Outstanding overview addresses nonsense in Davis's paper

[chew8]

Isn't he really just saying that it is perfectly possible to be able to 'say' words in an unfamiliar language by knowing the 'code' even without knowing what they mean? This seems to me to be utterly self evident and, at the same time, completely irrelevant to the debate about teaching reading. After all, one could equally well learn a number of foreign words as 'wholes' and also not know what they mean!

Seidenberg’s Bar Mitzvah example is obviously an extreme one – there are particular cultural reasons why Jewish boys are taught to decode Hebrew even if they don’t understand it. Such reasons don’t apply to ‘normal’ learning-to-decode situations, however. Children normally know a lot about the spoken form of the language before being taught to decode, so they are in a much better position to understand what they decode than are children learning to decode Hebrew for Bar Mitzvah purposes. In very transparent orthographies (e.g. Finnish) they can probably decode many words that they don’t yet understand orally, but they can also decode a lot of words that they do understand orally, and if their school reading material consists largely of such words, there is no reason why they shouldn’t understand it.

So in spite of his extreme Bar Mitzvah example, I don’t think that what Seidenberg is saying is ‘completely irrelevant to the debate about the teaching of reading’, even if the irrelevance charge can be levelled at some of the other people who talk about children ‘barking at print’.

Jenny C.

[john walker]

Well, I have to say that I taught my youngest daughter how to read in German before she knew what the words meant. This was in her first year of grammar school and she'd only just begun the subject. Of course the pron. became more and more important as she continued to study - right up to GCSE, which she took last year.

[chew8]

Yes, people can obviously learn to read in a language before they know what any of the words mean, but it's surely more normal to learn meanings as one learns to read the words. At least, that was what happened with all the foreign languages I've learnt, including Afrikaans, which it was compulsory for English-speaking children in South Africa to start learning in the third year of primary school. It was also what happened when I did a one-year introductory course in classical Greek at university. We were taught the Greek alphabet (letter-names and sounds) and were then expected to learn a lot of vocab. from printed material.

I think, though, that the normal situation envisaged by Seidenberg is one in which children are learning to read in their own language and already understand the spoken form of that language.

Jenny C.

[Toots]

Good decoding is an easy skill to acquire in most other alphabetically-written languages because the first sound which children are taught to associate with each grapheme is usually the sound which will apply in all words they encounter from then on. This is clearly not the case in English, but isn't the sensible solution to take extra time and care over the teaching of decoding in English? That could ensure far less prevalence of the problem that ‘failing to decode accurately stands in the way of reading for meaning’.

What does this extra time and care consist of? There are certain problems in English orthography which will not be solved by extra emphasis on phonics for longer. It's not a case of there being more to learn than in shallow orthographies but a case of there being some aspects that cannot be learnt through phonics alone. The differences between the orthographies is not a simple quantitative one, where there is more of the same to learn, but a qualitative one, where there are different things to learn about the language.

[JIM CURRAN]

As Mark Seidenberg indicates reading problems are not easily explained by use of a reductionist argument and I agree. I feel however that we must acknowledge the huge impact poverty has in all countries and more especially in the US and the UK where levels of poverty and inequality are so high. This is not to deny how important good initial instruction is.

In the 2009 Pisa scores if we control for poverty,schools in America where less than 10% of the children were on Free or reduced school meals had reading test scores that put them in a clear second place behind only Shanghai and if we compared American schools with schools in other countries with similar Free school meals intake, this time between 10- 24.9% American students came third in reading behind only Shanghai and Hong Kong. It would appear that in America and there’s no reason to believe that the case is any different here, educators seem to be very good at educating a certain section of the school population but as someone once said “ poverty has introduced a flaw into the system that we haven’t been able to deal with”

There are 22,000 children living in homeless shelters in New York. The percentage of children living in poverty in the US is over 23% the second highest of 34 industrialized countries. In Finland the % of children living in poverty is 4%.

[john walker]

Toots said:

What does this extra time and care consist of? There are certain problems in English orthography which will not be solved by extra emphasis on phonics for longer. It's not a case of there being more to learn than in shallow orthographies but a case of there being some aspects that cannot be learnt through phonics alone. The differences between the orthographies is not a simple quantitative one, where there is more of the same to learn, but a qualitative one, where there are different things to learn about the language

This is simply not the case! Let's compare for a moment Spanish and English. The former can be said to have a shallow orthography because there are (and experts sometimes disagree on exactly how many sounds there are in the language for a variety of reasons which I won't go into here) only between around twenty-two and twenty-four sounds in the language. These sounds can be represented (spelled) by around thirty plus spellings. This is a simple system and it is easy to learn, which is why many schools in Spain don't bother to start teaching reading and writing until the second year of schooling, and most children learn is very quickly. On the other hand, as we know, English orthography is much more complex. For this reason it takes longer to teach. So, the difference is both quantitative (i.e. there is more to learn - around 175 common spellings for the 44 sounds, as well as many more [than in Spanish] two-letter spellings, as well as three- and four-letter spellings) and qualitative in that the conceptual structure of the way in which the sounds of the English language relate to the writing or spelling system needs careful instruction.

