The view that definite noun phrases are systematically accompanied by uniqueness implications is defended in this dissertation against rivals and counterexamples, with emphasis on the precise understanding of the phenomenon and factors which have obscured its generality. It is argued that all definites are 'unique', in a sense defined within the Kamp/Heim framework on the basis of discourse representations containing implicated, accommodated and contextually supplied material. Definites are contrasted with indefinites, including indefinites with numeral determiners. In quantified sentences, a definite may be unique relative to another element. For example, the use of the definite pronoun in Every chess set comes with a spare pawn. It is taped to the top of the box implies that there is a unique spare pawn per chess set. It is argued that uniqueness depends on the configuration of quantifiers and their scope; roughly, a definite A is unique relative to an element B if it has narrow scope relative to B. A general theory of uniqueness is proposed, and applied to a variety of examples. Then the 'asymmetric' reading of quantified sentences is discussed; that is, for example, the reading of Most women who have a dog are happy which is false in the situation where out of ten women, one has 50 dogs and is happy, and the other nine have exactly one dog each and are miserable. The asymmetric reading is not captured by the unselective quantification found in the Kamp/Heim system. An analysis of this reading is proposed, which finds support in the behavior of definite NPs in 'donkey' sentences, and which derives a cluster of properties of the reading from the asymmetry inherent to scope relations. Thus both uniqueness and the asymmetric reading are seen to reflect scope relations. The combined understanding and analyses of these phenomena are shown to reveal and explain systematic patterns in the behavior of donkey sentences, and illuminate puzzling speakers' judgements concerning these sentences.

This dissertation investigates the syntax and semantics of two types of wh-constructions in Japanese. One is the construction that involves so-called “indeterminate phrases”, such as dare ‘who’ and nani ‘what’. The other is what is known as the “internally headed relative clause construction”. In both of these wh-constructions, association of two non-local positions in a sentence takes place. The association has been most commonly analyzed in terms of invisible movement. This dissertation reexamines this assumption. Indeterminate phrases differ from English wh-phrases in that they occur in the universal construction as well. Previous studies have put primary focus on indeterminate phrases in the interrogative construction, and tried to extend their analysis to the universal construction. Chapter 2 shows that examining the universal construction provides a new window on the nature of association between indeterminate phrases and the question particle ka and the universal particle mo. As a consequence of a proposed semantics of the universal construction, a uniform analysis of the two constructions emerges that does not involve English-type wh-movement. It has been a long-standing puzzle that the association between indeterminate phrases and ka and mo can take place across islands for movement, except for wh-islands. This presents a challenge to any movement theory. The analysis presented in this chapter allows a switch of perspective on this long-standing puzzle, and the puzzle is shown to follow from the interpretative process. Chapter 3 examines the internally headed relative clause construction, with particular focus on new data that involve quantificational NPs and indeterminate phrases. The data provide arguments for representations in which the internal head remains internal at LF. Furthermore, it is shown that a kind of “argument sharing” observed in this construction is better analyzed in terms of anaphora rather than in terms of movement. Evidence is presented that the interpretation of this construction involves E-type anaphora in particular. An explicit mechanism for compositional interpretation is proposed, which also derives a restriction on possible internal heads. This study has the cross-linguistic implication that the constructions called “internally headed relative clauses” in various languages do not form a homogeneous class.

A long-standing controversy in the interface between phonetics and phonology involves the nature of sonority. This dissertation seeks to help resolve this problem by showing that the sonority hierarchy is both physically and psychologically real. This is accomplished by reporting the results of two rigorous and in-depth experiments. The first of these involves phonetic (instrumental) measurements of five acoustic and aerodynamic correlates of sonority in English and Spanish: intensity, frequency of the first formant, total segmental duration, peak intraoral air pressure, and combined oral plus nasal air flow. Intensity values are found to consistently yield a correlation of at least .97 with typical sonority indices. Consequently, sonority is best defined in terms of a linear regression equation derived from the observed intensity results. The second major experiment—this one psycholinguistic in nature—involves a common process of playful reduplication in English. A list of 99 hypothetical rhyming pairs such as roshy-toshy was evaluated by 332 native speakers. Their task was to judge which order sounds more natural, e.g., roshy-toshy or toshy-roshy. The data again confirm the crucial importance of sonority in accounting for the observed results. Specifically, the unmarked (preferred) pattern is for the morpheme beginning with the more sonorous segment in each pair to occur in absolute word-initial position. A generalized version of the Syllable Contact Law is utilized in the formal analysis of this phenomenon in terms of Optimality Theory. Finally, a complete and universal sonority hierarchy is posited by building on the findings of the two experiments as a whole.

