The Japanese language allows flexible word order (Shibatani, 1990) and allows missing noun constituents (Shibatani, 1990). The case-marking is realized with postpositional particles, e.g. the subject marker —ga and the direct object marker -o. However, these case markers are often omitted (Aida, 1993; Miyata, 2008; Rispoli, 1989). While Bates and MacWhinney (1989) and Sasaki and MacWhinney (2006) view case marking as the strongest cue in Japanese and animacy contrast as the second reliable cue, Ito, Tahara, and Park (1993) and Rispoli (1987) claim that Japanese children produce case markers with errors because they do not process them. It has not been clear how Japanese children acquire verb argument structure when there is little structural information in their input. The current study aims to account for first language acquisition of Japanese verb argument structure by means of corpus analysis. We hypothesize that, while case markers are often omitted, both mothers' production and children's production initially follow the canonical pattern where the word order cue and the animacy contrast cue form a coalition (Bates & MacWhinney, 1989; Sasaki & MacWhinney, 2006), which aids children to identify the grammatical relations of the verb arguments. The results indicated that the coalition we proposed is reliable information when the two nouns were realized. While both mothers and children produced the non-canonical form frequently even at the earliest age contrary to our prediction, children start from the canonical vs. non-canonical ratio that is similar to the maternal speech and then develop their own distribution as they get older. Moreover, we suggest that animacy contrast cue in Japanese is more important for children than it has been suggested in the past. The animacy contrast is almost always present in the utterances and it is the information the speakers and the listeners use to specify the grammatical relations of the noun arguments when the sentence involves the reordering of the constituents or includes only one noun. On the other hand, the case-marking cue only appeared in a redundant context, which makes it hard for Japanese children to acquire case-marking.

Native Arabic speakers often demonstrate exceptional difficulties reading in English (Thompson-Panos & Thomas-Ruić, 1983). Research suggests that they have difficulties processing English vowels, leading to further difficulties with word recognition and phonological processing. Research also suggests that in their L1, native Arabic speakers rely on consonants alone for word recognition. This is because of a unique feature of Arabic orthography most vowels are not normally included in text. If Arabic speakers use a reading strategy that focuses on consonants and uses context to fill in vowels, this would have implications for learning to read another language (English) without predictable vowels. This is especially relevant because L2 learners often transfer L1 reading strategies to L2 (e.g., Koda, 2007). This study used eye-tracking to investigate the difficulties that native Arabic speakers have reading in English, with native English speakers as a comparison. The influences of two variables were examined: word frequency and orthographic vowel ambiguity (whether an orthographic vowel sequence has more than one common pronunciation). Participants read sentences containing high- and low-frequency words that had ambiguous or unambiguous vowels while their eye movements were recorded. Results show that native English speakers are not influenced by frequency, but are consistently influenced by vowel ambiguity, with more processing difficulty on words with ambiguous vowels than unambiguous vowels. This shows that native English speakers access phonology deeply enough during reading to be affected by an ambiguous vowel. In contrast, the native Arabic speakers showed a strong frequency effect (with more difficulty on low than high frequency words) but were rarely affected by vowel ambiguity. These results suggest that native Arabic speakers do not access English phonology deeply during reading. This is likely the result of transferring an L1 reading strategy that does not depend on vowel information. If native Arabic speakers do not access English vowel information, as these results suggest, this may explain their reading difficulties. Accurate phonological processing is essential for the development of fluent English reading (Adams, 1990). Using written vowels also frees cognitive resources for higher-level processes such as comprehension. Implications for models of reading and pedagogy are briefly discussed.

