THE DIFFICULTIES EXPERIENCED BY LUO AND SWAHILI

CHILDREN IN LEARNING ENG LISH

by

DONALD OWUOR. B~ A~

A Thesis submitted to the Faculty of Graduate Studies and Research in partial fulfilment of the requirements for the degree of Master of Arts.

Department of Education, McGill University. Montreal. August, 1963. PREFACE

The purpose of the dissertation is to locate, analyse and define the diliiculties which Luo and Swahili children· in the schools of Kenya experience when learning English, and to make sorne suggestions regarding the elimination of those difficulties.

English is the official language of Kenya. It is also the language of instruc tion at alllevels of education above secondary school. It is important, therefore. that the teaching of English in the schools of Kenya be of the highest standard attainable. Sorne knowledge of the factors which influence the way African children learn Engllsh is essential if the standards of the teaching of English in Kenya are to be improved.

I wish to express my indebtedness to Mr. B. J. Spolsky for his valuable suggestions and guidance throughout this study.

I am also grateful to Prof. Reginald Edwards for the counsel and help I received from him. CONTENTS

1. CHAPTER 1 Introduction ...... Page 1

2. CHAPTER II Luo and Swahili children at schoo1 •••••••••••••• • 17 (a) The schoo1 system • • ••••• . . . . . 18 (b} The LuQ child in the system •••••••• ...... 28 (c) The Swahili child in the system ••• ...... 35 (d) Sorne of their ex.ternal handicaps ••• ...... 39

3. CHAPTER ID The sounds of English, Luo_ and Swahili • • • • • • • • • • • 47 (a) The consonantal phonemes. • • • • • • • • • • • • • • • • 50 (b) Consonant clusters • • • • • • • 60 (c) Vowel phonemes • • • • • • • • • • • • • • • • • 63

4. CHAPT ER IV The cultural factor in language •••••••••••• . . . • • 129 (a) Learning English culture • • • • • • • • • ...... 134 (b) Family relationships in English and Luo...... 139 (c) Family relationships in Swahili. • • • •• . . 146 (d) New ideas and abjects •••••••••••• . . . . . 148

5. CHAPTER V Conclusion • • • ...... 155 6. BIBLIOGRAPHY ...... 163 ILLUSTRATIONS -

1. The Phonemic Structure of English, Luo and Swahili•••• Page 54(a)

2. Family Relationships in English, Luo and Swahili • • • • • tt 141 CHAPTER I

INTRODUCTION

Like other countries which have only recently emerged from British colonial rule, for example, India and Ceylon, or are on the verge of attaining self-government, Kenya has had to face the task of choosing an official language. In her case the choice has been complicated by the prominence of two languages, namely, Swahili and

English.

Swahili, a Bantu-based language, is widely spoken and under- stood among the country1s population, whose languages belong mainly

to the Bantu family of languages. It is estimated that sorne 71 000,000 people throughout East Africa speak Swahili as their native language. 1

In Kenya it is the language of commerce and the medium of communi- cation between the educated and the uneducated, since the latter do not speak English, and between tribe and tri be throughout East Africa.

However, some tribes, such as the Baganda in Uganda, deign to speak

Swahili not at all, or only occasionally, although they understand it perfectiy, their intense pride in their history, culture and language making them see in Swahili a fornùess, upstart language which should give way to their own language on every occasion.

English, on the other hand, is the language used at the higher

1. Frank A. Rice ed. Study of the Role of Second Languages, (Center for Applied Linguistics of the Modern Language Associa­ tion of America, Washington, D. C., 1962) p. 70 -2- levels of administration and education, as weil as in normal busiœ ss transactions and social intercourse between Europeans and educated

Africans, or Europeans and Asians, or educated Africans from dif- ferent tribal backgrounds. lt is the language of the socially privileged 2 and most influential group in the country. For example, it is the language used in Legislative Council, the Nairobi Stock Exchange and in all Government departm.ents and offices. Education and the ability to speak English are synonymous in Kenya, for the one, under favourable circums tances, almost always leads to the other. And the educated are the doctors, lawyers, engineers, veterinary scientists, teachers, re- search workers and others on whom the welfare and progress of the country depend.

When the choice came to be made between these two languages re gard had to be had to the relative merl ts of each as a medium of communication not only among Kenya1 s nine million people, but also with the wider world community from which Kenya could not afford to eut herself. lndeed, the Minister of Education, in cautioning members of Legislative Counc:U against too precipitate an introduction of Swahili as the language of secondary schools and higher institutions of learning, used this very argument. He said, among other things, that

2. Ibid. p. 71 -3-

unless the people of Kenya are going to be eut off from the rest of the world in educational developrœ nt, and world knowledge in the cultural and scientific fields. it is quite obvious that English must remain the language of education. 3

Under these conditions the only logical choice was English.

However, from the point of view of the effort called for and the order of ability necessary for adequate m.astery, Swahili would have been the better choice because of the close links it has with the Bantu family of languages. For those people whose languages belong to this group. for example, the Kikuyu, Embu, Luhya, Kamba and Kisii,. mastering

Swahili is relatively easy since their languages and Swahili have certain morphological features in common. Even the Luo and the

Masai, whose languages belong to the Nilotic family of languages, or the Somali, who speak a Hamitl.c language, do not in general find it difficult to learn Swahili.

It was perhaps because of these considerations that Tanganyika,

Kenya's s·outhern neighbour whose entire African population speak languages of the Bantu group, decided to make Swahili, rather than

English, her official language. But even in Tanganyika, English re- mains the language of higher education.

ln Kenya it was not a question of adopting English as the country's official language, but rather of deciding whether to continue

3. Lesislative Council Debates, Official Report. Thursday, !3th December, 1962, (Government Printer, Nairobi) p. 266. -4- to have it as the official language or to substitute Swahili for i~ for

English had been Kenya•s official language since the establishment of

British rule in the country towards the end of the 19th century. In lndonesla, on the other hand, one foreign language, Dutch, was re- placed by another foreign language, English, as the official language of the country when Dutch colonial rule came to an end in 1948. 4 In other words, intense national pride and consciousness in Kenya made it essential that an indigenous language, rather than a foreign one, be considered as a possible official language of the country, whereas in Indonesia considerations of the greater access English would give

Indonesians to the intellectual, technical and commercial world, as compared with Dutch or Malay, made English a better choice than

Du.tch or Malay. Besicles, the odium attached by Indonesians to every- thing Dutch after the end of Du.tch rule militated against the retention of

Dutch as lndonesia1s official language.

Even in the United States, national pride demanded that that country develop its own system, rather than lmport everything from

Great Britain. As Noah Webster stated:

As an independent nation, our honour requires us to have a system of our own, in language as well as in Government. Great Britain whose children we are, and whose language we speak., should no longer be our standard; for the taste of her writers is already cor­ rupted and her language on the decline. 5 4. I am indebted for this information to Henry Van Eykens who served in the Dutch army in lndonesia between 1944 and 1948. 5. Noah Webster, Dissertations on the English Language, (Boston, 1789) p. 20. -5-

In Kenya the choice has been made, if in terms tb.at are not specifie. When the question of the country•s official language was debated in Legislative Council in December, 1962, members

were of the opinion that Swahili will one day be the official language of this country, and were satisfied that Swahili is the most suitable medium. for keeping the Government. the leaders and the people of this a:>untry in close contact. 6

This was an oblique way of stating the fact tb.at English~ for sorne seventy years Kenya1s official language, would remain so until re- placed by Swahili. It is the author•s belief tb.at this change will not be made within the next twenty year s, if it is made at all, for it will take the country1s educators about that long to improve the teaching of

Swahili in the schools to a point where the maj ority of the people would be able to speak understandable Swahili. For many people at present make do with broken Swahili of limited vocabulary range, as the Swahili-speaking Russian woman who accompanied the Soviet delegation to the Afro-Asian Solidarity Conference at Moshi, Tanganyika, in

February , 1962~ discovered.. rouch to her bewilderment and annoyance, when she tried to converse in Swahili with members of the Kenya delegation. 1 In any case,. if and when Swahili replaces English as the official language of Kenya, English will still remain the language of the secondary schools and uni versities.

6. Op. cit. Legislative Council Debates, p. 278.

7. Lugo Taguaba, "The Conference at Moshi11, East African Newsletter, Vol. n, No. 5, Fe bruary, 1963 .. p. 3. -6-

lt is fortu.nate that Kenya has a considerable history of

English language teaching in the schools. However,. for various rea­

sons. the chief of which is the unsound teaching to which beginners are subjected, the degree of proficiency attained has not always been

impressive. For example, it is estimated that 95o/o of the stu.dents

who iail in the Cambridge School Certificate Examinations do so be­

cause of failure in English, 8 and as English is a compulsory subject in these examinations, poor performance in it is usually positively correlated with performance in other subjects. Since these examini­ nations are taken at the end of secondary school, that is. after ten years of English, failure in the subject reveals an inadequate mastery

of the language.

What constitutes adequate mastery of ·English must depend on the number of years of English received, and on the occupations or pursuits of the people concerned. A student intending to go to university must have sufficient command of the language to enable him to embark on university work. As he will have had at least ten years of English, it is not too rouch to expect that he attain this degree of proficiency.

On the other hand, a student who leaves sÇhool at the end oi primary school, that is, at the end of the eighth year of school,

8. Figures supplied by the Ministry of Education• Nairobi• Kenya. -7-

should be sufficiently fluent in English to be able to make himself

understood in simple English1 in addition to being able to make

summaries of the chapter s of a book such as Robinson Crusoe or

The Hound of the Baskervilles. He will have had six years of

English,. and so the above criteria should not tax his language

resources beyond their limit. As this person generally trains as

a primary school teacher~ or becomes a junior clerk in a govern-

ment department. he would not be able to work efficiently if he did

not command at least the amount of English stipulated above.

For the student intending to go to university, a sign that

he had adequate mastery of English was a credit or distinction

in the subject in the Cambridge School Certificate Examina tions.

But even such a sign has not always automatically indicated pro-

:fi.ciency in the language. For example,. the Institute of Education of

Makerere University College, Uganda, finds it necessary to offer

remediai courses in English for its diploma students. 9 As these

students would all have obtained at least a credit in English in the

Cambridge School Certificate Examinations, that they should still need some remedial courses in English is symptomatic of the generally low standards of English teaching in East African schools, especially

9. Edward Fry. 11Reading .hnprovement in Africa", Nufiield Research Project in the Teaching and Use of English in East African Schools, (Makerere University College) Seventh Progress Report, June, 1962, p. 20. -8- at the primary school leve!.

If it is the clear duty of the schools, as weil as of a few interested agencies such as the Kenya Broadcasting Corporation, to 'leach English effective!y to the people of Kenya, it is no less their responsibility to determine the type of English that is to be taught. There are many varieties of English, even in Great Britain itself. Such a country as the United States has, broadly speaking, three major types of English. These are the variety spoken in the

New England States, that spoken in the Southern States and that 10 spoken in the rest of the country.

It may be that intime there will develop:in Kenya a peculiarly Kenyan type of English, just as in the United States three major types of English have emerged. However, the practice has hitherto been to teach what is generally known as Received Pro- nunciation. This has been the dominant variety of English in Kenya, for the majority of British teachers, as weil as civil servants and administrators, have for a long time been drawn from those educated in the 11public schools11 of England or at Oxford, Cambridge or

London. And Received Pronunciation, as Daniel Jones defines it, is that form which is

10. George P. Krapp, The English Language in America, {The (The Century Company, 1925) Vol. 1, pp. 35, 36, 41. -9-

usually heard in everyday speech in the familles of Southern English people who have been educated at the public schools. This pronunciation is also used (sometimes with modifications) by those who do not come from Southern England, but who have been educated at these schools ••• It is probably accurate to say that a majority of Londoners who have had a university education, use either this pronuncia­ tion or a pronunciation not differing greatly from it. Il

It is difficult to say whether this type of English will for long continue to be taught in Kenya. The country is now on the verge of independence and many British people are leaving or have left to settle in other parts of the world. With complete self-government this exodus may be expected to increase consider- ab!y.

The influence of Received Pronunciation is therefore likely to decrease.

The situation is further complicated by the influx of

American teachers arriving to help reduce the country's acute

2 shortage of teachers. I If. as is stated in the Nuffield Research

3 Project/ this trend continues, the effect of Am.erican English on the Schools of Kenya will definitely increase. Moreover, when the

6ooi4 Kenya students enrolled in Am.erican universities and colleges in the 196I-62 academie year begin to return home in large numberst the influence of American English is likely to be further strengthened. II. Dâniel Jones, English Pronouncing Dictionaryt (London: J. M. Dent & Sons• I956), llthed., p. xv. I2. Op. Cit. Nuffield Research Project; P. 2. 13. Op. Cit. Nuffield Research Project p. 2. I4. Figures supplied by the Office of the Adviser for East African Students in North America, the British Embassy, Washington, D. C. -10-

As more and more students succeed in obta.ining higher education in the United States,. rather than in the United Kingdom whose higher

educational establishments are considerably harder to get into, so will the influence of American English in Kenya increase and that of

Received Pronunciation diminish.

lt is to be hoped that the final outcome in Kenya, while dis- tinctively Kenyan in the same way that Scotland's brand of English is distinctively Scottish, will be one which the people of Kenya, as weil as educated English-speaking people the world over, will under- stand easily. It is a virtue of Received Pronunciation, as Daniel

Jones states, that nit is easily understood in aU parts of the English- speaking countries; it is perhaps more widely understood than any 15 other type. 11 Comprehensibility should be the guiding principle in determining the type of English taught in the schools of Kenya.

At the present time, since the new influences have not had time to make themselves felt, the only course of action left the schools would appear to be to continue to teach Received Pronunciation. It is essential that this be done in the most effective way possible, for it is only under those circumstances that it would be possible., or even wise, to dispense with the advice given in the following statement:

We do not • • • advise anyone who has not already acquired the speech characteristics of English RP in a natural way to attempt to acquire them artificially. We believe that an irritating affectation will surely result from any

15• Daniel Jones, An Outline of English Phonetics~ (Cambridge: W. Heffer &: Sons, 1918) p. 12. -11-

such attempt. l6

The introduction of another variety of English at this

time would create confusion among Kenyats school children al-

ready having sorne knowledge of English. It wruld also come up

against the country1s most serious problem in the field of education.

namely, the shortage of teachers in the schools. For the present.

then, RP will continue to be taught.

It is pertinent at this point to ask certain questions~ al-

though no full and final answers can be given to them in a dissertation

of this nature. First. in view of the time at the disposai of the

country•s secondary schools. what degree of proficiency in English

should be aimed at. and how is a reliable measure of this proficiency to be obtained? Related to the above is second: to what use will the

English thus mas te red be put by the students? W ould it not be more logical and less expensive to teach the amo);tnt of English appropriate for each leve! of occupation: simple English to budding mechanics and technicians, and more complicated English to those intending to follow academie pursuits? ln any case, how will the educational authorities determine or know which child is suited for which occu- pation or pursuit, and what is the amount of English that is nappro- priate11 for a student who wishes to be a bricklayer, or a laboratory technician or a marine biologist?

16. Thorleif Larsen and Francis C. Walker, Pronunciation, (Oxford University Press, 1930) p. 15. -1.2-

Ali these questions are asked merely to give sorne indication

of the !ines along which the educational authorities of Kenya will have

to think ü English teaching in the country is to be put on a surer~

firmer foundati.on. In order to equip students to be able to undertake

secondary school work with profit, they should be given a thorough

grounding in English in primary school. By the end of secondary school,

after a total of ten years of English, this foundation should be solid

enough to be comparable to that that an English boy in the six.th form

is expected to have, that is, he should be as rouch at home in English as an English student of the same age. For British universities cannot

be expected to lower their standards in order to accommodate African

students from Kenya, nor can the University of East Afrlca do so with-

out losing its recently-acquired place in the world of learning.

It is a difficult proposition. But the author sees no reasonable

course of action left other than to go back to fundamentals, that is, analyse the causes of the ineffectiveness of English language teaching

in the schools of Kenya, and then adopt measures designed to eliminate

those causes. For the level of English in the schools of Kenya can, as is stated in African Education,11 be raised a good deal above its present 17 standard. 11 The argument advanced in this dissertation is based on the

17. The Nufiield Foundation and The Colonial Office, African Education, (Oxford University Press, 1953), P. 83. -13- linguistic fact that knowledge of a pupil's native language back- ground enables a teacher to appreciate~ and so better direct. the adjustments the pupil bas to make in learning a new language. For example. John P. Hughes writesr

The language you already speak is bound to influence the language yon: are learning , and we cannot teach the latter with maximum effi.ciency unless we take into account the sort of interference the former is likely to create. 18

A Kikuyu child learning English bas trouble separating /r/ from

/1/. l9 for the latter does not exist in Kikuyu. A Land Rover therefore cornes out as a Rand Lover in Kikuyu speech.

The importance of one's native language, as Hughes states, is that it is nthe only one which we can be sure was learned without 20 interference from any other language. u The native language or first language then influences the way a new language is learnt.

Since Kenyats children setting out on their first English lesson among them speak about fifty different languages, 21 it is hardly sur- prising that these languages have s o far not been analysed with a view to determining the extent to which they interfere with the way in which these cbildren learn English. As the author•s own knowledge of the

African languages spoken in Kenya is limited to two only. namely. Luo

18. John P. Hughes, The Science of Language, (Random House. 1962) pp 240-241. 19. Throughout this dissertation the following signs are used as indicated: (i) slant lines / / enclose phonemic script; (ii) the letter L or S before / 1 means that the. script within the slant !ines is a Luo or Swahili phoneme. 20. Ibid. John P. Hughes, pp 19-20. 21. Estimate based on figures supplied by the Ministry of Education. Nairobi. Kenya. -14- and Swahili,. the argwnent is presented from the point of view of native speakers of Luo and Swahili.. tb.at is,. 1, 100 .. 000 and 800, ooo22 people respectively. How many of these are children learning English it is difficult to estimate. Ho·wever, an estimate based on the incom- plete school attendance figures supplies by the Ministry of Education would be 300 .. 000 and 100,000 respective!y.

A similar analysis will have to be made for each of the other African languages of Kenya, and an exposition made of the difficulties peculiar to those for whom they are the native tangues, together with suggestions about how those difficulties can be over- come. For English teaching has to be improved, not only for Luo and Swahili children, but also for Kikuyu, Masai, Giriama, Suk,

Luhya, Kalenjin and Kisii children, as weil as for those too many to list.

No attempt is made in this dissertation to evolve a perfect course of English for African schools in Kenya. Rather.. is it the authorts hope that by identifying and defi.ning the areas of difficulty wbich Luo and Swahili children learning English experience, it will be possible to design an English course which will meet the needs of these children. The task of designing such a course will have to be under- taken by people who know English and the theory of learning more

22. The Ministry of Constitu.tional Affaira and Economie Planning, Kenya Population Census, (Government Printer, 1963) p. 2. -15- thoroughly than the author can claim to do. lt is imperative that tbat task be done soon.

The author learnt English in Kenya and stu.died it at the university level at Makerere University Colle ge, U ganda. His experiences in Kenya schools, and his observations of the peculiari- ties of his own speech and of those of his friends at Makerere. have necessarily been relled on. However, the main source of infor- mation has been the extensive literatu.re on linguistics in which experts in linguistics have recorded the resulta of their stu.dies, especially those stu.dies that have dealt with the teaching of English as a second language. This source is, fully acknowledged in foot- notes and bibliography.

Politicians in Kenya never tire of telling the people that education holds the key to the country*s future. As English is the medium of instruction at alllevels of education above Standard VI, that is, from the seventh ; year of school onwards, the role of English in aU spheres of life is very great indeed. This role is likely to in- crease in importance, rather than diminish, as education becomes more widespread in the country. The role of English in Kenya can, in the 1ast analysis, be closely identified with that of formal education.

And formal education, as Dr. Busia states:

has now come to be regarded as essential not only for equipping the indivi dual to earn a living but also for equipping Africa's peoples for nationhood. The formal education introduced by Europeans has come to be accepted as an indispensable lever, for pro- -16-

gress in all aspects of life. Schooling is. in fact, increasingly becoming a part of African culture, a borrowed institution fitting into the pattern of living. 23

23. K. A. Busia,. The Challenge of Africa• {New York: Fred.erick A. Praeger, l962) p. 80. -17-

CHAPTERII

LUO AND SWAHILI CHILDREN AT SCHOOL

A brief survey of the school system of Kenya is in order, for an examination of how the countryts educational policy and practice have affected aU aspects of development in the country is instructive. With this awareness of what has happened in the field of education since the colonization of the country by the

British in the late !9th Century, predictions can be made about the future course of events with a reasonable degree of assurance.

What is even more important, suggestions can be made, in the context of the countryls educational structure and history, re garding the teaching of English to African children with a vlew to enabling them to attain a higher degree of competence in the language than most children have hitherto been able to achieve.

This~ after all, is the main aim of this dissertation.

However, the suggestions made, and the teaching of

English 1in Kenya as a whole, should be seen in proper historical perspective, regard being had to the progress made in education since the advent of the British, and the part English has played, and is llkely to continue to play, in the nationallife of the country. -18-

The SchoolSystem

The history of education in Kenya spans a period of no more tha.n sixty years, for British rule was not firmly established until the completion in 1901 of the railway from Mombasa, on

Kenyafs Indian Ocean shores, to Kisumu, on the shores of Lake

Victoria. 1 No doubt the various tribes living in the country before the arrivai of the British had a system of their own of giving their young the training in discipline,. group loyalties and affiliations,. tribal lore and tradition, and related skills and practices which pro- moted the welfare and survival of the tribe. However, none of this was carried out in schools or similar establishments, but was i:n- culcated in the young at adolescence by their elders, and culminated in the initiation rites and other practices tha.t signalled the atta.in­ 2 ment of full adulthood. Once a youth had become an adult he had little more to learn; bis tasks consisted cbiefl.y of defending the tribe against attack,. bringing up his children in fear of the tribal

gods and fetishes, and in general furthering the collective welfare of the tribe.

It was into this atmosphere of regulated stagnation, with its then familiar scourges of famine, droughts and floods, and its

1. United Kingdom Information Service, Canada,. Kenya, January 1960, p. 2.

2. Kikuyu initiation rites and other ceremonies are iully described in Jomo Kenyattats Facing Mount Kenya (Chatto and Windus, 1945) -19- endless cycle of tribal wars and general restlessness, that the

British burst, gave the country a name to markthe area within the artificial boundaries which they had set up, and ushered in a new era of progress and enlightenment, and what has since devel- oped into a strong national consciousness. Needless to say, they also brought with them some of the less endearing features of their institutions and customs, for example, their class system under which a small but highly privileged group ruled the country. Ail these were 'reflected in the new society which the British set about creating.

It is a mark of the humanitarianism of British and American missionaries that in the field of education in Kenya they were the pioneers, for the first schools in the country were established by 3 them. The school at Maseno was established by the Church Mis­ 4 sionary Society in 1906 and the oné at Gendia by Seventh Day

Adventist missionaries in 1907.. 5 This was tb.ree years before the 6 first school for European children was e stablished by the government.

