Monday, June 27, 2016

A word to the Wyse

Today I’m wondering if ‘Stand Up for Education’ has been newly created for people,  with the title of professor, to tell jokes about education, or whether they simply use it as a forum for displaying their ignorance about the process of teaching literacy to young children.

The latest post on SUFE is titled, ‘Phonics fanatics: politicians who think they know best’ and it’s been written by Dominic Wyse. Our Dominic has plenty of form when it comes to phonics. Basically, he’s part of a cabal which includes such people as Michael Rosen and Henrietta Dombey, phonics deniers all.
Phonics deniers avoiding angry parents

As with so many of these kinds of arguments, its kernel of distortion and misunderstanding is cloaked inside a patina of fact. Although it's a truism, Wyse informs us that phonics has been used for centuries (actually it’s millennia but never mind) to teach reading and writing in Latin. Of course, in this, he’s right! The Romans used phonic methods to teach children to read and write in Latin. I imagine that teaching basic literacy to young Roman children took about as much time as it takes to teach children to read and write in Spanish, both being almost entirely transparent: each sound in the respective languages being represented by one letter spellings. However, anything more complex than the simplicity of Latin and Spanish seems to send the professor into a tail spin of confusion.

He slips in the tendentious claim that the Chinese writing system is ‘based on pictures’. Why he lights on Chinese is really beyond me because the teaching of phonics really has nothing to do with Chinese or with pictures. In any case, Wyse is flat wrong! As Peter Daniels, an expert on the world’s writing systems, informs us, ‘Pictography is not writing because languages include many things that cannot be represented by pictures... It is thus necessary for a writing system to represent the sounds of language.’* One of the contributors to Daniels and Bright’s The World’s Writing Systems, and an expert of Chinese writing, Victor Mair, tells us that ‘a few philosophers still insist that the Chinese writing system is pictographic and “ideographic”, but their views have been effectively countered by empirical and historical evidence.’ So, not a good start after all, Dominic!

Wyse’s next move is to sow maximum befuddlement in the minds of those who have little specialised knowledge in this area: he plays the Randomised Controlled Test card. And, yes, he’s right: there is no strong RCT evidence that ‘any one form of systematic phonics is more effective than any other’. That is largely because the government is sitting on its hands and seems reluctant to initiate RCTs on a scale that would show beyond doubt whether phonics approaches work. The other problem is getting schools to agree to participate in an RCT – for fairly obvious reasons. For the record, any time an impartial academic wants to set one up, Sounds-Write will be there: any school, any time, any level of SES, any percentage of speakers of other languages.

Wyse next complaint against phonics teaching is that the SATs results haven't improved much over the period 2008 - 2015. It’s true and I know why! It's because a large majority of teachers are still using mixed methods, and because they haven't been trained to teach phonics properly and don't understand the relationship between the writing system in English and the sounds of the language. Some of the blame for this must be attributable to Wyse himself, who loses no opportunity for confusing the issue with disconcerting advice to teachers.

Are we surprised then that teachers' knowledge is so poor in this area? Not a bit! You only have to look at the errors made by Dominic Wyse in the piece to realise that if he, a professor at the prestigious Institute of Education, is so badly misinformed, how can we expect teachers to do better. Not only does he believe that Chinese is a pictographic system (It isn't!), he also believes that English isn't phonetic (It is!)? But aren't all words (no exceptions) comprised of sounds and haven't all sounds been assigned spellings at the point at which they entered the language? In addition, he doesn’t seem to know that the spelling < sh > is a digraph and not, as he indicates in the article a 'consonant blend’. It's in the detail that so much of his ignorance is revealed.

If a professor at the IoE has such a poor knowledge of what phonics is about and how it should be taught, we shouldn't be particularly mystified when teachers don't. That's why we need in this country a series of RCTs to demonstrate for once and for all that high quality phonics teaching all the way through primary school is essential to ensuring that all children are literate before they enter secondary school. If you want to put a stop to the constant sniping from these Wyse guys, you need to listen up, Mr Gibb!

