Leslie W. Duffen Remedial Education Vol. II, No. 3, 1976, p. 139-142.

Reprinted with the permission of the author

One of the dogmas of education, and of special education in particular, is that fluent speech must, and should, precede learning to read. I believe that this may be an error of almost tragic proportion. It is possible, in fact, that some children can only learn to speak fluently by first learning to read.

My reasons for believing this are based solely on my experiences in teaching my daughter to read. Sarah, a child with Down's Syndrome, learnt to read before she went to a normal school at the usual age of five. At the age of eight and a half she is still in a normal school and with a reading age rather greater than her chronological age.
     This account of our experiences is divided into two sections. The first is related directly to Sarah and the implications for children like her. The second is concerned with the implications, as I see them, of Sarah's experiences on some generally accepted theories of language development and on reading in particular.

Sarah
Reading Method
Sarah started to learn to read at about three and a half years old. It is a matter of lasting regret to me that she did not start two years earlier. She learnt to read by a long continued series of flash card drills, basically a look-and-say whole word method. The words were selected from, and the expansion of the vocabulary built up roughly in the order of, the Ladybird Keyword lists. Essentially the drills consisted of me saying the word and then Sarah repeating it while looking at the printed word, until she could say the word before I did. It was a lot more complicated than that, of course, but that was the essence of it. Concurrently she learnt to write, recognize and say, with their common phonetic pronunciation, all the letters of the alphabet. Also concurrently she started reading through the Ladybird a, b, and later the c series. (At the present date we have almost completed the total of 36.) Apart from the 'learning to read' reading, Sarah, once she had acquired some reading ability, started selecting her own books from the library to add to her own collection of 'reading for pleasure' books.
     Only some of the words that Sarah learnt to read were in her word comprehension vocabulary. Very few were in her active word use vocabulary. NONE of the sentences were in her active use vocabulary.

Pre-reading language
Sarah learnt to read before she could speak, I define read, here, as the ability to reproduce the sounds represented by the printed words and to show, by question and answer, that she has understood and remembered what is being read, though not necessarily in the same form. I define speak, here, as the ability to produce sentences of standard construction to communicate ideas. Sarah was still in the 'one word' sentence speech stage when she was reading and understanding sentences several years ahead of that stage in the usual sequence of language development. Though the discrepancy between speech output level and reading comprehension level has now been much reduced, she can still read at levels well ahead of her speech output level.

Post-reading Language
Sarah's reading ability has considerably helped the development of her speech, and this is one of the reasons we persisted with the rather monotonous drills. The critical discovery was that Sarah read, remembered and used later, in the correct context, sentences that she was quite incapable of remembering when she had just heard them. This was quite an accidental discovery, made when I put together flash cards of words that Sarah had learnt to read individually, into sentences. One of them was 'We went to Exeter'. A week after reading this sentence Sarah reproduced it in speech, voluntarily and in context, i.e. just after we had been to Exeter again. At that stage in Sarah's development she was quite incapable of reproducing a sentence such as 'It is my turn' even after many spoken repetitions from me, and even though she could repeat each individual word correctly, and even though she clearly understood, from her actions, what 'It is my turn' meant. In the early days of her reading development it was often possible to pick out items from her improving speech and relate them back in part or whole to sentences she had read. Indeed we used this ability to remember printed sentences for regular practice in speaking sentences. Now that Sarah has read so much, and her speech has improved so much, it is no longer possible to pick out, directly, correspondence between speech and reading, but I have no doubt at all that a major cause of her speech improvement is that she learnt to read first.

Intelligence
Sarah's early and wide reading experience has probably improved her intelligence. I believe that all intelligence tests measure levels of attainment in different but specific, and sometimes narrow areas, but with an overall controlling effect from attainment in language development. What we should be concerned with is why a specific attainment is low and what can be done to improve it, rather than be confused and defeated by an overall masking concept such as IQ. Nevertheless, whatever it is that intelligence tests measure, it is beyond dispute that a mastery of language helps children to do well in them. It is a sad and distressing fact that many Down's Syndrome children have a DQ of over 70 when below three years old. Almost none of them have a DQ of over 70 when over three years old. Sarah was given a prolonged test, using the Griffith Developmental Scales, when she was seven years and eight months old and scored a mean DQ of 83. I was present during the whole of the tests and I know that she scored below her best because she was tired and irritable from a long journey the previous evening. Nevertheless she clearly has maintained or improved on the best known levels for younger DS children, and this is very rare. I cannot, of course, prove that this is directly influenced by early and wide reading experience, but the increasing importance, with increasing age, of language mastery in the ability to do well in intelligence tests, makes it highly probable. (Some intelligence tests, such as the Millhill Vocabulary Test, actually correlate vocabulary directly with IQ.)

Value of Reading Ability
Even without the consequential effects of speech development and intelligence, Sarah's reading ability has practical and pleasure values which are beyond dispute. It is also reasonably certain that, without this ability, Sarah would not now be in a normal school. She would probably be in an ESN(M) and possibly an ESN(S) school, because she would not have been able to demonstrate her undoubted overall intellectual ability. Though I hope to make a new career in Special Education, I would not expect any parent to be glad that his child needed to go to a Special School.