3. UNESCO, Paris, World Survey of Education, Vol. II, 1958, p. 1092. 4. E.R.J. Hussey, Some Aspects of Education in Tropical Airica, Oxford University Press, 1936) p. 28. 5. Ibid, E. R. J. Hussey, p. 28. 6. -op:- cit. UNESCO, p. 1092. -20-

In the course of time other schools were set up throughout the country. The missionaries who built maintained and managed these schools belonged to a diverse array of religious denominations, from Roman Catholics• Anglicans and Presbyterians to Baptists,

Quakers and Congregationalists. In addition, there were schools controlled by Moslems, especlally the Koran type of school for Arab children in the Coast Region, and in cities and towns as these came into being.

Although there were private agencies and individ.uals main- taining and managing schools, African education was largely in the hands of missionaries. 7 Grants of money were made by the govern- ment to most of the Missions, especially those who managed their schools to the satisfaction of government inspectors. 8 The Depart- ment of Education, which was set up in 1911, became the Ministry of Education in 1954 with a Minister, Director of Education and other administrative personnel to manage and control education through­ out the country. 9

Missionary influence has not decreased appreciably since the re-organization of the Departrnent of Education, but the govern- ment is now participating more fully and effective in education than at any time prior to 1954. For example, it now advises on matters related to the curriculum, originates and implementa policy and provides financial support for the whole e:nterprlse. In addition,

7. L. S. B. Leakey, Kenya: Contrasta and Problems (Methuen, 1936) p. 150 8. Op. clt. UNESCO, p. 1092. 9. United Kingdom Information Service, Canada, Kenya, January, 1960 p. 12 -21- internal exam.inations, such as the Kenya African Prelim.inary

Exam.ination, taken at the end of the first eight years of school, are set under its auspices.

It can be conjectured to what degree the confiicting religious beliefs and aim.s of education of the different m.issionary com.m.unities vitiated the new spirit of progress which they had established with their schools, churches and hospital$. But, as the com.m.issioners who studied African education in East and Central A!rica for the

Nuffield Foundation in 1952 state:

The Churches have made the initial experimenta, have learnt how to bring success out of failure, have been able to show the way, and have then handed over their achievements to Governments vmo develop systems from these pioneer efforts. 10

lt was the Beecher Report of 1950 which recommended the reorganization of African education along the Unes still in operation.

A major recommendation which it made, and which was unfortunately adopted a little precipitately, was the establishment of an eight-year school, divided into four years of primary school and four of inter- mediate· school, followed by four years of secondary school terminat- ing with the Cambridge School Certificate. Since Makerere University

College, Uganda, the Royal College, Kenya, and University College..

Dar-es-Salaam,. Tanganyika, no longer admit students at the

10. Nuffield Foundation, African Education,. (Oxford University Press, 1953),. p. 63. -22-

Cambridge School Certificate level or equivalent, but at the Higher

Certificate level or equivalent, it means that to the twelve-year school of the period between 1952 and 1961 has been added the two additional years required for Higher School Certificate courses.

Fourteen years is a long time for a child to spend in getting through secondary school, and as most children do not start going to school until they are six or seven,. it means that those among them who eventually go to university cannat do so until they are twenty or twenty one years old, graduating at the age of twenty- three or over.

This is a degree of waste which a country like Kenya, with her meagre natural resources and heavy dependance on economie aid from the United Kingdom and other countries, cannat afford. Even the Soviet Unlon, whose awareness of the potentialities of education as an instrument of economie development and cultural growth, as weil as of general human progress and enlightenment, is perhaps more acute than that of most countries, sets aside only eleven years for education up to the end of secondary s chool, now that the school reforma inltiated by Mr. Khrushchev in 1958 are being introduced throughout the cru.ntry. 11

It may be said that the Kenya African child needs more

11. George Z.F. Bereday et al, eds, 'ihe Changing Soviet School, (Boston: Houghton lvfifflin Company, 1960,) p. 87. -23- time togo through primary, intermediate and secondary s

The number of children who go to school but never finish primary school can be described in no other terms tb.an as appalling. For example, of the 12 7 • 287 enrolled in the first class of the country's primary schools in 1950 only 36,552 reached the fourth class four years later. 12 It is estimated that the enrollment at the primary level is only 90o/o of the potential intake.13 Only

22-1/2o/o of these reach the intermediate leve!, 14 and the proportion of those who reach secondary school is only 11 o/o of the number in intermediate school. 15 This is a sombre picture ..

12. Op. clt., p. 77 • Nuffield Foundation. 13. Op. clt., United Kingdom Information Service, p. 11. 14. UNESCO, Paris, World Survey of Education, Vol. III. p. 1188. 15. ~p. 1188. -24-

There is no compulsory education for African children. although for European children this measure was introduced in

1942 for the ages seven to fifteen and for Asian chUdren at the same ti.me and for the same age group. 16 As this measure is unlikely to be introduced and enforced for African children in the foreseeable future, mainly because of lack of funds and a shortage of teachers, it can only be hoped that a gradù.al improvement in the qualiJications and training of teachers wUl pay dividends in the form of more and more children who successfully complete the first eight years of s chool.

Literacy in his own language. as well as the acquisition of some ability to manipulate numbers, constitute the goals towards which the efforts of the child are directed during the first four years of school. There may be, in addition to tb.ese. an introduction to

English. usually in the third year. 17 Some s chools, especially

those in the Coast Region and the cities1 start teaching Swahili at about the same ti.me. This may be an unduly heavy language burden on young children, and, indeed, it is not a few that collapse under it. 18 An examination taken at the end of primary school de- termines the number of those who get promoted to intermed.iate

16. Op. cit., UNESCO• Vol. 1. p. 706.

17. Op. cit. 1 Nuffield Foundation, p. 80 18. 20o/o of' primary school enrolment and 15% of intermediate school enrolment, according to figures supplied by the Ministry of Education. -25- school. Usually there are more chlldren quali:fi.ed to attend intermediate school than there are places for in these schools, in spite of the fact that large numbers never finish primary schools. Inevltably, many chlldren who are unable to gain ad- mission at intermediate schools, too young to go out seeking employrnent, repeat standard four hoping for better luck next time. 1 9 Othe r s gi ve up and stay home, or .find thel r way into the cities and there swell the ranks of the unemployed. It is an altogether unsatisfactory situation.

Those who survive the intermediate programme, devoted mainly to extending and consolidating the skills and techniques learnt in primary school, take the Kenya African Preliminary

Examination. This examination is written in English. 20 Again it is only thos e who do weil at this examination that are allowed to go on to secondary school. Those who are not so fortunate find jobs as junior clerks, unskilled artisans. or train as primary 21 school teachers.

The secondary school curriculum is in many respects similar to that of the English grammar school. There is the same emphasis on academie subjects, for these are primarily university- preparatory schools. However, in the secondary schools of Kenya,

19. Op. cit., Nuffield Foundation, p. 78. 20. Op. cit. • Nuffield Foundation, p. 82. 21. Op. cit., United Kingdom Information Service, p. 12. -26- unlik.e those of England and Wales, no classicallanguages or any of the modern European or Oriental languages are prescribed for study. There is little training in such skills as carpentry, brlcklaying and the like, all these being the speciality of technical and trade schools, such as the one at Kabete near Nairobi.

It is not surprising, therefore, that when a student finishes secondary school he has so little of what has been called

11marketable skillsu. 22 If he does not go to university, and only about 2% of those who finish secondary school achieve this distinc- tion, he may become a minor civil servant or train as an inter- mediate school teacher, a technician or a skilled artisan. 2 3

Kenya is an agricultural country. In 1958, for example, agriculture contrlbuted lu 68 million to the country1s income of lu 175 million, and provided exports valued at 1. 2.5. 9 million, or almost 90o/o of total exports. 2.4 There would therefore appear to be grounds for disquiet at the great emphasis placed on academie subjects in the countryts secondary school curriculum. There is no doubt that the country needs people educated and trained in ali fields of endeava}.r • but it would seem that Kenya's limited re- sources could be harnessed more usefully for the training of those who are most likely to contrlbute, in a most direct and effective

22. James B. Conant, The American High School Today (1961) p. 36. 2.3. Op. cit. , Nuffield Foundâtion. 2.4. Op. cit., United Kingdom Information Service, p. 5. -27- way, to the economie development of the country.

The situation at the present time is not unlike that which in the SoviEt Union gave rise to Mr. Khrushchev•s denun- ciation of those who would not nsoil their hands11 • a condemnation which led to the school reforms of 1959 and 1960 emphasizing produ- ctive labour as part of the education of ali. 25 What the Soviet Union could not afford Kenya can afford even less. And soit is not easy to disagree with the view of the compilera of African Education that:

Rural socieites can be developed, based on sound economy and at the same time providing social amenities and outlets for a wide variety of interests, aptitudes and occupations. A major aim of education in these (East and Central African} territories can therefore be con­ sidered primarily in relation to an agricultural en­ vlronment in which men and women can work, can be good citizens and can develop their talents, and it is by its success or failure in such an énvironment that education will be judged. 26

Judged by this criterio~ the ducational system of Kenya has only been partially successful, for as the Report goes on to state:

The aim of education must be to prepare Africans to live wellin their own country, and the system of education must not representa pale reflection of that given in England, where conditions are al­ together different, and where the native culture of the people is altogether different. 27

25. Robert UHch, The E ducati.on of Nations, (Cambridge: Harvard University Press, 1961), p. 277. 26. Op. cit., Nu:ffield Found àtion, p. 66. 27. Op. cit. • Nuffield Foundation, p. 67. -28-

The country1s urgent educational task, then, is to work out more realistic aims of education, realistic in terms of her needs, resources and the aspirations and talents of her citizens.

As the Report concludes this section of its observations, 11 What we seek in effect is a more liberal education for Africans based on their 28 own African environment and on their own way of !ife. n As Dr.

Julius Nyerere., the President of the Republic of Tanganyika, said in Nairobi on June 28, 1963, at the ceremonies mar king the opening of the newly-established University of East Afrlca, of which he was installed as the first Chancellor on the same dayr

W e are too poor in money and educated man-power to support an ivory tower existence for an intellectual elite••• This University must take an active part in 2 the social revolution we are engineering. 9

The Luo Child in the System

The Luo child enters primary s chool in much the same way as any other African child in the country does., that is, v.h en his parents or guardians feel thathe is old enough to benefit from the experience and if they can afford the school fees that have to be paid for his education each year. However, the willingness to send children to school is widespread among African people. Indeed,

28. Op. cit., Nuffield Foundation, p .. 67 .. 29. Africa Report, vol. 8, No. 7, July, 1963,. p. 18. -29-

ttAfricans are often ready to go without material advantages if

their children can be educated, n 30 believing as they do in the

universal e:fficacy of education.

As bas already been pointed ou~ the main task undertaken

during the first four years of school is literacy and elementary

arithmetic.. In the case of the Luo child. literacy means the ability

to read and write Luo. Each African child in Kenya's primary schools

is, as far as possible, taught in his own vernacular. The only

exceptions are those children who live in towns or cities. The diverse

tribal backgrounds from which children in this group come necessi­

tates the use of a language which ail can understand. Swahili is such

a language, and is the medium of instruction in urban primary schools.

The fact that African languages are used for instruction

in the country's primary schools and for tv.o years beyond that leve!,

is important in the educational progress of African children. not only because instruction nmust necessarily be in the vernacular, for the symbols of literacy must first be related to the mothe r tongue, tt31 but also because of the e:ffect of these langœ. ges on the way these children learn English, the language of instructinn from the seventh year ci.

school onwards.

Luo, then, is the Luo child•s native language~ It shares

30. Op. cit., Nuffield Foundation, p. 74.

31. Op. cit., Nuffield Foundation, p. 74. -30- with English certain features and characteristics which cannot but influence the way the Luo child learns English. For example,

Luo, lik.e English, uses the Roman alphabet. However,. this alphabet is shorter byf our letters in Luo than it is in English, for Luo does not have Q,. V,. X and z.

It is quite possible to teach or learn a language entirely by ear without any recourse to writing. This is how children learn language, for they me rely listen to, and imitate the speech of their parents and of people around them. This, however, is not the way

Luo children lea rn English, for speaking and wrl. ting the language are taught simultaneously. The written word is therefore at !east as important as the spoken word. Thel.uo alphabet is therefore a factor in.fluencing the Luo child's mastery of English, at least in its wrltten form.

The limitations which the Roman alphabet imposes on

English also beset written Luo. For example, this alphabet has no letters, for such English sounds as /tf,~ • 6 and f /, not to mention such vowels and half-vowels as /a, i,. ~ and 1\/. Re- course has therefore to be had to such deviees as --ch, th, and -sh. Similar deviees are also used in Luo. In some cases one letter is used to represent two different sounds• for example ~ is a long i, that is J ~/ in!:,:; , and a short_!., that is /1/ in ~ The se are Luo words for a ram and throw respectively. In English the same is -31- true, for a letter auch as e representa different sounds in men and me te.

There is nothing fatal about these shortcomings. The ideal alphabet, since it w ould have a single sign for each of the phonemes of the language concerned, is unnecessary. As a matter of fact~

Writing, is, in its essence, nothing but a means of recording language with sorne degree of efficiency• • • • A system of writing is good or bad according as it records, accurately or otherwise, whatever forrn of the language it is aiming to record. 32

Even more important than the characteristics which

English and Luo have in common are those which they do not share.

One example is spelling. Luo is spelt as it is pronounced. For instance, kelna komll(\ 'bring me that chair', is pronounced

/kelna komno/. A symbol auch as ! is always soft, and 1 is always hard. Even deviees auch as the symbols for 1 'C 1 and 1 Ô / are always differentiated. For example, dhano, •a person', is

/'6ano/ and the sign for /Il is dh. On the other hand, thim, •a forest' is 1 6i:m/ and the symbol for 1 e 1 is th.

In this respect, therefore, Luo contrasta sharply with

English, for English spelling is highly irregular. Therefore the Luo child learnlng Englishfinds English spelling very difficult to master, for he is used to the more or less phonetic spelling of his own language.

Fries observes of English spelling that·

32. John P. Hughes, The Science of Language,. (Random House, 1962), p. 120. -32-

the same symbol or letter stands for a m~mber of disti.nctly different sounds as. for example, the letter i in bite, bit, machine, or the letters ea in beat;-breath; h-;art, earth. On the othe r hand. the same sound is re~nted by a variety of symbols; for example, the vowel in sweet is represented by ea in beat, e in mete, i in machine and caprice, eoTn pe~ ie in belie,;;; and ei in recei ve. 33 -

The bewilderment of the Luo child, or indeed, any child, not excluding the English child, trying to pronounce such words as rough, bough, cough, dough. through and though, is uuderstandable• for the element that ali the se words have in common, ough, is pronounced differently each time, whereas the Luo child e:x:pects it to have only one pronunciation.

Bernard Shaw's deduction of the spelling of fish as

11 ghoti11 from the irregu.larities of English spelling was not entirely a pointless exercise of his malicious and destructive wit. It was these irregulariti.es and the difficulties they caused English children that prompted Sir James Pitman to devise a phonie alphabet which he believed wruld make reading English an easier and more pleasant task, for British children than it has hitherto been, and which John Downing, the Director of the Reading Research Unit at the University of London

Institute of Education has trl. ed to implement, with considerable suc cess. '•

33. Charles C. Fries, Teaching and Learnizg English as a Foreign Language, (Ann Arbor: The University of Michigan Press, 1945,) p. Il. -33-

lt is .John Downingts view that:

In theory, once he has learned the value of each conventional cbaracter of the alphabet and has mastered the skill of com.bining them into whole words, the phonie learner should. •• be able to discover from the print the meaning of any word written in the cbaracters of the alphabet at least to the degree that he under­ stands it in speech. 34

lt is clear, then, that the Luo child learning English is not the only persan for whom English spelling is a source of great anxiety. Nevertheless, these problem.s are very real to him.

He is fortunate that he does not have to learn a new alphabet when he learns English, for this would be an extra burden. A Russian child, for example, has to learn a new alphabet when he learns

English, if he is ever going to read or write it, for the alphabet used in writing Russian, the Gyrillic alphabet, differa, though not greatly, from the Roman alphabet. Any Oriental child learn:bg

English has to learn the Roman alphabe4 which is enti rely different from any of the alphabets used for writing, say, Hindi, Arabie,

Urdu or Persian.

At the time that the Luo child begins to learn English he is, of course, fluent in his own vernacular. He is taught Luo until his seventh year in school. Indee~ during this period he is tau ht all su b ects in Luo, with the exception of English. From the ~. .John Downing, Experimente with an Augmented Alphabet for Beginning Readërs in British Schools {Presented at the 2 7th Educational Conference, sponsored by the Educational Records Bureau in the City of New York, November 1 and 2, 1962). -34- sixth year English gradually becomes more prominent~ replacing

Luo in such subjects as Arlthmetic, Geography and History. At the beginning of the seventh year English~ introduced in the third year of school., finally replaces Luo as the medium of. instruction.

Lu.o itsel f is not taught after the seventh year of school.

The complete disappearance of the vernacular from the curriculum at the end of the seventh year of school is imposed on the school system by the exigencies of the situation. It is conceivable that Luo, or any of the other African languages of Kenya, could be the medium of instruction throughout the first eight years of school.

However, those children going on to secondary school would be unable to learn to any great depth any of the subjects on the curri­ culum~ such as Geometry~ Biology. Chemistry and Physics. owing to the limited vocabularies of most of these languages,. and the dif­ ficulty of carrying an abstract thought in them. Luo is the native language of the author, but he finds it impossible to tal k or think in terms of rectangles!.. obtuse angles, correlation coefficients .. magnetism. llght refraction and the llke in Luo. There are no equivalent terms in Luo.

It is obvious, then. that a language more highly developed than any of the African languages of Kenya has to be taught for use in secondary school and beyond. That language is English. Be aides

the vast vocabulary of English1 its flexibility and range as compared with, say. Luo. it is a fact that literature, technical and non-technical. - 35 - is more readily available in English than in any of the African languages. And as this literature is essential for secondary school work, students must rnaster English in order to be able to rnake use of it. The vernacular therefore gives way to English.

The priee that Kenya Africans will have to pay for rnaking their children abandon their languages so early in their schoollife is yet to be cornputed. As is stated in African Education:

To preserve the vernacular languages of Africa is to preserve the tribes that speak them and to strengthen the moral sanctions that rest on tribal rnernbership. If a distinctive African contribution is to be made to the world it must be based on the African•s love and respect for the mental inheritance of his people, and rnuch of this is enshrined in his language. 35

And so the Luo child gradually becornes bilingual, speaking his native language, Luo, and, in tirne, English. If he lives in a town, such as Kisumu, he adds Swahili to the languages he can speak. The pattern would be that he would speak English at School, Luo at horne and

Swahili everywhere else.

The Swahili Child in the System

The Swahili child is, in sorne respects, in a more enviable position than his Luo cornpatriot. His language, for exarnple, is the native language of sorne 7, 000,000 people ali over East

Africa. It is also the official language of the Republic of Tanganyika,

35. Op. cit., Nuffield Foundation, p. 80. -36- and is widely understood throughout East Africa, the Congo and parts of Nyasaland and Northern Rhodesi a.. The fact that Swahili has had a written form since the !5th century36 means that there has been enough time for a considerable literature ta develop.

This is mainly in the form of poetry and short stories. The Swahili child has access ta all these, whereas for the Luo child the only kind of literatl.lre available to hlm in his own language consista largely of a rich oral tradition, in addition ta a few books, mainly on Luo customs and practices .. 37

The language burden shouldered by the Swahili child is not as heavy as that carried by his Luo confrere. He has ta learn only one langu.age that is foreign ta hlm, that is, English.

He is taught Swahili in much the same way as the Luo child is taught

Lu.o. The Luo child. anthe other hand, often has ta learn Swahili in addition to English, the main foreign language. But since Swahili 38 is a 11 simple,. phonetic langu.age with.... a very large Bantu ingredient, 11 it does not create great learning problems for Giriama, Tiriki or

Kalenjir children, or for any other African children whose langu.ages belong ta the Bantu family of languages.

That the language question in Kenyats primary schools is a perplexing one ta educators and school chUdren alike is shawn by the views of sorne of the people who have thought and written about

36. Op. cit., L. S~ B. Leakey, p. 150. 37. For example, Paul Mboya's Luot Kitgi gi 'llmbegi (ULuo Custams and Practiceslt) (Gendia Press, 1937). 38. Op. cit., Nuffield Foundation, p. 81. -37- education in Kenya. For example, the members of the East and

Central Africa Study Group, whose report forma part of African

Education showed great concern at the effects of the language policy implemented in the schools. They wrote

The existence of Swahili in Kenya ••• and its place in school teaching is unfortunate, for it seems to have affected adversely the teaching both of the vernacular and of English. It cames in between the study of these two languages as an element confusing tre ed.ucational picture. 3 9

But they wrote this in 1952 and so could not have foreseen the intense nationalism and pride in everything of African origin that prompted members of the Kenya Legislative Council to pass, in December, 1962, two resolutions asking Government:

to recommend to the Regional Authorities that the teaching of Swahili be made compulsory in schools under their control; and

to take al! practicable steps to ensure a higher degree of proficiency in Swahili within the Pllblic Service. 40

To these legislators Swahili is not 11an element confusing the educational picture11 • On the contrary, they see Swahili as one da y replacing English as the official language of Kenya and, as has already been noted, regard it as 11 the most suitable medium for keeping the Government, the leaders and the people of this

39. Op. cit., Nuffield Foundation, p. 81. 40. See footnote 6 in Chapte r 1. -38-

country ln. c 1 ose contact. 41

If the language burden falls lighU y on the shoulders of the Swahili child in comparison with other African children in

Kenya, the Swahili child is not markedly better off than, say, the

Kikuyu child when it cornes to learning English. He. like the Luo

child, finds English spelling difficult to learn. Sorne English sounds,

especially such phonemes as/~/ and / ($../, do not exist in his language and so have to be taught to him as carefully as they are

taught to the Luo child, in whose language they also do not exist.

This aspect of English teaching is dealt with more fully in a later

chapter, but it is necessary to draw attention he re to the advantages

and disadvantages that Luo and Swahili children, by virtue of having

Luo and Swahili as their respective native languages, bring with

them to their first lesson in English.

Swahili, like Luo, uses the Roman alphabet. Therefore,

the Swahili child learning English does not have to learn a new alphabet. In this sense it is fortunate for the Swahili child that

Swahili is no longer written in Arabie script, as it was before the

colonization of East Africa by the British. 42 The present Swahili alphabet does not have Q and X, and so is shorter than the English alphabet by two letters.

41. See footnote 3 in Chapter 1.

42. Op. cit., L. S. B. Leakey, p. 142. -39-

The question to which an answer must somehow be found is how all African children attending school are to be made competent in English within the relatively short time which the schools have at their disposa!. For since English is introduced in the third year of school,. there is only a relatively short time before it becomes the medium of instruction in the seventh year.

Sorne of their External Handicaps

In addition to the difficulties that stem from the features and characteristics of the native languages of African chlldren learning English, there are problems that arise from the learning situation itself. The most outstanding of these are the qualifications of the teachers who teach English in the first eight years of s chool. the availabUity of reading mate rial suitable for young African learne rs and the opportunities these children have to speak English.

It has already been noted that not ail students who pass the Kenya African Preliminary Examination, which is taken at the end of intermediate school, are able to go to secondary school. This is simply a question o:f the insufficiency of secondary schools relative to the number of students qualified to attend them.

