No description available.

No description available.

No description available.

This thesis investigates within a generative framework a cluster of morphological and syntactic facts in the Old Babylonian variety of Akkadian, whose common property, despite many divergences, is the occurrence of structurally reduced nominal constituents. Akkadian nominals, whilst in their normal morphosyntactic form fairly rich both inflectionally and in their capacity to support complex nominal constituents, appear in various restricted syntactic contexts in reduced forms, entailing the loss of affixal expression of features (especially case, as well as gender and number), the barring of modifiers, as well as other restrictions. These phenomena may be divided into several major categories, one of which has clear parallels in well-studied Semitic languages (the nominal-internal 'construct state'), and others which do not have such parallels, including nouns in the construct-state morphological form as heads of relative clauses and the 'stative' (a peculiar form of nominal predication). Each of these phenomena is described, investigated and analysed in successive chapters, both individually and in terms of their interrelations and differences, and their possible implications for various aspects of generative syntactic theory are explored. For example, the investigation of the Akkadian construct state construction has important implications for aspects of the general generative theory of these constructions; construct-headed relative clauses both for this and for the theory of headed relative clauses, especially for the ongoing debate concerning the internal/external status of the head and some of the fine properties of the 'raising analysis'; the stative suggests the hitherto unrecognised existence of denominal incorporation to a verbal head producing a copula-like interpretation. This is the first extended generative study both of the Akkadian construct state and of the other phenomena mentioned above.

This thesis is a study into certain areas of Meroitic phonology. The Meroitic language was spoken in an area that encompasses modern day Nubia (southern Egypt to northern Sudan). Evidence for the Meroitic language is only known through the survival of its inscriptions, whereby two forms are used to write these: hieroglyphic and cursive, both heavily borrowed from the Ancient Egyptian writing system. The Meroitic language has only been partially deciphered; Griffith (1911) established approximations for the signs’ sound values, along with identifying a handful of lexical items. Progress into the decipherment of the language has been seriously hampered by the lack of any bilingual texts, and more importantly, a lack of evidence for a genetic affiliation with an existing language or language family. This thesis concentrates on investigating the traditional representations given for the phonemic values of the Meroitic signs. The methods used for investigating this are: firstly, through analysing the correlative phonemic values of signs taken from transcriptions from languages such as Ancient Egyptian, Coptic and Greek, where equivalent forms with Meroitic ones are evidenced. These transcriptions from other languages are given with their sources. Secondly, empirical and typological phonological evidence is used to support the proposed revisions to the phonemic values of certain Meroitic signs, and thirdly the investigation also analyses these proposals within a theoretical framework, principally Government Phonology. Through this investigation, I not only challenge the traditional representations of certain signs but also present revisions to them. I highlight that research into the Meroitic script has to take into account the level at which the script is encoding the Meroitic language, whether this is the phonetic or phonemic level.

A central concern of psycholinguistic research is explaining the relative ease or difficulty involved in processing words. In this thesis, we explore the connection between lexical processing effort and measurable properties of the linguistic environment. Distributional information (information about a word’s contexts of use) is easily extracted from large language corpora in the form of co-occurrence statistics. We claim that such simple distributional statistics can form the basis of a parsimonious model of lexical processing effort. Adopting the purposive style of explanation advocated by the recent rational analysis approach to understanding cognition, we propose that the primary function of the human language processor is to recover meaning from an utterance. We assume that for this task to be efficient, a useful processing strategy is to use prior knowledge in order to build expectations about the meaning of upcoming words. Processing effort can then be seen as reflecting the difference between ‘expected’ meaning and ‘actual’ meaning. Applying the tools of information theory to lexical representations constructed from simple distributional statistics, we show how this quantity can be estimated as the amount of information conveyed by a word about its contexts of use. The hypothesis that properties of the linguistic environment are relevant to lexical processing effort is evaluated against a wide range of empirical data, including both new experimental studies and computational reanalyses of published behavioural data. Phenomena accounted for using the current approach include: both singleword and multiple-word lexical priming, isolated word recognition, the effect of contextual constraint on eye movements during reading, sentence and ‘feature’ priming, and picture naming performance by Alzheimer’s patients. Besides explaining a broad range of empirical findings, our model provides an integrated account of both context-dependent and context-independent processing behaviour, offers an objective alternative to the influential spreading activation model of contextual facilitation, and invites reinterpretation of a number of controversial issues in the literature, such as the word frequency effect and the need for distinct mechanisms to explain semantic and associative priming. We conclude by emphasising the important role of distributional information in explanations of lexical processing effort, and suggest that environmental factors in general should given a more prominent place in theories of human language processing.

