I have never seen anyone deny that synthetic phonics is useful to learner readers.
I have seen some who believe it is enough, and many more who believe it needs to be combined with comprehension work, or contextual work, or other approaches to decoding – a pretty wide range of views.
But I have seen a few who, when faced with stories of problems in schools using a purely SSP approach, deny that the problems described are even possible.
There are two main reasons given when denying the possibility of problems:
- The person telling the story has misinterpreted what they have seen.
- The person telling the story has ideological reasons for disliking phonics, and their story cannot therefore be trusted.
It is possible that in some cases either or both of these things may be true. But I have heard stories from people I trust (including educational professionals I know personally), and the stories from other people sound similar enough for me to wonder whether there is a pattern of issues which needs to be confronted.
If something is potentially going wrong with SSP in some schools, denial is definitely not a sensible response.
So I’m asking, here, for specific anecdotes about any problems which you have come across, and any specific benefits you have seen, either as a parent/carer or as a teacher/teaching assistant, and which you would be happy to share and discuss via this blog.
NB I know SSP has all sorts of good characteristics, and others that are up for debate. I’m not asking for theoretical reasons why SSP might go wrong, or for reasons why it cannot go wrong, either: there’s a quite enough of that about as it is.
Miscellaneous Witterings 
This is one problem I have come across:
Children get stuck at the sounding out stage. They sound out every word even when the word is known to them and even to the extent of sounding out incorrectly but still identifying the word correctly. Sometimes they ‘read’ a whole phrase by sounding out each letter or GPC but not stopping to blend. Mostly these children don’t know what they have ‘read’ although an interesting subgroup sometimes paraphrase it. Also interesting sometimes children sound out irregular words using the basic code and then say the correct word “/a/ /r/ /e/, are”.
Of course this problem is not insurmountable but nevertheless it is a barrier which has to be dealt with and it seems to arise from the child perceiving the reading task as being about identifying the sounds, not the words and not the meaning, as if a conditioned reflex has set in to make sounds when presented with printed words.
Thanks for this: it rings true from my own experience (not that I’d otherwise doubt you – just that some seem to doubt all these anecdotes!). I had to have a little chat with one of the boys at the point where he was starting to be able to read words without sounding first. He was sounding out nearly every time but my gut feeling was that he didn’t always need to.
So I said to him that sounding out was about finding the word, and he could also sound it in his head, or not do it at all, if he could read the word without it. After that he read words he could read without sounding out, and sounded any that he couldn’t – sometimes out loud, sometimes in his head or just under his breath.
At the moment he rarely needs to sound a word, but when he needs to, he does so (in these cases he still often needs support though, because the unfamiliar words tend to be either multisyllabic or have irregular orthography).
I’ve seen others mention the sounding-but-not-blending problem. If this is occurring, then there could be all sorts of reasons why, but it’s pretty clear that SSP doesn’t present that as the goal.
So I wonder why it’s happening? Blending *is* a very particular skill, and I think I’ve said before that I think moving from a sequence of phonemes, however closely packed, to the actual sound-flow of a word, is a cognitive leap that not all will find easy. I wouldn’t expect it to come naturally – which would mean blending itself would need to be modelled.
I used to model blending a lot with the boys, and if they took time to hear a word, I’d go back over the sounding and blending to model again what they’d done. I did the same with phrases when the individual words had taken time to work out – in the spirit of ‘this is what you’ve just done!’, and as a lead in to the next phrase.
It doesn’t make sense to me to let a child leave a word just sounded and not blended, without modelling the blend for them if they can’t do it themselves. I think most of us would struggle to get the meaning if we heard words read out as separate phonemes like that.
I don’t have experience of what to do to help a child who simply cannot learn to blend – and I don’t know how prevalent this is as an issue. It would be really interesting to know, though.
