Institutional Scholarship :: Browsing by Author "Sanders, Nathan"

Browsing by Author "Sanders, Nathan"

Now showing 1 - 20 of 27

Applying Translation Theory to a Corpus-Based Analysis of The Count of Monte Cristo
(2014) Bao, Lisa Y.; Sanders, Nathan
Alexandre Dumas pere's well-known 19th-century novel, Lecomte de Monte-Cristo [The Count of Monte Cristo], has been twice translated from the original French into English, in 1846 and again in 1996. In this thesis, I analyze corpora composed of French and English editions of Dumas's novel through the lens of modern translation theory. I use quantitative and qualitative methods drawn from corpus-based translation studies to conduct both a word frequency analysis and a theoretically grounded close reading of selected passages, ultimately finding evidence for several hypotheses about translational strategies in English.
Computational Generation of Referring Expressions Containing Relational Descriptions
(2013) Abernathy, Celia; Sanders, Nathan
Referring expressions like the tall red cup are used to identify entities in many different language tasks, so referring expression generation (REG) is a crucial part of computational natural language generation. One of the challenges of REG is relational descriptions that refer to an object by relating it to some other object, as in the cup on the table. This paper presents a new computational algorithm, DR-REG, for referring expression generation using relational descriptions. DRREG generates relational referring expressions for complex objects by searching through syntactic and semantic information for a suitable expression. DR-REG is evaluated in a controlled experiment that shows its referring expressions lead readers to correctly identify a target object.
A Corpus Study of Conjoined Comparatives
(2013) Coke, Reed; Sanders, Nathan
In this work, I examine the behavior of conjoined comparatives and uncover some of the properties behind these structures. To the best of my knowledge, conjoined comparatives have not received attention in the literature and this thesis is a step towards filling this gap. Conjoined comparatives exhibit a number of preferences that derive from their commonalities with adjective sequences. Many of the factors that influence the ordering of adjectives in a sequence also apply to conjoined comparative, though in a different fashion. The differences between adjective sequences, adjective conjunctions, and comparative sequences are also examined, as they help isolate the effects of the conjunctive and comparative aspects of conjoined comparatives.
Correlations Between Sentiment Analysis of Movie Tweets, Film Critics Reviews, and Box Office Earnings
(2014) Bhattasali, Shohini; Sanders, Nathan
This thesis explores correlations between sentiment analysis ratings from movie tweets, film critics' ratings, and box office earnings. Two sentiment analysis tools, NLTK and Pattern, are used to collect and automatically extract the sentiments behind tweets about five movies released during summer 2013: Despicable Me 2, The Bling Ring, Man of Steel, Monsters University, and World War Z. The average ratings for each movie are compared to the scores for each movie on Metacritic and Rotten Tomatoes. Overall, for most of the movies the results from the sentiment analysis ratings and film critics were similar, although in some instances they did vary considerably. Additionally, Pattern corresponded better with the film critics' ratings than NLTK. Furthermore, through graphical and statistical analysis, correlations between these sets of ratings and the box office earnings are studied. The analysis did not result in significant correlations to make concrete conclusive predictions.
Despite the great distance, existence unites the two: Translating the poetry of Fadwa Tuqan
(2012) Huntington, Franklin; Sanders, Nathan
English Loans in German and the Borrowing of Meaning
(2012) Lane, Mary; Sanders, Nathan
There are many English loans in German, which fall under the categories of (i) phonetic and semantic borrowings (e.g., Computer 'computer', downloaden 'to download'), (ii) phonetic borrowings (e.g., Handy 'cell phone', Long-Drink 'mixed drink'), and (iii) semantic borrowings, or calques (e.g., Sinn machen 'to make sense'). All of these loans are naturalized into German and undergo phonological and morphosyntactic changes. I provide an overview of the integration process that loans undergo and a description of how these words appear in German. When words are borrowed into German from English, secondary or metaphorical meanings of the English are not necessarily borrowed. Most often, the primary, literal meaning of the English word is the meaning the loan takes on, e.g., the loan Baby means 'baby, infant' in German, but not 'coward, wuss,' as it can in English. I report the findings of a survey of native German speakers, investigating the borrowing of meanings of English words. Through the examination of seven loans and the comparison of their meanings in English and German, I show that generally, only the primary meaning of an English word is transferred to a loan in German.
