BA-Project, RUC, Autumn 2007

Teaching Second-Language Phonetics to Adults

Prominent Second-Language Acquisition Theories and their Practical Implications

By: Ganna Vasylyna Hansen Superviser: Hanne Pernille Andersen

Table of Contents

Resume................................................................................................................2 Introduction.........................................................................................................4 Why Teach Pronunciation? ...............................................................................4 The Field of Research .......................................................................................5 Methodological Reflection .................................................................................6 Part I: Critical Period Hypothesis ......................................................................7 Introduction .......................................................................................................7 Age and Language Acquisition..........................................................................8 Critical or Sensitive?..........................................................................................9 Practical Implications.......................................................................................10 Part II: Contrastive Analysis Hypothesis ........................................................11 Introduction .....................................................................................................11 Language Transfer ..........................................................................................12 Contrastive Analysis on Supra-Segmental Level.............................................13 Second Language Learners and Interlanguage ..............................................15 Experience in Second Language Acquisition ..................................................16 Practical Implications.......................................................................................17 Part III: Speech Learning Model ......................................................................18 Introduction .....................................................................................................18 Age Factor in Second Language Acquisition...................................................19 Category Formation.........................................................................................20 Acoustic Cues .................................................................................................21 Speech Learning Model and Teaching Process ..............................................21 Part IV: Other Views on Second Language Acquisition ................................23 Introduction .....................................................................................................23 Perceptual Assimilation Model ........................................................................23 Native Language Magnet Model......................................................................24 Part V: Second Language Acquisition in Practical Terms ............................25 Introduction .....................................................................................................25 Formal Instruction in Second Language Acquisition........................................26 Counteracting Age Constraints in Second Language Acquisition ...................27 Making Use of the Adult Learners’ Cognitive Development ................................................. 27 ‘Where there’s a will there’s a way’: Motivation in Second Language Acquisition ............... 29

Fighting Negative Effects of First Language Influence ....................................30 Setting Priorities Straight ...................................................................................................... 30 Beginners versus Advanced Learners.................................................................................. 32

Conclusion ........................................................................................................34 Bibliography......................................................................................................37 Appendix ...........................................................................................................40

1

Resume

1

Formålet med dette projekt er at besvare følgende spørgsmål: på hvilke måder kan de mest fremstående teorier, der beskæftiger sig med tilegnelsen af fremmedsprog (andetsprog), hjælpe med at reducere de problemer, som voksne kæmper imod i processen af tilegnelse af Engelsk udtale. For at kunne opnå dette mål er to undersøgelsesniveauer involveret i projektet: (1) introduktion og diskussion af teoretiske positioner indenfor tilegnelse af andetsprog og (2) disse teoriers praktiske betydning.

Til brug i den teoretiske undersøgelse blev følgende tre indflydelsesrige teorier udvalgt: Critical Period Hypothesis (inklusivt Sensitive Period Hypothesis) (CPH/SPH), Contrastive Analysis Hypothesis (CAH) og Speech Learning Model (SLM). Gennemgang og diskussion af disse teorier blev suppleret med en kort introduktion af to relativt nye tilgange til området af andetsprogstilegnelse, nemlig Perceptual Assimilation Model (PAM) og Native Language Magnet (NLM). Disse er inkluderet med henblik på at give et bredere perspektiv i forbindelse med viden akkumuleret gennem undersøgelser i voksnes sprogtilegnelse.

Gennemgang af de teoretiske tilgangsvinkler har gjort det muligt at sammentrænge den teoretiske viden i to hovedpointer. For det første, har det fundamentale forarbejde for den videre undersøgelse været etableret på basis af introduktionen af de vigtigste synspunkter og mest repræsentative undersøgelser indenfor voksnes andetsprogstilegnelse. For det andet, har det for at kunne identificere de mest centrale problemer, som voksne oplever i tilegnelse af Engelsk som andetsprog, været nødvendigt at fremlægge de forskellige teoriers syn på de karakteristiske træk, der findes i voksens andetsprogstilegnelse.

De to hovedpointer har gjort det muligt at konkludere, at de mest centrale problemer for voksnes andetsprogstilegnelse ligger i perceptionen og produktionen af fremmedsprogs lyde (fonemer). Desuden har det været tydeligt, at ikke alle teorier har været ligeså opmærksomme overfor begge problemområder. Mens CPH/SPH og CAH primært er optaget af undersøgelsen af produktionen af fremmedsprogs lyde, ser SLM (samt PAM og NLM) det for relevant at undersøge fonetisk perception som et led i produktionen af lyde.

Derudover, hjælper gennemgangen af de teoretiske indfaldsvinkler i identifikationen af de faktorer, som teorierne mener at være årsager til de voksnes vanskeligheder med perception og produktion af fremmede fonemer. En af de centrale indflydelsesfaktorer, nemlig alder spørgsmålet, stammede fra den overvejende enighed om, at børn er bedre til at tage imod fremmede sprog, end voksne er. Aldersforhold er forbundet til problemet ikke kun på baggrund af 1

For resume in English see Appendix 1

2

de fysiologiske konsekvenser, der gør voksne mindre modtagelige mod tilegnelsen af fremmede fonemer. Det er også forbundet med det faktum, at udsættelse for modersmålets lyde (udtale) vokser

med

alder.

Som

konsekvens,

andetsprogstilegnelsen

være

undersøgt.

bør På

denne

den

modersmålets

baggrund

må

de

påvirkning mest

på

åbenlyse

indflydelsesfaktorer i forbindelse med andetsprogstilegnelsen afgrænses til (1) alderens indflydelse og (2) modersmålets påvirkning på tilegnelsen af andetsprog.

Eftersom både de centrale problemområder i voksnes andetsprogstilegnelse og deres årsager var defineret, kunne den anden del af denne undersøgelse, nemlig teoriernes praktiske implikationer for undervisning af andetsprog, tilgodeses. Denne del af rapporten, bestående af eksempler på mulige metoder, der kan være anvendt i undervisningsprocessen, gav mulighed for at definere fem væsentlige undervisnings-strategier.

For at kunne modarbejde den negative indflydelse af alder på voksnes andetsprogs-tilegnelse, er to fremgangsmåder foreslået. De empiriske undersøgelser, der er præsenteret i forbindelse med teoriers gennemgang, foreslår, at, i modsætning til børn, har voksne mennesker en kognitiv fordel i form af bedre udviklede analytiske evner. Det vil, altså, være fornuftigt at bruge de metoder, der gør brug af denne fordel i sprogundervisning for voksne. En anden løsning, som ligger mere implicit i de empiriske undersøgelsers resultater, taler for motivationen som en anden af mulige faktorer, der har indflydelse på voksnes succes i andetsprogstilegnelse. Af den grund virker det sandsynligt, at anvendelse af metoder, som forstærker motivationen, kan medvirke til mere effektiv andetsprogstilegnelse blandt de voksne.

De tre resterende undervisningsstrategier har at gøre med den effekt, som modersmålets fonetik har på tilegnelsen af lyde i andetsprog. En af løsningerne på dette problem var, at modersmålets fonetiske karakteristika skulle tages i betragtning i forbindelse med valget af andetsprogs fonetisk materiale til voksnes undervisning. Den anden tilgang består i at undervisningen kun bør fokusere på de problemområder, som også var meget udbredte/vigtige fonetiske områder i andetsproget. Den sidste forslag man bør tage i betragtning i forbindelse med undervisningsprocessen er niveauet af sprogkundskaben opnået i andetsprog. De empiriske undersøgelser lader til, at de forskellige niveauer af sprogkundskab har sine karakteristiske problemområder, og i den forbindelse må undervisningsmaterialet være tilpasset disse specifikke problemer.

3

Introduction Why Teach Pronunciation? Pronunciation is undoubtedly an irreplaceable element of language learning and teaching, but is it necessarily an important element? Indeed, pronunciation is often what is being perceived as redundant and is therefore often neglected and avoided by many teachers of second language (L2) (Brown, 1991: 1; Piske, 2007: 308). Why do some still claim the importance of native-like L2 pronunciation? What is it that makes pronunciation vital for the L2 learner? It is the accent. And accent can have a number of negative effects:

“Foreign accents … may make non-natives difficult to understand, especially in non-ideal listening conditions…. They may cause listeners to misjudge a non-native speaker’s affective state …, or provoke negative personal evaluations, either as a result of the extra effort a listener must expend in order to understand, or by evoking negative group stereotypes….”

(Flege, 1995: 234)

We could also turn this argument around by looking at what mastering native-like pronunciation can contribute with. First of all, native-like pronunciation allows a L2 learner, who also is a speaker, to be “understood by a listener at a given time in a given situation”, or in other words makes the speaker intelligible (Kenworthy, 1987: 13). Secondly, intelligibility contributes to effective communication, that is to say the kind of communication that is not a source of irritation and frustration caused by numerous repetitions and attempts to rephrase what has been said (Kenworthy, 1987: 15). An immediate result of an effective communication can be illustrated by means of the following quote:

“Ron, at the next table, wasn’t having much more luck. ‘Wingardium Leviosa!’ he shouted, waving his long arms like a windmill. ‘You're saying it wrong,’ Harry heard Hermione snap. ‘It's Wing-gar-dium Levi-o-sa, make the “gar” nice and long.’ ‘You do it, then, if you're so clever,’ Ron snarled. Hermione rolled up the sleeves of her gown, flicked her wand and said, ‘Wingardium Leviosa!’ Their feather rose off the desk and hovered about four feet above their heads. ‘Oh, well done!’ cried Progfessor Flitwick, clapping. ‘Everyone see here, Miss Granger’s done it!’” (Rowling, 1997: 127)

Although taken from the world of wizards and fairies, this example can still be a good demonstration of what happens when the speaker is being unintelligible as opposed to the

4

miracles of successful communication. Rowling’s allegory of the world helps to see in more clear terms the immediate effects of incorrect pronunciation, the effects that may not be seen as vividly in reality as they appear in the world of fiction.

This allegory also helps demonstrate that it is not only what we say, but also how we say it that matters, and that a good pronunciation is one of those small details of the context that, if neglected, may result in some of those negative shades and hues (mentioned by Flege) that inevitably color our speech.

The Field of Research There were a number of things that triggered my interest in the area of pronunciation. First of all, being a foreigner myself and having studied English, Spanish and Danish as three foreign languages, pronunciation, foreign accent and intelligibility of my own speech have inevitably been serious issues for me at least during the last decade of my life. This personal interest has been if not the most decisive element in the choice of the topic for this project.

