Surprisal and Interference Effects of Case Markers in

Abstract

Based on the Production-Distribution-Comprehension (PDC) account of language processing, we formulate two distinct hypotheses about case marking, word order choices and processing in Hindi. Our first hypothesis is that Hindi tends to optimize for processing efficiency at both lexical and syntactic levels. We quantify the role of case markers in this process. For the task of predicting the reference sentence occurring in a corpus (amidst meaning-equivalent grammatical variants) using a machine learning model, surprisal estimates from an artificial version of the language (i.e., Hindi without any case markers) result in lower prediction accuracy compared to natural Hindi. Our second hypothesis is that Hindi tends to minimize interference due to case markers while ordering preverbal constituents. We show that Hindi tends to avoid placing next to each other constituents whose heads are marked by identical case inflections. Our findings adhere to PDC assumptions and we discuss their implications for language production, learning and universals.

What Determines the Order of Verbal Dependents in Hindi? Effects of Efficiency in Comprehension and Production

“…There have been some previous corpus-based investigations on word order variation in Hindi (e.g., Husain et al, 2013;Ranjan et al, 2019); (also see, Vasishth, 2004). For example, Ranjan et al (2019) investigated the role of case-marking on word order choices in Hindi and found evidence for the Easy-First and Reduce Inference principles of the Production Distribution and Comprehension (PDC) hypothesis (MacDonald, 2013). Production ease was operationalized in Ranjan et al (2019) as low n-gram/dependency surprisal value (Hale, 2001) and interference was operationalized as casemarker similarity between preverbal nominals.…”

“…For example, Ranjan et al (2019) investigated the role of case-marking on word order choices in Hindi and found evidence for the Easy-First and Reduce Inference principles of the Production Distribution and Comprehension (PDC) hypothesis (MacDonald, 2013). Production ease was operationalized in Ranjan et al (2019) as low n-gram/dependency surprisal value (Hale, 2001) and interference was operationalized as casemarker similarity between preverbal nominals. Jain et al (2018) investigated the Uniform Information Density hypothesis (Jaeger, 2010) with regard to predicting Hindi word order in corpus sentences vs random sentences and found no support for UID in capturing such a distinction.…”

Locality and expectation effects in Hindi preverbal constituent ordering

“…The reference sentence above reflects the predominant pattern in the corpus of reference sentences i.e. adjacent constituents tend to have non-identical case marking (Ranjan et al 2019). showed that adjacent nouns with identical case inflections are prohibited, adhering to the Obligatory Contour Principle (OCP) constraint on Hindi word order.…”

“…The success of parser surprisal arises because it factors in case marker features and also ignores the constituent lengths. Empirical evidence for this claim emerges from Ranjan et al (2019). They used a similar setup for referent sentence prediction (amongst variants) and showed that an artificial version of Hindi sans case markers resulted in a dip in prediction accuracy of 7% for dependency parser surprisal vis-a-vis 2% for trigram surprisal compared to natural Hindi.…”

“…Another approach would be to compute the percentage of times the reference would be ranked first in comparison to all other variants as in our earlier work(Ranjan et al 2019). Though our results could easily be upacked in this format and explained, due to space limitations we do not do so.…”