Abstract not available

This work proposes a series of ternary scales which make certain phonological distinctions traditionally made with binary or privative features. The dissertation focuses on the Inherent Voicing scale, which has the values Voiceless Obstruent, Voiced Obstruent, and Sonorant. This scale replaces the traditional features (voice) and (sonorant). Other scales proposed are the Consonantal Stricture scale (stop, fricative, approximant/vocoid) and the Vowel Height scale (low, mid, high), which replace (continuant), (consonantal), (high) and (low). Applied within Optimality Theory, the ternary scales framework provides natural explanations for a number of phonological processes which are opaque in binary models. A ternary scale groups together certain phonological classes, while also making a statement that some values on the scale are closer to each other than others. Specifically, some values are adjacent on the scale and others not. This statement is impossible to make in binary features, but is necessary to capturing certain phonological phenomena. Once assimilation constraints and faithfulness constraints are allowed to make reference to the order and adjacency of the scale values, a natural explanation emerges for previously puzzling processes, such as chain shifts, attraction (as when voiceless obstruents voice before sonorants, or low vowels become mid before high vowels) and coalescence (as when low vowels and high vowels coalesce to mid). Other, more apparently binary processes (such as voicing assimilation or neutralization in obstruents) are also analyzed on the ternary scale. Such processes are part of a larger class of phenomena which call for a ternary analysis. Chapter 2 considers assimilation and attraction on the Inherent Voicing scale, showing that voicing assimilations in obstruents are a subset of assimilations occurring on the full ternary scale. Chapter 3 analyzes chain shifts (where voiceless obstruents voice and voiced obstruents become sonorants) as one-step movements along the ternary scale, caused by faithfulness constraints which require that the output stay near the input. Chapter 4 examines the effects of the markedness of voiced obstruents (including coda neutralization) in the context of the Inherent Voicing Scale. Chapter 5 turns to the Consonantal Stricture Scale and Vowel Height Scale, applying analyses developed in earlier chapters.

This dissertation develops the hypothesis that morphologically-related words are required to be phonologically identical by ranked and violable constraints. Pairs of surface forms are linked by a transderivational or output-to-output (OO) correspondence relation. Through ranking, constraints on the OO-correspondence relation may force a derived word to deviate from the canonical surface patterns of the language in order to be more like its output base. This theory obviates the traditional analysis that deviant phonology in complex words is the product of cyclic derivation. Given transderivational relations, cyclic effects are produced by constraint interaction in nonprocedural Optimality Theory. Cyclic effects are better understood as misapplication identity effects, similar to the over- and underapplication phenomena observed in reduplicated words. Phonological processes may overapply (take place where they are not properly conditioned) or underapply (fail to apply where properly conditioned) to achieve surface identity of paradigmatically-related words. Constraints that demand identity in paradigms interact directly with phonological markedness constraints and input-output faithfulness requirements. When OO-correspondence constraints take precedence, phonology misapplies. Three case studies are presented. The Austronesian language Sundanese shows an overapplication pattern, and Tiberian Hebrew demonstrates underapplication identity effects. In both cases, paradigmatic identity is achieved at the cost of greater markedness in surface forms. Both of these languages also show that paradigmatic identity is sacrificed when it would produce too marked a structure, providing support the claim that OO-correspondence constraints are ranked in a fixed, monostratal grammar. The study of English paradigms presents a theory of phonological classhood. Two arbitrarily-defined classes of affixed words participate in different transderivational identity effects. Each affix class triggers a distinct OO-correspondence relation governed by its own set of faithfulness constraints. All class-specific phonological behavior follows from the ranking of the two sets of OO-correspondence constraints. In this tranderivational theory, phonology is sensitive to morphology because phonological faithfulness relations hold over paradigmatically-related words. There are no cycles or levels of derivation. Complex words, like simplex words, are derived in a parallel grammar, without any intermediate stages.

This dissertation proposes a hypothesis on syllable structure and examines resyllabification of word final segments in Caracas Spanish and in standard Portuguese. The most important aspect of the hypothesis is that the Onset and the Nucleus of the syllable are a constituent $\sigma \sp\prime$, lower than the constituent that includes postnuclear segments. $\sigma\sp\prime$ is optimally CV, where C is a consonant or consonant cluster, but it can contain a prenuclear glide. Maximizing $\sigma\sp\prime$ prevents the syllable from incorporating postnuclear segments into all of the possible postnuclear syllabic positions. Syllabification at the post-lexical level, which is called resyllabification, can alter the syllable structure of words by adjoining the syllable final margin or Coda of a word to the syllable initial margin of a following word if this syllable has an empty Onset. It is the position of this dissertation that glides occupy an Onset position, thus resyllabification does not take place if the following syllable starts with a glide. The first chapter of the dissertation is a review of the notion of syllable, a review of possible syllables in human languages and a presentation of the prosodic units for Spanish and Portuguese. It is proposed that although there is variation among languages all present a strong tendency to have syllables of the type CV. The second chapter deals with previous studies on the syllable, from traditional to generative phonology. The hypothesis is presented in detail in the third chapter and the fourth chapter is a sociolinguistic study of the behavior of word final consonants in Caracas Spanish. The study is supported by a computational and probabilistic analysis of the data using GoldVarb. The results of the analysis confirm the prediction about resyllabification and show that this process takes place post-lexically, after all other processes have taken place at the word level. To confirm the lack of resyllabification of the syllable final margin when it is a velar nasal or when the following syllable starts with a glide, a questionnaire is added and the results, once again, support the initial hypothesis.