My interest grew stronger after reading the compendium in connection with the series of introduction seminars offered for our inspiration in the beginning of the semester. There were two topics in particular that helped me to narrow down my focus. A seminar on English as an Academic Lingua Franca set me off thinking of the relation between intelligibility and accented speech. But more important to my inspiration was the text on Age of Acquisition and a Second Language in connection with the issue of language pedagogy.

One of the issues raised in this text was that according to some linguists the best way to master the pronunciation perfectly was to start foreign language acquisition early in one’s life. Yet there were linguists who did not believe that age mattered.

In connection with the issue of age in second language acquisition this text added also some other perspectives to these considerations. One of them was the issue of cognitive development in a child with the further reflections about language perception and its connection to the second language acquisition both in a child and in an adult.

The issue of learning languages developed into the interest in a related area: teaching phonetics. My prevailing interest in the adult second language acquisition produced the question of whether there are ways of teaching pronunciation that would give tools to an adult learner, so that s/he can master accent-free pronunciation in the second language. These further perspectives,

5

combined with my initial interest (which emerged from my own experiences with learning foreign languages), resulted in the series of questions: 1. What theories on phonetics teaching are there? Which of them are the most prominent? 2. What difficulties in phoneme acquisition do these theories focus on? 3. How and who do the theorists categorize as an adult L2 learner? 4.

What teaching strategies can be elicited as a result of reviewing of these theories? How can these strategies help eliminate the difficulties experienced by adult L2 learners?

These questions and considerations developed at last into an idea to examine the most prominent theories on second language acquisition in order to extract the knowledge that would be most applicable in the context of teaching phonetics to adults. In order to achieve this, the following question will need to be answered:

How can the existing theories on teaching English phonetics help eliminate the difficulties faced by the adult L2 learners in the process of phoneme acquisition?

Methodological Reflection This report is an account of a meta-theoretical inquiry into the most prominent theories on second language acquisition. By means of a careful investigation of the existing research in the field this work attempts to condense the knowledge into the most appropriate strategies of teaching English phonetics to adult learners of second language.

In order to achieve this aim three major and two less influential theories are chosen as a theoretical basis. The major theories are Critical Period Hypothesis (CPH), Contrastive Analysis Hypothesis (CAH) and Speech Learning Model (SLM). The choice of these theories is based on several grounds. Since 1950s they have been the predominant theoretical positions, which have been reflected in the amount of empirical work that the proponents of these theories have produced within the field of second language acquisition (L2A).

Three other concerns played a central role in the choice of the theories introduced and discussed in this report. These concerns are closely related to the very topic and the angle of this project. Dealing with the area of phonetics acquisition by adults, this project’s natural focal points has to be (1) age factor in acquisition of second language phonetics, (2) influence of first language phonology on the acquisition of second language, and (3) the mechanisms in relation to phonetic perception and production in adult learners of foreign language.

6

The additional two theories, which are Perceptual Assimilation Model (PAM) and Native Language Magnet model (NLM), are chosen due to their overlap with several of the major theories. These models provide deeper explanation of the results achieved through the research in the field of L2A.

The results provided by these prominent theoretical positions lead to the conclusions about the three concerns listed above. These conclusions then provide the frame for suggestions concerning the process of teaching phonetics to the adults. After presenting several concrete practical considerations, follows the conclusion on the five general teaching strategies with the focus on phoneme acquisition by an adult L2 learner.

Part I: Critical Period Hypothesis Introduction It is widely believed that children are better and faster learners of foreign languages than adults. It is also often assumed that the older a person becomes the harder it is to learn a second, third, etc. foreign language (Abello-Contese et al., 2006: 7). These assumptions are made especially in connection with foreign language’s pronunciation, which is often experienced as impossible to master by adult L2 learners on the level that would be equal to that of a native-speaker (Scovel, 2006: 32). One of the most contradictive and therefore much discussed theories within the field of second language acquisition (L2A or SLA) based on this assumption is the so-called Critical Period Hypothesis (CPH).

The following chapters are going to give an overview of CPH and its practical implications on the process of teaching English phonetics to adult learners. These are the questions that are going to be answered in the process: 1. What is a Critical Period (CP)? 2. Who does CPH define as an adult L2 learner? 3. What is a Sensitive Period (SP) and how does it differ from a Critical Period (CP)? 4. What are CP/SP’s implications for phonetics teaching?

7

Age and Language Acquisition With its roots in neurology, neurosurgery and neurolinguistics, CPH has made its way into linguistics, where it has been one of the most influential theories ever since the beginning of 1970s (Abello-Contese et al., 2006: 8). The early CPH proponents, Lenneberg, Penfield and Roberts, focused on the children’s first language acquisition (L1A), especially what concerned 2

“non-normal primary language acquisition” . The important basis of CPH argued for by its

prominent supporters can be summed up in the following quote from Second Language Acquisition and the Critical Period Hypothesis by Birdsong: “Because of progressive lateralization of cerebral functions and ongoing myelization3 in Broca’s area4 and throughout the cortex5 the neural substrate that is required for language learning is not fully available after closure of the critical period” (Birdsong, 1999: 3).

The mentioned lateralization process – the differentiation of brain functions as a result of brain’s development and ageing – is typically expected to be finished around the age of puberty (Singleton, 1989: 161). Children’s remarkable ability to learn more than one language without a foreign accent is attributed to the neuroplasticity6 which is exactly what is lost as a result of brain maturation process. According to Scovel it is neuroplasticity that allows children to learn the neuromuscular skills, which are responsible for gaining native-like pronunciation (Scovel, 2006: 34). 7

8

Nowadays technology including PET , MEG and other forms for brain scanning allow language researchers to form a picture of the brain activity under the influence of different stimuli avoiding direct surgical intrusion. Brain scanning in the recent neurolinguistic research has shown that when languages are learned before the age of 11, the foreign language pronunciation recognition happens in the same place within the Broca’s area, whereas when language acquisition happens after the age of 11 two different places within this brain area are involved in the process.

“Forklaringen er formentlig den, at sprogtilegnelsen inden 11-årsalderen for visse lydes vedkommende blokerer for udfoldelsen af andre lyde efter 11-årsalderen.” [The explanation is 2

Abello-Contese et al., 2006: 8 Myelination: The formation of the myelin sheath (covering) around a nerve fiber. Myelin: The fatty substance that covers and protects nerves. Myelin is a layered tissue that sheathes the axons (nerve fibers). This sheath around the axon acts like a conduit in an electrical system, ensuring that messages sent by axons are not lost en route. It allows efficient conduction of action potentials down the axon. (From MedicineNet.com) 4 Broca’s Area: an area in the human brain involved in language acquisition (Elbro, et al., 2004: 15) 5 Cortex: a part of audio-analyzing system of the brain, which is responsible for transformation of audio-waves into neural signals. (my translation of the term from Almuhanova ed al., 2007: 193) 6 Scovel, 2006: 34 7 Positron Emmissions Tomografi (Elbro, et al., 2004: 19) 8 Magnetoencefalogram (Elbro, et al., 2004: 22) 3

8

that where some sounds are concerned, the language acquisition before the age of eleven blocks 9

the attainment of other sounds after the age of eleven.] (Elbro, et al., 2004: 24)

This explanation is in line with another argument for the CP, which sees the cause of the CP in a developmental change in the brain leading to a natural reprioritization of particular skills and abilities. According to this view, after reaching puberty the language skills become less important or even redundant. Consequently language acquisition skills used primarily for L1A are so costly to the human body that its natural reaction is to get rid of the “lion’s share of genetic features” responsible for this faculty of the brain (Birdsong, 1999: 6).

So far the CP has been defined in terms of a terminal and irreversible change in the brain, which – once occurred – makes it impossible for an adult

10

L2 learner to acquire native pronunciation,

as well as other aspects of language, such as morphology and syntax. Yet there is a less deterministic viewpoint on the matter, which will be presented in the following chapter.

Critical or Sensitive? Even the most ardent proponents of the CPH had to admit and account for the existence of exceptions from the rule of lost neuroplasticity (Scovel, 2006: 36). An attempt to do so was made through the following argument so well, yet with its share of skepticism, described by Birdsong:

“deriving from “use or lose” the inference that if language learning faculty is used it will not be lost, this “exercise hypothesis” can also accommodate anecdotal accounts of individuals who start L2 acquisition early and continue to acquire foreign languages successfully into adulthood.” (Birdsong, 1999: 6)

Another viewpoint is based on a claim that after reaching puberty (a certain stage of brain maturation) an L2 learner starts to experience a decline in the abilities to learn L2 with a native pronunciation (Birdsong, 1999: 68). Thus, in this line of thought, there is no “well-defined window 11

of opportunity”

or the CP, but instead a sensitive period (SP). One of the otherwise strong

supporters of the deterministic CPH, Scovel, admitted the following:

“there must exist exceptional learners who can overcome the loss of neuroplasticity and acquire native-like pronunciation even as adults. Over the decades, some SLA researchers have gone to great lengths to document the existence of adult second language learners 9

My translation of an age after puberty 11 Birdsong, 1999: 68 10

9

whose second language spoken proficiency is indistinguishable from that of native speakers.”

(Scovel, 2006: 36)

An interesting study supporting this statement was conducted by Bongaert, et.al. (1997), where two groups of Dutch late learners of English (onset age of about 12) and a control group of English native-speakers were recorded reading out loud. These recordings were then judged for foreign accent by English native speakers. The judges were asked give a rate on a scale from 1 to 5. One of the groups consisted of very successful adult learners of English, primarily university students and university teachers of English. The results of this research showed that this group’s pronunciation was judged to be extremely close to that of the control group, the group of nativespeakers: the scores of the successful native Dutch learners of English ranged from 4.18 to 4.93 compared to the scores of control group ranging from 4.67 to 4.94 (Bongaerts, et al., 1997: 455456).

Another study from 2006 concerned with the age effects on L2 learners’ phoneme acquisition tested three different age groups of Basque schoolchildren, who received equal amount of hours of formal instruction in English language (Gallardo del Puerto & Lecumberri, 2006: 118). This study concluded the following: 1. With consideration to the CPH’s position of early advantage

12

the learners with an earlier

onset age did not perform better than the ones who started learning English around puberty. On the contrary the cognitive development of older learners resulted in an advantage where phoneme discrimination was concerned (Gallardo del Puerto & Lecumberri, 2006: 125). 2. The stable and continuous exposure to the formal instruction allowed older beginners to outperform the younger ones (Gallardo del Puerto & Lecumberri, 2006: 127).

With these studies in mind it is evident that the more deterministic view on CP can be seen as somewhat inadequate when talking about age influence on L2A. A sensitive period hypothesis (SPH) seems to point towards more and broader possibilities regarding practical implications of the CPH, which will be laid out in the next chapter.

