Search

This work investigates the distribution of “wordhood issues,” in which a morpheme behaves like a word on one subset of wordhood parameters but like a bound item on another. The empirical focus is on the exponents of definiteness, case, indexation, and tense in 60 unrelated languages from five macro-areas. The methodological basis for the wordhood analyses is a set of eight parameters of phonological and morphological wordhood. The main result is that grammatical markers (“grams”) retain the ability of morphological words to co-occur with members of different syntactic categories even after being integrated into larger phonological word domains. Meanwhile, grams that show the syntagmatic behavior of affixes but prosodic freedom on at least one parameter are less frequent and limited to contexts in which the relevant stem domain is highly morphologically complex. These data can largely be explained by the diachronic models formulated in Bybee (2001, 2015) and Croft (2000b).

The thesis addresses the placement of `second position' clitics in a linear order in preposed predicates in Croatian. In particular, I propose a uniformly syntactic analysis of the `second position' effect in Croatian by analyzing the discontinuous AP predicates in root clauses and neutral discourse from Diesing\&Zec (2017). The motivation for the syntactic analysis of second position becomes inevitable with the evidence of novel data with the raising to subject construction presented in Chapter 3. The raising data with split and unsplit AP predicates forces the analysis where AP predicates are `split' earlier than post-Spell-Out (namely, in the syntactic component). This new evidence shows that AP predicates are not motivated by prosodically conditioned lowering of a clitic into the predicate in a post-syntactic component of the grammar, as most recently assumed by Diesing\&Zec (2017). In Chapter 3, I motivate the mechanisms that are involved in predicate `splitting': complex predicate formation and predicate inversion (Moro 1997). The former evacuates the head of the AP to spec BeP. This creates a remnant AP constituent which undergoes predicate inversion to spec TP (a la Moro (1997)). Second, the raising data force us to conclude that AP predicates prepose to spec TP position, in the same way as canonical DP subjects do. This result is consistent with the analysis of predicate initial copular constructions in (Moro 1997). In Chapter 2, I show that the `second position effect', more generally, involves movement of the closest XP to a specifier of a root node in a root clause, around a clitic complex. The clitics in the citic complex are a sequence of separate head constituents situated in a region of the clause higher than spec TP from where they do not move further. The analysis of the clitic complex in a stable region of a clause has been proposed in Romance (Kayne 1991), and the analysis where movement via attract closest occurs to the `1st position' in the root clause has been inspired by the root phenomenon in continental Germanic.

This dissertation analyzes subject pronouns in Bora, an endangered indigenous language of the Amazon. Bora uses a series of overt subject clitics to express the subject of clauses, which indicate cross-clausal coreference in many cases. Using data collected during personal fieldwork trips, I investigate the distribution of these clitics, and analyze their properties with respect to theories of control and binding. This accomplishes important research goals of (i) performing research on the Bora language, especially given its status as an endangered language, and (ii) using data from understudied languages to inform formal linguistic theory. After establishing a basic analysis of Bora syntactic structure, I review literature on binding theory and control structures, with the goal of showing how and whether the Bora data fit into existing linguistic theory. Regarding binding, I review canonical binding theory in generative grammar, as well as an alternative proposal that relies on reflexivity as a means of licensing anaphora. Regarding control clauses, I review analyses of PRO serving as a null, or in exceptional cases, an overt instantiation of a controlled subject. I compare this analysis with a theory of control that dispensed with PRO and instead analyzed control clauses as the result of syntactic movement. I conclude that, although Bora subject clitics appear in embedded clauses that would constitute control in other languages, such clauses in Bora are not control structures. Rather, every clause is a finite clause with an overt subject, with very few exceptions. I go on to show that overt subject clitics have anaphoric properties. For 3rd person subjects, an embedded proclitic i= (3COR) indicates coreference with the structurally next highest clause subject. The Speech Act Participant (SAP) clitic me= attaches to any verb with a 1st or 2nd person non-singular subject, and also indicates coreference with a higher clause subject when it appears without an overt subject NP. While the 3COR marker’s antecedent must appear in a higher clause, the SAP can take either a local or non-local antecedent, depending on its coreferent properties. The non-local nature of the antecedents of the 3COR marker and, in some cases, the SAP suggest that they constitute instances of long distance anaphora. This differs from the properties that I show for the 1st and 2nd person singular proclitics, which I argue to have properties of pronominals. After comparing the Bora data to instances of long distance anaphora in other languages, I determine that the Bora data has many similarities to these other languages. I argue that the ability of some of these clitics to be bound by non-local antecedents is similar to analyses of long distance anaphora in other languages. I first establish that Bora proclitics share properties with other types of long distance anaphors (being monomorphemic, occurring in restricted environments, having subject antecedents, and being subject to a Blocking Effect). I then provide an analysis based on similarity with Mandarin Chinese whereby the Bora anaphoric proclitics undergo raising at Logical Form in order to be bound by their antecedent, which must be the structurally next highest c-commanding subject in the sentence.

This dissertation approaches the idea of lexical types such as word, clitic and affix from an oblique angle. Starting with Cardinaletti & Starke's (1999) diagnostics for the Weak Pronoun, I deconstruct the category of clitic, breaking it down into two binary qualities: the syntactic primitive of being linked to a head of a different basic phrasal category, and the phonological primitive of being internal to the phonological word of the host. I argue that the syntactic behaviors of clitic-words, in particular pronominal clitic-words, are predictable if we assume that there is an operation that allows Xmin/max elements to fuse with X0 nodes during a syntactic derivation. This creates a complex X0 node that selects a clitic-type spell-out form. This fusion process—m-merge, adapted from Matushansky (2006)—is the same process that drives the incorporation part of head-movement. I also propose that the features identified as phonologically clitic-like—reduction, lack of a syllabic nucleus, no coordination and morphological simplicity—result from a similarity to inflectional affixes. Instead of arguing for a distinct clitic phonology I suggest that clitics—when they are simple feature bundles—and agreement affixes are functionally identical and therefore are likely to undergo spell-out in the same way, in particular by being realized internally to their hosts' phonological word, and being able to determine the shape of their host to a greater extent than we expect word-level phonological processes to allow. The evidence supporting these claims shows that many strange grammatical phenomena can be explained simply if we adopt the above principles. The operation m-merge—fusing an Xmin/max to an X0 node—offers a straightforward account of the WH as C pattern in interrogative relatives in Middle Bavarian and Lake Constance Alemannic. Considering how my model affects the diachronic trajectory of word-to-clitic-to-affix and adopting a restricted version of Roberts (2010) Move via Agree provides insight into the behavior of French subject pronouns in Old French, Standard French, and Contemporary Colloquial French. Most dramatically, my model offers a straightforward analysis for eight highly distinct types of φ marking in Middle Welsh, even to the point of predicting that three very different syntactic contexts will have the same realization for a φ-marked element. In sum, I argue that although clitics look and behave quite differently from canonical independent words and inflectional affixes, with a thorough understanding of the operations underlying head-movement and vocabulary insertion for paradigmatic elements, clitics are a predicted part of the model. Instead of requiring extra apparatus to explain, clitics offer us ways of simplifying our approach to syntax and the syntax-morphology interface as a whole.

This study addresses whether/how monolingual and bilingual Spanish-speaking children differ in their acquisition of grammar by examining direct object clitics in contexts where either proclisis or enclisis is possible (Lo voy a ver vs. Voy a verlo). The current study examines variable clitic placement in sociolinguistic interviews conducted with 21 Spanish-English bilingual children of Mexican descent, ages 5-11, and 71 monolingual children in Mexico, ages 6-11. All direct object clitics in variable contexts were extracted (N =140 tokens, .2% of total word count). Both the monolingual and bilingual children produced high rates of proclisis (76%, 75%, respectively). It is concluded that naturalistic production data do not support the view that English impacts bilingual childrens patterns of variable clitic placement in Spanish. Instead, their patterns of use are the same as those found among monolingual children.

In light of recent attempts to revive the operation of syntactic head movement and clitic movement in Phase Theory (Roberts 2010a, 2012), we argue that object clitics are underlyingly determiners in the syntax. Clitics engage in probe/goal relations to value and delete their uninterpretable Case features, and upon Agree, cliticize to their host via head-to-head incorporation. Although this account adopts the bare phrase structure theoretic mechanism employed by Ian Roberts to instantiate head movement (i.e., `defective goals'), the work outlined here diverges from the details of Roberts's account, most crucially by positing Abstract Case features on clitics. Based on clitic constructions from Standard Spanish, and various dialects, it will be demonstrated that the behavior of clitics, like that of other nominal elements, is governed by general abstract conditions on movement, namely Relativized Minimality (Rizzi 2013), Case Theory and the Phase Impenetrability Condition (Chomsky 2001, 2004, 2008). After a careful rethinking of well-known intervention and impenetrability effects (i.e., islands) involving clitics in Spanish, it is claimed that their movement, although unique in being both maximal and minimal, otherwise conforms to the standard conditions imposed on determiner phrases more generally. Contrary to recently influential Base Generation accounts, this work makes a case for distinguishing clitic movement from the movement of doubles, through a detailed study of Exceptional Case-Marking (ECM) constructions, where multiple clitic arguments can raise to object (Chomsky 2013). The complex array of possibilities involving clitic placement in these structures exemplifies the interaction of clitics with Case assignment and distinguishes the minimal nature of clitic head movement from XP movement of doubles. Finally, Chomsky's theory of Inheritance (2008) figures crucially in this account, as it is used to explain the order of clitics in clusters of two and three. Inheritance is also used to explain island effects that block clitic climbing. This study concludes by making the case that while in certain dialects, such as Los Angeles Vernacular Spanish (Parodi 2009a, 2011), clitics have apparently evolved into agreement/object markers, in most dialects, including the Standard, both direct object (DO) and indirect object (IO) clitics are argument pronouns that move to their derived positions in the syntax. Such pronominal clitics are contrasted with truly base-generated `morpheme' clitics, including `inherently' reflexive clitics and `speaker' ethical dative clitics (Strozer 1976), which cannot be doubled or related by the syntax to a corresponding stressed argument. The account that fellows then, although firmly within the movement tradition of clitics (Kayne 1975, Quicoli 1976) is intended to complement morphological approaches to clitic clustering with non-argument clitics (Cuervo 2013), and to shed light on the workings of the interface that relates the narrow syntax to the phonological component of the grammar.

The present work is a collection of studies on the Tocharian languages that focuses onthe phonological properties of units larger than the word. The first chapter involves a studyof the segmental properties of external sandhi in Tocharian A couched in the framework ofOptimality Theory. A philological study involving the Tocharian br ̄ahm ̄ı alphabet revealsthat there exists a correlation between external sandhi and the orthographic strategy usedto render a word-final consonant. The second chapter provides an analysis of the prosodicrelationship between clitics and their hosts, showing that these elements form a prosodicconstituent intermediate between the word and the phrase. This is the case both in TocharianA as well as in Tocharian B. The final chapter addresses two different aspects of the internalstructure of Tocharian Wh-words. First, the prosodic analysis employed for clitics andtheir hosts is extended to Wh-words in Tocharian B, making it possible to account fortheir otherwise aberrant accentuation as well as some of their segmental properties. Thesecond part of the chapter investigates the nature of indefinites in Tocharian A, showingthat, contrary to the descriptions in most handbooks, the attested forms do not belong toa single paradigm. Instead, we are dealing with two separate sets of Wh-words doublingas indefinites, a fact that has been obscured by a phonological process that renders one setidentical to demonstrative pronouns.

This dissertation presents a formal theory of Serbo-Croatian grammar. The theory predicts acceptable form/meaning pairs for a substantial chunk of Serbo-Croatian. In particular, we analyze Serbo-Croatian declarative and interrogative main clauses, embedded clauses, a couple of different types of nominal modification, control and predication, as well pro- and encliticization. Linguistic expressions are represented as triples of typed terms, with each typed term modeling one of the following sets of properties of a linguistic sign: semantic (i.e. truth-conditional meaning), tectogrammatical (i.e. syntactic combinatorial properties), and, finally, phenogrammatical properties which specify the expression's linearization possibilities. The focus of our work is on word order in Serbo-Croatian, which is very free in some respects but extremely rigid in others. With phenogrammar and tectogrammar as distinct components, we can isolate word order problems from tectogrammatical and semantic combination, and state theory-internal phenogrammatical generalizations. This is particularly important for the analysis of 2P enclitics, whose placement cannot be adequately characterized tectogrammatically.The most elaborate component is phenogrammar. We postulate many different phenogrammatical types and modes of combination. This enables us to create islands of fixed word order, while still allowing free reordering of higher-level phenogrammatical objects. Of special significance are phenogrammatical terms which denote sets of strings. Such terms represent possible pronunciations of expressions which can be linearized multiple ways without a change in meaning. Essentially, we are modeling semantically insignificant reordering as phenogrammatical indeterminacy. Our choice of grammatical architecture is empirically motivated, but methodological in nature. This dissertation purports to show that a decent theory of Serbo-Croatian word order can be given in a framework which does distinguish phenogrammar and tectogrammar; not that a comparable (or superior) theory cannot be given in one which does not. We do, however, believe that this project brings attention to languages with complex word order patterns and their role in delineating a realistic categorial grammar framework for linguistic analysis.

This dissertation has three interrelated goals: The main goal is an analysis of Czech clitics, units of grammar on the borderline between morphology and syntax with rather peculiar ordering properties both relative to the whole clause and to each other. We examine the actual set of clitics, their rather rigid ordering properties, and finally the properties of so-called clitic climbing. The analysis evaluates previous research, but it also provides new insights, especially in the position of the clitic cluster and in the constraints on clitic climbing. We show that many of the constraints regarding position of the clitic cluster suggested in previous research do not hold. We also argue that cases when clitics do not follow the first constituent are in fact not exceptions in clitic placement but instead unusual frontings. The second goal is the development of a framework within Higher Order Grammar (HOG) supporting a transparent and modular treatment of word order. Unlike previous versions of HOG, we work with signs (containing phonological, syntactic and potentially other information) as actual objects of the grammar. Apart from that, we build on the simplicity and elegance of the pre-formal part of the linearization framework within Head-driven Phrase Structure Grammar. Finally, the third objective is to test the result of the second goal by applying it on the results of the first goal.

This dissertation investigates the acquisition and the syntax of pronominal clitics in Croatian and Spanish, two languages that represent the distinction between second position (Wackernagel) and verb-adjacent (Tobler-Mussafia) clitics. It is proposed on the basis of both child and adult grammar that the same syntactic processes (in the case of pronominal clitics, verb movement) determine major aspects of clitic placement, and that surface differences (second position vs. verb-adjacent) can be attributed to independent, parameterized requirements. I investigate in detail clitic acquisition in the two languages from the onset of clitic production (around age 1;8) until children reach adult-like mastery (at age 2;9, with above 90% of adult-like production). I analyzed naturalistic data from six children, three speakers of each Croatian and Spanish (from the CHILDES database). I considered all the production of pronominal and anaphoric clitics, focusing on general structural properties and language particular characteristics that affect the syntax of clitics in the two languages. I argue that the acquisition of cliticization in the two languages is very similar, in terms of its path and speed. This is indicated by the early acquisition, around the same age (2;0), of all the major complexities in the clitic system of both languages, with a very low rate of errors (average 8%). I also evaluated several syntactic approaches to cliticization in Croatian and Spanish, comparing them regarding their success in accounting for clitic acquisition, focusing mainly on clitic placement. Given the results, I argue that clitic acquisition is to a large extent the same and results from the common properties determining cliticization in the two languages. The most relevant property is the independently motivated parameter of verb movement. I further explore the possibility that the language particular differences in cliticization can be attributed to other parameterized properties, including possibly order of adjunction (yielding clitic-verb or verb-clitic order), a second-position, and a verb-adjacency parameter. I argued here that the overall properties of cliticization in child and adult grammar result from the application of linguistic principles and parameters that apply cross-linguistically, and are not exclusive to either of these two languages.

This dissertation is a descriptive account of cliticization in four contemporary Slavic languages--Serbo-Croatian, Bulgarian, Polish, and Czech. The individual chapters provide a basic overview of the inventory and usage of clitic forms in each separate language. Two of the languages are South Slavic (Serbo-Croatian, Bulgarian) and two are West Slavic (Polish, Czech). East Slavic dialects are not systematically treated here because the inventory--and consequently the grammar--of clitics in Russian, Ukrainian, and Belorussian has been radically reduced. The South and West Slavic dialects, on the other hand, still have intricate grammars of cliticization, grammars which lend themselves naturally to an extensive synchronic investigation. Although the dissertation is primarily descriptive, selected comparative observations are also included. Given the fact that little comparative research on Slavic clitics has been carried out (although Slavic clitics themselves have been studied extensively), this descriptive effort seems to represent a legitimate and useful approach. The comparative observations suggest that all of the languages under investigation have a common core, but each displays interesting variations that one might not expect in languages so closely related. In particular, each of the languages which the dissertation treats has some auxiliary and pronominal clitics, the ordering of which is very similar in each language. Language-specific variations include differences in the clitic inventories, minor variations in the ordering of clitics, and some divergence in the use of the interrogative clitic li. Data were collected mainly from published grammars and specialized literature; native speakers of the languages under investigation also contributed invaluable information. The survey focuses primarily on the literary standard, but, in some few selected cases, colloquial data have been included. Mainly for the reasons of space, dialectal data have been kept outside the realm of the investigation. An attempt at a general overview such as that undertaken in this dissertation represents a convenient point of departure for future linguistic research. A map of cliticization may function as a useful basis for further theoretical discussion and for determining the underlying principles of the grammar of cliticization, both in synchrony and diachrony.