Synthetic phonics does not start with whole printed words. It starts with single letters and the sounds the letters represent. As soon as the children have been taught a few letters and sounds, including one or two vowels, they are taught to look at words, produce a sound for each letter (no digraphs are included at this point) and then blend the sounds all-through-the-word into normal word-pronunciations. This is significantly different from most National Literacy Strategy (NLS) activities, where the emphasis is on the analysis of spoken words supplied by the teacher. This is clearly important, but for spelling rather than reading. The focus in this paper is firmly on synthesising (sounding out and blending) for reading, as this is where the NLS is weak.
Increasing numbers of regular words can be blended as the number of letter-sound correspondences taught increases, e.g. at, dog, hen, spot, bend, hill. All the time the teaching is building up in a logical sequence. Almost from the beginning the children understand that there is a code to reading, and that unknown words can be worked out. This gives the children, and particularly the boys, a great deal of confidence. The more complicated and irregular words are gradually introduced later, and in a structured sequence. Although teachers read stories to the class, the children are not expected to read books for themselves, and certainly not books that contain digraphs and irregular words that have not been taught.
The next stage is teaching the digraphs, two letters making one sound. As the digraphs are taught the children practise blending regular words that use these digraphs e.g. boat, sleep, shout, sport, boil, sister. Regular practice enables the children to become skilled at blending words with digraphs.
Once the children are used to blending, then less regular keywords are tackled, usually two or three new ones a week. Initially the children blend them as far as possible; if blending results in inaccurate pronunciation, they are told the word and then look for the irregular part. Some part of the word will be regular and by concentrating on the irregular part the children are more able to get it into their long-term memory. For example, with the word 'do' the 'd' is regular but the 'o' has an /oo/ sound.
This thorough grounding prepares the children for the more complicated task of reading books for themselves. Initially it is important that the texts should be decodable, that is they use words that the children can read by sounding out and blending. Inevitably there have to be some less regular words but these should be kept to a minimum and should consist of the ones that have already been taught in class. Once there is fluency in the reading then the children can cope with far more irregular words. It is fluency in the blending that is the key to successful reading. This then leads to improved comprehension. If you cannot read it, you cannot comprehend it. The mechanics come first.
In the Quality and Curriculum Authority (QCA) publication 'Standards at Key Stage 1 English and Mathematics 2001' it was reported that the children who achieved Level 2B or above at Key Stage 1 Standard Attainment Tasks were good at blending but the children below that were particularly poor at it. Sounding out and blending is the strategy which good readers use when they encounter an unfamiliar new word. Beginners cannot do all the things that proficient readers do, but this happens to be one thing which they can do if text is appropriately graded and which will stand them in good stead forever. They encounter unfamiliar printed words far more frequently than proficient readers, but they can still use the same strategy, provided that the words incorporate letter-sound correspondences that they have been taught. This is why decodable texts are very important for beginners. They provide the revision of letter-sounds and give the blending practice which is necessary if the children, particularly the bottom 20%, are to become fluent at reading. The Office for Standards in Education (Ofsted) reported that 20% of children were so poor at reading at KS1 that they were not able to achieve the necessary level at the end of KS2. The NLS, with its lack of concentration on blending, lets this 20% of children down. They cannot blend because the NLS has not emphasised it enough.
Synthetic phonics builds up from small to large, in a structured way, and teaches blending as the first, and virtually only, strategy for the word-identification aspect of reading. The NLS goes from large to small, uses analysis rather than blending, and encourages a mixture of strategies most of which amount to guessing. This is why synthetic phonics is far more effective than the phonics in the NLS.
The differences between the NLS phonics and Synthetic phonics.
The impression given by the NLS is that words are either to be recognised as ‘sight words’ or to be guessed at from initial letter, context or pictures. This impression is given by the following:
(a) the word-lists in the ‘Framework’ are headed ‘High-frequency words to be taught as “sight recognition” words (p. 60),
(b) there is no reference, in the text below this heading, to the fact that some of the words can and should be read by sounding out and blending, and
(c) children are expected, from the start, to read books in which most of the words cannot be decoded on the basis of what they have been taught.
This is where the damage begins. The children with a good visual memory will cope well with this start, although not as well as with synthetic phonics, but for the many children who have a poor memory and find it difficult, or even impossible, to learn a few words, it is a bad start. Forty-five words to learn by sight must be a nightmare for them. Straight away the bottom group have a feeling of failure. They can't remember the words from their shapes. There seems to be an endless number of words to learn. In order to try and understand how difficult it is, we could imagine ourselves trying to memorise the shapes of Greek, Arabic or Hebrew words. Even though we know that there is an alphabetic code, and have probably a better memory for symbols than that bottom group of children, it would be difficult for us to learn these words by their shape. It is important to remember that our letters are just as odd to children as the Greek, Arabic or Hebrew ones are to us (my apologies to any literate readers of these other languages).