Practical Implications The above statement supporting that the adult L2 learners can be at least just as successful foreign language learners as children does not necessarily rule out the possibility that some of the adult learners can prove the deterministic CPH to be true, at least as far as their particular cases 12

Gallardo del Puerto & Lecumberri, 2006: 125

10

are concerned. A teacher of English language (and phonetics in particular) has to take this into account, yet not automatically concluding that there is nothing else to offer for these adult learners.

As seen in the previous chapters CPH founds its claims mainly on the biological facts of brain development process concentrating primarily on natural conditions for L2A. Since human beings are dependent on their physiology, there should simply be no hope for adult L2 learners according to CPH. Thus, teachers simply should not bother to teach adults foreign languages.

Yet, the studies by Gallardo del Puerto & Lecumberri (2006) and Bongaerts, et al. (1997) suggest that, just like in George Bernard Shaw’s Pygmalion, there is a possibility for a good teacher to make a pronunciation expert out of an Elisa Doolittle. In this case a teacher needs to concentrate on the adults’ cognitive advantage and analytical skills as well as supply the adult L2 learners with tools that they can use in order to perfect the native pronunciation on their own. Furthermore, it can be argued that motivation could play a role in the success achieved in L2A by adult learners. Thus, if the design of the pronunciation teaching for adult L2 learners is to be built around SP theory, then even though the most fruitful period for L2A is over, an adult L2 learner can still be successful in learning native-like pronunciation, provided that efficient methods of teaching pronunciation are found and applied.

Part II: Contrastive Analysis Hypothesis Introduction Another crucial line of thought within the field of phonetics, and especially within phonetics teaching, is Contrastive Analysis Hypothesis (CAH). Being influenced by behaviorism and structuralism, the works of the theory’s pioneers, such as Fries and Lado, appeared around 1950’s (Brown, 2000: 207-208). In the following chapters this theory will be presented while answering the following questions:

1. What is CAH’s theoretical position? 2. Who does CAH define as an adult L2 learner? 3. What levels of comparison/contrast are relevant for contrastive analysis of two languages? (What difficulties do they try to eliminate?) 4. What are CAH’s implications for teaching phonetics?

11

Language Transfer The main idea of CAH is summed up in the following quote:

“We assume that the student who comes in contact with a foreign language will find some features of it quite easy and others extremely difficult. Those elements that are similar to his native language will be simple for him, and those elements that are different will be difficult.”

(Lado, 1957: 2)

In accordance with CAH the problems that an L2 learner may face in the learning process are rooted in the direct transfer between learner’s first language (L1) and the L2. Such problems occur when

“[w]e tend to transfer to … [foreign] language our phonemes and their variants, our stress and rhythm patterns, our transitions, our intonation patterns and their interaction with other phonemes.” (Lado, 1957: 11)

In case of pronunciation acquisition, if transfer between the L1 and the L2 is positive it does not result in a foreign accent in L2. By the same token CAH theorists conclude that the errors in language acquisition must be the result of the negative transfer between the two languages (Brown, 2000: 208).

There are two reasons for the negative transfer in phonetics. Firstly this may happen when the L1 sound system does not contain some sounds that exist in the L2. These sounds will result in difficulties for an L2 learner, e.g. in case of a Spanish learner of English, the sound /dȢ/ as in jam does not exist in Spanish and thus can be expected to be substituted with sounds /tȓ/ (as in chica) or /j/ (as in yuyo) (Kenworthy: 153). The other reason is that the sounds existing in both the L2 and the L1 may be used differently in the L2 as opposed to the L1 (Brown, 1992:10). For example, a Spanish learner can have difficulties with differentiating between English /d/ and /ð/. Both sounds exist in Spanish, but whereas in English it is the orthographical representation that helps differentiate between these two phonemes (in words, like those vs. dose), in Spanish the position of letter d in a word or phrase/sentence is decisive for the pronunciation. Thus the same letter d can be pronounced as /d/ in the beginning of a word díme or as a fricative /ð/ in the middle as in cada (Lado, 1957: 14).

In the process of language acquisition the creative mind of an L2 learner produces a distinctive language form called interlanguage (IL). Thus IL is a unique combination of learner’s personal

12

positive as well as negative transfers created in the gradual progress from L1 towards target language (TL)

13

(Beebe, 1984: 51-52).

It is CAH theorists’ belief that in order to lessen the negative transfer to IL, it is important for a teacher of foreign language to compare the language that s/he teaches (L2) with the learner’s native language (L1). In Linguistics Across Cultures: Applied Linguistics for Language Teachers Lado gives a recipe for comparison of sound systems on different levels: contrast of phonemes, syllable structures and patterns of stress, intonation and rhythm, etc. (Lado, 1957: 43-50). He claims that to conduct a careful contrastive analysis one needs to go through a three-stage process including:

“linguistic analysis of sound systems, comparison of sound systems, and description of troublesome contrasts.” (Lado, 1957: 12)

Such a comparison will make it possible for the teacher to find the potential difficulties that a learner can be expected to have in the process of L2A thus providing the teacher with the necessary tools for eliminating those difficulties (Lado, 1957: 2).

Contrastive Analysis on Supra-Segmental14 Level Much of the research within this field concerns the segmental (phonemic) level due to that

“prosodic errors in stress placement or rhythm, may contribute less to FA [foreign accent] than do segmental errors.” (Flege, 1992: 589)

Nonetheless CAH proponents claim the importance of contrastive analysis on prosodic (suprasegmental) level: a comparison of syllables, stress patterns, intonation, etc. between L1 and L2 (Lado, 1957: 43-50, Tarone, 1984: 63, Mairs, 1989: 260 and Young-Scholten, 1995:115). Although many of the prosodic patters may have an influence on L2 pronunciation, in this chapter the focus will be on the comparison of syllables and stress patterns in English as L215.

In The role of the syllable in interlanguage phonology Tarone points out that little attention has been paid to this issue in earlier works on contrastive analysis (Tarone, 1984: 63-64). She argues for the importance of researching the syllable structures by referring to the study by Briére (1966). 13

In this part of the report (Part II) the concept of L2 will be used interchangeably with a concept of target language (TL). According to Aoyama and Guion “…suprasegmental aspects of speech involve multiple levels above phonetic segments, including syllables, words, phrases, and sentences. …Prosodic features consist of multiple perceptual features such as pitch, stress, and length.” (Aoyama and Guion, 2007: 282) 15 See pages 25-26 (this report) 14

13

In this study English phonemes /Ȣ/ and /ŋ/ were investigated. Niether of these two phonemes occur word initially in English, yet American subjects in the study showed difficulty with pronouncing only /ŋ/ word initially. The possible explanation is that as opposed to /Ȣ/, which occurs syllable initially in English (for example, in pleasure), /ŋ/ never occurs syllable initially (as in singing). This is why only /ŋ/ presents problems for Americans attempting to pronounce the Vietnamese ngao. Thus, the results of this study suggest that it is not so much the position within a word as the position within a syllable that has an influence on language transfer (Tarone, 1984: 64).

Tarone then proceeds with the question of whether the learner transfers syllable structure directly from L1 or whether there are universal syllable structures influencing the transfer. She analyzes observational data from Korean, Portuguese and Chinese L2 learners as well as several earlier conducted studies

16

and concludes that

“…there is observational and experimental evidence in second-language acquisition research to suggest that the universal intrinsic structure of the syllable has some influence on the shape of 17

IL phonology. Further, there is evidence that transfer of NL

syllable structure rules into TL18

sequences does occur, and is also influential in shaping the IL phonology.” (Tarone, 1984: 71)

As for contrastive analysis of stress patterns a study carried out by Mairs (1989) can be mentioned. She analyses 30 interviews with Spanish learners of English with a special attention to the word stress (Mairs, 1989: 260-261). Mairs continues thoroughly analyzing Spanish versus English syllable boundaries and stress rules followed by comparison of these rules with the data gathered in the experiment. She finds, similarly to Tarone, that where syllable structure is concerned the learners’ L1 tendencies influence L2A (Mairs, 1989: 282).

Her findings on stress patterns though are cardinally different:

“The results of this analysis further suggest that while transfer of constraints on nativelanguage syllable structure may be important in developing interlanguage grammars and influencing a number of different levels of grammatical structure, the tendency to transfer stress rules from the learners’ native language may be less pervasive.” (Mairs, 1989: 282)

The data showed that Spanish L2 learners correctly stress antepenultimate or initial syllable in words like executive and magazine, which contradicts the stress rules of Spanish language, 16

Tarone 1984: 65-69 native language (NL). In this context means the same as L1. 18 target language (TL). In this context means the same as L2. 17

14

where stress would fall on the last (in case of ending with a consonant) or penultimate (ending with vowel) syllable (Mairs, 1989: 269).

Second Language Learners and Interlanguage In contrast with CPH, which claimed the age before puberty to be the most productive period for 19

L2A , CAH does not seem to focus on the clear age groups of the L2 learners. Although there are instances of mentioning adult L2 learners as opposed to children (for example, Gussmann, 1984: 33 and Beebe, 1984: 57) CAH theorists neither offer a specific definition of an adult L2 learner, nor do they argue for children’s exceptional ability to acquire languages.

What is being discussed in relation to an L2 learner is the process of interlanguage (IL) production. Whereas the inventor of IL concept, Selinker, sees it as a sum of “learner’s attempts 20

to produce a target language” , later proponents of CAH are more inclined to see IL as a

creation of an active and resourceful constructor, an L2 learner (Gussmann, 1984: 33).

In this constructivist perspective the explanation to learners’ negative transfer

“[m]ay be either that they constructed the correct representations but failed to implement them in the nativelike manner … or they constructed a representation close to the correct one but not close enough.” (Ard, 1990: 254)

This assumption is based on the results of earlier research conducted by Beebe (1984). Her study, based on audio-material from twenty-five native speakers of Asian languages (Japanese, Mandarin, Korean, Thai and Indonesian) learning English as an L2 on the intermediate level, clearly showed that many of the CAH’s claims did not correspond to the picture resulting from Beebe’s empirical data (Beebe, 1984: 54-59).

Firstly Beebe points out that in the experiment the learners’ errors rarely were results of phonemic substitution (a direct transfer of the L2 phonemes to the L1). For instance, only 3% out of 46% of errors were due to the substitution of /l/ with an r-like phoneme in the words containing English /l/ (Beebe, 1984: 55).

Next she disagrees with the belief that the errors are caused by the negative transfer of phonemes from learners’ L1 to IL. Beebe observes that there were

19 20

See pages 8-9 (this report) Beebe, 1984: 51

15

“two main types of errors in the present study which lead to ‘original’ variants: approximations and composites.” (Beebe, 1984: 56)

Composites were combinations of two separate phonemes pronounced one after another, for example both Chinese and Japanese subjects used a composite of /θs/ for /s/. As for approximations Beebe gives an example of Thai and Chinese subjects’ numerous rapproximations produced in attempt to pronounce the English /r/, yet neither of these approximations are phonemes in Thai or in Chinese (Beebe 1984: 57).

