Date of Award

Spring 4-1-2013

Document Type

Restricted Thesis

Terms of Use

© 2013 Jacob B. Phillips. All rights reserved. Access to this work is restricted to users within the Swarthmore College network and may only be used for non-commercial, educational, and research purposes. Sharing with users outside of the Swarthmore College network is expressly prohibited. For all other uses, including reproduction and distribution, please contact the copyright holder.

Degree Name

Bachelor of Arts


Asian Studies, Linguistics

First Advisor

K. David Harrison


Reduplication is found in an unparalleled thirteen percent of the lexicon in Sora, a Munda language of eastern India with 310,000 speakers. Most cross-linguistically attested forms of reduplication are present in Sora, from full to partial, faithful to unfaithful, fixed segmentism, triplication and ordered reduplication. Reduplication in Sora is often onomatopoeic and occurs at higher rates in children's language. However, reduplication is not productive; many of the base forms are not distinct lexemes in Sora. Previous models like the morphological doubling model (Inkelas & Zoll 2005) cannot successfully derive reduplication that is not semantically-driven. Other models, like Optimality Theory (Prince & Smolenksy 1993; McCarthy 2006) cannot successfully account for all forms of reduplication with a single set of constraints. However, the precedence relations model (Raimy 2000) is a loop-based model which accounts for all forms of reduplication in Sora, regardless of their semantics. Full reduplication, like [baŋ 'baŋ] 'to be strong', is derived from a precedence loop placed at the coda /ŋ/ that leads to the onset /b/, repeating the entire base. Partial reduplication, like [dʒu-'dʒud] 'to lull to sleep', requires that the beginning or the end of the loop be altered to encompass a portion of the base. Triplication, like [ke-ke-'ke] 'the scream of the peafowl', requires repetition of the loop and ordered reduplication, like [da-'daŋ-da-'daŋ] 'the sound of cutting wood', requires the interaction of distinct precedence loops. My intent is to provide a typology of reduplication in Sora and show that a single model can account for all forms.


Recipient of the Alice L. Crossley Prize in Asian Studies, first prize, awarded in 2013