To read or not to read: decoding Synthetic Phonics

In England, current government policy on children's reading is strongly prescriptive, insisting on the delivery of a pure and exclusive form of synthetic phonics, where letter sounds are learned and blended in order to ‘read’ text. A universally imposed phonics ‘check’ is taken by all five year olds and the results are widely reported. These policies are underpinned by the claim that research has shown systematic synthetic phonics to be the most effective way of teaching children to read. Andrew Davis argues that there is a basic problem with this claim. Whatever it is that empirical researchers take themselves to be doing when they investigate synthetic phonics, they are not investigating a specifiable method of teaching reading. This is for two reasons. First, there are no such things as specifiable methods of teaching. Teaching is a vastly complex human activity involving contextual and reactive practical judgments that are responsive to the myriad contingencies of classroom life. The idea that teachers might proceed by way of prescribed methods rather than practical judgments is simply a fantasy. Second, teaching children to correlate letter combinations with sounds, and to blend sounds into sequences, is not teaching them to read. Reading is a matter of grasping meaning conveyed by text. While sustained attention to letter-sound correspondences can be helpful to some novice readers, we should neither assume that it is helpful to all nor confuse mastery of such correspondences with the ability to read. Davis's challenge to government policy on the teaching of reading, and to the empirical research that supposedly underpins it, is timely, radical and compelling. The zeal with which synthetic phonics is championed by its advocates has been remarkably effective in pushing it to the top of the educational agenda; but we should not mistake zeal for warrant.