Beebe’s third point is that the L2 learners’ phonological repertoire

21

does not expand throughout

22

the process of L2 acquisition , but changes depending on the level of acquisition:

“With training, they [L2 learners] may add to their phonetic repertoire, and a large number of approximations may occur as they advance to the intermediate level. Finally, however, as they progress to an advanced level of proficiency, they will realize TL norms more frequently and eliminate some of the approximations they formerly relied on.” (Beebe 1984: 57-58)

As a final point Beebe argues that the issue of sociolinguistic variation is relevant when talking about IL. The data in her experiment shows that Thais’ use of /l/ and variations of /r/ strongly depends on the social context where these sound are being produced, not on the ability or disability to pronounce these sounds (Beebe, 1984: 59).

Beebe’s concluding statement is that a contrastive analysis can never be detailed enough to be able to account for all of the existing variations in pronunciation (Beebe, 1984: 60).

Experience in Second Language Acquisition A number of studies conducted by Flege and his colleagues have shown an interesting tendency amongst L2 learners having to do with the proficiency level of the L2 learners. In 1995 Flege, Munro and Fox conducted an experiment with two groups of native Spanish (NS) speakers of English and one control group of native speakers of English (NE). The NS subjects were divided into two groups on the basis of their proficiency in English language: NS-1 group consisted of experienced speakers of English while in NS-2 group were the inexperienced ones. The subjects in these three groups were to discriminate between Spanish and English vowels organized into triads. These included the Spanish /a/ and English /e/, /Ǻ/, /i/, /a/, /ǡ/, /æ/ and /ǫ/. 21

Beebe 1984: 57-58 Some of the CAH proponents believed that the range of IL sounds expands throughout the process of L2 acquisition (Beebe, 1984: 57). 22

16

The comparison of the results shows considerably higher level of discrimination of /a/-/æ/ and /a//ǡ/ contrasts among the members of NS-1 group compared to the members of NS-2. At the same time the percentage of correct discrimination of contrasts like /a/-/Ǻ/ and /a/-/i/ was almost as high as that of NS-1 and control group. These findings contradict CAH’s assumptions concerning the difficulties presented by the L2 sounds non-existent in learner’s L1. First of all, the results suggest that not all of the L2 sounds different from L1 present a difficulty in discrimination. Secondly, with practice and experience in L2 learners’ discrimination of foreign sounds may improve considerably23. A further conclusion that may be drawn from this is that contrastive analysis may be useful for predicting difficulties faced by inexperienced learners (beginners), while more advanced L2 learners may need predictions based on other theoretical grounds.

Practical Implications For the practical purposes CAH seems to offer wide possibilities for the process of teaching phonetics. To begin with, by means of conducting a contrastive analysis of the L1 and the L2 a fair share of predictions about an adult L2 learner’s difficulties can be made. But these predictions will often prove to be too numerous and imprecise due to that, as Flege’s research shows, not all of the L2 sounds without an equivalent in L1 inventory will be problematic for a L2 learner. Furthermore, the results obtained by both Beebe (1984) and Flege (1995) suggest that the L2 sounds that do not exist in L1 inventory do not necessarily present a permanent problem for L2 learner, meaning that L2 sounds do not get permanently substituted by the L1 equivalents. Experience with L2 allows adult learners to progress in their production of L2 phonemes. Thus a teacher has to bear in mind, that contrastive analysis may be more equipped for making assumptions for the L2 learners on the beginner level.

It is also clear from the above mentioned studies that the different levels (segmental versus supra-segmental) of contrastive analysis should be prioritized according to the influence that these levels have on the successful L2A. For instance, whereas stress patterns of two languages in contrast may be of little practical value, the phonemic and syllabic levels of contrast seem to have more importance for the acquisition of L2 sounds and therefore for the teaching process. Thus the studies by both Beebe (1984) and Mairs (1989) show that there is more to the interlanguage transfer than a simple transfer of L1 rules onto the L2 claimed by CAH. There seems to be several more intriguing processes

23 24

24

that may prove to be responsible for the accents

These results are in conformity with the results achieved by Beebe (1984), see page 16 (this report). See pages 20-21 and 23-25

17

in L2, and a mere comparison between L1 and L2 does not seem to be able to explain or account for them.

Finally, there is an important issue that has not been explored adequately by CAH’s theorists. Before attempting to pronounce a L2 sound, a learner needs to hear/perceive it. Perception is therefore an inseparable part of the process of L2 phoneme acquisition. It is therefore only natural to attempt to understand the mechanisms behind the L2 sound production through getting insight into the processes behind perception of a L2 sound.

Part III: Speech Learning Model Introduction Contrastive analysis has contributed considerably to the research within the field of L2A not only by developing methods for comparison of L1 with L2, but also by raising the question of relationship and influence between the two (Brown, 2000: 7). Contrastive analysis’ theorists may not have answered the question of how exactly L1 influences L2A and how the process of L1A differs from that of L2A, yet they generated background knowledge allowing a new line of research within L2A to appear and develop (Brown, 2000: 7-8). L2A researchers came to the conclusion that

“a comprehensive model of L2 phoneme acquisition must integrate not only a theory of second language acquisition and a theory of phonological representation, but also a theory of speech perception.” (Brown, 2000: 7)

This was the dawn of Speech Learning Model (SLM) with such scholars as Flege and Munro in vanguard (Brown, 2000: 7-8). To review the hypotheses and research within SLM the following questions will be answered throughout the subsequent chapters:

1. How does SLM differ from CAH and CPH? What are SLM’s hypotheses? 2. What are SLM’s views on early versus late L2 acquisition (age factor in L2A)? 3. What is the difference between L1A and L2A, if any? 4. What do SLM’s findings mean for the process of phonetics teaching?

18

Age Factor in Second Language Acquisition SLM model addresses the question of age’s influence on “the ultimate attainment of L2 pronunciation” by opposing Scovel’s proposal of CP. Flege claims that

“Adults may be as able as children to imitate foreign sounds.” (Flege, 1995: 236)

This is due to the fact that the sensorimotor processes allowing a child to learn L1 do not deteriorate as a result of brain maturation process suggested by CPH (Flege, 1992: 591). Although his argument suggests an adults’ continuous ability to learn languages in any ages, Flege cannot help admitting to the following paradox:

“At the age when children’s sensorimotor abilities are generally improving, they seem to lose the ability to learn the vowels and consonants of an L2.” (Flege, 1995: 235)

Flege attributes this transformation in child’s language perception to a change in strategies applied in the process of L2A. Based on a study conducted in 1989

25

Flege argues that while a

child of five to seven years perceives phonemes based on the vowel context and syllable size, late L2 learners are more inclined to concentrate on separate phonemes. Younger children’s sound production is more characterized by coarticulation than that of an older child or an adult. Thus with age children become more segment-focused in their sound perception and production (Flege, 1992: 592-593).

However, the results of a recent study by Schmidt (2007) contradict this assumption at least where sound perception is concerned. Twenty monolingual native speakers of English listened to the recordings of 228 Korean consonant-vowel (CV) sequences. The subjects were asked to write an English consonant sound representation that matched to that heard on the tape as well as grade the proximity of the sound heard to the English equivalent on a scale from 1 to 5 (Schmidt, 2007: 189). The results showed that, for example, two Korean phonemes, lax /s/ and tense /s*/, were identified as English /ȓ/ before /i/, but as English /s/ in combination with the rest of the vowels (Schmidt, 2007: 195). Thus adults also use vowel context to determine proximity of L2 phonemes to phonemes in their L1 (Schmidt, 2007: 198).

To explore the similarities and differences between child and adult language acquisition yet further, another important topic, namely categorization, needs to be addressed. 25

This study compared children of three to seven years to adult L2 learners in their production of English fricatives /s/ and /ȓ/. The researchers noticed a “decrease with age in the effect of vowel context on the fricative spectra” allowing to conclude that younger children tend to perceive whole syllables rather than separate phonemes, whereas older children and adults tend to perceive the latter (Flege, 1992: 592-593).

19

Category Formation Apart from the sensorimotor processes there is also another process which remains active and productive throughout a person’s life according to SLM. It is called phonetic category formation (Flege 1995: 239; Brown, 2000: 9). It is a process that starts already in the early childhood together with the first speech attempts and basically means that when a child attempts to pronounce his/her first words

“the phonetic/phonological properties appear consistently allocated to the same spot(s). … the mere fact that specific phonetic/phonological information is activated repeatedly in such a way that the same features appear consistently allocated to the same spot in the lexical item … triggers a process that groups those activities together that need to be coordinated to produce the desired effect.” (Wode, 1992: 623)

This allocation of phonetic material to the same phonetic space is what Flege calls creation of phonetic categories (Flege, 1992: 593-594). With time these categories’ boundaries expand including different (if any) allophones of the same phoneme, so that there is less and less uncommitted or unoccupied phonological space left. On the basis of this suggestion Flege proposes that in the process of L2A only those L2 sounds that are within unoccupied L1 phonological space will result in creation of new phonetic categories. These are the L2 sounds which Flege classifies as new (Wode: 614). It is Flege’s belief that only the formation of separate phonetic categories for the new sounds will result in an authentic pronunciation (Flege, 1992: 597).

SLM also operates with two other types of sounds: similar and identical (Wode, 1992: 614). These are the sounds which are most likely to be placed within learner’s already existing L1 sound categories. According to Wode

“Identical elements are perceived by the L2 learners as being the same; consequently, the L2 target is substituted by the L1 equivalent.” (Wode, 1992: 614)

As for similar sounds, the categorization will happen by means of the mechanism of equivalence classification, which allows the L2 sounds to be put into L1 category on the grounds of perceived similarity, yet despite the differences in phonetic features of the phonemes (Flege, 1995: 239).

A recent study by McAllister (2007) supports this hypothesis, at least with regards to the mastering of /s/-/z/ contrast by the Swedish learners of English. In this study 17 Swedish students were judged by 8 native speakers of English (either British or American) on their pronunciation of /s/ and /z/ in 8 words incorporated into sentences (McAlister, 2007: 156-157). In total 68 attempts

20

to pronounce /z/ were made, and only 15 of them were successful (McAlister, 2007: 160). This result can be explained by the absence of /s/-/z/ contrast in Swedish language. The apparent closeness of /s/ and /z/ allows Swedes to place these two phonemes within the same phonetic category obstructing the formation of a new sound category for /z/ (McAlister, 2007: 164).

