The 9th meeting of Formal Approaches to South Asian Languages (FASAL-9) takes place March 16-17, 2019, at Reed College in Portland, Oregon. The meeting includes paper and poster presentations on the languages of South Asia from within any subfield of theoretical linguistics.
Linguistics Collegiate Fellow Savithry Namboodiripad is of three invited speakers at FASAL-9. Savi will present “An Experimental Approach to Flexible Constituent Order: Insights from Malayalam.” Her talk discusses constituent order and language contact in Malayalam, which is a language Savi grew up speaking and the language she primarily focused on in her 2017 thesis work.
An Experimental Approach to Flexible Constituent Order: Insights from Malayalam
Languages vary as to how flexible their constituents (subject, object, and verb) are, as has long been of interest to linguists working on South Asian languages (see the edited volumes by Dayal & Mahajan 2004 and Butt, King, & Ramchand 1994). While progress has been made in terms of accounting for this flexibility theoretically (e.g., Bhatt & Dayal 2007, Mohanan 1982), flexibility in constituent order remains under-described for many languages, mostly for (I argue) methodological reasons. In the first part of this talk, I present an operational definition of flexibility as introduced in my thesis work (Namboodiripad 2017) and developed thereafter (Namboodiripad, Kim, & Kim 2019): measuring the extent to which speakers prefer the canonical order and disprefer non-canonical orders in discourse-neutral contexts, via formal acceptability judgment experiments.
I focus on Malayalam, in which all six logical orderings of (subject, object, & verb are grammatical, attested, and have the same truth-conditional meaning, and the canonical (discourse neutral, intonationally unmarked) order is SOV. I will discuss the methodology and results of two separate experiments involving Malayalam speakers living in Kerala, in which subjects rated the six variants of transitive sentences, with animate subjects and inanimate objects, on a 7-point scale.
Next, I discuss language contact, another topic of particular relevance to linguists studying South Asian languages. Malayalam itself is a language borne out of contact between Dravidian and Sanskrit, and the language as spoken in Kerala has been in contact with English for generations, resulting in changes at the level of the lexicon and phonology (Kala 1977, Asher & Kumari 1998, Namboodiripad, Garellek, & Bakovic in prep.). Constituent order is particularly diffusable and susceptible to contact effects (Bickel et al. 2017), though the process by which this occurs remains a fertile area for further exploration.
I consider how experience with English affects constituent order in Malayalam: It could be that Malayalam speakers with a high level of English contact borrow the canonical SVO order from English wholesale, resulting in a relative preference for SVO. Alternatively, high-contact Malayalam-speakers could be less tolerant of movement in general, resulting in an increased relative preference for the canonical order in Malayalam, SOV.
The data and methodology presented will serve as a starting point for testing typological claims about constituent order, including investigations of the role of contact, in inflectional morphology, and case-marking as correlates of flexibility. I conclude by putting these experiments in the context of translated replications of this methodology in Avar, Korean, and English, showing that this measure can yield meaningful comparisons within and across languages, both enriching linguistic descriptions and helping to motivate observed typological distributions.