xiii, 145 p. : map. A print copy of this thesis is available through the UO Libraries. Search the library catalog for the location and call number. / Nominalization and its various functions is a topic of considerable current interest in Tibeto-Burman (TB) studies and has both typological and historical implications. This thesis documents and discusses data of nominalizing velar prefixes in the different branches of the TB language family. Based on the reconstruction ofa Proto-TibetoBurman (PTB) 'adjectival prefix' *gV- suggested by TB wide-scale comparativists such as Wolfenden, Shafer, Benedict, and Matisoff, this study incorporates extensive data on velar prefixes covering other functions that are, just like deriving adjectival modifiers, typically associated with nominalization in TB. The various pieces of evidence thus suggest that the existence of a PTB *gV- nominalizer is the best explanation for the distribution of forms and functions of the respective prefixes in TB languages. / Committee in Charge: Dr. Scott DeLancey, Chair; Dr. Spike Gildea

This dissertation uses syntactic, semantic and morphological evidence from English nominalization to probe the interaction of event-structure and syntax, develop a typology of structural complexity within nominalization, and test hypotheses about the strict ordering of functional items. I focus on the widely assumed typology of nominalization found in Chomsky (1970). In particular, I show that derived nominals are structurally more complex than nominal gerunds; this has long been assumed to be the opposite. I provide a structural and morphological account of these forms of nominalization. In doing so, I explore a number of disparate topics such as: the importance of syncretism in apparently unrelated morphological elements for theories like Distributed Morphology; the role of prepositions in allowing or preventing binding relations and NPI-licensing, the exact nature of root-object union that allows idiomatic interpretations; the morphological reflexes of Case in the nominal system; the syntactic structure of verb particle constructions; the nature of events in nominalization; and the role syntactic operations play in determining morphological regularity. The dissertation also explores the nature of the English verb particle construction, arguing that it has (at least) three distinct structural configurations. Using these three distinct structures I am able to explain a number of distinct behaviors from predicate-object relationships, particle modification and argument loss in particle construction. I also discuss the relationship between particles (and results) and the different forms of nominalization. In particular, I show that apparent co-occurrence restrictions between nominal types and particles are not due to event-structure or other semantic restrictions. Rather, these differences are tied solely to the particular, idiosyncratic morphological properties of the constructions. The dissertation shows that certain functional projections may only appear once with a given root, but that there is some freedom of ordering of projections relative to the root in some cases. This work provides a window into the interaction between syntax and event structure as well as the nature of ordering within functional projections.

This thesis presents an analysis of participles in MA which show verbal and nominal features but are not nouns or verbs. Participles pattern with verbs, combine with adverbs and take objects. Like nouns, they partially agree with their subjects, are negated with mu or inflected ma and cannot appear in VSO order nor do they allow subject-drop. I propose that without the functional projection vP, bare VPs are not fully verbal. When participles occur in a finite present-tense sentence, they act like non-verbal predicates and the resulting copula construction conforms to Benmamoun’s (2008) framework of verbless sentences in Arabic. The existence of VP explains the verbal properties, and the absence of vP explains the nominal ones. The lack of vP explains lacking full agreement and using non-verbal negative particles with participles. Viewing participles as bare VPs is consistent with Croft’s (1991) de-verbalizing hierarchy where verb types range from being fully finite to completely nominalized forms.