Acoustic Cues There are other sound features which an adult L2 learner may use in categorization of nonnative phonemes.

“Mismatches between L1 and L2 sounds may also be described in terms of differences in the acoustic cues used to contrast phones.” (Flege, 1995: 265)

A good example of such use of acoustic features in L2 phoneme discrimination can be found in 26

earlier mentioned study by Schmidt (2007) . The results of this experiment were compared to a similar study by Schmidt from 1996, which allowed her to contrast the strategies used by Korean listeners of English sound to the English listeners of Korean. In case of categorizing nasal stops Korean listeners tended to use VOT

27

as an indication of nasality in production of English /b/ and 28

/d/, thus categorizing these sounds as Korean /m/ (Scmidt, 2007: 192). At the same time

“English listeners’ labeling of Korean nasals was associated both with talker, consonant, and vowel differences related to use of Korean nasal allophones.” (Scmidt, 2007:192)

This suggests not only that English and Korean listeners are using different acoustic clues when labeling the L2 phonemes, but that they base the choice of these clues on the acoustic particularities of their respective native languages (Scmidt, 2007: 199).

It is now time to sum up SLM in relation to the two other theories as well as putting SLM into perspective of phonetics teaching.

Speech Learning Model and Teaching Process To sum up, SLM does not absolutely disagree with the fact that early learners achieve better results in L2A. Yet, although SLM has been able to shed more light on the difference between

26

This study was second in the series of reciprocal researches conducted by Schmidt. The first study, conducted in 1996, entailed Korean subjects categorizing and grading similarity of English sounds (Schmidt, 2007: 187). 27 Voice onset time 28 There are no voiced stops in Korean (Scmidt, 2007: 192).

21

child and adult perception and LA process, the question of why children are better learners of L2 than adults still remains unanswered.

As opposed to CPH, claiming the cutting point of a critical period being around the age of eleven, SLM talks about changes happening in child’s perception already at about the age of five to seven. However SLM’s explanation as to what these changes entail seem to be inadequate.

Furthermore, SLM does not believe in the terminal and irreversible changes in a mature brain, depriving an adult L2 learner of possibility to improve his/her L2 pronunciation. Hence, as mentioned earlier, it is possible for an adult L2 learner to achieve accent-free pronunciation by means of linguistic experience and practice. The learner’s progress will in turn depend on those methods that the teacher will have to choose during the teaching process.

With respect to CAH’s conclusions, several new views on L2A were offered by SLM. Whereas CAH’s prediction was that all of the sounds in L2 that are non-existent in learner’s L1 will constitute a problem in the process of L2A, SLM proposes that this is only true with respect to the beginning L2 learners. Yet, the new sounds will most likely result in new category formation and will be mastered to perfection with time and experience. As to the experienced learners, it is rather the similar than new sounds that will be problematic due to the mechanism of equivalence classification. The identical sounds will be reproduced both by beginners and experienced learners close enough to the corresponding L2 sounds without disrupting intelligibility.

With this in mind the main focus of the teaching process will have to depend on the level of learner’s experience with an L2. The teacher’s aim will be to assist the learner in his/her attempt to discriminate the phonetic nuances of the L2 sounds in order to trigger the formation of new categories.

There are several issues, though, that have not been addressed or explained sufficiently deep by SLM. Although SLM suggests that the process of category formation plays a significant role in the perception of L2 sounds, the model does not address the question of how exactly the categories are formed or how stable their boundaries are. The fact that Flege’s research suggests that with experience the discrimination of sounds by adult L2 learners improves suggests that category formation is a dynamic process, and categories themselves do not have rigid boundaries.

Finally, audio-perception may not be the only factor playing a decisive part in the acquisition of native-like L2 pronunciation. There is a possibility that especially adult L2 learners (who have had extensive training in writing their L1) may be influenced by the L1 sound-spelling contrasts, if the

22

29

learner’s L1 is based on alphabetic principle . It is also plausible that foreign sound production may be influenced by articulatory habits connected to pronunciation of L1 sounds.

Part IV: Other Views on Second Language Acquisition Introduction In order to give a fuller account of the existing theories on L2A the following two chapters will contribute with a short introduction to two additional models, which (as well as SLM) focus on perception in L2A. These two models, namely Perceptual Assimilation Model (PAM) and Native Language Magnet model (NLM), each in its own way are related to SLM. While PAM provides an alternative way of looking at phoneme perception and establishing of phonetic categories, NLM gives a description of the mechanism behind these processes.

Perceptual Assimilation Model A different view on the relationship between the mature phonological system and L2A can be found in Perceptual Assimilation Model (PAM) developed by Best (Brown, 2000: 8). Best claims that the assimilation of sounds into native categories happens on the basis of

“…the spatial proximity of constriction locations and active articulators and by similarities in constriction degree and gestural phasing…” (Best, 1995: 194)

In other words, if learner’s L1 includes bilabial, alveolar and velar stops, but does not have dental ones, it will be more natural for him/her to attempt to pronounce dental stop of L2 with the place of articulation closer to the alveolar ridge (Best, 1995: 193-194). This will happen because the L2 learner will fail to hear discrepancies as well as similarities between the sounds of L1 and L2 (Best, 1994: 190).

Best lists four major patterns for assimilation of two L2 sound into L1 categories: 1. Two Categories assimilation (TC): L2 phonemes are assimilated to gesturally similar L1 categories, in which case the perception is expected to be very close to L2 target (Best, 1994: 191; Best, 1995:195).

29

“The most basic concept is that our [English] orthography is alphabetic, that letters relate primarily to sounds, not to syllables or meanings.” (Dickerson, 1992: 104)

23

2. Single Category assimilation (SC): both phonemes are assimilated to one L1 category, but at the same time both are equally distant from L2 target phoneme. Inadequate discrimination will be expected (Best, 1995:195). 3. Category Goodness is when “both non-native sounds are assimilated to the same native category, but they differ in discrepancy from native ‘ideal’ (e.g. one is acceptable, the other deviant” (Best, 1995:195). The expected discrimination is “from

moderate to good” (Best, 1995:195). 4. Nonassimilable type (NA): both L2 sounds are perceived as nonspeech sounds, and as a result cannot be assimilated into any of L1 categories (Best, 1994: 191; Best, 1995:195). The level of perception can vary, but is expected to be good (Best, 1994: 193; Best, 1995:195).

The following chapter will present an attempt to explain the mechanism behind the category formation process.

Native Language Magnet Model As well as SLM Native Language Magnet (NLM) theory claims that the exposure to the sound system of L1 results in phonetic category formation and by means of these categories alters language perception. NLM’s proponents also believe that this change happens already during the first year of infants’ life (Kuhl, 1995: 123). The model sees the input that the infant gets from his/her caretakers as a decisive factor for speech perception and production as well as for formation of gross categories of sounds (Kuhl, 1995: 139). The categories are formed around the so-called prototypes or “perceptual magnets”

30

which metaphorically speaking draw related

sounds towards them. If one were to visualize the model, it would resemble a magnet with a magnetic field around it: the magnetic field is strongest when very close to the magnet and 31

weakest further away from the magnet, at the boundaries of the magnetic field . By this principle:

“Perceptual magnets warp the acoustic space underlying phonetic distinctions by shrinking the perceived distance between good instances and surrounding stimuli, and stretching the perceived distance in the region of the phonetic boundary. This will cause certain perceptual distinctions to be maximized (those near the boundaries between two magnets), while others are minimized (those near the magnet attractors themselves).” (Kuhl, 1995: 142)

30 31

Kuhl, 1995: 123 http://www.coolmagnetman.com/magfield.htm

24

Although primarily based on the infants’ native-language magnet effects, this theory claims that the same magnet mechanism applies to adults’ sound discrimination (Kuhl, 1995: 137). With reference to Flege’s findings on category formation32 Kuhl agrees that

“it is possible to increase performance on discrimination of foreign-language contrasts in adults with extensive training.” (Kuhl, 1995: 142)

Although the exposure to the L1 phonetic information is essentially responsible for the perceptual changes in children as well as adults, the abilities to discriminate the phonetic nuances do not deteriorate with age (Kuhl, 1995: 142).

By drawing on the principle of magnetic field this model provides a more explicit explanation of how exactly the sound categories influence speech perception and production, and why the similar sounds pose most of the difficulties for the L2 learners (Kuhl, 1995: 142).

The two additional models, PAM and NLM, briefly presented here are included into this report only for the purpose of giving a fuller view of this work’s theoretical basis. Therefore in connection with practical implications that will follow in the next part these models will be used to a limited extend.

Part V: Second Language Acquisition in Practical Terms Introduction Taking the point of departure in the earlier presented theoretical views the following chapters will present and discuss a number of suggestions of the approaches which can be used for teaching English phonetics to adult learners.

This part will mainly focus on teaching strategies in connection with teaching of English phonetics on the segmental level. The reason for focusing mainly on phonemes is based on the limited 33

influence of suprasegmental elements on intelligibility of L2 learners . There will be instances, though, of mentioning teaching syllables and stress patterns of English due to that among other

32 33

See page 20 (this report) See page 13 (this report)

25

34

suprasegemental elements these are the most relevant to intelligibility

(Dickerson, 1992:111-

112; Tench, 1992: 97).

On the segmental level there will not be a uniform focus on acquisition of either vowels or consonants: both areas will be treated in form of examples in connection with proposals of teaching strategies. This wide scope is in conformity with the findings of CAH and SLM (also NLM and PAM) suggesting that the variety and the nature of L2 Learners’ difficulties are influenced by a number of factors including the particularities of learner’s L1, learner’s achieved level of proficiency in L2, the purposes of learning L2, etc. (Kenworthy, 1987: 3; Piske, 2007: 310).

Formal Instruction in Second Language Acquisition Before proceeding with suggesting possible teaching strategies, a significant issue must be discussed: is formal instruction at all fruitful in relation to L2 phoneme acquisition by adult learners?

The studies presented in the earlier chapters point, although indirectly, in different directions concerning this question. On one hand the empirical work in connection with CPH suggests clearly that pronunciation cannot be mastered on a native-like level after the age of around 11. Also SLM proponents are of the opinion that children of age even younger than 11 show an increasing difficulty with mastering certain L2 sounds. There is therefore no reason why formal instruction should show any positive result in sound acquisition by adult L2 learners.

On the other hand such studies as by Beebe (1984), Bongaert, et al.(1997) and Gallardo del Puerto & Lecumberri (2006) suggest that progress in phoneme acquisition can be achieved by means of formal instruction. Also Flege’s conclusions in connection with the study from 1995 show that experienced learners discriminate L2 phonemes better than the inexperienced ones. If formal instruction is considered a part of learners’ experience with L2 then this study also suggests that L2 learning in a classroom can contribute to the learner’s progress.