Monday, May 30, 2016

The Oxford Roald Dahl Dictionary and the Scripps National Spelling Bee

As it’s half-term, I thought I’d draw readers attention to a couple of items you might have missed.

First up, the Oxford University Press have just published the Oxford Roald Dahl Dictionary. It’s been produced by Susan Rennie, a lecturer in English and Scots
language at the University of Glasgow, where she works on a wide and very interesting range of projects, which include a Scot's Thesaurus and work on the newly re-discovered Boswell's Dictionary of the Scots Language. You can read more about her work here and here.

The Dahl dictionary is a cornucopia of rich vocabulary, both real and made up by Dahl. Just about any word you can think of that has turned up in one of the Dahl books is contained in this dictionary, along with the sentence, a definition and the book in which it appeared.

And, for those of you who are sick and fed up with people trying to foist on you heaps of ‘alien’ words, look no further than the almost endless variety produced by Dahl’s inventive mind. Whether you’re a muggle-wump who likes mudburgers, a maidmashing meatdripper partial to mouldy muckleberries, or a nincompoop after a nishnobbler, you will be sure to tickle the interests of children and provide yourself with an endless array of grist to your phonics mill.

And, talking of strange words, you might be interested in the Scripps National Spelling Bee, which for third time in succession has resulted in a tie.

Reported on by USA Today, the winners this year were Nihar Jangar from Texas and Jairam Hathwar from New York. From an entry list of 285 competitors, the two contestants had fought their way through thirty-nine rounds of the competition to share the trophy and prize money of $45,000 in cash. Their winning words were: ‘feldenkrais’, which is apparently a somatic educational system designed by Moshé Feldenkrais; and, ‘gesellschaft’, which is German for ‘community and society’.

These days competitors are so well rehearsed in spelling words with Greek and Latin roots that the organisers of the event have introduced more obscure words that are less well embedded in English, such as Afrikaans, Irish Gaelic, Danish and even Mayan.

Believe it or not, Jairam’s favourite way of relaxing is to play golf with his dad. Both boys, whose parents are recent migrants from southern India, continue an astonishing run of success by Indian American entrants and both are aiming to become doctors.

If you have time to read more about the Spelling Bee, I've posted a number of pieces about it in the past, which you can select from here.

Friday, May 13, 2016

Linguistic v traditional phonics- an afterword

Thanks to reminder in a tweet from Maggie Downie, I’m adding a further point to my previous post ‘Linguistic v traditional phonics’.

As I’m always arguing, linguistic phonics gives primacy to the spoken language. The reason is because all children grow up learning the spoken language naturally, which is not the case with written language!

As I argued in the other posts, teaching children that words are comprised of the sounds in their language and that every one of those sounds has been assigned a spelling at some point in the past makes perfect psychological sense: “Oh that’s what this game is all about! Why didn’t someone tell me that before?” is the response I often get when teaching older pupils for whom this is revelation.

Now, here’s a claim I want to make: linguistic phonics is perfect for teaching any variety of English, regardless of accent (!) and this is why: it all depends on one simple idea - many spellings represent more than one sound. Moreover, it’s a very easy concept to understand. Show a four-year-old child a picture of a circle and ask them what it could be. They’ll give you innumerable examples: an orange, a ball, a face, a pizza, etc, etc. If they can understand that, they can understand that the spelling [ea] can be /ee/ in 'seat', /ae/ in 'great', and /e/ in 'bread'.

If you understand that, you can adapt your teaching of phonics to any accent of English:
If you’re teaching the word ‘bath’ in North Staffordshire, the spelling [a] represents the sound /a/ in ‘hat’. It is /a/ for the people living there and also in many other places in the UK. However, for a large number of other people who live across a great swathe of the South, the [a] represents the sound /ar/ as ‘father’.
So, it can be /a/ and it can also be /ar/. If one doesn't work, what do you do? Try the other! As long as teaching is grounded in the forty-four sounds of the language, it’s an absolute cinch to teach: does the [a] go in the /a/ category or in the /ar/ category?