An Examination of American Pediatric Policy on Language for Deaf and Hard of Hearing Children
(2014) Bristow, Sarah; Sanders, Nathan
When natural language input is not accessible to a child during the critical period for language acquisition, linguistic deprivation can occur. This can result in numerous negative effects on the child, including an inability to use any language at the level of a native speaker and subsequent social and emotional deficits and decreased abilities in other cognitive areas. Many children who are deaf or hard of hearing cannot adequately access spoken language for their cognitive needs to be met during the critical period. A sign language, however, can provide accessible natural language input and allow a deaf or hard of hearing child to avoid linguistic deprivation, and should therefore be considered medically necessary for members of this population. This thesis seeks to examine the treatment of language options for deaf and hard of hearing children in pediatric public policies set forth by the American Academy of Pediatrics (AAP). While the AAP does emphasize the importance of linguistic abilities for deaf and hard of hearing children, their policies do not adequately present the possible consequences of linguistic deprivation in a speech-only environment, implying that language options are purely a parental choice that will have no detrimental effects on the child's development.
Finding Remo: A Preliminary Phonetic Analysis of the Language
(2012) Cheng, Andrew; Sanders, Nathan
The Remo language is a Munda language in the Austro-asiatic family that is spoken by the Remo tribe of eastern India. With fewer than nine thousand native speakers, Remo is one of the highlights of what the Living Tongues Institute has classified as the Greater South Asia Language Hotspot. Between 1950, when Remo was first extensively written about, and the present day, very little attention has been paid to this community. In 2005, the Living Tongues Institute for Endangered Languages began a project to document Remo, to increase both awareness of and access to Remo on the Internet. As a part of this project, I was drawn to studying the unique aspects of this language. There is a paucity of published sources and recent linguistic research on the language of Remo. What it really needs is a complete and up-to-date reference grammar and lexicon, which will move it beyond the status of an under-documented language and into the common cognizance of the linguistic community. This thesis makes progress toward this goal by filling in gaps left by previous work on Remo in the areas of its phonetics and phonology. My hope is that others may use it in the future for the purposes of investigating Munda phonology, writing the aforementioned Remo reference grammar, or even learning how to speak this hidden treasure of a language. 1 Introduction
Heritage Languages Among South Asian Americans
(2014) Doraiswamy, Karuna; Sanders, Nathan
This thesis explores second generation South Asian Americans' relationships with their heritage languages; specifically, the ways in which these relationships might challenge current convention regarding the categorization of heritage and non-heritage learners along a linear spectrum. This thesis also examines the extent to which the decision to (re)leam one's heritage language might be considered a necessary step to earning cultural legitimacy in the diasporic community, a symbolic reclamation of one's history, or even a performative act intended to better match oneself to the dominant perception of what it means to be South Asian.
Historical phonology should be conducted with a basis in phonetics
(2016) Zhou, Z.L.; Sanders, Nathan
In this thesis, I demonstrate that Articulatory Phonology, a theory of phonological representation, can be fruitfully combined with Optimality Theory, currently the most widespread theory of phonological processes, to describe sound change. I argue that this combination is suited for explaining sound change because it is firmly grounded in the physical movements of the mouth - meaning that only those sound changes which are natural can occur. As an example of this approach, I analyze the development of Icelandic preaspirated stops from Old Norse unvoiced geminates as an example of constraint promotion and interaction, and find that just a few, wholly natural constraint rankings are necessary to produce the typologically rare subsystem that can be observed in the Icelandic oral stops.
