Saturday, March 05, 2011

Greg Brooks: no change in the spelling system! Well, just a bit.

While I agree with Greg Brooks in his piece in the Guardian yesterday, where he argues that changing English spelling simply won’t work, the way in which his argument is couched seems to contradict the initial message.
There are, as he points out, forty-four or so sounds – it depends to some extent on accent – in the language; and, although I have never counted them, he states that there are five hundred sound-to-spelling correspondences if you count all the infrequent ones. I have to say that this figure seems unnaturally high and doesn’t take into account the redundancy factor.
He then goes on to tell us that there are ten ways to spell the sound 'e'. He's right! Though in the context of his discussion, which is to do with the problems in teaching the sound-to-spelling system to children in school, there is just one spelling of 'e' (the letter e) you'd teach to reception-aged children before, shortly afterwards, adding two more, the ea and the ai spellings of 'e' in words like 'bread' and 'said'. After that, you’d bring in the rest of the spellings of 'e' in the context of the subject curriculum or as they come up in the round.
When it comes to the issue of spellings representing more than one sound, Greg Brooks is not clear about how the writing system works. He says that we pronounce the letter a in various ways, as if letters somehow mysteriously have sounds of their own. Even in the phonics community, this a point not always well understood. However, if children think that letters 'make' sounds, instead of representing sounds that we make in our speech, they may never get to understand the fundamental logic of the writing system: that letters are symbols for sounds. So, Brooks has got the thing back to front here. It should be that the letter a, like many spellings, represents more than one sound. In the word 'mat' it is 'a'; and, in the word 'baby', it is 'ae'.
Brooks is also less than explicit on a couple of other things: for example, the spelling a in 'about' and in 'village' are both schwas, or weak vowel sounds.
Apparently, according to Brooks, spellings in the words ‘should’ and ‘enough’ are ‘absurdities’. Why? The spelling oul for the way most people in the south of England pronounce the oo in ‘book’ only occurs in a small number of words and isn’t difficult to teach or learn. In ‘enough’, what is the problem with the sound ‘u’ being represented by the spelling ou or the sound ‘f’ by gh? After all, we have the words ‘cousin, ‘young’, ‘couple’ and ‘trouble’ and then there’s ‘tough’ and ‘rough’, which contain the gh spelling of ‘f’.
Moreover, to say that there’s not a ‘high level of illiteracy in the UK’, as Brooks claims, is simply disingenuous. It’s probably true that there are relatively few people who cannot read or write at all. However, there are huge numbers of illiterates, as defined by the OECD and various other bodies, including the Moser Report. And, where is the evidence for his claim that there are ‘also’ (Methinks he does contradict himself here!) low levels of literacy in countries with transparent orthographies?
Brooks ends his piece by suggesting we make the task of learning to read and write English ‘less difficult’, though he doesn’t say how. So far, his drift has been that we shouldn’t change the spelling system but then he flips about to a ‘Oh go on then - but only a bit!’ position.
What can we do to make sure that children become properly literate? Train the teachers to understand how the writing system works, and then get them to teach children in school from simple to complex, by which I mean:
  • teach them one-to-one correspondences and give them lots of practice in blending, segmenting and manipulating sounds in words; 
  • teach them that a sound can be spelt with two-letter spellings; after that teach a limited number of ways of spelling a sound; 
  • then teach that many spellings represent different sounds; 
  • and finally teach them how to put all the above together in polysyllabic words and how to recognise schwas or weak vowel sounds in polysyllabic words. Simples!
And yes, children should be reading texts that are commensurate with where they are in the teaching programme, as well as being read a wide variety of rich, literary texts.


valerie yule said...

You are right in saying that there are more illiterate people than Greg Brooks admits, but your 'Simples' help to explain the difficulty many people have. There are so many spellings for one sound; so many sounds for one spelling. We could try another way that would be less confusing:

a. Beginners start with the spellings for 44 sounds - the ABC plus digraphs needed for other English sounds, such as th and ow. This is also made the Pronunciation Guide in dictionaries. Formal pronunciation is the base, not casual speech with its schwas.

b.Then beginners learn by rote 37 common irregular words which make up 12% of everyday text – all almost always among are as come some could should would half know of off one only once other pull push put they their two as was what want who why, your, and word-endings -ion/-tion/-sion. This is not too much to learn by rote, with the aid of the clues from the Pronunciation Guide.

c. Next, grammar and units of meaning – s for plurals and tenses, consistent spellings for final vowels, and the ‘silent e' tactic for long vowels, for an optional spelling system for those who cannot get further to spell.

d. Allow up to 4 variant spellings for 9 vowels and 4 consonants, for spelling that most people could read. These are existing patterns. 149 less familiar spelling patterns are unnecessary.

3% of letters in words are changed in ordinary text, and 6% omitted as surplus letters useless for meaning or pronunciation.

e. Then present spelling can be readable with a little gesswork.

It would not require re-learning or reprinting books. English culture would still be transmitted.

B. A beginning for those presently literat to simplify spelling. Cut surplus letters in words that are no use to represent meaning or pronunciation. These are the main reason why most people cannot spell. This also cuts half the trouble with ‘spelling demons’.
An International English Spelling Commission monitors and implements research, on the lines of other languages’ national academies.

John said...

Thanks for your comment, Valerie.
Just a few things I'd like to take up:
Firstly, you can't assume that beginners start with 44 sound-spelling correspondences. In the Sounds-Write programme, we introduce: all the basic one-to-ones, plus four two-letter spellings (ff, ll, ss, zz). Time is spent on blending, segmenting and phoneme manipulation practice until children can read and spell CVC, VCC, CVCC, CCVC, and CCVCC words to a high level of skill. After that we introduce four spellings of the sound 'ae', then four spellings of the sound 'ee', and so on. All of these sound-spelling correspondences are introduced in the context of real words - so none are introduced in isolation.
In the S-W programme, as few high frequency words are introduced as possible so that we're not teaching two strategies that run counter to one another. In any case, why would you teach words like 'off' and 'push' as hfw when they are very easily decodable? Similarly, if you teach from sound to print, the way we spell the sound 'o' after the sound 'w' is very frequently with the letter a: thus, was, what, want, swap, swan, wasp, etc, etc.
Grammar is a completely different aspect of teaching from teaching reading and spelling skills.
On the question of spelling reform: the fact is that many have tried and none have succeeded to any particularly meaningful degree. Much better is to train the teachers how to teach it from day 1 of reception. By the end of Key Stage 1, if it's taught for half an hour every day five days a week, children become super readers and spellers.
To see the order in which we teach the code go to: on the Sounds-Write website.
There is there also a huge amount of data on how successful Sounds-Write is in teaching pupils in KS1 to read and spell:
Best wishes,