45Language testing: Current trends and future needsTEASIG Conference Proceedings, Granada 2014

METHODS OF TESTING COMMON

PHONOLOGICAL COMPETENCE

Prof. Irina PavlovskayaSt. Petersburg State University, Russia

What is common phonological competence?

The title may look perhaps too advanced: canwe really start to discuss some common abilityfor pronunciation of speech sounds shared bypeople of all (or some) language groups? Arewe polilingual enough? The classical definitionrooting back to Veinreich and Scherba reads:Multilingualism (bilingualism, poliglotism,polilingualism) is the existence in the headof an individual speaker/writer or a languagecommunity and use in their ver-bal behaviour of two or morelanguages.It could be of several types, ac-cording to: language competence – recep-

tive, reproductive, productive;

learning or acquisition of thelanguage – natural (acquiredfrom environment) or artificial(learnt at school);

relationship between the con-tact languages –coordinative(pure), or equipollent (mixed).

A more recent definition fromCEFR gives a different picture,being very enthusiastic at the end:

Multilingualism : “…the plurilin-gual approach emphasizes thefact that as an individual per-son’s experience of language inits cultural contexts expands,from the language of the hometo that of society at large andthen to the languages of otherpeoples (whether learnt atschool or college, or by directexperience), he or she does notkeep these languages and cul-tures in strictly separated men-tal compartments, but ratherbuilds up a communicativecompetence to which allknowledge and experience oflanguage contributes and inwhich languages interrelate andinteract.”

Since 1992, Vivian Cook has argued that mostmultilingual speakers fall somewhere betweenminimal and maximal definitions. Cook callsthese people multi-competent. However, com-mon competencies may vary in different areasof the language – grammar, functions, lexis,discourse, phonology. People invent grammarsand words, sounds are mainly biologicallybased. So phonologically, we are more multi-competent by virtue of our nature.In the table of Common pronunciation difficultiesby Gerald Kelly (2011) which takes into consid-eration 13 languages (English, Russian andChinese included) it may be seen, that the areaof difficulties is no bigger than the area of non-difficulties (there are 67% of empty boxes in thetable). (Fig.1 shows one page out of three in thetable).

46Language testing: Current trends and future needsTEASIG Conference Proceedings, Granada 2014

Therefore, what is the consequence of all that topronunciation? Shall it tend to be a kind of mix-ture of different features of contact languagesand phonetic teachers will never more botherabout perfection of pronunciation? Alternatively,shall we return to the sensor-motoric stage of ababy, who can pronounce all sounds of all lan-guages of the world, without knowing that theyare pronouncing “phonemes”? The first is un-wanted; the second is unlikable.

What can we realistically discuss is a commonphonological competence, shared by certaingroups of people speaking the same number oflanguages in certain language areas, and whatwe should do now, is to work out definite criteriahow to assess their phonological competence.Multi-linguals’ strategies in speech are differentfrom those of monolinguals and traditional bilin-guals, their brain works differently (according toevidence from neuroscience), and our methodsof dealing with the situation should also change.

Evidence from research

In St. Petersburg State University we havesome experience of designing a trilingual curric-ulum for International Department at the Philo-logical Faculty. The aim of the research is toinvestigate the domain of phonetic interferencebetween the three languages (Chinese-Russian-English) and to find adequate methods of test-ing pronunciation in such a challenging peda-gogical context.

1. The top priority for a Chinese learner is totrain the pronunciation of sonorant sounds atthe end of a word (team, animal, onion). Notall sounds of the Chinese language can com-bine with each other freely. In the final posi-tion sonorants are impossible for a nativespeaker of Chinese. For a teacher working ina Russian-speaking audience, such problemsare unusual. Russian students can easilypronounce final sonorants. Their attentionshould only be paid to the fact that they mustbe longer after lax vowels.