Many students in this situation undergo training for two years and become primary school teachers. Considerable numbers -40- of these teach in intermediate school~ such is the shortage of teachers in Kenya as a whole. For example, in 1956 there were

11,053 teachers in this category teaching in African primary and intermediate schools throughout the country, out of a teaching force of 13,371. 43

As these teachers teach all the subjects on the curri- culum, it means that they are also responsible for English teaching in primary schools, and in intermediate schools also whenever the staff does not include a teachers with higher qualifications.

This is an act of faith on the part of the Ministry of Education, for, implicit in the employment of these teachers is the hope that they will be able to cope with all the work to be done in these schools.

As far as the teaching of English is concerned, it is doubtful whether these teachers can be expected to perform in an:y outstanding manner. They themselves have had only six years of

English. and although this is augmented with courses taken during the two years of training, it is not enough to enable them to teach

English at the intermediate level, althoug it may, under certain cir- cumstances, be adequate for primary school English.

It is vitally important that children be given a good start in English, that is, such basics as pronunciation, should be intro-

43. United Nations, Special Study on Educational Conditions in Non­ Self Governing Territoriest 1960, p. 51; Progress of the Non-Self Governing Territories Undër the Châ:rter. Vol. 4, Edûcational Conditions, 1961, p. 95. -41-

duced only by people who know English weil. The author found at

Makerere that many af the peculiarities of his speech were really

the pronunciations he had learnt from his primary school teachers,

and he can attest to the difficulty of eliminating those peculiarities.

Some he probably will never eliminate.. As Nelson Brooks writes,

"The learning of a language involves behaviour patterns that the learner cannat create on his own, nor derive from books. They must be modelled .for hlm. 44 The only persan who can do this

11modelling11 satisfactorily is one who has complete mastery of the language, at least up to the level at which he expects to teach it, and knows something about the way children learn languages.

Teachers with low qualifications, then, handicap Luo and Swahili children, indeed, all African children in primary school, in a way that can only be remedied by the employment of prpperly qualified and trained teachers. It is these teachers that should in- troduce children to the basic features of English: its sounds, in- cluding stresst intonation and pitch, its grammar and ward arder, and the way its words are built up.

That these children at best get only an imperfect grounding in these fundamentals can be deduced from the fact that even at university leve!, at Makerere University College, it is necessary

44. Nelson Brooks, Language and Language Learning, {Harcourt, Brace and World, 1960), p. 141. - 42 - to institute courses for freshmen designed to eliminate the solecisms in their English and improve their English in every way.

As for reading material, there is no lack of it. It may also be said that this literature in the long run serves a most useful purpose: it broadens these children• s outlook on life and enables them to perfect such skills as comprehension, summary writing, memorization of appealing poems and vocabulary ex­ pansion. Most important from the point of view of this dissertation, it strengthens the se children' s grasp and deepens their knowledge of the language as a whole.

One fact about this literature is that it is usually designed for the English child living in England. Stories such as those about Robin Hood are naturally placed in an English setting.

African children reading them may understand and enjoy them, but their understanding and enjoyment cannot be as complete as an English child' s, since they are completely unfamiliar with the social and cultural background against which these stories are set.

In addition to the allen nature of the setting and idiom of the se stories, it has to be taken into account that by the time an African child has learnt enough English to be able to understand them, he is in his seventh year -43- of school~ bas bad about five years of English and is thirteen or fourteen years old. These stories, for example, Cinderella and

'Ihe Black Tulip, general!y have little, if any, emotional appeal for children of that age. The author himself remembers reading

Alice in Wonderland at the age of fourteen and finding it thoroughly boring. However, a second reading ten years later completely reversed that judgement.

Even such weil known nursery rhymes as Hickory Dickory

Dock, Little Jack Horner and Mary, Mary, Quite Contrary are in­ troduced when the African child is too old to fu.lly enjoy them. It is a pity, because they thereby miss much that English children absorb and live with ali their lives.

It was such considerations that led to the establishment of the East African Literature Bureau in Nairobi in 1948. Local talent was to be recruited, and stories with an African setting and in African idiom were to be written, not only for schools" but for the whole reading public. A few books have already appeared.. The

Bureau itself publishes Arrow, a periodical for schools which first appeared in 1958..

Other possibilities still to be explored are the writing down and translation into English of the wealth of tales, myths and songs that exist in Luo, Swahili, Kikuyu and other languages. In these, and in others still to come, the African child learning English -44- will at last find reading material designed specifically for him.

Of the third type of handicap under which Luo, Swahili and other African children learning English labour, tbat is, lack of opportunity to speak English or hear it spoken outside the school, it can be said that it is gradually being overcome. For example, there are now many Africans who have had their university education overseas or in East Africa. 45 These people, because they are professional people, use more English than they do their native languages. Their children would therefore have the opportunity d speaking, and hearing the language spoken, at home. However, for the vast majority of African children, the native language is the language spoken at home. Even at school

English is not spoken all the time. At Makerere, for example,

Airican students relapse naturally into Swahili or their native languages as soon as lectures end.

As more and more Africans buy radio~, their children will be able to hear English spoken in their homes, even if the parents themselves do not speak the language. And now that television has arrived in the country, 46 it will greatly augment the language learni:rg opportunities that radio has hithertb provided.

With the expansion of education in the country, mo:œ and more chil- dren will have parents with secondary school or university education,

45. United Kingdom Information Service, Canada, Kenya, January, 1960, p. 12. 46. It started working in October, 1962, and is managed and controlled by the Kenya Broadcasting Corporation. -45-

and s o will have the opportunity of speaking English at home.

Teachers and other people will also have to be called upon to help in this field. When a survey was made of African

schools for African Education in 1952, a distnrbing degree of

neglect of opportnnities to practise English outside the classroom was discovered. As was reported:

Observation bas shown that women ed.ucationalists often do not use English regularly as the medium of intercourse outside the classroom. It is claimed that they have neither the time nor the patience to slow down conversation, comment, or command to the pace of English understanding when they know the vernacular. This happens even in senior secon­ dary schools. 4 7

When it is added that these 11 educationaliststt are those for whom

English is the native language, and who for that reason present

African children with their best change of hearing English spoken weil, it is to be regretted that they do not Q. ve these children that chance. It may be that theyt too, wish to master the Afrlcan langages they are learning, and so are an.xious to practise them whenever they come into contact with native speakers of those languages. It should be possible to work out some arrangement whe reby both groups would benefit.

These, then, are the major difficulties and problems which confront the Kenya African child learning English. In some

47. Op. cit., Nuffield Foundation, p. 113. -46-

of them he has to work out his own salvation. For example. it is up to him to make use of the opportunities of practising his

English that present themselves. However, in order to overcome other difficulties he needs thehelp of other people: ordinary

English-speaking people~> writers, broadcasters and the community.

Above ali, he needs the help of his teacher. -47-

CHAPTERID

THE SOUNDS OF ENGL!SH, LUO AND SWAHILI

Bilingualism has be en defined as 11 di vided linguistic

1 allegiance .. ·~ It is difficult to estimate the number of African children that are bilingual, but it is kn.own that many of them are.

If they live in towns and cities they usually speak Swahili in addition to their own native languages. If, on the other hand, they live in the country they speak their native languages and the language of a neighbouring tribe. lt is not infrequently that some in the second group have a perfect command of two or more such languages.

Jomo Kenyatta, for example, spoke Kikuyu, Meru, Kamba and

Masai before he went to school. However, in his case the fact that his father was a Kikuyu and his mother a Masai gave him initial advantages in this respect that most African children do not have.

If, as has been shown, 2 Kenya is committed to giving a thorough training in English to ail her children attending school, it seems that the most effective way of doing so, àl ort of employing English governesses and tutors for them, would be to inculcate and .encourage in them habits most likely to produ.ce the

1. André Martinet in the Preface to Uriel Weinreich•s Languages in Contact, (Linguistic Circle of New York,. 1953), p. 3.

2. See Chapter 1. -48-

11 desired degree of 11divided linguistic allegiance • The second or third languages that many African chlldren speak, with the exception of Swahili, are usually learnt through informa! contacts, such as those provided by the market place, the soccer sta.dium and the village fair. Swahili, though presented more systematically at school, is also learnt to a considerable extent in this fashion. English, on the other hand, is presented in school.

The understanding and mastery of the basic sounds of

English is an essential first hurdle that has to be surmounted.

Failure or only partial success in this will produce students who at best speak English with a foreign accent, or at worst English that is hardly intelligible. Inasmuch as a foreign accent is 11merely the carry-over into a second language of habits of articulation from onets first language", 3 it is a factor which militates against comprehen­ sibility. Therefore the desirability of its elimination need not be argued.

Better still it would be economical, in terms of time and effort, if from the sta.rt students learnt to prod.uce English sounds as they should be produced. lt is Bloomfield who observes that

3. John P. Hughes, The Science of Language, (Random House, 1962), p. 241. -49-

When a foreign speaker reproduces the phonemic values of our language so asto make himself understood, but do es not dis tri bute the non­ distinctive features in accordance with our habit. we say that he speaks our language well enough, but with a foreign 11accentf1• 4

In this dissertation, all references to English sounds and charac- teristics are in relation to Received Pronunciation, for this is the only varlety of spoken English with which the author is su:fficiently familiar for discussion purposes.

The sounds of English present the Luo or Swahili

child with certain problems. In the opinion of Fries:

The view of most naive speakers, that in the language of foreigne rs "aside from a few striking differences ••• the sounds they use are the same as those they aré familiar withu, is very far from the trulli. and this mistaken view often interferes greatly with any effective mastery of a foreign language. 5

However, it would not be difficult to demonstrate that a considerable number of Luo and Swahili phonemes are either practically the

same as their counterparts in English, or not too dissimilar to them.

This will become clear in the course of the discussion.

A phoneme is tta minimum unit of distinctive sound- feature11 • 6 For example, bet and pet are similar in ali respects except for the initial sound in each word. Since this difference be- tween the two words distinguishes them from one another,. the two

4. Leonard Bloomfield, Language (Henry Holt & Company, 1933) p. 80.

5. Charles C. Fries, Teaching and Learning English as a Foreign Language, (Ann Arbort The University of Michigan Press, 1945) pp. 14-15. 6. Ibid• Bloomfield, p. 79. -50- initial sounds~and _g_ are phonernes, and are ~ refore transcribed thust /b/ and /p/.

The Consonantal Phone mes

There are both differences and similarities between sorne of the consonantal phonemes of Luo and English. For the Luo child learning English certain difficulties arise because of these sirnilarities, and certain others because of the nature of the dif­ ferences between Luo and English consonantal phonernes.

The English /d/ as in dim , for example, is a voiced dental stop. The Luo /d/ as in dirn, 1hit the target', is also a voiced dental stop pronounced with the tangue in the sarne position as in the English /d/. The two words are in fact pronounced the same, except that the Luo word has a longer vowel sound than the

English word.

Similarly /p/ as in pill, has the same sound value as

L/p/, as in pirn, 1take a measure of1• Both phonemes are voiceless bilabials. The difficulty ex.perienced by the Luo child is that although in the above exarnples the English /p/ and the Luo /p/ are identical, there are phoneti.c differences between different /p/'s in English, depending on their position in a word, whereas the position of the Luo

/p/ in a word does not affect its sound value. For example, a /p/ in the initial position in a word, as in pill, is pronounced with a puff -51- of air whose force can be noticed if a sheet of paper is held about six inches fro:..n t.he mouth as the word is pronounced. The /p/ in the central position, as in captain, is not pronounced with a similar plosion; in fact, no effect is noticed on the paper b.eld before the mouth. The same applies to /p/ in the terminal position as in leap; in this case there is a stop after the closure of the lips for the p-sound, but the sound itself often merges into the following word if there is one. Ali these different kinds oi /p/ are said to be allophones of the phoneme /p/, and there are phonetic differences between them.

ln Luo, on the other hand, there are no allophones of

/p/ or of any other consonantal phoneme. It means therefore, that on discovering that the g of pen is the same as the p of his own language, the Luo child uses this /p/ even in places where other allophones of /p/ are appropriate. This constitutes a failure to

11 distribute the non-distinctive features in accordance with our habit11 which Bloomfield mentions. 7

An example of differences between Luo and English consonantal phonemes can be seen in the Luo / e/ as in thim

/ e i:m/, •a forest1, and the English / e / as in~ /6in/.

Whe reas the English / e / is a dental fricative, the Luo phone me

7. See footnote 4. above .. -52-

is a dental affricate; the first is made with the breath escaping

between the tongue and the upper teeth, whereas in the second the

front teeth are pressed fairly firmly on the tongue and the breath

is then allowed tp pass with a sudden release between the tongue

and the upper front teeth. These therefore are not the same sounds.

However, the Luo 1 8 1 is not so different from the

English 1 8 1 that each time he uses it instead of the English phoneme he will not be understood. It is just ove r a year ago when

the author1s attention was drawn by a friend to his pronunciation

of thatch (B~tf 1,. He had been using the Luo 1 e 1 al! along but no o ne had failed to understand him.t although he must have sounded peculiar to English-speaking people, as he did to this particular

friend.

As Uriel Weinreich observes~

The practice of the same phonetic habits in both languages is an efficient way of easing one 1s burden of Hnguistic deviees. As a matter of fact~ it requires a relatively high degree of cultural sophistication in both languages for a speaker to afford the structural! luxury of maintaining separate subphonemic habits in each. 8

It will be seen, then, that the difficulties with which the Luo child learning English has ta contend are of three kinds:

Those which stem from the close similarity between certain Luo

8. Urie! Weinreich, Languages in Contact, {Linguistic Circle of New York, 1953), p. 24. - 53 - and English phonemes, although this similarity does not extend to the allophones of these phonemes, since Luo phonemes do not have allophones; those which stem from the differences between certain Luo and English phonemes which at first sight may appear to be identical; and those that are the result of the absence in Luo of certain English phonemes.

It is indisputable that;

the average individual, listening to speech sound in any language, identifies phonemes only within the framework of the language he already knows, his native language. 9

Or as Roger Brown states it, but in different terms:

Most of the variations that occur in speech are not used to dist inguish one referent from another •••• However, languages do not all recogni.ze the same distinctions and so a person learning a second language has serious difficulty "hearing" that language as it is intended to be heard and pronouncing it as it is intended to be pronounced. 10

An exhaustive but selective examination of the phonemic structure of

Luo and Swahili will greatly clarify the nature of the difficulties which the consonatal phonemes and other phonemes of English present to Luo and Swahili children learning English. For this purpose the chart on page 54 (a) will be referred to as needed.

9. Op. cit. .John P. Hughes, p. 264.

10. Roger Brown, Words and Things, {The Free Press, 1958) pp 16-17. -54-

As can be seen from the chart, there are a nurnber of significant differences between the phonemic structure of .ù.l.o and English on the one hand, and Swahili and English on the other.

As for the differences between the phonemic structure of Luo and

Swahili, these are not significant, l.imited as they are to the absence in Luo of the Swahili phonemes /V /./z /and /j~ /, and the absence in Swahili of the Luo phonemes /e/ and /o/. It is beœ. use of the close phonemic similari ty between Luo and Swahili, in addition to other linguistic feature s that they have in common, for example, vowel distribution, that mak.es it possible to consider these languages together in a discussion involving the linguistic and other features of English.

For the Luo child learning English, /v/, /Z./, 1 [ /, and/ 3 /, among the consonantal phonemes, will be completely

new sounds. These he must be taught1 for, left to his own deviees, he will pronounce them in a manner betraying his Luo language

habits. For example, the English lv11 as in vote, usually comes out as /b/ in Luo speech, so that the Luo child pronounces ~

/vout/ as /bout/. There is a peculiarity about the vowel sound of this word, as pronounced by the Luo child, vmich will be considered later.

It is not only the English /v/ that becomes /b) in

Luo speech; the Swahili /v/ is similarly rendered. A Swahili -54- (a)

- ...r:: J ~ ...r: 0 ..c

"J) p a p 1 ::! ~ oo-- 0 () 11 CD "::f :r-E. 0 0" ? '.) (/) {) ~.._ Ç) :S cr / 8 Ç> ~ .:L .J. ;i 11) ID < l l() ct - ? ."? QI "'?- ct1 { iJ} t-f •.J .7lJ ~ d tv ~ •...J w :F ~ 0 H ·-' ..J 0... ~ tt 1-1 ..1 rf) ' b ~ - -:J - ~Q r: ..... IY (/) (]. (Le( ~ ..1 > (ff __ , ~ L z: ~0 [.. 0 oJ w o..l. l- <( u.. N. 0 t- "7J ï[TJ ç:- N 6 __J ~ r.J ;::: (/.) (l ~ -' (ji r:: Cf) 1 < -f-1 ~ +l --' ~ 7 ~ .\0 JtfJ tt ! <[)

word such as viti /vi ti /, 'chairs' therefore becomes /bi ti /

in Luo speech. ln this respect it is useful if a Luo child setting

out to learn English already speaks Swahili, for then he will have

learnt to pronounce /v/ and so "WOuld not say /bout/ when he

meant /vout/ but only when he meant boat.

As for the English /z/ as in zero, and / j /as in

shield, the Luo child usually replaces them with / s/. He the re-

fore pronounces the above words as /si rou/ and /si ld/. As

far as the English /z/ is concerned, it is not surprising that the

Luo child should substitute /s/ for it, for Luo bas the phoneme /s/,

an unvoiced alveolar fricative, but lacks its voiced variant /z/.

The Luo child will therefore have to be taught to voice his /s/ when-

ever he has to the / z/ phoneme in English.

It is not so easy to explain why the English !J /

becomes /s/ in Luo speech. The Luo /s/ and the English /s/ are Il identical; both are the refore 11 grooved sibilants11 • Therefore

the Luo /s/ is not very far horizontally from the English f.J /, and so is probably a reasonable substitute for it, from the point of

view of the Luo child. He has to learn to produce the / J / sound

by placing his tongue in a position slightly further back than it

normally occupies for the production of the Luo /s/.

The only new sound of the consonantal phonemes that

!:S. Urie! Weinreich, Languages in Contact, (Linguistic Circle of New York, 1953); p. 24. -56- a Swahili child learning English has to master is the English

1 3 1 as in leisure /Le3a). His difficulty with this sound is not as serious as is the Luo child1s with the same sound, for Swahili has the. unvoiced palato-alveolar fricative, / J /, whereas this sound does not exist in the sound structure of Luo. Ail that the

Swahili child has to learn to do in order to produce /3/ is to voice the / J / of his own language. The Luo child, on the other hand, has to proceed more indirectly, starting from his own Luo /s/ to the English IJ /and finally to the English /3 /.

The second group of English atounds among the con­ sonantal phonemes which Luo and Swahili children find difficult to learn includes all the English phonemes which are close to th.eir Luo and Swahili counterparts but are not quite like them.

The example of/ 6 / has already been discussed. Other examples are / â /, /r/, /h/, /t /and/ j /. The Luo /~ / differs from the English 1 '0 / in rouch the same way as the Luo 1 e / differs from the English 1 e /. The Luo 1 G 1 is a dental affricate whereas the English / '() / is a dental fricative. When the Luo child pro­ nounce s this / (!:, ls/ he uses the voiced dental affricate of his own language rather than the correct dental fricative of English.

Rather like the difference between the Eng lish / e/ and/ 'Ô 1 and the Luo ones is the way the English phonemes

/tf 1 as in chair, and 1 j 1 as in jest differ from their Luo -51- equivalents. The Luo /tj J is made with the tangue firmly on the alveolar ridge so that no breadth passes through. The tangue is then lowered suddenly at the same time that the

'breath is released, making it come out in a little explosion.

The Luo sound / J / is made the same wa y except that it is voiced, whereas !tJ / is unvoiced. These phonemest then. are palato-alveolar affricates which are made further forward than their English counterparts. When the Luo child learns the words chair and jest. , he uses the Luo /tf / in chair and the Luo / J / in jest rather than the appropriate English phonemes.

Also resembling their English equivalents are the Luo phonemes /r/ and /s/. However, the l..uo /r/ is not qJ.ite like the English /r/, since it is made with severa! taps of the tangue on the alveolar ridge. Usually the more such taps the better, in contrast to the English /r/ which does not require as many tapa. lt is this /r/ which the Luo child uses in pronouncing words such as run /rl\n/, rate /reit/ and the like.

The Luo /h/ as in hombo/hombo/ •a plea• is also not

.quite like the English /h/ as in head /hed/. Like the English /h/ • the Luo /h/ is a glottal fricative but is strongly aspirated, in contrast toits English equivalent. It is therefore rather like the -se-

Russian phoneme at the beginning of such words as k.b.orosho 12 •weil' and khozyain •host• • In pronouncing the word head, therefore, the Luo child uses the strongly aspirated glottal fricative of the sound scheme of his own language, rather than the English /h/ which is only slightly aspirated.

As will be seen from the chart, Swahili shares with

Luo those phonemic features which have been considered as causing the second type of difficultvies to Luo and Swahili children learning

English. The Swahili child therefore has to make the same adjust­ ments to his habits of articulation that the Luo child has to make in

or der to produce the English /8, Ô , tj 1 j, r, h/.

Consonantal phonemes never occur in isolation in

English' they always come before a vowel phoneme or another

consonantal phoneme or a series of these in arder to make up a word. For Luo and Swahili children, certain difficulties arise when two or more consonantal phonemes follow one another in a word without a vowel phoneme separating them. For example, street has an s, t, r cluster of consonants, and obstruction a b, s, t, r cluster of consonants. Such clusters are discussed below.

12. My own transliteration. - 59 -

Consonant Clusters

There is a very small number of consonants that can occur together in Luo or Swahili. In Luo, for example, the letter ~ never occurs in isolation but always together with h to form ch /t / as in chak, 'milk'. This, however, is not a consonant cluster since the essence of consonant clusters is that each consonant in the cluster is pronounced separately. In

Luo .!_and~ can occur in juxtaposition, as in kelna, 1let me have'.

Other examples of consonant clusters in Luo are lk as in kel ka, fbring it heret, mp as in tem piyo, 'try quickly' mb as in mbaka, •conversation', ~ as in temruck, 1 temptation', and rr as in herruok, 'affection•. It will be noticed that Luo consonant clusters consist mainly of two consonants. There are no clusters consisting of more than two consonants.

Of importance, too, is the fact that most Luo con­ sonant clusters occur in the initial or medial position in words, and very few in the terminal position. ln fact, of the examples listed above only ~ can occur in the terminal position, as in demb, fhold carefully•. One not listed above is nd as in dind, 1pack tightly1, which occurs also in the initial and medial positions.

ln Swahili, as in Luo, consonant clusters consist only of two consonants. These are mainly a combination of rn with other consonants. For example rn bele, 'forward', mdogo, t small' -60-

said of a person or an animal, mfuko, •a pocket•, Mganda1, a persan from Uganda, •.'Mhindi•, an'Indian1 , mji, 1a town or

city• and so on to mzee. 'an old man*. The other consonant with which severa! other consonants form clusters is n. For example, nd as in kuandika, •to write',. !::f as in kuingia, 1to enter•, nj as in njaa. ' hunger', and~ as in mnene, 'plump1•

Swahili consonants never occur in the terminal position, and the same is true of consonant clusters in Swahili. This position is always occupied by a vowel. It means theref1Dre that in Swahili there can be no consonant clusters made up of a terminal consonant in one ward and an initial one in the following ward. There are examples of auch consonant clusters in Luo. as has been pointed out.

It will be seen, then1 that Luo and Swahili children learning English will find it difficult to pronounce English consonant clusters. not only because most of these do not occur in Luo or

Swahili, but also because even the ones which do occur in these languages do so in positions in which they usually do not occur in

Luo or Swahili.

