Theses (Linguistics)
Permanent URI for this collection
Browse
Recent Submissions
Item Production of verbal morphology in heritage speakers of Tamil(2025) Srikanth, Shankhalika; Bird, SonyaThis study investigates the differences in the verbal morphology of heritage Tamil speakers and L1 Tamil speakers with the ultimate goal of providing applicable pedagogical insights for heritage Tamil teachers and learners. This work is centred within my community of heritage Tamil speakers. Tamil is a Dravidian language that is dominant primarily in South Asia. Tamil heritage speakers, like heritage speakers of other languages, often face issues of social stigma, shame, and feelings of exclusion when speaking their heritage language, in addition to linguistic barriers because of the lack of pedagogical materials and research on heritage Tamil acquisition. Nevertheless, many adult Tamil heritage speakers have a strong motivation to learn their heritage language. The long-term goal of my research is to identify and implement strategies to support heritage Tamil learners who want to (further) develop their oral proficiency in the language, as well as to raise awareness for heritage Tamil being a legitimate variety of Tamil in its own right. For the purposes of this thesis, I limited my scope to verbal morphology, and the question I investigated was how heritage Tamil speakers and fluent L1 Tamil speakers differ in their production of verbal morphology. I used a combination of games and stories to elicit verb forms with a range of tense and person-number-gender (PNG) affixes, making sure to use methods that could be reapplied in the classroom as teaching tools. The heritage speakers I worked with used a combination of target-like forms as well as different strategies of overgeneralization, notably overmarking, to simplify irregular morphological paradigms that are present in L1 tense and PNG morphology. The findings from this study demonstrate that heritage Tamil speakers have a deep awareness of Tamil’s complex verbal morphology, while also highlighting clear points of divergence between L1 Tamil and heritage Tamil grammars that parallel those seen in other heritage languages cross-linguistically and which can be useful for heritage language teachers in determining what types of forms to focus on in the classroom.Item The efficacy of explicit instruction of discourse connectives on Chinese EFL learners’ argumentative writing performance across different writing proficiency levels(2025) Yuan, Qin; Huang, Li-ShihArgumentative writing involves cognitively sophisticated thinking and reasoning skills (Taylor et al., 2019). Using connectives to indicate the logical reasoning process is one of the challenging aspects of learning writing for English learners (Snow & Uccelli, 2009). Discourse connectives (DCs) are important for discourse coherence and “such coherence is often marked by using discourse connectives” (Zufferey et al., 2015, p. 390). Regarding the use of DCs in IELTS argumentative writing, many Chinese undergraduate English learners in China with different proficiency levels do not perform well (Yao, 2014). What is more, according to the researcher’s teaching experience in English writing classes, many Chinese students could not properly employ DCs in their argumentative writing. Therefore, due to these reasons, more empirical studies are needed in order to investigate the effects of explicit instruction of DCs on Chinese undergraduate EFL learners’ argumentative writing performance across different writing proficiency levels. Using qualitative and quantitative research methods, this study investigated the efficacy of explicit instruction about DCs involving 96 undergraduate Chinese EFL learners across different proficiency levels (i.e., elementary proficiency level, n = 30; intermediate proficiency level, n = 32; upper intermediate proficiency level, n = 34) in their English argumentative writings. Conducted through a four-week treatment, the study used a pre- and post-test design to measure participants’ use of DCs and their writing abilities. Further, participants’ post-treatment survey provided qualitative data about participants’ use of DCs during the treatment. Through both qualitative data (i.e., content analysis) and quantitative (i.e., descriptive statistics, paired-sample t-test, and Pearson’s correlation), the results showed that the participants had stronger awareness and use of DCs and writing performance after the treatment. In the total frequency and appropriateness of using DCs, there were statistically significant differences between pre- and post-treatment writing tests across Elementary Group (EG), Intermediate Group (IG), and Upper Intermediate Group (UIG). As for the frequency and appropriateness of using three types of DCs, there were statistically significant differences on Type I DCs (e.g., in addition, finally, and similarly) and Type III DCs (e.g., because and to conclude) in EG, IG, and UIG. In Type II DCs (e.g., nevertheless, although, or, in fact, and on the other hand), there was a statistically significant difference in IG. However, there were no statistically significant differences in EG and UIG. As to their writing performance, participants in EG and IG made significant improvements but not in UIG after the treatment. Correlational analyses showed that there were significant correlations between the total appropriate use of DCs and participants’ writing scores in EG and UIG, but not in IG after the treatment. The analysis of the qualitative data revealed that explicit instruction of DCs could facilitate EG, IG, and UIG participants to overcome their challenges on using DCs in argumentative writing.Item Developing speaking strategies among adult