2. The second problem for Chinese speakers isto pronounce combinations of plosive conso-nants+sonorants at the beginnings of wordsor syllables (blind, play, little). Such syllabicstructures do not exist in Chinese and learn-ers insert an intervocalic /ə/. In Russian, thiscombination is not ruled out, but consonantsassimilate not as much as in English.

3. For Chinese it is problematic to articulateEnglish diphthongs correctly. The vowel partof a syllable, or final, in the Chinese language

can be expressed with a monophthong, adiphthong, or a triphthong. In spite of a bignumber of diphthongs and tripthongs in Chi-nese, it does not help, because the connec-tion of vowel elements (nuclei and glides) indiphthongs occurs according to certain rules.In Russian, there are no diphthongs, but usu-ally no such problems arise, because vowelelements in Russian follow each other in anon-restricted way. Although the teacher hasto emphasize the rising character of an Eng-lish diphthong (glides are non-syllabic).

4. At the beginning of words, Chinese speakersdo not differentiate voiced and voiceless con-sonants, which never happens in Russian.

5. In Chinese, as well as in Russian, there areno such consonants as (/w/, /h /, /ϴ / и / ð / and learners have to get accustomed tothem.

6. Chinese learners do not differentiate /r/- /l /.

7. There is much in common in the intonationsystem, Chinese tones have very similarcounterparts among English intonation con-tours.

To add more difficulties to the work of a Russianteacher, there are mistakes which occur as aresult of interference with the first foreign lan-guage of the learner, the language of the envi-ronment (Russian):diphthongization of some monophthongs (/o/is / /uo/ in Russian);all noised consonants become voiceless at theend of the words: bag / bæk/;consonants are palatalized before front vowels:meal / m,il/;underdifferentiation of long and short vowels:/u:- , A:-ᴧ, e-æ, i:-ı, ᴐ:- ɒ, ɜ:-ə /.

Methodology

I believe that such pedagogical challenges canbe dealt with by phonosemantic approach,which helps both to find a medium of communi-cation in a trilingual situation and to turn to lin-guo-cultural peculiarities of the target audience(Cross 2006).

The ideas behind methodology are the following(Pavlovskaya 2002):

1. There is a direct link between sound andmeaning which is pre-linguistic and not attachedto phonological or other linguistic units(morphemes, lexemes, syntax structures)

2. In the process of perception sensory chan-nels can overlap and produce double or triplememory traces (synesthesia).

3. Sound iconism is universal and teachers can

ᶷ

47Language testing: Current trends and future needsTEASIG Conference Proceedings, Granada 2014

use it in teaching and testing languages of dif-ferent origin and structure (colour-sound corre-spondences included).

Phonological Testing: Task TypesPhonological tasks types may be divided intoarticulatory tasks and perceptive tasks, taskswith visual representation and tasks with audialrepresentation, tasks of linguistic knowledgeand tasks of communicative skills.

1. An articulatory test task with visual represen-tation in traditional mode most commonly willbe: “read and say”.Less traditional: “say it with colour”.We have measured the sound-colour associa-

tions of our subjects in psycho-linguistic experi-ments and the results are shown on Fig. 2.

So when we want Chinese students to say theEnglish sound /u:/ we ask them to make thissound blue. The pages of the teaching or test-ing material are colour-coded accordingly.An articulatory test with visual representationmay look like this (pictures are borrowed fromAnn Baker’s popular book “Ship orSheep?” (2008): Fig.3.

In this kind of test students perceive an ideo-

graphic image of a word and then say it. ForChinese students it is preferable because of thepictorial, ideographic nature of their native lan-guage.

2. A receptive skills test task with audio repre-sentation in traditional mode will be:

“Listen to the recordings of phonemes, wordsor sentences, identify the difference in minimalpairs” (the Listening part in Dave Alan’s OxfordPlacement Test is a very good example).

Less traditional:Recordings have sound track (background mu-sic), the rhythm of which requires certain syllabi-fication and provides memory trace for learning(Pavlovskaya, Radievskaya 2001).