As I said, it is not insurmountable as a problem and one to one tutoring will always help. I noticed this problem when I was supporting readers as a volunteer helper. I would worry about the children who do not have the benefit of the sort of attention
I was able to give and whose reading lessons consisted of group work on sounding and blending with teachers adamant that they should not guess words.
Yes, I would worry about that, too. But that comes back to the apparent ban on ‘context’ which turns out not to be a ban at all… If there’s no reason why context can’t be used to help a child ‘hear’ a word in order to understand it (even, sometimes, to blend it completely), there ought not to be any reason why such support couldn’t be given in group work as well. Are people worried they’d be doing the wrong thing by referring to the context, do you think?
In SSP use of context in decoding is acceptable if it is to help decide between homographs (read/read, wind/wind etc), but in other circumstances is not encouraged as a tool to support decoding.
I don’t think it is the ban on using context which is the major problem here. I think it is the belief that reading is decoding, at this stage at least, which leads to teachers and therefore children checking and counter-checking the letter/sound correspondences and blending them without thinking much about meaning. After all SP fans say that children can identify the meaning of the words without difficulty; they really don’t see comprehension as being a problem at this stage. Teachers hearing this might be inclined therefore to not worry too much about the comprehension side of things. Whereas, perhaps comprehension becomes more of a problem with the SP process, which stands as a barrier to hearing and understanding the complete word, phrase, sentence.
This is why Debbie Hepplewhite’s concept of ‘tweaking’ is so interesting to me. As I’ve wondered before, how can ‘tweaking’ occur in the absence of meaning? If the tweaking is necessarily taking place with reference to latent vocabulary, then meaning has been accessed in some form, and there is potential for the meaning of that one word to connect in the child’s mind with the meanings of previous words. It might not happen, of course, but it could.
I’m finding it hard to imagine that, English orthography being what it is, children could sound and blend *plausibly* without connecting with meaning. Or is it, as in your sounding-only example, possible that teachers are sometimes happy for children to approximate the pronunciation without achieving a truly plausible-sounding word? That in this case, too, the process of ‘reading’ is only partially completed?
Does the structured nature of phonics schemes sometimes lead people to place a little too much trust in the system over their own involvement, perhaps? This seems to happen so easily in big institutions – people can feel a bit alienated by a lack of personal choice, and end up abdicating responsibility to the system itself – even when the effectiveness of the system actually requires personal engagement from the people in it.
I think the idea of ‘tweaking’ is a solution to the problem that a child might not get the exact pronunciation through SP and would need to use some judgement to identify the exact word ie arrive at something familiar. Is it about using context or about the child trying to fit the word to their oral vocabulary? It seems to me that this might or might not be about meaning. I’m really not sure if SP fans want children to decode the word by thinking about the meaning or simply about whether it sounds familiar.
SP fans also make quite a lot of the idea that every new word (new in its written form) is, in effect, a nonword, for young readers.
So what are we to conclude from this? Either children believe that they will know the words they are reading and therefore ‘tweak’ a word that they don’t recognise into one that they do, or they are satisfied to think that the word might not be one they don’t know and stick with their first phonic attempt.
It all seems a bit confused.
Children can sound and blend *plausibly* without thinking about meaning – this is what they are expected to do in the phonics check. But they will not always be able to sound and blend *correctly* without thinking about meaning and the context. The worry is that SP actually encourages children to sound and blend plausibly rather than correctly, because of its banning of context and its emphasis on decoding by primarily using phonic knowledge.
Sorry, in the third paragraph there it should have read, “Either children believe that they will know the words they are reading and therefore ‘tweak’ a word that they don’t recognise into one that they do, or they are satisfied to think that the word *might be one they don’t know* and stick with their first phonic attempt”.
Thanks – this is why I’ve spent time trying to understand what’s actually going on at the tweaking stage. Without proper research, though, it’s impossible to know – which is frustrating. Just to clarify, I meant that the sound of the word should be ‘plausible’ to the adult listener (ie that it would sound to the adult as if the child has read and understood the word well enough to get the pronunciation right), not that the sounding/blending was plausible merely to the child, or that it was merely phonetically plausible. I know that the PSC seems to use plausibility in roughly that latter sense – but what’s plausible to a child or as a *possible* word is not necessarily what’s plausible to an adult in relation to the lexis, so I’m using ‘plausible’ to mean something slightly different: plausibility in relation to individual words in the actual lexis, that is, it is credible that the child has read well enough to read the word as itself, not as a phonetic approximation.
So I’ll try and stick to ‘credible’ in the future, to try and be clearer what I mean.
I’ve wondered elsewhere in this blog how children are supposed to know when to tweak – ie how do they know that there aren’t two words, one beGAN and one BEgan (or whatever). I do think that although in many cases the question doesn’t arise, it ought to arise often enough for the listening adult to have a pretty good idea from the child’s pronunciation as to whether the child understands the words they are reading.
I really don’t think that, without a transparent alphabet and beyond the CVC stage, children *can* pronounce all the words they read entirely *credibly* without reference to latent vocabulary. And the means by which they refer to their vocabulary for disambiguation *must* relate, although perhaps in a non-linear and intermittent way, to the context of the text they are reading.
If I were to create a proposition for testing, for research purposes, I’d posit that:
*Children who pronounce words credibly after sounding and blending are doing so with regular reference to their latent vocabulary and possibly also the textual context. Children who produce an approximation via plausible pronunciations which are not credible to an adult listener are doing so mostly or entirely without reference to vocabulary or context.*
Now, if only someone would pitch up with some funding…. 😉
I’m following your substantial pieces of the last few days, together with Nemocracy’s replies, with much interest. They’ve prompted reflections on two themes –I don’t pretend there’s anything new in them.
Like you, I’m still trying to to get clearer about some tricky issues to do with reading. They are important for the long-standing controversy over attempts to impose on teachers a particular approach to early reading that may be interpreted in a way that denies them professional flexibility in the classroom.
Theme One. The Simple View of Reading
Your reflections show up once more that the so-called “Simple view of Reading” should, in fact be dubbed “The Over-Simplistic view of Reading”. SVR says that reading (by which I mean, of course the process that results in obtaining the meaning of the relevant text) is the product of Listening Comprehension and Decoding. LC relates to those words which we hear and understand as real words. SVR is often strongly defended by synthetic phonics enthusiasts.
But what emerges when we think through examples, as you have, is that Listening Comprehension and decoding interact, sometimes in complex and ‘iterative’ ways for reading to occur. Often, pronunciation itself cannot be fully established without drawing on meaning. Decoding and Listening Comprehension are not conceptually independent of each other in the way that the Simple View of Reading seems to imply.
Theme Two. Reversibility.
This is the thought that decoding and segmenting are reversible processes. You will find this claim made explicitly in the Rose Report and in some SP materials. It looks as though the Government evaluation of ‘Sounds-Write’ are endorsing this when they say ‘Similarly the reversible processes of segmenting and blending were well explained.’ This observation is made alongside the relevant criterion, which reads “how words can be segmented into their constituent phonemes for spelling and that this is the reverse of blending phonemes to read words.” But we need to be very careful about how the reversibility claim should be interpreted.
One way of understanding it is that segmenting and reversing are in a 1-1 relationship in any given word. Now I think this works for those special pseudo-words which only admit of one pronunciation. I ‘decode’ ‘vit’ to produce a sound, and could split it up again into its constituent letter sounds.
When dealing with ‘real words’, it often isn’t a 1-1 relationship.
The vast number of homophones and heteronyms are one of the reasons for this.
I decode ‘wind’, and have a choice of a result that rhymes with ‘tinned’ or with ‘lined’. I might use the context of the rest of the sentence to choose the sound that goes with the real word of interest to weather forecasts. But if I start with the real word ‘wind’ of interest to weather forecasters and ‘segment’ it for the purposes of spelling, I am not simply reversing what happened with the decoding. For I’m now starting with a real word sound that is just one of the two choices I had when I was decoding ‘wind’. And another thing. Suppose I hear a word that rhymes with ‘lined’ and am required to segment it. Well -its spelling could be ‘whined’ rather than ‘wind’. This often happens, doesn’t it.
I decode ‘his’, and has been pointed out, I need context to rule against a pronunciation that rhymes with ‘hiss’, and to opt instead for one that rhymes with ‘fizz’. When a child hears the word ‘his’ and segments it for the purpose of spelling, again this is not simply reversing the decoding that helped but did not finally settle the pronunciation of the real word ‘his’.
But we have to look at each case on its merits. Decoding ‘bus’ should rhyme with ‘fuss’ rather than ‘buzz’ – again, context would be needed to decide on the appropriate pronunciation. Once we have this, that’s fine. I can hear the sound for the real word ‘bus’ and simply segment it into constituent sounds.
Anyhow, let’s have a few more cases that challenge straightforward reversibility. I decode ‘prints’. My resulting sound matches what we hear when someone says ‘prints’ or ‘prince’. So you can’t just shout ‘reversibility’ to obtain the original sound constituents and thus the spelling, can you. Similar points can be made about ‘dents’ (because of ‘dense’), ‘mints’ (because of ‘mince’), ‘bass’ (because when it’s the singer we also have ‘base’), and so on.
Then there’s accent. A southerner decodes ‘fast’. The middle sound in her result is similar to the middle sound in ‘farm’. A northerner decodes ‘fast’. The middle sound in her result is likely to resemble the middle sound in ‘bat’. It’s reasonable to think about reversibility here, perhaps, but a northerner’s reversibility differs from a southerner’s. That’s an interesting situation to deal with in a class with a range of accents, or where the teacher’s accent differs from at least some of her pupils.
I think it’s really important to be sensitive to the sound-worlds that a particular child may be navigating, especially when parents, teacher and child have different accents or first languages. One child I know adapts his speech depending on which parent he is speaking to: for instance, he says ‘bahth’ to his southern mother, but ‘bath’ to his northern father. At one point he had a Scottish teacher, so although he read in his own accent his standalone phonics sounds were Scots.
To me this is evidence of the potential for children to hear with great subtlety, even if they are not always aware of what it is they are doing; but it also suggests that some – many – children are navigating a more complex set of options than 40+ clearly-defined phonemes (which of course are themselves capable of division into finer distinctions in some cases).
When it comes to spelling, such children may have two or three possible pronunciations of a word in mind, and as you say, some of those pronunciations may be more ‘reversible’ than others.
My cousin, who is big enough and old enough to cope with being mentioned in a blog, had a father from Michigan and a mother from England. They lived in Texas for most of his life, and one of his defining memories of school was being asked to spell ‘farm’ in a test. The teacher read out the words, and he heard her say a word which sounded to him like ‘form’. He was the sort of child who got things right in tests, and the class and teacher’s response to him getting something ‘wrong’ has stayed with him.
Even apparently tiny moments like this can mean a lot to a child, given how desperately so many little ones want to get things ‘right’. I worry that if a system is not sensitive to these complexities, and teachers are not given support in adapting the system to their own classes’ needs – or are even prevented from doing so at all – this could have a negative effect on some children’s confidence.
It would be really interesting to hear about the ways in which teachers deal with EAL and accent issues in SSP approaches to reading and spelling.
A further thought in relation to privileging sound over other possible cues is the effect on children with hearing or speech difficulties – including temporary issues such as glue ear.
I could see that, again, sensitivity is important (a focus on sounding might actually lead to earlier diagnoses of problems), but also flexibility – children with either hearing or speech difficulties (or both) would not always be best helped by using their own sounding out of words as the route to comprehension.