English-as-an-additional-language (EAL) learners in performing the IELTS speaking tasks, mediated by audio-recorded and video-stimulated individual verbal reflection(2024) Alhemaid, Amjad; Huang, Li-ShihImproving speaking skills is a considerable challenge for many English-as-an-Additional-Language (EAL) learners striving to achieve oral proficiency and succeed in high-stakes standardized language tests. This study examined the reported and observed speaking strategies employed by 24 intermediate-level EAL learners performing IELTS (International English Language Testing System) Speaking tasks. The study explored the efficacy of integrating individual verbal reflection through Video-Stimulated Recall (VSR) and audio-voice recorders to facilitate EAL participants’ development of task-specific strategies and oral language production. Over a six-week period, participants engaged in a non-credit bearing speaking course, with a specific focus on the IELTS Speaking tasks. Participants were randomly divided into three groups: experimental group A (n = 8), experimental group B (n = 8), and comparison group C (n = 8). All groups received identical weekly lessons from the same instructor, followed by 30-minute post-task activities: group A engaged in VSR verbal reflections, group B in audio-recorded verbal reflections, and group C in non-reflective activities. In the sphere of strategy use, data were collected and analysed from three sources: weekly instructor-facilitated reflective group discussions, strategies reported weekly during reflection sessions, and the researcher’s observations of participants’ weekly performances. In the sphere of oral production, pre- and post-test scores, along with weekly task performance scores determined by two raters, were used to measure changes in participants’ oral production. Qualitative data were gathered via a post-study perception questionnaire to explore participants’ views on verbal reflection. Descriptive statistics revealed that participants used a wide range of strategies, with six strategy categories and a total of 2,038 individual strategies, including 84 unique individual strategies, identified across all data sources. Nonparametric tests indicated a significant improvement in oral production for group B, suggesting that audio-recorded verbal reflection may positively impact oral language development. However, the differences between groups A and C did not reach statistical significance. Interestingly, the lack of significant differences between groups A and B suggests that both reflection modalities may similarly facilitate the development of strategy use and oral production, meriting further research. The correlational analysis uncovered significant relationships among different variables. For example, cognitive strategies in Week 5 and metacognitive strategies in Week 6 positively correlated with performance, while communication and social strategies showed negative correlations in Weeks 1 and 4. Notably, eight individual strategies demonstrated positive correlations with performance scores, while four individual strategies showed negative correlations. Content analysis of the responses to the perception questionnaire highlighted key considerations for future research and pedagogical practices. The findings and their implications are discussed, offering practical pedagogical recommendations for implementing individual VSR and audio-recorded verbal reflections.Item Traditional ecological knowledge in Indigenous language revitalization(2024) Moffat, Anna; Urbanczyk, Suzanne ClaireTraditional Ecological Knowledge (TEK) is the knowledge and understanding of the complex systems of local ecology. Indigenous Language Revitalization (ILR) is a movement against the shifting use of a language within its particular community. Both languages and TEK are passed by intergenerational instruction and carried by each specific culture. This thesis explores how ILR and TEK are interconnected in many ways, including in language lessons (both method and content), in understanding worldviews which provide conceptual foundations in language, in language reclamation, and in understanding the land. This thesis follows an Indigenist paradigm and uses the structure of Parker (2012) to answer the following questions: how do communities include TEK in their language revitalization work? What are some of the effects of including TEK in Indigenous language revitalization work? What about TEK is important to language revitalization? To answer these questions, the thesis includes a review of the literature, interviews with Indigenous experts, a website survey and finally, a usable resource. The literature review contains analysis of extant literature. Interviews with experts who have been involved in the work of incorporating TEK in ILR in four Indigenous languages brings additional insight through their greater depth of knowledge, experience and perspective. The website survey contains an analysis of community ILR websites which correspond to the languages spoken by the interviewees. Finally, the creation of a resource ensures that this research is reciprocal. This study contributes to our knowledge of how TEK and ILR are intertwined, and underscores the importance of incorporating, respecting, and recognizing TEK in ILR.Item Learning from Qʷi·qʷi·diččaq ‘Makah’ story: Collaborative analysis and emergent linguistic lessons(2024) Hashimoto, Erin; Lukaniec, Megan; Czaykowska-Higgins, EwaArchival and legacy materials have become important resources for many language revitalization and reclamation programs to support the reawakening of sleeping languages (baird 2013; Baldwin & Costa 2018; Lukaniec 2022) or to maintain relationships with past generations