[john walker]

Of course, Jenny, you're quite right to say that

it's surely more normal to learn meanings as one learns to read the words.

This is the issue at stake when we're talking about comprehension in the way that Davis, Rosen et al mean.
However, their perspective on this is a huge red herring, in my opinion. The number of words a child knows by the age of five or thereabouts vastly exceeds the words the child is likely to be introduced to through the medium of phonics teaching. Of course, what 'knowledge' of words meanings constitutes can be problematical inasmuch as definitions differ: is 'going' to be regarded as a separate word from 'go'; are we talking about productive language or receptive language; are there qualitative difference in the kinds of definitions children give for a word when they are seven years old and when they are eleven years old.
Chall puts the average vocabulary of a child in the USA at around 4,000 words by the time they enter Grade 1, though, as we know, children are different and come from different backgrounds, with some children's vocabularies exceeding others by factors of ten or more. Nevertheless, the fact that the vast majority of children, by age four or five, have such extensive vocabularies is why Stanovich can state that '[t]here is no research evidence indicating that decoding a word into a phonological form often takes place without meaning extraction, even in poor readers. To the contrary, a substantial body of evidence indicates that even for young children, word decoding automatically leads to semantic activation when the meaning of the word is actively established in memory'. It is also widely agreed that by the age of four or five years, 'most children give the impression of having assimilated at least three-quarters of all the grammar there is to learn' (Crystal).
The problem has always been one of decoding, and what good quality phonics teaching does is to teach children to decode and encode successfully, whether Davis and co agree or not.

[chew8]

Hi John -

What I said in my last message about learning meanings as we learn to read words applied specifically to learning to read foreign languages, as you had mentioned teaching your daughter to read German before she knew the word-meanings and this struck me as unusual (but see last paragraph below).

I’m not sure whether I understand exactly what you mean when you say that ‘this is the issue at stake when we’re talking about comprehension in the way that Davis, Rosen et al. mean’ and that ‘their perspective is a huge red herring’. Do you mean that the parallel that they have in mind is the one with children learning to read in a language that they don’t already know orally, in which case phonics may enable them to produce the spoken forms of the words but does not go the extra step of teaching them meanings? This is certainly possible (it happened when you taught your daughter to decode German), but I’ve always assumed that what Davis et al. had in mind was that children did already know the spoken words and their meanings in a conversation context, but in a reading context would often decode written word to spoken word without accessing the meaning. If this is what Davis et al. mean, then they are failing to take account of the Stanovich point which you mention. If, on the other hand, they are thinking in terms of children decoding words that they don’t already know orally, then they are assuming that this is a far larger part of learning to read in one’s native language than it actually is.

After posting my last message, I remembered a sort of in-between situation in my forays into foreign languages. My son is fluent in Russian. When I visited him while he was studying in Russia in the 1990s, he insisted that I learn what sound to say for each of the Cyrillic letter-shapes. From then on, as we were sight-seeing, he would make me decode words on shop-signs etc. At first I was irritated, thinking that it was pointless – I might be able to produce the spoken forms of the words, but it was laborious and I’d be none the wiser in terms of comprehension. I soon realised, however, that he was choosing words which were close to English words in sound and meaning (e.g ‘tabak’ was ‘tobacco’), so having decoded them I did get the meaning. There were relatively few words of this type, but these were the only ones he got me to decode. When we teach English-speaking children to read in English, the number of words whose meaning they already know is obviously far larger.

Jenny C.

[Toots]

Toots said:

What does this extra time and care consist of? There are certain problems in English orthography which will not be solved by extra emphasis on phonics for longer. It's not a case of there being more to learn than in shallow orthographies but a case of there being some aspects that cannot be learnt through phonics alone. The differences between the orthographies is not a simple quantitative one, where there is more of the same to learn, but a qualitative one, where there are different things to learn about the language

This is simply not the case! Let's compare for a moment Spanish and English. The former can be said to have a shallow orthography because there are (and experts sometimes disagree on exactly how many sounds there are in the language for a variety of reasons which I won't go into here) only between around twenty-two and twenty-four sounds in the language. These sounds can be represented (spelled) by around thirty plus spellings. This is a simple system and it is easy to learn, which is why many schools in Spain don't bother to start teaching reading and writing until the second year of schooling, and most children learn is very quickly. On the other hand, as we know, English orthography is much more complex. For this reason it takes longer to teach. So, the difference is both quantitative (i.e. there is more to learn - around 175 common spellings for the 44 sounds, as well as many more [than in Spanish] two-letter spellings, as well as three- and four-letter spellings) and qualitative in that the conceptual structure of the way in which the sounds of the English language relate to the writing or spelling system needs careful instruction.