By contrast, synthetic phonics avoids any suggestion of sight-word learning at first, emphasises sounding-out and blending as the first and virtually only strategy to be used for the word-identification aspect of reading, and provides texts which make this feasible.
It is also worth noting that the vast majority of children arrive at school with no sight words, apart from perhaps their name. It is a myth that they can easily read words like McDonalds, Coca Cola, Bob the Builder etc. Take these words away from their settings or logos and the children cannot read them at all. They are not reading the letters, only the colours/pictures/building surrounding them.
The NLS then encourages the teaching of most of the alphabet letter sounds so that the initial letter can be used to help with reading unknown words.
With synthetic phonics this would be completely discouraged. Guessing words from the initial letter is notoriously inaccurate and starts the children in the bad habit of guessing, which is a very hard habit to break.
The NLS continues by teaching the short vowels a,e,i,o,u, but in the NLS Progression in Phonics (PIPs) programme this does not take place until Step 4, which is the last step before Year 1. Therefore, without any vowels there can be no blending before this stage. The children are expected to read books or texts from Step 1 but the blending skill is left until the latter part of Reception, and then hinted at rather than explicitly being taught. This reinforces the impression that the children are meant to read by learning words by sight and not by blending. This is why the NLS is not a synthetic phonics approach.
In Year 1 the letters v,w,x,y,z are taught and the main digraphs. The children play games with these letter-sounds (PIPs). Out of the 28 games, only three are truly blending games. The other games are very much about identifying the sounds in spoken words, which is the skill needed for writing.
Synthetic phonics teachers find that these games are very time-consuming and fail to devote the necessary time to the essential skill of blending. However, another major problem is the fact that these games are played in isolation and not related directly to the task of reading and writing texts: the incompleteness of phonics teaching when not applied to the reading and writing of texts was brought out by the National Reading Panel in the USA. In the NLS blending is always the last strategy that children are expected to use when reading texts (it is the first strategy in synthetic phonics). You only have to look at the Early Literacy Support video to realise this. You never see a teacher saying "blend it" or "sound it out". You see the teachers encouraging the use of picture cues, initial letter and guess, or just plain guessing from the context (even when the word is easy to decode). The implication in the NLS is that the children do not blend when reading texts and this, in my view, is the worst fault. The more able children, usually the ones with the good memories, understand the alphabetic code from the games and are able to apply it to reading texts and writing. These are the children who will get Level 2B or above at KS1 (If they had been taught with synthetic phonics they would achieve even higher results). The less able children are bewildered by the many strategies, they can't remember words by sight, and they become demoralised. The number of new words to learn seems endless.
If this method of teaching was applied to maths, teachers would be expected to start with sums: holding up cards with sums on them, such as 2+3=5, 2+4=6 etc., and chanting what they say. Then they would follow this by playing some games that teach the value of numerals, and, as with the NLS and Progression In Phonics, expect the children to make the connection between sums and numerals by themselves. Once again you would find some children manage to do it but the others would become totally lost. This is roughly what we are doing to the children when we ask them to learn whole words without blending skills or sufficient letter-sound knowledge, and then expecting them to understand the code of reading by playing games that are not connected to texts. The cart is still before the horse and this is why the phonics in the NLS does not work as well as systematic synthetic phonics.
*Reading and writing skills in their various stages are measurable. The scientists should be involved in helping to decide what is the most effective and inclusive good practice.
Editor’s comment: Following the promotion of the Reading Recovery Book Bands catalogue, particularly through the NLS Early Literacy Support programme, schools have invested in further reading books to address the apparent needs of beginning readers. On the following page are examples of full texts from the lowest level of Book Bands and these are frequently given to children who have only been introduced to some of the letter-sound correspondences. You can see that predictable repetitive texts are still the flavour of the day. The children are expected to learn the words by sight. In reality many, if not most, are unable to do this and cope by memorising the sentences and using the pictures for any new words. Cover up the pictures and the children cannot read. The RRF know this type of text is entirely unhelpful to children when in the early stages of learning to read. Children taught by synthetic phonics can read independently an extremely large number of words at an early stage and books should reflect this decoding ability whilst experience and fluency develops. In contrast the texts on p.28 have a very limited number of unjustifiably difficult words for a beginner: