Linguistics Theses and Dissertations

Permanent URI for this collectionhttp://hdl.handle.net/1903/2787

Browse

Now showing 1 - 20 of 118

Future reference 'without' future morphology
(2024) Mendes, Jéssica Viana; Hacquard, Valentine; Santorio, Paolo; Linguistics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
In some languages, present morphology can be used to refer to non-scheduled future events. Since this form of future reference is constrained to certain subordinate environments, like conditional antecedents (‘If John gets a new job, he played his cards right’) and relative clauses (‘Everyone who gets invited to this party is very lucky’), I propose to call the phenomenon Subordinate Future (SF). Two factors have hindered our understanding of the SF: First, the SF often occurs in modalized sentences, which makes it difficult to tease apart its contribution from that of the environment. Second, present morphology in English can express several readings; therefore, the appearance of this future is not particularly informative. This dissertation brings new intra- and cross-linguistic evidence to bear on the nature and the meaning of the SF. I observe that, in addition to temporal displacement, the SF also introduces modal displacement. Then, I argue that the source of this modality is a subjunctive mood morpheme, which is silent in English, but pronounced in Portuguese. I proceed to decompose the semantics of the subjunctive. I propose that the subjunctive should be treated as a Heimian indefinite (Heim, 1982) ranging over situations. Simply put, the role of the subjunctive is to introduce a situation variable. The motivation for my proposal comes from the behavior of the subjunctive in relative clauses, and from the anaphoric pattern of sentences with the SF. In relative clauses, the SF blocks a specific reading of the DP. Besides that, the SF seems to be able to ‘bind’ the situation variable of predicates outside of its domain of c-command, giving rise to modal donkey anaphora. These two facts would be difficult to reconcile with a quantificational treatment of the subjunctive. I then turn my attention to the temporal interpretation of the phenomenon. As Crouch (1993, 1994) observed, this future is also able to anchor the temporal interpretation of clauses outside of its domain of c-command. I propose that this effect is a byproduct of modal donkey anaphora, and demonstrate how casting my proposal in terms of situations provides a natural account of the phenomenon. I conclude with a comparison between my proposal and existing accounts.
A world without words: A non-lexicalist framework for psycho- and neuro-linguistics
(2024) Krauska, Alexandra; Lau, Ellen; Linguistics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
In standard models of language production or comprehension, the elements which are retrieved from memory and combined into a syntactic structure are “lemmas” or “lexical items”. Such models implicitly take a “lexicalist” approach, which assumes that lexical items store meaning, syntax, and form together, that syntactic and lexical processes are distinct, and that syntactic structure does not extend below the word level. Across the last several decades, linguistic research examining a typologically diverse set of languages has provided strong evidence against this approach. These findings suggest that syntactic processes apply both above and below the “word” level, and that both meaning and form are partially determined by the syntactic context. This has significant implications for psychological and neurological models of language processing as well as for the way that we understand different types of aphasia and other language disorders. As a consequence of the lexicalist assumptions of these models, many kinds of sentences that speakers produce and comprehend - in a variety of languages, including English - are challenging for them to account for. In order to move away from lexicalism in psycho- and neuro-linguistics, it is not enough to simply update the syntactic representations of words or phrases; the processing algorithms involved in language production are constrained by the lexicalist representations that they operate on, and thus also need to be reimagined. This dissertation discusses the issues with lexicalism in linguistic theory as well as its implications in psycho- and neuro-linguistics. In addition, I propose a non-lexicalist model of language production, the “WithOut Words” (WOW) model, which does not rely on lemma representations, but instead represents that knowledge as independent mappings between meaning and syntax, and syntax and form, with a single integrated stage for the retrieval and assembly of syntactic structure. Based on this, the model suggests that neural responses during language production should be modulated not just by the pieces of meaning, syntax, and form, but also by the complexity of the mapping processes which link those separate representations. This prediction is supported by the results of a novel experimental paradigm using electroencephalography (EEG) during language production, which observes greater neural responses for meaning-syntax and syntax-form mapping complexity in two separate time windows. Finally, I re-evaluate the dissociation between regular and irregular verbs in aphasia, which has been used as supporting evidence for a distinction between the grammar and the lexicon. By training recurrent neural networks and measuring their performance after lesioning, I show that the observed clinical data can be accounted for within a single mechanism. By moving away from lexicalist assumptions, the non-lexicalist framework described in this dissertation provides better cross-linguistic coverage and aligns better with contemporary syntactic theory.
MODELING ADAPTABILITY MECHANISMS OF SPEECH PERCEPTION Nika Jurov
(2024) Jurov, Nika; Feldman, Naomi H.; Idsardi, William; Linguistics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
Speech is a complex, redundant and variable signal happening in a noisy and ever changing world. How do listeners navigate these complex auditory scenes and continuously and effortlessly understand most of the speakers around them? Studies show that listeners can quickly adapt to new situations, accents and even to distorted speech. Although prior research has established that listeners rely more on some speech cues (or also called features or dimensions) than others, it is yet not understood how listeners weight them flexibly on a moment-to-moment basis when the input might deviate from the standard speech. This thesis computationally explores flexible cue re-weighting as an adaptation mechanism using real speech corpora. The computational framework it relies on is rate distortion theory. This framework models a channel that is optimized on a trade off between distortion and rate: on the one hand, the input signal should be reconstructed with minimal error after it goes through the channel. On the other hand, the channel needs to extract parsimonious information from the incoming data. This channel can be implemented as a neural network with a beta variational auto-encoder. We use this model to show that two mechanistic components are needed for adaptation: focus and switch. We firstly show that focus on a cue mimics humans better than cue weights that simply depend on long term statistics as has been largely assumed in the prior research. And second, we show a new model that can quickly adapt and switch weighting the features depending on the input of a particular moment. This model's flexibility comes from implementing a cognitive mechanism that has been called ``selective attention" with multiple encoders. Each encoder serves as a focus on a different part of the signal. We can then choose how much to rely on each focus depending on the moment. Finally, we ask whether cue weighting is informed by being able to separate the noise from speech. To this end we adapt a feature disentanglement adversarial training from vision to disentangle speech (noise) features from noise (speech) labels. We show that although this does not give us human-like cue weighting behavior, there is an effect of disentanglement of weighting spectral information slightly more than temporal information compared to the baselines. Overall, this thesis explores adaptation computationally and offers a possible mechanistic explanation for ``selective attention'' with focus and switch mechanisms, based on rate distortion theory. It also argues that cue weighting cannot be determined solely on speech carefully articulated in laboratories or in quiet. Lastly, it explores a way to inform speech models from a cognitive angle to make the models more flexible and robust, like human speech perception is.
Topics in the Syntax and Semantics of Coordinate Structures
(1993) Munn, Alan Boag; Hornstein, Norbert; Linguistics; Digital Repository at the University of Maryland; University of Maryland (College Park, MD)
This thesis is concerned with developing a syntax for coordinate structures which is compatible with both the syntactic behaviour of conjunction structures and with their semantics. It argues that coordinate structures are asymmetrical, hierarchical structures that conform with X-bar theory. The conjunction head projects a phrase which is adjoined to the first conjunct. This provides an account of a number of syntactic asymmetries in conjunct ordering including agreement and binding asymmetries and provides a principled analysis of Across-the-Board extraction as instances of parasitic gaps. It further argues that the Coordinate Structure Constraint cannot be a syntactic constraint, but rather must be a condition on conjoining identical semantic categories. This provides an account of unlike category coordination which is shown to be freely possible if semantic identity is preserved and no independent syntactic constraints are violated, a result which follows from the adjunct nature of the coordinate structure. In order to account for the semantic identity, it is proposed that at Logical Form, each conjunct is a predicate in an identification relation with the conjunction head, which raises to take scope over all the conjuncts. Assuming theta role assignment at LF, only the conjunction head receives a theta role; none of the conjuncts does. Because each conjunct is in a predication relation with the conjunction head at LF, the semantic identity constraint follows directly. The fact that the conjuncts do not receive a theta role accounts for their inability to act as antecedents for reflexive binding and for fact that modal adverbs can appear inside conjoined NPs. The proposed analysis assimilates coordinate structures directly to plurals, and argues that a consequence of the proposed LF is that all natural language conjunction and disjunction is group forming rather than propositional. All semantic ambiguities between distributed and collective coordination can then be derived with the appropriate logical representation for plurals in general, rather than having a separate semantics altogether for coordination.
GENERATING AND MEASURING PREDICTIONS IN LANGUAGE PROCESSING
(2023) Nakamura, Masato; Philips, Colin; Linguistics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
Humans can comprehend utterances quickly, efficiently, and often robustly against noise in the inputs. Researchers have argued that such a remarkable ability is supported by prediction of upcoming inputs. If people use the context to infer what they would hear/see and prepare for likely inputs, they should be able to efficiently process the predicted inputs.This thesis investigates how contexts can predictively activate lexical representations (lexical pre-activation). I address two different aspects of prediction: (i) how pre-activation is generated using contextual information and stored knowledge, and (ii) how pre-activation is reflected in different measures. I first assess the linking hypothesis of the speeded cloze task, a measure of pre-activation, through computational simulations. I demonstrate that an earlier model accounts for qualitative patterns of human data but fails to predict quantitative patterns. I argue that a model with an additional but reasonable assumption of lateral inhibition successfully explains these patterns. Building on the first study, I demonstrate that pre-activation measures fail to align with each other in cases called argument role reversals, even if the time courses and stimuli are carefully matched. The speeded cloze task shows that “role-appropriate” serve in ... which customer the waitress had served is more strongly pre-activated compared to the “role- inappropriate” serve in ... which waitress the customer had served. On the other hand, the N400 amplitude, which is another pre-activation measure, does not show contrasts be- tween the role-appropriate and inappropriate serve. Accounting for such a mismatch between measures in argument role reversals provides insights into whether and how argument roles constrain pre-activation as well as how different measures reflect pre-activation. Subsequent studies addressed whether pre-activation is sensitive to argument roles or not. Analyses of context-wise variability of role-inappropriate candidates suggest that there are some role-inappropriate pre-activations even in the speeded cloze task. The next study at- tempts to directly contrast pre-activations of role-appropriate and inappropriate candidates, eliminating the effect of later confounding processes by distributional analyses of reaction times. While one task suggests that role-appropriate candidates are more strongly pre- activated compared to the role-inappropriate candidates, the other task suggests that they have matched pre-activation. Finally, I examine the influence of role-appropriate competitors on role-inappropriate competitors. The analyses of speeded cloze data suggest that N400 amplitudes can be sensitive to argument roles when there are strong role-appropriate competitors. This finding can be explained by general role-insensitivity and partial role-sensitivity in pre-activation processes. Combined together, these studies suggest that pre-activation processes are generally insensitive to argument roles, but some role-sensitive mechanisms can cause role-sensitivity in pre-activation measures under some circumstances.
The Learning and Usage of Second Language Speech Sounds: A Computational and Neural Approach
(2023) Thorburn, Craig Adam; Feldman, Naomi H; Linguistics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
Language learners need to map a continuous, multidimensional acoustic signal to discrete abstract speech categories. The complexity of this mapping poses a difficult learning problem, particularly for second language learners who struggle to acquire the speech sounds of a non-native language, and almost never reach native-like ability. A common example used to illustrate this phenomenon is the distinction between /r/ and /l/ (Goto, 1971). While these sounds are distinct in English and native English speakers easily distinguish the two sounds, native Japanese speakers find this difficult, as the sounds are not contrastive in their language. Even with much explicit training, Japanese speakers do not seem to be able to reach native-like ability (Logan, Lively, & Pisoni, 1991; Lively, Logan & Pisoni, 1993) In this dissertation, I closely explore the mechanisms and computations that underlie effective second-language speech sound learning. I study a case of particularly effective learning--- a video game paradigm where non-native speech sounds have functional significance (Lim & Holt, 2011). I discuss the relationship with a Dual Systems Model of auditory category learning and extend this model, bringing it together with the idea of perceptual space learning from infant phonetic learning. In doing this, I describe why different category types are better learned in different experimental paradigms and when different neural circuits are engaged. I propose a novel split where different learning systems are able to update different stages of the acoustic-phonetic mapping from speech to abstract categories. To do this I formalize the video game paradigm computationally and implement a deep reinforcement learning network to map between environmental input and actions. In addition, I study how these categories could be used during online processing through an MEG study where second-language learners of English listen to continuous naturalistic speech. I show that despite the challenges of speech sound learning, second language listeners are able to predict upcoming material integrating different levels of contextual information and show similar responses to native English speakers. I discuss the implications of these findings and how the could be integrated with literature on the nature of speech representation in a second language.
ON SERIAL VERB CONSTRUCTIONS
(1992) Wu, Daoping; Hornstein, Norbert; Linguistics; University of Maryland (College Park, Md); Digital Repository at the University of Maryland
The term serial verb construction (SVC) refers to a construction in which more than one verb are not connected by any lexical device such as conjunction and punctuation, etc.. This construction is quite popular in Chinese, Caribbean creoles, West African languages, and Dravidian languages. Structurally, the SVCs may be compounds, clauses or phrases. The clausal SVCs have been attested in all the serialized languages. The compound SVcs are reported in Chinese and Edo. Only a few instances of phrasal SVCs have been found in Dravidian languages. The compound SVCs in Chinese can consist of two verbs or a verb plus an adjective. The productive compound can only have the following structures: V trans. +V intrans., V trans + A and V intrans. + A. Among the three compounds, the head of the compound must link to both the external and the internal arguments when it is transitive. The nonhead adjective can link to either the external argument or the internal argument, while the nonhead verb can only link to the internal argument, and in most cases, the intransitive verb must be unaccusative. Feature Percolation Convention and the Case requirements generate all the grammatical compound SVCs and rule out all the ungrammatical ones. These rules also can well account for the differences between the VV compound and the VA compound. This conforms to what the Feature Percolation Convention and the Case requirement predict. It is suggested that the syntactic SVCs under discussion are clausal because of facts concerning their binding properties. This analysis is first proposed by Bickerton & Iatridou(1989) in their study of the Caribbean creoles. With this analysis, if the first verb in an SVC is in the matrix clause, the second verb is in an adjunct clause attached to the V' position of the matrix clause. This analysis obtains support from Chinese. In this respect, the only difference between Chinese and the Caribbean creoles is that the adjunct clause has two adjoining positions. Furthermore it is proposed in this thesis that the adjunct clause is a CP rather than an IP.
Island Constraints: What is there for children to learn?
(2022) Hirzel, Mina; Lidz, Jeffrey; Lau, Ellen; Linguistics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
This dissertation presents behavioral studies that target the early syntactic representations of wh-movement during infancy and early childhood. Previous studies show that by 20 months-old, infants represent wh-movement and use this knowledge to respond to wh-questions during language comprehension tasks (Gagliardi 2012; Gagliardi et al., 2016; Seidl et al., 2003). Studies probing the nature of early representations of wh-movement show that by around 4 years-old, children represent island constraints (e.g., de Villiers et al., 1990; de Villiers & Roeper, 1995a, 1995b; Fetters & Lidz, 2016; Goodluck et al., 1992). It remains unclear how knowledge of wh-movement develops. What is the source of this ‘empirical gap’ between the onset of knowledge of wh- movement, and the observation that children respect island constraints? One possibility is that knowledge of island constraints is a component of Universal Grammar (e.g., Chomsky 1965, 1986; Hornstein & Lightfoot 1981). In this case, the ‘empirical gap’ in the linguistic abilities of infants compared to young children isn’t indicative of their linguistic knowledge, but rather the difficulties with testing infants and toddlers on complex syntax. Another possibility is that knowledge of island constraints is acquired via experience (e.g., Pearl & Sprouse, 2013). In this case, the ‘empirical gap’ reflects a knowledge gap, and there’s no evidence for knowledge of island constraints during infancy because it has yet to be acquired. Experiment 1 shows that by 19 months-old, infants have knowledge of wh-movement, and use this knowledge during language comprehension. Results are consistent with recent work which shows that 18 month-olds, but not 17 month-olds, know that wh-phrases co-occur with gap positions in wh-object questions (Perkins & Lidz, 2021). Experiment 2 shows that 3 year-olds respect locality constraints on wh-movement in wh- questions, and Experiment 3 shows that adults behave as expected on this task. Experiments 4 and 5 test children and adults on locality constraints on wh- movement in relative clauses, but these results are inconclusive (likely due to difficulties with moving the task online). The results of Experiment 3 raises challenges for learning hypotheses of island constraints which emphasize the role of linguistic experience. Learning models which propose that linguistic experience is the key factor in the acquisition of island constraints must consider these behavioral results when estimating the amount of data that the learner needs to solve the acquisition problem. These behavioral results are consistent with the hypothesis that knowledge of island constraints is innate, but further work is needed to close the ‘empirical gap’ between the onset of knowledge of wh- movement and the onset of knowledge of island constraints.
What could go wrong? Linguistic illusions and incremental interpretation
(2022) Muller, Hanna; Phillips, Colin; Linguistics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
The systems underlying incremental sentence comprehension are, in general, highly successful - comprehenders typically understand sentences of their native language quickly and accurately. The occasional failure of the system to deliver an appropriate representation of a sentence is therefore potentially illuminating. There are many ways the comprehender's general success could in principle be accomplished; the systematic pattern of failures places some constraints on the possible algorithms. This dissertation explores two cases of systematic failure, negative polarity illusions and substitution illusions (sometimes called "Moses illusions") with the goal of identifying the specific circumstances under which the illusion arises, and, as a consequence, the specific constraints placed on possible implementations of linguistic knowledge. In the first part of this dissertation, I explore the profile of the negative polarity illusion, a case in which a sentence containing an unlicensed negative polarity item and a preceding, but not structurally-relevant licensor is perceived as if it is acceptable, at least in early stages of processing. I consider various proposals for the grammatical knowledge that determines the restricted distribution of negative polarity items, and possible algorithms for using that grammatical knowledge in real time to process a sentence containing a negative polarity item. I also discuss possible parallels between negative polarity illusions and superficially-similar illusory phenomena in other domains, such as subject-verb agreement. Across sixteen experiments, I show that the profile of the illusion is more restricted than previously thought. Illusions do not always arise when an unlicensed negative polarity item is preceded by a structurally-irrelevant licensor, and the circumstances under which they do arise are quite specific. These findings suggest that the negative polarity illusion may be meaningfully distinct from other illusory phenomena, though this conclusion does not necessarily require stipulating a separate mechanism for every illusion. I discuss the implications of these findings for possible real-time implementations of grammatical knowledge. In the second part of this dissertation, I turn to the substitution illusion, a case in which a word in a trivia fact is swapped out for another word, making the sentence a world knowledge violation, but comprehenders do not consciously detect the anomalous nature of the sentence. Here I attempt to develop specific and testable hypotheses about the source of the illusion, paying particular attention to how the same mechanism that "fails" in illusion sentences (in that it does not allow the comprehender to detect the anomaly) serves the comprehender well in other circumstances. I demonstrate that the substitution illusion, like the negative polarity illusion, is more restricted than previously thought - some stimuli yield very high illusion rates while others yield very low illusion rates, and this variability appears to be non-random. In seven experiments, I pursue both a correlational approach and an experimental manipulation of illusion rates, in order to narrow the space of possible explanations for the illusion. These investigations collectively demonstrate that occasional errors in comprehension do not necessarily reflect the use of "shortcuts" in sentence processing, and can be explained by the interaction of the linguistic system with non-linguistic components of the cognitive architecture, such as memory and attention. While neither illusion phenomenon is ultimately fully explained, the research presented here constitutes an important step forward in our understanding of both domains and their broader implications.
Are you asking me or telling me? Learning clause types and speech acts in English and Mandarin
(2022) Yang, Yu'an; Hacquard, Valentine; Lidz, Jeffrey; Linguistics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
Languages tend to have three major clause types (declaratives, interrogatives, imperatives), dedicated to three main speech acts (assertions, questions, commands). However, the particular forms that these clause types take differ from language to language, and have to be learned. Previous experimental results suggest that by 18 months old, children differentiate these clause types and associate them with their canonical speech act. This dissertation investigates how children learn to identify different clause types and speech acts. To learn clause types, children need to identify the right categories of clauses (the "clustering problem") and figure out what speech act they are canonically used for (the "labeling problem"). I investigate the extent to which learners need to rely on pragmatic information (i.e., knowing what speech act a given utterance of a sentence is conveying), to solve not just labeling, but the clustering itself. I examine the role of pragmatics computationally by building two Bayesian clustering models. I find that morpho-syntactic and prosodic information are not enough for identifying the right clause type clustering, and that pragmatics is necessary. I applied the same model to a morphological impoverished language, Mandarin, and found that the model without pragmatics performs even worse. Speech act information is crucial for finding the right categories for both languages. Additionally, I find that a little pragmatics goes a long way. I simulate the learning process with noisy speech act information, and find that even when speech act information is noisy, the model hones in on the right clause type categories, when the model without fails. But if speech act information is useful for clause type learning, how do children figure out speech act information? I explore what kind of non-clause type cues for speech act information are present in the input. Even if children must rely on clause type information to figure out speech acts, they could have access to additional information that is unrelated to clause typing, but informative for recognizing speech act type. When speakers perform speech acts, because of the conventional functions of these speech acts on the discourse, the performance might be associated with certain socio-pragmatic features. For example, because of questions' response-elicitation function, we might expect speakers to pause longer after questions. If children are equipped with some expectations about the functions of communication, and about what questions do, they might be able to use these socio-pragmatic cues to figure out speech act. I explore two cues that could potentially differentiate questions from other speech acts: pauses, and direct eye gaze. I find that parents tend to pause longer after questions, and attend to the child more when asking questions. Therefore it is in principle plausible that there are some socio-pragmatic features that children can use, in addition to their growing knowledge of clause types to infer the speech act category of an utterance. This little bit of information about speech act could then be used to provide the information that the child needs in order to get the clause type clusters identified accurately.
Explorations in Diagnosing Competence and Performance Factors in Linguistic Inquiry
(2022) Liter, Adam; Lidz, Jeffrey; Hornstein, Norbert; Linguistics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
This dissertation presents a series of case studies concerned with whether the signal in a given set of measurements that we take in the course of linguistic inquiry reflects grammatical competence or performance factors. We know that performance and competence do not always covary, yet it is not uncommon to assume that measurements that we take of linguistic performance do transparently reflect the underlying grammatical competence that is the target of inquiry. This has been a very useful and fruitful assumption in the vast majority of cases. Nonetheless, there are certain cases where more careful consideration of the linking hypothesis between the underlying competence of interest and the measurements of linguistic behavior (i.e., performance) that one takes might be warranted. This dissertation presents three case studies that try to model such consideration. How performance and competence might interact is highly dependent on the phenomenon being investigated as well as the method being used to investigate it, so there is no one-size-fits-all approach to these kinds of considerations. The goal of this dissertation is to model such consideration and to encourage more of it. In Chapter 2, we investigate English-acquiring children’s non-adult-like productions of medial wh-phrases. On the basis of experimental data showing a correlation between an independent measure of cognitive inhibition and the production of such examples, we will argue that the best explanation of these productions is that children fail to inhibit the pronunciation of the wh-copy at the intermediate clause boundary due to an underdeveloped executive function and that children do have the target adult-like English grammar with respect to the formation of wh-dependencies (contra, e.g., Thornton 1990, McDaniel, Chiu, & Maxfield 1995, de Villiers, de Villiers, & Roeper 2011). Then, in Chapter 4, we investigate the status of island violations under sluicing (i.e., TP ellipsis). Sluicing apparently improves the acceptability of island violations contained inside the ellipsis site (see, e.g., Ross 1969). Whether we should understood this improved acceptability as indicative of such examples being grammatical is an open question (cf. Ross 1969, Chomsky 1972, Lasnik 2001, Fox & Lasnik 2003, Merchant 2005, 2008b, 2009, Temmerman 2013, Griffiths & Lipták 2014, Barros 2014a, Barros, Elliott, & Thoms 2014, 2015). We investigate the status of such examples with several 2 × 2 experiments, an experimental paradigm discussed in detail in Chapter 3. The idea of the experimental design is to use differences between acceptability ratings and subtraction logic afforded by the linking hypothesis between acceptability and grammaticality to try to more directly get at grammaticality. Our results from this chapter are ultimately somewhat inconclusive, but for potentially methodologically informative reasons. Finally, in Chapter 5, we use the same kind of experimental paradigm to investigate the status of Bulgarian examples with multiple wh-dependencies, where one of the wh-dependencies crosses an island and the other does not. Bulgarian is a language with multiple fronting of wh-elements, and it has been observed that examples where one of the wh-dependencies spans an island but not the other are improved in acceptability (see, e.g., Richards 1997, 1998, 2001). Such examples have thus been taken to be grammatical, though they do still exhibit some degree of unacceptability. We use the same sort of experimental paradigm to try to ascertain the grammaticality status of these examples. We find evidence that such examples are indeed grammatical, which reaffirms the importance of ensuring our syntactic theories can account for such examples.
All about alles: The syntax of wh-quantifier float in German
(2021) Doliana, Aaron Gianmaria Gabriel; Lasnik, Howard; Hornstein, Norbert; Linguistics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
This thesis offers an in-depth investigation of “wh-quantifier float” of the quantifying particle ‘alles’ in German. 'Alles' (etymologically, ‘all’) appears in wh-questions like 'Wen alles hat die Mare eingeladen?' (‘Who-all did Mare invite?’). The thesis focuses on the syntactic distribution of 'alles'. 'Alles' enjoys a wide distribution in the clause. It can occur both ‘adjacent’ to its ‘associate’ wh-phrase, and ‘distant’ from it, in various positions of the clause. I address three questions: What determines the distribution of 'alles'? Are adjacent 'alles' and ‘distal alles’ the same category? What licenses distal 'alles'? I answer these questions by arguing for a stranding analysis of distal 'alles': 'alles' and its associate form a first-Merge constituent, which is optionally separated in the course of the derivation through a process that involves movement ([WH alles] ⇒ [WH. . . [[WH alles]. . . ]]). The conclusion is compatible with prior analyses that argued for or assumed (a) constituency, and (b) a movement dependency in overt syntax. The conclusion is at odds with adverbial analyses, which assume that distal 'alles' is an adverbial. I provide two main empirical arguments. First, I argue against the idea that distal 'alles' and adjacent 'alles' are separate lexical items, or have different lexical content. Second, I argue that the “Chain Link Generalization” is the most accurate generalization for the distribution of 'alles': Given a derivation involving 'alles' and a licit associate, 'alles' may appear in any position which hosts an Abar-chain link of the associate, and in no other position. I show that 'alles' has “no distribution of its own in the clause”. Rather, the distribution of 'alles' depends on the potential distribution of its associate and can be predicted by the associate’s category, the associate’s base-position, the derivation that the associate undergoes in a given sentence. Conceptually, I argue that a stranding analysis is favored by simplicity as most generalizations established in this dissertation are directly entailed by it.
Phasehood and Phi-Intervention
(2021) Thivierge, Sigwan; Preminger, Omer; Polinsky, Maria; Linguistics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
This dissertation investigates the notion of phases in syntactic theory, and offers a reanalysis of certain phases as instances of phi(φ)-intervention. Under the standard view, phases are syntactic structures that, according to the Phase Impenetrability Condition, are opaque to operations originating outside of the phase (Chomsky, 2000; 2001). Phasehood was linked to certain heads such as C and (transitive) v, but several issues arise once the empirical domain is broadened beyond English. As more work has turned to unrelated languages, a less stipulative alternative has presented itself: phases are intervention effects, and are reducible to a more general locality issue. In Rackowski and Richards’ (2005) account of Tagalog wh-movement, for example, CPs act as phases because they constitute the closest goal. In Halpert’s (2019) account of Zulu hyper-raising, CPs do not act as phases due to the cyclic nature of Agree. In Keine’s (2017) account of Hindi long-distance agreement, vPs do not act as phases, which I argue is because v is not a φ-goal. In Georgian, vPs act as phases because v0, in contrast, is a φ-goal (as I will argue in this thesis). These languages show that XPs act as phases only when they are potential goals for a syntactic operation. These languages also illustrate two ways of diagnosing phasehood as φ-intervention: via movement out of the domain, and via agreement into the domain. These results suggest that phasehood is an epiphenomenon, and that the interior of the ‘phase’ is accessible even after the phase is complete. In this dissertation, I argue that certain instances of phasehood derive from the ‘phase’ head bearing a φ-probe: the φ-features on the probe intervene for φ-agreement, which results in phase-like effects. The empirical data in favour this claim comes from the Georgian agreement system. I show that subjects in Georgian are base-generated in different positions, depending on whether they fall under the basic agreement paradigm or the inverse agreement paradigm. In the basic, subjects are introduced above v and are the closest goal for Agree operations that originate outside the vP domain. In the inverse, subjects are introduced below v; in this case, the φ-features that are associated with the φ-probe on v constitute the closest goal for Agree.
Identity Conditions on Ellipsis
(2021) Ranero Echeverría, Rodrigo; Polinsky, Maria; Preminger, Omer; Linguistics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
This dissertation presents a new perspective on the identity condition underpinning ellipsis in natural language. It argues that the condition is irreducibly syntactic—at least in part—but the way this syntactic component works is different than previously thought. First, instead of simple identity of structures/features, the condition relies on non-distinctness. For example, a privative feature present in the antecedent but not in the ellipsis site (or vice-versa) does not constitute a violation of identity. Nor does a functional projection present in one but not the other. Second, the identity condition includes a component that pertains to √ROOTs. Unlike the component requiring featural non-distinctness, √ROOTs in the ellipsis site and the antecedent must be strictly identical. After providing an overview of the core research questions surrounding ellipsis, the dissertation builds its initial case in chapter 2 on the basis of novel data from Kaqchikel (Mayan). In contrast to the pattern familiar from languages like English, Kaqchikel bans certain voice mismatches under sluicing, but allows others. To account for that, I argue that clauses in the Agent Focus voice—which can mismatch with active and passive clauses—lack the VoiceP layer. The proposed identity condition which relies on non-distinctness captures this newly-established pattern. The empirical scope is expanded in chapter 3, where I consider mismatches above VoiceP in several languages. I show that the proposed identity condition can account for the observed generalizations regarding tense, polarity, illocution, and modality mismatches, which remain unexplained under other proposals. Chapter 4 zooms into the nominal domain and discusses mismatches in grammatical gender under nominal ellipsis in argument and predicate positions. I present cross-linguistically recurrent patterns of well-formed and ill-formed mismatches and argue that the proposed identity condition (coupled with the independently motivated mechanism of repair-by-ellipsis of morphophonological gaps) is necessary and sufficient to account for the attested patterns. I also argue that certain configurations satisfy the identity condition but are ill-formed for other reasons; in particular, ellipsis cannot repair encyclopedic gaps. Extensions of the proposal are discussed in chapter 5, including voice mismatches under sluicing in Austronesian languages, Chung’s generalization, and vehicle change phenomena.