Agreed, my point being that it is the qualitative difference which causes the problems. Quantitative differences can be met by extra teaching along the same lines, but qualitative differences need to be met by different teaching strategies. The careful instruction necessary would involve taking care to acknowledge and compensate for the limitations of phonics in the context of the English language.

[maizie]

The problem has always been one of decoding, and what good quality phonics teaching does is to teach children to decode and encode successfully, whether Davis and co agree or not.

But you must know, from years of teacher training and 'debating' the topic on the internet, that a very major problem we have is the firmly entrenched belief that children can decode and blend words with which they are familiar in their expressive or receptive language and not attach any meaning to them. This is usually presented as a major problem, the implicaton being that it affects most children who are learning to read.

Jenny has posted while I was writing this and what she says about Davis et al is true, but I do feel that their view is taken by most people to be 'the truth' (and I am sure that toots will agreee with them) and that it is a very difficult view to shift just by the application of reason!

[Susan Godsland]

I think a lot of misunderstanding about synthetic phonics is a result of the following (incorrect) belief, clearly expressed by Eddie Carron:

As long as a child is decoding text, the brain is preoccupied and cannot assimilate the intellectual content of text.

[maizie]

And that's another 'belief' that is near impossible to shift, Susan!

Mind you, it can be found in a number of research studies, so I suppose that there is thought to be 'scientific' evidence for it..

[chew8]

Re. the argument that 'As long as a child is decoding text, the brain is preoccupied and cannot assimilate the intellectual content of text': I'm prepared to accept that there is some truth in this, in the sense that at the stage at which lifting the words off the page is laborious, it's hard for learners to take in the meaning of more than a few words at a time.

BUT... I would say that

1. this is true regardless of the method of lifting the words off the page (phonic decoding, onset-rime analogy, look-and-say, use of cues from pictures and/or context);

2. practice at phonic decoding means that learners speed up quite quickly and the process becomes sub-conscious, at which point all conscious attention can be devoted to understanding meaning.

Jenny C.

[Debbie Hepplewhite]

http://teachingbattleground.wordpress.c ... al-debate/

Although another thread has been started with reference to Old Andrew's blog posting above, I thought it would be good to provide the link here also be because it makes reference to Davis's IMPACT paper which is the topic of this thread.

That said, I do think the arguments used by phonics denialists are simply terrible, so terrible, that I do wonder how comparison with, say, climate change denialism could be considered unfair. Two recent examples stick in my mind. The first was actually in a pamphlet published by The Philosophy of Education Society of Great Britain. Here’s the key argument:

[Davis writes:]

In this short book I am going to argue that research into the teaching of reading involves some fantasies. These take the form of imagining that specific teaching methods could be identified, and that their efficacy would be open to empirical investigation. I show that if any schools were actually implementing such strategies, the adults responsible would have abdicated their role as teachers. In reality, implementations of SP in any one school will not and should not precisely resemble those in other schools and in any case, current research into SP ‘effectiveness’ is not informed by a detailed blow by blow description of what actually happens in the classrooms concerned. Hence, it is never really made clear what the research is actually investigating. If teachers are actually teaching, there will be and should be nothing common to all SP programmes. The effects of drugs or fertilisilisers can, of course, be investigated using orthodox scientific methodologies, but we lack the equivalent here in terms of teaching approaches.

Now, the limits of scientific methods to isolate and evaluate what happens in the classroom is a real issue. I’m certainly sceptical about a lot of education research for that reason. However, the claim that we could never, even in theory, objectively evaluate a teaching method is as extreme a denial of science as anything you will hear from homeopaths or creationists (who are also often prone to claim that science cannot hope judge their claims). The claim that all the research in an entire field (not just the hundreds of studies on phonics, because this argument applies equally to all teaching methods) is a particularly extreme one. It entails that all those who have conducted empirical research in teaching methods were mistaken, and all those who found statistically significant results were deluded.

[john walker]

Hi Jenny,
You could say that teaching child who is just embarking on learning German formally in school how to read each sound-spelling correspondence in the words she was learning was just such a 'foray'. The huge advantage here is that so many German words already correspond pretty closely to their English 'equivalents'.
Still, the point I was making, and I shouldn't have continued to muddy the waters by referring to the teaching of an L2, was that when young children are learning to read and spell for the first time in their L1, they already have within their spoken repertoire a (comparatively to what they are able to read) very large repository of words they already comprehend. So, the point made by Stanovich was really the one that Davis doesn't seem to take heed of.
On the other point you took up about decoding and comprehending, I agree with you on both counts. On the second, though, I'd add that Hatte and Yates in their recently published Visible Learning and the Science of How We Learn, make explicit exactly why learning to decode and trying to comprehend simultaneously is very difficult: the cognitive load is too great, which is why multitasking is not really possible unless, as you suggest, one of the two things one is doing is being done automatically. To overcome this, we phonics practitioners present our material, a limited number of items to be learned, in short steps in the form of worked examples, comprising part of an extending schema, and follow it with lots of repetition.