(in English): Typology of verbal suppletion on the ground of Indo-European languages is the focus of the thesis Verbal suppletion in Indo-European languages. The diachronic approach is of special interest. Description of verbal suppletion in the oldest stages of the IE languages and reconstruction of their Proto-Indo-European forms is thus aimed for (for each of the ten well-described IE language branches, one sample language is chosen, and its verbal system and suppletive verbal paradigms are described). Besides the descriptive data from individual language branches, the outcome of the thesis comprises the evaluation of the data by typological means from diachronic and synchronic view-point and its assessment in relation to frequency, semantic fields and Semantic relevance hierarchy. The thesis delivers a review of Proto-Indo-European roots in suppletive paradigms and their mutual relationships.

Ce travail se propose d’étudier la manière dont l’écrivain goannais de langue portugaise Francisco Luís Gomes (1829-1869) aborde dans son roman à thèse Les Brahmanes (1866) le christianisme, le système hindou des castes et le colonialisme. Libéral et catholique convaincu, Gomes fut très marqué par les idéaux de la Révolution française et aurait pu être inscrit dans le courant du romantisme « réformateur » dont faisaient partie Lamartine, Lamennais et Hugo. Inspiré par un fait historique (la révolte des Cipayes de 1857, le premier mouvement insurrectionnel indien d’envergure contre la domination anglaise), Les Brahmanes pourrait être considéré comme étant non seulement le premier ouvrage de fiction à attaquer les castes de l’Inde mais aussi le premier roman de la littérature moderne à dénoncer les abus du colonialisme. Le discours de Gomes comporte néanmoins quelques problèmes, car l’auteur finit par légitimer dans son livre le fait colonial. L’action des Brahmanes ne se déroule pas en Inde portugaise, d’où était l’écrivain, mais en Inde britannique, la véritable cible du roman. Or, celui-ci contribue aussi bien au mythe selon lequel le Portugal aurait agi, en tant que conquérant de peuples, d’une manière « douce » et égalitaire qu’au « style occidental de domination, de restructuration et d’autorité sur l’Orient » appelé « orientalisme » (Edward Saïd). / [?] / Este trabalho propõe-se a estudar a maneira como o escritor goês de língua portuguesa Francisco Luís Gomes (1829-1869) aborda em seu romance de tese Os Brahamanes (1866) o cristianismo, o sistema de castas hindu e o colonialismo. Liberal e católico convicto, Gomes foi bastante marcado pelos ideais da Revolução Francesa e teria podido ser inscrito na corrente do romantismo “reformador”, da qual faziam parte Lamartine, Lamennais e Hugo. Inspirado em um fato histórico (a Revolta dos Cipaios de 1857, primeiro movimento inssurrecional indiano de envergadura contra a dominação inglesa), Os Brahamanes poderia ser considerado como sendo não apenas a primeira obra de ficção a atacar as castas da Índia mas também o primeiro romance da literatura moderna a denunciar os abusos do colonialismo. O discurso de Gomes comporta entretanto alguns problemas, pois o autor acaba por legitimar em seu livro o fato colonial. A ação de Os Brahamanes não se passa na Índia portuguesa, de onde era o escritor, mas na Índia britânica, o verdadeiro alvo do romance. Ora, este contribui tanto para o mito segundo o qual Portugal teria agido, enquanto conquistador de povos, de uma maneira “doce” e igualitária, quanto para o “estilo ocidental de dominação, de reestruturação e de autoridade sobre o Oriente” chamado “orientalismo” (Edward Saïd).

