1 I MPACT OF INSTRUCTION ON LEARNER S ACQUISITION OF VARIOUS FRENCH PHONEMES By EDITH OHENEWA PARE A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS UNIVERSITY OF FLORIDA 2011
2 2011 Edith Ohenewa Pare
3 To my parents Emmanuel and Kate Parry
4 ACKNOWLEDGMENTS First and foremost, I wish to express my sincere appreciation to my advisor Dr Theresa A. Antes, for her contin uous support and without whom I would never have received immense knowledg e and training in the field of second language a cquisition and without whom this study would never have come into fruition Her guidance helped me throughout the time of research and also during the writing of the thesis. As my teaching mentor and supervisor, I am deeply indebted to her for believing in me and helping me become a good and confident teacher. Many thanks also go to Dr James Essegbey, who was a member of my thesis commi ttee for his encouragement, insightful comments and for painstakingly revising my thesis. T hanks go out t o Dr Carol Murphy Dr Rori Bloom and Dr Anny Mavambu for allowing me into their various classrooms to recruit participants for my study. I also ex tend thanks to all the participants in the study for their interests and willingness in participating in the study. My sincere thanks go to all other faculty and staff in the department who always asked about the progress of my study and also for their h elp and encouragement throughout my studies. I am also indebted to my colleagues Audrey Viguier and Fabien Cappelli, who helped me rate all the speech samples of my study. My love and appreciation goes to my family for their caring, spiritual support and unconditional love throughout my studies.
5 Last but certainly not least, it is a pleasure to express my gratitude to my husband, Ebenezer Ferguson Laing, for his love and encouragement. He has made available his constant support in a number of ways.
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 TABLE OF CONTENTS ................................ ................................ ................................ .. 6 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF ABBREVIATIONS ................................ ................................ ............................. 9 ABSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 12 2 LITERATURE REVIEW ................................ ................................ .......................... 15 ........................... 17 ................................ ................................ ......... 19 Age and Ability to Acquire Second Language ................................ .................. 20 Attitude and M otivation as Contri buting Factors in Second Language Acquisition ................................ ................................ ................................ ..... 22 Technology and its impact on Second Language Acquisition ........................... 23 Perspectives on Sec ond Language Teaching ................................ ......................... 24 How Pronunciation Should Be E xamined ................................ ............................... 27 3 METHODOLOGY ................................ ................................ ................................ ... 29 Participants ................................ ................................ ................................ ............. 29 Instruments ................................ ................................ ................................ ............. 30 Judges ................................ ................................ ................................ .................... 31 Sound s Examined ................................ ................................ ................................ ... 32 Data Collection ................................ ................................ ................................ ....... 33 Preparation of Speech Samples ................................ ................................ ............. 34 Description of Curriculum ................................ ................................ ........................ 34 Hypothesis ................................ ................................ ................................ .............. 35 4 RESULTS AND ANALYSIS ................................ ................................ .................... 37 Differences between the Experimental and Control Group ................................ ..... 44 Comparison between the Experimental and Control group ................................ ..... 44 5 CONC LUSIONS AND DISCUSSIONS ................................ ................................ ... 52 Limitations ................................ ................................ ................................ ............... 56 Recommendations and Implications ................................ ................................ ....... 57
7 APPENDIX A FRENCH TEXT FOR PRE AND POSTTEST ................................ ........................ 60 B QUESTIONNAIRE ................................ ................................ ................................ .. 61 C DATA FOR CASES ................................ ................................ ................................ 62 LIST OF REFERENCES ................................ ................................ ............................... 63 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 68
8 LIST OF TABLES Table page 4 1 Native languages of participants ................................ ................................ ......... 48 4 2 Descriptive statistics for the experimental group ................................ ................ 48 4 3 Descriptive statistics for the control group ................................ .......................... 48 4 4 Difference in accuracy between the phonetics and composition classes for phoneme /y/ ................................ ................................ ................................ ........ 49 4 5 I ndependent samples test for /y/ ................................ ................................ ......... 49 4 6 Difference in accuracy between the experimental group and the control group phoneme /u/ ................................ ................................ ................................ ..... 49 4 7 Independent samples tes t for /u/ ................................ ................................ ........ 50 4 8 Difference in accuracy between the Phonetics and Composition class (nasal vowels) ................................ ................................ ................................ ............... 50 4 9 Independent samples test for nasal vowels ................................ ........................ 50 4 10 List of words generally articulated wrongly ................................ ......................... 51
9 LIST OF ABBREVIATION S CMC C omputer mediated c ommunication IPA International Phonetic Alphabet L1 The first language a child speaks L2 Any language learned after the first language or mother tongue
10 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Arts I MPACT OF INSTRUCTION ON LEARNER S ACQUISITION OF VARIOUS FRENCH PHONEMES By Edith Ohenewa Pare August 2011 Chair: Theresa Antes Major: French and Francophone Studi es The purpose of this study is to examine the effectiveness of explicit instruction on pronunciation accuracy in a French intermediate class. Another objective of this study was to compare two groups of students: students who concurrently received phonet ic training and took other French classes (generally French literature and /or sociolinguistics ) against students who took a Composition and Stylistics class (often in addition to French literature or sociolinguistics ) but did not receive phonetic training. Participants in both groups were students of the University of Florida and participated in this study in the Fall of 2010. T his study investigates how learners of French are able to acquire second language phonology, with focus on certain French vowels wh ich are generally considered to be difficult for second language learners Although some researchers have stated that acquiring native pronunciation may not be considered a realistic objective in the foreign language classroom, discrimination of phonemes i s a crucial factor in foreign language both for comprehension and intelligible purposes. Participants were tested on the recognition and production of the French phonemes /u/ and /y/ as in the following words: and rue which are phonetically transc ribed as [/ R u/, / R y /] respectively. We also examined the ability for learners to make a distinction
11 between oral and nasal French phonemes found in seau and son transcribed [/ so/, / s /] This research generally consisted of only one task : reading a s hort French text aloud in both pre and posttests. After analyzing our data, we found significant differences between the two groups with respect to the /u/ and /y/ phonemes. On the other hand, our participants had little or no difficulty distinguishing be tween oral and nasal phonemes. These findings did no t support the hypotheses of the study with and therefore did not reveal the effectiveness of explicit instruction on our experimental group Nevertheless, we found that there were certain factors such as certain that could have influenced the Our findings did not corroborate studies of researchers who found th e importance of explicit instruction in the acquisition of second language phonology in their studies.
12 CHAPTER 1 INTRODUCTION Learners of French at the college level have the opportunity to take a Phonetics course in their fifth or sixth semester in orde r to improve pronunciation in the target language. P ronunciation in the foreign language is important for comprehensibility purposes. Llisterri (2003) no needs to control when learning a foreign language Generally, if words are not accurately pronoun ced, comprehensibility and intelligibility problems arise The native language sometimes causes a breakdown in communication therefore one can on ly imagine the im pact pronunciation inaccuracy in a foreign language has on both native speakers and learners as a whole. However there has been scarce research on factors affecting the acquisition of the second language phonological system (Elliott, 1995 ). There are generally four skills which are vital for language instruction and in recent years most institutions have integrated them into their curricula. These skills are listening, speaking, reading and writing but occasionally, due to time constrain ts, educators may not be able to assign equal attention to each of these skills. For example, Tarone (1978) indicated that more attention is not given to developing pronunciation because there is general conviction on the part of second language acquisi tion researchers, second language teachers and students, that pronunciation of for exa such that ning tapes to enable them to improve on both their listening and speaking skills. They are therefore required to listen to these
13 tapes on their own before and after classroom work but one may wonder how many of these students utilize these tools. Researche rs like Dansereau (1995) have proposed teaching only basic phonetics at this level in order not to overwhelm students In other words, phonetic symbols can be confusing at the initial stages of learning a foreign language since many of these symbols repre sent sounds different from the corresponding grapheme (for example, the French phoneme /y/ corresponds to the grapheme whereas others like Elliott (1997) have argued that learners need to receive explicit instruction to be able to acquire near native pronunciation. We realize that there are conflicting views on what and how to teac h pronunciation even though researchers acknowledge that it must be taught as part of language instruction. Specifically examined questions : 1. What are the effects of explicit in struction on pronunciation accu racy? 2. measured by pre and posttests? 3. Are there certain contexts in which phonemes occur in a sentence that affect The purpose of this present study was to examine whether or not explicit phonemes which are sometimes considered by educators to be difficult for non natives. This investigation ex amined the effectiveness of phonetic training on pronunciation accuracy. For this reason, two groups of learners of French at the intermediate level were recruited. While one group received phonetic training, the other did not. After examining previous lit erature concerning this subject most researchers support the explicit instruction of Phonetics in language classrooms but the findings of this study evidenced otherwise Most p articipants who received phonetic training went from
14 satisfactory to worse pron unciation by the end of the instruction period whereas the participants who did not receive this specific instruction generally did better over time. The findings therefore supported the position of Arragon (1985) who did not see the importance of explicit instruction. The literature review is presented in the next chapter and the focus of chapter three is the methodology of the study Results are then r evealed and discussed in detail in the subsequent chapters.