'I Am a Perpetual Underdog': Lady Gaga's Use of Creaky Voice in the Construction of a Sincere Pop Star Persona
(2016) Esposito, Lewis; Sanders, Nathan
Creaky voice. a vibratory phonation produced when the cartilage from the front and back of the larynx are pressed closely together (Zirnrnan 2013), has received significant media attention in recent years, particularly for its prevalence in the speech of female American pop stars, such as Kesha, Brimey Spears, and Lady Gaga. Recent studies (Yuasa 2010; Podesva 2013) have found that the phonation is more pervasively employed by American women than men, which is in contrast to earlier studies showing its associations with upper-class males in the UK (Henton-Bladon 1988; Esling 1978; Stuart-Smith 1999). Ye~ until recently (Zirnrnan 2015; Levon 2015), few smdies have extensively analyzed its potential social meanings from an intraspeaker perspective. Additionally, none examine its usage among the women noted above. Firmly situated within the realm of third-wave intraspeaker variation studies, this thesis considers Lady Gaga's use of creaky voice across four different speech samples - two interviews and two speeches - to discern her possible stylistic usage of the phenomenon. ] find that the calculated percentage of creak is highest in the most personal of the samples and lowest in one of the speeches. Considering these results along with the conversational contexts in which creak is most prevalent, I argue that she employs creaky voice to take an affective stance of intimacy and convey character traits of openness and sincerity. Further, I argue that these meanings arise out of creak's iconicity with a 'contained' emotional state. This thesis (1) adds to the growing body of literature examining the indexical meanings of creaky voice in American English, (2) offers evidence for stance-based persona construction, and (3) emphasizes the importance of considering variation in interaction.
Improving Langscape's Text-based Language Identification Tool
(2016) Koukoutchos, Melina A.; Sanders, Nathan
Text-based language identification (LID) is the task of determining the language a piece of text is written in. Although modem LID tools achieve high accuracy using the widely-accepted n-gram method, there are several areas of LID that remain more difficult, particularly the task of distinguishing between closely related languages. Langscape, a project of the University of Maryland's Language Science Center, has an LID tool that uses a variation on the n-gram method. In this thesis, 1 propose and test a modification to Langscape's LID tool to improve its ability to distinguish between closely related languages.
Intraspeaker Variation: Stop Devoicing in Elderly Black Speakers
(2016) Weissler, Rachel Elizabeth; Sanders, Nathan
The current study looks at intraspeaker variation in the usage of final stop devoicing in elderly speakers of African American Vernacular English. The speakers were interviewed in one-on-one situations and a group-style situation. The results indicated that participants used final stop devoicing significantly more when they were in a group interview than when they were in a one-on-one interview. In addition to these findings, one of the speakers exhibited apparent differences between quoted speech and speech "as self." However, upon analysis, the difference in usage of final devoicing by itself was not statistically significant. While the feature alone may not be statistically significant, it could be a part of a larger pattern that is significant. Finally, when confronted with questions about language attitudes, participants varied in apparent metalinguistic awareness, discussing educational status as being a factor in how they were treated earlier in life, with one participant asserting that they recognize situations where they need to "speak well," and change their speech accordingly.
Kanza and Osage: Language Materials, Revival and the Necessity for Phonetic Analysis
(2013) Sear, Victoria; Sanders, Nathan
In this thesis I explore the possibility, and practicality, of revival of Kanza, a dead Siouan language from Oklahoma. In particular, I examine the pedagogical language materials that have been created via existing documentation of the language by focusing on the three most comprehensive materials created to date: a reader, a board game and a dictionary. This project, inspired by my internship at the Kaw Nation Language Department (KNLD), was conducted by examining these materials, interviewing individuals involved in their creation and referencing existing literature on language revitalization. In addition to examining pedagogical materials produced by the KNLD I address whether they can appeal to both a community- and academia- focused audience at the same time; to do this I reference Carolyn Quintero's Osage Grammar (2004) and Osage Dictionary (2009) that tries to tread this line. I then segue into an explanation of how phonetic analysiswhich Quintero was never able to conduct-will add to the corpus of Osage documentation, the creation of further pedagogical materials and new modes of classroom instruction. I then conclude with my own preliminary phonetic analysis of the vowels /0 i 0 u e/ that occur in Osage.