This is a study of processes of structural and functional diversification of the uses of three cognate verbs across the Indo-Iranian language family: “do/make”, “be/become”, and “give”. First, this study identifies over sixty distinct construction types in which these verbs are used, including complex predicate constructions, nominal predication constructions, serial verb constructions, and several distinct auxiliary constructions. Since the sets of verbs studied here are cognates, and share a common source, crosslinguistic differences in their uses are the result of grammatical change, and especially shared and parallel innovations of similar uses. Then, this study presents a taxonomy of different complex predication types with “do/make”, and shows that there are general patterns in the deployment of different types of complex predication to express different types of situations. These patterns exhibit “transitivity prominence” previously identified by typologists with “heavy” or “lexical” verbs. This study then shows that these patterns are the result of several distinct pathways of grammatical change, often motivated by analogy to existing constructions, giving raise to different types of N-V complex predication constructions. Then, this study shows that despite the fact that Indo-Iranian speakers can potentially deploy distinct constructions to encode each of the six nominal predication functions, sets of such functions are often co-expressed by the same structural coding means, especially clauses with cognate “be/become” verbs. This study uses a novel method, based on bipartite network graphs, to compare of the degree to which nominal predication functions are co-expressed in different languages. Finally, this study shows that the three sets of cognate verbs are more likely to be used similarity within branches and subbranches of Indo-Iranian than across branches. The scope of this branches, however, is different for different verbs: “do/make” and “give” behave more similarly in languages which belong to the same major branch, Iranian or Indo-Aryan, but “be/become” clusters are at different levels of subbranching. This is the result of the different types of innovations attested with these verbs: reanalysis and actualization motivated by analogy with “do/make” and “give”, and metaphorical and metonymy extensions with “be/become”.

This thesis presents a study of Pakistani documentary, with a focus on the ways in which it represents Pakistan’s national identity and history. The study examines three sources of documentary production – state media, commercial television channels, and independent filmmakers – as three distinct voices of Pakistani documentary. The study argues that the discourses of these institutions are governed by their respective ideological, political, and economic priorities. These factors result in two competing approaches to Pakistan’s national history and identity: right-wing and left-wing. The Islamic ideology of the state governs the discourse of state-sponsored documentaries. The commercial television documentaries take an anti-establishment position, however, they remain faithful to Islamic ideology of the state to a large extend. The independent filmmakers, on the other hand, offer a liberal perspective of history and a secular identity of Pakistan. Hence, they offer a critical view of the state’s Islamic ideology as a governing principle of historiography and identity formation. The notion of representation entails the issues of authenticity, credibility, and truth-value, associated with the various methods adopted by the filmmakers. Hence, attention is paid to the styles and modes of documentary, with a reflection on the documentarian’s individual approaches to realism. The documentaries have been placed within historical and political contexts considering Pakistan as a postcolonial state, which also functions as a critical framework of this study.

The Proto-Indo-European causative/iterative suffix *-ei̯e- was inherited by Old Iranian and persists in almost all Middle and Modern Iranian languages as -aya- and -ēn- (-Vn-) respectively. Comparably, in the Indic branch -aya- functions as a causative suffix in Sanskrit beside another suffix -āpaya which became the productive causative suffix -āvē- in Middle Indic and still used in Modern Indic today. Evidence shows eight Eastern Iranian languages- †Khotanese, †Khwarazmian, Parachi, Wakhi, Munji, Pashto, Ormuri, and Yidgha- using the morphological causative suffix in addition to the expected Iranian one -aya- or -Vn-. This alternative causative suffix is reconstructible as *-au̯ai̯a- and its attested reflexes have the forms -VwV-, -Vv-, and -wV-. Moreover, in two dialects of the Northwestern Iranian language Gilaki, Dakhili and Langaroudi, the causative suffix is not -Vn- but is rather -be(ː)- in the present tense. In this study I examine the synchronic function of the Gilaki causative suffix -be(ː)- as well as its diachronic origins. I show that Gilaki -be(ː)- primarily functions as a causative suffix and that it is a form which cannot be explained as an innovation within Gilaki itself through phonological or analogical change. As a matter of fact, I demonstrate that this suffix is better explained as deriving from PIr.*-au̯ai̯a- and is connected to the aforementioned Eastern Iranian suffixes. I also argue the reason for realization of /p/ and */u̯/ in -āpaya and *-au̯ai̯a- is phonological and probably goes back to some stages of PIIr.

