The effectiveness of pronunciation teaching to Greek state school students Eleni Tsiartsioni Aristotle University of Thessaloniki E-mail: [email protected].

The effectiveness of pronunciation teaching to Greek state school students

Eleni TsiartsioniAristotle University of Thessaloniki

E-mail: [email protected]

14th GALA Conference,Thessaloniki 14th -16th December 2007

2

REVIEW OF RESEARCH ON PRONUNCIATION TEACHING

• ‘the teaching of pronunciation has been considered almost a luxury in the ESL/ EFL curriculum’ (Goodwin, Brinton and Celce-Murcia 1994)

• ‘an orphan in English programs around the world’ (Gilbert 1994)

• ‘pronunciation has been marginalized within the field of applied linguistics. As a result, teachers are left… with little direction’ (Derwing and Murno 2005)

3

Pardo’s (2005) conclusions on pronunciation teaching

• There is a positive effect of well-planned, quality pronunciation training (out of the 25 studies reviewed, 23 reported improved pronunciation after instruction).

• Use should be made of specific teaching techniques; pronunciation is not simply ‘picked-up’.

4

Age of onset in Foreign Language (FL) contexts

The BAF Project (University of Barcelona):• Age of onset in FL contexts and exposure

were not determinants for perceiving and producing L2 sounds in a native-like manner.

• Later starting age tended to result in better pronunciation and perception

• Increased input and explicit phonetic instruction beneficial in FL contexts.

(Fullana 2006)

5

THEORIES OF PHONOLOGICAL ACQUISITION

• Theories of Markedness (Eckman 1987)Unmarked features: voiceless stopsMarked phenomena: aspiration, vowel duration differences

• ‘New’ vs ‘similar’ sounds, ‘equivalence classification’- ‘Speech Learning Model’ (Flege 1986, 1991, 1997)Present study: the acquisition of word initial stops involves a ‘similar’ contrast.

6

TYPES OF MEASUREMENTS:

Voice Onset Time (VOT): the time interval between the release of the stop’s closure and the onset of voicing, which may precede or follow the burst

Aspiration: the voiceless noise interval between the release of the stop and the onset of glottal vibration (sounds like a puff of air)

Vowel duration before word final stops

7

Differences between the Greek and English stop system

GREEK:• Initial stops:

-Voiceless unaspirated /p, t, k/, short VOT

-/b, d, g/ fully voiced, negative VOT

ENGLISH:• Initial stops:

-Voiceless aspirated /p, t, k/, long positive VOT-/b, d, g/ voiceless, or partially voiced, unaspirated, short VOT

• Vowel length: vowels longer before voiced stops than before voiceless stops

8

QUESTIONS OF THE STUDY:1. Is pronunciation teaching to children

effective in a foreign language environment?

2. Does age play a role in the acquisition of pronunciation in a foreign language environment? For example, are students of younger age more favourably predisposed to acquiring FL pronunciation, as happens in naturalistic second language settings?

3. Are some phonological features acquired more successfully than others? To what extent are the theoretical models confirmed?

9

METHODOLOGY OF THE STUDY (1)

Control group (N=12): no pronunciation teaching

Experimental group (N=12) 12 pronunciation lessons embedded in

the English classes at school (20-40 minutes each)

Subjects:o 4 ten-year-old children (Age-group A)o 4 thirteen-year-old children (Age-group B)o 4 fifteen-year-old children (Age-group C)

10

METHODOLOGY OF THE STUDY (2) Subjects’ proficiency level: A1, A2, B1

Onset age of FL learning:9 years old

Recordings before and after the teaching intervention (Time 1,Time 2)

Recordings of L1 Greek and English by 6 native Greek and 6 native English children (2 of each age group)

11

Recording materialsfor English and Greek

• For English: Pack, Tab, Cab, Bag, Dab, Gab Gap-Gab, Bat-Bad, Back-Bag

• For Greek :Πάσα /’pasa/, Τάσα /’tasa/, Κάσα /’kasa/, Μπάλα /’bala/, Γκάμα /’gama/, Ντάμα /’dama/

• 3 repetitions of each word

12

Framework for pronunciation teaching (Celce-Murcia, Brinton and Goodwin 1996)

1. Description and analysis of features.

2. Listening discrimination activities3. Controlled practice and feedback4. Guided practice and feedback 5. Communicative practice and

feedback

13

RESULTS : Initial /p, t, k/- aspiration (1)

Mean VOT (in ms) for initial /p, t, k/ in Greek and English L1

52,39

64,42

32,03

11,8615,79

69,64

0,0010,0020,0030,0040,0050,0060,0070,0080,00

/p/ /t/ /k/stops

VOT

(in m

s) Greek L1

English L1

14

RESULTS (2)

VOT -voiceless stops at Time 1, age group A

15,4418,38

30,09

10,3315,08

34,33

0,00

10,00

20,00

30,00

40,00

50,00

60,00

70,00

/p/ /t/ /k/stops

VOT

Experimental group A

Control group A

VOT-voiceless stops at Time 2, age group A

18,14

36,36

11,1114,58

51,02

30,50

0,00

10,00

20,00

30,00

40,00

50,00

60,00

70,00

/p/ /t/ /k/stops

VOT (

in ms

)

EXPERIMENTALGROUP ACONTROL GROUP A

15

RESULTS (3)

VOT - voiceless stop at Time 1, age group B

9,99

16,08

27,23

15,0812,21

26,00

0,00

10,00

20,00

30,00

40,00

50,00

60,00

70,00

/p/ /t/ /k/stops

VOT

(ms)

Experimental Group B

Control Group B

VOT-voiceless stops at Time 2, age group B

37,37

46,66

64,27

12,1716,00

29,50

0,00

10,00

20,00

30,00

40,00

50,00

60,00

70,00

/p/ /t/ /k/stops

VOT (

in ms

)