The three studies in this dissertation bring together quantitative and qualitative methods in order to understand L2 learning of derivational morphology. By using measures of derivational knowledge developed from L1 research, Study 1 provides a foundation for in-class research by assessing what L2 learners know and do not know about derivational morphology in comparison to adult native English speakers and how factors such as L1 background and L2 proficiency shape L2 knowledge of derivational morphology. Results show that L2 learners have poor knowledge of derivational morphology regardless of L1 background or L2 proficiency. Study 2 follows from these results and investigates the effects of input-processing versus pushed output instruction on the development of productive and receptive morphological abilities. The results of this study support the hypothesis that instruction is beneficial for L2 derivational learning; however, results do not support the hypothesis that pushed output instruction leads to better immediate and long-term learning than the input-processing condition. In fact, results suggest that equivalent learning occurs between the two conditions across all measures of derivational knowledge. Finally, Study 3 was a qualitative investigation of learners' attitudes, actions, and motivations towards the learning of derivational morphology over the course of Study 2. Using activity theory, this study describes how students' initial actions, which were not aligned with the goal of morphological learning, were transformed over the course of the study as students came to realize the importance of derivational morphology for their success in English. The results of this study are also important because they offer an alternative explanation for why the hypotheses in Study 2 regarding the effectiveness of output were not supported. Specifically, many participants in this study became aware of derivational morphology for the first time as a result of this study; therefore, a "novelty effect" (cf. Tulving & Kroll, 1995) may have overridden any potential benefit of the output treatment over the input treatment.

Previous research involving the Aspect Hypothesis (AH) has shown that second language (L2) learners are sensitive to lexical aspect when applying grammatical markers, associating perfective-past marking with telic verbs and imperfective past marking with atelic verbs (Andersen, 1991; Andersen & Shirai, 1994). Some studies, however, report that in the initial stages of learning, L2 learners may assign a default past tense form across lexical aspect categories, suggesting that beginning learners may not initially adhere to the AH (e.g., Salaberry, 1999). The primary purpose of this dissertation is to investigate the degree to which L2 learners of Russian at various levels of proficiency adhere to the AH. This dissertation addresses two primary issues: (1) the degree to which L2 Russian learners of varying proficiency levels adhere to the AH, and (2) the conditions under which the AH may or may not be supported. Data from written narratives (N=42) and oral narratives (N=42), elicited using a film clip (Modern Times), and oral proficiency interview conversations (N=33) collected from classroom learners of Russian (L1 English) at various proficiency levels were analyzed for lexical aspect and tense-aspect marking with comparable data from native Russian speakers (N=18). The results indicate that the AH is supported to varying degrees dependent on task type and proficiency level, and that tasks involving lower planning levels (oral narratives and conversations) were generally more supportive of the AH, compared to the written narrative task that involved a higher level of planning. The results also show that beginning-level learners of Russian prefer the imperfective form in the past tense across the different task types. The results constitute preliminary support for the default past tense hypothesis in that there is a preference for a default marker (whether imperfective or perfective) in the past tense, at least in the beginning stages of language learning by L1 English classroom learners of aspectual languages such as Spanish or Russian. Two factors discussed as possible explanations for the use of default markers are the role of instruction and L1 influence.

Thesis (Ph.D.)--New York University, School of Education, 2007. / Typescript. Includes bibliographical references (leaves 200-210). Also available in electronic format on the World Wide Web. Access restricted to users affiliated with licensed institutions.

Thesis (Ph. D.)--University of Queensland, 2002. / Includes bibliographical references.

<p> Though previous research has documented the Southern Vowel Shift (SVS) in Alabama and Tennessee, none has focused on Mississippi. Also, the majority of research has focused on European-Americans. In this study, data was collected from women from northern and central Mississippi, with central residents evenly recruited from urban and rural areas. Of these, 15 were European-American and 19 were African-American. Participants read a word list including target vowels in the <i>b_d</i> frame. F1, F2, and vector length were analyzed to determine to what extent participants exhibited the SVS and Back Vowel Fronting. For the SVS, there were effects such that central residents shifted more than northern, rural residents shifted more than urban, and African-American residents shifted more than European-American. European-American women fronted /u/ and /o/ more than African-American women. These results suggest that African-American women from Mississippi do participate in the SVS but are not fronting their back vowels.</p>