This also allows for interesting classroom discussions: the spelling [oo] can be the sound you hear in ‘could’ in words like ‘cook’ and ‘book’ in many accents; it can be the sound you hear in ‘moon’ and ‘soon’ in others.

Linguistic phonics is simple and logical because it is always anchored in the forty-four or so sounds in the language, whether you come from Manchester in England, Glasgow in Scotland, Austin in Texas, USA, or Perth in Western Australia, teachers can adapt their phonics instruction to the accents of the children they are teaching. 

With 359 million native speakers in the world, why isn't everyone teaching linguistic phonics? You know it makes sense!


Thanks to Andrew Weldon for the cartoon image: http://andrewweldon.com/

Tuesday, May 03, 2016

Linguistic phonics v traditional phonics

Given that for many researchers working in the field of beginning reading and writing it is axiomatic that teachers should be adopting a synthetic phonics approach, the next question is: should that approach be graphemic, as Letters and Sounds is; or, should it be phonemic, as Sounds-Write, Sound Reading System, and That Reading Thing are?

Although I've written about the differences here and here, this post is a response to the almost constant, recent requests I've received for clarification of the differences between a truly phonic approach and a graphemic one.

I would love to find common ground with the people who advocate traditional (i.e graphemic) phonic approaches but the differences between them and phonics approaches start right from the off: in the normal course of things and whatever the graphemic programme, teachers present children with simple, three-sound words and point to the individual spellings and ask, “What sound does this letter make?” Or “What sound does this letter say?” When I hear this, I want to scream “Letters don’t ‘make’ or ‘say’ sounds!! We do! People do! Children do!” We are the active agents in this game. Speech sounds precede the invention of writing by tens of thousands of years. Writing is a relatively recent invention and was developed to record the sounds in people’s speech. Spellings comprised of one-, two-, three-, or four-letter spellings are symbolic representations of sounds.

If writing systems are invented, what are we to learn from this? That if you are going to teach young children to read and write, you should teach it in the direction of how it was meant to be taught. If you don’t, you quickly become entangled in the most ridiculous contortions to explain your teaching. These are many and varied and include such notions as ‘silent letters’, ‘kicking /k/ and curly /k/’, ‘hard sounds and soft sounds’, teach letter names rather than sound-spelling correspondences, and so on. The result is likely to be confusion, at least in the minds of some young learners.

So, not a good start, although you might still be thinking that most pupils will ‘get it’ if a teacher teaches that letters ‘make’ sounds, and if the teacher doesn’t bother to teach the children to say sounds precisely, or even if the teacher throws in letter names along with sounds. And, perhaps, many children might find ways of finding enough logic in what is being presented to be able to make sense of reading simple, one-letter spelling words. On the other hand, there are almost certainly already a number of children who don’t ‘get it’ and are beginning to wonder how this reading and writing thing works. And, again, maybe I can exclude the writing bit because lots of teachers don’t teach reading and writing together anyway!
Linguistic phonics practitioners surveying the chasm.

So, let’s move on a step. Where the differences between the two approaches suddenly develop into a chasm is when we get to the teaching of the double consonants. It was reported to me the other day that children in a class of young learners were being told to sound out the word ‘miss’ by saying ‘missssss’. This, presumably, because the teacher had no understanding that we spell sounds with two letters! In fact, the double consonant provides the perfect opportunity to teach that we can spell a sound with two letters: “It’s two letters but it’s one sound!” This is a concept that children are going to be getting to grips with a lot because, sometime later, after the four double consonants, pupils are going to encounter the spellings [ sh ], [ ch ], [ th ]. A linguistic phonic approach enables the teacher always to be consistent and without having to fall into muddled explanations.