Finding modal force
(2021) Dieuleveut, Anouk; Hacquard, Valentine; Williams, Alexander; Linguistics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
This dissertation investigates when and how children figure out the force of modals, that is, when and how they learn that can/might express possibility, whereas must/have to express necessity. Learning modal force raises a logical “Subset Problem”: given that necessity entails possibility, what prevents learners from hypothesizing possibility meanings for necessity modals? Three main solutions to other Subset Problems have been proposed in the literature. The first is a bias towards strong (here, necessity) meanings, in the spirit of Berwick (1985). The second is a reliance on downward-entailing environments, which reverse patterns of entailment (Gualmini & Schwarz, 2009). The third is a reliance on pragmatic situational cues stemming from the conversational context in which modals are used (Dieuleveut et al., 2019). This dissertation assesses the viability of each, by examining the modals used in speech to and by 2-year-old English children, through a combination of corpus studies and experiments testing the guessability of modal force based on their context of use. I show that negative and other downward-entailing contexts are rare with necessity modals, making them impractical on their own. However, the conversational context in which modals are used in speech to children is highly informative about both forces. Thus, if learners are sensitive to these conversational cues, they, in principle, do not need to rely either on a necessity bias nor on negative environments to solve the Subset Problem. Turning to children’s own productions, I show that children master possibility modals very early: by age 2, they use them productively, and in an adult-like way. However, they struggle with necessity modals: they use them less frequently, and not in an adult-like way. Their modal uses show no evidence for a necessity bias. To assess how children actually figure out modal force, and which of the available cues children use to figure out modal force, I then examine which aspects of children’s input best predict their mastery of modals. Preliminary results suggest that negation is predictive of children’s early success with necessity modals, and that frequency of modal talk, but not of particular lexemes, also contributes to their early success.
The Psycho-logic of Universal Quantifiers
(2021) Knowlton, Tyler Zarus; Lidz, Jeffrey; Linguistics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
A universally quantified sentence like every frog is green is standardly thought to express a two-place second-order relation (e.g., the set of frogs is a subset of the set of green things). This dissertation argues that as a psychological hypothesis about how speakers mentally represent universal quantifiers, this view is wrong in two respects. First, each, every, and all are not represented as two-place relations, but as one-place descriptions of how a predicate applies to a restricted domain (e.g., relative to the frogs, everything is green). Second, while every and all are represented in a second-order way that implicates a group, each is represented in a completely first-order way that does not involve grouping the satisfiers of a predicate together (e.g., relative to individual frogs, each one is green).These “psycho-logical” distinctions have consequences for how participants evaluate sentences like every circle is green in controlled settings. In particular, participants represent the extension of the determiner’s internal argument (the cir- cles), but not the extension of its external argument (the green things). Moreover, the cognitive system they use to represent the internal argument differs depend- ing on the determiner: Given every or all, participants show signatures of forming ensemble representations, but given each, they represent individual object-files. In addition to psychosemantic evidence, the proposed representations provide explanations for at least two semantic phenomena. The first is the “conservativity” universal: All determiners allow for duplicating their first argument in their second argument without a change in informational significance (e.g., every fish swims has the same truth-conditions as every fish is a fish that swims). This is a puzzling gen- eralization if determiners express two-place relations, but it is a logical consequence if they are devices for forming one-place restricted quantifiers. The second is that every, but not each, naturally invites certain kinds of generic interpretations (e.g., gravity acts on every/#each object). This asymmetry can po- tentially be explained by details of the interfacing cognitive systems (ensemble and object-file representations). And given that the difference leads to lower-level con- comitants in child-ambient speech (as revealed by a corpus investigation), children may be able to leverage it to acquire every’s second-order meaning. This case study on the universal quantifiers suggests that knowing the meaning of a word like every consists not just in understanding the informational contribu- tion that it makes, but in representing that contribution in a particular format. And much like phonological representations provide instructions to the motor plan- ning system, it supports the idea that meaning representations provide (sometimes surprisingly precise) instructions to conceptual systems.
Coordination without grammar-internal feature resolution
(2021) Lyskawa, Paulina; Polinsky, Maria; Preminger, Omer; Linguistics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
Morphological agreement with coordinate phrases involves a computation that takes as its input a set of features from each conjunct and outputs exactly one resolved set of features (number, person, grammatical gender/noun class, commonly labelled phi-features). Such resolution is typically taken to be grammar-internal because it relies on other grammar-internal ingredients (phi-features, agreement, coordination), and at least in some instances seems to follow systematic rules that may be captured by familiar grammatical operations. Exceptions to these apparent rules, if not ignored altogether, have received disparate analyses depending on the language, framework, and the particular features involved. In this thesis, I argue that it is such exceptions that are illuminating, and that the appearance of rigid rules is misleading. Treating variation in agreement with coordinate phrases as exceptional with respect to the otherwise deterministic output rules either delays the task of explaining the surface data, or risks weakening the language competence theory by adding the baroque stipulations that a purely grammar-internal treatment would require. Phi-agreement with coordinate phrases is subject to inter- and intra-speaker variability and ineffability; such variation is widespread in the world’s languages, even the ones with limited phi-agreement morphology like English. I therefore reject the grammar-internal approach to agreement with coordinate phrases and argue instead that the agreement morphology we observe on the surface is due to grammar-external mechanisms being recruited to determine the resulting agreement morphology. Under this approach, systematicity in agreement with coordinations is only apparent and can be manipulated. The reason a grammar-internal mechanism is unavailable is because it would have to take place on the agreeing head (e.g., Infl0 or v0), and what we know about agreement between a syntactic head and multiple arguments (e.g., from omnivorous agreement and the Person Case Constraint) renders it ill-suited for the task of coordination resolution.