Au cours du premier millénaire avant notre ère, les bibliothèques, qui apparaissent avec le besoin d'organiser la conservation des textes, sont immédiatement confrontées aux difficultés de l'indexation. Le titre apparaît alors comme une première solution, permettant d'identifier rapidement chaque type d'ouvrage et éventuellement de discerner des ouvrages thématiquement proches.Alors que dans la Grèce Antique, les titres ont une fonction peu informative, mais ont toujours pour objectif d'identifier le document, l'invention de l'imprimerie à caractères mobiles (Gutenberg, XVème siècle) a entraîné une forte augmentation du nombre de documents, offrant désormais une diffusion à grande échelle. Avec la recrudescence des textes imprimés, le titre acquiert peu à peu de nouvelles fonctions, conduisant très souvent à des enjeux d'influence socioculturelle ou politique (notamment dans le cas des articles journalistiques).Aujourd'hui, que le document soit sous forme électronique ou papier, la présence d'un ou de plusieurs titres est très souvent constatée, permettant de créer un premier lien entre le lecteur et le sujet abordé dans le document. Mais comment quelques mots peuvent-ils avoir une si grande influence ? Quelles fonctions les titres doivent-ils remplir en ce début du XXIème siècle ? Comment générer automatiquement des titres respectant ces fonctions ?Le titrage automatique de documents textuels est avant tout un des domaines clés de l'accessibilité des pages Web (standards W3C) tel que défini par la norme proposée par les associations sur le handicap. Côté lecteur, l'objectif est d'augmenter la lisibilité des pages obtenues à partir d'une recherche sur mot-clé(s) et dont la pertinence est souvent faible, décourageant les lecteurs devant fournir de grands efforts cognitifs. Côté producteur de site Web, l'objectif est d'améliorer l'indexation des pages pour une recherche plus pertinente. D'autres intérêts motivent cette étude (titrage de pages Web commerciales, titrage pour la génération automatique de sommaires, titrage pour fournir des éléments d'appui pour la tâche de résumé automatique,).Afin de traiter à grande échelle le titrage automatique de documents textuels, nous employons dans cette étude des méthodes et systèmes de TALN (Traitement Automatique du Langage Naturel). Alors que de nombreux travaux ont été publiés à propos de l'indexation et du résumé automatique, le titrage automatique demeurait jusqu'alors discret et connaissait quelques difficultés quant à son positionnement dans le domaine du TALN. Nous soutenons dans cette étude que le titrage automatique doit pourtant être considéré comme une tâche à part entière.Après avoir défini les problématiques liées au titrage automatique, et après avoir positionné cette tâche parmi les tâches déjà existantes, nous proposons une série de méthodes permettant de produire des titres syntaxiquement corrects selon plusieurs objectifs. En particulier, nous nous intéressons à la production de titres informatifs, et, pour la première fois dans l'histoire du titrage automatique, de titres accrocheurs. Notre système TIT', constitué de trois méthodes (POSTIT, NOMIT et CATIT), permet de produire des ensembles de titres informatifs dans 81% des cas et accrocheurs dans 78% des cas. / During the first millennium BC, the already existing libraries needed to organize texts preservation, and were thus immediately confronted with the difficulties of indexation. The use of a title occurred then as a first solution, enabling a quick indentification of every work, and in most of the cases, helping to discern works thematically close to a given one. While in Ancient Greece, titles have had a little informative function, although still performing an indentification function, the invention of the printing office with mobile characters (Gutenberg, XVth century AD) dramatically increased the number of documents, which are today spread on a large-scale. The title acquired little by little new functions, leaning very often to sociocultural or political influence (in particular in journalistic articles).Today, for both electronic and paper documents, the presence of one or several titles is very often noticed. It helps creating a first link between the reader and the subject of the document. But how some words can have a so big influence? What functions do the titles have to perform at this beginning of the XXIth century? How can one automatically generate titles respecting these functions? The automatic titling of textual documents is one of the key domains of Web pages accessibility (W3C standards) such as defined in a standard given by associations about the disabled. For a given reader, the goal is to increase the readability of pages obtained from a search, since usual searches are often disheartening readers who must supply big cognitive efforts. For a Website designer, the aim is to improve the indexation of pages for a more relevant search. Other interests motivate this study (titling of commercial Web pages, titling in order to automatically generate contents, titling to bring elements to enhance automatic summarization).In this study, we use NLP (Natural Language Processing) methods and systems. While numerous works were published about indexation and automatic summarization, automatic titling remained discreet and knew some difficulties as for its positioning in NLP. We support in this study that the automatic titling must be nevertheless considered as a full task.Having defined problems connected to automatic titling, and having positioned this task among the already existing tasks, we provide a series of methods enabling syntactically correct titles production, according to several objectives. In particular, we are interested in the generation of informative titles, and, for the first time in the history of automatic titling, we introduce the concept of catchiness.Our TIT' system consists of three methods (POSTIT, NOMIT, and CATIT), that enables to produce sets of informative titles in 81% of the cases and catchy titles in 78% of the cases.

