Formal Approaches to South Asian Languages (E-Journal)
Not a member yet
116 research outputs found
Sort by
Egocentric questions: The view from Bangla and Hindi-Urdu
Full text linkIn Bangla and Hindi-Urdu, we find a kind of question that is grammatically restricted to being about a particular event. This kind of a question consists of a demonstrative pronoun followed by a plain question. We refer to such questions as egocentric questions and to the demonstrative pronoun they con- tain as the egocentric pronoun. The egocentric pronoun picks out an event and the question is about this event. Since the speaker and the hearer need to pick out the event the question is about, such questions cannot be used in a state of speaker ignorance. This differentiates them from plain questions where speaker ignorance is the default. We show that various properties of egocentric questions follow from the need to be able to assign a reference to the egocentric pronoun and from the nature of access the speaker has to the event that the egocentric pronoun picks out. 
Relative clause constructions in New Indo-Aryan languages: Hierarchies of macro roles
Full text linkSince the seminal work by Keenan & Comrie (1977), typological studies have shown that languages vary with respect to the range of arguments that can be relativized on. In this study, we systematically examine what can be relativized in five New Indo-Aryan (NIA) languages: Hindi-Urdu, Nepali, Early Nepali, Sinhala, and Bengali. Inspired by typological studies on relative clauses, we conducted our examination using a novel systematic methodology. First, we examined both headless and headed relative clauses. Second, we examined relativization on arguments for each of the macro roles S, A, P, T, and R. Lastly, we examined every participial strategy for relative clause constructions when a language had different participles for tense or aspect. Our investigation showed that there are both similarities and differences in the relativizability of NPs in relative clause constructions in the five NIA languages examined. On the one hand, in each language examined, arguments of the same range of macro roles can be relativized on in both headed and headless relative clauses. On the other hand, the five languages differ as to which macro roles can be relativized on. Based on this difference of the relativizability of NPs and our novel methodology, we propose hierarchies of relativizability for these NIA languages. The hierarchies are the onset-oriented hierarchy {S} > {A} > {P, T, R} for relative clause constructions by imperfective/nonpast participles and the termination-oriented hierarchy {S, P, T} > {A} > {R} for those by perfective/past participles. We explained these hierarchies in terms of viewpoint, localist metaphor, and a metonymy relationship. 
Dependent dative case in Hindi-Urdu
Full text linkWhile dative case has traditionally been analysed as a case assigned to a DP by a head (Chomsky 1981, 1986; Woolford 2001, 2006), Baker & Vinokurova (2010) and Baker (2015) have argued that dative case in Sakha is a dependent case in the sense of Marantz (1991). Following Baker & Vinokurova (2010)’s analysis of Sakha, this paper proposes a dependent case analysis of dative case in Hindi-Urdu, based on crucial evidence from the causativised ingestive construction. This account is novel support for the view that dative case may be a dependent case in some languages
Kannada through the lens of the NP/DP parameter
Full text linkBoškovi?’s (2008, 2012) NP/DP parameter suggests that languages with definite articles are DP language and languages without definite articles are NP languages. However, more recent literature on the topic demonstrates that some article-less languages may be DP languages (see, e.g., Syed and Simpson 2017 on Bangla; Dees 2020 on Dholuo). This paper explores Kannada, an article-less Dravidian language, based on a number of Boškovi?’s NP/DP-divide generalizations. The results demonstrate that Kannada patterns like the NP languages from Boškovi? (2008, 2012). It is then illustrated that Kannada may lack certain movements within the nominal domain that have been associated with other article-less languages which have been proposed as DP languages. These results provide necessary details for better understanding what the NP/DP ‘divide’ looks like cross-linguistically. 
The internal syntax of -iva clauses in Vedic Sanskrit
No full textIn this paper we argue that a proper characterization of the synchronic internal syntax of clauses of comparison in Vedic Sanskrit can clarify issues of interpretation. We also briefly explore aspects of the diachrony of these structures. 