<p> The theories of biological folk taxonomies have been discussed in the anthropological literature since the 1960&rsquo;s with several researchers such as Brent Berlin and Eugene Hunn devoting many articles and even books to the subject. Despite many examinations of the naming systems present in languages all over the Americas, there have been few, if any, works about the linguistic principles behind the two major theories of naming. This paper frames the linguistic bases for the two theories using data drawn from a corpus of Anishinaabemowin plant names and describes the linguistic basis for both Berlin&rsquo;s theory of a morphological (in the biological sense) basis for a hierarchical system of naming and Hunn&rsquo;s theory of use-based names. This paper will demonstrate that the theories of folk taxonomies can be greatly improved if theories of morphological (in the linguistic sense) preference are considered along with theories based in biological morphology or cultural usage.</p>

<p> Function words are known to be shorter than content words. I investigate the function words <i>BE</i> and <i>HAVE</i> (with its content word homonym) and show that more reduction, operationalized as word shortening or contraction, is found in some grammaticalized meanings of these words. The difference between the words&rsquo; uses cannot be attributed to differences in frequency or semantic weight. Instead I argue that these words are often shortened and reduced when they occur in constructions in which they are highly predictable. This suggests that particular grammaticalized uses of a word are stored with their own exemplar clouds of context-specific phonetic realizations. The phonetics of any instance of a word are then jointly determined by the exemplar cloud for that word and the particular context. A given instance of an auxiliary can be reduced either because it is predictable in the current context or because that use of the auxiliary usually occurs in predictable contexts. The effects cannot be attributed to frequency or semantic weight. </p><p> The present study compares function word production in the speech of school-aged children and their caregivers and in inter-adult speech. The effects of predictability in context and average predictability across contexts are replicated across the datasets. However, I find that as children get older their function words shorten relative to content words, even when controlling for increasing speech rate, showing that as their language experience increases they spend less time where it is not needed for comprehensibility. Caregivers spend less time on function words with older children than younger children, suggesting that they expect function words to be more difficult for younger interlocutors to decode than for older interlocutors. Additionally, while adults use either word shortening or contraction to increase the efficiency of speech, children tend to either use contraction and word shortening or neither until age seven, where they start to use one strategy or the other like adults. Young children with better vocabulary employ an adult-like strategy earlier, suggesting earlier onset of efficient yet effective speech behavior, namely allocating less signal to function words when they are especially easy for the listener to decode.</p>

Speech sounds contrast on many acoustic dimensions. The constellation of acoustic "cues" defining a given sound contrast is language-specific, such that the "same" sounds in different languages are actually realized slightly differently. Furthermore, even within the same language group, speakers and listeners exhibit considerable variability in their use of acoustic cues. This work explores acoustic cue use in stop voicing contrasts across languages (Spanish, English, and Korean) and modalities (production and perception). A first group of experiments target "baseline" cue weights, or how speakers and listeners use multiple acoustic dimensions to define native and foreign sound contrasts in production and perception. A second set of experiments investigates how listeners modify their definitions of these categories in order to accommodate to changes in the input (e.g. a speaker with an "accent"), and in particular, how baseline cue-weighting strategies can influence and direct these adaptation patterns. Along with comparison of cross-linguistic differences in cue use, all of the studies focus on variability within language groups and examine the relationship between perception and production on an individual level. Taken together, the studies provide a detailed comparison of the way speakers and listeners make use of multiple acoustic cues cross-linguistically. The experiments give a comparative picture of the extent of between- and within-language variation in "baseline" cue weights across three languages and provide insight into the processes by which listeners adapt or "tune" their phonetic categories when confronted with changes in what they are hearing. The work highlights the complexity of the perception-production interface, as well as the interplay between language-specific knowledge, general learning mechanisms, and general auditory factors in how listeners use acoustic cues to define and modify their speech sound categories.