Lexical Acculturation in Siletz Dee-ni
(2012) Johnson, Jennifer; Sanders, Nathan
Lexical acculturation (the creation of new words as a result of contact with other cultures) in the highly endangered Oregon Athabaskan2 language of Siletz Dee-ni has been primarily influenced by contact with English since European arrival in the Pacific Northwest and California at the beginning of the 19th century. Indian removal from southwest Oregon to the Coast Reservation following the Rogue River War of 1850-1856 profoundly influenced the development of the Siletz Dee-ni language. Siletz Dee-ni exhibits a preference for basing new lexical entries on native linguistic resources (Campbell and Grondona 2011), though it exhibits both primary and secondary forms of lexical acculturation as described by Brown (1996). Whether Siletz Dee-ni speakers choose to adopt a loanword, extend an existing Athabaskan term, or create a morphemeor nominal-based neologism for a new lexical item is consistent with naming patterns displayed by pre-contact words. Siletz Dee-ni speakers have made good use of the language’s affixally polysynthetic morphology: for instance, creating the word me’-naa-draa-‘a’ ‘telephone (lit. inside/into one speaks)’, in the same spirit as the native word me'-drvlh-t’es ‘cookhouse (lit. inside one cooks)’, through a passivization of the verb and the use of the preposition me’. In articulating the process involved in lexical acculturation in Siletz Dee-ni, it is my goal to provide a resource to this speech community and support members’ efforts to create new words in Siletz Dee-ni in ways that are both practical and culturally authentic. 2 Alternately
Lexical Variation in Italian Sign Language
(2012) Silver-Swartz, Amira; Sanders, Nathan
Italian Sign Language, or LIS, is a language rich in lexical differences. One person might sign AUGUST by passing a thumb across their brow; someone else might articulate it as two hands fanning the signer. There is anecdotal evidence that lexical variation is especially rich in the month and color signs; this thesis seeks to record and catalog the variations collected from 12 signers during a reunion at the Tommaso Pendola School in Siena, Italy. The similarities and differences among variants for a single concept/sign (for instance, RED) are discussed. The background of LIS and Italian deaf education is important to this research. The trends in European Deaf education are explored, starting with the Spanish monk Pedro Ponce de Leon and the French Abbé de L’Epée. The influence of the Milan Congress’s decision that oralism, not the teaching of sign language, was the preferred method for teaching deaf students, and the state of deaf education in Italy are both described in order to give some sociological context to the data; the influence of oralism also accounts for the common practice of mouthing parts of Italian words as the LIS sign is being articulated. The historical changes in the morphology and phonology of LIS are also described in order to pick out trends in the data; specifically, the tendency for signs to move from having contact on the face or hands to being signed in neutral space is mentioned. All of these are preludes to the core of the work, which is contained in the variation charts for each month or color. Several signs are highlighted and discussed in particular because of their interesting behavior; some variants for the signs PURPLE and GREEN are mentioned because they are similar to each other. Both use initialization of the letter V as their iconic root. The influence of the Church and other sociolinguistic factors in variant form and choice is not to be ignored; there is strong evidence that cultural references factor into the iconic roots of several signs, notably the Christmas-themed variants of DECEMBER. The data is analyzed using several statistical methods to examine sociolinguistic trends, paying particular attention to how many times a pair of interviewees uses the same sign variants. The geographic origin of the interviewees is shown to have some statistically significant effect on their lexical choices, and Tuscan interviewees are the only ones who use certain signs, for example in the case of ORANGEB. Overall, this thesis is an exploration of the extent that LIS varies in this specific set of signs, celebrating the beauty and complexity of the language.