Such English consonant clusters as ~as in dents,

~as in~ 1 !) k 1 as in sin.k.f 1 ~ 1 as in bulge, mpt as in

empty 1 mps as in glimpse do not occur in Luo or Swahili. Luo and Swahili children therefore place vowels between s ome of the consonants in the clusters in order to make them easier to pronounce. Sometimes such a vowel is placed in the terminal position~ especially in the speech of Swahili children, -who do not have consonants in the tenn inal position, as has been pointed out.

For example, bulge \/ 1bi\ l' J 1 in Luo speech becomes

/ r br\lij/, and /b 1\ lji/ in Swahili speech. In Luo speech glimpse

/ glimps] becomes / glimpis/, and 1 glimpisi/ in Swahili speech, whereas empty /empti/ is /empiti/ in the speech of both groups.

These are very real difficulties, for it is doubtful whether an

Englishman hearing / empti/ pronounced /empiti/ would recognize it as the English word empty. Another English-spealdng Luo or

Swahili individual probably would recognize the word even if so pronounced, possibly because he would be aware that no auch word exista in his own language, but mainly because he wruld know that extra vowels usually appear in the speech of his people whenever they encounter unusual English consonant clusters. But since Luo or Swahili children should not, and indeed cannot, expect to rely on the ingenuity of a listener for the unravelling of the mean- ing of what they say, these extra vowels have to be eliminated.

English consonant clusters consisting of more than three consonants acquire two or more extra vowels in the speech of Luo and

Swahili children. The vowels are usually /i/ or /e/. For example, closed street has the zd str cluster o! consonants. In Luo and -6Z-

Swahili speech, therefore, /klouzd strit/ becomes /klouzdi

sitrit/ Forked bridge, /f 0 :kt brij/ becomes f:::>: kti brij/.

Consonant clusters, then, constitute pronunciation probler23.s

for Luo and Swahili children which, left to their deviees, they

solve by using extra vowels. This makes it difficult for people

to understand what they say.

Vowel Phoneme s

Tho se English vowel phoneme s which have no

equivalents in Luo or Swahili are more difficult for the speakers

of these languages to produce then the corresponding consonant phonemes. English vowel phonemes in general require more movement and flexibility of the sound-producing organs than

either tle Luo child or his Swahili compatriot needs for the production of the vowel phonemes of his own language. ln the minds of Luo people it is something borde ring on the vulgar for one to open one•s mouth wide when talking. The only tLme when this is overlooked is when one is shouting. The Luo child there­ fore starts with this disadvantage, and will, for that reason, find it difficult to produce English vowel phonemes.

There is, however, a more fundamentalle\e 1 at which the difficulties of Luo and Swahili children with English vowel phoneme s multiply considerably. - 63 -

For exarnplet an exarnination of the chart would not lead one to expect the Luo-speaking child or the Swahili-speaking child to find it difficult to produce the English high front vowels fi/ as in beat /bit/, and /I/ as in bit /bit/. Luo has similar vowels: /i/ as in~ /rito/ 1to wait for•, in whlch /r/ is

1/r/ and ois 1/o/, and /I/ as in timo fti:m;)f, 'to do•.

Swahili, too, has similar vowel phonemes, for example, /i/ as in kiti/kiti/ •a chair', and I as in kikapu/klkapu/, •a basket•.

And yet, contrary to what one would expect, Luo and Swahili children have difficulties with the English /i/ and

/I/. There is a tendency among Africans, not just Luo and

Swahili chlldren, to use the phoneme /i/ when they should use

/I/. For example, instead of saying live /lx v/ they say /liv/ whlch is confusing to their listeners, who think they mean leave.

The meaning of the statement may be revealed by the context if

/i/ is used instead of /I/, but one can imagine occasions when even the context would not reveal what is rneant. For exarnple, an

African rnight say I don•t know whether 1111/liv/ when he meant

/l:tv/, and if he were ill in hospital, it would not be immediately clear exactly what he meant.

On the other hand there is no corresponding tendency among Airicans to substitute /I/ for /i/. It is always /i/ that is -64- substituted for /1/. For example. meal /mil/ is always pronounced with /i/ and never as /m 1/, whereas mill/mD./ generally cornes out as /mil/. lt seems, then, that, for Luo and Swahili children. as weil as other Africans who confuse

/i/ and /I/, it is the phoneme /i/ which causes most of their difficulties with the high front vowel phonemes.

The Luo/i/ is often used to contrast minimally with the Luo/I/, that is, as in the English words meal/mil/ and mill/mn/ or scene /sin/ and sin /s!n/, it is /i/ and /l/ which distinguish the pairs kich /kitj /, 1bees1 , and ~k!tf /, tan orphan•. However, there are not many such word pairs in

Luo, so that to confuse /i/ and /1/ in Luo changes the meaning of only a few words. This contrasta sharply with English, in which such word pairs are numerous; for example, seal and sill, feel and fill, field and filled, ~and fit, feast and fist, heal and hill, heat and hit, ~eap and lif>, leak and lick, seat and sit, weak and wick, and many others. When these word pairs are jronounced, /i/ and /I/ contrast minimally and so keep their meaning apart.

The Luo or Swahili child confuses such English word pairs because /!/and /!/ are used to distinguish meanings only to a limited extent in Luo and Swahili. As Fries states:

If a sound feature is used in my language to distinguish meanings, then it is easy for nE to hear that feature when it is used in a foreign -65- language. But if in my language a difference between two sounds is ne ver used to distinguish meanings, it is difficult :br me to hear that difference in another language when it is thus used.l3

Another reason for the way Luo and Swahili children confuse the Engllsh /i/ and /I/ is that in both Luo and Swahili /i/ and/!/ are often used interchangeably. For

example, the Swahili word Kilemba, •a turban• 1 may be pronounced either /kllemba/ or /k:ilemba/. It m.akes no dif- ference whether /i/ or /!/ is used. although the word is usually pronounced with /!/. Similarly, such Luo words as bilo, 'to taste1, rido, •to tighten', and diro, •to throw•, may ali be pronounced with /i/ or /I/ without giving rlse to the disorientation of the listener. These children obviously transfer this habit into

Engllsh and thereby confuse words which should not be confused.

It is clear, then, that Luo and Swà:iJli children confuse the English /i/ and/!/ mainly because these phonemes., although existing in both Luo and Swahili, do not contrast minimally in Swahili, and only to a limited extent in Luo. The Luo child has to be taught that the difference between peel and pil1 is very rouch like that between the Luo pair !:J:_/bi/ •come•, and bi/b!/'wring', and that /i/ and /!/ are never used interchangeably in E:rg lish, as they are in Luo. The Swahili child. too, has to learn tba. t the two sounds are distinctive.

13. Op. cit. Charles C. Fries, p. 16. -66-

The extent to which Luo and Swahili spelling adds to the difficulties Luo and Swahili children have with the English

/i/ and /1/ is not easy to determine. lt will have been noticed that the two phonemes are spelt with the letter .!:_ in bath Luo and Swahili. This contrasta with the practice in English, for

English has a wide variety of letters for each of the two s ounds ranging from ------ea» ee, ei, eo» e, ie and i for /i/, as in -neat, feet_ receive,. people, mete, field and sardine, to!.! ;: and ui for /1/, as in bill, women and build. Luo and Swahili speiling, then, does not prepare those speaking these languages for the great variety of ways in which /i/ and /1/ are spelt in English.

As Robert Lado states:

The student often mispronounces words because of influences from•••• wri ting systems. When bath the foreign language and the native. language use the same alphabet, the problem may be traceable to one of two possible causes ••• the same symbol mightr.epresent two different sounds in the two languages. In such a case the student tends to transfer the native language symbolisation to the foreign language. 14

To the extent that the Luo or Swahili child automa- tically says /i/vihenever he has to pronounce the letter i, and since in his own language it makes no great difference whether he pronounces the letter as /i/ instead of /1/, his difficulty with

14. Op. cit. Robert Lado, p. 20. -67- the English /i/ may be attributed to the English letter ~ itself when the sound /I/ is spelt with this letter. Therefore miss and kin become /mis/ and /kin/ in Luo and Swahili speech because ~ is identified as /i/ by Luo and Swahili children.

Another English vowel phoneme which Luo and

Swahili children find difficult to prodll.ce is the vowel sound in auch a word as~· There is no auch pheneme in the sound system of Luo or Swahili. The Luo or Swahili child usually substitutes his own /a/ for the English / (f.../. Words .such as man/ m ~ n/ -bat/b?i...t/ and clan/kl- ~ n/ therefore come out as /man/. /bat/ and /klan/ in Luo and Swahili speech.

Although the meaning will generally be clear) the Luo and Swahili

/man/ being rather like the Lowland Scottish pronunciation of the same word,. there are a number of English words which are bound to be mistaken for others if the Luo or Swahili child does not pronounce them with their correct 1 è:J:.. / pheneme.

For example, if the English/ ~ / is replaced by the Luo and Swahili /a/ in such word.s as tan and ban the result will probably be close m ough t o ~and barn to be mistaken for them. It is Robert Lado who states: -68-

Experience shows that when the foreign language uses a phoneme which does not exist in the learner's native language, that is, v.h en the re is no phoneme in the native language that could be transferred to the foreign language and actually function as the phoneme in que'Stion~ the student will not be able to produce that phoneme readily in learning the foreign language. He will sub­ stitute some other pheneme from his native stock. IS

As the chart shows, the Luo and Swahili /a/ is the pheneme closest in sound value to the English / ~ / in the sound system of English and the two African languages. Therefore when the Southern

Englishman pronounces the word ~ it is their own /a/ that

Luo and Swahili children hear. They have to be taught to hear

/~/, for it is intermediate between /e/ and /a/, both of v.hich exist in Luo and Swahili.

The last set of vowels which Luo and Swahili children confuse is made up of the phonemes /u/ as in pool/puJ/ and /ü/ as in pull/pul/. Luo has both phonemes, for example, lnm/lum/ 'grass•, and huxn./hum./•to marvel•. The same is true of Swahili also, for exa.m.ple, ~o/umbo/lphysical appearance•, and Umeme/umeme/'lightning'. It is surprising, therefore, that these children should have any difficulty with the English phonemes, for these are practically identical to the Luo and Swahili equivalents.

15. Op. cit., Robert Lado, P• 13. -69-

What in fact happens is that in the mind of the

Luo or Swahili child no definite boundary exista between /u/ and

1 0 /, ee that words with / 0 1 are invariably pronounced with

/u/ by Luo and Swahili children. For example, pull/pul/ become s

/pul/ as if it were spelt pool, and bull/bO 1/ becomes /bul/ as if it were spelt 11booll1•

It would seem, then, t'hat the substitution of /u/ for

/0 / by Luo and Swahili children learning English is not unrelated to the replacement by the same children of /I/ with /i/, instances

of which have already been examined. In both pairs of vowels it is the short vowel that causes sorne pronunciatinn difficulty,. and the long one which is substituted for it.

As with the /i/, /I/ pair; the phonemes /u/ and /0/

can be used interchangeably in Swahili, that is, a word auch as umeme, 'llghtning1, can be pronounced /ume me/ or / (..)meme/ without any violation of the accepted rules of Swahili pronunciation.

However the same does not apply to the Luo/u/ and / U / ..

In Luo the distinction between /u/ and / U / exista, as it does in Engllsh. It means, therefore, t'hat the Luo child

cannot pronounce the word uno, •a rope', as / Uno/ instead of

/uno/ without being told to speak better Luo, for there is no such word as / U no/ in the langt:~ge. An even more clear-cut distinction

exiats in certain pairs of words which, like similar English pairs, -70-

have to be given the correct vowel value ü the proper meaning

is to be conveyed. For example, ruyo means •to hurry aomeone

up• when pronounced /ruyo/ and 1to sit up late at night•, or •to

remain awake all night1, when pronounced /ruyo/. Similarly

luro/luro/ means •to approach atealthily' and luro/luro/ means

•to gut an animal'. There are many auch pairs in Luo.

Corresponding to these Luo pairs are the English

words -fool /ful/ and .füll/ful,- and -pool/pu!/ and-- pull/pol /. Already mentioned is the bull, Ubool11 contrast, although in this

instance 11booln is not a word in English.

The problem for the Luo chUd is to learn which

English words have the /u/ phoneme and which the / 0 /. ln this

he will be helped by English spelling to sorne extent, for~ is g enerally pronounced /u/. On the other hand.. book is not /buk/

as he might deduce from its spelling but /buk/. In this respect

English contrasta with Luo, which does not provide elues as to pro-

nunciation by means of spelling, so that ruyo, for example, is spelt

exactly the same whether it is pronounced with /u/ or / U /, as has

been demonstrated.

Wi th re gard to the phonemes /u/ and / U /, then

the Swahili child's difficulties are more serious than those of the Luo

child. He has to learn that these phonemes, unlike their equivalents -71- in his own language, are not interchangeable in English. He bas to learn to separate them, and know where each is appropriate.

As neither Luo nor Swahili bas equivalents of any of the English central vowels /a:/, as in bird, /a/ as in the final sound in father, and /A / as in eup, it would correctly be surmised tb.at this would be an area with special problems for

Luo and Swahili children learning English. The problem here does not involve learning to use in new situations a set of sounds already existing in the native language, but learning to hear sounds that are entirely new and then being able to produce them accurately.

With regard to the difficulty of hearing phonemes tb.at do not exist in one1s language, it is easily enough identified..

For example• a Japanese learning a language with /l /, such as

Italian, faces the same problem. Japanese does not have the phoneme / l /, but it bas /r/, and a Japanese learning the word Mussolini for example substitutes the /r/ of his language for the ! l /in this word, rendering it as /musorini/. 16 As

Lado states:

When a phoneme in the foreign language does not exist in the native language the stud:ent will tend to substitute the native phoneme that seems nearest within the whole structure of his native language.17

16. Fosco Maraini, Meeting with Japan, (The Viking Press, 1959) p. 167 17. Op. cit. Robert Lado, p. 27. -72-

It would seem, then, that the new phoneme for which the nearest native equivalent is substituted by the learner is never heard by him as it is~ that is, he does not register it in the same way that a person the phonemic structure of whose language it is a part registers it. To the Japanese the Italian

/ L 1 does not sound like 1 l / at ali, but like the Japanese /r/.

In rouch the same way do Luo and Swahili children solve the problems posed by the English phonemes 1 ~: f:. / ;::) / and / 1\ /.

For example, when an Englishman pronounces the word earth

/ o: e / air/ea/ and cup/k 1\ p/ the Luo of Swahili child hears no difference between the three vowels. So similar do they sound to him, in fact, that he uses one phoneme as a substitute for ali three. Since in the Luc and Swahili sound schemas /a / seems closest to the English central vowels, it is this phoneme that is substituted for the English / è): /, / a / and 1 1\ /. The three words mentioned above the refere be come / q; e 1. / e.ct / and

/kap/ in the speech of Luc and Swahili children.

It will be recalled that Luc and Swahili children sub­ stitute their own /a/ for the English /~/as in~; These children therefore use the same vowel in such different words as cap, eup, carp and kerb. This shows the degree of differentiation of vowels which Luo and Swahili children have to make before the risk of being misunderstood, or not understood at all, can be eliminated. -73-

In order that the first steps towards the solution

of this problem may be taken, the Luo child will have to be

prepared to make lip, tongue and jaw movements whlch. because

of the attitude of Luo people to auch mouth movements, he never

rnakes :i:n.apeaking his own language. According to Jones,. the mis-

take of using /a/ instead of/~: 1 can be corrected..

by taking care not to open the mouth too wide; in fact it is often advisable to practise the sound 1e; / with the teeth kept actually in contact. 18

It will be necessary to teach these children to make the / è): /

sound, further forward and a little higher than the position at

whlch they make the /a/ of their own language,. for L/a/ is made

at the back of the throat with the mouth. open but not too wide.

The Swahili /a/ is made in a similar manner.

The difference between /~:/and /-al. as Jones

explains, 19 is one of length: /

if ali the members of the /a/ group of phonemes are taken

into consideration. It is therefore interesting to note that the

Luo or Swahili child• once he has learnt to produce the basic

sound, finds it easier to pronounce words with. the a-sound,. and

its variants, than tho se with the at sound. It is the length of the

latter that constitutes the pronunciatiœ. problem. In general

Luo and Swahili vowels are short. Long vowels in other languages

18. Daniel Jones, An Outline of English Phonetics, {Heffer,. 1918), p. 90. 19. Ibid, Daniel Jones, p. 92. -74- the refore constitute pronunciation problems for Luo and Swahili children learning those languages.

As for the English / 1\ / as in eup / kAp/, the Luo or Swahili child1s substitution of his own /a/ for it is not as confusing to his listeners as is his use of the same vowel in the place of the English / Ç): /. Jones himself observed that

11there is ••• no objection to using a more a-like sound,. as long as the 1fronting1 is not overdone. n20

However, a certain degree of precision would appear to be necessary,. for,_ as has already been pointed out, Luo and

Swahili children pronounce ep , eup and carp all as /k.ap/ using their own all-purpose /a/ vowel. Whereas some of the time 1hese children will mean eup when they say /k.ap/,. it may be assumed that at other times they will mean other words. As Bloomfield states:

When we try to speak a foreign language or dialect,. we are likely to replace its phonemes by the most familiar phonemes of our own language or dialect. Sometimes our native phoneme and the foreign one overlap,. so that part of the ti.me our reproduction is correct,. but part of the time it falls outside the range of the foreign sound. 21

Luo and Swahili children therefore have to be taught to modify the /a/ of their respective languages to sound more like the English //\ /. As Jones states: 19. Ibid. Daniel Jones,. p. 92. 20. ~ cit. Leonard Bloomfield, p. 81. 21. Op cit. Daniel Jones, p. 88. -75-

It is a good plan to learn / ç:,: / before learning /!\ / ••• / A/••• is intermediate between /o: / and / o. /, and in practi c• it is found that / 1\ / may often be taught by directing the learner to make a sound about half-way between / ~: / and /a/.22

The langth of English vowels, as has been mentioned, is another area of difficulty for Luo and Swahili children. In neither Luo nor Swahili doœ vowellength change meanings.

This is also true of English, but in English there are certain words which are liable ta be confused with certain others if they are not pronounced with the correct vowellength. For exam.ple, bark /ba:k/ has a long vowel be cause / r / is silent.

In Lowland Scotti.sh English /r/ is sounded and the length of the vowel correspondingly reduced. The author had a Lowland

Scottish teacher in secondary school and noticed this peculiarity.

Luo and Swahili children someti.mes do not make the vowel in bark long enough. Whenever that happens the result is

/bak/ which is close enough to buck /b 1\k/ ta be mistaken for it.

In the same wa y ~ /ka:t/ may sound like cut/k !'\ t/, and hart /ha:t/ like hut/h f\ t/.

Other /a/ sounds are long without the aid of a silent

/ r /. For example, the /a/ in father /'fa:~~/ is long. It has a definite ah sound. It is the length of /a/ in dance which further

22. op. cit. Daniel Jones, p. l03. -76- distinguishes it from dunce. whereas in Luo or Swahili speech the two words sound alike because /'a/ is not made long enough.

Luo and Swahili children learning English have to be taught to recognize the importance of vowellength in spoken English.

Diphthongs

Diphthongs do not as a rule constitute pronunciation di:f:ficulties for Luo and Swahili children. in spite of the absence of equivalents in Luo and Swahili of a number of English diphthongs.

Since most of the se diphthongs are made up of vowel phonemes that exist. or have phonemes wbich sound very much like them~ in both Luo and Swahili• the comparative ease with which Luo and Swahili children learn to produce them is easily accounted for. There are, however. a few exceptions and these have to be taught carefullyJ but before discussing them,. there are a few points that have to be cla:rified.

It will be noticed from the chart that both Luo and

Swahili have two sets of /e/ and two of /o/. The unmarked /e/ in both cases is similar to the English / e/ as in get / get/ and head /hed/. The luo word lemo/lemo/ 1to lickt and the Swahili

1 word meza /meza/ . a table 1 have the same /e/.

However. the second / e/~ in each case marked with

L or S to show that it is a special Luo or Swahili sound is not - 77 - quite like the first.. The L/e/ phoneme and the S/e/ phoneme are a little more central than the first / e/ and the tangue is a little tenser wh en they are pronounced than when / e/ is pronounced.

L/ e/ like S/ e/, resembles the initial sound of the vowel in the word late /leit/.

The re is a vital difference between L/ e/ and S/ e/.

Whereas in Swahili S/ e/ is non-distinctive and can be used interchangeably with the unmarked / ef, in Luo L/ e/ is distinctive and has to be used correctly if confusion is to be avoided. For example, the Luo word quoted above, lemo, means 1to lick' if it is pronounced /lemo/ and •to pray' if it is pronounced

/leme/, where / e/ is L/ e/.

The relevance of the foregoing to English diphthongs as approached by Swahili and Luo children becomes clear when it is noted that the English diphthong 1ei/ has special düficulties for these children. For example, most Airican children can hear the difference between let/let/ and late/leit/. In pronouncing the word late, however, the Luo or Swahili child does not give the diphthong its full value but uses his own L/ e/ or S/ ef as the case may be. The use of L/ e/ or S/ e/ instead of the full/ ei/ diphthong does not cloud meaning. However, in the interests of accurate pronunciation, these children should be taught to add

1if to their own L/ e/ or S/ e/ in or der to get the correct pron­ unciation of words like late, eight, bake and others like them. - 78 -

The second phonerne of which both Luo and Swahili have two sets is /f>/, as can be seen on the chart. The rernarks already made with regard to / e/ apply to / o/ as well. In both Luo and Swahili, / ::::> / is close enough to the English / ;:) / to be sirnilar to it. The word hall, the Luo word ~~ 1dazzle', and the

Swahili word shona ali have the sarne / .J / vowel phonerne.

Differing from / ;:) / as L/ e/ and S/ e/ düfer from / ef are the phonernes transcribed as L/ o/ and S/ of. Both are sirnilar phonet ically and resernble the initial sound of the vowel in the word note. Ag ain the phonerne L/ of is used to contrast rninirnally with

/ ::::> /, just as L/ e/ contrasta rninirnally with / e/.

For exarnple, the Luo word chor rneans 'dazzle• when pronounced /tf::;, r/, and 'push' when pronounced with L/ o/.

Sirnilarly, /l-:Jï/ rneans •corne down•, 1 alight1 whereas /lor/ with

L/ o/ me ans t shut'.

In Swahili, on the other hand, a word auch as moto/moto/ with S/o/ rneans tfire' even if/ ;:::, / substituted for S/o/. Sirnilarly kiboko/ kiboko/ rneans • a cane' or 'a hippopotarnus' whether S/ of or

/ ::::> /1 s used. ln short, S/o/ does not contrast rninirnally with

/ 0 /, that is, neither phonerne is used as the sole rneans of dis­ ting uishing rn eaning s.

In the sarne way as the English diphthong / ei/, the siphthong - 79 -

/ou/ is never given its full sound value by Luo children. For

exarnple, instead of saying boat /bout/ they usually say /bot/ using

their own L/ o/. This is not so far from the correct forrn as to be

rnistaken for a different word. Once more, in the interests of

accuracy of pronunciation, these children should be taught to rnodify

their L/ o/ by adding a fleeting / u/ so that they would get the correct

pronunciation of such words as float., rote, shouhler, road and the like.