Ineducable Children
Sarah is a child to whom the group label of 'ineducable' would have been automatically applied only a few years ago. In fact many Down's Syndrome children have learnt to read, though none perhaps so well and so early as Sarah. Even if Sarah is unique in her attainments, I am quite sure that it is only because many other such children have not been given the opportunity. In particular many other children have been deprived of the opportunity to learn to read solely on the basis of poor speech development. This is quite as illogical as preventing a child from listening until he has developed fluent speech.

Language development
This second section is based on my experiences with Sarah and my study of accepted ideas of language development, both before and since I became a mature student at Saint Lukes College of Education.

Reading Readiness
The concept of 'reading readiness' is harmful nonsense. It has no more validity than a concept of 'listening readiness'. To say that a child must have certain levels of skills, experiences and abilities before she can begin to learn to read is to deprive her of one of the major methods by which she can acquire those skills, experiences and abilities. If a child can see, hear and repeat sounds correctly, without necessarily understanding (whatever that means) then that child is ready to learn to read, and will probably do so with benefit.
     Sarah, with a reading age greater than her chronological age of eight and a half, still cannot pass some of the accepted tests of reading readiness. If I had not disregarded, through ignorance, the concept of reading readiness, I would have not taught Sarah to read. If she had not been able to read before she went to school, many, if not most schools, would not have taught her to read either, again because of this dangerous concept.

Language Input
There are two verbal inputs to language development, not just one. These inputs are, of course, reading and listening. Reading has two advantages over listening. The first is that the visual memory develops earlier and becomes more efficient than the auditory memory. The second is that the rate of input and repetition of input can be controlled to a much greater extent by the reader than by the listener, particularly the pre-speech reader/listener. Reading has one disadvantage. The visual output mechanism corresponding to reading, i.e. writing, requires equipment other than that existing in the human body, and is relatively slow when compared with speech. But, overall, reading is a superior input to language development when compared with listening. For some children, where the auditory input and memory mechanisms are greatly inferior to the visual, reading may be the only effective method of language development.
     'Reading' is one of those blanket terms, like 'intelligence' and 'mathematics', which confuse more than they define. We do not teach a child 'mathematics', we teach him 'adds' and 'takes' and many years and sequential steps later we teach 'calculus'. Similarly we do not teach 'reading', we teach first the 'adds' and 'takes' of reading. This can be defined as 'the ability to recognize, remember and associate a specific graphic with a (moderately specific sound, and, possibly, but not invariably, with a) very much less specific, meaning'. It would be possible to use an even more simple definition, by omitting the words within brackets, which would eliminate the other input mechanism of sound but, in the absence of contrary evidence, it seems reasonable to suppose that simultaneous input to visual and auditory memories will, at the least, have an additive effect on retention.
     If we define reading in this elementary way, and listening in the same way, then all the other skills, experiences of language, definitions, understanding, etc, which are commonly included in definitions of reading or of reading readiness, can be placed, where they properly belong, with the development of language control in general. The evidence of this development is shown in the increasing refinement and complexity of the outputs of speech and writing, and, certainly, the improvement of the outputs feeds back to improve the efficiency of the inputs, but reading and listening are, equally, inputs of language development and not products of the development of the outputs.

Reading Methods
If children have little or no language it is a waste of time and possibly harmful to try to teach them to read by any method which relies heavily on the skills acquired only by mastery of language. Almost all methods of teaching reading assume that children already have a considerable control of language and that reading is simply a transfer process. Naturally if there is little or nothing from which to transfer then the child will not learn to read by that method, nor, after an initial failure, by any other method either. But it is surely a very circular logic which conclusively proves that an absence of language mastery leads to failure, in a situation where only presence of language mastery can lead to success, and then disguises this as a method of learning to read, or rather as a method of failing to learn to read. The parallel in mathematics would be to teach fractions to children who had not learnt adds and takes. The natural, i.e. the easiest, sequence of language development is reading, listening, language dominated thinking, speech and writing, but this is a circular expanding development with a positive feedback at all stages.
     The fact that reading is usually taught out of sequence, as an output of early language development and not as an input, may partially explain the considerable difficulties that some children have. Lateness in teaching/learning reading may be cause of difficulty rather than a result.

Value of Reading
Reading is a more easily acquired and efficient input to the learning processes than listening. It is not a middle class cultural value by which children learn to read Dickens in the original. Parents and teachers who believe that auditory inputs and the impermanent visual images of the television screen can replace reading as a learning input may be doing their children a grave disservice. By the same token parents and teachers who do not teach their children to read at the earliest possible age—and this is much earlier than five—may be allowing them to waste the most receptive and important learning years.

Summary and Conclusions

1. If a child has little or no language this may be a good reason for starting and accelerating the teaching of reading rather than postponing it. Such children can be taught to read, read well, and read at an early age. Even if this does not have consequential benefits for development of speech and intelligence, the practical, pleasure and social adjustment values of fluent reading are beyond dispute.
2. The earliest possible acquisition of reading fluency should be a first priority in all schools, but particularly in the reception classes of primary and special schools.
3. Reading is the most easily acquired and most efficient input mechanism to language development and learning, not the difficult end product of language mastery, provided that it is learnt early enough and by a suitable method.