Search

A doctoral dissertation on general clitic theory and Finnish particle clitics in particular looks at the mixed linguistic status of clitics and proposes that they are not as mysterious as reported, but should be viewed either as bound words (bound lexemes) or phrasal affixes. An initial chapter introduces the Finnish particle clitics and their characteristics and status. A second chapter presents a syntactic account of (sentential) connective and epistemic adverbs and the closely related class of conjunctions, within the framework of generalized phrase structure grammar, to illustrate how Finnish particle clitics are related to independent words that function as sentential adverbs. A third chapter identifies syntactic, semantic, and prosodic properties common to the sentential adverbs and particle clitics, and another chapter reviews the literature on cliticization. Finally, a discussion of the morphosyntactic status of the particle clitics is presented, offering some external evidence for the bound word status of second position clitics. Appended materials include notes on bound words in the Finnish lexicon and a brief comparison of Finnish and Lappish morphology. (MSE)

The distribution of adverbs is particularly difficult to account for, given the amount of variation it encompasses. Not only are adverbs typically optional, but any adverb may also appear in several different positions relative to other constituents, with placement differing according to adverb type and language. As a result, although adverbs are not essential clausal mainstays, the way they are incorporated into the syntax has crucial implications for an overall understanding of clause structure. Some recent accounts of adverb distribution, most notably Cinque (1999), require a highly articulated clausal cartography, where each adverb fits into a specific syntactic position. The placement of adverbs is determined by their semantic properties inasmuch as their specified positions correspond to semantic classes. The ordering of these positions is syntactically predetermined, supposedly with no little or no semantic input. More semantics-based accounts of adverb distribution, as exemplified by Ernst (2002), do not restrict adverbs to specific positions. Rather, any adverb may adjoin to any projection, as long as its individual semantic requirements are satisfied. Such theories of distribution thus depend on adverbs’ semantic interactions with each other and other constituents. The differences between these ‘syntactic’ and ‘semantic’ approaches have led to questions about the nature of verb movement, functional projections, and adjunction. The debate over adverb distribution also raises the issue of what contribution semantics makes to the syntax, and what is syntactically primitive. The aim of this dissertation is to develop an account of adverb distribution that neither requires the introduction of new functional projections, nor attempts to shoehorn an external semantic hierarchy onto a pre-existing syntactic one. It will focus on the position of adverbs in relation to other constituents rather than their order with respect to each other. In this thesis I will review previous theories of adverb distribution, giving special attention to Cinque’s (1999) ‘functional specifier’ approach and Ernst’s (2002) ‘semantic adjunction’ approach, as well as some alternatives, especially the VP-remnant analysis proposed in Nilsen (2003). I will then look at the little-discussed phenomenon of ‘Adverb Climbing’ (AC), in which an adverb precedes a verb that takes an infinitival complement, but is interpreted as modifying the embedded rather than the matrix verb. Taking the varying availability of AC with Control and Raising verbs as a starting point, I will develop a theory of adverb licensing that determines where an adverb may adjoin according to its location in relation to a particular projection. Specifically, I will propose that an adverb must c-command the projection it modifies, and must have access to that projection either in the same phase or at the edge of a lower phase. Based on this analysis I will argue that AC is in fact an indicator of restructuring, and that control and raising verbs take different sizes of infinitival complement. I will also examine the distribution of ‘verb-modifying’ adverbs. Drawing on previous ‘split VP’ proposals (e.g. Ramchand 2008; Travis (2010)), I will contend that the varying distribution of agentive, subject-oriented, and manner adverbs indicates that each is distributed in relation to a different projection within the vP, and that some postverbal adverbs are complements of VP. This proposal will require the introduction of crosslinguistically parameterised restrictions on the order in which adverbs and feature-checking elements may be merged to a single projection. Moreover, I will argue that the array of positions available to agentive adverbs indicates that English has head movement within the vP which bypasses a head, violating Travis’s (1984) Head Movement Constraint (HMC). I will then posit a new analysis of head movement which allows for this violation while still precluding the instances of ungrammaticality that the HMC was meant to rule out. I will finally discuss the distribution of adverbs and negation in the IP range, giving special attention to Pollock’s (1989) classic data from English and French. I will develop an analysis of negation which will allow me to explain the distribution of both sentential adverbs and negation without splitting the IP. Further refinement of the ordering restrictions on multiple merge will also provide an explanation for the ungrammaticality of an adverb between a subject and the highest verb in French, and between do and not in English. This dissertation will serve to situate the study of adverb distributionwithin Chomsky’s (1995) Minimalist framework while providing fresh insight into the extent to which adverb distribution may be used as an indicator of clause structure and movement of other constituents.