The Swahili children on the other hand use, not their own

S/ of in such words as boat/bout/ and goat/gout/ • but the ordinary

J :> 1 phonerne,. pronouncing the above words as /b;::, tf and 1g :::::> tj.

It is clear how confusing this can be, for these words, so pronounced

are actually bought and got. It is necessary therefore to teach Swahili

children to use their own S/ of instead of / .;:::, /, and from there

advance to the addition of / u/ in order to produce the English diphthong

fou/.

English spelling does not always give the foreign learner

sorne indication as to which vowels are diphthongs and which are not.

There is,. for instance,. no way of knowing that the vowel in host /houst/ is a diphthong whereas the vowel in lost/1 :::> st/ is not. These words

are spelt exactly the sarne except for the initial consonant, but a foreigner would have to hear them pronounced before realizing that there was a difference in their pronunciation. Since Luo and Swahili spelling leads Luo and Swahili children to expect words to be pronounced the way they are written,. they are perhaps in greater need of guidance in English - 80 - pronunciation than children who do not have this initial expectation.

French children, for example, would not expect words to be pron- ounced the way they are written, since French is not spelt phonetically.

The general caution given by Jones in his discussion of the

English diphthong /ou/ is one which Luo and Swahili children would do well to heed. He states that:

Foreign learners should avoid overdoing the diphthongal character of ou or replacing it by forma like au, all of which may be heard in London and other dialects. It is better to use the continental o: than these exaggerated forma. 23

And it is reassuring to know that "Speakers of Received Pronunciation

24 do not •••• all use the same variety of diphthong. u

The remaining English diphthongs may be commented upon briefly since they present no special problems for Luo and Swahili children. In general it may be said that there is one major difference between Luo and Swahili diphthongs and English ones. Jones states that:

For the purpose of practical language teaching it is convenient to regard a diphthong as a succession of two vowels, in spite of the fact that, strictly speaking, it is a gliding sound. 25

24. Op. cit. Daniel Jones, p. 99

25. Op. cit. Charles C. Fries, p. 14 -81-

Luo and Swahili d.iphthongs., on the other hand., are really two

separately pronounced vowels, and not ngliding sounds11 at all.

This point will be discussed more fully in connection with stress

and intonation_. but a few examples will suffice to give an indication

of this characteristic of Luo and Swahili diphthongs.

Whereas in English a succession of two vowels may

or may not constitute a diphthong_. in Luo or Swahili auch an

arrangement of vowels never constitutes a diphthong. In ·English,

for example, the vowels ~in shoulder constitute a diphthong,

whereas ~in~ do not. In Swahili. on the other hand, two vowels

sim.ilarly juxtaposed have the sound value of each of the vowels

concerned, that is, each vowel of the pair is pronounced separately.

In fact, the more separately such vowels are pronounced_. the better

is the pronunciation thought to be, 11 better" in the sense of being

accurate and clear. The same applies to a succession of two vowels

in Luo.

For example, the Luo ward aim/aim/, 1 should I caver',

is prononnced with what may be mistaken for a diphthong. There is a juncture between /a/ and /i/, thus produclng a staccato utterance.

However, there is no objection to giving the ward a smooth and un­

broken delivery. In English, on the other hand, aim/eim/ is pronounced with a diphthong. In Swahili the vowels ae in aeleze.

•that he may explain', are pronounced separately.

Such words as pool and peel are not pronounced with a diphthong but with /u/ and /i/ respectively,that is -82-

/pul/ and /pU/. Now in Luo and Swahili even double vowels are pronounced separately~ in contrast to English where this is never done. Luo never uses double vowels, except when these have sOine syntactical significance. For example, the Luo word /;:)'ol/ pronounced with L/ o/ means 'he or she is tired•. It is not really a ward in the sense that ~ is a word in English, for the first / o/ is the sign of the third persan singular, and /ol/ means 'is tired•.

The double vowels in the word therefore differ in function from the doubel vowels in pool. The double vowels in ool are pronounced with a junctu.re to mark off the first / o/ from the second.

In Swahili, on the other hand.. double vowels may be used with or without any syntactical significance. In either case the vowels are pronounced separately. For example, kitambaa is a single independent word which means a 'handkerchief'. Other similar words are -koo, 'throat1, kutii •to obey1 and mguu,- 1 foot*. At the same time there are certain ward-complexes in Swahili which have double vowels for syntactical reasons in the same manner as ool in Luo. For example, tu.tembee, 1let us walk1 or *let us go for a walkf, has a second l_e/ in the terminal position because this is a sign of the subjunctive. Broken down into its various constitu.ents, tu.tembee becomes : tu., •we1 or •us1, tembea, 'walk',

'go for a walk', and~ , as has been stated, is a sign of the sub- junctive. It is clear, then, that tu.tembee is not really a word in the same sense that peel is a word in Eng lish. -83-

In general, then, Luo has a succession of two vowels

mainly in cases where this has syntactical connotations. ln any

case, even such words as achiel/atj:I: el/ and ria/ riO./ meaning,

tone' and 1widowhoodf respecti.vely, have two vowels in juxtaposition but are pronounced, not with 11a gliding sound", that is, a diphthong, but with each vowel separate and distinct from the others. ln Swahili, too, two vowels may be next to one another with or without any syntactical significance, and in neither case do the vowels form true diphthongs in the English sense.

It can therefore be concluded from the foregoing that

Luo and Swahili children will show a marked tendency to pronounce both parts of English diphthongs with equal emphasis, transferring their own language habits to the English situation. They have to be taught that only one part of an English diphthong need be stressed, and that they should aim at a much smoother delivery than they normally give diphthongs in tb.eir own languages.

With this accomplished, it would be a relatively easy step for these children to master the pronunciation of the few

English words which have a succession of more tb.an tvf> vowels.

For example, choir/kwatè / has three such vowels. Such a word wou.ld be difficult for Luo and Swahili children to pronounce, not because there are no words with three vowels in succession in their own languages, but because of the absence in their languages of the vowel phoneme / à /. -84-

English words with three vowels, following one another do not as a rule present as many difficulties of pronuncia­ tion,. for Luo and Swahili children as do those with only two.

Progress from the second group to the first can therefore be expected to be relatively smooth.

Stress, Intonation and Pitch

lt hardly need be pointed out that mastery of the sound system of a language is by itself not enough to make the learner competent in that language, although it is an essential first step in that direction. A persan learning a new language has to know, in addition to the sound value of each of the phonemes of that language,. what happens to those phonemes in a piece of continuous utterance.

For example, it is commendable if a foreigner learning

English discovers that the word ~e is pronounced /hi/. However, this knowledge is not enough, for the word has that pronunciation only in thee:nphatic form_. or when it is uttered in isolation. ln continuous speech at a moderate speed /hi/ becomes /hl/ and at a very fast speed /hi/ gets reduced to /1/. Both /hl/ and /1/ are said to be weak forms of /hi/.

It is essenti.al, then, that a learner know something about stress, intonation and pitch in the language he is learnirg. ln fact, the more th.orough the grounding he receives in these -85-

aspects of language, the greater is his mastery and control of

the spoken form of that language likely to be.

Luo and Swahili children learning English bave as

their objective the same aim by which the Spanish-speaking students under the direction of Charles Fries and his colleagues at the

University of Michigan are inspired, that is,

the recognition of distinctive sounds as they occur in the actual speech of native speakers of English and the accurate production of these sounds so that they can be recognized by native English speakers.Z6

To which one might add,. ail educated English-speaking people, whether English, American or African.

The Luo child generally finds stress and intonation in

English speech very difficult to get correct. He brings with him to the English situation knowledge of stress and intonation in his own language, and as these are entirely different from stress and intonation in English, they consti'b.lte a handicap which requires great patience as weil as considerable skill on the part of the teacher to overcome.

lt is Robert Lado who states:

When learning a foreign language we tend to transfer our entire native language system in the process ••• our phonemes and their variants, our stress and rhythm patterns, our transitions, our intonation patterns and their interaction with other phonemes. 27

26. Op. cit. Robert Lado, p. 13.

27. Op. cit. Charles C. Frles, p. 15. -86- lt is some of these features of Luo that the Luo child has to

learn not to transfer to English when learning that language.

One characteristic of Luo phonemes has already

been discussed in connection with Luo diphthongs. lt was pointed

out that Luo vowels not separated by a consonant are always

pronounced singly1 and that the same applies to double vowels of any kind. There is none of the elision or traUing off in Luo which

characterise English utterances. ln fact, clarity in Luo is

assessed in terms of the degree of precision and distinctness with which every single phoneme in an utterance is delivered. ln English, on the other band, clarity is assessed in terms of

the precision with which stressed syllables are stressed, and unstressed ones given their proper value.

lt will now be seen tba.t stress in English is a very

real problem for the Luo child learni~ English. What is true

of vowels not separated by a consonant and double vowels ir4 Luo,

is true also of a single vowel forming part of a syllable. For

example, neno/nen:J /, •to see1, has two syllables, each of which

is pronounced distinctly. The same word in a piece of contl.nuous speech may lœe the last vowel/:) /, but only if the initial phoneme of the word following is a vowel. For example, aneno adita,. 'I see a basket', -87-

when said at the normal speaking speed becomes /anen adlta/.

Elision, then,. is also a difficult feature of English

which the Luo child learning the language has to overcome. In

English,. unlike in Luo, elision is determined.. not so rouch by the nature of the phoneme coming before or after the phoneme or phonemes to be elided_. as by the speed and emphasis with which the particular word or utterance is spoken. For example, the

Luo child has to realise the degree to which such a word as and /

f~ncl/ 1 gets reduced when such a phrase as spick and span is uttered at the speed at which most people normally speak. In such a phrase /::~!..nd/ contracta to a mere residual/n/, a form greatly popularized by American films and jazz in such shapes as /Urock•n•rollu for rock and roll, Uwash'n'weartt for wash and wear, meaning clothes that drip dry and need no ironing, and nspick•n•span11 for spick and span.

Even normal contractions such as isn1t for is not, shouldn't for should not and~ for I am are not normally found in the speech of Luo children. Not only do they say it is where the Englishman would normally say- it1s; even when they come across --it•s, don•t and they're in their reading, these children use the long form of these expressions if they read them aloud, rendering them as~, do not and they have.

There is another aspect of stress in English which generally intimidates the Luo child. Although Luo uses stress to sorne -88- extent to distinguish meaning between words written exactly the same, greater use is made of this method in English than in Luo. In

English, for example, perfect is an adjective when pronounced

/'parfekt11 with the stress on the first syllable, and a verb when pronounced /pa:1fekt/ with the stress on the second.

In Luo, on the other band. words whose meaning can be changed according to the position of the stress are not many.

The author knows only five such words. And even in these, stress is not' the sole factor determining meaning, for other factors such as pitch and tone are used in conj:.1nction with stress to determine the meaning oi such words.

For example, kendo/kendo/ with L/o/• means1'to marry11 if pronounced with the stress on the first syllable, whose vowel is also made quite long, and the whole word uttered with a high pitch dropping slightly on the second syllable. The same word means ta fireplace' or ta stove' when pronounced with a low pitch falling even lower on the second sylla He. It means 'a gain' when the first syllable is made short with a rising pitch on the second syllable.

Where a complete utterance is involved, for example,

He gave me this key, the Luo child is usually at a losa as to the kind oi stress he needs for the particular meaning he wishes to put across.. He usually settles for the ordinary stress pattern which merely affirma the statement. He says 11 'hi geiv tmi ~t.s 1ki/ -89-

This stress pattern is quite appropriate for this kind of affirmation. However, the Luo child generally does not know and is unable, on his own, to see that other stress patterns are needed for other kinds of affirmation. For example, if he wi shes to emphasize that it was He, and not .Joan or the first eleven soccer team then the Luo ochild needs to know that he should pronounce

He with extra stress.

However, if he wishes to emphasise the fact that he did not obtain the key by sorne underhand means, if he wishes to declare his innocence, so to speak, then he needs to know that it is on the word gave that the strongest stress should fall. In short~ the Luo ,;child has to be taught a vital function of stress and intonation in English, that is. to attract the attention of the listener to the most significant part of what the speaker has in mind.

Swahili has a slightly better organized stress system than Luo has.. Therefore the Swahili child would not be ex.pected to have as many difficulties with English stress as are encountered by his less fortunate Luo compatriot. For example. stress is not used to distinguish meanings in Swahili, as it is to sorne extent in

Luo. But even more important, Swahili words have stresses falling in a certain definite pattern. for example, on the first syllable if -90- if the word is made up of two syllables, on the second if it is made up of three, and on the penultimate syllable if the word is made up of more tban three syllables. A word such as ku.tetemeka, •to tremble• , is made up of five s ")illa.bles • but the stress falls only on the last syllable but one. A word such as knfika, 1to arrive• , is stressed on the second syllable, and a monosyllabic word auch as ~ •not is stressed strongly.

The conclusion can therefore be drawn that

Swahili is a very stress-conscious language, but does not use it as a means of distinguishing meaning as is extensively done in

English. Thi s stress -consclousness on the part of Swahili­ speaking people is, however, not altogether to their advantage, for they are likely to assume, when they come to learn English, that English words have the same stress patterns as do Swahili ones. Therefore they have to be taught that ·in a piece of continuous speech in English they have to stress certain syllables or words and mumble others so as to leave no doubt in the minds of those listening to them to which syllables or words they wish to draw particular attention. For in Swahili, tone rather than stress is used for this purpose, a word such as kufik.a being strongly stressed on the second syllable regardless of the meaning it may have in a piece of continuous speech. - 91 -

In general, the advice which Fries gives his Spanish-

speaking students learning English applies equally well to Luo and

Swahili children in a similar situation. He writes that the Spanish

student

will have to learn a new mode of speech production••• avoid his usual staccato utterance, an utterance which gives full value to each syllable, and he will adopt a more undulating rhythm giving sorne syllables more prominence and reducing others. 28

Swahili children in particular will have to learn to avoid

transferring to English the high-pitched endings which they generally

give utterances in their own language. .High-pitched endings are

appropriate for questions in English, but not for any other kind of

utterance. Luo children, for their part, will have to learn to speak

English with the flat tones with which Luo is usually spoken, but with the appropriate intonation and undulating rhythm.

Correct intonation and pitch can only be achieved through

careful attention to the speech of people whose native language is

English, and by constant practice. Jones is certainly right to

emphasize the importance of ear-training for foreigners learning

a new language. 29

Mastery of aU the aspects of English sounds considered in this chapter should provide Luo and Swahili children learning English

28. See Footnote 27. 29. Op. cit. Daniel Jones, p. 21. -92- with the basic skills which will ensure for them an increasingly bigh degree of competence in spoken English. On this firm foundation they can coniidentl.y build a surer knowledge of the grammatical structure of the language, and let the enlarge­ ment of their vocabulary follow on naturally.

They will have to keep in mind the fact that they are learning 11a new mode oi speech production# n30 and that it will be neither possible nor desirable for them 11to learn the precise articulation of each separate sound and then remember to reproduce that articulation in each instance of its occurrence. u31

All these skills will have to assume the nature of habits so that they get to be used automatically. For an unnecessary degree of consciousnessof one1s speech may lead to an affected, elocutionary style whose shallow preciosity may offend.

30. See footnote 27. 31 .. Op. cit. Charles C. Fries, p. 15. -93-

CHAPTERIV

PROBLEMS OF GRAMMATICAL STRUCTURE AND VOCABULA.RY

Grammatical structure in this dissertation is under- stood to mean 11the systematic formal deviees used in a language to convey certain meanings and relationshipsn. 1 The view of traditional grammarians that grammar comprises a set of rules. permanent and final, which govern the way in which the words of a language are put together to form meaningfu.l utterances~ or parts of such utterancest is considered only for the effect it has had on the teaching of English to Luo and Swahili children.

As has already been observed~ Luo and Swahili children setting out to learn English already speak their respective native languages. They are nine or ten years old at this time, or even o~der if they started going to schoollate in life. As lAlo and Swahili each has its own set of 11 systematic formal devicesn, that is~ its own

grammatical structure~ it is to be expected that learning the ndevicestl used in English wœld present many difficulties, since these deviees are totally different from Luo and Swahili ones, with the exception of

one or two which will be discussed presently.

1. Robert Lado, Linguistics Across Cultures. {Ann Arbor: The University of Michigan Press, 1957) p. 52. -94-

The aspects of grammatical structure which are of

crucial importance in learning English are word order • in:flection,

correlation of forms and function words. Already considered are

stress, intonation and junctu.re.

Word Order in Luo and English

The Luo-speaking child is fortunate in one respect

when he comes to learn English.. His language has in many respects a word order similar to word order in English, and word order has

the same significance in the two languages.

In English. when a sentence such as he saw th•~ watchman

is presented, any English-speaking person understands at once that

it was he who did the seeing, and not the watchman. Another

rather similar example, Freda sent them a letter,. means that it is

Freda who did the sending of the letter and they to whom i.t was sent.

The meaning of both statements is implicit in the order of the words,. with the person performing the action coming first in the sentence.

In Luo the person,. thing or agent that performs the action,. that is, the subject, also comes first in a sentence, or at any rate precedes the person thing or agent that the actinn is performed upon,. that is, the object. For example, the first of the two sentences above -95- would be ne oneno jarito in Luo.. ln this version ~ is the sign of the past tense. the first;: in ~ is the sign of the third person singular and is the subject of the sentence, neno is the finite verb of the infinitive neno,.- pronounced with L/ol rather than with / / 1 as in the first case. It means •to see' and jarito is •a watchman 1 the object of the sentence ..

The second sentence above becomes Freda ne ooronigi barua in Luo .. ln this case the acting agent. the subject, precedes the sign of the past tense because she is specifically, named. If he in the first sentence were Otieno, a Luo name for a boy born at night, the word order of the sentence would be Otieno ne oneno Jarito. ln the second sentence ooronigi is really a number of words put together. The first;: is again the sign of the third person singular, the subject of the sentence, ~ means tsend1, ni means tto• and gi means 'them•, the object of the sentence, and barua, a word annexed from Swahili, means •a letter•, the indirect object of the sentence.

However, it is only in these sentences in which the subject comes first, followed by the verb and object that Luo word order is similar to the pattern in English. Luo, for example,. does not use the passive voice, except for a form which Luo students who have learnt English and Swahili are attempting to transfer to Luo from these languages, with little noticeable success so far.. A sen- tence such as The chief was succeeded by his son can be translated -96- into Luo only in a form which means 'the chief1s son succeeded the chief•. But a statement sucb as The bouse was built quickly becomes ot ne ogerore piyo, that is, 'the bouse built itself quickly1•

Therefore it is not simply because of word order that some word arrangements in English bave no similar patterns in Luo; auch arrangements are rendered in another way in Luo, as, for example, the use of reflexive verbs to express certain forms of the passive voice.

There is another instance in whicb Englisb word order differa significantly from the Luo pattern. For example, a guest bouse means 'a bouse for guests•, since guest modifies bouse.

The fact that guestis the modifier is signalled by its position in the expression, which is before bouse, the word it modifies. In Englisb modifiera usually precede the words they modify. If, on the other band, the word order is changed to a bouse guest, the meaning of the phrase would be changed to •a guest of the bouse•, because bouse would be the modifier and guest the word modified.

This kind of construction is not used in Luo.. Instead of relying on word order alone for determining the meaning of a phrase auch as the above, Luo, like French, uses the "of" construction, the noun preceding "ofl1 always being modified by the noun following it.

For example, the Luo version of a guest bouse is otmar welo, that is, •a bouse for guests1, whereas a bouse guest is wendo marot - 97 -

in Luo, that is 'a guest of the house1 • In Luo phrases of this nature mar may be used to mean noun-of, for example,

1hous e of'. Another form of ot mar welo is od welo. Both mean 1 a house for guests•. In this example, both otmar and od mean 'a house of1 or 1 a house for', and mar, which means 'for' or 'of' becomes superfluous when od is used.

It will be seen, then, that word order in Luo differs from word order in English in many respects, the only area of similarity being in the subject-verb-object construc tion. The Luo child learning English therefore has düficulty with such patterns as she was seen off by her father, tending to use the form her father saw her off instead. This may on occasion be a better form stylisti­ cally, but it representa the inability of the Luo child to recognize the passive form because there is no such form in his own language.

Whether educated Luo people will succeed in popularizing the passing form they have adopted from Swahili and Engllsh, it is still too early to tell. However, at present the use of this new form is confined to the educated, and not all of them are enthusiastic about it, the author among them.

The elimination of these düficulties will enable the Luo child to use more natural patterns of word order in English than he uses when he transfera the patterns of his own language to English. -98-

Word Order in Swahili

Swahili is to sorne extent an agglutinative language,

that is, a number of paradigms are frequently joined together to

form an utterance or a series of utterances. For example,

ilipoonekana is not a wo rd in the sense that ~ or he at is a word

in English. Broken down into its various paradigms, ilipoonekana,

'when it was seen', although literally it means 'when it saw itself1,

becomes i, 'it1, li is the sign of the past tense, po, 1when1,

onekana, 'was seen•.

It will be seen that this composite word is buUt up with

its various parts following one another in a definite order. The

third person singular, .!! is the subject of the sentence and there­

fore cornes first, followed by the tense sign and the finite verb.

The same order applies in a complete sentence. For example, nUimkimbiza means l made him run1• The subject of the sentence, ni, that is, •rr, cornes firs~ followed by ~ the sign of the past tense, rn which in this case means 'him1 and then cornes the finite verb.

As bas already been pointed uut, word order in English

sentences usually follows the subject-verb-object pattern. The

Swahili child, in spite of the fact that the verb in Swahili usually

cornes after the object, does not bave much difficulty understanding word order in English. It is true that in a statement such as niliposikia kelele nUiamka. 'when I heard a noise l woke upt, the -99-

word order could be changed to niliamka niliposikia kelele,

•1 woke up when I heard a noise•. However1 the emphasis would

not be the same, for in Swahili1 as in English1 the most important

idea or concept always comes first1 except when certain effects.

such as suspense. are desired.

Other patterns of word order in English have equivalents

in Swahili. The paœive form. for example1 is highly developed in Swahili. A statement such as the tree will be eut down tomorrow

becomes mti utakatwa kesho in Swahili where mti means tthe tree•, 1 - utakatwa:. 'will be eut down•. and kesho •tomorrow1• It is wa in

utakatwa which is the sign of the passive form. A rather similar

statement. the eup is broken becomes kikombe kimevunjdika in in Swahili. Kikombe means 'eup'•- ki in kimevunjika is the prefix for verbs used with words which begin with~or ch,. ~ is the sign

of the perfect tense, k:uvinja, the infinitive from which the finite

verb vunjika derives1 means •to break•,. and the reflexive nature

of the verb is indicated by its -~ ending. The statement as a whole

does not mean just v.bat its English equivalent means; it means also

that 'the eup has broken itself•. and that no one can 1herefore be held

responsible for the damage.

In this particular instance, then, the passive form in

Swahili conveys slightly more information than does the English

passive form1 and thereforethe Swahili child needs to be taught -100- the boundaries within which the passive form operates in English.

A major area of difference between word order in

English and Swahili is that which includes such phrases as a watch pocket and a pocket watch, a roof garden and a garden roof.

