Dorothy Bishop · Michael Rosen

Rosen unwound

I see that Michael Rosen has been ‘sounding off’, if you’ll excuse the pun, on his blog about phonics again.

It is really rather baffling that someone who is clearly very inquisitive and who actually listens carefully to the people he interviews and talks to on his Radio 4 programme ‘Word of Mouth’ should be so unwilling to listen to the arguments put forward in favour of phonics.
He claims that ‘using phonics in the teaching of reading is fine’, only to undercut the statement with a long series of objections. What always emerges from those objections is that he doesn’t understand the relationship between the sounds of the language and writing system. He appears to believe that we have a part phonic, part logographic system which necessitates the learning of (presumably) large numbers of (undecodable?) words.
Of course, all words are comprised of sounds and all sounds have spellings assigned to them. What Mr Rosen either doesn’t seem to appreciate or doesn’t believe is that, however complex the system, it can be taught successfully if it is taught from simple to complex.
To teach from simple to complex, teachers must understand how the alphabet code works. They need to know that there are forty-four sounds, as well as how those sounds are spelt in English words. Admittedly, when we arrive in the realms of highly specialised and esoteric language, this can be difficult even for the best spellers. However, for the most part, the task of teaching reading and writing is easily within the capabilities of a literate teacher.
Knowing how the code is structured also requires that teachers understand it conceptually. The need to impart to their pupils that sounds can be spelled with more than one letter, that sounds are spelled in more than one way, and that most spellings can represent more than one sound.
Mr Rosen understands some of this. What he doesn’t understand is what synthetic phonics teaches and why. For instance, he says that ‘many words in English cannot be entirely decoded using synthetic phonic methods’ and gives as an example the word ‘wound’. Any teacher of SP worth their salt knows that the spelling can represent the sounds ‘oo’ (in ‘soup’), ‘u’ (in ‘trouble’), ‘ow’ (in ‘sound’), and ‘oe’ (in ‘mould’). Mr Rosen knows this too. Where his understanding of what phonics teaching is fails him is in his seeming to assume that phonics teaching is conducted in a hermetically sealed capsule, when, of course, nothing could be further from the truth. It takes place in a context, the context of everyday language usage.
Good quality phonics teaching always takes places within the context of words, sentences and texts in English. That is its purpose! So, if a child is reading the sentence, ‘Peter wound up his watch last night’ and they read the word ‘wound’ as ‘woond’, or ‘wund’, or ‘woend’, their brain should be telling them that none of those alternatives makes sense in the context of that particular sentence! When they try the ‘ow’ alternative, it suddenly makes perfect sense. However, the child needs to know the alternatives in order to try them.
Teaching this might sound more difficult than simply trying to get the child to recognise by ‘sight’ the whole word and that is why whole word approaches seem so attractive, ‘promising  everything and delivering nothing’, in the words of Diane McGuinness. The problem is that it is impossible to learn the million or words there are in existence in English. On the other hand, it is entirely possible to teach the, roughly, 175 common spellings for the forty-four sounds, to teach the requisite skills of blending, segmenting and phoneme manipulation, and to teach how the writing system works in relation to the sounds of the language conceptually.
In teaching the above, the knowledge and skills can be generalised and applied to any word in the language. Spellings are symbolic tools that have a deliberate as opposed to a spontaneous character in the teaching process. They can be systematically taught because they are systematically organised around the forty-four sounds in the language, which give the system its basis. And, they can be generalised and applied.
What we have here is a functional linguistic system of the sort Halliday (much admired by Mr Rosen, it seems) would have approved.
There’s more to disagree with in Mr Rosen’s posting, not least of which is his aversion to the government’s Y1 phonics screening check. He appears not to be aware that the check isn’t punitive; it’s being conducted to find out something, that something being whether a child can decode successfully or not. If they can’t, someone needs to teach them because if a child can’t decode, they can’t read. Which brings us back to a point Dorothy Bishop made recently in a blog posting about Rosen’s views: ‘you can’t read for meaning if you can’t decode the words’.