15 CHAPTER 2 LITERATURE REVIEW Objective s in language teaching might have advanced over a decade at the university level but instruction on pronunciation has been more often than not a n aspect of language learning that seems to be neglected or found irrelevant by some instructors. It is undoubte dly a fact that there is a direct connection between and his pronunciation and these two features are inseparable but as foreign language educators, much attention must be given to pronunciation teaching so as not to break down com munication. Should instruction on pronunciation be inculcated into students at the high school level? Should it be viewed as important as grammar or vocabulary for language studies? Which criteria should be adopted in selecting what to teach at both the be ginning and intermediate university levels? In this chapter, I review literature on some relevant and prominent empirical findings in an effort to measure several hypotheses regarding the acquisitio n of phonological features in second or foreign language Second language pronunciation has historically proved to be a difficult skill to sometimes contradictory results (Elliott, 1995). Could the lack of success in this area of pedagogy be due to the way pronunciation is taught? For example, a French Corrective Phonetics course is offered to students in their fifth or sixth c ollege semester, and one may wonder why this is taught so late in their French studies. Scovel (1988) argued that the acquisition of new phonological features in a second language is a critical element aspect of language performance tha and as a result
16 will inevitably have a non much attrib uted to neurological factors, as reiterated by Scovel. On the other hand, Neufeld and Schneiderman (198 0) run s counter to that of Scovel, and argue that like proficiency in the prosodic and articulatory features o (105) after receiving phonetic instruction. Lado (1964 ) noted that some of the method s of teaching pronunciation for many decades included the direct approach. This is when students repeat word for word after their teachers and the lea lingual approach, which drills students in the use of grammatical sentence patterns. Students are made to hear a model dialogue and repeat each line of the dialogue. The cognitive code appr oach, in contrast, focuses on how the brain processes language and in turn makes the student aware of how the language system works. It is true that in second language acquisition, cognitive and affective factors can affect the learning process and that i s why learners may share common traits and others may have individual characte ristics. In this era therefore we see several educators follow the communicative approach in the classroom encouraging day to day comm unicative experiences. The communicative ap proach is not entirely focused on pronunciation but on oral proficiency as a whole All to this aspect of language teaching is insufficient. Morley (1991) stated that pronunciation should be viewed in the same light as grammar and syntax and must be seen as more than correct production of phonemes. He added that the goal of
17 more reali stic and convincing goals of developing self confidence, functional intelligi bility functional communicability, speech monitoring abilities and speech modification strategies intended for use beyond the classroom setting (500) e Language Transfer on L2 Acquisition According to Kenworthy (1987), some of the factors that inhibit or enhance the acquisition of native level of exposure to the target language, in add his sense of identity and the refers to when a learner applies forms and rules of his native lan guage to a foreign or a second language. learning proc ess of an L2 He over into the L2 9). Also, t here are lots of online journals and websites such as forvo.com and fonetiks.org which provide online resources in order to improve pronunciation skills a great deal. However it has not yet been confirmed by research whether this practic e has actually improved the acquisition of new phonemes with regards to transfer from L1 Most researchers have studied isolated sounds on which educators can train their students but have not studied the effects of explicit instruction on pronunciation i n its entirety (Cenoz & Lecumberri, 1999, Walley & Randazza, 1998). Cenoz & Lecuberri analyzed the effect of training on the perception of English vowels by native speakers of Basque and Spanish. One of the research questions aimed at analyzing the effect of
18 training on the discrimination of English simple vowels (long vowels /i:/ and /u:/, short vowel /a/) and diphthongs. After a t t est was carried out, the result s indicated that training ha d a significant influence on the perception of English simple vowe ls and diphthongs with the exception of the long vowel /u:/ Indeed, the mean obtained before training was lower than the mean obtained after a few hours of specific instruction in phonetic discrimination. The findings showed that the perception of some En glish vowels presented a level of difficulty for native Spanish and Basque speakers and even though transfer was not the only factor affecting phonetic development in second language, according t o Odlin (1989), it is often believ ed that the categorization of foreign language sound with respect to the first language phonemic inventory elucidates the problems in phonetic discrimination. Dulay, Burt and Krashen (1982) suggested in their study that there are two possible explanation s for L1 influence on the pro cess of L2 acquisition. The sociolinguis tic perspective looks at the occurrence of language interactions when there are two language communities in contact with each other. This could be through borrowing from the L1 or through code switching. The second e xplanation, according to them, is from a psychological standpoint, which looks at the influence from old practices when new ones are in the process of being learned. Furthermore, learners of French at the beginning level are confronted with the challenge of the link between orthography and pronunciation because written words are necessarily not pronounced the way they are written. Even intermediate learners who are introduced to IPA symbols become overwhelmed because these symbols do not always correspond to the way they are pronounced. pronunciation errors
19 might therefore emanate from bad habits that have been formed while trying to translate from print to sound (Dansereau, 1995). In other words, bad pronunciation habits may be developed through transfer of skills from written production to oral expression in the Does this transfer of skills also confirm the idea that pronunciation errors in the target language are due to the fact that learners extend their knowledge and skills in the L1 to the L2? Having foreign language knowledge in reading alone is not sufficient to help ency in or der to communicate, as Lado (1964) noted that students taught only to read were bound to the mispronunciation encouraged by spelling and were not able to free themselves from the mistaken idea that pronunciation is a poor expression of the correct written form, when actually it is writing that constitutes a poor representation of speech (5) All l earners have the same ability to acquire second language phonology but in a language classroom where students may have different native languages interference may vary due to L1 transfer. For instance in English, there is an allophonic difference between nasal and oral vowels whereas in French, there is a distinctive phonemic difference The question is from what languages should examples be drawn in order to make the m accessible to all? Assuming English and some Spanish, a not uncommon occurrence in American classroom s today, how should pronunciation be taught to meet each learner It therefore becomes a challenge for many educators who are confronted with how to handle pronunciation in a language class due to its multicultural and multilingual nature (which can be especially
20 problematic for learners ) Occasionally, t here i s an issue between orthography and phonology. An example is a sound which is spelt one way but p ronounced differently and occurs in both French and Spanish but is somewhat absent in English. An example g/. I n pronouncing thi s word, the is pronounced [g] in French but [ k ] in English It is then easy for learners whose L1 is Spanish to produce a near native French pronunciation because there is a pronunciation similarity rs will not comprehend was not assimilated Therefore, in this instance, the Spanish speakers may be at an advantage. On the other hand, the dynamics change when Spanish speakers are asked to nintelligible in some cases due to the fact that Spanish speakers tend to pronounce /b/ in place of /v/. L. Rodriguez (1993) proposed a four stage pedagogical progression in which the linguistic diversity of the learners will no longer be an obstacle to la nguage learning but serve as a source of reflection and a didactic tool speech Also, educators need to monitor the source language habits shared by learners into target language prac tice (comparative activity). In addition, corrective exercises should be set up in the target language (articulatory and acoustic method) and finally, ancestral linguistic speech habits should be integrated into language teaching (6). Age and Ability to Ac quire Second Language There is conflicting research with regards to child versus adult proficiency in the acquisition of phonology. Researchers like McLaughlin (1977) and Tarone (1987) found an important age (between six and t welve years) for the native li ke acquisition of phonology, though there have been some exceptions (Neufeld, 1980). Neufeld
21 co ncluded from his data that or near native competence and rece Scovel (1 988) suggested that acquiring other aspects of language is fundamentally different from learning pronunciation, due to the fact that morphosyntax, vocabulary and other pho nology He also claimed that those who begin to be exposed to an L2 after the age Scovel was however quick to note that t here are rs who coun ter the predictions of this critical period hypothesis. Also in their study of Dutch learners, Olson & Samuels (1982) found that elementary age children scored more poorly in German pronunciation than both junior high and college age participants and argu Hoefnagel Hohle (1977) also found adults to be superior to children in the area of sound reproduction. Schumann (1975, 1978) pointed out that sociopsychological factors ar e said to favor younger learners over older learners. According to his data, there is the tendency for younger learners to be motivated to reach a near native level. Theo Bongaerts et al. (1997) conducted a study with three different gro ups of participants made up of five nativ e speakers of British English ( control group), ten Dutch learners of English who had an extensive knowledge of British English, and a group composed of twelve Dutch university students of English at different stages of aptitude. None of the participants in the third group had received instruction in English before the age of twelve nor had they been exposed to English input before that age. Participants in the first two groups were