The aim of this paper is to analyse the ways in which a Swedish translation of an English text intended as a manual for health practitioners needs to be adapted in order to function as an informative text aimed at patients, their families and other laymen interested in the subject. Focus lies on adaption of terminology, style, and target reader perspective. For this paper, selected parts of an English source text were translated into Swedish, and different adaptions prompted by the change in genre, intended target reader and purpose of the text were noticed and subsequently analysed. In the analysis, it became clear that translation of terminology can be treated through addition, omission or generalization where subject specific terms are not relevant for the intended readers. Under the subject of style, formality level defined through the usage of nominalizations and passives was analysed, and here, modulation and transposition proved useful for adapting the formality level to better fit the purpose of the text. Under the category of target reader perspective, omission, generalization and modulation of different kinds were translation strategies used to turn the focus from the intended ST reader to the intended TT reader. Defining or re-defining the subject of different actions was also necessary to shift the perspective from one intended group of readers to another. Finally, the analysis showed that different kinds of adaption may be purposeful in different situations.

Nominalization as a matter of great concern of linguistic research has been analyzed widely over the last forty years from different theoretical viewpoints with fluctuating aims and formalisms. The process of nominalization refers to the use of a verb or an adjective into a noun, with or without morphological transformation, so that the word can act as the head of a noun phrase, e.g. develop→ development, close → closure, write → writing, etc. In present-day English, -er, -or and -ee nominalizations show a wide scope of semantic, morphological and syntactic features. Thus the aim of the present study is to analyze the derivational, transformational, and semantic peculiarities of -er, -or and -ee nominalizations. Like many corpus investigations, this study deals with both qualitative and quantitative methods. The scope of the present study is 1 225 nominalizations selected from “Oxford Advanced Learner’s Dictionary” by A. S. Hornby [CD-ROM] (2005). The content of the communication is a situation, which does not refer directly to an extralinguistic reality which exists in the real world, but rather to the speaker’s conceptualization of it. The constituents of this conceptualization of reality are semantic functions: processes, participants and circumstances. The pivot of any situation is the process which is expressed by a verb. The following types of processes are distinguished: material, behavioural, mental, verbal, relational, and existential. When the process is... [to full text] / Nominalizacija yra plačiai analizuojama per pastaruosius keturiasdešimt metų skirtingais teoriniais požiūriais. Nominalizacijos procesas remiasi veiksmažodžių arba būdvardžių keitimu daiktavardžiais, su arba be morfologinėmis transformacijomis, tad žodis gali veikti kaip daiktavardžio frazė, pvz. plėsti → plėtra, uždaryti → uždarymas, rašyti → rašymas ir t.t. Dabartinėje anglų kalboje ,-er,-or ir -ee nominalizacijos turi plačias semantines, sintaksines ir morfologines savybes. Taigi šio darbo tikslas yra išanalizuoti derivacinius, transformacinius ir semantinius -er,-or ir -ee nominalizacijų ypatumus. Šiame tyrime nagrinėjama tiek kokybiniais, tiek kiekybiniais metodais. Šio darbo apimtis yra 1 225 nominalizacijos surinktos iš "Oxford Advanced Learner's Dictionary" A.S. Hornby [CD-ROM] (2005). Darbe nagrinėjamos semantinės funkcijos: procesai, dalyviai ir aplinkybės. Būtų galima daryti išvadą, kad -er,-or ir -ee nominalizacijos kildinamos iš procesų.

