THE PRIMARY MOTIVATION for this essay is my continued interest in the linguistic behaviour in diglossia communities during the past decade. I was born and brought up in such a community, and it is clear to me from whatever introspective insights I possess that the learning strategies, attitudes to expression and points of emphasis, as well as the choice of criteria for norm specification, are significantly different in these communities from those in other communities which, for the sake of contrast, may be loosely called 'non-diglossic'. There have been many advances (at any rate, changes) in the teaching methods employed, and learning strategies presented, in the classroom since the time I was a schoolboy, but a language classroom in a south Asian diglossic environment today is little different from what I remember of it from my school days. The way I was taught to read and write and write compositions in Sinhalese and the way Sinhalese school children are taught these crafts today both differ drastically from the comparable learning activity of, say, my children whose first language is English.

It must be said, however, that my observations, which are made in the body of this essay stem, not from any personal reminiscences, but from a fair number of case studies I have made during the past few years. As far as possible, I have refrained from allowing my memories to influence these case studies, and I have always refrained equally from pre-judging the issues; my finding in recent years have, on the other hand, enabled me to recover more vividly my own involvement in the classroom in a diglossic setup. (While I am on this impertinently personal note, may I be permitted to be immodest enough to say that my own Sinhalese writings have never been criticized for grammatical inaccuracies, etc., and (perhaps) therefore, this essay must not be understood as containing an apologia for any of my own frustrations as a Sinhalese language learner in the class-room or as an adult Sinhalese writer. The significance of this comment will become clear when the reader finds that my findings on literacy seem to disfavour diglossia).

In 1974-5, I spent some nine months in India, and about six weeks in Sri Lanka, working on a project to which I gave the title "An investigation into problems of literacy in diglossic communities which particular reference to Kannada, Sinhalese and Tamil". Although during this research I was fortunate enough to associate with people engaged in teaching literacy to illiterate adults in India, and to have the benefit of taking part in their workshops and seminars, I was never directly concerned with the very valuable task of devising techniques for facilitating universal adult literacy in the sub-continent. Teaching literacy to adult illiterates is a theme that is being pursued now more conscientiously than everbefore.Mostprotagonists of adult literacy campaigns are motivated, quite rightly, by their social conscience. Notice, for instance, the effect in Britain of the recent discovery of two million illiterate adults. That quick measures to eradicate this problem should be sought in a country infested with an illiteracy figure of seventy-five per cent is therefore, neither surprising nor objectionable.

There is another side to this literacy campaign, namely, the need to discover the causes of illiteracy by asking the sociological question as to why there is so much illiteracy in a world which is technologically so advanced. We may ask, likewise, why the figures for India are even worse than the world average. For a country which had been known in the distant past for its pioneering intellectual activity in the arts as well as the sciences, twenty-five percent literacy in India, but this should not deter research.

One very important factor that must be noticed and explained is the high rate of literacy claimed in the neighbouring island of Sri Lanka which contrasts dramatically with the low rate in India. I bring Sri Lanka into the picture because, as will be seen subsequently, my research involves that Island. Although the social order in Sri Lanka is not identical to that in India, they are comparable and share many salient features. They are culturally similar; both are largely peasant economies; both are caste-based in social structure; and so forth. It might be that the smaller size of the Island has something to do with the prevailing situation. It might also be that the law governing compulsory primary education is enforced more diligently and effectively in the smaller island than in the enormous sub-continent. To account for bulk-literacy, we must take note of as many factors as we can glean from evidence and assess the relative contribution of each factor, as far as feasible, to the maintenance of literacy standards as we know them. Until this research is done, the only satisfactory answer to the question "Why is there so much illiteracy" is "We simply don't know".

Another social and psychological question that is worth asking is "Do we want universal literacy in our country?" Such questions are not favoured by the dedicated social worker, but it might well be the case that in certain communities the illiterate adults does not particularly want to be literate. Even the aspirations of such schemes as functional literacy are not always well received in all communities. In a way, all literates do not particularly want the entire communities to become literate. Let me illustrate this with one example. It has been proposed time and time again as a quick and simple solution to mass illiteracy, that if each literate person took under his wing one illiterate adult and saw him through, universal adult literacy could be achieved in India in five years. An attractive suggestion, indeed, but one, which is apparently more rhetorical than practically viable; for, if the solution is as simple as this proposition seems to suggest, why has it never been taken up as a serious venture in India, or anywhere else? When the organization of a society and its attitude to education are such that education-based élitism becomes a divisive minority concern, altruistic endeavours as advocated by the above suggestion become far less practicable than they appear at first sight. Might it be the case, then, that the hierarchical social structure in India somehow contributes to the perpetuation of illiteracy? But, then, what about Sri Lanka, where the social hierarchy is very similar? I am only illustrating how ignorant we are of the causes of illiteracy in our countries. As long as we cannot say for certain what impact each possible cause may have on the perpetuation of illiteracy, we must continue to explore, taking note of each factor that might seemingly be a cause and analyzing with some care its impact upon the society with a view t estimating the relative significance of that cause. This is precisely what I have attempted to do during my research year, and the factor I chose was diglossia.

Kannada, Sinhalese and Tamil communities are diglossic. (For more details see later chapters.) Translated into the school or the classroom situation, by 'being diglossic' is meant that these communities have accepted norms of linguistic excellence, the teaching of which is the purpose of the language teaching curriculum in the school. This normative variety of language is the one that people are expected to write. It is distinct from the various spoken dialects in grammar, lexis and phonology in spite of the shared features which make them mutually comprehensible to some extent. Similar situations are known in almost all language teaching activities everywhere, but, as I shall venture to demonstrate in due course, the linguistic values associated with diglossia are different from the overtones of 'good usage' that all teachers of all languages attempt to inculcate in the learners. With reference to literacy, my investigation had one motive, namely, to discover if this cleavage was in any way responsible for making literacy difficult to achieve. If the investigation showed that the cleavage made the learning of literacy difficult, then, the related question naturally crops up, namely, "Is diglossia responsible for the high rate of illiteracy in India?" Here, once again, we are thrown into confusion when we, inescapably, compare the India figures with figures from Sri Lanka, where the learning of literacy is apparently unaffected by the diglossic character of the Sinhalese community. Could it, then, be the case that the literacy statistics are not comparable in the two countries? Or, perhaps more realistically, might it be that the concept of literacy has different meanings in two countries? And, following from this, might it be that any differences in the definition of literacy in India and Sri Lanka have been motivated by different attitudes to diglossia in the two countries? It is the type of answer we can get for these and other related questions that will enable us to assess the definitions of attitudes to, and the social motivations for, literacy. I have handled some of these issues in my case studies which will be discussed in summary form in due course.

Let us, for a moment, turn our attention on the educational systems the world over. A fair proportion of the world's population is not only unenthusiastic about literacy but is also unable to derive the desired benefits form the schooling systems operative in their respective countries. It has been suggested that the middle-class oriented (see Bernstein: 1973, etc.) 'banking' concept of education (see Freire: 1970) disfavours the rural, peasant learner. The prevailing educational systems in all parts of the world are geared towards linguistic elaboration and abstract participation. It is Bernstein's belief that, by virtue of their social background, the middle class child enters school equipped with the tools necessary for this activity. Bernstein suggests that

the typical, dominant speech mode of the middle class is one where speech becomes an object of special perceptual activity and a 'theoretical attitude' is developed towards the structural possibilities of sentence organization. This speech mode facilitates the verbal elaboration of subjective intent, sensitivity to the implications of separateness and difference, and points to the possibilities inherent in a complex conceptual hierarchy for the organization of experience. (Bernstein 1973: p.78)

Bernstein maintains that this is not the case for members of the lower working class

The latter are limited to a form of language use, which although allowing for a vast range of possibilities, provides a speech form which discourages the speaker from verbally elaborating subjective intent and progressively orients the user to descriptive, rather than abstract concepts. (op. cit. p.79)

Bernstein concludes that

Between the school and community of the working-class child, there may exist a cultural discontinuity based upon two radically different systems of communication. (op. cit. p. 166)

I do not wish to engage in a critical appraisal of Bernstein's work in this essay. It might well be that social class is not the only dominant variable to consider. Poverty is another important variable, and poverty and membership of lower classes are not coterminous. I want, however, to assume for the present purpose that Bernstein must be making sociological sense insofar as social class is at least one variable that is worthy of consideration.

In his Language in the Secondary Classroom Barnes (1969,……1974) reports on a number of case studies in order to assess the amount of communicability in the classroom situation. Unlike Bernstein whose ultimate objective is to discover remedies to help the deprived, Barnes attempts to observe how much a child can do in the classroom. Without invoking social stratification as a variable, Barnes discovers that the teachers' attitude to communication in the classroom does not motivate pupil participation as readily as should be the case and does in this sense generate a cleavage, between the classroom use of language and the pupils' normal linguistic usage. The 'banking' concept of education is, perhaps, partly to blame. In his Pedagogy of the Oppressed (Freire 1970), and elsewhere, Paulo Freire severely criticizes the banking concept and advocates the principle of participatory learning particularly as a device to sustain the interest of the under-privileged learner.

I have talked about this notion of school education as a middleclass, banking system with a purpose. In India, might it be the case that this character of the school system fails to motivate the rural, working-class or 'backward' child to avail himself of the educational opportunities? I am not a student of sociology of education in any significant way, but the South Asian evidence is strong enough to persuade one to uphold the view that the present educational systems are, perhaps, not in the best interests of the 'backward' communities in the South Asian villages. It is conceivable that the linguistic content is particularly in their disfavour. It is a fact that, the world over, more middle-class children benefit from school systems (or, fit the school systems) than working-class ones; India and Sri Lanka are not exceptions. Linguistic cleavage may not be the only reason for this. Poverty may turn school education into a sheer luxury that the starving millions cannot afford, but it must be recognized that even in areas in which education is valued so high that a parent would pawn his belongings to send a child to school, the school system forces some children to stagnate or drop out. The school curriculum does not motive the child to see its relevance to his kind of living. Some recent tribal education statistics for India (Education in India 1963-64) show that while 72 percent of the tribal children enrolled in schools in the age group 6-11, their number dwindled away rapidly, going down to 28 in the age group 11-14 and to 13 in the age group 14-16. This is not a tribal matter: a similar picture is given by the statistics of stagnation and drop-out for the whole country (Ahuja 1975) The failure of Government's efforts might be due to the educational philosophy that demands from the lower working-class or the peasant child the abilities of the middle class child which the latter acquires not through his intellectual capacities but through his more conducive social circumstances.

If, following Bernstein and other, we were to conclude that linguistic deprivation is largely responsible for the low rate of education in rural communities, it would be reasonable to assume that the school requirements in diglossias would present a further handicap to children of such rural communities. In some sense, the book language in diglossias is twice removed from the lower working class or rural child: he has to cope, first, with the school-oriented communication modes and, second, with the special requirements imposed upon him by the book language. In such circumstances, the inhibitive effects of diglossia must be considerable, though not necessarily quantifiable. My findings show that the presence of a high variety with its social implications inhibits people in their writing activity. In being made to learn features which are alien to the normal daily usage-as if they are the real elements of his language, which, devoid of them, is incorrect, the learner is forced to emphasis form rather than content: embellishment rather than essence; imitative ability rather than creativity. These features are widely known in the school writings in South Asia.

Now for a few brief comments on the uses of the term 'Literacy'. It seems to be relevant to make a distinction between necessary literacy and sufficient literacy, at least for the type of statement I shall be making in the body of this essay.

My own interest in literacy is, needles to say, an academic one. As I have said before I have never been involved in the pedagogic aspects of literacy either for the non-literates or for the neo-literates. Being involved in education, I am naturally concerned with literacy in its widest sense, but in that sense literacy and education are virtually synonymous. In my academic research, my interest lies in discovering through empirical observation how far people's linguistic repertoires and their societies' idealization of literacy are compitable, and also, how for the acquisition or, at least, the awareness, of literacy can condition people's idealization of language and linguistic behaviour. The impact of literacy on linguistic evoluation, linguistic conservatism, and, even, linguistic chauvinism, come within the preview of my investigation. In short, my primary interests are the sociology of literacy and its impact on human language behaviour at large. I shall, however, attempt to draw from my understanding of literacy in the context of diglossia, at least some facts that might not be altogether irrelevant for practical purpose. From among the pedagogues involved in literacy programming, the definition of literacy that is closest to my kind of research purposes in the one held by the proponents of the language experience approach to literacy education. I shall attempt to rely heavily on the language experience approach as my pedagogic model in the belief that it is the one that is most relevant for adult education purposes also. My references above to Bernstein, Barnes, Freire and others were made in this spirit.

As my work is on diglossic communities and, therefore, linguistic styles have an important role to play in my research, let us assume that language is but a conglomeration of styles with their conventional functions. Literacy naturally presupposes language. Also, literacy, in any real sense, is expedient only in a language of which the learner is a fluent speaker. Within the so-called 'same language' there can be styles which are unknown to the learner, or which are not within the repertoire of the learner's linguistic functions. Just as literacy learning is inexpedient in a foreign language, even so literacy learning is irksome in foreign styles: Bernstein and Barnes deal with this aspect specifically. It may also be said that literacy is, in some sense, a potential creator of further styles, for literate people, through their association with the expressed styles of writers, evolve way of expressing themselves which might not have been possible without acquiring literacy. Literacy is thus a device for both unification as well as diversification, at different levels. It is with the acquisition of sufficient literacy that the individual is able to play these conflicting roles. The distinction of necessary literacy and sufficient literacy is, therefore, a useful one. If literacy in this second sense is virtually synonymous with education, it might be argued that the notion of sufficiency is inapplicable: for, there is never a sufficiency mark in learning. I do use the term advisedly, however. There is a point at which a person's learning is sufficient to grant him admission into the educated class or the literati. Although he does not necessarily refrain from any further learning, he may, in this regard, be thought of as having reached a sufficient standard for such admission. This is my use of the term 'sufficiency' in this context.