22 lecturers who taught English at a Dutch university or a teacher training institute The tasks for all the participants were to provide a spontaneous speech sample and read out loud a brief English text, ten sentences and a list of twenty five English words. The different samples were rated for accent by four linguistically inexperienced native speakers of British English. The most outstanding result of their study was that it was challenging for the judges to differentiate the group of highly successful learners from the native speaker control group. These fi ndings thus undermine the assertion that only younger learners can attain a near native pronunciation of a second language. Attitude and motivation as Contributing Factors in Second Language Acquisition Elliott (1995) experimented with what he classified a s multimodal instruction (a method that appeals to individual learning strategies and learner preferences (oral, aural and visual) with the view of using formal instruction to improve L2. Findings from f the factors related to pronunciation acc uracy; however it was not a major predictor of improving pronunciation. It is often common in language classes to have differing attitudes towards the acquisition process especially at the beginning levels. There a re learners who find themselves in the language classroom because it is an academic requirement and therefore tend to focus more on their major courses. For some of these students, their goal is to pass the class (excel in exams and have a good grade at th e end of the semester). Often times, there is no tendency to continue wi th language study therefore learners lack moti vation and drive to maintain a positive attitude towards language ent to which the individual works or strives to learn the language because of a desire to do so and the Ushida (2005) also investigated the role of
23 r Spanish in pilot language online courses (LOL) and the findings from the study indicated that motivated students studied frequently and productively in order to take every opportunity they get to improve their language skills. Technology and its impact o n Second Language Acquisition Computers play an important role in langua ge teaching and learning and their popularity has increased rapidly especially in the 21 st century. As technology captured the world, there was another science, known to the linguistic s and language w orld as CMC. CMC is described by Herring (2007) based human to human interaction mediated by networked compu CMC includes, but is not limited to, cyberchat (real time: Instant Messenger Skype, Hotmail IM, Yahoo Messenger, Facebook chat, etc). Additionally, there are virtual chat room s for forums and discussions that students can subscribe to in order to communicate with native speakers of the foreign language that they are studying. Lik ewise, Computer Assisted Language Learning (CALL) has received a lot of attention especially in the field of Second Language Acquisition (SLA). Podcast and audio books are some of the CALL tools where learners of a foreign language can listen to a wide ran ge of words even on their mobile phones (which are very accessible) and listen repeatedly when they think they heard something different from what they pronunciation over a period of t ime. Some of the limitations of CALL, though, are that some learners assume there is no need for classroom instruction and tend to replace the latter with the use of technological tools. However, Chapelle (2001) acknowledged that many scientific discipline s, including education have been improved due to the
24 introduction of communication technology and addressed issues such as the effective use of software in language learning. However helpful these tools may be, they cannot be a replacement for explicit ins truction. Additionally, though there are lots of software to enhance second language pronunciation, there has been little literature about how these technological tools can be used to complement teaching in the language classroom (Lord, 2008). Perspective s on Second Language Teaching McCandless and Winitz (1986) provide evidence that during the beginning stages of L2 acquisition, broad acoustic input can improve pronunciation of the target language. Dansereau (1995) on the other hand, argued that advanced phonetics (use of phonetic symbols) should be kept out of the beginning and intermediate level classroom because students there are already weighed down by the fact that they are learning a new language. She rather encouraged the teaching of consonants, vo wels and In contrast to the observation, Elliot (1997), in his study, found that input alone did n cit instruction (technical terms pronunciation. There are several differing views on what to teach at these levels in the language cla ss even though all researchers agree that the different aspects of language pronunciation to be taught should be highly selective. Arragon (1985), in his study, did not find the importance of phonetics for intelligibility or improvement in communicative sk ills. However, those who adopt a communicative method believe that learners will
25 enough to be understood, and will be able to work on difficulties later, should they wish to continue with their language studies. Undoubtedly, there is a direct link between the attempt to attain better pronunciation in the L2 and state of mind. Stevick (1978) pointed out that in all too often, self conscious ness leads to tension, tension leads to poor performance, poor performance leads to added tension, and so on around a downward spiral pronunciation errors. More often than not when students are corrected in the presence of their counterparts, they become embarrassed and this in turn can be viewed as counterproductive. Krashen (1982) underlined useful guidelines which sho uld be used in deciding when ing to him, educators need to correct errors that impede communication or break down the clarity of th e message. Also to be corrected are stigmatized errors that may cause the most negative reactions and errors which occur repetitively He further mentione d that educators can provide learners with the correct form when correcting errors or may adopt the inductive approach, where the learner is guided to discover the right form himself. Hadley (2001) pointed out that when educators want to get learners focus ed on mastery of particular features of the language such as pronunciation, it will be more beneficial to give fairly direct and immediate feedback on correctness. Educators thus need to employ these feedback mechanisms to promote the most favorable growth in aptitude. However simple this may seem, it must be emphasized that due to differences in personality, learning styles and learning preferences, educators are often caught in
26 No netheless, p ronunciation errors when not corrected become embedded in the second language to a level that cannot be reversed after a long period of time as noted by Hammerly (1973) Thus, it appears that Ha dley was correct in arguing for direct and timel y intervention. On the other hand, what happens if learners are persistently corrected yet with no positive results? Should educators continue to model words or expressions repeatedly? Hammerly proposed that educators should let learners know exactly what the error is because simple repetition denies the learner of pieces of information they need to know in order to correct their errors efficiently. T his can Ellio tt (1995) also argued that correction should be immediate and consistent and should be done by the instructor but occasionally by fellow classmates as well since this may make them feel comfortable and more receptive to corrections because they are peers However, during oral presentations, learners should not be interrupted for corrections because by doing so students may not concentrate on form but on pronunciation alone. It would not be a bad idea if e native language phonology in order to help identify individual problems. Wong (1987) noted that of predicting what might be difficult for learners, but more importantly for helping us In this instance, e ducators would therefore be able to understand why a Spanish speaker, for example, would pronounce all consonants at the end of French words.
27 How P ron unciation S hould B e E xamined Pronu nciation, like any other aspect of language teaching, when taught, must be tested and the commonest form of test used in the language class is aural tests. This therefore involves collecting speech samples from learners in a laboratory setting or quiet env ironment with tasks ranging from reading alou d sentences or isolated words to participating in conversations. Linguists, in their research, record samples from participants and have scorers listen to them in order to rate them. Likewise, in a language clas sroom, students who take phonetics classes are made to record samples of individual words, then sentences, then short passages and conversations in the form of a dialogue with their partners. We hardly find learners in a language class at the beginning lev el being asked to record samples, perhaps because of time constraints and the fact that most educators presume that learners at the beginning level have not yet mastered pronunciation, therefore it would be irrelevant to give such a task to them. A new tes t was proposed a decade ago by Koren (1995) of pronunciation, including stress and intonation which reflect a foreign language researchers who study the connection between foreign language pronunciation and students at universities and teacher training colleges who specialize in a foreign language and more especially those who would be interested in teaching the foreign language they are studying in the future need to be familiar with a means for testing s and researchers will have to focus on a particular feature of pronunciation to be evaluated. In high schools and elementa
28 ages been reading aloud either individual words or sentences in a book, and then errors are corrected by teachers. For example, if learners are tested on the sentence below in order to verify t he in French, this sentence can be used not only for phoneme recognition and production but also to check intonation and stress. E ll e e st plus grand e qu e r e e st la plus p e tit e . It has however not yet been proven by researchers that the mere recognition and production of correct sounds in the target language can make learner s sound like native speakers. Conversely, Koren argues that when instructors give feedback on par ticular In other words, he argued that this way of testing pronunciation th eoretically risk s being misleading because although learners are able to rectify their errors, recognize and make a distinction between discrete phonemes, this alone would not make them attain a native like pronu nciation. Lado (1961) disagreed with Koren a nd argued that there is a difference between recognition and production, therefore these aforementioned aspects of pronunciation need separate tests. In his study, participants were scored for pronunciation, for stress and for intonation with his argument being that these are the three main characteristics of articulation.