A meta da presente tese é analisar as orações não-finitas em Karitiana. Esta é uma língua Ameríndia falada atualmente por 400 pessoas que habitam a Terra Indígena Karitiana. A reserva Karitiana está localizada a 95 quilômetros ao sul da área urbana de Porto Velho, Rondônia, Brasil. A língua é a única representante do ramo Arikém, família Tupi. A presente pesquisa investiga as diferenças entre as orações subordinadas que funcionam como orações subordinadas adverbiais, relativas e completivas, bem como a nominalização oracional e as orações infinitivas marcadas pelos sufixos e , respectivamente. Apesar de a língua não apresentar morfologia de tempo, modo ou concordância em subordinadas, ela possui várias outras características, tais como núcleos aspectuais, voz, variação na ordem de palavras (SOV ou OSV), morfologia de foco do objeto, o que sugere que se trata subordinação e não nominalização (Storto 1999, Storto 2012, Vivanco 2014). Mostramos ainda que as orações encaixadas Karitiana podem ser modificadas por advérbios, negação e evidenciais que estão associados a orações ou sintagmas verbais. Esta pesquisa levantou uma discussão que tem sido amplamente abordada na literatura sobre as orações subordinadas nãofinitas em línguas Ameríndias. A literatura sobre essas orações não-finitas mostra que muitos autores que assumem que elas são nominalizações utilizam dois argumentos: (1) a falta de traços finitos e (2) o fato de algumas línguas exibirem marcas de caso. Os argumentos com base no item (2) supramencionado não parece muito convincentes para assumirmos uma análise de nominalização em Karitiana, visto que orações subordinadas finitas em várias línguas podem ser usadas como complemento verbal e receber marcas de caso. A nossa análise apresentada nesta tese assume que a falta de traços finitos não significa necessariamente que estas orações sejam nominalizadas porque elas exibem em suas estruturas várias características de orações ativas, tais como núcleos funcionais de voz, aspecto, advérbios, negação e evidenciais. Tipologicamente, esses núcleos funcionais estão associados a orações ou a sintagmas verbais. E, internamente à língua, eles estão também relacionados a orações ou a sintagmas verbais tanto em ambientes finitos quanto em ambientes não-finitas. / The main aim of this dissertation is to analyze non-finite clauses in Karitiana. Karitiana is an endangered Amerindian language spoken today by approximately 400 people who live in a reservation located 59 miles south of the urban area of Porto Velho, the capital of the state of Rondônia, Brazil (Amazonian region). The language is the unique representative of the Arikém branch, one of the ten linguistic groupings identified inside the Tupian family. The current research investigates differences among embedded clauses that function as adverbial, relative, and complement clauses in Karitiana, as well as nominalization and infinitival embedding marked respectively by the suffixes and . Even though subordinate clauses in Karitiana do not display any finite morphology of agreement, tense, or mood, it is true that the language shows many other functional heads such as morphemes of causativization, passivization, and object focus construction, as well as aspectual nuclei, and word order variation (SOV, OSV), suggesting that they are clauses and not nominalizations (Storto 1999; Vivanco 2014). Furthemore, we show that Karitiana embedded clauses can be modified by adverbs, negation and evidentials that are associated with clauses or verbal phrases. The literature on non-finite clauses in Amerindian languages shows that many specialists in these languages have claimed that these clauses are nominalized based on two arguments: (1) lack of finite traits and (2) the fact that some of these languages display case-marking. The argument that the presence of case-marking in subordination characterizes them as nominalizations seems to be unconvincing because finite subordinate clauses in several languages can be used as verbal complement and can be marked with case. In our analysis, the lack of finite features also does not necessarily mean that these clauses are nominalizations, since Karitiana subordinate clauses exhibit other properties of active clauses such as functional heads: voice, aspect, adverbs, negation, and evidentials. Typologically, these functional heads are associated with clauses and, internally to the language, they are also correlated with clauses and verbal phrases, functioning either in matrix clauses or in subordinate clauses.