Correct pronunciation of spoken language requires the appropriate modulation of acoustic characteristics of speech to convey linguistic information at a suprasegmental level. Such prosodic modulation is a key aspect of spoken language and is an important component of foreign language learning, for purposes of both comprehension and intelligibility. Computer aided pronunciation teaching involves automatic analysis of the speech of a non-native talker in order to provide a diagnosis of the learner's performance in comparison with the speech of a native talker. This thesis describes research undertaken to automatically analyse the prosodic aspects of speech for computer aided pronunciation teaching. It is necessary to describe the suprasegmental composition of a learner's speech in order to characterise significant deviations from a native-like prosody, and to offer some kind of corrective diagnosis. Phonological theories of prosody aim to describe the suprasegmental composition of speech for a specific language. It is argued here that the suprasegmental composition of the speech of a non-native talker can be highly influenced by mother-tongue interference thereby rendering a language-specific phonological representation of prosody inappropriate. Moreover, languages vary in the way acoustic characteristics of speech are modified to manifest prosodic aspects of speech and the only secure means available to describe prosody for foreign language teaching therefore lies at an acoustic-phonetic representation. The automatic prosodic analysis of speech presented in this thesis aims to provide such an acoustic-phonetic representation. The prosodic aspects of speech are described in a syllabic domain which is synchronised with a phonetic segmentation. An algorithm is presented which groups acoustic-phonetic segments into syllabic units.

We present a new type of analysis for scientific text which we call <i>Argumentative Zoning</i>. We demonstrate that this type of text analysis can be used for generating user-tailored and task-tailored summarises or for performing more informative citation analyses. We also demonstrate that our type of analysis can be applied to unrestricted text, both automatically and by humans. The corpus we use for the analysis (80 conference papers in computational linguistics) is a difficult test bed; it shows great variation with respect to subdomain, writing style, register and linguistic expression. We present reliability studies which we performed on this corpus and for which we used two unrelated trained annotators. The definition of our seven categories (argumentative zones) is not specific to the domain, only to the text type; it is based on the typical argumentation to be found in scientific articles. It reflects the attribution of intellectual ownership in articles, expressions of author’s stance and typical statements about problem-solving processes. On the basis of sentential features, we use a Naive Bayesian model and an ngram model over sentences to estimate a sentence’s argumentative status, taking the hand-annotated corpus as training material. An alternative, symbolic system uses the features in a rule-based way. The general working hypothesis of this thesis is that empirical discourse studies can contribute to practical document management problems: the analysis of a significant amount of naturally occurring text is essential for discourse linguistic theories, and the application of a robust discourse and argumentation analysis can make text understanding techniques for practical document management more robust.

In this thesis I investigate the mechanisms and processes underlying figurative language comprehension. I attempt to determine whether there might be something unique about metaphor due to the interpretation processes involved, or whether metaphoric interpretations are in fact processed in the same way as other non-literal uses of language, such as hyperbole. Various theoretical accounts of figurative language interpretation from different pragmatic and psychological processing models are examined from an empirical perspective as a way of exploring the cognitive basis of their claims. Previous empirical research investigating metaphor comprehension is critically discussed not just from a psychological perspective, but also in relation to pragmatic accounts of figurative language. There is a plethora of past and current theoretical literature on metaphor, which over the last few decades has been discussed in relation to psycholinguistic research investigating metaphor processing. In contrast, despite recent unified pragmatic accounts of figurative language, which posit a unified account of metaphor, hyperbole, and other loose uses, there has been little, if any empirical research looking at hyperbole or other tropes. This leaves us with an important question; can what we know about the processing of metaphor be generalised to other tropes such as hyperbole? With this question in mind, I will present a series of on-line and developmental experiments, aimed at further exploring the processes and mechanisms underlying metaphor comprehension, and directly contrasting the processing of metaphor and hyperbole. The results of these experiments have implications both for psycholinguistic research on non-literal language processing, and for lexical pragmatic accounts of figurative language comprehension, but also for developmental research investigating children’s pragmatic capacities. As well as shedding light on the cognitive processes involved in constructing metaphoric and hyperbolic interpretations, the findings of this thesis give us some indication of the cognitive mechanisms that need to have developed in order to arrive at a non-literal interpretation of an utterance.

This thesis throws new light on issues debated in the experimental literature on neutralisation. They concern the extent of phonetic merger (the completeness question) and the empirical validity of the phonetic effect (the genuineness question). Regarding the completeness question, I present acoustic and perceptual analyses of vowel/zero alternations in Bedouin Hijazi Arabic (BHA) that appear to result in neutralisation. The phonology of these alternations exemplifies two neutralisation scenarios bearing on the completeness question. Until now, these scenarios have been investigated separately within small-scale studies. Here I look more closely at both, testing hypotheses involving the acoustics-perception relation and the phonetics-phonology relation. I then discuss the genuineness question from an experimental and statistical perspective. Experimentally, I devise a paradigm that manipulates important variables claimed to influence the phonetics of neutralisation. Statistically, I reanalyse neutralisation data reported in the literature from Turkish and Polish. I apply different pre-analysis procedures which, I argue, can partly explain the mixed results in the literature. My inquiry into these issues leads me to challenge some of the discipline’s accepted standards for characterising the phonetics of neutralisation. My assessment draws on insights from different research fields including statistics, cognition, neurology, and psychophysics. I suggest alternative measures that are both cognitively and phonetically more plausible. I implement these within a new model of lexical representation and phonetic processing, the Variability Field Model (VFM). According to VFM, phonetic data are examined as jnd-based intervals rather than as single data points. This allows for a deeper understanding of phonetic variability. The model combines prototypical and episodic schemes and integrates linguistic, paralinguistic, and extra-linguistic effects. The thesis also offers a VFM-based analysis of a set of neutralisation data from BHA. In striving for a better understanding of the phonetics of neutralisation, the thesis raises important issues pertaining to the way we approach phonetic questions, generate and analyse data, and interpret and evaluate findings.