The reading wars began when the first research appeared showing that phonological awareness correlates with later reading ability (eg Chall 1967). (Phonological awareness is the ability to distinguish spoken phonemes by ear – to hear the constituent sounds of language. It correlates well with learning literacy in that children who are good at it usually go on to learn to read quicker and better than children who do not. The question is what, or how much, does that mean?) The debate has been much bedevilled by weak or absent definitions – for example, what is meant by reading? Is it simply word recognition, an easily measurable ‘barking at print’ concept, or is there more to it? Is phonological awareness merely correlated with reading ability (whatever that is), does it predict it, or does it actually cause it? Many commentators suggest the last; in other words that phonological awareness is an essential, causative prerequisite of reading. The research evidence does not support so strong a claim, however (see Castles and Coltheart 2004 for a review). There are many axes being ground and many personal and political agendas confusing the issues.

Many research articles have demonstrated the correlation between phonological awareness and subsequent reading ability (good accounts are given in Bradley and Bryant 1993, Goswami and Bryant 1990 and Stanovich 2000). However, correlation is only correlation and shows, of itself, nothing more than correlation. For example, wealth correlates well with health – the more money you have, in general, the healthier you are. The well-off tend to be much healthier and live longer than the poor. However, nobody suggests that this observed correlation means that money per se brings health – that cash perhaps radiates health rays so that the more of it you have the healthier you become. What the correlation indicates is neither more nor less than that there is a positive relationship between the two variables, between wealth and health. More of one tends to go with more of the other. Once we reject the idea that money directly causes health we are obliged to seek the reason, or reasons, for this real enough relationship elsewhere. All that money, or lack of it, must be affecting the lives of people in some way to produce such a strong correlation so consistently, but the proximal cause, or causes, of the correlation lie elsewhere, among the complexities of social realities which wealth (or poverty) indicate. The correlation is simply a signpost towards some interesting ideas; the signpost may be real but it is only a signpost, not the destination. We must do more digging, and thinking, if we are to find truth.

Indeed, while engaged in this sacred sceptical work we need to consider our direction of travel. Are we moving forwards, or round in a circle? Perhaps our ability to discriminate among the sounds of our language and our ability to read, write or spell it are but two aspects of the same thing? Perhaps, like Pooh and Piglet, we are simply trudging with increasing excitement round and round a single tree? The number of footprints we discern in the snow ahead of us rises fascinatingly for every circuit we make but, in reality, this signifies approximately nothing. ‘…the relationship between [phonological] awareness and reading performance has strong implications not of causality but of circularity’ (Uppstad & Tønnessen 2007 p. 162) and perhaps ‘It is therefore, in light of the philosophy of science, high time to shout ‘the emperor has no clothes!”’ (op cit p. 163).

And it is a similar story when it comes to phonological awareness and later reading skill. That there is a correlation few deny (though it may be less robust and indicate less than we think - see Paris 2005 and later in this chapter); children with good phonological skills tend to do better at learning literacy than those without. The debate is about what that correlation means. Many say that phonological awareness is the most powerful predictor of reading ability, and a real correlation does indeed make for a real predictor. Phonological skills do predict literacy performance, at least to a certain degree. However, many researchers point out that there may be, must be, other influences which more or less co-vary with phonological awareness. These other influences could be as important, or even very much more important, to reading acquisition than phonological awareness appears to be. Why does a child who exhibits good phonological skills do that? And once you have asked that question you begin to see that there will, of course, be a plethora of influences intertwined in the child’s history which will predict his phonological skills, but this plethora of influences will also predict his reading ability. For example, the child who has been talked to a great deal, and been encouraged to respond in clear language, will reach school with better phonological skills, but he will almost certainly also arrive with all the other pre-literacy (or even early literacy) skills and practices which make it likely that he will ‘take off’ quickly (e.g. Hart and Risley 1995). He will probably have had many years of highly individual educational experiences which will have left him with a range of skills and appetites which are bound to affect his interface with literacy once he reaches school. These experiences are mostly unknown, and unknowns are what experimenters call confounding variables - all those myriad and mysterious other influences on our results which will affect them more or less powerfully but in unknown, perhaps even absolutely unknowable, ways.