29 CHAPTER 3 METHODOLOGY The present study investigates the effects of explicit instruction on development of pronunciation accuracy over a period of time, as measured during a reading alou d task. (See Appendix A). This section discusses the research methodology as well as some background information about the participants in the study. In addition, there will be a description of data collection procedures and treatment. The corpus was collec ted solely through reading samples and not interviews. Data will be analyzed both quantitatively and qualitatively, and issues pertaining to validity and reliability will also be discussed. Participants Participants fo r this study were undergraduate studen ts recruited from two different 3000 level (Intermediate) French classes at the University of Florida in the Fall semester of 2010. The age range of participants in this study was between 19 and 22 years with the exception of one 60 year old woman in the P honetics class. Very typical of French classrooms at the college level, females dominated males in this study, with a ratio of 6:1. The experimental group took Corrective Phonetics (FRE 3780L), a class that met twice a week (2 credits) and took other Frenc h classes concurrently (generally French sociolinguistics or French literature). These classes were taught in French. The control group, unlike their counterparts, took Composition and Stylistics (FRE 3320), which was also taught in French, and met three t imes a week (3 credits) in addition to the other French courses aforementioned excluding explicit phonetic training. In other words, the latter did not receive explicit phonetics training. All students from both classes were invited to participate in the study; I succeeded in finding 12 participants for
30 the experimental group and 8 for the control group. Since this was a comparative study on how French learners in the United States acquired French pronunci ation, the info rmation that was controlled was lea exposure to French outside of the classroom L earners who had been to France and other Francophone countries were exempted. It was assumed that all students had at least the same minimum proficiency level in French due to the fact that in the French department, students must satisfy certain prerequisites before being enrolled in higher level classes. It is possible that there were students who were at a higher proficiency level than others considering the fact that in a language class, there a re differing abilities. However, no further test of proficiency was conducted. In addition, all participants in this study were asked to sign an informed consent form which was approved by the Institutional Review Board (IRB) of University of Florida. Inst ruments Surveys were conducted after which students signed consent forms and completed questionnaires (Appendix B). Consent forms stated the risks and benefits to participants and also elucidated the reasons for the research but never addressed the specifi c variables being studied. The questionnaire was designed to give the researcher were either native speakers or who had French speaking relatives. For aptitude and affecti grammar, culture, vocabulary, etc) were to them. The participants recruited were eith er native speakers of American English or native Spanish speakers There were two other
31 participants who had Arabic and Telegu (a language spoken in India) as their L1 respectively. All these participants had neither been to France nor other francophone co to ensure that they had learned to pronounce French words solely from taking French classes in the United States. Judges A total of four raters were involved in this st udy: two native speakers (1 female, 1 male) and two non native speakers (the researcher and her thesis director). The two native speakers were selected based on the fact that they were both current residents of the United States and also graduate students who had some experience with the English language. The female rater had been in the US for five years and the male rater, for nine months and both had been exposed to the use of the French language in classrooms. Both were French teaching assistants at th e university where the research was conducted and had received formal training in linguistics as well. The researcher had been studying French for twenty years and the thesis director had also been teaching French at the college level for more than two dec ades. The corpus was initially rated by the researcher and then by the male rater following the same criteria. Rating was done by mark ing an item as incorrect each time it was realized incorrectly and indicating in parenthesis the exact phoneme realized by the participant. On the other hand, if the phoneme was realized accurately, it was scored as correct Participants were rated blindly to avoid any bias Due to the fact that there were differences after evaluations, the female rater stepped in to help val idate the results. When inter rater reliability was inconsistent for particular phonemes, two of the raters listened to specific
32 Sounds Examined The sounds selected for this present s tudy are among those that are believed to be most difficult for non natives, especially na tive English speakers. The study specifically focused on the pronunciation and linguistic discrimination of the French phonemes /u/ and /y/ and the production of Fren ch nasal vowels which are absent in th e English language. Waltz (1980) reports that though English has slightly nasalized vowels and thus are not distinctive in English. Learners o f French tend to encounter problems with the difference and production of the French vowel system with /u/ vs /y/ and nasals being very difficult to produce at the beginning levels The reason could be that because /u/ is present in both Spanish and Englis h, it is produced mostly to replace /y/ which is totally absent in both langu ages. Learners therefore mistakenly produce /u/ for most words that have the French front rounded vowel /y/ and sometimes vice versa. Both vowels were tested in initial, medial an d po st consonantal positions, for example [une, calculatrice, culture] respectively or [ouvert, cours]. Some vocabulary used in the short text to elicit the sounds tested included 1. [y ] chaussures, rue, calculatrice 2. [u ] coucou, doux, roue 3. [ ] matin, ital ien, sain 4. [ ] bonjour, rpondent son 5.  enfants, manquer, dans In total, seventy nine tokens of phonemes found in a short text were t ested: /u/, /y/ and // The focus was on the contrast between oral and nasal vowels and not whether participants were able to realize specific nasal vowels.
33 Data Collection D ata collection occurred in two phases; the first phase began the second week of classes and during this period, pretest data was collected. T he researcher visited each class in order to introduce and discuss the study in general terms and to emphasize the fact that participation was voluntary and anonymous and also that the task involved two recordings (at the beginning and end of the semester). Most participants in the experimental group recorded thei r reading samples just before their corrective phonetics class since the venue for that class was the language laboratory and those who got there early agreed to record their samples before class started. Those who could not record their samples made an a ppointment with the researcher to record their samples on different days within the first three weeks of the semester. Participants in this group did not need instruction on the use of the recording equipment because they were already familiar with it. Par ticipants were advised to read at their own pace and as if they were reading out loud in class. Recordings were saved in a folder which was later transferred to a compact disk for evaluation. On the other hand, participants in the control group were ins tructed on how to use the recording equipment since they had not taken classes that required their presence in the language laboratory. Collection of data for this group also took place within the first 3 weeks of the semester. All participants chose code names in order to help the researcher identify samples for analysis purposes. The pronunci ation pretest was administered several weeks before the experimental group received training in the targeted sounds while the posttest for both groups was administer ed during the fifteenth week of a sixteen week semester. For the purposes of comparability, the same text was used for both pre and posttests, and
34 recorded in the same language laboratory. The number of weeks between pre and posttest was twelve, thereby reducing the possibility that the outcomes evidenced would be attributed to memory of the passage. Preparation of Speech Samples Each judge listened to a tape that contained two sets of speech samples (pre and posttests); each consisted of the same short text with approximately sixteen sentences read by all twenty participants. There were fifteen tokens of [y], twenty six tokens of [u] and thirty eight words with nasal vowels that raters had to evaluate. Raters were advised to focus only on specific target s and not the entire text. The goal here was not to determine learner the target phoneme had been properly pronounced. Points were assigned as follows: incorrect pronunciation was scored as zero and the exact phoneme prod uced by the participant was indicated each time. A properly pronounced phoneme was awarded one point Raters chronologically evaluated the tape as it was recorded without knowing which sample belonged to either the corrective phonetics group or the compos ition and stylistics group. Scorers somet imes listened to the same token three to four times when in doubt, especially for the /u/ and /y/ tokens. Description of Curriculum T he Composition and Stylistics curriculum dealt with the study of style in written French. Throughout the semester, students wrote essays on various subjects (description, narration, summary and text explication). By the end of the semester, it is expected that students should have been taught to analyze both literary and non literary wo Generally, revision and in depth knowledge of grammar was essential for this class.
35 The general textbook used for the clas Fauve l which has all kinds of exercises ranging from comprehension activities to report dictionary, specifically Le Petit Robert was also highly recommended for the class. Evaluat ion for this class was as follows: At home essays 45%, In class compositions 15%, Exercises and homework 10%, Exams 20% and Participation 10%. T he Corrective Phonetics class placed particular emphasis on vowels, nasalization, rhythm, diphthongs a nd the French portion of the IPA ( International was designed purposely for individualized instruction. There were both oral and written exercises and students had to reco rd and listen to themselves and were then graded by their instructor. At the beginning of the semester, they studied phonetics theoretically before applying it directly to activities, but the majority of the class period was used for practice. I visited th e class quite often and for the most part, the instructor insisted that they produce sounds, both together and individually, paying attention to the place of pages ac cording to the syllabus and prepare before class. Occasionally, students were required to listen to French songs and figure out what they heard and also transcribe portions of newspapers. Evaluation for this class was as follows: Quizzes, Homework and Reco rdings 40%, three tests 15%, Transcriptions of newspapers 30% and Participation 15%. Hypothesis This study compared the performance over a period of sixteen weeks between two groups of university students taking intermediate French classes. The goal of this study
36 was to rate the accuracy of reading pronunciation pre and post instruction. The research sou ght to answer questions such as 1. What are the effects of explicit instruction on reading pronunciation (comparison between two groups composed of an experimental and control group)? 2. When students read the same French text, is there a significant difference in pronunciation accuracy between pre and posttests between these groups? 3. Does the context in which phonemes occur in a sentence affect pronun ciation? I hypothesized that there would be an overuse of the /u/ phoneme and an underuse of the /y/ phoneme due to the fact that all the particip ants in this study stated that English was used in their day to day activities (their residence, at school, etc) and also because the /y/ phoneme is totally absent in the English language. According to Elliott (1995, 1997) and other researchers mentioned previous ly in the literature reviewed above, explicit instruction plays an important role in acquiring L2 ph onological patterns. My hypothesis was therefore that the experimental group (Phonetics class) would produce more accurate pronunciation of the selected vowels especially by the end of the semester compared to the control group (Composition class). Due to instruction over the course of the semester, I expected both groups to make progress in French, which would also be expected to have a positive effect on their pronunciation. I fully expected that the experimental group would impro ve their pronunciation sk ills by the end of the semester because that was the main focus of this class.