of speakers (Hill 2024). This thesis centers around one such legacy recording in Qʷi·qʷi·diččaq (Makah language) which shares a well-known Makah oral story, “Raven and Crow,” as told by first language speaker, Ruth Claplanhoo. Working alongside Maria Hita·ʔa·ʔoƛ Pascua in her role with the Makah Language Program and Makah Cultural & Research Center, this research transforms the original story recording into accessible materials that can be adapted for current and future speakers, learners, and teachers while simultaneously addressing the limited availability of analyzed stories and texts in Makah. By developing a method for Collaborative Analytical Storywork (described in Chapter 2), Maria and I were able to create transcriptions, translations, morphological analyses, and time-aligned resources which respond to the variety of interests, challenges, and language-learning goals within the Makah language community (Chapter 1). In Chapter 3, I connect the conversations that emerged through my and Maria’s engagement with the “Raven and Crow” story to existing Wakashan literature and present the lessons about Qʷi·qʷi·diččaq that this story teaches us (or questions that it poses). Chapter 4 offers potential applications of this research in Makah language revitalization and suggests new topics and possibilities for future researchers.Item The effect of cross-cultural content schemata on ESL reading comprehension of beginners(1993) Zhang, XiaThis study investigates the effect of cross-cultural content schemata on reading comprehension of ESL beginners. It attempts to examine whether ESL beginners comprehend and recall significantly better a text which is culturally familiar than a text which is culturally unfamiliar. Subjects were a homogeneous group of ESL Chinese upper beginners at Camosun College. All of them were literate in their first language. The actual number of subjects participating the test was twelve. Two stories representing Chinese and Western cultures respectively were chosen as the testing materials. Because the language used in the original texts was too difficult for the subjects, adaptation was necessary. The adaptation of the materials was based on the idea that both texts should contain the same number of propositions, the same type of discourse structures, and should be at the same level of linguistic complexity. The subjects were asked to read the two stories and then recall them in Chinese. To better measure their comprehension of the texts, they were also asked to answer fifteen multiple-choice and true or false questions on each text. Recall protocols were analysed for the number of propositions and inferences. Two methods of analyzing text propositions were employed. Kintsch's propositional analysis was used at a micro level, focusing on the detailed information of the texts. Brown & Yule's discourse analysis was employed for a macro level of analysis, focusing primarily on the gist information of the texts. The analysis of inferences included elaborations and distortions. Dependent t-tests were used to analyze the data. The statistical analysis of results showed that the subjects recalled and comprehended the culturally familiar text significantly better than the culturally unfamiliar text. They made more culturally appropriate elaborations and less distortions from the Chinese story than from the Western story. Moreover, they spent less time in reading the familiar story than the unfamiliar story. Finally, a close look at the data revealed high individual differences. The results of the study indicated that cultural content schemata play a significant role in reading comprehension and recall of ESL beginning students. And having cultural background knowledge of the content of an L2 text may compensate for ESL students' low level of language proficiency. However, significant individual differences suggest that while cultural content schemata of a text may facilitate comprehension and recall of an L2 text, the degree to how much the prior knowledge can facilitate the understanding and remembering of the text seems to depend on the reader's second language competence and reading skills. Implications for educational practice are that reading teachers of ESL upper beginners should provide the students with appropriate background knowledge, especially cultural background knowledge. And they should attempt to choose reading materials which are culturally familiar to the students. Additionally, the importance of the linguistic aspects of L2 should be noted.Item Chinese-English bilingual education in China : the implication of content-based instruction(2002) Yang, JingboBilingual education programs as an alternative approach to teaching foreign languages have only recently been introduced in China. To date, no research has been done on this area of study from a Chinese perspective. This evaluation study is the first to investigate the effectiveness of bilingual programs in China. This study was conducted in a school in China. Toe author collected data from three different grade levels of bilingual and matching non-bilingual classes by using three vocabulary tests designed by the author, midterms and final exams provided by the school and two questionnaires. Both the statistical analyses of the quantitative data and the students' opinions derived from the qualitative data indicate that the pro gram was highly effective. Based on these results, I conclude that a well-designed bilingual program like the one in this evaluation study is a worthwhile replacement to a regular EFL programs and should be encouraged within the Chinese school system.Item The genetic association of Ainu(1994) Yanagizaki, YumiOne of the most interesting issues concerning the Ainu