The DfE’s guidelines for buying phonics materials – some questions.

This document, which the DfE posted on Twitter for reference prior to a Q&A with primary teachers on #primedchat, is supposed to help schools to evaluate phonics teaching materials prior to purchase. As seems often to be the case, some very important aspects of reading have been missed out or confused – or perhaps lost in translation between those people who gave advice and the officials who drew up the document.

NB I am addressing this document specifically, rather than all objections to phonics! One big question which this doc does not touch on is the issue of lock-in, where schools will be tied to particular schemes which might become costly over time – or might disappear, leaving schools with the SSP equivalent of Betamax. Something to ponder….

It’s a document from 2010, but it’s worth looking at it now, I think, since the DfE apparently believes that it is still the best advice – or perhaps the only fully government-approved advice – around.

To be charitable, I’m going to assume that somewhere along the line things got muddled by accident, rather than that this document really is meant to say what it appears to say.

Working through the list of criteria for choosing a scheme, the first potential problem lies in the criteria relating to visual prompts. The document states that the chosen scheme should use:

A multi-sensory approach so that children learn variously from simultaneous visual, auditory and kinaesthetic activities which are designed to secure essential phonic knowledge and skills.
But that
Children should not be expected to use strategies such as whole-word recognition and/or cues from context, grammar, or pictures.

My feeling is that there is perhaps some confusion about the role of visual prompts in SSP, and this document does not do anything to clarify the situation. If schools are going to be guided in their choice of reading/phonics materials, it seems important to make sure that they can make the best choices possible in relation to the visual (including pictorial) aspects. It seems to me in some ways slightly arbitrary to allow visual prompts when learning discrete phonemes, but not when deploying those phonemes during reading. I think that at the very least, if this document is updated or replaced after May 2015, it should explain where the use of visual prompts is considered helpful, where it would not be, why this is, and also include some guidance as to where the distinctions might be expected to fall in practice. It might help clarify these aspects of the SSP approach, especially if the theoretical basis for the distinction (if there is one) were to be made explicit.
There also seems to be a certain lack of clarity in stating that the scheme should ensure

That children are taught high frequency words that do not conform completely to grapheme/phoneme correspondence rules

while also stating that children should

Practise by reading texts which are entirely decodable for them, so that they experience success and learn to rely on phonemic strategies.

My question here hinges on how one defines ‘decodable’ words. To what extent are high-frequency irregular words decodable by children with early-stage knowledge of the SSP-style alphabetic code? To what extent should these words be included in early reading books? If they are included in early reading books, will such books still count as ‘fully decodable’? Teachers who use phonics methods would ideally have available reading books which contain high-frequency words in an order specified by the scheme that they are using, so that children will come across words that they have already learned; but these words are not quite ‘decodable’ in a normal phonics sense, at least at the stages when children would be expected to read them for the first time, so I think a bit of clarification there would be helpful as well.

[Of course, there are various approaches to teaching high-frequency words. You can do it in various ways as groups of ‘sight’ words, where the whole word is recognized but not deconstructed; you can do it by deconstructing each HF word into its constituent grapheme-phoneme correspondences. Personally, as a lover of language and its history, I would always prefer that some form of the latter is used in the process of learning these words. I see the value of helpiing children understand that there is still the same code operating, just a less familiar part of it. However, given the low frequency in the wider language of some of the sounds in these high-frequency words, it’s obviously the case that for most children these words will be in effect, once learned, exception words which they will tend to recognize as discrete little groups of letter/sounds, because whereas in most phonics work children progress from constituent sounds to a wide range of examples, in these words the example and the constituent sounds are tied much more closely together.These words are mostly their own example. A child is much more likely to come across ‘have’ than any other word with this particular pronunciation for -ave. If they sound and blend it they are likely to sound it out to rhyme with ‘gave’, and realize – from context, including the grammar of the sentence – that it must be ‘have’. It seems pretty obvious that they will be doing this with reference to the specific and familiar whole word ‘have’, with all its constituent sounds, rather than sifting through their knowledge of rare GPCs and reapplying that rule instead.]

This brings me to the part of this document which I’m afraid I find simply baffling. Explanatory note no. 7 says that children should not use

Cues from context, grammar, or pictures.

Setting aside the various subtle roles that pictures might play (see here too – my thoughts on this are in process!), and the fact that no phonics practitioner that I have ever spoken to has ever seen context as inadmissible as long as sounding and blending comes first, I’d like to focus on the document’s specific exclusion of grammar.

I have various questions about this, which it might be useful for the DfE to think about before they draft any more phonics-related advice:

1. How can we, as adults listening to children read, know when they are using ‘grammar’ to help them and when they are not?

2. How is it possible to exclude grammar from the reading process?

3. If it is, how is it possible to do so without doing damage to the coherence of the text for the child, and the child’s understanding of grammar in the long term?

4. Why does it help a child to exclude grammar as a prompt (assuming this is possible, which to be honest I doubt).

5. What is the logic of excluding grammar from early reading, while putting a huge emphasis on grammatical terms and analysis elsewhere in the literacy curriculum?

Actually I’m expecting that as with other forms of context, SSP does not in fact, exclude the use of grammar at all, but expects it to operate as a secondary prompt in cases of disambiguation, and/or to help match the blended phonemes with the stored word in the child’s own vocabulary. My initial impression is that this is indeed the case.

So, in summary, this document seems to me to do much to hinder the introduction of phonics, since it implies a much more limited – and limiting – system than is actually used anywhere.

Some core aspects of how the DfE represents phonics teaching need to be rewritten to take account to the fact that far from excluding ‘context, grammar, or pictures’:

a) Visual prompts *are* used in SSP: clarify which sorts are considered helpful, and which a hindrance to children learning to read under this system? [In particular, explain why it is acceptable to have (eg) a picture of a book to help with learning the phoneme ‘oo’, but not to help with matching ‘b-oo-k’ to the word book and its meaning.] {Edited to add: even this parameter is too narrow for SSP use of pictures; there’s nothing wrong with relating the word ‘book’ to a picture of a book – simply that the sound/letter/aural vocabulary relationship comes first. See my further witterings here.}

b) Context should never excluded, but should be used in combination with, and *after*, the initial attempt at sounding and blending (for instance in the case of homographs or words which a child has not come across before in their reading).

c) Grammar (which is in fact a form of context) should also not be excluded, but, again, should be regarded as working in combination with sounding and blending, *after* the initial attempt to sound out and blend.

I do think that if the DfE were to be more accurate in its understanding and presentation of SSP methods, this would go a long way to bridging the divide that currently exists: anyone, I think, can see that although there are big differences from previous methods, there is actually no exclusion of context or grammar, just a more specific use of them; and the use of pictures, I think, needs further thought because pictures, for a child, fall into the ‘context’ category too, and the arguments for using them in a similar way are strong.