37 CHAPTER 4 RESULTS AND ANALYSIS Data collected from the reading text was rated and statistical analyses using SPSS were run on these vowels [/y/, /u/] as well as nasal versus o ral vowels. For the purposes of determining equality of means, t tests were also conducted. Based on raw data, an overview of the distribution of tokens will be presented and discussed. As mentioned in the section on data collection, the text contained 15 tokens of the phoneme /y/, 26 tokens of /u/ and 38 tokens of nasal vowels. Table 1 illustrates the distribution of Each token was awarded one po int if the phoneme was properly pronounced, and zero points if it was not. In this way, an accuracy rate was calculated for each participant for each vowel, on both the pretest and the posttest. There were 12 participants in the phonetics class. As shown in table 2, the mean accuracy rate for the phoneme /y/ decreas ed from 13.25, (SD = 1.603) in the pretest to 12.17, (SD =3.689) in the posttest. This means that over time improv ement was not achieved, in that the mean score in the posttest was considerably lower compared to the pretest. This trait was very surprising given the fact that more attention was given to phonetics and the general accuracy of pronunciation among this group and also due to the fact that students were engaged in lots of pronunciation practice in and out of class. On e would wonder what accounted for these findings but it must be stated that one participant (Awesome) went from producing 11 out of 15 tokens accurately in the pretest to only 2 out of 15 correctly in the posttest. While som e participants showed less improvement and others showed no change at all at the end of the semester, it would be right to point out that individual variation is an
38 important factor in these evaluations. It is worthy to note that students also began the se mester with differing abilities and expectations and thus performed at different rates though quite close in certain instances. One would realize instantly from the raw data ( both at the begin ning and at the end of the semester. This, however, is not surprising since this happens to be one of the features which is somewhat common in a language class at the intermediate level. As previously mentioned, participant which deserve scrutiny She went from 73 .3% accuracy on the pretest to only 13.3% accuracy on the posttest in the production of /y/ Does this imply that she was better off not taking a phonetics class since she yielded poor results? Or was it that this particul ar phoneme was very difficult for this participant to produce? Not only did she not improve but she became markedly worse in pronouncing this phoneme. Even though instruction plays an important role in language learning, there are other factors such as the attitude toward the target language that affect the acquisition process. A critical look at her performance with respect to the other phonemes tells that she improved over time with the /u/ phoneme and correctly pronoun ced all the nasal vowe ls but got worse with the /y/ phoneme he communicates with her family and friends She only speaks French in class and hardly interacts with French speakers on campus. According to the questionnai re she completed before the recording, when she finds herself among people who speak French, she speaks the target language half of the time. However, she pointed out that pronunciation was very important to her French
39 studies. She stated that she did not listen to French news at all but sometime s watched French movies. It would not be exact to attribute this deficit to the influence of L1 on the target language, even though she falls within the group which has English as their native language For validati on purposes, t languages, but there was no significant confirmation to indicate that there was an influence of L1 in the findings even though it could be a reason for some of our participants. Could it be that t his participant and others became astounded by the phonetic training they received du ring the course of the semester? In the pretest, Awesome read at a normal pace and seemed at ease although she had lots of pauses and hesitation. She often repeated word s and also made a few self corrections but with the posttest, she showed confidence while reading but made a few back and forth self corrections. She made more self corrections in the pretest than in the posttest. However in the posttest, corrections usual ly were not in the right direction. In other words, she got worse with her corrections. She read at a fast pace but hesitated with the word Given that this participant will be observed throughout her French studies, one may hope that there would be an improvement in her pronunciation because she may have m ade modifications especially after receiving phonetic training and would therefore be able to distinguish between the front rounded vowel /y/ and the back rounded /u/ without slip ups With reg ards to the /u/ ph oneme we observe similar trends to those in the case of /y/. Participants in this group started off well at the start of the semester but ended up performing poorly by the end of the semester. On the whole, the mean accuracy rate for the pretest was 24.42 with a standard deviation of 1.165 indicating that there was little
40 variation among participants. This is confirmed by the range of scores: 22 to 26. Having observed this performance in the pretest, my expectations were that participants would be accurate in the posttest considering the fact that / u/ is present in English and also because they were near perfect in the pretest. However, this expectation was not met. If we look at the performance of for the pretest, even though she read with a low tone and at a normal pace, for the /u/ phoneme, she pronounced 25 out of 26 accurately but got worse in the posttest with only 15 tokens produced accurately However, in the posttest, although she exhibited confidence while reading, she made pauses before certain words that had the /u/ phoneme, which may indicate that she was self conscious and wanted to be correct but ended up hypercorrect ing certain words and thereby not producing the phoneme accurately like she did in th e pretest Instead of the expected improvement we observe that participants in the experimental group performed better in the pretest with regards to the /u/ phoneme than in the posttest. There was a decline in mean accuracy rating from 24.42 (SD = 1.165 ) to 22.17 (SD = 4.549). Two qualifications should however be made concerning the demand further examination There was a decline in accurate pronunciation of this phoneme from 23 to 19 and 24 to 16 respectively for these participants. Blue performed poorly over time with regards to /u/ and /y/ but showed great improvement with regards to the nasal vowels. This participant was uncomfortable during the first recording. She read wo rd after word with caution and had lots of pauses. Generally, t here were errors in the entire reading but because the focus of this study was not on the entire reading but only on
41 the elicited phonemes we only rated particular phonemes. I hoped that after instruction, Blue would be confident, but on the contrary, she still read with care and still at a slow pace with hesitations throughout. My presumption is that she naturally reads slowly and for her, it was difficult to tell if she benefited from the pho netic training she received or not but it is fair to point out that she did improve with nasal vowels Royal Blue also got worse over the course of the semester in the accurate pronunciation of /u / and /y/ but pronounced all nasal vowels correctly. This participant showed more confidence in the second recording than in the first, where he read with a low tone and seemed unsure of his pronunciation. Confidence seems to be an attribute which learners gained from receiving explicit instruction because there was a new sense of empowerment due to the fact that throughout the semester, students were introduced to phonetic symbols that made them aware of the place and mode of articulating these phonemes. That is to say, before instruction, learners had a level o f knowledge about pronunciation but during instruction, they became cognizant of how and where these phonemes we re produced Finally, there was undoubtedly no diffi culty in the correct realization of nasal vowels (Table 4 2) We know that there is an allo phonic difference between oral and na sal words in English. That is to say those vowels preceding nasal consonants are nasalized while those occurring elsewhere are not On the other hand, in French, there is a distinctive phonemic difference between the t wo. We gather from the results that participants had no problems with nasal ph onemes Students may have established the differences between oral and nasal vowels in the early years of study at the college level but did not master t he difference between /y/ and /u /. There is no doubt therefore
42 that the mean for nasal vowels 37.08 on the pretest and 37.83 on the posttest, is stable and near ceiling. It was hypothesized that the Composition class would maintain a level of pronunciation which was lower than their counterparts in the phonetics class considering the fact that they were not receiving explicit instruction and might not even consider pronunciation a very important aspect of L2 acquisition. For the control group, the mean accuracy rate for the /y/ phoneme in the pretest was 10.75, a mean which was lower than what the phonetics class exhibited in their pretest. It must however be stated that 6 out of the 8 participants produced 12 or fewer tokens of this phoneme accurately. Also, with regards to t he control group, responsible for the low mean in the pretest for /y/ since they both accurately produced only 7 and 6 tokens respectively. This brings up the issue of the effect of differing individual abilities in language classrooms. Lumos spoke both Spanish and English at her residence and stated that she had French classes daily, often interacted in French when she was among French speakers and found pronunciation in the target language extremely important. Red Girl, on the other hand, only spoke English and did not interact with French speakers at all and even when she was among students who spoke interacted in French even when she was addressed in the ta rget language. Even though she found pronunciation to be very important, she hardly watched French movies and never listened to French news. It is therefore expected that a learner with such an attitude towards the target language will not do well. However both participants improved markedly over time.
43 Nonetheless, what the data indicates is that overall, the participants in this group improved their performance toward the target pronunciation of the /y/ phoneme over the course of the semester with even th e aforementioned participants producing a total of 11 accurate tokens each. Considering the fact that the composition class is not designed to particularly focus on pronunciation, one is inclined to believe that continuous exposure to the target language i s beneficial for the accurate pronunciation of certain phonemes otherwise perceived to be difficult for learners. Some learners, such as however showed no change. In addition, using the recordings analyzed here as a measure of th eir general pronunciation abilities, they showed improvement in the correct pronunciation of /u/ over the course of the semester as well, with a mean of 22.13 in the pretest and a mean of 23.25 in the posttest. Again, Lumos produced 17 out of 26 occurrence s of the phoneme accurately in the pretest but improved to 23 which could indicate that there are learners who randomly produce phonemes without reall y knowing the mode and place of articulation hence these results. As observed in the experimental group, the control group also had very high and stable results for the nasal vowels: 37.00 in the pretest and 37.63 in the posttest. It seems learners are ab le to distinguish between oral and nasal vowels over time as they are exposed to the target language. This is true, at least in this study, where we have two intermediate classes who have taken first and second year courses in French at the college level w hich were not necessarily directed towards mastering nasal vowels (with the exception of the experimental group which did receive phonetic training ) but
44 somehow the participants figured out how to accurately pronounce most of them. My intention however, w as to look only at the difference between nasal and oral vowels and not to evaluate whether learners were using the right nasal vowel. Differences between the Experimental and Control Group Clearly, students in the Composition class outperformed their cou nterparts in the Phonetics class. T tests were performed to determine whether differences in mean between posttest and pretest for phoneme /y/ were significant or not. While participants in the control group (group B) had a positive difference in mean of 1 .75, showing improvement over time, the experimental group (group A) had a negative difference in mean of 1.08, reflecting their decline in acc uracy from pre to posttest ( Table 4 4 ). Recall that the latter had a mean of 13.25 before instruction, but scor ed 12.17 after instruction. On the other hand, the control group started off with a mean of 10.75 and ended the semester with a mean of 12.50 which is a striking change. When t tests were conducted for the pretest we found that the difference between the two groups with respect to /y/ was significant at p=.049 (t = 2.244, df = 9.947) In other words, the experimental group produced this phoneme better in the pretes t than the control group. Conversely, in the posttest, although there is evidence that the c ontrol group produced /y/ better than the experimental group ( a mean of 12.50 to 12.17 ) there was no significant difference, p = .791. Comparison between the Experimental and Control group As can be observed ( Table 4 5 ), an independent samples t test was conducted to compare the performance of the Phonetics class and the Composition class. Confidence levels were set at the 95% interval. A two tailed t test measuring for