This kind of word order already discussed to reveal the Luo child1s approach to it. confuses the Swahili child. For example. a watch pocket means 1a pocket for a watch*, since. as has been seen, watch. the nounc which precedes the only other noun in the phrase. modifies pocket. A reversa! of the word order produces a pocket watch. ln this case pocket is the modifier. and the phrase means

1 ta watch for the pocket 1 as distinct from a wrist watch or sorne other kind of watch.

ln Swahili, as in English, such a phrase as a watch pocket could not possibly be confused with another like a pocket watch. The first phrase in Swahili is mfuko wa saa, and the second saa ya mfuko. ln the first phrase mfuko means 1pocket1,

~'of or for• is in agreement with mfuko, and is used with ail words which begin with~ in the singular. And saa means •a watch' or 1a clock1• ln the second phrase ya •of or for•, agrees with saa because~ belongs to a group of words which remain unchanged in singular as in plural.

lt can be seen, then, that the Swahili child recognizes at once from word order, and also from the form of~ or for, -101- which noun is the modifier and which the head of the construction.

For him the difficulty of an English phrase such as a pocket watch stems from the absence of a word such as -of or for,-- word order alone being deemed suificient to indicate meaning. He has to be taught that, whereas in Swahili the noun which precedes

~or ya, or other forms of 2!... or for, is modified by the noun which follows it, in English the nounc which cornes first in the above phrase, and others like it, modifies the noun,:. which follows it.

An example of complete reliance on ward order to

convey meaning in Swahili could be used to illustrate for the

Swahili child the function of word arder in English. If a con-

struction similar to tho se discussed above had nouns belonging

to the same class of words in Swahili, it follows that the word for

of or for would be the same for both nouns. For example~,

•a watch• or tclock1 and meza, ta table•, belong to the same class, which includes ali words which remain unchanged in singular

as in plural. If the Swahili child wanted to say a table clock in

Swahili he would have to say saa ya meza. In this case, ya could agree with both ~and~ since these nouns belong to the same

class. In actual fact it agrees with ~· It is therefore word arder alone that indicate s which noun is the modifier, and which the head

of the construction. The word order reversed would produce -102- rneza ya saa. 'a clock table'• and again it is the order of the words which indicates the functions of the two nouns. Word order per­ forrns a sirnilar function in English.

Questions in Luo and Swahili

Questions in English usually take the forrn of Is John an alderman? That is. the auxi.liary cornes first. followed by the noun or pronoun, and then the rest of the sentence. The whole is spoken with rising pitch from a low ls to a high -man. Any

English-speaking person wou.ld recognize the sentence as a question if so uttered.

The sarne sentence with the word order altered slightly produces the staternent .John is an alderrnan. This rnatter-of-fact staternent can be changed into a question by sirnply uttering it with a rlsing pitch. There are therefore several deviees which signal the question forrn in English: word order, with the verb, the auxiliary or sorne deviee such as ~ corning first. and the rest of the staternent following in the usual order; and rising pitch.

ln Luo and Swahili, questions do not have a special forrn of the kind that English has in such a sentence as ls John an alderman? Instead of using word order and pitch as English does in the above e:xample, Luo and Swahili use pitch alone to indicate -103- a question, that is, a rising pitch of voice at the end of a sentence with the arder of words remaining the same as in an ordinary stateme nt. For example, ls John an alderman? would be John ni mwaziri? in Swahili, and --John en jabtua? in Luo, both statements being uttered with rising pitch. If no rising pitch is used the se would be ordinary statements of fact in bath languages, that is, John ni mwaziri, and John en jabura,

'John is an alderman1•

As one way of changing statements into questions in English involves the simple use of pitch, ward arder remaining the same as in an ordinary statement, Luo and Swahili children

can be expected to find this method easy to understand, since a

similar method is used for the same purpose in their respective languages. These children actually use rising pitch far more frequently than the native :English-speaking persan does in his everyday speech. It seems, therefore, that although they recognize ls John an alderman? as a question, Luo and Swahili children need to be encouraged to use the form more often than they actually do.

A characteristic of English verbs involved in questions to which Luo and Swahili children have to pay careful attention is the disappearance of the final~ in the third persan singular.

For example, he cames he re every day becomes does he come ••• -104- and not does he come, as Luo and Swahili children expect.

They have to realize that in this case does makes super.fl.uous the final s in the third person singular verb.

lnflection in Luo and English

The major features of language considered under in.fl.ection are the sign of the plural and the sign of the past ten se.

The Luo child learning English proceeds from a language in which plurals are formed in highly irregular ways to one in which there are definite patterns followed in forming plural s. In his own language. the absence of defini te rules governing the formation of plurals makes this perhaps the most difficult aspect of an otherwise simple language. It may be that it is not so much the absence of firm 11Ües as the great multi­ plicity of them which makes for the excessive irregularity of plural forms in Luo~ so that most words and their plurals have more or less to be learnt singly even by the Luo child. Ad.ults, too, occasionally use inappropriate plural forms~ especially those of rare words.

For example, the plural of dero, •a granary', is deche. On the otherhand, the plural of lero, 'a head wound'~ is not 11leche11 as one would expect because of its similarity to dero, but lere. Although leche is a word in Luo meaning -105-

'veins•, the possibility of confusion between ~e, 'veinst, and 11lechen, 'head wounds', could easily be avoided by giving the words different intonation, for intonation alone distinguishes a number of such words in the language. It is therefore not just because of the possibility of confusion that the plural of lero is not n1eche11 •

Again the plural of dhiang', ta cow', is dhok, whereas the plural of tiang1, 'sugar cane', is tiange. It seems, therefore, that the spelling and pronunciation of two words may be the same, except for one phoneme, without their plurals being simi lar. On the other hand, certain words share the above features and yet have similar plurals except for one phoneme.

For exampl~, the plural of got, •a hillf or 'mountain' • is gode, and the plural of rot, 'a narrow passage in a fence•, is rode.

Sometimes the plural in Luo bears no recognizable resemblance to the word whose plural it is. It is easy,. for example, to see that wan, 'we' or fust is in some way derived from~~ 111 or 'me1• It is not so easy to see the connection between udi, 1houses1, and ot, •a house' .. or between mon,

'woment, and dhako, ta woman•. These are ali plural forms peculiar to the language and have to be learnt individually.

All these examples are cited to show the great variety of ways in which plurals are formed in Luo. When the Luo child comes to learn English, therefore, his initial difficulty will be to -106- realize that a terminal .!. is enough to change most English words into plural, for exarnple, book and books, and bouse and houses. A few words such as leaf and tomato will present problems, but these are easily solved by changing! of leaf into!.. and adding ~ thus producing leaves, and by adding

~ to tomato to produce tomatoes.

Other plural forms which the Luo cbild has to be taught are those which are formed by vowel mutation, as in man and men and !ouse and lice, and the form which remains unchanged in singular as in plural, for example,. sheep. This last the Luo child understands readily enough since the re is a similar form in his own language. For example, nyamburko, ta motor vehicle',. a surprisingly genuine native word for sozre thing which is foreign and relatively new, remains unchanged in singular or plural.

The Past Tense in Luo

The Luo child moves from a relatively simple system of past tense formation in his own language to a rather complicated one in English. He learns, for example, that most English verbs adda terminal -d or -ed in order to form the past tense. Therefore he manages the farm weil becomes he managed the farm weil. -107-

However, there are many English verbs which do not form their past tense by the simple addition of a terminal

-d or -ed. For example, the past tense of see, in Southern

English at any rate, is not 11 seedu but~ and the pas t tense of catch is caught. Other verbs which have irregular past tense forms are buy, weep, and write. There are many such verbs in English. All these the Luo chUd has to learn singly.

It is the simpliclty of the past tense form in Luo, in contrast to the irregularity of the past tense form of many

English verbs, which makes the Luo child find it difficult to learn past tense forms in English. For ex.ample, ali that is needed to change a Luo statement into the past tense is the insertion of~ before the subject-verb. No verb by itself has a past tense form in Luo, in contrast to English in which it is

only the verb which can be changed into the past tense. For

exar:tl.ple, she sings beautifully is ower mamit in Luo, and she

sang beautifully is ne ower mamit.

However, the speaker has to mak.e a decision regard- ing the time limit covered by ~· In English, for example, the past tense applies to any action or event which happened any time between a moment ago and countless years ago. In Luo,

on. the other hand, what took place a moment or a few hours ago ha s to be referred to witb.- aye or nende, rather than ne. -108-

For ex.ample, if she sang well only a few moments ago the

Luo statement would be nende ower mamit. lft however• it is something that happened yesterday or the day before yesterday, appropriate words would have to be used~ in this case nyoro and nyocba respectively, for example, nyoro ower mamit and nyocba ower mamit. It is only to the period earlier than the day before yesterday that ~the true past tense form in Ùl.o, applies.

It will be seen, then• that Luo past tense forma are quite simple, and that ail that is necessary is an awareness of the time at which an action took place, that is, how long ago it happened, and the appropriate word referring to that particular time can then be used. The Luo child bas to be taught the time factor in English past tense forma, as it is more extensive in

English than in Luo. Whereas the Englishman may say I saw him just now,the Luo person does not use the past tense but the perfect tense. The above statement would therefore be nende anene sani in Luo. -109-

In.fl.ection in Swahili and English

Of the languages with which the author is at !east to sorne extent acquaintedt Swahili is the only one in which substantives are grouped into classes according to the first letter or the first two letters of their plural forms. It follows, therefore. that the plural of a substantive has to be known before it can be assigned to its proper class, and it is only if its class is known that ali the words used in agreement with it can be given their appropriate prefixes.

For example, the plural of mezat ta table• is meza.

This word therefore belongs to the fifth class, which includes ali the words which remain the same in singular as in plural.

In meza yangu kubwa, 'my big table•, - angu is the rootf>r possession in the first person singular and y- is the prefix for the fifth class. Therefore yangu indicates possession of an article belonging to this class of words. And kubwa, 'big'• needs no prefix when used with the fifth class.

If in the same phrase eup is substituted for table, it becomes kikombe changu kikubwa. ln this case~and ki- are the prefixes for words used with members of the third class. which is the group to which kikombe belongs.. The plurals of the two phrases are meza zanœ;t kubwa and vikombe vyangu vikubwa respective!y. -110-

The characteristics of the prefixes used with different classes of substantives are revealed by plural forms. For example,. yangu changes to zangu to indicate plural in the first phrase.. whereas~and ki- in the second change to vy- and !!:._ in the plural, so that changu kikubwa becomes vyangu vikubwa. Kikombe itself becomes vikombe in the plural.

As there are eight classes of words altogether, it means that there are eight different ways of forming plurals in Swahili. However,. the rules governing plural formation are quite simple and are therefore easy to follow. For example, ali words beginning with :::_- or ~ , ii they are the names of warm-blooded animals,. form their plurals by changing :::.:.._ to~·

Ali such words belong to the first class. For example, the plural of mwalimu,. •a teachert, is waalimu. Similarly, the plural of mtoto, •a child', is watoto. On the other hand, the plural of mwaka, •a year•, is miaka because this word belouga to the second class and is not the name of a warm-blooded animal. Similarly, the plural of mti, •a tree•, is miti. Words of the second class form their plurals by changing m-or mw- to mi-.

The initial difficulty which a Swahili child learnirg

English experiences in relation to plurals is that in English the sign of the plural,. !!_ occupies the terminal position in Engllsh, -III- in cont rast to the Swahili practice of putting the sign of the plural in the initial position. However, he becomes accustomed to the terminal s ül.irly quickly and then has to solve a rather more difficult problem# namely, how to know which English words have the -sor -es ending in the plural, and which have different plural forms.

It helps the Swahili child enormously if he is made to realize that the=!... or~ ending applies to most English words in the plural. His attempts at fincling plural forms for such words as loaf, sky and ~can then be properly clirected if he is taught that !_and :X change to ~and 2:_ in the plural, so that the plural form of the above words are loaves, skies and wives. His attention can then be drawn to the seven words which form their plurals by vowel mutation, these being man, woman~ goose. foot• tooth, lause and mouse, their plurals being

~ women, gee se, feet, teeth, lice and mice.

The last group of words to which the attention of the Swahili child can then be drawn includes those words which remain unchanged in plural as in singular. For example• sheep and deer may be used in the singular or plural. The Swahili child, already acquainted with such wo rds in his own language, usually finds this plural form relatively easy to understand.

However, he has to learn them individually, for, as in Swahili, -112-

there is no other way of recognizing the words as belonging

to this group.

The Swahili chUd has to learn that in English

the sign of the plural in one word may change a whole statement

into plural~ whereas in Swahili each word in a statement has to

have the appropriate sign of the plural prefixed to it in order

to change the statement into plural. For example, the plural of my big eup is my big cups, whereas in Swahili kikombe changu kikubwa becomes vikombe vyangu vikubwa, in the plural.

The Past Tense in Swahili

In much the same wa y as the Luo child,. the Swahili

child finds English past tense forms, with the exception of the terminal..=!_or -ed, relatively difficult to learn. He has fi.rst of ali to grasp the fact that in English it is only verbs and auxiliaries that have past tense forms, in contrast to Swahili which, like Luo.. uses a prefix to change a whole statement into the past tense. For example the past tense of knock is knocked and that of is is was. In Swahili verbs as such do not have past tense forma, but if preceded by -~ in a statement have -113- a past tm. se meaning.

For example. in alifika mapema~ •he arrived early'• !: means lhe't ,!!_ is the sign of the past tœ. se.

~ is the finite verb of the infinitive kufika• 1to arrivet. and mapema means •early'.

As in the case of the Luo past tense:. a decision has to be made regarding the ti.me to which ~ refers. In

Swahili any action or event which took place yesterday or earlier may be referred to in the past tense. However, some qualification may be needed in order to indicate more speci­ fically whether the event took place yesterday, and not. for example, last year. The sign of the past tense is sti.llli, but jana. 'yesterday1, may be added, or wiki jana , •last week' if it is necessary to be more specifie.

ln cases where the action described happened today or some hours ago -~ is used instead of -li-. This is the equivalent of the perfect tense in English. Therefore the

Swahili child has to learn that where -me- is used in his own languaget the past tense is used in English. For example, one may say in English I saw her just now, whereas in Swahili one would have to say nimemwona sasa hivi, ... ~- indicating that this is something that has just happened. 114 -

Other procedures are the same for the Swahili child as for the Luo child. For example, he has to learn the

English verbs which have irregular past tense forms. To him catch and caught appear as unrelated as they do to the Luo child.

Such verbs, and all others that have irregular past tense forma, he has to learn individually.

Function Words in Luo and Swahili

These are words used as grammatical signais.

For example, who in English signala a question if it cornes at the beginning of a sentence, in isolation or if uttered in a high-pitched volee. The Luo equivalent is ng1 a/lJa/ or ng 1 ano/r:r~nof, and the

Swahili equivalent is nani. Therefore who is that? is mano ng'a? in Swahili.

Luo and Swahili function words, th en clos ely parallel

English ones, and so the latter do not in general create learning problems for Luo and Swahili children. It will be noticed in the above example that word order in the Luo and Swahili versions is different from the pattern in the English statement. Sometimes ng1ano and nani come first in a sentence, and at other times they, like who, come at different places, depending on the -115- meaning of the sentence.

The absence of grammatical gender in Luo and Swahili, as in English, has made it possible to dispense with a cumbersome load of words and deviees which would other- wise be needed to signal this grammatical distinction. In French and Russian, for example, grammatical gender exists and makes necessary the use of function words and other deviees appropriate for each gender, as the author discovered, much to his frustration, when he attempted to make himself acquainted wi th those languages.

A great source of d.ifficulty for Luo and Swahili children learning English is the article in English. In Luo or

Swahili there is no article, whether the definite article, the indefinite or .zero article. English uses ail three kinds. The article, as John Hughes states, 11 is a ward whose almost sole purpose is to convey information about structure, and which hardly conveys content-iniormation at all. 112 These children•s problem is their inability to tell whe re the defini te article~ should be usedt where the indefinite -a~ and where zero article. For example, they understand readily enough that a is used when introducing something, and the when referm ce is

2. John P. Hughes, The Science of Language, (Random House, 1962), p. 162 -116-

again made to what has just been introduced. Fo r example.

there was a butterfiy in my room and the butterfiy was brightly

coloured. This, however, is elementary usage of the definite

and the indefinite articles, and Luo and Swahili children experience no difficulty in understanding these forms ..

There are other functions of a and the. For

example, ! used with a substantive performs the function of

selecting one member of the same class of substantives to

represent the whole class of substantives in question. 1 saw a man on the roof, for example, provides no information regard.ing

the identify of the indivi dual man se en, but the fa ct that a is used with ~ indicates that it was one of a large number of

beings generally referred to as~ and not a squirrel or a~

Therefore ! d.istinguishes one member of a group without identify­

ing hlm beyond that point. It is 11indefiniteu.

It is the. the definite article, that identifies particular ind.ividuals or types from the whole group to which these

belong. For example, 1 am going to the airport is quite clear,

since the airport can only refer to one particular airport which everyone knows or has heard about. ln he reads the bible every

evening,. the bible refers, not to the copy in the chape! or the one his grandmother gave hlm. but to a copy of a book of scriptures known as the bible. That is, the bible, like the President, is not -117- one of many~ but the only one of its kind~ or the only one in mind. It is this sense that the is "definite".

:"·Zero article is used with objects or names which are unique. For example, Montreal Harbour need not have an article before it~ that is, it has zero article because it is unique in the sense that there is no other Montreal Harbour anywhere else. To say a Montreal Harbour would therefore be inappropriate, as it would imply that there were more than one Montreal Harbours in existence.

AU these articles cause di:fficulties to Luo and Swahili children. For example, they use the definite article where they should use zero article, or the indefinite article where the definite article would be more appropriate. They

see no difference between people don•t like it and the people don1t like it. Zero article in the above statement generalizes it~ whereas the definite article makes it more particular. If the Leader of the Opposition said the people don1t like it he would clearly be referring to the electorate. If. on the other hand, a salesman said people dontt like it he would be referring to people in general.

These distinctions have to be made clear to

Luo and Swahili chUdren, for the absence of the article in their languages makes them find it difficult to understand its function in English. -118-

Other General Points

The remaining aspects of grammatical structure, namely, intonation, stress and pause, have already been discussed.

Other general features of English which Luo and Swahili children learning English have difficult in understanding may now receive brief mention. These features raise learning problems for these children mainly because they differ in some way or other from the corresponding features in Luo and Swahili.

Possession in English is indicated by ~ 1 for example, John's book. In Luo possession is indicated in various highly irregular ways. For example, od Onyango means •the house of Onyango', Onyango being the Luo name for a boy born between sunrise and noon. As has been pointed out, od is a con­ traction of otmar, •house of•. On the other hand, Onyango's

~ is wuor Onyango in Luo. In this case, there is no special possession form for 11 shoe o:f11 , wuor, •shoe•, being sufficient to convey this meaning.

The Luo child has little difficulty mastering the

•s form in English, although there are variations of it which he may find a little difficult, for example the mayor of Nairobi•s hat.

His attention has, however, to be directed to the arder of words in the English possession form. This is different from the -119- arrangement in his own language. In English ~ after a word or phrase indicates that possession refers to the word or phrase, as the œ. se ma y be.

The Swahili child in general experiences dif­ ficulties similar to those encountered by the Luo child with regard to possession in English. Swahili, lik.e Frene~ uses the 11 of" construction by the prefix for the equivalent of of is determined by the class of the word to which it refera. For example, John1 s book is kitabu cha John, which literally is

1 the book of John•~ and Sam1 s box is sanduku la Sam, 1the box of Sam•. Whereas in Engllsh it is possible to say the man who came here yesterday's box, in Swahili such a statement would be rendered llthe box of the man who came here yesterday1, that is. sanduku la mtu ambaye alikuja hapa 3ana.

The teaching of English grammar to Luo and

Swahili children needs to be re-examined, in view of the length of time spent on it and the amount of actuallearning which takes place. If, after being taught English grammar three times a week for twelve years any Luo or Swahili child still needs to do some remedia! work on the subject at college, then the methods used in teaching grammar to these children have to be reviewed and over­ hauled. -120-

John P. Hughes, discussing the state of

:English language teaching in the United States. writes:

Traditionallanguage teaching went on the (unproved) assumption that if the student were brought to comprehend the principle governing the selection of forms, andmade to commit the forms themselves to memory, he could• for any occasion, selecih.e proper forms in accordance with the principle. 3

Some acquaintance with Luo and Swahili grammar on the part of the teacher is a definite asset when he cornes to teach Englishl grammar to Luo and Swahili children. It will then be easy for him to understand why Luo and Swahili children invariably put English adjectives, both predicative and attributive, after the nouns they qualify. This is the position adjectives occupy in Luo and Swahili. For example, instead of saying a hard chair they say a chair hard. The same phrase in Luo is kom matek, and kiti kigumu in Swahili. The attributive adjectives matek and kigumu come after their respective nouns in Luo and Swahili. The assumption by Luo and Swahili children that :English attributive adjectives also come after the nouns they qualify takes time to eradicate.

However, there are instances when Luo and Swahili children would be correct to put the attributive adjective after, rather than before, the noun it qualifies. For example, in a zebra slain -121- the adjective cornes after the noun it qualifies.. The meaning of a slain zebra and a zebra slain is not quite the same.

Ali these are points which bave to be taken into consideration when teaching English grammar to l.llo and

Swahili children..

Vocabulary

The fact that Luo and Swahili are completely un­ related to English has both advantages and disadvantages. One advantage is that there are no words in Luo and English.. or

Swahili and English, which are spelt the same but bave different meanings in the two languages. Such words are a source of confusion among, for example, Englishmen learning French.

These people encounter many French words, such as large and libraire, which resemble English words, in these instances large and library, but which in actual fact mean different thlngs., wide and bookseller in the above examples.

However, the disadvantages of this absence of re­ lationship far outnumber the advantages.. lt means, for example, that Luo and Swahifi chlldren have to learn every single word in

English individually, or at least as many as they find adequate for their needs, for every word in English is a new word to these -122- children. When a French child learning English comes across auch wo rds as education'" communication and table he knows what they mean if he has met simllar words in his own language.

Luo and Swahili children on the other hand. having no assistance from their respective languages in this respect, have to proceed more labouriously with such help as they may get from their teachers, dictionaries and glossaries.

Recent times have seen the beginning of word borrow­ ing in both directions. Naturally.. Luo and Swahili have borrowed more heavUy from Engllsh than Engllsh has done from either, for the coming of the British brought with it many concepts, objecta and processes, for which there were no native words in

Luo or Swahili, and so English names for them were borrowed by these two languages.

The re are, for example, many English words which have now been fi.rm.ly incorporated into Luo. Instances are skul, buk, osiptal, ingi, petroli, the months of the year, and many others. These have been adopted from school, book, hospital,

~nd petrol respectively. English, too, has annexed some Luo words, at least in the context of East Africa. For example, ngege, a Luo name, for a type of fish found in Lake Victoria, is part of

Engllsh terminology in :East Africa. -123-

There are borrowings. too,. from English to

Swahili and vice versa. Sometimes words are borrowed from

English even when perfectly adequate native words exist in

Swahili.. For example, ~ is the Swahili word for school.

However, skuli,. adapted from the English word, has practically displaced the native word in everyday speech. English, for its part, has annexedisuch Swahili words as safari and, in the context of East Afrlca, shauri,. •matter• or 1affair• jembe and panga, which are garden implementa, and a few others.

However, the process of borrowing has nqt had time to go very far. Therefore Luo and Swahili children still have more difficulty with vocabulary in English tb.an French children in a similar situation would be expected to have, since English and French have such a large number of words in common..

