Syntactic Alignment

El índice del presente paper es el siguiente:

1. La familia lingüística chon
2. La lengua selknam
3. Aspectos de la fonología del selknam
4. El género y la lengua selknam
    4.1. Generalidades sobre el género
    4.2. El género en selknam
            4.2.1. Índices enclíticos de género
            4.2.2. Prefijos de animacidad
                        4.2.2.1. Prefijos de animacidad en verbos
                        4.2.2.2. Prefijos de animacidad en posposiciones
                        4.2.2.3. Prefijos de animacidad en interrogativos
            4.2.3. Sufijos de género
                      4.2.3.1. Sufijos de género en verbos
                      4.2.3.2. Sufijos de género en numerales
4.3. Género y morfología derivacional
4.4. Interacción entre género y alineamiento
      4.4.1. Los prefijos de animacidad y el alineamiento
      4.4.2. Los sufijos de género y el alineamiento
5. A manera de conclusión
6. Referencias bibliográficas.
7. Ejercicios

Although there is some literature on compound and complex verbs (Boyé 1999, Pokharel 1999, Lohani 2003), as well as on finite verbs (Genetti 1999), no work has been done on three-argument verbs and the ways they code arguments in Nepali. This paper is the first work which describes three-argument verb constructions in Nepali. The linguistic data for this paper was mostly collected following the Questionnaire on Ditransitive Constructions (Comrie et al. 2009) and the ditransitive paper (Malchukov et al. 2009). It describes and analyses the coding properties and different types of syntactic alignments. It also addresses the effect of animacy and definiteness in differential object marking in Nepali.1 Moreover, it also deals with the behavioural properties of three-argument constructions with regard to a number of syntactic constructions such as passivization, incorporation, constituent questions, reciprocalization, reflexivization, relativization and Quantifier float.

The topic of this PhD thesis is a classification and description of one-argument intransitive verbs in Georgian in a typological perspective. The analysis consists of two steps: firstly, the classification of the various types of verbs and secondly, a critical investigation of the complexity of the system under study and on the concept of intransitivity in general. The proposed classification begins by separating verbs not derived from other verbs (‘primary intransitives’) from those whose morphology indicates a process of de-transitive derivation (‘derived intransitives’). Primary one-argument verbs, in turn, are divided into two classes, each consisting of verbs with the same morphological, as well as semantic properties: telic verbs (such as “to die”) and atelic verbs (e.g. “to live”). Two-argument verbs are briefly described according to a transitivity scale. Discussed next are derived one-argument verbs which may have up to four readings: passive, decausative, autocausative and antipassive, and finally, verb pairs whose members, one transitive and one intransitive, are both morphologically marked. One-argument intransitivity in Georgian has not yet been investigated thoroughly; this particularly applies to some of its aspects such as the antipassive, of which a detailed study is offered here. The present work is intended to contribute to Kartvelian studies and at the same time, to recent discussion in general linguistics on the concepts of transitivity, valency and voice. Indeed, owing to its large number of verb classes and the close connection between morphosyntax and semantics, Georgian offers excellent data for the study of these concepts.

This paper focuses on the morphosyntax as well as the semantics of relativisation in Wobzi Khroskyabs, a Rgyalrongic language spoken in Sichuan, China. Different strategies of relativisation are presented, especially the nominalisation strategy. Wobzi Khroskyabs exhibits an innovative relativisation strategy with the genitive marker =ji, which is rarely found in other Rgyalrongic languages. Several hypotheses are put forward to account for the evolutionary pathway from genitivisation to relativisation, showing that genitive =ji probably followed an ergative pattern to enter the relativisation of core arguments.

Neste trabalho, revisitamos o sistema de referência de pessoa em três línguas da Família Tupi-Guarani, a saber: Apyãwa (tradicionalmente conhecida por Tapirapé), Tupinambá e Guaraní-Mbyá.  Partimos de descrições existentes das referidas línguas, bem como de trabalhos comparativos dessa Família linguística para analisar o comportamento morfossintático dos índices pessoais e o funcionamento da correferência gramaticalizada (nos termos de Wiesemann, 1986) em predicados verbais independentes e dependentes, nomes e posposições. Nosso objetivo é reunir as descrições do sistema referencial das três línguas, compará-las e reanalisá-las a partir de categorias tipológicas mais recentes. Inicialmente, buscamos averiguar a natureza argumental dos índices de pessoa e defender a dupla expressão do argumento nessas línguas, com base em Siewierska (2001), Dixon (2010), Schultze-Berndt (2011), Kibrik (2011) e Haspelmath (2013). De modo a facilitar a comparação entre os sistemas referencias compostos por esses índices, distinguimos três mecanismos que regem a marcação de pessoa nessas línguas: cisão intransitiva (Klimov, 1974), Hierarquia Referencial (Monserrat & Soares, 1983) e correferência (Wiesemann, 1986). O motivo por traz dessa tripartição está em evidenciar o comportamento da referência de pessoa de acordo com dois fatores: i) o tipo de construção; ii) os macro papéis sintático-semânticos que envolvem os argumentos. Assim, o sistema referencial de predicados monovalentes independentes é regido pela cisão intransitiva; o sistema dos predicados divalentes independentes é regido pela Hierarquia Referencial; e o sistema das construções dependentes verbais e não-verbais (nomes e posposições) é regido pela correferência. Dentre os três sistemas, o menos coeso é o correferencial, isso devido a alterações de alinhamento sintático sofridas pela marcação de pessoa em línguas Tupi-Guarani. Alguns estudos investigaram essas alterações e propuseram as seguintes hipóteses: i) um grupo de línguas Tupi-Guarani (subconjuntos I e VIII) estenderam o sistema referencial das construções independentes para as dependentes (Jensen, 1990); ii) a marcação de pessoa do pré-proto-Tupi-Guarani se alinhava a um padrão ergativo-absolutivo (Jensen, 1998); iii) quanto mais erodida for a ergatividade de uma língua, menos produtiva será a ocorrência dos prefixos correferenciais (Jensen, 1998); iv) a partir dos critérios propostos pela teoria da gramaticalização, reconstrói-se uma marcação de pessoa  nominativa-acusativa para o pré-proto-Tupi-Guarani (Gildea, 2002). Propomos um diálogo entre essas hipóteses de modo a identificar as correlações entre a referência de pessoa, o alinhamento sintático e a correferência. Por fim, motivados pela hipótese do “caminho da acusativização” proposto por Givón (2001), e atualizado por Queixalós (2004), buscamos identificar quais construções sofrem as mudanças de alinhamento primeiro, e qual o papel desempenhado pela correferência nesse processo.