This study outlines a methodological framework for reconstructing linguistic history within a dialect continuum and applies this methodology to an under-described, controversial, and complex subgroup of New Indo-Aryan (NIA)—the Kamta, Rajbanshi and Northern Deshi Bangla lects (KRNB). ¶ Dialect continua are characterised by non-discrete boundaries between speech communities, and as a result previously divergent lects may undergo common innovations; the result is the familiar picture of overlapping dialectological isoglosses. The sequencing of these innovations and the historical relations between the lects involved are often highly ambiguous. Given the right sociohistorical conditions, a widespread innovation may be more recent than a localised innovation—the very opposite sequencing to that implied by the splits in a family tree. ¶ Not surprisingly, discrete application to the NIA continuum of traditional methodologies—including the Comparative Method, etymological reconstruction and dialect geography—has yielded unsatisfactory and at times chronologically distorted results. Historical studies, therefore, have chosen between: (a) only studying the histories of NIA lects with written records; (b) reconstructing using the chronology suggested by the shape of a family tree; or (c) settling for a ‘flat’, non-historical account of dialect geography. ¶ Under the approach developed here, the strengths of each of these traditional methods are synthesised within an overarching framework provided by a sociohistorical theory of language change. This synthesis enables the linguistic history of the KRNB lects to be reconstructed with some detail from the proto-Kamta stage (1250-1550 AD) up to the present day. Innovations are sequenced based on three types of criteria: linguistic, textual and sociohistorical. The old Kamta stage, and its relation to old Bangla and Asamiya, is reconstructed based on linguistic Propagation Events and Speech Community Events—two concepts central to the methodology. The old Kamta speech community and its language became divided into western, central and eastern subsections during the middle KRNB period (1550-1787 AD, dates assigned by attested sociohistorical events). During the same period, KRNB lects also underwent partial reintegration with NIA lects further afield by means of more widely propagated changes. This trend of differentiation at a local level, concurrent with reintegration at a wider level, also characterises the modern KRNB period from 1787 AD to the present. ¶ This account of KRNB linguistic history is based on a rigorous reconstruction of changes in phonology and morphology. The result is not only a reconstruction of historical changes, but of the proto-Kamta phoneme inventory, hundreds of words of vocabulary, and specific areas of nominal and verbal morphology. The reconstruction is based on data collected in the field for the purposes of this study. Phonological reconstruction has made use of the WordCorr software program, and the reconstructed vocabulary is presented in a comparative wordlist in an appendix. ¶ The methodology developed and applied in this study has been found highly successful; though naturally not without its own limitations. This study has significance for its contribution both to the methodology of historical linguistic reconstruction and to the light shed on the linguistic prehistory of KRNB.

This dissertation explores the nature and extent of retroflex consonant harmony in South Asia. Using statistics calculated over lexical databases from a broad sample of languages, the study demonstrates that retroflex consonant harmony is an areal trait affecting most languages in the northern half of the South Asian subcontinent, including languages from at least three of the four major families in the region: Dravidian, Indo-Aryan and Munda (but not Tibeto-Burman). Dravidian and Indo-Aryan languages in the southern half of the subcontinent do not exhibit retroflex consonant harmony. In South Asia, retroflex consonant harmony is manifested primarily as a static co-occurrence restriction on coronal consonants in roots/words. Historical-comparative evidence reveals that this pattern is the result of retroflex assimilation that is non-local, regressive and conditioned by the similarity of interacting segments. These typological properties stand in contrast to those of other retroflex assimilation patterns, which are local, primarily progressive, and not conditioned by similarity. This is argued to support the hypothesis that local feature spreading and long-distance feature agreement constitute two independent mechanisms of assimilation, each with its own set of typological properties, and that retroflex consonant harmony is the product of agreement, not spreading. Building on this hypothesis, the study offers a formal account of retroflex consonant harmony within the Agreement by Correspondence (ABC) model of Rose & Walker (2004) and Hansson (2001; 2010). Two Indo-Aryan languages, Kalasha and Indus Kohistani, figure prominently throughout the dissertation. These languages exhibit similarity effects that have not been clearly observed in other retroflex consonant harmony systems; retroflexion is contrastive in both non-sibilant (i.e., plosive) and sibilant obstruents (i.e., affricates and fricatives), but harmony applies only within each manner class, not between them. At the same time, harmony is not sensitive to laryngeal features. Theoretical implications of these and other similarity effects are discussed.