Tendo em vista que a análise de unidades lexicais neológicas tem muito a contribuir para os estudos de formação de palavras, procuramos, neste trabalho, com base nos modelos teóricos de Bybee (1988) e Langacker (1987, 1991) descrever um aspecto da formação de palavras no português brasileiro: os chamados nominais (substantivos abstratos derivados de verbos). Abordamos as tendências e restrições de uso de cada sufixo nominalizador (enfatizando -ção, -mento, -agem, -da e -nc(i)a) bem como da chamada derivação regressiva. Também descrevemos certas características semânticas (polissemia, especialização semântica etc.), sintáticas (exigências sintáticas, uso de verbos-suporte etc.) e discursivas (emprego em textos mais ou menos formais, mecanismos de coesão textual etc.) desse tipo de formação. Em seguida, buscamos mostrar como os sufixos nominalizadores e os nominais são descritos nos dicionários de língua do português brasileiro. A partir de uma base de dados de formações neológicas coletadas em textos jornalísticos entre os anos de 1993 e 2000, analisamos cerca de 170 nominais neológicos, com a finalidade de verificar se tais formações obedecem às tendências e restrições de uso descritas, e se é possível detectar tendências novas. Observamos que: o sufixo -ção, o mais freqüentemente empregado, tem se unido a verbos da segunda conjugação, fenômeno não-atestado em períodos anteriores da língua; os sufixos -ção e -agem vêm apresentando conotações familiares ou jocosas, com destaque para o chamado -ção iterativo; as formas -nça e -ncia vêm adquirindo conotações divergentes, o que pode apontar para que não mais sejam analisadas como formas alomórficas do mesmo sufixo; entre outras observações. Por fim, propusemos uma forma de descrever esses nominais neológicos, a ser utilizada no Dicionário de Neologismos do Português Brasileiro Contemporâneo (Década de 90). / Bearing in mind that the analysis of neological lexical units has much to contribute to word-formation sdudies, we attempt, in the present work, based on the theoretical models of Bybee (1988) and Langacker (1987, 1991), to describe an aspect of Brazilian Portuguese word-formation: the so-called nominals (abstract nouns derived from verbs). We address the usage tendencies and restrictions of each nominalizing suffix (with emphasis on -ção, -mento, -agem, -da and -nc(i)a), as well as the so-called “regressive derivation". We also describe certain characteristics of this kind of word-formation, such as: semantic (polysemy, semantic specialization etc.), syntactic (valency, support-verbs etc.) and discursive (text formality, textual cohesion etc.). Then, we attempt to show how the nominalizing suffixes and the nominals are described in Brazilian Portuguese dictionaries. Based on a neologism database collected from journalistic texts published between 1993 and 2000, we analyze about 170 neological nominals, in order to verify if these formations follow the above described usage tendencies and restrictions, and if it is possible to detect new tendencies. We have observed, among other minor characteristics, that: the suffix -ção, the most frequent nominalizing suffix, can be combined with second-conjugation verbs, a phenomenon attested only very recently in the Portuguese language; the suffixes -ção and -agem have presented colloquial or playful connotations, especially the so-called iterative -ção; the forms -nça and -ncia are being used with diverging connotations, which can indicate that they should not be analyzed as allomorphic forms of the same suffix. Finally, we propose a way of describing these neological nominals, to be used in the Dicionário de Neologismos do Português Brasileiro Contemporâneo (Década de 90) (Dictionary of Neologisms of Contemporary (90’s) Brazilian Portuguese.

Nesta tese, avaliam-se as dificuldades de uso das nomonalizações em português segunda língua (L2) expressas por estrangeiros falantes do inglês (L1) e suas devidas implicações nas áreas de Morfologia, Sintaxe e Semântica. Nela apresentase a teoria da análise contrastiva (AC) e cria-se uma tabela de correspondências entre a interlíngua (I) e a língua portuguesa (L2), considerando os errors e mistakes devidos aos fenômenos da transferência e interferência através do método indutivo. Para tanto, tomaram-se por apoio os conceitos teóricos de transferência, interferência, erro e norma, entre outros. O trabalho apresenta-se dividido em duas partes: uma teórica, a análise contrastiva a priori, em que se estabelecem semelhanças e diferenças entre o inglês e o português: outra prática, a análise contrastiva a posteriori, em que se estudam as formas correspondentes da interlíngua (I) e da L2 detectando as razões das falhas e suas possíveis causas Comprovados os problemas das nominalizações, acrescentam-se, ao final da tese, algumas sugestões pedagógicas, com vistas ao ensino-aprendizagem de português para estrangeiros. / In this thesis, the difficulties related to the use of Portuguese nominalizations, expressed by English-speaking foreigners, are evalued considering the respective areas of Morphology, Syntax and Semantics. First a Contrastive Analysis (CA) theory is presented and afterwards a table of correspondences between Interlanguage (I) and Portuguese (L2 ) is created, taking into account the presence of mistakes and errors due to the transference and interference phenomena by means of an inductive process. Consequently it was based on the theoretical concepts of transference, interference and norm, among others. The work is divided into two parts: a theoretical one which is established by a contrastive analysis a priori discussing the similarities and main differences between English and Portuguese. Then, comes other part a practical one where the corresponding forms of the Interlanguage and L2) are studied and followed by the inferences of failures and probable reasons underlying them. After confirming the nominalization problems, some pedagogical suggestions were added to facilitate the Learning of Portuguese as a second language.