The use of gradable adjectives and adverbs represents an important means of expressing comparison in English. The grammatical forms of comparatives and superlatives are used to express explicit orderings between objects with respect to the degree to which they possess some gradable property. While comparatives are commonly used to compare two entities (e.g., “The blue whale is larger than an African elephant”), superlatives such as “The blue whale is the largest mammal” are used to express a comparison between a target entity (here, the blue whale) and its comparison set (the set of mammals), with the target ranked higher or lower on a scale of comparison than members of the comparison set. Superlatives thus highlight the uniqueness of the target with respect to its comparison set. Although superlatives are frequently found in natural language, with the exception of recent work by (Bos and Nissim, 2006) and (Jindal and Liu, 2006b), they have not yet been investigated within a computational framework. And within the framework of theoretical linguistics, studies of superlatives have mainly focused on semantic properties that may only rarely occur in natural language (Szabolsci (1986), Heim (1999)). My PhD research aims to pave the way for a comprehensive computational treatment of superlatives. The initial question I am addressing is that of automatically extracting useful information about the target entity, its comparison set and their relationship from superlative constructions. One of the central claims of the thesis is that no unified computational treatment of superlatives is possible because of their great semantic complexity and the variety of syntactic structures in which they occur. I propose a classification of superlative surface forms, and initially focus on so-called “ISA superlatives”, which make explicit the IS-A relation that holds between target and comparison set. They are suitable for a computational approach because both their target and comparison set are usually explicitly realised in the text. I also aim to show that the findings of this thesis are of potential benefit for NLP applications such as Question Answering, Natural Language Generation, Ontology Learning, and Sentiment Analysis/Opinion Mining. In particular, I investigate the use of the “Superlative Relation Extractor“ implemented in this project in the area of Sentiment Analysis/Opinion Mining, and claim that a superlative analysis of the sort presented in this thesis, when applied to product evaluations and recommendations, can provide just the kind of information that Opinion Mining aims to identify.

El superlativo absoluto en español puede construirse mediante el adverbio muy o el sufijo –ísimo. A pesar de que la reintroducción de la forma morfológica ha sido datada de finales de la Edad Media, investigaciones más recientes apuntan al siglo XIX como fecha clave para su popularización. Con el fi n de analizar su distribución dialectal en el español peninsular a lo largo del siglo XX, hemos extraído datos del Atlas lingüístico de la Península Ibérica (ALPI) y de otros posteriores. De acuerdo con estos, observamos que, efectivamente, el sufijo –ísimo está circunscrito a un área geográfica determinada, mientras que muy ocupa la zona restante, sin dar apenas ocurrencias de la solución morfológica. A ello hay que añadir el uso de mucho (muncho en ciertos enclaves) como grado del adjetivo, siendo este fenómeno una reminiscencia del estado lingüístico del paso del latín al español.

Este trabajo presenta un nuevo marco interpretativo para analizar la gramaticalización del adverbio sí como adverbio de polaridad en español antiguo. El artículo muestra que existen diferencias sintácticas importantes en la expresión de la polaridad positiva entre el español antiguo y el español actual, pues en el primero esta no solo se encomendaba al adverbio sí, sino que guardaba relación con el movimiento del verbo, el foco, la posición de los clíticos, la elipsis del SV y las propiedades del Sintagma Polaridad. Se defiende que la reestructuración de la periferia izquierda en español clásico afectó a la expresión de la polaridad positiva y posibilitó la gramaticalización de sí como adverbio de polaridad, así como la aparición de estructuras sintácticas desconocidas en la lengua medieval, como el empleo de sí proforma oracional o la estructura sí que. Todos estos cambios se vinculan a las propiedades del núcleo Polaridad, cuyo carácter fuerte o débil condiciona la expresión sintáctica de la polaridad en español antiguo.