The discrepancy between these studies’ results suggest that: 1. Progress in sound acquisition by adult L2 learners can be achieved by means of formal instruction; 2. This progress must be dependent on a number of factors.

34

For example, such word as present will change its grammatical category and meaning depending on which syllable is stressed. Thus the intelligibility will depend on the correct usage of stress in such a word.

26

In the following chapters I will consider the factors/problems that appear to be of importance in L2A as well as come with suggestions for the teaching strategies to counterbalance them.

Counteracting Age Constraints in Second Language Acquisition Making Use of the Adult Learners’ Cognitive Development Looking back at the preceding overview of L2A theories, an adequate amount of empirical data has contributed to the overall agreement that early L2 learners prove to be the most successful ones in L2A in the majority of cases. The most obvious conclusion drawn from here is that in order to succeed in L2A a learner has to start as a child, and the earlier in childhood the better.

Yet it is not an option for a person who decides to start learning L2, say, at the age of thirty. As we could see there are at least four theoretical positions (SPH, SLM, NLM and PAM) which maintain that age may be an important, but not a decisive factor in L2A. The Sensitive Period theory, for example, even suggests that compared to a child an adult L2 learner has an advantage which is his/her analytical skills.

One way of making use of this advantage is by introducing phonetic symbols to the teaching/learning process. As Davidsen-Nielsen put it:

“For this purpose, it is pedagogically extremely useful to know the phonemes of the foreign language, for once the learner knows and masters the phonemes, she controls all linguistically relevant sound contrasts and will consequently be able to communicate freely.” (Davidsen-

Nielsen, 1994: 22)

To begin with a teacher can emphasize a clear distinction between sounds as opposed to their orthographical representations (letter ‘a’ being pronounced as /ei/) as well as sounds in connection with their occurrences in words (‘a’ in cake /keik/ as opposed to cat /kæt/). This can be achieved by an alphabet exercise described by Tench where letters with their respective sound representations are organized by the learners into a table, which reveals several tendencies of English pronunciation (Tench, 1992: 98).

The knowledge of English phonemes and phonetic script supplies the adult L2 learner with an invaluable key to the correct pronunciation of new English words. Provided that most of the English dictionaries give a phonetic transcription of the words they contain, a L2 learner can always check the pronunciation, and thus has a freedom to learn the new words, which s/he may come across outside the classroom (Tench, 1992: 96).

27

However, the phonetic script, as key to pronunciation, has its limitations. In order to achieve native-like pronunciation, a L2 learner needs also to distinguish between different allophones of a single phoneme. For instance, Danish learners often fail to pronounce dark /ǻ/ before the consonants, because of the missing distinction between a clear /l/ and a dark /ǻ/ in Danish (Mees and Collins, 2003: 175). A phonetic transcription found in most of the dictionaries will not supply a learner with details on use of allophones.

Thus, when deciding on such method as teaching pronunciation through phonetic transcription, it would be reasonable to consider the learners’ level of proficiency. For example, an introduction to English phonemes through minimal pairs is more suitable for the beginners, as it provides the knowledge of L2 phonemes which are new to the learner. For an advanced learner such method will be insufficient, as the discrimination between different allophones is a more relevant objective 35

for the advanced learners of English as a L2 . Besides being insufficient for some L2 learners teaching phonetic alphabet and consulting transcription in dictionaries can also be considered a “time-consuming and impossible strategy” (Kenworthy, 1987: 97).

Some adult L2 learners may find orthographical clues to pronunciation a more efficient way of learning English pronunciation. The efficiency of this strategy is in the fact that

“By memorizing a modest number of simple rules, adult learners can in fact become their own teachers.” (Dickerson,1992: 103)

This approach can be especially appreciated by the learners who have Spanish or Russian as their native tongue. The pronunciation in these languages builds on a principle where one sound corresponds to one letter. These learners may therefore see more sense in learning orthographical cues to pronunciation than the phonetic transcription, which is practically redundant in their L1’s.

Even though English language is known for its inconsistencies in correspondence between spelling and pronunciation, the alphabetic principle

36

provides a fair amount of clues as to how

words are to be pronounced. For example, seventeen out of nineteen English consonants each have one corresponding sound (Kenworthy, 1987: 102). 35

The importance of this objective will depend, though, on how native-like a learner wants his/her pronunciation to be (Kenworthy, 1987: 3) 36 “It [alphabetic writing system] follows the convention that there is a correspondence between letters and individual sounds.” (Kenworthy, 1987: 94) Chinese, for example, uses a writing system which does not provide any clues as to pronunciation.

28

However a teacher needs to watch out for overgeneralization/overanalyzing of orthographical clues to pronunciation. A final ‘e’ rule

37

, for example, can be useful for learning pronunciation of

vowels in monosyllabic words. Yet, a learner may use the rule in the areas where it is not applicable resulting in further mistakes. Kenworthy gives an example of French learners making such a mistake in a word like impressive. Since the word ends with ‘e’, a learner naturally assumes that the last syllable is pronounced with a diphthong /ai/ (Kenworthy, 1987:100).

Finally, a well defined, clear structure of each individual session/lesson with frequent reviews of earlier learned material could also contribute to making a better use of cognitive advantage of an adult L2 learner thus improving perception and recollection of new knowledge by an adult L2 learner (Chandler, 2006: 70). A regular material revision could be a good tool to use in order to satisfy adult L2 learners’ need for stable and continuous exposure to the formal instruction 38

allowing older beginners to do even better than the younger ones .

‘Where there’s a will there’s a way’: Motivation in Second Language Acquisition It is not completely unthinkable that some adult L2 learners may perceive phonetics as an unnecessary and uninteresting part of language learning resulting in lacking motivation for learning it. As Joanne Kenworthy put it:

“if you don’t care about particular task or don’t see the value of it, you won’t be motivated to do well. Learners may also be unconcerned because they simply are not aware that the way they speak is resulting in difficulty, irritation or misunderstanding for the listener.” (Kenworthy,

1987: 8) 39

Depending on the overall L2 learners’ aims with the language , a teacher of pronunciation can increase learner’s motivation level through, for example, introducing a discussion about the reasons for learning pronunciation to begin with, but also about reasons for achieving native pronunciation (Kenworthy, 1987: 54).

Another way to make phonetics learning more motivating is to make it less monotonous by spicing it up with more visual and audio material. For example, to make learning English phonemes more fun a teacher can show the class cards with phonemic signs, for instance /ð/, asking the class to come with possible examples of words containing this sound (Tench, 1992: 37

This rule states that a letter ‘a’ /ei/ will be pronounced as /ei/ (just like in alphabet) in a monosyllabic word ending with a silent ‘e’. The same rule applies to the rest of the English vowels in the described context. 38 See page 10 (this report) 39 A L2 learner may only be concerned with the written proficiency, if s/he only uses the language for business correspondence or the like (Kenworthy, 1987: 3).

29

98) or playing audio-material with native speakers of English pronouncing the words, like this, although, etc, simultaneously with the card being shown. 40

There are also different interactive internet media and tutorials concerned with phonetics , which could be of great help to a teacher as well as more fun for the learner. Tutorials like these can also be useful for the learner outside of the classroom as they allow the learner to master sounds without teacher’s help.

The importance of motivation as a factor in L2 phoneme acquisition by adult learners can be disputed, though. The few studies conducted in this area (which because of limited space are not presented in this report) do not lay sufficient grounds for any strong conclusions about adequate positive effect of motivation on acquisition of native-like pronunciation in L2 (Piske 2007: 309310). But some studies, for instance, earlier mentioned Bongaert, et al. (1997), indirectly suggest that motivation can prove to be a significant factor for L2 phoneme acquisition at least for some adult L2 learners. Being students or teachers of English on a university level, this study’s most successful subjects were naturally highly motivated adult learners of English with respect to all of its aspects, including pronunciation. These L2 learners “reported that it was very important for them to be able to speak English without a Dutch accent” (Bongaerts, et al. 1997: 462). It is

therefore reasonable to assume that motivation played a significant role in their excellent results.

Fighting Negative Effects of First Language Influence Setting Priorities Straight Several theories, for instance CAH and SLM, assume that L1 influences the acquisition of L2 in one way or another. Whether it is the substitution of L2 sound for one in learners L1 (CAH) or it is the lacking discrimination between L1 and L2 sounds that are similar (SLM and PAM), it is plausible that an adult L2 learner will face difficulties at least with some of the phonemes in the L2. Predicting L2 learners’ mistakes can be of help to a teacher of English as a second language (ESL), and a contrastive analysis of learner’s L1 and L2 can be of help in several ways.

First of all, a contrastive analysis may be of help in identifying exactly what kind of problems a learner faces and through that help assess the focus of teaching methods in relation to these problematic areas. For example, a learner may not be aware of the principles of word stress, if his/her native language is Chinese, where pitch fluctuation is a more meaningful suprasegmental element than stress (Kenworthy, 1987: 130). Contrastive analysis may in such instance provide a deeper insight into the distinctiveness of Chinese in comparison with English. As a result, a 40

For example, http://hctv.humnet.ucla.edu/departments/linguistics/VowelsandConsonants

30

teacher may find out that there may be other, more fundamental, principles that need to be introduced to a learner before the stress problems can be addressed. For example, it may be beneficial to start out by explaining the alphabetic principle of English to a Chinese learner and only then proceed by introducing stress rules (Dickerson, 1992: 104).

On the other hand, this example also emphasizes a negative aspect concerning CAH’s usability in identifying and eliminating of the learner’s problem areas. CAH stresses the importance of conducting a very detailed comparison of the L1 with L2 with the purpose of eliciting all of the learner’s potential difficulties. Yet, even though the learner may prove to have the predicted difficulties with the pronunciation of certain sounds, these sounds need to be assessed in terms of importance to L2A. As Kenworthy puts it:

“Some problems learners have need to be given high priority because they are vital for intelligibility; others do not affect intelligibility and can be given low priority.” (Kenworthy,

1987: 123)

An example of problems which can be given low priority in the teaching process is contrast between /u:/ and /Ț/. There are only a very limited amount of minimal pairs contrasting words containing these two phonemes in English, hence there is little reason in spending valuable 41

teaching time on discrimination of these sounds (Brown, 1991: 218) .

On the other hand in assessing the importance of the learner’s problem a teacher also has to take into account the specifics of the L1 itself. If we consider the earlier mentioned example with a Chinese learner of English, he or she is more likely to experience the difficulties with English stress patterns than a native Spanish learner of English (as, for example, can be seen in the 42

research by Mairs (1989) ). Difficulties with stress in the case of Chinese learner will need to be given a high priority, whereas a Spanish learner may face other challenges in connection with English pronunciation. An example of these could be the discrepancies between English spelling and pronunciation. A Spanish learner, used to predominant letter-sound correspondence with a couple of strict rules to account for the few inconsistencies, can find the irregularity of English pronunciation rather difficult to deal with. A high priority in such case needs to be given to the material helping the learner to distinguish the orthographical clues to pronunciation43.