Este trabalho descreve diversos aspectos da formação de perguntas complexas, orações subordinadas e ordem de palavras no Karitiana, uma língua Tupi da família Arikém. Primeiramente, descrevemos a formação de perguntas bi-oracionais como perguntas indiretas e perguntas de longa-distância. No primeiro caso, argumentamos que perguntas indiretas strictu sensu não existem em Karitiana, pois pronomes interrogativos não são em geral permitidos dentro de orações subordinadas. Já perguntas de longa-distância são possíveis na língua, mas sua formação inclui uma estratégia conhecida como pied-piping de larga-escala, na qual a oração subordinada inteira contendo o pronome interrogativo é movida para a posição inicial da sentença. Argumentamos que estes dois fatos podem ser capturados se assumirmos uma estrutura para orações subordinadas (1) que não contém certos núcleos oracionais, como C; e (2) que é a projeção de um núcleo nominalizador n. Essas propriedades seriam capazes de explicar o comportamento ambíguo destas construções, que ora se comportam como orações, ora como sintagmas nominais. Além disso, essa proposta também explica outros fenômenos aparentemente não relacionadas, como a ausência de tough-constructions, a extração com verbos factivos, a cliticização em subordinadas e a presença de relativas de núcleo interno. Propomos ainda que este núcleo n de caráter nominal teria como uma possível realização fonética um sufixo -a, que pode ser detectado em diversos ambientes nominais da língua e que possivelmente teria conexões históricas com o nominalizador -a presente em outras línguas Tupi. Na segunda parte deste trabalho, discutimos ainda construções com o morfema de voz inversa ti-, reanalisadas aqui como configurações de redobro de clítico. Essa proposta é capaz de explicar diversas propriedades deste morfema, como sua emergência em diferentes construções, o padrão de concordância excêntrica disparado por ele, a distribuição complementar com a passiva e a leitura pressuposicional do tema nestes ambientes. Por fim, discutimos ainda dois fenômenos prosódicos - o sufixo -o e a intonação de relativas de objeto com ti- - que poderiam constituir argumentos adicionais para as análises desenvolvidas neste trabalho. / This dissertation investigates aspects of question formation, embedded clauses, and word order in Karitiana, a Tupi-Arikém language spoken in Brazil. Firstly, the formation of bi-clausal questions such as indirect questions and long-distance questions is described. Regarding indirect questions, our claim is that there are no indirect questions in Karitiana, as interrogative pronouns are usually disallowed inside embedded clauses. On the other hand, the language has long-distance questions, but they obligatorily involve a strategy known as clausal/large-scale pied-piping - i.e., fronting of the whole embedded clause containing the interrogative pronoun. These two facts can be explained if one assumes a structure for embedded clauses that does not contain certain clausal projections, such as CP, and that includes a nominalizing head n. These properties also account for the mixed behavior of these constructions, which exhibit a clausal template while manifesting several nominal features. Moreover, the proposed analysis also explains other apparently non-related properties of the language, such as the lack of tough-constructions, extraction with factive verbs, obligatory cliticization, and internally-headed relative clauses. Additionally, we claim that this nominal head has a phonetic realization as a suffix -a, detected in several nominal phrases and which is possibly linked to a Tupian nominalizer of the same form. In part II, we discuss constructions with the inverse voice morpheme ti-, claiming that these can be reanalyzed as cases of clitic-doubling. This analysis accounts for several properties of this preffix, such as its presence in cases of WHmovement, the eccentric pattern of agreement that it triggers, the complementary distribution with the passive, and the pressupositional reading of the theme in these constructions. Finally, two prosodic phenomena - the suffix -o and the intonational pattern of object relative clauses with ti- - are describedin detail and it will be shown how these provide additional evidence for the analyses developed in this dissertation.