Recent studies have begun to redefine the idea of accuracy in second language acquisition to include not only grammatical correctness, but also native-like selection. This is an exploratory study aimed at identifying areas of nonnative-like selection of adverbs, such as sentence position, semantic category preferences, frequency of use and breadth of word choice. Using corpus-linguistic methods it compares the writing of nonnative English speakers at an intermediate and advanced level to both American college students’ writing and published academic writing. It also conducts in-depth case studies of three of the most commonly used adverbs. It finds that while advanced students are grammatically accurate, there are still several ways in which their use of adverbs differs from that of native speakers.

This article develops a diachronic analysis of the Latin adverbs máxime and inclusive in Spanish. This study situates their first chronological and textual appearances in our language and proves their spread; and, at the same time, the successive manifestations of their value and the transitions between their consecutive functions are described. Lastly, their behavior in Spanish is examined from a comparative perspective that takes into account as well other elements of language. The historical analysis shows that máxime and inclusive share not only their status as Latin adverbs but also their linguistic evolution. After a discreet incorporation into the 15th century Castilian, these adverbs transcended diaphasic and diastratically: from formal to informal language, and ?though in different degree for each one? from the most cultured speakers to users of varied socio-cultural level.

This thesis explores adverb distribution in Iquito, a Zaparoan language spoken byapproximately 25 people in the northern Amazon Basin of Peru. The syntactic distributions of Iquito adverbs correspond to four semantic classes: time, manner,epistemic, and an intensifier. Time adverbs have the broadest distribution, occurring before the topic of a topicalized sentence, between topic and subject, after the verb, and after the object of a transitive sentence. Manner adverbs have a similar distribution, but are not found before topic. Epistemic adverbs have an even narrower distribution, never occurring sentence-initially (whether the sentence is topicalized or not) and rarely occurring between topic and subject. The intensifier adverb has the most restricted distribution, as it only occurs before adjectives or other adverbs. These distributions can be used to classify "atypical" adverbs, namely infinitival verbs that are used adverbially. Furthermore, these distributions shed light on the phrase structure of Iquito. Adverbs are analyzed in the literature as adjuncts, and the allowable positions are explained either as the result of adjunction to different constituents (Ernst 2002; Iatridou 1990) or movement between adjoined positions (Cinque 1999). The pre-verbal positions of Iquito adverbs, particularly in irrealis and negated constructions, raise questions for these analyses. The data demonstrate that adverbs can occupy non-adjoined positions, namely the object position in an irrealis (SOV) construction and possibly negation, forcing a reevaluation of the current treatment of adverbs. The research also expands the existing documentation on Iquito.

Nominal stems ending in t are very rare in Proto-Indo-European. Despite this fact Ancient Greek displays a series of nominal stems that end in t. Their presence in Ancient Greek suggests that the insertion of t in words of PIE origin must be a process which occurred in Greek itself. From a diachronic perspective these stems fall into two categories. The first is represented by words which display the t in their stems in all the historical records we have, both in epigraphical evidence-including Mycenaean-and in literary sources. For this category we cannot trace the moment and the place when this “older” t entered the paradigms. The second category, however, contains words that show an allomorphic variation between t-ful stems and t-less stems. Given the fact that Homer uses in the majority of cases the t-less stems one can see, in post-Homeric dialects, how the stems with this “intrusive” t coexist with the t-less stems and win eventually over. The main goal of the dissertation is to see how this allomorphic variation evolved through historical Greek. Its second purpose is to see what this allomorphic variation can tell us about the origin of the intrusive t. The perfect active participle shows up in historical Greek as a t-stem. Nevertheless, scanty evidence from Mycenaean suggested that in this dialect of Greek the participle may have been t-less. The work addresses this problem again and comes up with the hypothesis that the forms Mycenaean displays might be only adjectives derived from former t-less participles, but not t-less participles. Consequently, the t-ful participle may still be a creation of Common Greek. A last issue regards the –ti adverbs in Greek, which have been often considered to be former locatives of t-stems. The study shows that this hypothesis does not seem to be true and that they are more likely to be the result of an analogical process, which occurred in Greek and not in PIE.

This thesis presents detailed descriptions of the English adverbial of degree (e.g., very, quite,rather,extremely) and the adverbial of extent (e.g., much, some, at all, excessively).