In this way, pupils gain, through practice activities, a firm grasp of the idea in different contexts, such as with vowel digraphs, other consonant digraphs. Over time of course, pupils will butt up against the inadequacy of this concept in that they’ll be trying to read words with three-letter and four-letters. This takes us into the territory of extending understanding. After practice with one- and two-letter spellings for sounds, the idea that spellings can contain three-letters and four-letters is a mere bagatelle: the ground has already been well prepared. Imagine now that if you are a child and someone has already made clear that sounds can be spelled with two letters and you get to the word ‘light’ in a text you are reading. The teacher, knowing that neither spellings of the sound /ie/, nor three-letter spellings have yet been covered, runs her chopstick/pencil under the [ igh ] and says, “This is three letters but it’s one sound. Say /ie/ here.” And the child reads “/l/ /ie/ /t/, ‘light’.”


A sound to print approach is logical because there are a limited number of sounds in the English language and they don’t change. The hard bit about teaching the writing system is that it contains so many complexities. Now, I’m not claiming that every single child will, if taught in this logical and consistent way, eventually learn to read and write without making errors. What I am saying is that, although there are probably too many less frequently encountered spellings for anyone to become a perfect speller (or even reader, in the case of a few words), this approach makes perfect sense. Taught well, from simple to more complex, with plenty of opportunities for extensive practice, almost every child can learn to read and spell to a very high level of proficiency.

Saturday, April 30, 2016

One sound, different spellings: the Sounds-Write way

I’ve just been asked by someone in Australia why it is that the Sounds-Write programme aims to teach to young children  multiple spellings of a sound at the same time - the concern being about overloading children’s memories.

This is unquestionably the hardest aspect of learning how to read and spell for every single one of us. Why? Because there are so many different ways of spelling the 44 or so sounds in English. Even a person who regards him/herself as an almost perfect speller will admit that reading and spelling words they have never seen before and that contain less frequently encountered spellings is very troublesome. For example, could you read accurately or spell the word ‘laodicean’, /l/ /ae/ | /oe/ | /d/ /i/ | /s/ /ee/ | /schwa/ /n/, not having seen it before? I chose this seemingly random word because someone actually did! It happened in 2009, when Kavya Shivashankar from Kansas triumphed over 292 other entrants to win that year's Scripps Spelling Bee by spelling the word correctly.

So, why do we believe it necessary to introduce more than one spelling of a sound as soon as we have taught the basics. There is good reason for this. If a teacher teaches one spelling for a sound and three months later teaches another spelling for the same sound, followed, say, a few months later by another spelling of the sound, the connection between the sound and the spellings is lost! Furthermore, taught in such a piecemeal way, young children are likely to form the impression that there are endless spellings of a sound and that, therefore, there is no point in trying to learn them. In other words, such an approach is demotivating.

By the time they arrive in school, children have already acquired the sounds of their own language. This is our starting point and the anchor for our teaching of the code. If you tell the pupil that this [ ay ] spelling is the way we spell the sound /ae/ in ‘play’ but that there are other ways of spelling the sound and these others are introduced and are taught in the context of whole words, the idea isn’t difficult to grasp. From a conceptual point of view, children are easily able to identify different types of car or flower but still know perfectly well that they all fall under the superordinates of ‘car’ or ‘flower’.


The anxiety some teachers have when we tell them that we teach four spellings of a sound to begin with is that we expect the child to remember them and be able to use them in their writing immediately. This isn’t at all the case! For pupils to be able to use the spellings of sounds correctly in words, they still need to have lots of exposure from reading. For this, they need to be trained to notice which particular spellings are being used in the words they read, which they get from having lots of deliberate practice. By helping them to internalise the key points of learning, what we are doing is creating mental representations of the spellings of sounds in words.