language is the uncertainty of its genealogy. Ainu is often classified as a language isolate whose genetic affiliation to other languages or language families has not been conclusively proven, as is the case with Japanese, Ainu's neighbour. The most widely debated and studied hypothesis is the one which assigns Ainu to the Altaic family, assuming the subgrouping of Japanese-Korean and Ainu together within the family. Today, the most time-honored hypothesis seems to be the one relating Ainu to the Austronesian languages. From a linguistic point of view, the recent attempt to demonstrate the affiliation between Ainu and the Austronesian languages was made by Murayama (1992) following Gjerdman (1926:29-84), yet fully-fledged linguistic investigation has not been made involving various dialects of Ainu and recent materials for Proto-Austronesian. It is the purpose of my thesis to provide conclusive evidence relating Ainu to the Austronesian languages. The study is proceeded by setting up Ainu roots for comparison with Proto-Austronesian. A considerable number of cognate sets and regular sound correspondences between Ainu and Proto-Austronesian are found as a result of the comparison.Item Hardening and weakening in Kwakiutl(1976) Wilson, Peter JohnThis thesis demonstrates that the Kwakiutl processes known as hardening and weakening are based on two synchronic phonological processes; assimilation and deletion. Hardening sees stem-final voiceless obstruents assimilate glottalization, while weakening sees them assimilate voicing. The following suffix-initial segments that cause this assimilation are subsequently deleted in certain environments. Rules are presented which generate the appropriate hardened or weakened form. To this end, the phonological processes involved in hardening and weakening, particularly assimilation and deletion are examined and analysed. Previous work, done mainly by Sapir, assumed that hardening and weakening are not based .on synchronic phonological processes. Instead, he hypothesized an historical solution which assumed that assimilation and deletion were the cause of hardening and weakening, but failed to note that the processes are synchronic. This thesis, therefore, furthers previous work by showing that hardening and weakening are based on synchronic processes, and begins the task of constructing an encompassing analysis. Unfortunately, I am not able to show that every occurrence of hardening or weakening is predictable. In the Conclusion, I speculate that the system of relative phonological strength proposed by Foley may account for this irregularity. It is also possible that the non-predictable data is historically based. In this thesis, however, I concentrate mainly on the predictable forms and leave aside the more obscure and less general problems for later work.Item On a certain tension in linguistics : Noam Chomsky and Roy Harris(1993) Williams, Kenneth EvanThis thesis is about a certain kind of tension existing between two quite divergent concepts of language: that of the American linguist, Noam Chomsky, and that of the British linguist, Roy Harris. According to one conception, language is considered to be a cognitive capacity-the 'language faculty'. The methodological approach of this view is to abstract some notion of language away from language use by way of the competence / performance distinction and other idealisations. Language is then conceived to be a formal mechanism, consisting of a system of rules and principles, instantiated in the brain. The general form of this mechanism (universal grammar) is said to be biologically innate and when exposed to some linguistic environment, virtually causes intact human beings to acquire one or more languages. The other conception also considers language to be a cognitive capacity which, no doubt, depends upon the brain, among other factors. However, according to this particular conception, certain obvious everyday aspects of language-e.g. that it is very much a diverse form of social interaction, that it is a complex normative practice, etc.-are thought to be paramount. This view holds that it is a mistake for any study of language to artificially contrive a distinction between some abstract notion of language, on the one hand, and language use, on the other, and then estrange the one from the other, treating the abstraction as fundamental. The former vision is that of Noam Chomsky and the latter vision is that of the Oxford linguist Roy Harris; and they are seen by their respective advocates as being mutually incompatible (in principle) in their aims for linguistic analysis. But what is most interesting about these two ostensibly incompatible conceptions of language and linguistics is that some of what they both stand for seems to make sense-this, of course, intensifies the tension. I am to some extent inventing this 'tension', as I call it, between Chomsky and Harris, since there is currently no dialectic debate in the discipline in which they are involved in scholarly exchange. And yet they have both widely published their respective views on language and its investigation, which are largely in radical disagreement. But in spite of this lack of dialectical connexion between Chomsky and Harris, there are definitely substantive issues here involving contrasting sets of ideas worthy of examination. Notwithstanding this divergence, however, aspects of both of their views seem reasonable and considered together, in contrast, may provide some greater understanding of linguistic communication. This thesis is aimed at examining this contrast.Item A phonology of Gitksan, with emphasis on glottalization(1974) Wickstrom, Ronald WilliamThis thesis consists of a phonemic analysis at the word level