One tendency that Luo and Swahili children have to guard against is their fondness for long,. polysyllabic words in the mista.ken belief tb.at these impress. They are more likely to say. for example,. the conflagration is extinguished tb.an the simpler and more natural the fire is out. This is asnnch a matter of style as it is of vocabularly. There may be occasions when long words are appropriate, as when one is ridiculing a speakerts bombastic oratory. In this case the effect would be comic. lt was this same effect that Sir Winston Churchill -124- achieved in the House of Coxmnons when he apologi.zed for an error and termed it a ttterminological inexactitude". However.

Luo and Swahili children have to be taught the virtu.es of sim- plicity, in terms of natu.ralness. clarity and grace.

Luo and Swahili children come up against the problem. not only of having to learn English words individually, but also of learning what these words mean in certain contextst even when they know the dl.ctionary definitions of the words. For example, the se children may know all the meanings of do listed in the dlctionaryt and yet fail to understand such statements as we shall do away with the law, and he has to mak.e do witb. only half his allowance. In a similar way, difficulties crop up over the different meanings of running in a running motor, a running feud and a candidate running for election. lt is Fries who states:

a language is much more than words themselves ••• communication demands the use of severa! types of meaning in addition to that which can be expressed by dictionary definitions. 4

The main cause of difficulties of this kind is the way in which most words in Luo and Swahili are restricted to single meanings. ln English, on the other hand• most words can be used to have a wide variety of meanings in different contexts.

It is this greater flexibility of words in English that Luo and

Swahili children have to learn to cope with.

4. Charles C. Fries, The Teaching and Learning of English as a Foreign Language, (The University of Michigan Press, 1945) p. 39. -125-

For example~ if the Luo child cornes across a statement auch as KOkoth chiemo, it means to him that Okoth, a Luo name for a boy born when it is raining, is eating. That is to say, the act of eating is adequately described by chiemo.

This word has only that meaning. However, its English equivalent, eats, has severa! meanings, for example, an acid eats into metal, and a bully and a braggart can be made to ~his words by a stronger man.

A Swahili word such as kupiga, 'to beat• or 1 strike•,. has a wider range of meanings in certain contexts than many words in Luo. For example, kupiga hodi means 1to knock. at a do or•, kupiga kelele means •to make a noise• or •to shout•, kupiga magoti is •to kneel' and kupiga chapa is •to print•. There are many other meanings which this word can have depending on the context in which it appears. In general.however, most Swahili words, like most Luo, ones, are restricted to single meanings.

Another area of difference between Luo and Swahili on the one han~ and English on the other, becomes clear when the name s • for sorne parts of the body are considered. For example, ~ is mkono in Swahili. Now the Swahili word for~ is also mkono, so that if a

Swahili chUd says mkono wangu he may mean either 'my handt or -126-

'my arm•. In Swahili raise your hand is inua mkono wako. reference being made to the whole arm~ rather than just to the hand.

Similarly,~is mguu in Swahili, and~s also mguu. Therefore mguu mkubwa may mean ta big foott or ta big legt. As for the digits of the hands and fee~ they are counted from the thumb to the little finger, the thumb being the first or main~ finger and the little finger the last.. The same applies to the digits of the foot; the big toe is the first or main digit and the rest follow in their normal order.

The Swahili child~ like the Luo child in whose language a system similar to the Swahili system is used, has to understand that hand and arm refer to different parts of mkono_. and that the first finger in English is not the thumb, as it is in Swahili, but what is called 11the second finger11 in Swahili.

For Luo and Swahili children, as, indeed, for all foreign children who have no direct contact with native English speakers~ the best procedure would seem to be wide reading consolidated and reinforced by attempts to use new words, or new meanings for words already known, as accurately as possible. However, they have to bear in mind the fact that meanings change, and that new ones are constantly being added to those already in use. For example, nice, which in the time of Shakespeare meant precise, exact, now means pleasant. -127-

Also idiomatic expressions, which may have been fresh and meaningful when first used. become hackneyed in the course of time and degenerate into clichés. For example, it no longer l".l:e ans much to say that something happened at the psychological moment.or that the plan to commit a robbery was nipped in the bu.d by an alert policeman.

Luo and Swahili children in particular have to be cautioned against auch expressions, for to them each expression they come across in their reading is new and therefore fresh. They have no way of gauging the currency of such phrases.

Other shortcomings are revealed by the extensive misuse of such words as beside and beside sometimes and sometime, ~and really and many ethers. In addition, these children sometimes use strange words such as irregardless instead of the proper words• regardless in this case. All these are problems of. vocabulary~ since they are the result of an inadequate understand.ing of what certain words mean.

There are several ways of. forming verbs from nouns and adjectives in English. For example, the prefix en- may be used, as in enrage, or the prefix~ and the suffix ~may be used, as in embolden, or the suffix -fy may be used, as in beautify.

Luo and Swahili children have to be cautioned against allowing their -128-

enthusiasm for this method of verb formation to go beyond boundst

for it often leads them to form verbs rather indiscriminately.

The author has heard such words as Hmisplanttt and Hdirtify''• and himself used them many years ago. The two above are supposed to mean 1to plant in the wrong place1 and 1to make dirty•. These may be examples of commendable attempts at word-building, but the conclusion is inescapable that they are used in the erroneous belief that the wo rds actua.lly exist.

Luo and Swahili children have :to work almost as l:ard on vocabulary as on the sounds of E;nglish. Both these aspects of

English cause these children a great variety of difficulties. -129-

CHAPTER V

THE CULTURAL FACTOR lN LANGUAGE

Cultures, Robert Lado states, are Ustructured systems of patterned behaviourtt, l and as the relationship between one such 11 systemtt and the language in which it is expressed is peculiar to the particular language community in which both operate, it follows that no two language com- munities have exactly the same way of looking at reality.

For language is a way of looking at reality. As ·· Ho~jar observes:

The languages of human beings do not so much determine the perceptual and other faculties of their speakers vis-à-vis experience as they influence and direct these faculties into prescribed channels. Intercultural com­ munication, however wide the difference between cultures may be, is not impossible. It is simply more or less difficult, depending on the degree of difference between the cultures concerned. Z

For the would-be bilingual, the implications of this singularity of cognition are manifold. H9wever, his major task is to learn to understand and appreciate the cultural milieu in which the native speakers of the language he is learning operate. He will

1. Robert Lado, Linguistics Across Cultures (The University of Michigan Press, 1957) p. 111 z. Harry Hoijer ed. Language in Culture (The University of Chicago Press, 1954) p. 94. -130-

fi.nd that mastering the sound structure# granunar,. syntax and

vocabulary of the language, though essential for accurate speech production,. will not always enable him to make sta.tements that

native speakers of the language understand. He will find the

dictionary very useful, but he will also need ta know the situations

and circumstances which govem the use of certain expressions or

words or formsof address.

It is the application of this knowledge that is behind

the exhortation of teachers to their puplls to "think in EnglishH,

or in Frene~ if these are the foreign languages being learnt,.

rather than think in the native language and then translate the

thoughts into the new language.

Differentiation in English, Luo and Swahili

It is Roger Brown who states:

ln any comparison of two languages (A and B) there seem always to be ranges of experience more differentiated in A than in Band other ranges more differentiated in B than in A. 3

For example; dog is a generic term for aU members of the canine

family of animais. However• particular breeds of dogs have,. in

English, special names by which they are further identified, when-

ever this may be necessary. Labrador, Alsatian.,. bull terrier and

3. Roger Brown, Words and Things, (The Free Press, 1958) p. 234 -131- poodle are ali names for special breeds of dogs. ln Luo there is no such differentiation of dogs; ali dogs are just dogs~ that is,

guogi. lt means therefore that if a Luo child or adult visited a

Dog Show~ the great variety of dogs he would see there would ali be classified by him under the generic term guogi. An Englishman• on the other hand• would classify them by breed~ with each breed appropriately named.

The fact that Luo has no name s for the different breeds of dogs does not mean that Luo children or adults cannot describe these dogs when they see them. It does mean that what

Luo has for dogs are phrases describing them~ rather than special names as English has. For example, the English bulldog would be identified by its pug nose. and the poodle by its characteristic coat and facial contours. If the Luo child saw a borzoi he would probably cali it guok matiendene boyo, that is, •the dog with the long legs•.

ln Swahili, too• mbwa is the generic term for ali dogs, and there are no special names for the different breeds.

There is the same tendency in Swahili, as in Luo, to use phrases to describe the various breeds of. dogs .. singling out their most outstanding features.•

Differentiation obviously serves a special purpose.

The Englishman, or any other person who uses a similar kind of -132- differentiation$ finds it essent:l.al to classify the different breeds of dogs and name each breed because he has to know the various functions for which different breeds can be trained. For examplet he has to know which breed makes a good police dog, which a sheep-dog and which a loyal family pet. No one would. for example. attempt to train a poodle to be a blind man's guide orto take part in dog races. It does not have the necessary attribu.tes. This particular deficiency distinguishes the poodle from the Labrador, which may be trained to be a blind man's guide, and the greyhound. which may be trained as a racing dog. These distinctions are very real to the Englishman, and the various names ior breeds of dogs have the signiiicance of marking those distinctions.

In Luo and Swahili, on the other hand, such distinctions are not as important as they are in English. In any case, none of the breeds of dogs so far menti.oned was known in Kenya before the advent of the British. However, the absence of Luo or Swahili names for these breeds is not to be attributed solely to the relative recency of their introduction into the country. for there are many objecta and substances which are equally new, but which have acquired Luo or Swahili names. The main explanation for this absence of native names for the different breeds of dogs is that the differences and characteristics which distinguish breed from breed are oi relatively little significance to Luo and Swahili people .. -133-

It is clear~ then, that in the matter of dog types tb.ere is a greater degree of differentiation in English than in either Luo or Swahili. Therefore the problem. for Luo and

Swahili children learning English is to learn to differentiate more specifically tb.an they do in their respective languages the various breeds of dogs. For example, the Swahili child has to learn to say Alsatian whenever appropriate~ instead of thinking it sufficient to say a large dog as he does in his own language. A statement such as he was bitten by an Alsatian translated into Swahili becomes aliumwa na mbwa mkubwa, where mbwa mkubwa, ta large dog', is a rather inadequate description of an Alsatian.

At the same time, there are certain differentiations which Luo and Swahili make. but which do not exist in English.

Here the problem for Luo and Swahili children learning English is to guard against transferring such differentiations to English.

For example, in English certain of the animals collectively known as cattle are also known as cows. A young cow is known as a calf, and an older calf as a heifer. This would appear to be ali the differentiation of cows that exista in English.

In Luo, on the other hand, not only are there equivalents for ~~~and heifer, dhiang' nyaroya and dwasi respectively; -134-

there are names also for different kinds of cows, for examplet those that have just calved, those that give mUk steadily, those that have weaned their young, those that no longer calve and those that are sterile. In this instance, then, the importance which Luo people attach to milk bas resulted in a high degree of differentiation of cows according to their milk-producing and calving potential. No doubt milk is at least as important to Englishmen as it is to the Luo, bu.t it forms the basis on which cows are differentiated in Luo, whereas in English it does not.

Therefore when speaking English the Luo child or adult has to be aware that this particular kl.nd of differentiation

of cows does not exist in English. English farmers may classify cows, too, bu.t they do this b)'fbreed, such as Angus, rather than by capaclty in producing milk. The Luo person should therefore not expect that there will be ·English equivalents of the

Luo names for different kinds of cows. He bas to get accustomed to a smaller degree of differentiation in English in this instance than exista in his own language.

Learning English Culture

Differences between Luo and English, and Swahili and English can be seen on a much wider front than that indicated -135- by the degree of differentiation of the sa.me phenomenon or experience in each of the three languages. Thesedifferences ca.n be seen, for exa.mple, in family rela.tionships, in the concept~s and ideas that are dominant in ea.ch language com- munity, and in concept of time and space, to mention only a few.

If all these factor are trea.ted in English in a. way that differa significantly from the way in which they are treated in Luo or Swahili, then it would be poor tactics on the part of

Luo and Swahili children to lea.rn these items piecemeal. For these differences exemplify the differences between Luo and

Swahili cultures on the one ha.nd, and English culture on the other.

It seems, then, tha.t the best procedure for these chUdren to follow would be to lea.rn as mucl:l· about English culture as possible, for it is only when they have this knowledge that they would be able to use the English language as English people use it.

Edward Sapir states that

language does not exist apa.rt from culture, tha.t is, from the socially inherited assemblage of practices and beliefs that determines the texture of our lives. 4

The forms of greeting in Luo, Swahili and English illustrate the influence of different cultures on the speakers of these languages.

In English, for example, one may say to a friend, 11 Hullo. How are you ?" The usual answer to this is "Very well. thank yoo.. 11 The

4. Edward Sapir, Language .. {Harcout, Brace & World, 1921) p. 207. -136- pointis that this particular answer is given even if the person giving it is not reaily weil. The Englishman who says 11How are you?l* is not really ask.ing for information on the state of health of the person to whom he says it. If this person were to respond with a catalogue of the various ailments with which he is affl.icted it would be considered bad m.anners.

In Luo,. on the other band, Ingima? whlch literaily means tare you weil?' is really meant to elicit the information asked for. The person so addressed therefore answers Angima.

I am weil. if he is,. or else meticulously ennwnerates ail his current allments together wi1h their symptoms and the amount of pain and discomfort they are causing him. His friend then ofiers his sympathy and. in his turn. describes his own aches and pains, or those of someone he knows. AU this is news. and whatever is news is carefully attended to. Not to impart this news when asked, orto do so perfunctorily when it is appropriate togo into detail. is regarded as very bad manners. And as ceremony and ritual characterise much of the life of the Luo. a breach of manners of this k.ind offends deeply.

The problem for the Luo child learning English is obvious. He has to refrain from boring his English friend with accounts of his state of health when his friend asks him "How are you ?JI This does not rule out the few occasions when this -137- form ceases to be a mere formula and becom•3S the equivalent of the Luo ~gima? For example, an Englishman visiting a friend in hospital will say "How are you ?'1 and expects to be given an account of the friend1 s real state of health.

In Swahili Hujambo? is the equivalent of the

English nHow are you ?n although literally it means •is it that you are not well'. The answer to this is always Sijambo kidogo, •I am :':lOt weil a little•, that is, I am a little unwell. The curious fact about this answer is that its accepted meaning is •1 am quite well1, although literally it means the opposite. Swahili people, like the English, answer 111 am quite we. 111 to the question, 11How are you11 • However, Swahili people arrive to this answer by stating its opposite.

It should be noted that Hujam bo? is but one of a number of forma that make up the set of questions and responses of greeting in Swahili. The next question in the series is Uhali gani? Which means 1what is the general state of things ?t This question has to be asked, and at its prope r place in the series.

The answer to it is always Njema, that is, 'well1, •excellent•.

There is in Swahili, as in Luo to some extent, a marked ceremoniousness about greeting that prolongs the pre­ liminaries considerably. One has to go tb.rough the motions. It is -138- considered bad form to plunge into conversation, even with a friend, without first uttering these formulae of greeting. And it is unspeakably bad manners merely to nod or grunt as English people do when they meet friends on the street, at the club or at a pub.

It is after the forms of greeting have been gone through that the Swahili person, if he really wishes to know how his friend is, will ask hlm aboutit speclfically. But he has to do this in a roundabout way, for example, by saying something like

Lo! Umenona! that is •Goodness, you've put on weight! • or

Nasikia hall haikuwa njema hivi karibuni, that is, •I understand things have not been going too welllately. • Even these oblique feelers may only be put out after the conversation bas gone on for

some time, that is. for five minutes or so. By English or North

Am.erican standards Swahili people must seem a decidedly leisurely people.

The problem for Swahili children learning English is like that of Luo children in a similar situation. They have to learn to suppress their rather effusive natures and adopt the

generally more reserved nature of English people when they greet their Erg lish friends. Swahili children, for instance. have to learn that the English response to 11How are you.,'f11 tb.at is, HQuite well, -139- thank. you11 is unlike its Swahili equivalent in that it is not taken to mean much. whereas the Swahili sijambo kidogo is part of the solemn ceremony of greeting.

What Luo and Swahili children have to remember is that in the culture of the English the forma of greeting are in the main mere formulae. whereas in Luo and Swahili they are not.

A form such as nHow do you do ?1 1 does not exist in either Luo or

Swahili because in these languages greeting is part of the ceremony and ritual that play such a significant part in the lives of Luo and Swahili people, and is therefore taken seriously.

As the best way Luo and Swahili children can learn English is to use as their madel the speech of educated native speakers of English, it follows that they have to learn what English people say and in what circumstances, if they are to make themselves understood. For the use of English translations of their own Luo or Swahili forma will not always be understood outside Luo and Swahili circles.

Family Relationships in English and Luo

Differences in family relationships and organization among English people on the one handt and Lu.o and Swahili people on the other. create learning problems for Luo and Swahili chUdren and interfere with the way they learn English. -140-

The family unit in England is normally considered to consist of a man, his wife and their children, if any.

Such people as the man's father, mothert brothers and sisters, as wèll as those of his wife, though they may be close to hlm emotionallyj are really not part of the family unit of which he is the head. They are his parents or relatives, that is, unhles, aunts, cousins, nephews, nieces, father-in-law, mother-in-law and the like.

The way these relatives are distinguished from one another shows the inflœ nee of English culture. For example, uncle is the brother of one's father or mothert and aunt the sister of one•s father or mother. On the other band, cousin is either the son or the daughter of one•s uncle or aunt. In other words, there is no great differenti.ation of cousin in Eglish.

Onels nephew is the son of one1s brother or sister, and one•s niece the d.aughter of one•s brother or sister.

As for the parents of one•s wife, they are barely brought into the family circle by the in-law deviee. Although the parents of one•s husband are also one•s in-laws, they are not as emotionally removed, from one's immediate family circle as are the parents of one's wife. The mother-in-law, that is, the mother of one1s wife, is by tradition a figure of fun and the butt of many jokes. -141- (J.Irt :z:: (- z ~ ~ ~ ~ 8 l 1 1 ---.: 1 1 ~ 1

J I t( ?: 1/j

0 :f _J r 0 j (!}z V.! r.

- I. (/)z 0 r­ e( ..1 w t dj sl :? 0 r u .J E u.< -141-

Luo family relationships contrast sharply with

:English ones. The family unit among Luo people does not consist only of a man, his wife and their children as it does in English society; it includes the man's parents, grandparents, if any, as weil as his brothers and their wives and children and his sisters if they are still unmarried. His wife's parents are also members of the family if they live close enough to make their influ­ enèe felt~. And the head of this unit is not the man but the oldest

~ in the group, that is, his father or grandfather. The older women in the group also exert considerable influence..

It will be seen, then, that when a Luo individual makes reference to his family he does not mean only those people the

Englishman has in mind when he refers to his family; the Luo individual refers to a large group of people whom the Englishman does not regard as part of his iamily.. If this Luo individual is learning English, he bas to be taught that familyhas different connotations in English and Luo.

Of greater significance than the connotations of family in English and Luo are the different ways in which the various relatives are identified and ~ .named. For example, uncle in Luo is not the brother of one1s father but the brother of one1s mother. One•s mother's cousin is also one1s uncle, cousin in the English sense and only if the cousin is a male. The brother of one•s father is onets

"little father11 , that is wuora matin, a legacy of an old Luo custom -143- whereby this individual married his brother•s widow as soon as she became one. If he was already married it did not matter, since polygamy was, and still is, part of Luo culture and was actively encouraged.

On the other band, aunt is not the sister of one•s mother but the sister of one•s father. One's fathers cousin, again in the English sense and if female, is also one•s aunt. The sister of one•s mother is one's Ulittle mothern, that is mama matin, a legacy of another old Luo custom whereby this woman, if her sister died, married her sister•s husband as soon as he became a widower. If she was already married a younger sister took her place.

As for one's cousins in the English sense, they are all one's brothers and sisters in Luo. It means. therefore. that if a Luo child says Ouma en owadwa, •Ouma is my brother•, he may mean that Ouma, a Luo name for a boy born upside down, is his brother or cousin. If it is necessary to be more spea: ific he may add maodwa to the statement,. that is, •of our house• meaning that Ouma is his brother, or madire, •of the side', meaning that Ouma is his cousin in the English sense. Similarly, Auma en nyamera may mean that Auma, a Luo name for a girl born upside down, is my sister or cousin in the English sense. Again, if it is necessary to be more specifie, the same words maodvta and -144- madire convey the necessary information.

It is a little strange that there is no native Luo word

for mother or my mother 1 although there is a word for mother­ hood. The word for my mother, marna,. was borrowed from

Swahili. Mothe r in Luo is minwa,. which means •our mothel.

There are variations of this for your mother, his mother and their mother, and other variations for the plurals of these.

One point likely to cause difficulties for Luo children learning

English is tb.at minwa,. •our mother1 means more tb.an mother does in English. Any woman the age of one•s mother or above may be addressed as minwa in Luo,. and she always takeitas a compliment when so addressed.

The high esteem in which the mother of one•s wife is held is indicative of the position which the parents of one's wife occupy in the wide family circle to which one belongs. While there is no specifie word in Luo for one•s father-in-law. this individual merely being described as 11 the father of the bride", that ist ~nnyako ,there is a specifie word for one's mother-in­ law. She is known as maro, a formidable personage whom one never addresses directly but only in oblique phrases and expres3ions, and whom one is forever placating with well-timed presents and other marks of deference.

If~~ one•s mother-in-law, is a venerated figure in the Luo scheme of things, her sons and daughters, tb.at is, -145- the brothers and sisters of onets wife, are not treated with rouch lesa respect. One 1s brother-in-law is a very important person. He is known as~ This is not the same term for the brother of one•s husband, this individual being known as yuoro. which is the same term for the sister of one1s wife.

In both cases yuoro is the person who, ii male and his brother dies~ marries his brother•s widow, and if female and her sister dies, marries her sister•s husband.

lt is sufficiently clear by now that Luo family relationships are vastly more complicated than English ones.

For the Luo child learning English, the important point to bear in mind is that English terms for the different relatives are not always applied to the people he would describe by roughly equivalent terms in his own language.

For example, my nephew is wuoda in Luo, that is» tmy son•. This is the same term for my wife•s son and mine. There is apt to be a considerable amount of confusion in the mind of the Luo child learning English when he uses the English word brother thinking that it includes the same people that its Luo equivalent can be applied to. These potential sources of confusion have to be clarified for the Luo child before he can begin to be more accurate in his use of

English words. As Robert Lado states:

Individuals tend to trans fer the fo rms and meanings of their native language and culture - both pro- -146-

ductively when attempting to speak the language and to act in the culture, and receptively when attempting to grasp and understand the language and culture as practised by natives. 5

FamUy Relationships in Swahili

Much of what has been stated concerning Luo family organization applies also to :family structure in Swahili society. For example, ndugu means 'brother• bu.t, like its

Luo equivalent, includes also the son of the brother of one's father and the son of the brother of one1s mother. Similarly, dada• •sister•, includes one's female cousin in the English sense. Whenever it is necessary to be more specifie, one usually adds marna moja, 1the same mother1,. meaning that the person referred to is a brother or sister.

ln the same way mjomba, •brother-in-law, • is of wider application than the English term. Unlike tts Luo equiva- lent, bu.t like the English term, mjomba in Swahili applies to both one•s wifets brother and one's husband1 s brother. However,

Swahili people, lik.e the Luo, usually do not wait until a marriage has taken place before they begin to apply the term mjomba to the appropriate people. Any boy or young man witb a sister of marriageable age may be addressed as mjomba in Swahili or~ in Luo. In addition, tbese terms may be used in addressing any

S. Ibid. Robert Lado, p. 2. -147-

any male who is younger than oneself• much as son is used

in American English.