MODALS AND THEIR COMPLEMENTS IN DUTCH AND BEYOND
(2020) van Dooren, Annemarie; Hacquard, Valentine; Polinsky, Maria; Linguistics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
In this dissertation I investigate the syntax and semantics of modals like can and must and their counterparts in other languages. Modals like can and must can be used to express both obligations (as in employees must wash their hands, called deontic modality) and possibilities given what is known (as in John must be home; his car isn't in the parking lot, called epistemic modality), but previous work shows that the availability of these different 'flavors' of modality are constrained by their syntactic environment. My main claim is that in all languages discussed, modal meanings are specifically restricted by their complement size. For English modals, which are treated as functional items that are part of the functional projections from the verb, this is often captured by having modals appear in different positions in the functional projection of the verb based on their modal flavor: Epistemic modals are located high, above tense, while non-epistemic modals, such as deontics, are located low, below aspect (Cinque 1999, Hacquard 2006, 2008, a.o.). I argue that in Dutch, modals are verbs (following Aelbrecht 2010), and as such, they host their own functional projections. Despite this, some of the same syntactic restrictions on the availability of modal flavor hold, which argues for a recasting of the cross-linguistic generalizations not in terms of position of the functional projection, but in terms of complement size. I claim that cross-linguistically, different flavors require different types of complements: epistemics need a complement the size of a Tense Phrase (in line with Cinque 1999, Hacquard 2006), deontics need the size of an Aspectual Phrase (building on Rubinstein 2012), while other non-epistemics can combine with a smaller-sized complement. I will provide two case studies in favor of the claim that complement size restricts the availability of modal flavors: In chapter 3, I will discuss the interaction between tense and modality, and in chapter 4, I will discuss the case of modals with non-verbal complements.
THE COMPUTATION OF VERB-ARGUMENT RELATIONS IN ONLINE SENTENCE COMPREHENSION
(2020) Liao, Chia-Hsuan; Lau, Ellen; Linguistics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
Understanding how verbs are related to their arguments in real time is critical to building a theory of online language comprehension. This dissertation investigates the incremental processing of verb-argument relations with three interrelated approaches that use the event-related potential (ERP) methodology. First, although previous studies on verb-argument computations have mainly focused on relating nouns to simple events denoted by a simple verb, here I show by investigating compound verbs I can dissociate the timing of the subcomputations involved in argument role assignment. A set of ERP experiments in Mandarin comparing the processing of resultative compounds (Kid bit-broke lip: the kid bit his lip such that it broke) and coordinate compounds (Store owner hit-scolded employee: the store owner hit and scolded an employee) provides evidence for processing delays associated with verbs instantiating the causality relation (breaking-BY-biting) relative to the coordinate relation (hitting-AND-scolding). Second, I develop an extension of classic ERP work on the detection of argument role-reversals (the millionaire that the servant fired) that allows me to determine the temporal stages by which argument relations are computed, from argument identification to thematic roles. Our evidence supports a three-stage model where an initial word association stage is followed by a second stage where arguments of a verb are identified, and only at a later stage does the parser start to consider argument roles. Lastly, I investigate the extent to which native language (L1) subcategorization knowledge can interfere with second language (L2) processing of verb-argument relations, by examining the ERP responses to sentences with verbs that have mismatched subcategorization constraints in L1 Mandarin and L2 English (“My sister listened the music”). The results support my hypothesis that L1 subcategorization knowledge is difficult for L2 speakers to override online, as they show some sensitivity to subcategorization violations in offline responses but not in ERPs. These data indicate that computing verb-argument relations requires accessing lexical syntax, which is vulnerable to L1 interference in L2. Together, these three ERP studies allow us to begin to put together a full model of the sub-processes by which verb-argument relations are constructed in real time in L1 and L2.
Investigations on salvation and non-salvation by deletion
(2020) Ribeiro Mendes Junior, Gesoel Ernesto; Howard, Lasnik; Hornstein, Norbert; Linguistics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
I use salvation and non-salvation by deletion as a window to understand computational and lexical resources in natural languages. I revisit previous findings from the literature and present novel data of repair under ellipsis from Polish and Nupe. Salvation by deletion is then used to analyze verb-echo answers in Brazilian Portuguese, which, I show, require a word order that is not available in the language. The phenomenon is also used to compare two types of approaches to phasal domains with date from Nupe. I show that salvation by deletion does not obtain in cases of intervention in A-movement and head movement locality, which implies that these are derivational constraints. Finally, salvation and nonsalvation by deletion are used as a way to distinguish two types of lexical gaps with data from Brazilian Portuguese, Russian and English.

Browse

Recent Submissions