Cover Image
close this bookInvestigating Bilingual Literacy: Evidence from Malawi and Zambia - Education research paper No. 24 (DFID, 1998, 99 p.)
View the document(introduction...)
View the documentDepartment for International Development - Education Papers
View the documentList of other DFID education papers available in this series
View the documentOther DFID education studies also available
View the documentAcknowledgements
Open this folder and view contents1 Introduction
Open this folder and view contents2 Background: Malawi and Zambia
Open this folder and view contents3 Teaching reading in Malawi and Zambia
Open this folder and view contents4 Reading research: 1992
Open this folder and view contents5 The 1994 research
Open this folder and view contents6 Individual reading sessions
Open this folder and view contents7 Individual reading sessions
Open this folder and view contents8 Discussion
View the documentReferences
View the documentAppendix A - Approaches to teaching initial reading
View the documentAppendix B - Transcript of two lessons
View the documentAppendix C - Texts for two lessons
View the documentAppendix D - Extracts from English ''word find'' reading test, 1992
View the documentAppendix E - Extracts from Chichewa and Nyanja reading tests 1992
View the documentAppendix F - Extracts from English reading test, 1994
View the documentAppendix G - Extracts from Chichewa and Nyanja tests, 1994
View the documentAppendix H - Histograms of 1994 test results
View the documentAppendix I - Text for Chichewa/Nyanja individual reading

Appendix A - Approaches to teaching initial reading

The following is a brief characterisation of some of the main approaches to the teaching of initial reading. In practice most teachers employ an eclectic approach, and also attempt to promote a positive attitude to reading through activities not mentioned here, for example by reading stories aloud. It should also be born in mind that many U.K. teachers and reading specialists now believe that pre-reading phonological awareness is a causal factor in learning to read. In crude terms this means that initial readers are helped if they already have an appreciation of the fact that words are made of different sounds. Such appreciation is probably fostered by rhymes, songs and word play through minimal pairs (e.g. shells, bells) or contrastive addition (e.g. row/grow) which alert learners to the phonemic system. Thus, irrespective of the language concerned or the eventual reading instruction approach, the development of initial reading skill is partly a function of pre-reading experiences, which is not directly connected with written language. (see for example, Bryant and Bradley 1985; Goswami and Bryant, 1991).


This method proceeds from the conventionalised "sound values" of letters - the letter c being given the value "kuh", for example, and the word "cot" being analysed to "kuh" "oh" "tuh" and then synthesised to "cot". The main advantage of this approach is that it enables learners to "build up" by sounding out, and hopefully recognising, words that they have not met previously in printed form. It is sometimes referred to as the "phonetic" method, although phonetic symbols are not used with the learners.

One obvious disadvantage of the method is the lack of consistent letter-sound relationships in
English spelling. Another disadvantage is that there is often a difference between the pronunciation of letters in isolation and the sounds represented by the same letters in a word. Thus in the previous example, neither "c" nor "t" are pronounced in the same way in isolation as they are in the word "cot". In order to identify the word, English speakers will be helped by already knowing the word "cot". Clues gained from the "sounding out" of "kuh" "oh" "tuh" provides learners with sufficient information to enable them to identify the written word with the item "cot". Clearly if the reader does not know the word, arriving at an appropriate synthesis of the "sounded out" letters will be difficult.

Onset and Rime.

This is arguably a technique rather than an approach, deriving from the phonological awareness principle (see above). It is based on the view that syllables can be broken down into onsets and rimes, the rime being the part that allows the word to rhyme with others, and the onset being the initial consonants. Thus in dog the onset is d- and the rime is -og. The rime is also found in words like jog and frog, where the onsets are j- and fr- respectively. It is believed that the manipulations of onsets and rimes in written word games and activities, help children to recognise and develop analogies. It may be regarded as the equivalent for English of the syllabic approach.


The syllabic approach is widely known in Malawi and Zambia, but seems to be used exclusively to teach reading in African languages. It is based on "consonant-vowel" sequences e.g. ba, be, bi, bo, bu; ka, ke, ki, ko, ku, etc. From these, teachers prepare written "syllable charts", as in the following example (from Ms Bernadette Zulu, Kabwata Open School, Lusaka):































The chart provides the basis for various activities, particularly making up different words from the chart (e.g. kalulu, hare; sukulu, school; amai, mother). Such word play activity appears to be very popular with learners, and alerts them to the fact that words are composed of sounds, and sounds are (albeit not always in a perfect one-to-one relationship) represented by letters.

This syllabic approach is well suited to Chichewa/Nyanja, since many of their words, in common with other languages of the Bantu family have a consonant-vowel phonological structure e.g. ka+lu+lu = kalulu (hare); ma+lu+ba = maluba (flowers). (There are digraphs such as mb or ns but these represent single consonant sounds.) The syllabic method does not of course lend itself to English, but the "onset and rime" approach is analogous.

Whole Word and Whole Sentence

This is also referred to as the "look-and-say" method. Here, learners are presented with the written versions of whole words, phrases or sentences, which are read aloud by the teacher, often through the use of flash cards or words written on the blackboard. Pupils are expected to memorise them through repetition, and recognise them as wholes. The claimed advantage of this is that it facilitates rapid recognition of whole units, rather than depending on a laborious letter-by-letter strategy, and as such, that it approximates more closely to the fluent reading of a proficient reader. The disadvantage is that it does not help learners to work out for themselves words that they have not already met in print.

A further point to note is that for native speaker readers attention to meaning will not normally be crucial, or even necessary, since such learners will by definition understand what they are repeating. In a second/foreign language situation however, there is a clear danger that learners will simply repeat without understanding.

Language Experience

This is an integrated approach to both reading and writing which exists in different versions, of which the best known is Breakthrough to Literacy (Mackay et al, 1979). The latter has been adapted by the South African based Molteno organisation for indigenous languages in various southern African countries. A typical classroom procedure is first that the learner decides what to write - usually a single sentence. This is then constructed out of words already printed on cards, or provided by the teacher. The child then reads the sentence - facilitated, of course, by the fact that the child created the sentence in the first place. The child then copies the sentence into an exercise book, and is again able to read it by itself. Both the phonic and whole word methods may be incorporated into this approach.

The advantage of this method is that the child will immediately be able to attribute meaning to what he or she says. The disadvantage is that the approach may be cumbersome to use with a large class, although it can be adapted to such situations. The method assumes the child knows enough language to be able to express itself, as would normally be the case with native speakers for whom the approach was developed.