What happens when we 'let the wrong one in'.

Kenneth Goodman and Frank Smith’s whole language theory of reading, whereby decoding and word recognition are eschewed for the pursuit of meaning with the implication that learning to read is akin to learning to speak, still pervades the world of reading instruction. California still struggles to unpeel its vicelike grip; balanced literacy (with its attempted and fatal compromise of contradictory approaches) maintains considerable traction and Marie Clay’s Reading Recovery programme (recovering from her whole language instructional model with more of the same…just much more) has condemned generations of struggling New Zealand readers to semi-literacy (more here).

In England, Goodman, Smith and Clay’s theories were never imported as a wholesale approach but snuck in through a side door and ended up being integrated into the Searchlights model of instruction as part of the National Literacy Strategy (DfEE, 1998) . That side door was opened, and the spectre was ushered in by one of the foremost educational academics of the day: Emeritus Reader in Education at the Institute of Education, Margaret Meek.

Meek’s (1982) book, ‘Learning to Read’, although written for parents, had a significant influence on the teaching profession. It espoused her version of the whole language theories and maintained from the outset that, ‘right from the start, learners have to behave as if they mean to become readers’ (1982: 15) and throughout encourages Goodman’s (1970) guessing technique as well as introducing the concept of ‘readerly behaviours’. The emergent reader is encouraged to say aloud what they think is written and approximate text by using the pictures to help them guess the words. In this way, she maintained, the beginner teaches themselves to become more and more exact.

Meek believed that phonics instruction was ‘highly inefficient’ (1982/2012: 42) as English contained so many exceptions that it had drifted from its phonic and alphabetic constructs. She posited that the use of phonic rules constricted a child’s curiosity about words and therefore their ability to develop orthographic processing and thereby a love of reading and the opportunity to become literate. If a child guessed a word rather than sounded it out, she suggested that this was the individual child’s decision, should be encouraged as their preferred approach and teaching adapted to the individual’s style. The implicit link to ‘learning styles’ is evident. If a child were unable to guess the word, then the word should be read for them. The child should never be instructed in word recognition. Approximation of the word was perfectly acceptable. The teacher, she stated, should not adopt a devised system but respond to the individual child’s efforts and approaches and promote these. By avoiding any cognitive exertion by the child, the instruction avoided and sniff of drilling and promoted these ‘readerly behaviours’.

Quite how she came to these conclusions is difficult to fathom. Her book, ‘Learning to Read’ (1982), is written with such self-assurance and certitude that one would expect it to be grounded in significant research and contain refutation of Chall’s (1967) research which had concluded that ‘Most children…are taught to read by…a meaning emphasis method. Yet the research from 1912 to 1965 indicates that a code-emphasis method – ie. one that views beginning reading as essentially different from mature reading and emphasizes learning of the printed code for the spoken language – produces better results…’ (1967: 307). Furthermore, there was no response to ‘The Simple View of Reading’ (Gough and Tunmer, 1986) in later editions (more here). The only referenced academics in the book are, inevitably, Goodman and Smith.

However, Meek did carry out some very significant and influential research with a longitudinal study involving a group of secondary teachers and some struggling adolescent readers. ‘Achieving Literacy: Longitudinal studies of adolescents learning to read’ (1983) was seen as a triumph for the whole language movement. It is perhaps the greatest evidence of the catastrophic nature of the whole language approach and one of the saddest accounts of teacher failure (driven, it should be noted, by beneficence) of vulnerable pupils.

Each secondary teacher on the project selected a pupil whom they liked (this was to ensure student motivation but undermined the study from the outset) and whom they considered a poor reader. They worked with them on a one-to one basis for three years. The pupils, it was noted, could sound out words as an attack strategy but not recognise words. Phonic methods were explicitly stated as having failed the participant pupils and were thus avoided and demonised from the get-go (although teachers did feel it necessary to revert to initial phonemes during the project). Indeed, phonics was implied as the reason for the pupils’ struggles with reading and the cause of their dislike of reading. Meek (1983) clearly advanced her own agenda to influence the teachers and the approaches of Goodman (1970) and Smith (1975) were adopted and utilised in the study. The teachers, having read Goodman’s ‘Reading: A Psycholinguistic Guessing Game’ (1970), appear alto have experienced a damascene revelation and articulated their relief at their new understanding that pupils no longer had to rely on what was on the page. Official reading tests were eschewed as assessment and progress was measured on the adoption of ‘readerly behaviours’ (1983:111). For much of the time, pupils were read to by the teachers and the pupils did no reading. This even shocked Goodman on his visit to the project (presumably at the invitation of Meek), who advised that the students needed to do at least some reading. Teachers spent a whole year on pre-reading activities: talking with students and recording the conversations and reading Dr Seuss (the irony of Theodor Geisel penning the Cat in the Hat books after reading Flesch’s ‘Why Jonny Can’t Read is lost on them – more here) and Mr Men books (they noted the usefulness of the pictures for guessing) to the point that one student who is being cajoled into reading by praise, explodes in exasperation that he’s been doing this for a whole year, “and nothing has happened!” The teacher expresses guilt, but a colleague likens the failure to a doctor giving a patient a pill: “it isn’t your fault that the pills haven’t worked.”

Where the teachers did have some success, and almost suggest that they have stumbled across a silver bullet is through the students reading their own compositions and retellings of stories. These are dictated to the teacher (heaven forefend that the students should have to do any actual work) and then read to the teacher. The fluency of the reading is much improved, but the teachers note that this is because the student is aware of the words he has used and had scribed but they mention this as a benefit rather than an issue. Much is made of the student’s growing understanding of narrative but a colleague who teaches him is unconvinced that he can actually read. Perhaps the student states it most clearly, describing reading as a game concluding with, “I don’t have to read everything.”

And the outcome of three years of one-to-one whole language reading instruction with a teacher who likes you? Well, they all exhibited greater ‘readerly behaviours’. But could they read? One of the students eventually got the motorbike they had been dreaming of as a result of getting a job (arranged through one of the teachers). Another, it was noted, had begun to read for pleasure and purpose. The others? ‘We have no evidence about the others.’ (1983:231). No evidence? After three years of one-to-one instruction? And the conclusion? ‘We now know, in great detail, that inexperienced readers in secondary school who want to learn to read have to subject themselves to a particular kind of metaphysical distress.’ (1983: 231). It is perhaps ironic that a project that sought to promote reading for pleasure concludes that not being able to read causes distress. Perhaps that distress is caused by not being able to read the words on the page.


It is interesting that all of the students in the study were apparently able to decode and blend but were unable to recognise words instantly; even words they had previously encountered. This may suggest the phonological dyslexia advanced by Berninger and Richards (2002) who propose that it is the point at which orthographic (word) and phonological information become linked that is impaired leading to difficulties with fluent word recognition (more here). Whole language approaches would have been even more disastrous for these students.