45 differences between the experimental and control group indicates that the observed di fferences in mean are significant at p = .02 (t= 2.555, df = 17.830). Equally with regards to the phoneme /u/, the experimental group got worse from pre to posttest, with a negative mean of 2.25 and a standard deviation of 4.535, as is seen in table 6. This group was evidently outperformed over the course of the semester by the control group, who did not receive explicit instruction in phonetics. The control group, on the other hand, not only showed improvement in pre and posttest but performed better than the other group with a positive mean of 1.13. Once again, the results indicate that, contrary to early assumptions students in a phonetics class did not always succeed in improving pronunciation in the short term In the case of /u/, an independent samples t test does not confirm a significant difference between the two groups, (t = 1.921, p = .07, df = 17.767). However, there is reason to believe that if we had a larger data set, the result would be significant, given the near significance achieve d with only 20 participants. With regard to this phoneme, the pretest had the following means: the experimental group with 24.42 and the control group with 22.13 .This means that the experimental group performed better at the beginning of the semester than the control group thus when independent samples test was carried out, there was a significance of .054 The posttest had means of 22.17, SD = 4.549 and 23.25, SD = 1.488 respectively, which means the control group was better this time around than the expe rimental group. However, independent samples test concluded that the difference was insignificant (.456). Finally, with respect to nasal vowels, the results emerging from each group showed a high degree of consistency and demonstrated clearly that students had little
46 or no problems with the accurate pronunciation of nasal vowels. The mean accuracy rate for the phonetics class was .75 and that of the composition class was .63, indicating that there was a slight difference between these two groups. However, a n independent samples t test indicated that there was no significant difference between these groups (t= .402, p = .691, df = 17.879). As mentioned earlier, there was no significant difference between the two groups on both tests: the pretest showed that t he experimental group performed slightly better than th e control group (37.08 to 37.00 respectively ) p = .816. Likewise, the posttest showed no significant difference, p = 484 though both groups did better in the posttests than in the pretest s Regarding arget language ( Table 4 1) 58.3% of the participants in the phonetics class had English as their native language against 41.3% who had other languages (not English) as their native languages. The composition group on the other hand was made up of 62.5% native English speakers and 37.5% with other languages as their L1. While English speakers got worse by the end of the semester, those who did not have English as their native language improved considerably. How ever, when t tests were run for the various phonemes examined with respect to the L1 of all participants our findings did not indicate that those who had English as their L1 were less able to produce the /y/ phoneme than speakers of other native languages (in this case Spanish) neither did the results show that English native speakers produced the /u/ phoneme more accurately than others who did not have English as their L1 beca use of the existence of /u/ in English. 1 In other words, no role can be attribu ted to L1 interference in this study. 1 Spanish has /u/ and not /y/, however, just like English
47 Furthermore, we acknowledged that those who had French as their L3 generally performed better in the acquisition of the phonemes than those who were studying French as an L2. We presumed that perhaps they had the exper ience of learning other languages in addition to their L1 and thus were exposed to certain strategies in language learning. However, when t tests were run, the results did not show significant differences even though there were differences in means with re spect to whether Generally, the most common words that were difficult for our participants in all three categories are shown below (Table 4 10)
48 Table 4 1. Na tive languages of p articipants Native Language Phonetics C lass % Composition Class % English 7 58.3 5 62.5 Other 5 41.7 3 37.5 Total 12 100 8 100 Table 4 2. Descriptive statistics for the experimental g roup Group A Number of Participants Minimum Maximum Mean Std Deviation Y 1 12 10 15 13.25 1.603 Y 2 12 2 15 12.17 3.689 U 1 12 22 26 24.42 1.165 U 2 12 12 26 22.17 4.549 N 1 12 35 38 37.08 .793 N 2 12 37 38 37.83 .389 Total 12 Group A = Phonetics Class, y 1 = /y/ pretest, y 2 = /y/ posttest, u 1 =/u / pretest, u 2 = /u/ posttest, n 1 = // pretest and n 2 = // posttest Table 4 3. Descript ive statistics for the control g roup Group B Number of participants Minimum Maximum Mean Std. Deviation Y 1 8 6 14 10.75 2.866 Y 2 8 10 15 12.50 1.773 U 1 8 1 7 25 22.13 2.748 U 2 8 21 25 23.25 1.488 N 1 8 36 38 37.00 .756 N 2 8 36 38 37.63 .744 Total 8 Group B = Composition Class
49 Table 4 4 Difference in accu racy between the phonetics and c omposition classes for phoneme /y/ Group N Mean Std. Deviat ion Std. Error Mean Y Diff A 12 1.08 3.118 .900 B 8 1.75 1.832 .648 Table 4 5. Independent samples t est for /y/ Levene's Test for Equality of Variances t test for Equality of Means 95% Confidence Interval of the Difference F Sig. t df Sig. (2 tailed) Mean Difference Std. Error Difference Lower Upper Y Diff Equal variance s not assumed 2.555 17.830 020 2.833 1.109 5.165 .502 Table 4 6. Difference in accuracy between t he experimental group and the control g roup phoneme /u/ C N Mean Std. Deviation Std. Error Mean U Diff A 12 2.25 4.535 1.309 B 8 1.13 3.314 1.172
50 Table 4 7. Independent samples t est for /u/ Levene's Test for Equality of Variances t test for Equality of Means 95% Confidence Interval of the D ifference F Sig. t df Sig. (2 tailed) Mean Difference Std. Error Difference Lower Upper U Diff Equal variances not assumed 1.921 17.767 .071 3.375 1.757 7.070 .320 Table 4 8. Difference in accuracy between the Phonetics and Composit ion class (nasal vowels) C N Mean Std. Deviation Std. Error Mean N Diff A 12 .75 .866 .250 B 8 .63 .518 .183 Table 4 9. Independent samples t est for nasal vowels Levene's Test for Equality of Variances t test for Equality of Means 95% Confid ence Interval of the Difference F Sig. t df Sig. (2 tailed) Mean Difference Std. Error Difference Lowe r Upper N Diff Equal variances not assumed .403 17.879 .691 .125 .310 .526 .776
51 Table 4 10. List of words generally articulated wron gly List of /y/ phoneme List of /u/ phoneme List of Nasal phonemes puces doux machin chaussures roue pull vous rue cours sr une culture couture
52 CHAPTER 5 CONCLUSIONS AND DISCUSSIONS The purpose of this study was to investigate the effectiveness of explicit instruction on pronunciation accuracy in an intermediate level college French class. Results from the study indicated that explicit instruction did the elicited French phonemes (especiall y /y/ and /u/) which we considered to be difficult for learners of French The conclusio ns drawn from this study call for the development of more elaborate phonological acquisition theories of L2 learning and also serve as a foundation for recommendations for teaching practice. One may want to argue that f oreign language pronunciation should be incorporated in language curricula at the beginning of French studies, which means more emphasis will be put on correct pronunciation at the i nitial stages rather th an later. However in this study, we cannot propose that pronunciation should be a vital feature of language learning in the early stages eve n though there is slight evidence from the data that generally, our participants had no difficulty distinguishing be tween oral and nasal vowels. It is possible that learners figured out this distinction at the initial stages of learning French because for both groups, there was little or no problem with the nasal phonemes. The findings of this study actually showed tha t phonetic instruction did not improve pronunciation based on the selected phonemes but one could speculate that if basic phonetic training is integrated into future textbooks for beginning French classes it will be valuable since it is quite evident that learners become aware of certain tendencies in the target language even before they get to higher levels. In other words, it may be necessary for language institutions to include and focus more on pronunciation in French curricula at the introductory stag es of language learning.
53 As discussed in the chapter regarding previous literature, Scovel (1988) on, noting that if acquisition of L2 does not begin before puberty, learners cannot achieve a nea r native accent. T he overall results from this study support his position because all our participants were in addition did not show improvement in the acquisition of the elicited phonemes. Nonetheless taking into y years old it may be fair to argue that her results somewhat corroborat e Neufeld and Schneiderman who found in their study (1980) that adults are also able to achieve native like proficiency in the articulatory features of the second language. Results of p art that she improved over time in the accurate pronunciation of the /y/ phoneme but got worse in the accurate pronunciation of t he /u/ phoneme, while no change was observed over time with the nasal vowels. Generally in the experimental group, our participants performed better in the pr etests and got worse in the posttests. This tendency may be at tributed to the fact that students b egan the semester with some knowledge on how to pronounce certain phonemes correctly but during instruction, they became aware of the place and mode of articulation of these phonemes, therefore in an attempt to accurately pronounce them, they became overwh elmed and perhaps It can be hoped that after students have had some time for assimilation, their pronunciation will again get better this
54 but we partially observe that these factors may have contributed to the performance of our participants especially given the variations in pronunciation by our participants An (73.3% in the pretest against 13.3% in the posttest). However, this factor may not be the main predictor of L2 acquisition of phonological features in our study, perhaps du e to the nature of our elicited data [/u/ /y/ and //. Also, considering the fact that educators are responsible for correcting learners in a direct and timely manner, and more importantly in a corrective phonetics class, could these findings concur with ( 1978 ) that found that all too often, learners become self conscious when educators intervene and this in turn leads to tension and poor performance? It is probable that constant intervention affected some learners especially in the short term. Despite the fact that the results were not statistically significant enough to prove the hypotheses, the study suggests that explicit instruction in pronunciation claim (1991) that g pronunciation to realistic goals like developing self confidence. The findings of this study r un counter to the hypotheses of overall improvement over time of par of the selected French phonemes. This observation was especially striking when it turned out that those in the experimental group became worse over time compared to those who did not receive explicit instruction in Phonetics. T s position (19 85) that there are no benefits to explicit phonetic Elli ott (1997) who argued vehemently that basic instruction in the foreign language
55 ion, whereas explicit instruction yielded considerable results. T he context in which phonemes occur can affect pronunciation accuracy. We made this observation with regards to a question that appeared in the short text participants read for both the pre and posttest tasks Most participants had difficulties with the accurate pronunciatio n of /u/ when reading the inverted form of asking questions as we find i n the example below: vous One wonders if learners at the beginning years of learning French are mostly accustomed to th e construction (subject + predicate) because for most of our participants, when the verb preceded the subject in a question form they used the /y/ phoneme as opposed to /u/. Again, we noticed that when the sub question, some of our participants use d the /y/ phoneme instead of /u/ an example is subject). In the same light, when students read this sentenc vous vous allez souvent . most of our participants do not favor questions with stress in French with regard s to pronunciation. Other findings from this study concur with those of Ell is (1994), who considered a In the case of the /y/ phoneme, we saw that those who had English as their native language generally had troubles pronouncing the phoneme accurately.