Gender Agreement in a Tamil-Hindi Bilingual Situation: The Role of Feature Valuation
Full text linkWithin the Generative paradigm, variation is understood in terms of features. A crucial mechanism for this is the Borer-Chomsky Conjecture, explained in Baker (2008) as: ‘All parameters of variation are attributable to differences in the features of particular items in the lexicon.’ This paper is an attempt to understand a case of asymmetric bilingualism of Hindi-Urdu (Indo-Aryan) and Tamil (Dravidian) and to explain the resultant changes in gender agreement in terms of featural configurations of functional heads. The empirical core is set in New Delhi, with simultaneous bilinguals acquiring Tamil (L1) at home and Hindi-Urdu (L2) from the external environment. While both grammars mark gender information on the verb, Hindi-Urdu has grammatical and biological gender. In Tamil, on the other hand, gender is purely semantic. The Hindi- Urdu grammar of this bilingual population appears to find gender agreement challenging. This paper adopts a representational approach; the loss of Gender is analysed as the deletion of an uninterpretable valued feature on a functional head. The first approach is to posit the uninterpretability of the feature as the cause of deletion. The inability of this claim to hold up empirically is then taken to mean that the explanation for the deletion of the feature lies in its other property: Value. Reanalysis of the change in contact situations reveals that losing valued features could simply be a strategy adopted by languages in an effort to be more parsimonious. 
DOM in Kodava takk: a complex interaction among multiple factors
Full text linkThis paper presents novel data from Kodava takk (Dravidian), also known as Coorgi, which exhibits the well-attested syntactic phenomenon of Differential Object Marking (DOM). Crosslinguistically, objects which are differentially marked tend to be associated with features such as specificity and/or definiteness, humanness, animacy, or a combination of these. Well- known examples of specificity-driven DOM include Turkish (von Heusinger and Kornfilt 2005) and Senaya (Kalin 2018), whereas direct objects in Spanish (Ormazabal & Romero 2013) and Hindi (Dayal 2011, Bhatt & Anagnostopolou 1996) receive differential marking on the grounds of animacy/humanness and specificity. As will be illustrated, this phenomenon is most definitely present in Coorgi, as the accusative case-marker does not always appear on direct objects. However, on the surface, there is no clear-cut featural split between objects which do and do not receive this case-marker. Instead, this differential marking is triggered by a complex interaction of multiple factors: animacy, specificity, number, humanness, and inherent lexical properties of verbs. This paper outlines the interactions which derive Differential Object Marking in Coorgi and offers a formal analysis to capture the empirical facts, which modifies Kalin’s (2018) account where DOM is a result of nominal licensing. This paper not only provides complex novel data from an understudied and endangered language, but also deepens our understanding of this crosslinguistic phenomenon, and calls into question the role grammatical Number plays in Differential Object Marking. 
\u27and-a-half\u27 Numeral constructions in Hindi
Full text linkComplex numerals combine via addition and multiplication in the syntax from a sequence of simplex numerals. In this paper, we investigate a novel class of numerals labeled ‘and-a-half’ numerals which can combine with simplex numerals via addition resulting in a simplex numeral. But across languages, the presence of ‘and-a-half’ limits the formation of complex numerals to only via multiplication. Further addition of another numeral in this structure considerably degrades the construction. This paper focuses on Hindi data and seeks to explain this restriction placed by ‘and-a-half’ by investigating its pragmatic role in setting standards of precision. The analysis presented here predicts that the planning component in communicating standards of precision is encoded at the phrasal level where once you set a low standard of precision you cannot arbitrarily raise it - which is exactly what happens when an additive component is introduced in the structure
Acquisition of Hindi\u27s laryngeal contrast by Meiteilon speakers
Full text linkThough Meeteilon does not have phonemic contrast in voicing, native speakers can accurately recognize voiced stops and aspirated voiced stops in word initial positions and categorically distinguish these from voiceless stops and aspirated voiceless stops. However, they are unable to perceive any of these laryngeal contrasts in word-final position. We explain these facts by proposing that tone and aspiration being phonemic in Meeteilon, these cues from L1 can be re-recruited by native speakers for learning laryngeal contrasts in a second language like Hindi. Since these cues from L1 cannot be used to perceive laryngeal contrasts in word-final position, the contrasts are not perceived in these positions
Coordinated on the context: the many uses of Marathi =ts
Full text linkSeveral Indo-Aryan languages, including Bangla, Gujarati, Hindi, and Marathi contain a discourse clitic whose uses overlap with those of English particles like exclusives only/just, anaphoric indeed/that very, intensifiers really/totally, precisifiers right/exactly/absolutely, and scalar additive even without corresponding perfectly to any of them. This paper focuses on the Marathi variant =ts and offers a detailed empirical picture of a subset of its uses – uses involving discourse salience and noteworthiness or unexpectedness. I put forward the hypothesis that =ts conventionally signals that interlocutors are in mutual agreement that the proposition denoted by the prejacent is uniquely salient among alternatives in the current question. That is, =ts conveys that the proposition expressed by the prejacent offers a schelling point (or focal point) for the interlocutors to coordinate on