BRITISH ENGLISH VOWELS(FREQUENCY OF OCCURRENCE)

HmÁu Vlve¡rco T.

Area de Inglés

It has already been stated that the selectionof phonological ite,ms to be included in the

teaching of pronunciation must be based onvarious criteria (Montero, S. and Vivanco,.C.,1975). The rnost important of these, as faras vowels are concerned, are: a) interferenceof the Spanish phonological system; and b)frequency of occurrence of the items*.

Interference has been the subject of sever-

al studies, while the frequency of occurrencehas been neglected. Thus, investigating thelatter seemed to be of particular interest andusefulness.

I took some dialogues, recorded by nativespeakers of British English (RP), and ana-lysed the vowels used in them. These dialoguesare employed for teaching purposes at the

English Section, Modern Languages Depart-ment, University of Chile, Santiago. Thesetexts are a good sample of the type of Engiishour students, mostly future teachers of Englishas a ,Foreign Language, are faced with.

The investigation I am describing coveredapproximately 17,200 words, reaching a totalof 20,000 vowels**.

rWhen dealing with the pronunciation of Englishcon§onants, other factors must be considered. Amongthese, distribution is specially relevant.

r'In .1976 a group of our students made a firstáttempt at working on RP vowels. They covered a

limited sample: 1,500 vowels and 2,300 consona¡rts.The members of this group were Miguel Soto, Pilar-Tronco¡o and Inés Wiegand.

Table I shows the frequency of occurrenceof each RP vowel and glide. The results ofProf. Fry's study (rFry, 1947) are also in-cluded in it, so that the reader can comparethem. A. C. Gimson's symbols are usedthroughout (Gimson, 1970).

An analysis of the data provided in TableI will show that the percenrages given by rFryand those in our study are considerably similar- The percentages corresponding to twelvevowels vary less than t point; those corre-sponding to six vowels vary more than ,l butless than 2 points. These eighteen vowel§constitute 90/o of. our sample. Consequentüy,there are two sounds in which iFry's figuresand ours differ more noticeably. These are

lt I and la l.Fu /r/ Fry gives.Zl.24i/o andwe record 18.35i7o. The difference is, then,2.E9. For I a I Fry gives 2l.39/o and we have20.55'%. The difference is greater this time:6.84.

It is significant that rhe greatesr differenceis found in the most frequent vowels. It is awell known phonetic characteristic of Englishvowels in unstressed position th¿t they arereduced more radically than unstressed vowel§in other languages*** an4 that in unstressed

***See Delattre, 1969, who gives the followingperce[tag$ of vowel reduction in four languages:English, 17 1g%; French, 8.69/o; C*rman, 6.391o;

Spanish, E.65/o.

§ee Vivanco, 1976, for vowel reduction in Spanish

ard hglish.

.rü

CORE Metadata, citation and similar papers at core.ac.uk

Provided by Revistas Académicas de la Universidad de Chile

20,000 r00.007o 100.0a7o

positions it is usual to find I a I or I r I al-most exclusively. A. C. Gimson stater:

"It will be seen that totally unstrersed

syllables are associated particulary with von-els of a central or centralized quality (ora syllabic consonant) , i.r. I e l, I I l, *dlu/ (though [u] in a weak situation is

normally replaceable by I a / or may be

reduced b /w/ before a following von-el) ". (Gimson, 1970).

With respect to this point the two gener-

aüvirts, Chomsky and Halle, 1968, indicate:

"We have referred several times to the

well-known fact that lax vowels reduce to

a cc¡tr,al, high, or mid unrounded "neu-

tral" vowel in English when they are suf-

ficiently weakly stressed, in some way thatmust be made explicit. \\¡e have been rep-resenting this neutral vowel as [a]. Thee:ect phonetic realization of I a ] does notconcern us. For any particular dialect, thefeature specifications and üe appropriatephonetic rules can be established. For ease

of exposition, we simply make the assump-

tion here that I a ] is distinguished from allothe¡ vocalic segments".

Daniel Jones, 1962, defines a weak form inthis way:

"A weak form of a word is generally dis-tinguished from a strong form either by adifference of vowel-sound, or by the ab-

sence of a sound (vowel or consonant), or,by üe difference in the length of a vowel.

When the forms differ in vowel quality, itis generally found that the weak form has

a where üe strong form has some othervowel".

R.-Nf. S. Heffner, 1969, shares the point ofview already mentioned, when he says:

"In English almost all unstressed vowels

tend to become [a ], though certainly nota,ll of them arrive at that end as yet. Manyspeakers preserve something of the qualityof the stressed vowel in the unstressed syl-

lables of words like affliction, adult, . . . ,

enfold, explode, or use an unstressed Ir ]rather than [a] in words like deceil, define'. . . , added, credit. The vowel [a] is as

nearly an unarticulated vowel sound a¡ isto be found in human speech, and it seems

to ,be used by most languages in some, at

least, of their unstressed syllables".