56 it is pronounced differently in English and most of our participants even those who received phonetic training pronounced the word as it is pronounced in English and not in French [kuty R pronounced not in the French way [kylty R u R ] was o ften closely likened to On the whole, the most difficult phoneme for our participants was /y/ which is nonexistent in English. th hesitati ons and self corrections and our participants were very cautious, which may indicate that they acknowledged the fact that these were French words but their pronunciation leaned towards English and not French. Limitations The objective of this study was to examine the efficiency of explicit phonetic instruction. By doing so, we had to compare the performance of an experimental group and a control group over a period of time. First and foremost, one of the major concern s of the study wa s the size of the sampl e. We only succeeded in having twenty participants in total a number which appeared to be small for such a study. Evidently, the larger the sample, the more stable the results would be across similar samples. Small samples may sometimes not yield great r esults. Also, some tables showed relatively high standard deviations due to the considerable variance among the different participants. Perhaps, larger samples would help rectify this kind of variance thereby providing more definitive statistical results. Taken in their entirety, one may wonder how to really interpret the significant differences between groups, in the case of this study, of learners of French. What evidence is there to indicate that learners who improved over time on these tasks are
57 percei ved by native speakers as having adequate knowledge in phonological features? Equally, what evidence is there to suggest that those who performed poorly or did not improve over time lack the ability to acquire these new phonemes in an L2? Future research c an be conducted as to how a non native can combine phonetics and other aspects of the language in order to achieve native like p ronunciation Additionally, at the initial stages of evaluating samples, there was some level of inconsistency due to the fact t hat the researcher is not a native French speaker but the second rater is. However, to authenticate ratings, another native speaker listened to and evaluated samples and when there were still discrepancies, we resorted to the e research, greater training should be provided to different raters at the same time so that ratings would be done independently. However, there is no guarantee that this would address the issue of validity of ratings especially because this way of rating samples is often subjective. Even though the use of a spectrogram is highly recommended for such ratings, it is time consuming. Conversely, its use can address the discrepancy issues. Furthermore, what is lacking in this study is feedback from students who were in the phonetics class. A follow the explicit instruction they received. Even though instructors do not always receive constructive criticisms during student evaluations, it would be very intere sting to read comments written by students especially to help us know if they found the class and the material useful or not. Recommendations and Implications It would be fair to assume that most students went into the Phonetics class (an elective) with a good and positive attitude but due to the evidenced variations in
58 pronunciation, attitude would be one area of interest to study in future research. From uninteresting because s tudents routinely had to record samples at the beginning or at the end of class. Students may also think that taking a class with phonetic training the entire semester is demanding, therefore instructions should be fun and should take different forms such as periodic short reading quizzes based on the contents of the textbook, listening to the same selected songs over and over again for familiarity purposes and not only fo r the purpose of memorization. We believe that this can have a positive effect on pron unciation because the learner will be exposed to the target language and due to the friendly and relaxed environment in the classroom or language laboratory, studying will be more fun. language learning and ca n also serve as a linguistic exercise without learners feeling the urgent need for competition because of the relaxed atmosphere (Frelen, 1996). Also, phonetic training may be incorporated into French curricula and educators should introduce learners to ph onetic symbols but certainly not go into details like transcriptions which can be more overwhelming and astounding to students at those early stages of learning a foreign language. Finally, it will be very interesting if this study is conducted in a differ ent geographical area, perhaps the mid west or the northern part of the United States and it is probable that there would be different results especially considering the fact that in Florida, most students are exposed to foreign languages such as Spanish a nd Creole. For example, if this same study is conducted in a state such as New Jersey where there are lots of descendants of Italian immigrants or in Nebraska where there are German
59 settlers, we c ould have different results although we did not find the inf luence of L1 as a main factor that led to the findings of this study. We also had participants who were either bilingual or multilingual therefore it would be of interest to separate participants according to their native languages and only recruit partic ipants who were monoling uals learning a second language. To conclude, this study reveals that the use of explicit instruction on pronunciation merit s further exploration and study, including, perhaps, longitudinal follow up with learners to determine long term effects of instruction.
60 APPENDIX A FRENCH TEXT FOR PRE AND POSTTEST Je suis alle au march aux puces le week end dernier dans la rue Machin. Ma un pull trs doux, mais je lui ai achet un plat trs dur. Mon beau calculatrice pour son cours d Tout le monde arrive au march pour acheter de nouveaux articles mais mon amie Danielle va au march pour se promener. Tt le matin, je mange des aliments sains avec mon coucou James mais nos voisin s mangent avec leurs enfants. Bientt, ils vont dmnager et iront vers le centre du pays. Ils vont beaucoup nous manquer. Nous leur Et vous, vous allez souvent au march aux puces ? Vous aimez la vraie culture ou vous prfrez la haute couture ? Avez vous des voisins aussi ? Sont ils gentils avec vous ? Vous leur dites bonjour tous les matins ? Vous rpondent ils ?
61 APPENDIX B QUESTIONNAIRE Thank you for taking the time to answer this questionnaire. As a graduate student, I pronunciation is acquired in the United States. That is the reason why I ask you to take some time to answer the following questions. The answers will stay anonymous, and will only be used in my thesis, please g ive as much details as you can. Again, thank you for your help. __________________________________________ ____________________________ Code Name: Gender : M F 1. What language(s) do you speak at your current residence and/or with your family? 2. Ho w often do you interact with French speakers? Daily Once a week Sometimes Hardly Not at all 3. When you are among people who speak French, how often do you interact in French? Always Often Half of the time Sometimes Never 4. Have you e ver travelled to a French speaking country? If yes, indicate which one (s), when and for how long. 5. How important is vocabulary to your French studies? Extremely Very Neutral Somewhat Not really 6. How important is pronunciation to your French studies? Extremely Very Neutral Somewhat Not really 7. How important is grammar to your French studies? Extremely Very Neutral Somewhat Not really 8. How often do you watch French movies? (exclud ing homework)? Daily Once a week Sometimes Hardly Not at all 9. How often do you listen to French news on the internet? Daily Once a week Sometimes Hardly Not at all 10. Are you a member of any French conversation club?