To strengthen this, we make sure that pupils spend two weeks focusing on and practising, through a variety of activities, the spellings of target sounds in the context of words, sentences and texts. This gives them time to transfer the new knowledge into long term memory and to further build the schema for sounds and spellings we have helped them develop from the beginning of the programme.

After a sound and a limited number of spellings have been covered in the two-week period, they are recycled over the next two weeks when we introduce another sound and its spellings. The process is further reinforced a short time later, this time in the context of teaching polysyllabic words.

So, back to my Australian correspondent: after such rigorous exposure to the common spellings for each sound, taught over two years, we find that almost all children are able to cope very easily with multiple spellings of a sound for both reading and spelling.

Monday, April 25, 2016

What do young children learning to read and write have in common with the young Mozart?

When Matthew Syed, former table tennis champion, Times correspondent and author, trained his sights on what it is that makes someone a champion, he tackled the question head-on by dealing with one of our most cherished enigmas: in his superb book Bounce (2010), he asked the question 'How do you solve a conundrum like Mozart?'

The reason he asked this particular question was because, although many of those interested in the field of how people acquire expertise in any domain, Mozart appeared to contradict the so-called 10,000 hours rule. This ‘rule’ states that it takes around that amount of time to become an expert in any particular field. Never mind that this figure requires a large degree of qualification, the fact is that the acquisition of expertise requires a great deal of knowledge, practice and expert tuition and that this takes time

So, how was it that, at the age of six, Mozart was wowing the courts of eighteenth century Europe with his compositional skills, especially when, as Syed points out, he’d ‘scarcely even lived ten thousand hours’! Perhaps the only other explanation was that Mozart was a child prodigy, a genius.


Syed is quick to build the contrary case: Mozart’s father was, even for the standards of his time, a ‘tiger’ parent, who put his young (three years of age) son through a rigorous course of instruction in composition and playing. By the time he was six, according to the work of psychologist Michael Howe in his book Genius Explained, Mozart had put in 3,500  – still a long way from 10,000 – hours of practice before he was six.

Syed also points out that lots of other children were regarded as being child prodigies, until that is, investigators began to dig a little deeper into their backgrounds. There, they found that many children, who, through their interest in a specific domain, had had the benefit of a huge amount of practice under the tutelage of very experienced and knowledgeable teachers, turned out to be outstanding in their chosen enterprises. 

So far, so good! However, there was always a lingering doubt. Notwithstanding the rigour of the tuition and the amount of practice, the case made by Syed still, for some, does not necessarily explain Mozart's extraordinary ability. Could it also be attributed to the fact that, to be a musician of such remarkable ability, one needs to be born with 'perfect pitch'?

What Matthew Syed wasn’t able to know at the time he wrote his book was what has emerged in a new contribution to this debate in K. Anders Ericsson and Robert Pool's new book Peak: Secrets from the New Science of Expertise is some very interesting, new research.K. Anders Ericsson’s

Wolfgang Mozart did indeed possess the, for musicians, invaluable ‘gift’ of perfect pitch, which is, apparently, an ability that only one in ten thousand of us have and, until recently, was thought to be an innate talent. Thanks to what Pool and Ericsson call a ‘beautiful experiment’, conducted at the Ichionkai Music School in Tokyo by the Japanese psychologist Ayako Sakakibara, we no longer regard perfect pitch in the same light.

Sakakibara trained twenty-four children, aged between two and six years, to identify ‘by their sound, various chords played on the piano’. The sessions were short and delivered for only a few minutes, four or five times a day over a period of between twelve and eighteen months. At the end of this training, all children – every single one! – had developed perfect pitch and were able to identify notes on the piano. This was an unprecedented discovery!

Mozart worked for very much longer and very much harder than any of the children involved in Sakakibara’s training. It is hardly surprising then that he developed the talent at such an early age. What is likely is that Mozart wasn’t born with perfect pitch but, as Pool and Ericsson put it, was born with ‘ability to develop perfect pitch’, a gift almost all of us have, given the right kinds of exposure and teaching.