of the phonology of Gitksan, a language spoken by the people of the Hazelton, Kiwiox, Kitwanga, and the Skeena Crossing area of British Columbia. The study includes chapters on attestation, allophony, phoneme distribution, distinctive features, and glottalization. The section on glottalization includes reproduced spectrograms of Gitksan Glottalized consonants made on the Kay Electro-Sonagraph at the University of Victoria. Included also are appendices consisting of a word list and a sample text.Item Language, literacy and the developing world(1992) Whitney, John ChristopherItem Textual cohesion : an analysis of ESL students' oral narratives(1985) West, Elizabeth NorahA growing body of linguistic research, in the study of both first and second language, examines language beyond the sentence and focuses instead on discourse or text, either oral or written, as the unit of analysis. Textual cohesion, the semantic relations between sentences, provides some of the resources that a speaker or writer draws upon when constructing a text. These devices, termed cohesive ties, include reference, substitution, ellipsis, conjunction and lexical cohesion. Drawing on the research in both first and second language cohesion studies, this research reports the use of cohesive devices by adult English as a Second Language (ESL) students. The questions addressed are (1) What cohesive devices do these subjects use to maintain the flow of spoken discourse? (2) What , if any , is the difference among the levels of proficiency? (3) What problems arise? In order to address these three questions , oral narratives from twenty-eight Quebecois students in the summer English Language Program at the University of Victoria were elicited and analyzed for cohesive devices.Item Orthographic realizations of a selected sample of students of English as a second language(1976) Warbey, MargarettaThis research investigated the problems adult non-native speakers experience in their acquisition of English language spelling strategies. The problems inherent in the spelling system of English were examined by the writer. The research of other investigators in the areas of applied linguistics and language skills acquisition was examined for insights. To further clarify the areas of spelling difficulty encountered by second language learners, an experimental spelling test, incorporating several potentially troublesome segments, was given to 102 adult language students enrolled in English language classes. The results of this experiment were analyzed in the context of applied linguistics theory and current pedagogic practice to determine what, if any, conclusions could be drawn about the second language learner's internalized knowledge of the rules governing English spelling. Strategies for preparing useful teaching materials utilizing the theories of applied linguistics were considered with reference to the spelling abilities of non-native students as determined by their performance on the experimental test.Item Using pauses to improve listening comprehension for second language learners(2000) Van Loon, JohnThis thesis investigates the effectiveness of using pauses moral texts to improve listening comprehension for second language learners of English. The subjects in this experiment listened to three different radio news stones. One story was unmodified and two others were modified with pauses at specific junctures. After listening to each news story, the test subjects were given a written short answer test to measure their comprehension. The experiment parallels earlier work by Blau (1990), which showed that pauses placed m an oral text resulted in a significant improvement in the listening comprehension of second language learners. Blau describes the use of pausing as a means of giving the listener additional processing time without disturbing the natural features of the normal flow of speech (1990 749). As well, Griffiths notes that pauses are assumed to increase comprehension by (a) providing processing time, and (b) grouping words into syntactic constituents ( 1990 56). He also notes that slow speech is slow mainly due to the frequency and length of pauses. The design of this study parallels the Blau study, but also considers the effects of using much shorter pauses and different pause locations. The methodology and research from pausology studies was incorporated into the study design. This study is focused on two basic questions (1) Does editing oral text by inserting 500 millisecond pauses at juncture points improve the comprehension test scores for second language (henceforth L2) learners of English? (2) Does the location of an inserted pause have an influence on the comprehension test scores? The results from this study show that editing an oral text with 500 millisecond pauses at intonational and syntactic junctures does improve comprehension test scores for second language learners of English. There are also indications that the actual location of the pauses has an influence on comprehension test scores. This study concludes that there is some evidence suggesting that pauses can be a useful tool m facilitating listening comprehension for second language learners who are working with authentic oral texts. This seems to be especially true for lower level learners.Item An algorithm for the assignment of lexical stress in converting text to speech(1989) Urbanczyk, Suzanne C.This thesis describes the development and testing of an algorithm to assign main stress to English words. It has been developed for inclusion in a text- to- speech conversion system. Within the field of speech technology, the correct assignment of lexical stress is important in ensuring that the correct rules of prosody are applied to