However a real classroom context may not allow such differentiation of teaching focal points to take place. It is likely that the class will consist of learners with different L1 backgrounds in 41

Brown calls this an assessment of “functional load” (Brown, 1991: 212). See pages 14-15 (this report) 43 See page 28 (this report) 42

31

combination with only a limited time to spend on pronunciation teaching. In practice a teacher may end up with having time for focusing only on the difficulties that all of the students in a given class face, or only the pronunciation mistakes most relevant to the given students in a given 44

context .

Furthermore, the fact that it is practically impossible to account for all of the details of 45

pronunciation differences , a contrastive analysis is also a very time-consuming procedure. Conducting a contrastive analysis may take a lot of unnecessary effort and precious preparation time from a teacher, especially, if it turns out that a student does not, in fact, transfer certain patterns/rules of her/his L1 onto the L2. It may therefore be more effective first to identify the precise problem areas for the student, asses the functional load of the problem, and only then conduct a comparison of the languages, but only within the limits of the problematic areas.

Beginners versus Advanced Learners As the studies by Beebe (1984) and Flege (1995)

46

showed, the progress achieved by an

advanced L2 learner can in itself be another factor influencing the difficulties faced by the learners and in turn the necessary teaching methods. In other words teaching methods relevant for beginners should have different focal points than those more appropriate for the advanced learners of English.

For a beginning learner the sounds that are non-existent in his/her L1 are more likely to pose problems. A Danish learner of English, for instance, may substitute a sound like /θ/, which does not exist in Danish, with /t/ or /f/. This sound needs to be given high priority, because the lacking discrimination between /θ/ and /t/ or /f/, for example in words like three versus tree or free, will have a negative effect on the intelligibility of the learner’s speech (Mees and Collins, 2003: 174).

But according to SLM and NLM

47

(both emphasizing that the decline that can be observed in

adult’s L2 sound perception is not permanent) this sound has a good potential of being mastered. With more experience or phonetic training a learner begins to distinguish the difference between /θ/ and /t/ or /f/ and as a result establishes a separate category for the sound /θ/. Furthermore, looking at the mechanisms behind the category formation48, with the adequate L2 input also the discrimination of allophones can be trained. A beginning learner will most likely allocate both allophones of English /l/ (clear /l/ and dark /ǻ/) to the same phonetic space: that of Danish /l/. This 44

See page 30 (footnote 37, this report) See page 16 (this report) See pages 15-17 (this report) 47 See page 19 and 25 (this report) 48 See page 25 (this report) 45 46

32

mechanism will result in lacking discrimination between dark and clear l-allophones, both being pronounced as clear /l/. But an advanced learner can be more equipped to extend the existing 49

category for /l/ so that it will include dark /ǻ/ thus differentiating between the two allophones .

Especially relevant to this argument is the study by Neufeld (1979) (briefly mentioned by Flege) emphasizing the importance of mere listening for the sound perception and consequent sound production (Flege, 1995: 236). Along with SLM’s claims the study suggests that the use of audiomaterial can be efficient in order to focus the learner’s attention on the nuances of L2 pronunciation. It is needless to point out that the audio-material will have to consist of recordings of native English speech.

The use of L2 audio-material can be beneficial both for advanced and for less experienced learners. In case of a beginner, listening to the native speech will draw the learner’s attention to the new sounds, thus helping him/her to master the contrasts between L1 and L2 that affect intelligibility the most. As for an advanced learner, such material can help him/her to tune the perception towards the details of those similar sounds that are placed into L1 categories, yet need to be allocated to separate ones50. If an advanced learner is familiar with the phonetic script, s/he can also benefit from analyzing the transcription of a speech sequence before listening to it (Kenworthy, 1987: 116-117). This exercise can be useful for drawing more attention to the subtleties between similar sounds (and/or allophones). 51

If, as suggested by Best , an accented speech is viewed as based on L1 articulation influence, then a teacher may consider using head diagrams as a teaching material. Head diagrams are clearly a tool that would make use of the advantages of an adult learner’s cognitive development, as they will require a great deal of right brain side activity (the non-linguistic side if the brain) (Brown, 1992: 125). There are, though, two major disadvantages of this method. Firstly,

“[m]any of the head diagrams which have been published in books on phonetics and pronunciation teaching are seriously inaccurate in certain anatomical and physiological respects.” (Brown, 1992: 127)

And secondly, so far head diagrams have not proven to be very efficient for pronunciation teaching, due to that many adult learners find them rather confusing (Brown, 1992: 126). Rochet 49

See page 20 (this report) For example, in Danish /s/ and /z/ are allophones of the same phoneme, and are therefore, in Flege’s terms, both within the same extended category of Danish /s/. This results in that both /s/ and /z/ are often pronounced as /s/ by Danish learners of English (Mees and Collins, 2003: 174). In order to pronounce the English /z/ in a native-like manner, a Danish learner will have to establish a new category for /z/. 51 See page 23 (this report) 50

33

gives an example of native English learners of French making little sense out of articulatory instruction in pronunciation of French /y/:

“in spite of straightforward articulatory instruction – which consists of telling students to pronounce a vowel with their tongue bunched high in the front of the mouth and their lips rounded – L2 beginners often continue mispronouncing the target vowel /y/, substitute for it another vowel in imitation tasks, and usually fail to detect any difference between their incorrect pronunciations and the intended vowel…” (Rochet, 1995: 381)

Despite these weaknesses Brown argues that this method can be quite productive when used with advanced learners, and provided that several conditions are fulfilled. The main requirements are that the head diagrams and the attached terminology are kept as unambiguous and simple as possible (Brown, 1992:126).

Conclusion The aim of this research have been to find out in what way the most prominent theories within L2A can contribute to the process of phonetics teaching to adult L2 learners. In order to achieve this goal three major theories of L2A (as well as two less influential theoretical models) were accounted for and were put into a practical context of L2 phonetics teaching.

To begin with, on the basis of investigation of CPH/SPH, CAH and SLM it was possible to conclude that formal instruction can contribute to the successful acquisition of native-like pronunciation. Furthermore, this inquiry uncovered two central issues in the investigated area: the age factor and the influence of L1 on L2 phoneme acquisition. These issues are interdependent, especially with respect to those practical solutions that can be employed in the teaching process. The three theoretical positions agree that the onset age for learning L2 and the experience with L1 have an influence on the success in acquisition of L2 phonemes. The result of the influence can be observed in the difficulties with phoneme perception and production that are faced by most of the adult L2 learners. Explored by all three of the theories, phoneme production proved to be the predominant focal point of the research in L2A. As for perception, attention to this aspect has only been paid in studies within SLM (NLM and PAM).

Both perception and production were considered with respect to the possible teaching strategies. The main objective was to find the teaching strategies that would counteract adult L2 learners’ difficulties. As a result, it was possible to elicit five key strategies, two of which have to do with

34

counteracting age factor in L2A, and three with the L1 influence on the L2 phoneme acquisition. These results are summed up in the following table:

Influencing

Difficulties with

Possible teaching

factor

L2 sounds

strategies

Age

CPH/SPH:

• Exploiting adult

production;

learner’s cognitive

SLM (PAM, NLM):

advantage

perception and production

• Employing methods to increase motivation • Assessing the

Examples • Transcription; • Spelling vs pronunciation contrast • Visual material (for example, head diagrams) • Clear structure with regular material revision • Focus of teaching material will

L1 influence

CAH: production;

on L2A

SLM (PAM, NLM):

problems in terms

depend on learners L1

perception and

of L1

background (Only the actual

production

difficulties are addressed) • Assessing the functional load of

• Only most frequent phonemic contrasts are given high priority

the phonemes • Assessing the focus in terms of the proficiency level

• Audio material • Articulatory instruction and head diagrams • Transcription Beginners: • Focusing on new sounds Advanced learners: • Focusing on similar sounds

The findings suggest that the age constraints can be tackled by means of exploiting adult learners’ advantageous cognitive development. This will involve methods focusing on the rational, the systematic aspects of language, such as rules and consistent patterns in the sound system of English language. Also means of motivation improvement can prove useful in teaching phonetics to the adult learners of L2.

35

As for L1 influence, this problem can be helped by means of adjusting the teaching process with respect to several factors. The distinctive traits of L1 phonetics, the functional load of the problematic areas and the level of learners’ proficiency in L2 are the three central factors that need to be considered when choosing the appropriate methods of teaching. Filtering the problematic areas by means of these considerations can make teaching process more efficient.

The concrete approaches, presented in the table under examples, overlap in terms of the problems that they are meant to overcome. Thus there are teaching tactics that may prove to be efficient in overcoming both the age constraints and the difficulties caused by the influence of L1. The choice of the actual means to efficient L2 sound acquisition will depend on the strategies considered by the teacher.

From a more broad perspective it should be considered that the conclusions drawn here are based only on grounds of theoretical data. The theories used as a basis for conclusions are built largely on the empirical research in the field of L2A which has been conducted in simulated conditions, and may therefore prove to be of limited application to a real classroom situation. Therefore the assumptions presented above are not universal truths, but rather propositions whose efficiency needs to be proven in practical context.

36

Bibliography •

Almuhanova, A. et al. Bolshaya Psihologicheskaya Entsyklopedia (The Large Encyclopedia of Psychology) (2007), Moskva (Moscow), Russia: Eksmo

•

Aoyama, K. and Guion, S. G. (2007) Prosody in Second Language Acquisition. In O. Bohn and M. Munro (Eds.), Language Experience in Second Language Speech Learning: In Honor of James Emil Flege, Amsterdam, The Netherlands: John Benjamin Publishing Co.