Now my speculation is this: given the same kind of exposure to very high quality phonics teaching and at a relatively early age, virtually all children can be taught to identify the individual sounds of their own language – in this case I’m talking about English, or even the sounds of English as an additional language. After all, if our naturally evolved hearing, through which we are primed to pay attention to the sounds of our own language even before we leave the womb can be co-opted to identify and differentiate the individual sounds produced by a musical instrument, how much harder can it be to train children to identify and differentiate consciously the sounds of their own language in the context of learning to read and spell?

Saturday, April 16, 2016

The phonics achievement challenge

Following on from my last post in which I reported the results of a spelling test taken by a class about to begin Y2 in St George's Primary School in Wandsworth, this time I'm publishing the results from the same spelling test for a class just about to begin Y3. 

The reason I'm putting the posts back to back is that their juxtaposition will give you a good idea of the kind of progression you can expect to see from the extra year's tuition the pupils get from the Sounds-Write programme.

I’d make clear that the test (Dennis Young’s Parallel Spelling Test) was not developed for teachers of phonics. In fact, the test moves very quickly from simple two-sound and three-sound words made up of simple one-to-one correspondences, plus some initial consonant digraphs, to the greater complexities of the code: more than one way of spelling a sound.

After Y2, the class’s third year of phonics teaching, you can see that the effects of their exposure to the code and their practice of the skills, through the excellent teaching they receive at St George’s and a wide variety of reading, is now beginning to have a much more profound and qualitative effect on their spelling.

As you look at the results, bear in mind, too, that this school is situated in the most deprived area of London, south of the river. You can see that boys do as well as girls and that summer-born children do as well as children who entered school well after their fourth birthday.

You can also see that there is not a single child whose spelling age is lower than their chronological age. The same is true also of the previous year. On a test that has been normed and standardised on thousands of children, the achievements made by the staff and children of St George’s is remarkable.
What’s more, this is a challenge to all those who say that mixed methods and whole language are superior to a systematic, very high quality phonics approach. 

To all of those sceptics, I throw down the gauntlet.

St George’s Church of England Primary School

Year 3 Young’s Parallel Spelling Test September 2015

Male/Female
Raw Score
Age yrs.mo
Spelling Age
Standardised

M
28
7.5
9.4
120
+ 23 mths
M
27
7.7
9.3
116
+ 20 mths
F
36
7.4
11.1
133
+ 45 mths
F
32
7.4
10.2
125
+ 34 mths
M
29
7.8
9.6
117
+ 22 mths
M
36
7.6
11.1
130
+ 43 mths
M
23
7.10
8.6
108
+   8 mths
F
32
7.10
10.2
120
+ 28 mths
F
32
7.3
10.2
127
+ 35 mths
F
33
7.8
10.4
123
+ 32 mths
F
19
7.5
8.1
107
+   8 mths
F
38
7.9
11.9
>135
+ 48 mths
F



abs

F
17
7.3
7.8
106
+   5 mths
F
34
7.6
10.6
126
+ 36 mths
F
33
7.2
10.4
129
+ 38 mths
F
28
7.1
9.4
124
+ 27 mths
M
38
7.9
11.9
135
+ 48 mths
M
34
7.1
10.6
133
+ 41 mths
F
33
7.7
10.4
125
+ 33 mths
F
30
7.10
9.8
117
+ 22 mths
M
23
7.9
8.6
109
+   9 mths
M
19
7.3
8.1
109
+ 10 mths
M
32
7.1
10.2
130
+ 37 mths
F
724
7.1
8.8
118
+ 19 mths
F



abs

M
31
7.2
10.0
126
+ 34 mths
M
22
7.10
8.5
106
+    7 mths
F
39
7.11
12.5
135
+ 54 mths
F
20
7.0
8.2
112
+ 14 mths