Morphology of the Maung Language
(2016) Katz, Laura B.; Sanders, Nathan
This thesis aims to elucidate phenomena in the morphophonology of Maung, an Australian language with an estimated 260 speakers. Using cross-linguistic typological patterns as a framework, this paper discusses phonological phenomena that occur when two-consonant and two-vowel clusters form across a morpheme boundary. Major topics include consonant cluster reduction, sonority of onsets, and nasal place assimilation. Additionally, phonological processes are prioritized in a manner conducive to future Optimality Theoretic analysis.
A Phonetic Exploration of the English of Portland, Maine
(2014) Ryland, Alison; Sanders, Nathan
Linguists in the United States have been interested in the dialects of English-speakers from Maine for many decades (e.g., Perkins 1927, Heffner 1938, Lambert 1976), but few researchers have focused exclusively on Maine in more recent years. To contribute to this research, I studied the speech of residents of Maine currently living in the Portland area. Filling this gap in the literature is important because more recent data on the speech of Maine English is needed to provide a current view of the linguistic situation of Maine as well as a more complete geographic view of New England speech as a whole. I designed my study to be a survey of the current state of English in Portland, Maine. I focused on the phonetics of Maine residents, particularly the nasal short-a system, the NORTH-FORCE distinction, and the LOT-THOUGHT merger. Using these characteristics as a jumping-off point, I compared the general findings of Labov et al. (2006) and Nagy and Roberts (2008) to the results of my own speakers in a synchronic analysis of the current overall characteristics of Portland speech. While my analysis finds both similarities and differences between the current speech of Portland and what prior research suggests, it is my hope that this exploration will serve as a starting point for further indepth research on how people speak in Portland, Maine.
Pluralization in German Sign Language
(2012) Herbert, Marjorie; Sanders, Nathan
Like many other signed and spoken languages, German Sign Language (Deutsche Gebärdensprache; DGS) makes use of multiple strategies for the plural marking of nominal signs. The plural marker is realized in three ways. The first default realization is lateral reduplication, in which the sign is reduplicated as the hands move laterally through the signing space. The second is simple reduplication in the same position in space. The third case is zeromarking, with no overt realization of the plural marker. The realization of the plural depends critically on the phonological properties of the base sign, making this a case of phonologically triggered allomorphy. However, as is the case for almost all sign languages, DGS can make use of classifier constructions in conjunction with these nouns. Classifier constructions are available to all noun classes in DGS for a variety of uses, but in the case of underspecified nouns, or the nouns that display zero-marking in the plural, a laterally reduplicated version of the classifier handshape is also available. I argue that this type of classifier is more grammatically regular in its use than it is in other classifier constructions, which suggests that it is being used as an alternative pluralization strategy. I offer a detailed description of the criteria that divides nouns into phonological categories and how these nouns can be alternatively pluralized by means of classifier constructions and spatial localization, a phenomenon in which the articulation of the sign both introduces the noun into the discourse and designates it as an entity within the signing space.
Problems in Bantawa Phonology and a Statistically Driven Approach to Vowels
(2016) Vogel, Rachel; Sanders, Nathan
This thesis examines several aspects of the phonology of Bantawa, an endangered and fairly understudied Tibeto-Burman language of N epa!. I provide a brief review of the major literature on Bantawa to date and discuss two particular phonological controversies: one concerning the presence of retroflex consonants, and one concerning the vowel inventory, specifically whether there are six or seven vowel phonemes. I draw on data I recorded from a native speaker to address each of these issues. With regard to the latter, I also provide an in-depth acoustic analysis of my consultant's 477 vowels and consider several types of statistical models to help address the issue of the number of vowel contrasts. My main conclusions, based on the data from my consultant, are first, that there is evidence based on minimal pairs for a contrast between retroflex and alveolar stops, and second, that there is no clear evidence for a seven-vowel system in Bantawa. With regard to the latter point, additional avenues of research would still be needed to explore the possibility of allophonic variation and/or individual speaker differences.

Browsing by Author "Sanders, Nathan"

Results Per Page

Sort Options