In this work, we revisit the cross-referencing system in three Tupi-Guarani languages: Apyãwa (traditionally known as Tapirapé), Tupinambá and Mbyá-Guaraní. We started from descriptions of these languages, as well as comparative works of this linguistic Family, to analyze the morphosyntactic behavior of person indexes and the functioning of grammaticalized coreference in independent and dependent verbal predicates, names, and postpositions. Our object is to assemble past descriptions of the cross-referencing systems of all three languages and to reanalyze them by means of more recent typological categories. As person indexes are the basis to these systems, we begin by investigating their argument status and the possibility of arguments being doubly expressed in these languages ( Siewierska 2001; Dixon 2010; Schultze-Berndt 2011; Kibrik 2011; e Haspelmath 2013). Then, aiming a more straightforward comparison between the cross-referencing systems, we split them into three paradigms: split-S (Klimov, 1974), Person Hierarchy (Monserrat & Soares, 1983) and Correference (Wiesemann, 1986). This three-way system seeks to stress the correlation between cross-reference and two typological features i) the type of construction involved in person marking; ii) the macro syntactic-semantic roles of arguments. Thus: the independent monovalent constructions present a split-S system; the independent divalent constructions present a Person Hierarchy system; and the dependent/non-verbal constructions present a cross-referencing system based on coreference. Within theses three paradigms, the coreferential one is the less coherent and more heterogeneous. This behavior is assigned to cross-referencing and syntactic alignment changes on Tupi-Guarani languages. Four works looked into these changes and proposed the following hypotheses: i) some TG languages, from subgroup I and VIII, extended their cross-referencing system of independent constructions to dependent constructions (Jensen, 1990); ii) the pre-proto-Tupí-Guaraní cross-referencing system is reconstructed to be ergative-absolutive (Jensen, 1998); iii) the less ergative a Tupi-Guarani language is, the less productive will be its coreferential marking system (Jensen, 1998); iv) based on grammaticalization theory and its assumptions, the cross-referencing system reconstructed on pre-proto-Tupí-Guaraní turns out to be nominative-accusative (Gildea, 2002). We discuss these assumptions intending to single out correlations between cross-reference, syntactic alignment and coreference. To achieve this, we add the “accusativization path” hypothesis to the equation as we try to identify which construction changed its cross-referencing pattern first, as well as the coreference role on this process.

La familia lingüística chon incluía al menos cuatro lenguas: tehuelche y teushen de la Patagonia continental, selknam y haush de la Isla Grande de Tierra del Fuego. Solamente el tehuelche cuenta todavía con un último hablante de avanzada edad, mientras que del selknam (cuyos últimos hablantes fluidos habrían muerto en las décadas de 1970-1980) quedan, según Rojas Berscia (2104), unos pocos recordadores parciales.
Se dispone de descripciones gramaticales producidas por lingüistas profesionales de solo dos de las lenguas chon, el tehuelche (Fernández Garay 1998) y el selknam (Najlis 1973, Rojas Berscia 2014). Lamentablemente cuando las lenguas fueron registradas por estos lingüistas, ya estaban en períodos de obsolescencia más o menos avanzado. Hay vocabularios de lenguas chon desde el siglo XVI, e intentos de descripción gramatical de misioneros desde el siglo XIX.
El propósito del presente trabajo es mostrar que tanto en tehuelche como en selknam existen dos tipos de alineamiento que se superponen, uno nominativo-acusativo y otro ergativo-absolutivo.

When we read or listen to language, we are faced with the challenge of inferring intended messages from noisy input. This challenge is exacerbated by considerable variability between and within speakers. Focusing on syntactic processing (parsing), we test the hypothesis that language comprehenders rapidly adapt to the syntactic statistics of novel linguistic environments (e.g., speakers or genres). Two self-paced reading experiments investigate changes in readers’ syntactic expectations based on repeated exposure to sentences with temporary syntactic ambiguities (so-called “garden path sentences”). These sentences typically lead to a clear expectation violation signature when the temporary ambiguity is resolved to an a priori less expected structure (e.g., based on the statistics of the lexical context). We find that comprehenders rapidly adapt their syntactic expectations to converge towards the local statistics of novel environments. Specifically, repeated exposure to a priori unexpected structures can reduce, and even completely undo, their processing disadvantage (Experiment 1). The opposite is also observed: a priori expected structures become less expected (even eliciting garden paths) in environments where they are hardly ever observed (Experiment 2). Our findings suggest that, when changes in syntactic statistics are to be expected (e.g., when entering a novel environment), comprehenders can rapidly adapt their expectations, thereby overcoming the processing disadvantage that mistaken expectations would otherwise cause. Our findings take a step towards unifying insights from research in expectation-based models of language processing, syntactic priming, and statistical learning.