•

Ard, J. (1990) A Constructivist Perspective on Non-Native Phonology. In S. Gass and J. nd Schachter (Eds.), Lingusitic Perspectives on Second Language Acquisition, 2 Edition, New York, USA: Cambridge University Press

•

Beebe, L. (1984). Myths About Interlanguage Phonology. In S. Eliasson (ed.) Studies in Descriptive Linguistics: Theoretical Issues in Contrastive Phonology, Heidelberg, Germany: Julius Groos Verlag

•

Best, C. (1994) The Emergence of Native-Language Phonological Influences in Infants: A Perceptual Assimilation Model. In J. C. Goodman and H. C. Nusbaum (Eds.) The Development of Speech Perception, Massachusetts, USA: Massachusetts Institute of Technology Press

•

Best, C. (1995) A Direct Realist View of Cross-Language Speech Perception. In W. Strange (Ed.) Speech Perception and Linguistic Experience: Issues in Cross-Language Research, Baltimore, USA: York Press

•

Birdsong, D. (ed.) (1999) Second Language Acquisition and the Critical Period Hypothesis, London, England: Lawrence Erlbaum Associates, Publishers

•

Bongaerts, T., van Summeren, S., Planken, B. and Schils, E. (1997). Age and Ultimate Attainment in the Pronunciation of a Foreign Language. In A. Valdman and S. Gass (Eds.) Studies in Second Language Acquisition, 19, 447-465, USA: Cambridge University Press

•

Brown, A. (1992). Twenty Questions. In A. Brown (Ed.) Approaches to Pronunciation Teaching, volume 2, number 2, Singapore: British Council

•

Brown, A. (1992). Head Diagrams. In A. Brown (Ed.) Approaches to Pronunciation Teaching, volume 2, number 2, Singapore: British Council

•

Brown, C. (2000) The Interrelation between Speech Perception and Phonological Acquisition from Infant to Adult. In J. Archibald (Ed.), Second Language Acquisition and Linguistic Theory, Oxford, UK: Blackwell Publishers

•

Brown, H. D. (2000). Principles of Language Learning and Teaching, 4 Edition, New York, USA: Addison Wesley Longman, Inc.

•

Chandler, P. M. (2006). Critical Period or Language Learning Difficulties: Are There Two Types of Adult Foreign Language Learners? In C. Abello-Contese et al. (Eds.), Age in L2 Acquisition and Teaching, Bern, Switzerland: Peter Lang

•

Davidsen-Nielsen, N. (1994) An Outline of English Pronunciation, Odense, Denmark: Odense University Press

th

37

•

Dickerson, W. B. (1992) Orthography: A Window on the World of Sound. In A. Brown (Ed.) Approaches to Pronunciation Teaching, volume 2, number 2, Singapore: British Council

•

Elbro, C. et al. (2004). Hjernen og sprog, Denmark: Hjerne Forum

•

Flege, J. E. (1992) Speech Learning in a Second Language. In C. A. Ferguson, L. Menn and C. Stoel-Gammon (Eds.), Phonological Development: Models, Research, Implications, Baltimore, USA: York Press

•

Flege, J. E. (1995) Second-Language Speech Learning. In W. Strange (Ed.) Speech Perception and Linguistic Experience: Issues in Cross-Language Research, Baltimore, USA: York Press

•

Gallardo del Puerto, F. and Lecumberri, M. (2006). Age Effects on Single Phoneme Perception for Learners of English as a Foreign Language. In C. Abello-Contese et al. (Eds.), Age in L2 Acquisition and Teaching, Bern, Switzerland: Peter Lang

•

Gussmann, E. (1984) Contrastive Analysis, Substantive Evidence and the Abstractness Issue. In Eliasson, S. (ed.) Studies in Descriptive Linguistics: Theoretical Issues in Contrastive Phonology, Heidelberg, Germany: Julius Groos Verlag

•

Kenworthy, J. (1978) Teaching English Pronunciation, London and New York: Longman

•

Kuhl, P. K. and Iverson, P. (1995) Linguistic Experience and the “Perceptual Magnet Effect”. In W. Strange (Ed.) Speech Perception and Linguistic Experience: Issues in Cross-Language Research, Baltimore, USA: York Press

•

Lado, R. (1957) Linguistics Across Cultures: Applied Linguistics for Language Teachers, th 10 Edition, USA: The University of Michigan Press

•

Mairs, J. L. (1989) Stress Assignment in Interlanguage Phonology: an Analysis of the Stress System of Spanish Speakers Learning English. In S. Gass and J. Schachter nd (Eds.), Lingusitic Perspectives on Second Language Acquisition, 2 Edition, New York, USA: Cambridge University Press

•

McAlister, R. (2007) Strategies for Realization of L2-categories. In O. Bohn and M. Munro (Eds.), Language Experience in Second Language Speech Learning: In Honor of James Emil Flege, Amsterdam, The Netherlands: John Benjamin Publishing Co.

•

Mees, I. M. and Collins, B. (2003) Sound English, Denmark: Handelsskolens Forlag

•

Rochet, B. L. (1995) Perception and Production of Second-Langiage Speech Sounds by Adults. In W. Strange (Ed.) Speech Perception and Linguistic Experience: Issues in Cross-Language Research, Baltimore, USA: York Press

•

Rowling, J.K. (1997) Harry Potter and the Philosopher’s Stone, London, Great Britain: Bloomsbery Publishing Plc

•

Piske, T. (2007) Implications of James E. Flege’s Research for the Foreign Language Classroom. In O. Bohn and M. Munro (Eds.), Language Experience in Second Language Speech Learning: In Honor of James Emil Flege, Amsterdam, The Netherlands: John Benjamin Publishing Co.

38

•

Schmidt, A. M. (2007) Cross-Language Consonant Identification. In O. Bohn and M. Munro (Eds.), Language Experience in Second Language Speech Learning: In Honor of James Emil Flege, Amsterdam, The Netherlands: John Benjamin Publishing Co.

•

Scovel, T. (2006). Age, Acquisition, and Accent. In C. Abello-Contese et al. (Eds.), Age in L2 Acquisition and Teaching, Bern, Switzerland: Peter Lang

•

Singleton, D. (1989) Language Acquisition: The Age Factor, Clevedon, England: Multilingual Matters Ltd.

•

Tarone, E. (1984). The Role of the Syllable in Interlanguage Phonology. In Eliasson, S. (ed.) Studies in Descriptive Linguistics: Theoretical Issues in Contrastive Phonology, Heidelberg, Germany: Julius Groos Verlag

•

Tench, P. (1992) Phonetic Symbols in the Dictionary and in the Classroom. In A. Brown (Ed.) Approaches to Pronunciation Teaching, volume 2, number 2, Singapore: British Council

•

Young-Schilten, M. (1995). The Negative Effects of ‘Positive’ Evidence on L2 Phonology. (pp.107-123) In L. Eubank, L. Selinker and M. S. Smith (Eds.) The Current state of Interlanguage: Studies in Honor of William E. Rutherford, Amsterdam, The Netherlands: John Benjamin Publishing Co.

•

Wode H. (1992) Categorical Perception and Segmental Coding. In C. A. Ferguson, L. Menn and C. Stoel-Gammon (Eds.), Phonological Development: Models, Research, Implications, Baltimore, USA: York Press

Websites: http://www.medterms.com/script/main/art.asp?articlekey=33824 http://www.medterms.com/script/main/art.asp?articlekey=4477 http://hctv.humnet.ucla.edu/departments/linguistics/VowelsandConsonants http://www.coolmagnetman.com/magfield.htm http://hctv.humnet.ucla.edu/departments/linguistics/VowelsandConsonants

39

Appendix Appendix 1: Summary This report is concerned with the area of language acquisition by adult learners of English as the second language (L2). The aim of this work is to find out in what way the most prominent theories on L2 acquisition can help reduce the impact of those problems that adult learners encounter in terms of L2 sound acquisition. In order to achieve this goal, two lines of inquiry were involved in the process: the overview of existing theoretical positions on L2A and their practical implications.

In the course of the theoretical investigation three major theories on L2 phoneme acquisition were introduced, discussed and put into practical perspective. These three theories are Critical Period Hypothesis (including Sensitive Period Hypothesis) (CPH/SPH), Contrastive Analysis Hypothesis (CAH) and Speech Learning Model (SLM). Also two relatively new models, Perceptual Assimilation Model (PAM) and Native Language Magnet (NLM), are briefly introduced in order to provide a broader perspective with respect to knowledge accumulated in the field of L2A.

Conducting a theoretical inquiry provides the basis for condensing the theoretical knowledge to the two main focal points. First of all, the basic framework for the investigation is constructed by means of the introduction of the main claims and representative studies within the field of L2A. Secondly, with the purpose of identifying of the main problematic areas for an adult L2 learner the different theoretical views on an adult L2 learner are outlined.

These focal points in turn lead to conclusion that the main L2 learners’ difficulties addressed by the theories have to do with perception and production of L2 sounds. It is also clear that not all of the theories pay equal attention to both problematic areas. Whereas both CPH/SPH and CAH primarily focus on researching the production of L2 phonemes, SLM (as well as PAM and NLM) finds it relevant to investigate the perception of L2 sounds as well.

Furthermore, the investigation of the different theoretical views helps to find out what main factors causing difficulties with perception and production the theories pay the most attention to in their research. One of the central influencing factors, namely the issue of age, emerges from the overall agreement among the theoretical positions that the children are better learners of L2 than adults. The age factor is not only connected to the area because of the physiological effects of age on the adult learners’ success in L2A. It is also interconnected with the fact that with age the exposure to the sound system of L1 increases. Consequently, the effects of this exposure to L1 on the acquisition of L2 sounds are investigated, and as a result two most obvious aspects

40

influencing the acquisition of L2 sounds are identified to be (1) the influence of the age and (2) that of the learner’s first language (L1).

Once identified, the adult L2 learners’ problems and the factors causing them allow focusing on the other line of inquiry: the practical implications of the theories on the process of teaching phonetics to adult L2 learners. This part of the report, consisting of the examples of possible methods that can be applied in phonetics teaching, provides the foundation for eliciting of five central teaching strategies.

In order to counteract the negative influence of age on the acquisition of L2 sounds two approaches are suggested. The investigated theories provide empirical data suggesting that, as opposed to children, adult learners have a cognitive advantage in form of his/her analytical skills. Exploiting these therefore seems to be an efficient way of teaching L2 sounds to adult learners. Another solution, lying more implicitly in the results of the same empirical data, suggests motivation as a plausible factor having effect on L2A by adults. Thus, employing methods of increasing motivation seems a reasonable teaching strategy.

The three remaining teaching strategies have to do with the influence of the sound system of L1 on the acquisition of sounds in L2.

One of the solutions to this particular problem is the

assessment of L2 phonological material with respect to the learner’s L1 sound system. The other approach recommends to asses the L2 phonemes within the learner’s problematic areas with respect to how often these phonemes appear in L2, and give high priority only to those L2 phonemes most frequently used in L2. The last issue to be addressed in connection with L1 influence on L2A is that of learner’s proficiency level: the teaching material has to be adjusted with respect to the specific problems characteristic for different levels of proficiency in L2.

41

Appendix 2: Structure of the Report

Research Question

Theories: • Claims and research within the area of L2A • View on L2 learner: Learners’ difficulties • Discussion

Theory

Practical examples

Result / conclusions

CPH

CAH

SLM

PAM / NLM

Implications for teaching process

Teaching strategies

42