The fact that both lal and lrl occur rn

unstressed syllables and that both can be

étiaea if the vowel reduction goes a step fur-ther, makes it poosible to have variationc in

TABLE I

vrvAñfco rRY

VOWEL out of 20,000

uouels07/o

.t7/oi .,..ii..,

i:l

e

e4:

D

f,:

U

u:

3:

e

CI

EU

AI

EU

3I

IE

e3

ual

l,l7g3,670

1,432

1,004

472

956

646

342

906

992

2284,110

7701;144

1,262

890

30

254

1456

5.8e%tl&.Eflo7.16%5.02%2.E6%4.78%g.zEYo

t.7t%4.58%4.e6%

r.44%20.b8%

3.Wo5.72%

6.ElEo

r.eíYo0;tíYor.27Y"0.7+7"

o'oYo

4.20%2t.2+To

7.57'%

3.69%2.$rYo

t.4s%3.rs%2.'l9To

2.w%4.46%l'1Yo

27.E97o

4.8ú7o

E.&syo

4.667o

r.56%0.ErYo

0.5E7o

o'$Woo'lWo

36

the frequency of occurrence of these sound¡.

The type of language we analysed was col-loquial and fast. In many cases the reductionof a vowel meant its elision, as in the case

of "rather a ,bother', which was pronounced

/ra: óra boóa/ instead of /ro: óara boóa/,or in "history", pronounced /hrstrr/.

It is pedagogically important to considerthe facts as presented in Table II. In thisTable the vowels are ordered according tothe frequency of their occurrence in the twoanalyses under study.

Table II shows that in both counts the

same four sounds occupv the first four places.

TABLE II

VIVANGO FRY

VOWEL d/o ú

/o

They are positively the most frequent vowelsounds in English and must be given special

importance when selecting and grading the

phonological items to be included in a sylla-bus or les,son. The addition of the individualfrequencies of occurrence of /a, r, e, aI/ gives

60.86 (Fry) and 52.t7 (Vivanco), thus con-

stituting more than half of the vowel sounds

in any English utterance.Both studies are similar at the bottom of

the list, as well. The eight least used sountls

according to both authors are the same, allof them with a frequency of occurrence un-d,er fln.

The aim of this brief investigation has

been to provide the teacher of English withinformation which he oan use to give more

or less importance to some phonologicalitems. It is frequent to find syllabuses and

texts which distribute sounds along the unitsas if they were equally important.

The teacher must not forget, nevertheless,

that the frequency o[ occurrence of the

sounds is only one of the criteria he musttake into consideration when he organizes histeaching material. The degree of difficultythat the individual sounds pres€nt to ourstudents is another one. For instance, while

lel and. f arl are relatively easy for Spanish

speaken as there are similar vowels in thislanguage, lal and f rf arc extremely difficultto recognize and produce.

IIe can take our contribution as addition-al info¡mation to help hirn in the selection

of items to be included in a syllabus or unitand in the grading of these items. The moreinformation a teacher can handle, the bet-

ter he will perform his job.

J

U

IL

\

B

f0AD

A

d; io

I

áañl-,(')

II

('.L.

a

I

e

AT

,A'.:

CI

I:

EU

eD

J:

u:U

c:OU

5:

ae

IE

f,I

UA

e

I

e

AI

i:EU

aá,

D

u:CI

J:o:OU

(,

3:

IA

ta¡IUA

20.5570

t8.3ú7o

7'rWo6.31,70

5.8e7o

5.72%

ú.ozYo

4.967(,

4.7Wo

4:úl%3,85'%

3.ztYo

2.E6Yo

't.eúYo

t.7t'%1.490r.z1Yo

0,74%

5:rff"0.0170

27.3s%

2r.241%

7.57%4.ffi%4'4Wo4'\Wo+.2070

3.aYo3.6W"3.4Wo

3.tgYo

2"ffiY"2'lW"2.o\qot.bóY.r.\Yoo'$Wo0.5t%o'\WoOtlffo

VOWEL

37

Crcrrsnv, N. & He¿¿¡, M. 1968. The Sou¡il Patt¿r¡ol tr}ltglirh, Harper a¡d Rm, Ncs Yort,

I)t¡arr¡E, P. 1969. "An Aou¡tic and Artio¡latory§tudy of Vowel Reductio in Four rrt¡gu¡gc'',a IRAL, rrl. vrr, 4.

fny, D. B, 1947, "The Froqucncl of Occurrance of§peech Sounds in §outhern Eagliú", A¡chiutNée¡lodai*s & Phonétiqw Expérimaúth, vol.X:.

GrMsoN, A. C. 1070. A¡ l¡t¡oduciioa to thc Pro¡un-cüeion ol Engli§h, §econd Edition, Edward Ar-aold, Loodoo.

ll¡rrx¡r, R.-M. S, l9ii9. Gcneral Phonctics, Tla.e

Univenity of Wisconsin Press, Mad:son.

Jorrc, D. l*i?. An Outli¡e ol Eaglkh Pho¡tcrL¡,Ninth Edition, W. Heffer & Sons, Gambridgc.

Moxrrro, S. & Vrvero, C. 1975. "Criterioe para laSelección y Graduación de Itemes Fmológicor cnla Enseñanza de la Pronunciación a Alumnos deIa Carrtra de Pedagog{a en Inglés", en LctrgrnsModcrna, Na 2, Universidad de Chile, §anti4o.

VrvAN@, H. 1976. "Análisis EspcctrográBco dc leReducción Vocálica cn Erpa.ñol y cn Inglér", an

Lcnguos Modernas, No E, Univer¡idad dc Chile,

Santiago.

t¡ütlocnllrfY

38