This dissertation provides solutions for some problems of quantification and anaphora interpretation in VP ellipsis contexts. The introductory Chapter 1 begins with brief descriptions of the problems and some of the basic assumptions employed in this thesis. I will then discuss the 'identity problem' of VP ellipsis and argue that the relevant identity condition for VPs is determined at the level of LF but is oblivious to indices on anaphoric expressions. This conclusion leads to a question of how to constrain indexing of pronominal elements in VP ellipsis contexts so that unavailable interpretations are successfully eliminated. In Chapter 2, I argue that the distribution of indices in VP ellipsis contexts is constrained by a pragmatic principle, namely a principle of focus interpretation. It will be shown that some of the most recent development in the semantics and pragmatics of focus gives the right degree of restrictiveness on indexing. Chapter 3 is an investigation of a curious puzzle of quantifier scope in VP ellipsis contexts. When a quantificational NP is contained in the antecedent VP, its interpretation appears to be dependent on the quantificational structure of the ellipsis site. I will argue that the unavailable interpretations of quantifiers in those contexts are ruled out because they do not satisfy the condition of focus interpretation. The solution advocated here removes the need for special syntactic constraints on scope taking in ellipsis contexts. The core empirical issue in Chapter 4 is a set of examples in which a pronoun in VP ellipsis contexts allows a so-called 'sloppy' reading in a structural configuration that is believed to disallow variable binding. These examples threaten the standard view that a strict reading of a pronoun is brought about by the referential nature of the pronoun while a bound variable pronoun yields a sloppy reading under ellipsis. I will provide a solution for this puzzle, in which I will relate the problem to the semantics of indefinite NPs and pronouns anaphoric to them.

There are a variety of phonological asymmetries exhibited by segments which appear in perceptually or psycholinguistically prominent positions such as roots, root-initial syllables, stressed syllables, and syllable onsets. In such positions, segmental or featural contrasts are often maintained, though they may be neutralized in non-prominent positions. Segments in prominent positions frequently trigger phonological processes such as assimilation, dissimilation and vowel harmony; conversely, they often block or resist the application of these processes. The goal of this dissertation is to develop a theory of positional faithfulness which will both generate and explain the range of positional asymmetries attested in natural language phonology. Chapter 1 introduces the notion of positional privilege, as well as the fundamental aspects of Optimality Theory. Positional faithfulness constraints are introduced and demonstrated in an analysis of onset/coda asymmetries in Catalan voice assimilation. I argue that positional faithfulness provides an explanation for the attested onset/coda asymmetries that is not afforded by licensing alternatives. Faithfulness in root-initial syllables, a position in which prominence derives largely from psycholinguistic (rather than phonetic) properties, is considered in Chapter 2. Particular attention is given to the analysis of vowel harmony in Shona, and to the phonology of consonantal place in Tamil. Chapter 3 is devoted to the domain of stress, showing once again that positional faithfulness constraints unify and explain a wide range of phonological asymmetries associated with the positional prominence. The core of the chapter is an analysis of nasal harmony in Guarani; vowel reduction in Catalan is also examined. In Chapter 4, I turn to positional privilege effects which are sensitive to the distinction between root and affix. Such cases provide further support for positional faithfulness theory. Finally, in Chapter 5, a different type of positional faithfulness effect, that of positional maximization, is examined. I argue that constraints which favor maximal packing of prominent constituents are necessary. Such constraints are crucial in cases of prominence-driven ambisyllabicity, as in Ibibio. Positional MAX constraints also account for the appearance of complex syllable margins in prominent positions, though complex margins may be excluded elsewhere in the language.

In this work I investigate the syntax of determinerless (D$\rm\sb\phi)$ subjects and the interpretations associated with them. The main focus is on explaining the contrast in syntactic behavior and interpretation between languages of the Catalan type and languages of the English (and Germanic) type. Ultimately, the system proposed will account for the general crosslinguistic variability in the interpretation and syntax of these elements. I develop a system that derives their interpretive properties from a particular combination of (independent) syntactic properties. More concretely, I derive the (exclusively) existential interpretation of D$\rm\sb\phi$ subjects in Catalan, from the syntactic configuration that results from (long) V-movement and the use of Merge (instead of Move) to fill the Spec of the head where V moves. The different crosslinguistic patterns result from the combination of the V-movement properties in any given language and the use of Merge/Move to fill the (relevant) Spec: English, for instance, has short V-movement and Move; some Germanic languages (e.g. Icelandic) have long V-movement and Move. This work, thus, is part of a research program that investigates how syntax and semantics interact, in particular, how syntactic configurations limit the range of semantic interpretations crosslinguistically. As part of this approach, I will claim that the interpretation of (subject) arguments is not strictly tied to a particular position, but is dependent on the particular configuration in which the argument appears. In particular I will claim that what is crucial is its relative position with respect to the verb and the operators in the sentence. Thus, a single syntactic position may have different properties across languages depending on the particular configuration in which it appears (which in turn is determined by the specific syntactic properties of the language). Such configuration will, in turn, determine its interpretation in the interpretive component.