Multilingual testing: Tasks samples

Methodologically the problem of multilingualphonological testing may be solved in threeways:

1) by implementing top-down approach(from whole text to sound segment), prosodic

48Language testing: Current trends and future needsTEASIG Conference Proceedings, Granada 2014

features (intonation, rhythm, pauses) playingthe most important role;

2) by training receptive skills (listening) longbefore productive skills (pronunciation);

3) by using sound-iconic scaffolding for betterunderstanding and better pronunciation.

Testing of polilingual pronunciation can be divid-ed into three sequential steps.

Step 1. Language identification atsuprasegmental level, communica-tive type (expressed by intonation)identification, segmentation of speechflow.

Listen to the utterance. What lan-guage is spoken? Is it a question, astatement or an exclamation?

Listen to the utterance. How manystressed syllables do you hear?

What is the total number of syllablesin the utterance?

How many words are there in the ut-terance?

STEP 2. Language identification atthe level of a word

Listen to several words. Which ofthem are English?ма - ma – mā (mother)

Listen to the word. In what languageis it spoken?1) chat 2) чат 3) chá (tea)

Listen to the three words. Which isEnglish? Which is Russian and whichis Chinese?ю (Russian letter) – you – yǔ (Chinese “rain”)

Listen to the word. Is it pronouncedby a native speaker or not? If it ispronounced incorrectly, say it as itshould sound and write it down inphonetic script.1) я (me in Russian) 2) yá (tooth in Chinese) 1) yard 2) я́рд (Yard inRussian)

STEP 3. Identification of phonemesand their differences in the threelanguages.

Students hear the sound /ж/ (Russian) . Question: To what lan-guage does this sound belong?

Students hear the English sound /ng/.Question: To what language does thissound belong? Can you say it in yourown language (Chinese)? Can youpronounce it in Russian?

Listen to the words: wabbit, tisl, borx,read. What do they mean? Whichwords were pronounced incorrectly?Say them correctly and write themdown in spelling and in phoneticscript. Write down in phonetic scriptthe way they sounded originally.

The final issue is to decide in whatlanguage to give the instruction to thetasks. It depends upon the level ofstudent’s multilingual skills. The testtasks presented above are intendedfor A-1 level.

References

Baker, A. Ship or Sheep? An Intermediate

Pronunciation Course. Cambridge University Press,

2008. – 224 p.

Common European Framework of Reference for

Language Learning and Teaching. Strasburg:

Europarat, 2000.

Cross, J. Understanding and Improving Chinese

Learners’ Pronunciation of English. In: ”Speak Out!”-

IATEFL PRONSIG Newsletter, May, 2006, Issue 35,

p. 16.

Kelly, G. How to Teach Pronunciation. Pearson-

Longman, 2011.

Pavlovskaya, I. Yu. & Radievskaya, M.G. Tasks in

Practical Phonology of English: Vowels. (with a CD-

ROM) St. Petersburg, 2001. – 100 p.

Pavlovskaya, I. Yu. Phonosemantic Analysis of

Speech. St. Petersburg: St. Petersburg State Univer-

sity, 2002. – 291 p.

Pavlovskaya, I. Yu., Timofeeva E.K. A Beginner’s

Course in English Pronunciation: Phonosemantic

Approach. - St. Petersburg: Hippocrates, 2009. - 128

p.

49Language testing: Current trends and future needsTEASIG Conference Proceedings, Granada 2014

Prof. Irina PavlovskayaDoctor of Philology, Professor, lecturesin Methodology of Language Teaching,The Theory of Testing, Theoretical Pho-netics and Phonology, Teacher Trainingand Teacher Development programmesfor TKT and CELTA at the Faculty ofPhilology, St. Petersburg State Universi-ty, Russia. She is the chair of TestingSIG at the International PhilologicalConference at St. Petersburg State Uni-versity,has been a member of IATEFLsince 1993, was the founding presidentof SPELTA.pavlovskayairina2@gmail.com