Many people define literacy as the ability to read and write. The dictionaries, too, give this as one definition of the term. In language such as those in India there is a certain degree of correspondence between the spoken sounds and the letters we use to represent them on paper. This correspondence, where it obtains, is regular, and, if we want a name for this, we may say that our writing systems are to a certain extent phonemic. A writing system is said to be phonemic when the regularity of sound-symbol correspondence is easily discernible. In phonemic writing systems, the learning of reading and writing is a comparatively easy thing to do. This is why, for instance, the initial teaching alphabet, which is modified form of English writing -modified in the direction of phonemicity-has been accepted by many English schools as a good starter for the learner. The UNESCO adult literacy organization has devoted a full volume of their journal Literacy Discussion to discuss the advantages of the I.T.A. We in India do not, fortunately, have to evolve initial teaching alphabets; our alphabets are good teaching-learning systems.

In our languages, therefore, the acquisition of the ability to read and write should be comparatively easy. The writing systems in our languages, we can say, are adequately designed to enable minimum literacy. The ability to read and write is minimum literacy, or NECESSARY literacy.

When I once spoke to a group in England about my interests in literacy research, some English school masters asked me whether I would consider the ability to read the English Daily newspaper called the Daily Worker to be literacy. This is an interesting question. These schoolmasters obviously thought that, while all literate persons can read the Daily Worker, or to put it differently, the ability to read the Daily Worker is a necessary requirement for being literate, that ability alone would not be sufficient as the definition of literacy in a society in which most people can read and write. Their question was, then not about necessary literacy, but about sufficient literacy: not whether the ability to read the Daily Worker is necessary, but whether it is sufficient. I have said above that the definition of literacy as the ability to read and writer is a definition based on necessary conditions. As a definition based on sufficient conditions, notice the following definition of literacy from a book recently published on the subject:

In modern education literacy has to be conceived as including an ability to express oneself articulately for a variety of purposes, socially, intellectually and vocationally, both in speech and writing: command a capacity to read for information, enjoyment and enrichment: and to respond sensitively and intelligently to what is said as well as to what is written. (Goddard 1974: p. 21)

In third-world-societies there is need for programming literacy in both dimensions-both to impart the necessary abilities as well as to enable people to become sufficiently literate. As I have said before, the illiteracy figures for India is appallingly high, and, for those unfortunate many, sufficient literacy is even further than a dream world. Many have passed the necessary stage via schools and via adult education schemes, but in order to grant them all sufficiency in literacy the educational technology has to be radically overhauled: the desperate need, to my mind, is an overhauling in the direction of language experience. This is particularly so in diglossic communities with the notion of prestige associated only with the high variety of usage. In our overhauling efforts we should plan to introduce methods, which would eventually enable the bulk of the population to express themselves

articulately for a variety of purposes, socially, intellectually, and vocationally, both in speech and writing; command a capacity to read for information, enjoyment; to respond sensitively and intelligently to what is said as to what is written. (Goddard, op. cit)

It is one's observation, however, that such a degree of sufficiency is still beyond even the average university student or graduate: hence the need for a reappraisal of teaching strategies. The two requirements, namely the need for necessary literacy programmes as well as sufficiency programmes, must be met simultaneously, for, as John Dewey once said, the progress is not in the succession of studies but in the development of new attitudes towards, and new interest in, experience.

We have noticed that Indian writing systems are phonemic to a large extent and, therefore, not too cumbersome to learn. Those in the business of teaching literacy know, however, that the teaching of letters in isolation is too time consuming and dull. Many projects have succeeded in teaching literacy via whole words. It is equally dull and often futile to attempt to teach via words which have little or no semantic relevance to the learner. The learning of literacy becomes a meaningful operation only when it is associated with personal experience. For, as Nora Goddard points out,

Literacy in this sense must have its first beginnings in sensory and social experience, for it is not possible to understand fully what has not been either felt at first-hand or entered into through empathy.
(Goddard op. cit. p.21)

My comments on problems of literacy in diglossic communities must be understood as an attempt to observe critically the techniques of literacy acquisition in the communities under investigation, by a proponent of the language-experience approach briefly outlined above.