Of course, teachers, including those who have built the SSP systems, will be aware of these possibilities, but I thought it was important to mention them here, to add to the list of exceptions to the standard SSP approach.
One question might be: where does a list of exceptions begin to outweigh the perceived ‘norm’? The effects of geographical mobility are such that I imagine there are few schools with *no* issues to do with variant pronunciations; and there must be few schools with *no* children who have hearing or speech issues.
To what extent will the intensive standardization of approach from Sept 2014 make it easier, or harder, to diagnose and support children with these issues?
It can be said that children with hearing issues are served well by phonics. The continuously variable sound of normal speech is broken down into phonemes for these children and made *clearer* for them.
A member of my family, who was a late talker, came on in leaps and bounds when he was 4 years old and we began to teach him to read and write. It was as if breaking down the words into letters to write them and sounds to read them, suddenly made sense to him. Late talking was no longer a problem, and he knew all the letter sounds and simple words before starting school.
It is difficult to know if this was a coincidence with other cognitive developments going at the same time, and SSP, as a method, was not used. He certainly learnt to recognise the whole written words, cat, dog, I, am, etc, before we could break these down into separate letters for him. He needed to know the meaning of these words as well as the sounds of the letters before he could attempt to blend the separate sounds back to a real word.
Thankfully he was too old by 2012 to take the phonic check, because trying to say some sounds and blend them together without them meaning something to him, would have been a futile experience. He would have looked at the teacher with an “are you mad, am I dumb” sort of stare, switched off, lost his confidence and decided not to try again.
Thank you for this comment – it’s so useful to have different experiences and perspectives. Interesting that you see reading and writing as having been helpful to this particular child but feel that the nonsense words in the check would have seemed pointless.
I think your instinct about clamming up when faced with something new (especially when it’s possibly difficult for a child to see the point of it) might well apply to many children. Perhaps it’s a fear of this kind of response in a child as much as the specific skills required by the test which is leading to children having to practice nonsense words in class. I can completely see the logic of this! (But I also think; what a shame, when they could be reading stories…)
I am getting tired of this debate. Our school has been transformed by phonics teaching. Our kids can read and over the past ten years the numbers falling below the 10th percentile in tests has fallen by over 70%. The vast majority of those now below the 10th percentile are in the older classes, where phonics teaching was less ingrained in the school culture, and new children who have come to the school.
The debate seems to be older than most of the people having it, so I’m not surprised you’re tired of it. If you were able to find time to answer/comment on some of the questions in my first post on this blog, I think it would be really valuable in trying to bring the debate, eventually, to some kind of conclusion. Thank you! https://miscellaneouswitterings.wordpress.com/2014/06/17/reading-methods-defining-terms/
It’s great to hear success stories, and very tempting to think that these successes can be transferred to every school and ultimately every child in the land. This is what the government seems to want to believe, and it is beguiling. However, when we look at the research studies used to support SSP they are not fully backed up by valid research methodology, and more often than not they are carried out by individuals already sold on SSP.
We are in a situation now where teachers’ judgement is not being trusted and teacher expertise is being sidetracked in favour of the expertise of programme authors (see the voices listened to by the Rose Report). So
Next to schools which might choose, with good reason, to implement and follow carefully an SSP programme, there will be schools which have other priorities and might, with good reason, choose something different, or a different configuration of strategies including SSP, but are being denied any support in doing that under present policy. Government policy presents SSP as a method, whereas more flexible policy could present it as a strategy.
I’m really interested in the specifics of the success stories: what worked, what didn’t? How was change managed? How were changes integrated with/adapted to SEN & EAL support? What changed in terms of wider classroom practice (for instance, what knock-on effects did SSP have on other literacy work, but also was there anything else that changed)? How does that flow through the school – do teachers in the upper school find that they’re picking up on different issues post-implementation, and if so what are they? How is the PSC handled? How were parents informed of/engaged with the changes? That sort of thing….