62 DATA FOR C ASES Participant Code Name L1 Group Y 1 (15) Y 2 Y Diff U 1 (26) U 2 U Diff N 1 (38) N 2 N Diff Awesome E A 11 2 9 24 25 1 38 38 0 Blue E A 14 11 3 23 19 4 35 38 3 Blue 2 E A 14 15 1 24 22 2 37 37 0 Bugsy O A 15 15 0 26 26 0 37 38 1 Franklin J E A 12 11 1 25 25 0 38 38 0 Grasshopper E B 10 10 0 24 25 1 38 38 0 I love French O B 12 13 1 25 22 3 37 38 1 Livre O A 14 14 0 25 25 0 37 38 1 Lm E A 14 12 2 22 26 4 37 37 0 Lumos O B 7 11 4 17 23 6 37 38 1 Mercury O A 10 14 4 24 20 4 37 38 1 Midnight Blue E A 15 15 0 25 24 1 37 38 1 Noel E B 14 15 1 20 22 2 36 37 1 Major E B 12 12 0 21 25 4 38 38 0 Red Girl E B 6 11 5 22 24 2 36 36 0 Royal Blue E A 12 10 2 24 16 8 38 38 0 Sasha O B 12 14 2 23 24 1 37 38 1 Stars O A 14 15 1 25 12 1 3 37 38 1 Stripes O A 14 12 2 26 26 0 37 38 1 Grindylow E B 13 14 1 25 21 4 37 38 1 Y 1 = Pretest for /y/, Y 2 = Posttest for /y/ U 1 = Pretest for /u/, U 2 = Posttest for /u/ N 1 = Pretest for //, N 2 = Posttest for // L1 =Native language E = En glish O = Other languages (Spanish, Arabic and Telegu)
63 LIST OF REFERENCES Archibald, J. (1998). Second language phonology, phon etics and typology Studies in S econd L anguage A cquisition 20 ( 2 ) 189 211 Arragon, J.C (1985). The role of phonetics in language teaching. Some personal remarks Incorporated Linguist 24, 27 28 Beauvois M.H. (1997). Computer mediated c ommunication ( cmc ): technology for improving speaking and writing in Bush, M.D ( E d. ) T echnology en hanced language learning. The ACTL foreign language education series ( pp 165 184 ) Lincolnwood, IL: National Textbook Company Bergen, J. (1974). A practical framework for teaching pronunciation in beginning Spanish courses Hispania 57, 479 483 Bongaerts, T., Van Summeren C., Planken B. & Schi ls E. (1997). Age and ultimate attainment in the pronunciation of a foreign language Studies in Second Language Acquisition 19, 44 7 465 Caduner, S. & Hagiwara, M.P. (1982). La prononciation du franais international: A cquisition et perfectionnement New York: John Wiley & Sons Cenoz J. & Lecumberri, L.G. (1999). The effect of trainin g on the discrimination of english vowels International Review of Applied Lingui stics in Language Teaching 37 ( 4 ) 261 275 Ch ampagne Muzar, C. Sch nei der man E. & Bou r dages J. S (1993 ). Second language accent: The r ole of the pedagogical environment. International Review of Applied Linguistics in Language Teaching 31(2), 143 160 Chapelle, C. (2001). Computer appli cations in second language acqu isition: F oundations for teaching testing and research London : Cambridge University Press Chaudron, C. (1988). Second language classrooms : R esearch on teaching and learning New York: Cambridge University Press Colette, S G. La femme cache Paris: Galli mard, Editions Folio Dansereau, D. (1995). Phonetics in the beginning and intermediate oral proficiency oriented French classroom. French Review 68, 638 651. Dansereau, D. (2006). Savoir Dire: cours de phontique et de prononciation Houghton Mifflin : Seco nd Edition Dulay, H., Burt, M., & Krashen, S. (1982 ). Language t wo New York: Oxford University Press
64 Elliot A. (1995). Foreign language phonology : Field independence. Modern Language Journal 79 ( 4 ) 530 542 Elliot A. (1997). On the teaching and acquisition of pronunciation within a communicative approach. Hispania 80, 98 108 Ellis, R. (1994). The study of second language acquisition Oxford: Oxford University Press Flege J.E. (1984). The detection of French accent by American listeners. Journal of the Acoustical Society of America 76 ( 3 ) 692 707 Flege J.E & Hillenbrand J. (1984). Limits on phonetic accuracy in foreign language speech production. Journal of the Acoustical Society of Ameri ca 76( 3 ) 708 721. Gardner, R. C. (1985) S ocial psychology and second language learning : The r ole of a ttitudes and m otivation London: Edward Arnold Hammerly, H. (1973). The correction of pronoun errors. Modern Language Journal 57 ( 3 ) 106 110 Hammond, R. & Fl ege, J.E (19 9 0 ). The acquisition of second language phonological systems in a communicative framework The role of attitudes and experience. In B. Lawton & A. Tamburri ( Eds ) Romance Languages Annual 1989 1 ( pp. 671 676 ) Purdu e: Purdue Research Foundation Hardison, M. (2004). Generalization of computer assisted prosody training: Qua ntitative and qualitative findings. Language Learning and Technology 8, 34 52. Harlow, L.L & Muyskens, J. (1994). Priorities for intermediate level instruction. Modern Language Journal 78, 141 154 Herring, S.C. (2007). A faceted Classification Scheme for Computer Mediated Discourse Language @ Internet 4, article 1 Kenworthy, J. (1987). Teaching English pronunciation London, Longman Koren, S. (1995). Foreign language pronunciation testing : A new approach System 23 ( 3 ) 387 400 Krashen S.D (1982). Principles and practice in second language acquisition New York : Pergamon Press Kuhl, P.K (2000). A new view of language acquisition Proceedings of the National Academy of Science 97, 1185 0 11857 Lado, R. (1961). Language t esting London: Longm an Lado, R. (1964). Language t eaching: A scientific approach New York : McGraw Hill
65 Llisterri, J. (2003). La enseanza de la pronunciacin. Cervantes. Revista del Intitulo Cervantes en Italia 4 ( 1 ) 91 111. Long, M. H. (1983). Does second language instruction make a difference? A review of the research. TESOL Quarterly 17 ( 3 ) 359 382 Lord, G. (2005). Can we teach foreign language p ronunciation? On the effects of a phonetics class on second language pronunciation. Hispania 88, 557 567. Lord, G. (2008). Podcasting communities and second language acquisition Foreign Language Annals 4 1 ( 2 ) 364 379 Lord, G. (2010). The Combined Effects of Immersion and Instruction on Second Language Pronunciatio n. Foreign Language Annals 43 ( 3 ) 488 503. McCandless P. & Winitz H. (1986). Test of pronunciation following one year of comprehensive instruction in college Germ an. Modern Language Journal 70 ( 4 ) 355 362 McLaughlin B. (1977). Second language learning in children. Psychological Bulletin 84, 438 459 Misrachi, E. & Denney, N (19 79). Developmental study of foreign language pronunciation. Developmental Psychology 15 ( 4 ) 458 459 Morley J. (1991) T he pronunciation component in teaching English to speakers of other languages TESOL Quarterly 25 ( 3 ) 481 520 Neuf eld, G. (1980). nology. Teachers of English to s peakers of other Languages. TESOL Quarter ly, 14 ( 3 ) 285 298 Neufeld, G. & Schneiderman, E. (1980). Prosodic and articulatory feat ures in adult language learning In S Krash en, R Scarcella & M. Long (Eds). Research in second language acquisition ( pp 105 109 ) Rowley: Newb ury House O dl in, T. (1989). Language t ransfer: Cross linguistic influence in language learning Cambridge: Cambridge University Press Olson, L. & Samuels, S. (1982). The r elationship between a ge and a ccuracy of f oreign l anguage p ronunciation. In S. Krashen, R. Scarcella & M. Long (Eds.), C hild adult differences in second language acquisition ( pp. 67 75 ) Rowley, MA: Newbury House Omaggio Hadley, A. (2001). Teaching language in context Boston: Heinle & Heinle Pierrehumbert, J. (1990). Phonological and phonetic representation. Journal of Phonetics 18, 375 394
66 Piguet, Therese. (2001). Effects of phonetics/ phonics i nstruction on r eading p ronunciation (French Level 1 ). Florida International University, Miami, FL Pureell, E. & Suter, R.W (1980). Predictors of pronunciation accuracy: A reexamination. Language Learning 30, 271 28 7 Rerrell T. (1989). Teaching Spanish pronunciation in a communicative approach. In P. Bjart man & R. Hammond (Eds ), American Spanish p ronunciation (pp. 196 214 ) Washington, DC: Georgetown University Press Cahiers Franco 5 ( 1 ) Printemps 5 17 Schumann, J. (1975). Affective factors and the problem of age in second language. Language Learning 25, 209 225 Scovel, T. ( 1988). A t ime to s peak: A p sycholinguistic i nquiry into the c ritical p eriod for h uman s peech. Rowley MA: Newbury House Simoe s, A. (1996). Phonetics in second language acquisition: An acoustic study of fluency in adult learners of Spanish. Hispania 79, 87 95 Siskin, H. J., Krueger, C. & Fauvel, M. ( 2003 ) : French c omposi tion. Hough ton Mifflin : Second Edition Snow, C.E. & Hoefnagel Hohle M. ( 1977). Age differences in the pronunciation of foreign sounds. Language and Speech 20 (4), 357 365 Stevick, E.W. (1978 ). Toward a p ractical p hilosophy of p ronunciation: Another v iew. TESOL Quarterly 12 ( 2 ), 145 150. Suter, R.W (1976). Predictors of pronunc iation accuracy in second language learning. Language Learning 26, 233 253 Tarone, E. (1987). Some influences on the syllable structure of interlanguage phonology. In G Loup & S. Weinberger (Eds ), I nterlanguage phonology : T he acquisition of a se cond language sound system (pp. 232 247 ) New York : Newbury House. Tarone, E. (1978). The phonology of interlanguage. In J.C. Richards (Ed ), Understanding second and foreign language learning: Issues and approaches (pp. 15 33 ) Rowley, MA: Newbury House Tremblay, P. and Gardner, R. (1995). Ex panding the motivation construct in language learning The Modern Language Journal 79 ( 4 ) 505 518 Ushida, Eiko. (2005). The r ole of s a ttitudes and m otivation in s econd l anguage l earning in o nline l anguages c ourses. CALICO Journal 23 ( 1 ) 49 78.
67 Wong, R. (1987). Teaching p ronunciation: Focus on English r hythm and i ntonation Englewood Cliffs, NJ: Prentice Hall Regents
68 BIOGRAPHICAL SKETCH Edith Pare received her B.A. degree in French and Spanish from the University of Ghana, with other diplo mas from Alliance Franaise, Ghana. From 2004 to 2007, she worked in the banking sector before joining her husband in the United States to further her education. undergraduate French courses both at the begi nning and intermediate levels for the Department of Languages, Literatures, and Cultures. She received her M.A degree in French from the University of Florida in the summer of 2011.