Experimental group B

Control group B

16

RESULTS (4)

VOT- voiceless stops, at Time 1, age group C

12,2317,14

28,20

16,2526,42

11,00

0,0010,0020,0030,0040,0050,0060,0070,00

/p/ /t/ /k/stopsExperimental Group C Control Group C

VOT-voiceless stops at Time 2, age group C

28,8235,30

13,5017,92

30,92

64,46

0,0010,0020,0030,0040,0050,0060,0070,00

/p/ /t/ /k/stops

Experimental group C Control group C

17

Results: Within group comparisons (5)

VOT-voiceless stops Experimental Group A

15,44 18,38

30,09

18,14

36,36

51,02

0,00

10,00

20,00

30,00

40,00

50,00

60,00

70,00

/p/ /t/ /k/stops

VOT

(ms) TIME A

TIME B

VOT-voiceless stops Experimental Group B

9,9916,08

27,23

37,37

46,66

64,27

0,00

10,00

20,00

30,00

40,00

50,00

60,00

70,00

/p/ /t/ /k/stops

VOT

(ms) TIME A

TIME B

VOT-voiceless stops Experimental Group C

12,2317,14

28,2028,8235,30

64,46

0,00

10,00

20,00

30,00

40,00

50,00

60,00

70,00

/p/ /t/ /k/stops

VOT

(ms) TIME A

TIME B

18

RESULTS: Initial /b, d, g/-voicing (6)

Greek and English L1: mean VOT (ms)

-150-130-110-90-70-50-30-101030507090

stops

VO

T (

ms

)

Greek L1

English L1

Greek L1 -103,29 -107,16 -97,89

English L1 -1,03 -6,03 10,19

/b/ /d/ /g/

19

RESULTS (7): voiced stops in English L2, Age Group A

Control group: no improvement

Experimental group: improvement for individual subjects1 out of 4 subjects partly devoiced

66% of the repetitions for /d/ and /g/

1 subject target-like /g/

20

RESULTS (8): voiced stops in English L2, Age Group B

Control group: no improvement

Experimental group: some improvement for all subjects3 out of 4 subjects devoiced /d/1 out of 4 partly devoiced /d/ at

33% of repetitions2 out of 4 produced partly voiced

/g/ at 33% of the repetitions.

21

RESULTS (9): voiced stops in English L2, Age Group C

Control group: no improvement Experimental group: reduced

VOT at Time 22 out of 4 subjects target-like

/b/1 subject partly devoiced 33% of

the repetitions for /b/1 out of 4 subjects devoiced all

three stops at initial position

22

RESULTS: Vowel length before word final stops (10)

173,11

220,83

170,11

231,44

159,14

226,31

0 50 100 150 200 250

vowel length (ms)

Vowel length before word final stops in English L1

V/g/#

V/k/#

V/d/#

V/t/#

V/b/#

V/p/#

23

RESULTS (11)

Vowel length before word final stops, Age Group A

0 50 100 150 200 250

Control group ATime 1

Control group ATime 2

Experimentalgroup A Time 1

Experimental group A Time 2

vow el length (ms)

V/g/#

V/k/#

V/d/#

V/t/#

V/b/#

V/p/#

24

DISCUSSION: Relation to initial questions (1)

1. Is pronunciation teaching to children effective in a foreign language

environment?• Strategically planned pronunciation

teaching seems to be effective with children and teenagers even in a FL environment.

• Even though the teaching intervention was rather short, students’ pronunciation improved in two of the three features taught.

• No improvement for the groups that followed the regular school programme.

25

DISCUSSION: Relation to initial questions (2)2. Does age play a role in the acquisition of

pronunciation in a classroom environment?

The youngest age group did not learn the English pronunciation easier or faster than the older groups.13-year-old students showed the biggest improvement in pronunciation

• Their advantage over the 10-year-old group: cognitive development and maturation

• Their advantage over the 15-year-old group: fossilization has not occurred in their speech yet

26

DISCUSSION: Relation to initial questions (3)

3. Are some phonological features acquired more successfully than others?

• Aspiration showed the biggest improvement of all features it increased for almost all places of articulation for all age groups.

• It was also the first feature to be learnt compared with stop devoicing and vowel lengthening

• Possible explanations: Aspiration is perceptually more salient than

devoicing or vowel length It is easier to demonstrate with visual aids while

teaching The learning of a frication feature, such as

aspiration, may be easier that the learning of a time feature, i.e. vowel length

27

DISCUSSION: Relation to initial questions (4)

To what extent are the theoretical models confirmed?

• Our expectation that stop devoicing would be easier to acquire than aspiration due to markedness is not confirmed

• Markedness, however, can explain the lack of improvement for vowel lengthening (a marked feature)

• Evidence for Flege’s SLM : intermediate VOT values between Greek and English , possibly because of ‘equivalence classification’

• Exception : native-like production for /k/ by the 13 and 15-year-old groups.

28

CONCLUSION (1)

1. Aspiration was the first feature to be learnt and showed the biggest improvement of all features that were taught

2. 13-year-old group the biggest improvement, maybe due to cognitive maturation and lack of fossilization

3. Years of experience did not result in better pronunciation (no differences among age-groups at Time 1)

29

CONCLUSION (2)

4. Considerable effect of pronunciation instruction, even after 12 mini-lessons. Pronunciation teaching is not only applicable but also very effective

5. In FL contexts, improvement in pronunciation comes as a result of explicit and strategically planned pronunciation instruction.It appears that pronunciation is not acquired in FL contexts unless it is taught.

30

Acknowledgments

Dr Katerina Nikolaidou Dr Marina Matthaioudaki Participants of the study, their

teachers and their school headmasters

The Pedagogical Institute