Thursday book: Reading between illiterate lines

WHY CHILDREN CAN'T READ: AND WHAT WE CAN DO ABOUT IT

BY DIANE MCGUINNESS, PENGUIN, pounds 8.99

IF YOU think the age of sexual consent is controversial, try reading. For decades, experts have battled over the best way to teach it, with hapless parents caught in the middle. In one corner are the advocates of "real books" who believe that children learn to read simply by being exposed to books. In the other are supporters of "phonics" - crudely, the sounding out of words as in C-A-T for "cat". And somewhere in between are the proponents of "look and say", which involves memorising whole words. The official view is that a mixture of methods works best.

So Diane McGuinness's book is shocking. She insists that all these theories are wrong. She contends that there is only one scientifically proven way, based on the ability to hear individual sounds - phonemes. If children also learn to map each sound to its most probable spelling and to master alternative spellings for the 43 phonemes in the language, almost all can learn to read. Though most of her spleen is reserved for the "real books" advocates, she also dismisses "phonics". Phonics does not work, she says, because children start with a letter and match it to a sound.

Her review of 25 years of research leads her to other shocking assertions. Dyslexia, used as an explanation of "learning difficulties" and a label for children of normal or above-average intelligence who have difficulty learning to read, does not exist. Children fail to read not because of something wrong with their brain but because they have not been taught properly. Remedial programmes, such as Reading Recovery, pioneered in New Zealand and now being used by several English local authorities, are dismissed as expensive and ineffective.

There are shocks, too, for parents. If you listen to your children read and correct words or even sound them out without providing them with any way of correcting their own mistakes, you are wasting your time. Forget teaching children to memorise the alphabet, the names of letters and capital and lowercase letters: you may actually do them harm.

Ms McGuinness, a developmental psychologist at the university of South Florida, goes back to the Sumerians to explain why we have failed to teach around a third of the population to read. She offers a fascinating analysis of how writing developed to argue that speech sounds are the basic unit for all writing systems, and therefore for learning to read. We have, she says, 5,500 years of evidence to prove it. She offers research from two recent American programmes: Lindamood Auditory Discrimination in Depth and Phono-GraphixTM, both based on phonological awareness, to argue that everyone, including adults and children who have failed other remedial programmes, can be taught to read. Indeed, 97 per cent of poor readers can improve in 12 hours or less.

McGuinness's book is compellingly written, but do we believe her? It requires something of a leap of faith to accept that, after 400 years of failing to teach the English writing system, we are on the threshold of the Holy Grail. Can the solution to the nation's literacy problems really be that simple? Not long ago, Reading Recovery was supposed to provide the solution to illiteracy and ministers were pouring money into it. Can all teachers be trained to use the system outlined by McGuinness? To the lay eye, it looks dauntingly complicated.

Yet, the case for using speech sounds as a basis for teaching reading is powerfully made. Her story of Sumerian, Egyptian and Chinese efforts to translate talk into writing unfolds as enthrallingly as a detective novel. They help her to the conclusion that no child should ever be required to memorise whole words by sight because writing systems based on whole words don't work: people's capacity for memorising words is limited. Her argument that everything we have tried so far has failed is unassailable. Reading standards in this country have not risen since the war, despite educational see-sawing from one strategy to another.

Perhaps that is not surprising since they pre-date any scientific understanding of reading. This, by contrast, is a comprehensive scientific approach to literacy, yet some of the research into earlier remedial reading programmes looks dubious. The findings of Australian and American scientists showing that dyslexic children just have difficulty ungluing sounds are persuasive.

But American research into the two new reading programmes that are offered as a solution looks convincing enough. For a government committed to transforming reading standards over the next five years, it must surely be worth a try.