This dissertation explores the nature and extent of retroflex consonant harmony in South Asia. Using statistics calculated over lexical databases from a broad sample of languages, the study demonstrates that retroflex consonant harmony is an areal trait affecting most languages in the northern half of the South Asian subcontinent, including languages from at least three of the four major families in the region: Dravidian, Indo-Aryan and Munda (but not Tibeto-Burman). Dravidian and Indo-Aryan languages in the southern half of the subcontinent do not exhibit retroflex consonant harmony. In South Asia, retroflex consonant harmony is manifested primarily as a static co-occurrence restriction on coronal consonants in roots/words. Historical-comparative evidence reveals that this pattern is the result of retroflex assimilation that is non-local, regressive and conditioned by the similarity of interacting segments. These typological properties stand in contrast to those of other retroflex assimilation patterns, which are local, primarily progressive, and not conditioned by similarity. This is argued to support the hypothesis that local feature spreading and long-distance feature agreement constitute two independent mechanisms of assimilation, each with its own set of typological properties, and that retroflex consonant harmony is the product of agreement, not spreading. Building on this hypothesis, the study offers a formal account of retroflex consonant harmony within the Agreement by Correspondence (ABC) model of Rose & Walker (2004) and Hansson (2001; 2010). Two Indo-Aryan languages, Kalasha and Indus Kohistani, figure prominently throughout the dissertation. These languages exhibit similarity effects that have not been clearly observed in other retroflex consonant harmony systems; retroflexion is contrastive in both non-sibilant (i.e., plosive) and sibilant obstruents (i.e., affricates and fricatives), but harmony applies only within each manner class, not between them. At the same time, harmony is not sensitive to laryngeal features. Theoretical implications of these and other similarity effects are discussed.

A thesis presented on the political history of Fiji from cession to Britain in 1874 compares and analyses the country’s four political coups. A military coup occurred in 1987 by Lt. Col Sitiveni Rabuka. Six months later he staged a self-coup. In 2000 George Speight staged an armed civilian coup or putsch, and in 2006 Commodore Frank Bainimarama, head of Fiji’s military forces, overthrew the government of Laisenia Qarase. This paper is an internal comparison of the four coups of which the aim is to examine why coups occur in Fiji. The conclusion is that the level of influence of the country’s traditional paramount chiefs is a strong causal factor in events leading to the political overthrows. Issues such as ethnicity, constitutionalism, democracy, traditionalism, and modernity make the study of the Fiji coups complex. All of the major actors involved have been present or have been somehow linked to each coup. Questions of leadership arise as do issues regarding pluralism and multiculturalism. These issues are discussed in this paper. The end result is that if the question of traditional leadership is not addressed within a democratic framework then Fiji will continue to have coups.

The broad objective of this dissertation is an interdisciplinary study uniting historical linguistics, classical philology, and comparative poetics in an attempt to investigate archaic Greek poetic texts from a diachronic perspective. This thesis consists of two parts. The first part, “Etymology and Poetics”, is devoted to several cases where scantiness of attestation and lack of semantic information render traditional philological methods of textual interpretation insufficient. In such cases, the meaning of a word has to be arrived at through linguistic analysis and verified through appeal to related poetic traditions, such as that of Indo-Iranian. Chapter 1 proposes a new interpretation for the enigmatic word ἀάατο̋, the Homeric epithet of the waters of the Styx, which is shown to have meant ‘sunless’. Chapter 2 deals with the word ἀριδείκετο̋, argued to mean ‘famous’: this solution finds support in the use of the root *dei̯k- in the poetic expression “to show forth praise”, found in Greek choral lyric and the Rigveda. Chapter 3 investigates the history of the verbs ἰάπτω ‘to harm’ and ἰάπτω ‘to send forth (to Hades)’. Chapter 4 improves the text of Pindar (O. 6.54), restoring a form ἀπειράτωι. Chapter 5 discusses the difficult word ἀμαυρό̋, establishing for it a meaning ‘weak’ and proposing a new etymology. Finally, Chapter 6 places Alc. 34 in the context of comparative mythology, with the object of reconstructing the history of the Lesbian lyric tradition. The second part, “Grammar of Poetry”, shifts the focus of the inquiry from comparative poetics to the language of early Greek poetry and its use. Chapter 7 addresses the problematic Homeric aorist infinitives in -έειν, showing how these artificial forms were created by allomorphic remodeling driven by metrical necessity; the problem is placed in the wider context of the debate about the transmission and development of Homeric epic diction. The metrical and linguistic facts relating to the distribution of infinitives are further discussed in Chapter 8, where it is argued that the unexpected Aeolic form νηφέμεν in Archil. 4 should be viewed as an intentional allusion to the epic tradition, specifically, the famous midsummer picnic scene in Hesiod. / Linguistics