the speech output. Some of the problems associated with word stress in English have been addressed, including aspects of the morphological structure of English and a quantity- sensitive stress rule. The algorithm developed is as accurate or more accurate than other stress- assigning algorithms reported on in the literature. During the evaluation of the algorithm, some observations have been made that may have future implications for theories of lexical stress assignment for English.Item Chiac - an example of dialect change and language transfer in Acadian French(1986) Thompson, Jennifer A.The principal goal of t his thesis is to determine whether Chiac, a dialect of Acadian French spoken in Moncton, New Brunswick, forms a dialect in its own right, or is simply a stage in the evolution of the traditional Acadian dialect of the surrounding area. The basic difference between Chiac and the traditional dialect of the region is t he amount of anglicisation found in each. The Acadians were a very isolated people over a whole century, longer in some cases, and as a result their language has retained many features of t he seventeenth century French that their ancestors spoke. Over the past century more and more Acadians have been moving to the urban centres in New Brunswick, to be surrounded by anglophones. Living in cities that are over sixty percent English, as Moncton was once, the Acadians cannot avoid some interference of English in their French. In the case of Chiac, anglicisation accounts for at least twenty-five percent of the dialect, and probably more. In order to determine whether or not the changes to the urban dialect are enough for it to be considered a separate dialect from the rural speech, nineteen informants were asked to participate in tape-recorded interviews. Four were unilingual and from rural areas, three were bilingual and from rural areas, and twelve were bilingual and from the greater Moncton area. The informants were asked to read a passage from a newspaper or to answer questions that would elicit certain sounds in the responses so that the phonological features of each speech type could be examined for instances of change. After being asked to describe pictures or to name objects in the room, the informants were asked to discuss any topic of interest to them. The results of these parts of the interviews were examined for any regular changes or instances of language transference. There were morphological, lexica l, semantic and syntactic changes and transference that seemed to occur mainly according to age and level of bilingualism. The results of the interviews in general showed a marked difference in speech habits between the rural and the urban informants. The fact that the urban speakers, particularly the younger ones, had lost some of the features of the Acadian dialect and gained some from English and from Standard and Quebec French makes a good case for Chiac being considered as a separate dialect from the traditional speech of the region.Item The effects of loanwords on the phonology of contemporary Japanese(1979) Tatsuki, MasaakiItem Telling the truth : a linguistic analysis of narrative construction in the cross-examination of Canadian aboriginal witnesses(2002) Steel, Heather LindsayThis thesis contributes to a growing area of interest that considers courtroom interaction from a linguistically based, discourse analysis perspective. It examines how narrative is constructed during the cross-examination of Aboriginal witnesses in Canadian courts. The Canadian legal system operates on the idea that there is only one truth. The purpose of the trial is to seek out that truth through questions. This search for 'the' truth can pose a problem for members of marginalized groups, such as Aboriginals, who perhaps have a different perspective and, therefore, a different truth. Adhering to the traditions of courtroom discourse may mean that their narratives may not be conveyed in a way which matches their reality. To get a clearer understanding of how narratives are built in the cross-examination stage of the trial, this study examines testimony given in two cases involving Aboriginal rights. The first part of the study looks at how much discourse witnesses contribute during the cross-examination. With this understanding, the second part of the study explores the content of the discourse to determine patterns within the construction of narrative. The study goes on to discuss sociolinguistic reasons for why some witnesses contribute less to the narrative being constructed than others do. Finally, some recommendations are put forth to help bridge the cultural gaps which presently exist in the courtroom.Item Keeping tabs on idioms : a sociolinguistic study of the use of English idioms(2002) Smith, Mavis AnneThis study investigates sociolinguistic aspects of the use of English idioms in order to determine if there are age and gender differences related to idiom usage. Varying definitions of the term "idiom" exist in the literature; therefore, an integral part of this study is a critical review of existing definitions and the development of an operational definition of "idiom." I have used the ethnographic methodology developed by Hymes (1972) to examine the use of idioms among 319 native English speakers engaged in spontaneous conversation. Speakers were observed without the active participation of the researcher in order to eliminate the effects of the "observer's paradox." The results indicate an idiom frequency of 2.4 idioms per hour in 42 hours of speech, reveal a low rate of repeated idioms ( 4%), and suggest that male speakers in their forties and older make the most frequent use of idioms. Implications for second language teaching are discussed.