finds that “[i]n many languages ... subjects that are experiencers appear in the dative case, whereas those that are willful instigators appear in the nominative or ergative case”. It seems to me in English, ça me plaît in French, me gusta in Spanish, and es fehlt mir in German are all examples of constructions with a dative experiencer.
Case studies of non-canonical subjects have been carried out on a variety of languages, including several languages on the Indian Subcontinent belonging to different language families, and thus spoken in the same greater linguistic area as Balochi, e.g., Hindi-Urdu (Davison 2004), Bangla (also called Bengali) (Dasgupta 2004, Onishi 2001), Gujarati (Mistry 2004), Nepali, Kashmiri and other languages of the Himalayas (Bickel 2004), Kannada (Amritavalli 2004), Malayalam (Jayaseelan 2004) and Tamil (Lakshmi Bai 2004).
Information on non-canonical subject constructions are found in various grammatical descriptions of Iranian languages (see, e.g., Lazard 1992: 111‒112, Jahani and Korn 2009: 666, Edelman and Dodykhudoeva 2009: 804‒805). Haig (2008) discusses non-canonical subjects in Iranian languages from a diachronic perspective, particularly their role in the emergence of ergativity in the past temporal field. He also presents constructions with a non-canonical subject for one specific Kurdish dialect, Badīnānī (Haig 2008: 257‒263). There are also a number of theoretical studies on Persian that deal with, among other constructions, non-canonical subject constructions, e.g., Barjasteh (1983) and Sedighi (2001).
The language under study here, Balochi, belongs to the Iranian branch of the Indo-European languages. It is spoken in south-eastern Iran, south-western Pakistan, southern Afghanistan, as well as in the UAE, Oman and other places on the Arabian Peninsula, in Turkmenistan, in India and in East Africa. It is generally classified as a North-Western Iranian language, although the strict borderline between North-Western and South-Western Iranian languages has recently been questioned by Paul (see, e.g., Paul 2003: 71) and Korn (see, e.g., Korn 2003, 2005: 329‒330).
The aim of this article is to describe and classify constructions with a non-canonical subject in three Balochi corpora available to the authors, namely Behrooz Barjasteh Delforooz’s corpus of tales and life stories from Iranian and Afghan Sistan (BS), Maryam Nourzaei’s corpus of tales, a life story and a procedural text in Koroshi, a dialect of Balochi spoken by scattered communities in Fars and other south-eastern provinces of Iran (BK), and Carina Jahani’s corpus of modern short stories from Pakistan (BP). These three corpora comprise approximately 90 000 words.
Costa Mesa: Mazda Publishers , 2012. 196-218 p.