The central fact regarding Luo and Swahili family

relationshlps as compared with English ones is that the terms used in Luo and Swahili do not always coinclde with what may at

first sight appear to be their English equivalents. A Swahili

person address:b;g another in Engllsh as brother-in-law will

be understood even if his style may be regarded as deplorable.

But the same individual addressing• say. an Englishman as

Brother-in-law without this relationship actually existing

between them will not only not be understood• he will be th.ought

to be rude• presumptuous and offensive. Similarly a Luo

child who. meaning to be polite. addresses an Englishwoman

he thinks is his mother•s age or above as mother willlikewise

be misunderstood.

It is clear• than, that Luo and Swahili children

learning English have to be made to realize :from the start that

there are major differences between Luo and Swahili terms

for the various members of their familiest and English ones

for the corresponding people in English society. This will

eliminate a potential source of confusion among Luo and Swahili

children learning English. -148-

New Ideas and Objecta

lt is to be expected that objecta and ideas wbich did not exist in the normal !ife of Luo and Swahili people until they were introduced by the British would not have native

Luo and Swahili names or words describing them. This, to a large extent, is actually what the situation is.

ln Luo, for example, there is no native word for a book. Luo people had no books untU after the arriva! of the

British. What bas happened is that both the English word and the Swahili word for this object have been borrowed and are now part of the Luo language. ln Luo ma en buk matin and ma en kitabu matinmean 'this is a small book•, and both buga and kitapa mean •my book•. The Swahili word kitabu, ta book' is a loan word from Arabie.

There are innumerable examples of borrowing from

English to make up deficiencies in Luo. English words such as lock, tray, glass, picture, ink• hospital, cupboard, store and many others are now part of the Luo language. They mean in

Luo what they mean in English,. but are generally spelt differently.

For example, the Luo version of store is sito.

However, there are other words which have been borrowed from English but which have lost their English meanings and have now acquired Luo meanings. For example, the Luo word memba, adopted from member, means, not a person who -149- belongs to an organization, but 1 a rich man•.. Similarly, diro, adopted from drawer, does not mean a number ci

sections of a dresser which can be pulled out and pushed in, but ta bookcase•. It is words of this kind tb.at the Luo child learning English has to take care to use correctly. He may, on discovering that they are borrowed from English, use the:n to mean what they mean in Luo thinking tb.at they mean the same thin fifJ in English.

Swahili, because of its partial derivation from

Arabie, a language with a long history of development and

growth, is in a more favourable position than Luo with regard to the existence of native words for certain com...rn.on objects and ideas. The existence in Swahili of Arabie words for most of these objecta and ideas has correspondingly lessened the need to borrow words from English. The result is that there is probably less risk of a Swahili child misusing an English word whose borrowed version in Swahili has a different meaning th.an of a Luo child falling into a similar error.

That its links with Arabie have made possible a great degree of development in Swahili is shown by the fact that Luo, which had no similar links with a highly developed language, has had to borrow many words from Swahili. For example, to read is kusoma in Swahili. In Luo it is somo, an adaptation of the Swahili word. Education, on the other hand, is elimu in -150- in Swahili, whereas in Luo it is again somo.

Many other words show how heavily Luo has borrowed from Swahili over the last seventy years. The Swahili word for a train is gari la mosti, literally •a vehicle of smoke•. The

Luo word for the same object is gare, adapted from the Swahili word. A spoon is kijiko in Swahili and ojiko in Luo, and a table is meza in Swahili and mesa in Luo.

However, there are certain words which the Swahili child learning English has to be care:fu.l about, since they do not always have in English the meaning that they have in Swahili.

For ex.ample, gari, •a vehicle1, in Swahili includes objects not normally referred to as vehicles in English, such as a peram- bulâ.tor. The Swahili child, when translating gari as 'vehicle1, has to be careful to cover only the range of objects grouped under vehicle in English.

It is an accurate observation that "Between two languages one cannot expect to find a one-to-one word relation- ship, since the symbols used in each language are arbitraryn. 6

In other wo rds, each language community uses words to describe its own things, ideas, beliefs and customs, and as these are not always the same for any two language communities, there is

6. Edwin T. Cornelius, Jr. Language Teaching, {Thomas Y. Crowell, 1953) p. 131. -151- bound to be a certain lack of correspondance between the meaning of words as used by each language community.

Confusion is sometimes created in the minds of a people who use one of their own native words for a concept that may not correspond with the original meaning of the word. For example~ the Luo word hera means •love•.

When the Bible was translated into Luo charity was translate-ti as hera. lt happens that ..!2.!e was the original meaning of charity and therefore ~ was probably an accurate translation. Yet the meaning of charity has changed, that is, it is now used to mean the gi ving of alms for the purpose of relieving certain forms of disability or distress. There is no Luo word for this concept, for the practice is relatively new in the life of the

Luo people. The meaning of~ is now being extended to include this new sense of its English equivalent.

The way time is told is also a source of confusion for Luo and Swahili children learning English.. Watches and docks are fairly recent arrivals on the East African scene.

Before their introduction by the British, time was told only in such vague generalities as morning, afternoon and night• with sorne subdivisions of these, such as early morning, late after­ noon or midnight. However, with the introduction of watches and clocks Luo and Swahili people's ideas about time ceased to -152- be vague and became precise. There was, however,. one major difference between the Luo and Swahili time scheme and the English one. Instead of regarding 1 o'clock in the morning as the first hour of the day as English-sp~aking people everywhere do, Swahili people regarded 7 o1clock in the morning as their first hour of the day, and Luo people, in their turn, followed suit. In both languagest therefore, the house of the day are counted from 7 o•clock in the morning to 6 o 1clock in the evening,. and from 7 in the evening to 6 in the morning.

This contmsts with the pattern. in English, which is from 1 olclock in the morning to 12 noon,. and 1 in the after­ noon to midnight. There is therefore a difference of six. hours between the English patternand the Luo and Swahili one.

Morning,. for ex.ample, starts six hours earlier for the

Englishman than for the Luo or Swahili individual, that is,. morning for the Englishman is the time between midnight and noon,. whereas for Luo and Swahili people it is the time between sunrise at 7 o'clock and noon.

It is this difference between the time each group thinks the day starts that is the source of confusion for Luo and Swahili children learning English. For ex.ample, a

Swahili child saying ni saa tisa means •it is 3 o•clock•, for the Swahili word tisa means 'nine•. Very often Swahili -153-

children say in English it is nine o•clock when they mean

3 o'clock,. since they translate the saa tisa of Swahili directly

into the English nine o'clock..

If it is 3 o•clock in the afternoon,. the Swahili

individual. if he wishes to point this out, adds ya mchana to

the above statement. just as the Englishman,. for the same reason,.

adds in the afternoon to his statement. If. on the other hand,. it

is 3 o•clock in the morning, the Swahili individ.ual does not say

ya asubuhi, lin the morning1 as the Englishman does, bu.t ya usiku,

•at night•, for night for hlm is from 7 o 1clock in the evening to

6 o•clock in the morning. Therefore when speaking English the

Swahili child says it is 3 otclock at night he means 3 o 1clock

in the morning.

Luo children' difficulties with the English time

scheme are similar to those of Swahili children. They, too,.

say en sa ongachiel when they mean 3 o'clock. When they say

in English it is nine o'clock they mean 3 o'clock because 9 is the

digit used in Luo for 3 o'clock, and they translate the Luo word

for nine, ongachiel, directly into English and get nine o 1clock.

In the afternoon is mar odiechiengt in Luo. As

in Swahili,. night covers ali the hours of darkness, tb.at is, from

7 o 1clock in the evening to 6 olclock in the morning. 3 otclock

in the morning is therefore not mer okinyi t in the morningt, but mar otieno,. 'at night'. -154-

It is not only when speaking English that Luo and

Swahili children have difficulties with the English tim·e s cheme; they have the same trouble when listening to English spoken.

A statement in English that it is 9 otclock makes these children that it is 3 o•clock. for they translate nine into their language and get ongachiel and tisa respectively. And as has already been pointed out, sa ongachiel in Luo and saa tisa in Swahili mean

3 o•clock.

By the same token, 3 o'clock means to Luo and

Swahili children 9 o•clock• since, by a process similar to the above, 3 becomes adek in Luo and -tatu in Swahili. And sa adek in Luo or saa tatu in Swahili means 9 p"clock.

The basic cause of all the dif.ficulties discussed in this Chapter is .failure on the part of Luo and Swahili children learning English to 11 think in Englishn, that is, to use English words to represent the concepts that they symbolize in English, to see the reality behind the words of English. Direct translation from one language into another is a very inadequate means of gaining an insight into the inner processes of a new language. Such an insight is best gained through .familiarity with the culture of the native speakers of the new language, that is. an awareness of reality as seen by the native speakers of the language being s'flldied,. and of the words these people use to represent that reality. 155-

CHAPTER V

CONCLUSION

It is clear from the foregoing that Luo and Swahili children experience certain difficulties when learning English, and that these difficulties are caused by the linguistic and cultural differences between their native languages and English.

There are othe r factors which also influence the kind of progress made in English language teaching in the schools of Kenya. Sorne of these are the leve! of education and the amount of training which those who teach English in the schools have. the amount of time devoted to the study of English and the opportunities that are available for Luo and Swahili children to expa;nd and consolidate the knowledge of English acquired in school.

Ail these factors have been examined and discussed in the body of this dissertation. It remains to be seen whether any condtsions of wide application can be drawn from this awareness of the extent to which a1l theee factors influence the way Luo and

Swahili children learn English. The author has no illusions re­ garding the depth at which these difficulties have been analysed and defined in this dissertation. It is clear• for example, that a more detailed examination of the linguistic and cultural differences -156- which exist between Luo and English, and Swahili and English, will have to be carried out before there can be a feeling that what is known about these differences consti.tutes a firm enough foundati.on upon which better and more effective methods of teaching

English can be based.

In certain areas. such as the phonemic structure of English,

Luo and Swahili• special instruments may be needed in order to determine to what extent English phonemes differ from, or are similar to, the corresponding phonemes in Luo and Swahili. Ail the differences thus discovered and the concomitant learning diffi­ culties that they cause would then be taken into account in designing an English course for Luo and Swahili children, drawing up the curriculum to be followed and w:riting textbooks to be used by these children in the study of English. As has already been stated. the author does not have the knowledge necessary for the successful execution of this task, but certain suggestions can be made to indicate the lines along which the people who will undertake that task will have to think.

A basic requirement which will have to be met before there can be any improvement in the quality of English teaching is the qualifications and training of teachers. A decision has to be made regarding the level of education necessary for those who teach English -157- at the various levels. It is vitally important that those who teach

English in the primary schools have an intimate knowledge of the language, especially its sound systemand grammatical strudlre.

For it is they who introduce young children to English and on them therefore depends to a large extent the foundation on which the mastery of :English by these chiklren rests.

The stage at which English is introduced in the schools will also have to be set down specifically. At present it is done in the tb.ird year of school, but many schools, especially those in the outlying districts which do not have sufficient teachers, do not do so until the fourth or fifth year of school. Pxogress in the language is therefore uneven, and this affects secondary school work.

But apart from the stage at which English is introduced, it is important that the children start to learn the language when they are at that stage of development most conducive to language mastery.

For some, the third year of school may be the most suitable tim.e for contact with a new languageJ but for others this may be earlier.

Dr. Wilde;r Penfield, formerly the Director of the Montreal Neurolo- gical Institu.te, has argued for introduction of a second language

at a tender age when imagination, sensibility and memory are sufficiently flexible to receive and easily retain the impressions stam.ped upon them. 1

1. Canadian Conference on Ed~tion: Innovations in the Teaching

of Engllsh as a Second Language1 1961 p. 13. -158-

And it is the opinion of John Downing that

it is not unreasonable to suggest that the optimum period for leaning the skill of manipulating the visual code of the letters of the alphabet may closely follow the period in which children naturally master the auditory code of phonemes .. 2

The question to which an answer has to be found is whether or not the introduction of English earlier than is now done would benefit most children in the schools of Kenya in terms of giving them a grasp of the language sound enough to enable them to pursue with profit whatever goals. academie or otherwise, they may choose.

As to how !essons in English at the primary leve! could be conducted, a general outline may suffice. !t would., for example, be prudent if in the initial stages of Engllsh learning only those words with sounds existing in the language of the learner should be presented, leaving those with sounds that do not exist in the sound system of the learner•s language until confidence in the new medium has been established through familiarity with its known sounds. In this connection it may be weil to remember that

Experience has repeatedly shown that forcing a child to express himself in a second language before his ear has become attuned to the new sounds· is to torture him and risk discouraging him. 3 2. John Downing. Address p. 78. 3. ~d. Canadian Conference on E~tion p. 33. -159-

In teaching Luo and Swahili children, for example, such words as desk, ~~ pet, ink, table, door, and the like, could be taught at the start without making the children fee! that the new language is difficult, for ali these words have sounds which exist in both Luo and Swahili. The words could then be used in sentences auch as what is this, with the children responding it is a desk, a book and so on. This would be an example of the application of the knowledge of English and Luo sounds, or English and Swahili sounds, to the teaching situation.

It would be a definite asset if a list could be made of the words suitable for introduction in the first few months of

English. Another list could then be made of the words with sounds that do not exist in either Luo or Swahili, and which Luo and

Swahili children would, for that reason, find difficult to pronounce.

The only setback is that sorne of the most conunon words in Erg lish re long to the second group, so that it would not be wise to postpone their introduction for long. Examples are ~ bird and eup. ali of which have vowels which do not exist in Luo or Swahili.

However, the principle governing the choie e of words taught in the initial stages would have to be ease of pronunciation for the children concerned. It would be necessary to start with simple common words, preferably monosyllabic words such as pen and desk.

Unusual words could be left until familiarity with the common sounds of the language was established. -160-

It is important that the aim s of English language teaching at the primary leve! be clearly defined and adhered to as far as possible. The people who will design a course of

English for the primary grades rra y first have to formulate the aims of primary English,. for it would not be possible to know what to teach if there was no unifying aim or purpose b&hindit.

Progress would be severely hampered if these aims were not realistic in terms of the level of achievement possible in the primary grades.

For ex.ample,. there should be no attempt in the primary grades to teach English grammar. All efforts should be directed at producing a mastery of the basic sounds of the language,. especially those represented by some of the most common words in the language. This should gradually develop into the abUity to express in simple language thoughts and ideas of common occurrence,. for example., My naxne is Juma, ~ eight years old. Reading could then be introduced at a later stage.

The skills required in the middle grades, that is, from the sixth to the eigh1ityear of school~ necessitates an approach to the English language that is radically difference from the approach used in the primary, grades. Children in the middle grades have to read a considerable amount of literature and therefore need to be taught the skills useful in reading.. for example, speed in relation -161-

to the material read and comprehension.

At a more fundamentallevel. it is necesaary

that the k.ind of literature available to children in the middle

grades be that which can best give them the background they need for further study of English. It would be inappropriate,

for example, to have on the reading lists of these grades such books as Sir Walter Scott•s lvanhoe or Charles Dickens•

Pickwick Papers or George Eliot•s Middlemarch. These would frustrate children in the middle grades because of the difficult language in which they are written and the totally alien background in which the events in these books take place.

What is needed is a great supply of adven~ue books

suitable for children in the se grades and cast in an African setting.

Tom Sawyer is an enjoyable adventure book, but it is doubtful whether many African children in the middle grades would enjoy

reading it, asSWir.dng that they w:x:1 erstood it. Although interest in, and enjoyment of certain animal tales eut across cultures, most of the stories written for English children in the middle grades are as unsuitable for African children of the same age as they are for, say, American children of the same age. It is im­ perative that African writers produce adventure stories designed specifically for African children, for there is a wide gap between the material available to the adult reading public and that which -162- forms the bill of fare for African children.

At the secondary s chool leve! the re could be an increase of contact between students and English culture.

The purpose of this contact would be to deepen the students• knowledge of the English language and wlden their intellectual horizon. It would therefore be essential that the medium through which this contact is mainta.ined is chosen with care.

The best way of esta.blishing contact with a new culture, short of living in the new culture itself, is by reading the literature it has produced. Since African secondary school

students cannot be expected to have time to read ail the literature that England has produced, there would have to be a careful sele­

ction of the works to be read; the aim being to present these

students with a representative crosa-section of Engllsh thought at a level they can understand.

These, then, are the Unes along which experts will have to think if Kenya is to solve the problem of Engllsh language teaching in the schools. Any piecemeal approach to the problem is unlikely to pToduce worthwhile results. What can be done, for

Luo and Swahili children, by way of analysing the difficulties they experience in learning English, and adjusting their English programme to eliminate those difficulties, can be done also for ail the other African children in the schools of Kenya. It is a task that has to be undertaken. -163-

BIBLIOGRAPHY

1. Agaard.. Frederick B. and Dunk.el, Harold B., An Investigation of Second Language Teaching, (Ginn & Company, 1948)

2. Allen, Harold B. Readings in Applied English Linguistics (New York: Appleton-Century Cr&fts, 1958)

3. Anshen, Ruth N. ed. Language: An Enquiry into its Meaning and Function (Harper & Brothers, 1957)

4. Bernstein, Theodore M. Watch Your Language (London: G. Bell & Sons, 1914)

5. Block, Bernard and Trager, George L. Outline of Linguistic Analysis (The Linguistic Society of Ame rica, 1942)

6. Bloomfield, Leonard An Introduction to the Study of Language (London: G. Bell & Sons, 1914)

Language (New York: HenryHolt & Company, 1933)

7. Brown, Dorothy and Butterfield, Marguerite. The Teaching of Language in the Primary Grades (New York: MacMiliân, 1941)

8. Brown, Roger. Words and Things (Glencoe, illinois: The Free Press, 1958)

9. Bryan~ M. A. The Bantu Languages of Africa (Oxford University Press, 1959) 10. Cherry, Colin. On Human Communication (The Technology Press of the Massachusetts Institute of Technology and John WUey & Sons, 1957)

11. Cole, Robert D. Modern Foreign Languages and Their Teachlng {New York: D. Appleton Centry Company, 1937)

12. Cornelius, Edwin T. Language Teaching. (New York: Thomas Y. C rowell, 19 53)

13. Dean, Leonard F. and Wilson, KEnneth G. eds. Essaya on Language and Usage. (New York: Oxford University Press, 1963)

14. Ebeling, G.L. Linguistic Units (Mouton & Company•s-Gravenhage, 1960) -164-

Bibliography (Continued)

15.. Elliot, A. V.P. and Gurey P. Language Teaching in African Schools (Longmans, 1940)

16. Estrich. Robert M. and Sperber, Hans. Three Keys to Language (Rinehart & Company, 1952)

17.. Fife, Robert H. A Summary of Reports on the Modern Foreign Languages (Macmillan, 1931)

18. Fries,. Charles. Teaching and Learning English as a Foreign Language (Ann Arbor: The University of Michigan Press, 1945).

19. Gardiner, Alan H. The Theory of Speech and Language (The Clarendon Press, 1932)

20. Gleason, H.A. An Introduction to Descriptive Linguistics (Henry Holt & Company, 1955)

21. Gordon, Edward J. and Noyes, Edward S. Essays on the Teaching ... of English (New York: Appleton-Century-Croft, 1960)

22. Gowers, Ernest. The Complete Plain Words (Her Majestyts Stationery Office, 1954)

23. Graff, Willem L. Languase and Languages (New York: D. Appleton Gentury-Croft t 1932)

24. Gray, Lewis H. Foundations of Language (New York: Macmillan, 1939)

25. Gurrey, P. Teaching English as a Foreign Language (Longmans,. 1955)

26. Harris., Robert T. and Jarrett, James L. Language and Informa! Logic (Longmans, 1956)

27. Harris, Zellig, S. Methods in Structural Linguistics {The University of Chicago Press, 1951)

28. Henmon, V.A. G. et al. Prognosis Tests in the Modern Foreign Languages {Macmillan, 1929)

29. Hill, A. A. Introduction to Linguistic Structures (Harcourt, Bm ce and Company, 1958) -165-

Bibliography (Conti.nued)

30. Hockett, Charles F. A Course in Modern Linguisti.cs (New York: Macmillan. 1958)

31. Hodgson" F.M. Learning Modern Languages (London: Routledge & Kegan Paul, 1955)

32. Hoijer. Harry ed. Language in Culture (The University of Chicago Press, 1954)

33. Huebener. Theodcœ. How to Teach Foreign Languages Efficiently (New York University Press. 1959)

Audio-Visual Techniques in Teaching Foreign Languages (New York University Press, 1959)

34. Hughes, John P. The Science of Language: An Introduction to Linguisti.cs (New York: Random House, 1962)

35. Jespersen. Otto. }Iow to Teach a Foreign Lan-guage (George Allen & Unwin,. 1904) The Philosophy of Grammar (George Allen & Unwin.: 1924)

36. Jones, Daniel. An OÜ.tline of English Phonetics (Cambridge: W. Jeffer & Sons, 1918) The Phoneme: lts Nature and Use(Cambridge: W. Heffer & Sons, 1950) 37. Lado, Robert. Linguisti.cs Across Cultures (Ann Arbor: The University of Michigan Press, 1957)

38. Larsen, Thorleif and Walker, Francis C. Pronunclation (Oxford University Press. 1930)

39. Laban, Walter et al. Teaching Language and Literature (Harcourt, Brace & World, 1961)

40. Mackwardt, Albert H. American English (New York.r Oxford University Press, 1958}

41. Makerere College, Kampala, Uganda. Report of the Commonwealth Conference on the Teaching of English as a Second Language, 1961.

42. Mencken, H.L. The American Language (New Yorkt Alfred Knopf, 1936) .' -166-

43. Modern Language Association of America. Second LanFge Learning (Center for Applied Linguistics, Washington, D.C.. 1961)

44. Palmer• Harold E. The Oral Method of Teaching Languages (New York: World Book Company, 1926)

45. Purin, C. M. The Training of Teachers of Modern Foreign Laneges (Macmillan, 192 9)

46. Sapir, Edward. Language. (New York: Harcourt, Brace and World, 1921)

47. Smalley, William A. Manual of Articulatory Phonetics, Part One. (Committee on M.issionary Personnel, Division of Foreign Missions NCCC 1961)

48. Smith, Arnold. Grammar and the Use of Words (London: Methuen & Company, 192

49. Street, Lucie. SpoU the Child. (London: Phoenix House, 1961)

50. Strong, Barbara M.H. Modern English Structure (Edward Arnold (Publisher) 1962)

51. Ullman, Stephen. Semantics: An Introduction to the Science of Meaning. (Oxford: Basil Blâckwell, 1962)

52. UNESCO, Paris. The Use of Vernacular Languages ~·in Education, 1953

53. Upton, Albert. Design for Thinking (Stanford University Press, 1961)

54. Weinreich, Uriel. Languages in Contact; Fi:rdings and Problems (Linguisti.c Circle of New York, 1953)

55. Werner. A. The Language Families of Africa (Society for Promoting Christian Knowledge. 1915)