Tuesday, February 16, 2016

The ill-conceived idea of 'regular and 'irregular' spelling

What do people mean when they talk about ‘regular’ and ‘irregular’ spellings?
‘Regular’, as the dictionary definition suggests, means ‘arranged in or constituting a constant or definite pattern, … well ordered, well structured, perpetual, constant…’ The problem is that there is only one constant in the spelling system in English and it isn’t the spellings! It is the forty-four or so sounds in the language; it is sounds that drive the spelling system, not spellings. What people are confused about is the nature of the writing system by which we represent the sounds.

We can, fairly reasonably, say that Spanish, say, is almost ‘regular’. What this means is that most of the sounds in Spanish are represented by one-letter spellings. With around 23-25 sounds in Spanish and only around 35 spellings of the sounds, Spanish spelling is really easy to learn. It is true that there are a few complexities but these pale to insignificance in comparison with English.

Most English teachers who teach phonics have accepted the model of what is often referred to as a ‘basic code’. At Sounds-Write we call it the Initial Code, but no matter. A basic code looks very similar to what teaching reading and spelling in Spanish looks like. It begins with teaching children one sound-one letter spellings because they are easy to learn. Different programmes may have a different order of play - the order in which sound-spelling correspondences are introduced - but they all arrive at the end point having taught that the sound /a/ is spelt [ a ], /b/ is spelt [ b ], and so on.

Of course, in terms of regularity, a problem can be identified early on: if the teacher has taught that the sound /k/ is spelt [ c ] and then, a week or so later, they go on to teach that [ k ] can also represent the sound /k/, which one is regular? Later still, the teacher will go on to teach that the sound /k/ can be spelt still another way: [ ck ]. Does this alter our view of the idea of regularity?

And if we flip over from sound to spelling to spelling to sounds, we quickly encounter further ‘irregularities’. For example, the spelling [ c ] can be /k/ but it can also be /s/; [ f ] can be /f/ and it can be /v/; [ g ] can be /g/ and /j/; and so on. And that is only the consonants. The vowels are much more complex and that is where, in the minds of many teachers, all hell breaks loose because beginning readers and writers need to know that sounds can be spelt in different ways. For example, /ee/ can be spelt ‘seed’, ‘tea’, ‘funny’, ‘key’, ‘brief’, ‘ski’, ‘she’, and ‘receive’. Are all of these spellings in words ‘regular’? This is the point at which people begin to shift uneasily in their seats. Suppose that the answer to the latter is that they are 'regular' because they are all appear fairly regularly in English words. Then, what about the /ee/ in ‘archaeology’, or the /ee/ in ‘amoeba’? Are these encountered infrequently enough to be classified as ‘irregular’? Surely it depends on the domain of knowledge to some extent?

The thing is that the idea that spellings of sounds in words are regular or not regular simply doesn’t have any logic to it. What does make sense is to talk about the gradations between common (i.e. frequently encountered) spellings and highly unusual spellings. I would consider the spelling [ ae ] representing the sound /ee/ to be highly unusual in children’s KS 1 texts, informational or narrative. However, by Key Stage 2, I would expect it to crop up from time to time in certain words and be taught explicitly as it comes up in teaching (the Egyptians or Romans, for instance) at KS2.

As I’ve said, there are gradations. Is the spelling [ a ] in ‘was’ irregular? Clearly, many teachers think so. Why do they think so? Because ‘was’ is a frequently encountered word in children’s texts and they believe that, in the initial stages of learning to read, to teach that [ a ] can be /a/ in 'mat' and that it can also be /o/ in ‘was’ might confuse young learners. I don’t doubt that this is absolutely right, it might, but that isn’t reason to suggest that the spelling is ‘irregular’. If it were, we’d have an awful lot of trouble teaching ‘swan’, ‘want’, ‘swap’, ‘what’, ‘wan’, ‘wallop’, ‘wasp’, ‘quad’, ‘qualify’, ‘quantity’, etc. In fact, as can be seen, the spelling [ a ] representing the sound /o/ in words preceded by the sound /w/ is very common. So, perhaps the pattern isn’t so irregular after all.

Where does this leave us? It leaves us teaching a straightforward ‘Initial’ or ‘Basic’ code to begin with, followed by more frequent spelling patterns - always, I hasten to add - grounded in the sounds of the language. Our teaching can then be completed by introducing less frequently encountered spellings – usually as they crop up in the context of what is being taught in the curriculum. These can be added, as reminders, in the context of real words, to spelling posters on the walls of the classroom.


Nonetheless, we come back to the same old refrain: to be able to teach all of this, teachers need to understand how the code works and how to teach it from simple to more complex.

Monday, February 08, 2016

Letter names or sounds?

This post has been written as a quick response to a debate on Twitter about whether teachers should be teaching letter names or sounds or both to young children just embarking on learning to read and spell.

Until young children (Reception/Y1) are secure with sounds – i.e. they understand that letters are representations of sounds in words – they should not be introduced to letter names. Teaching letter names (as well as sounds) to young children is bound to cause many children confusion. What's more, this confusion very often lasts until secondary school, where pupils who can't read and write well are often thoroughly mixed up about what this code is all about.

But, understand that teaching letter names from the start isn’t the same thing as talking about sound symbols collectively as ‘letters’! So, for example, a teacher might teach the two-letter spelling [ ss ] and say, “This," pointing to the [ ss ], "is two letters but it’s one sound. It’s /s/. Say /s/ here.” Why is this different from telling the pupil that the spelling is ‘double ess’? For two very good reasons: first, the alphabet code is a code for the sounds in our speech and the two letter spelling [ ss ] in the example represents the sound /s/; second, because we want to orient pupils in the direction of understanding how the code works: we can spell sounds with one, two, three or four letters.

To take another example: the pupil is reading a book and stumbles over the spelling [ ea ] in the word ‘stream’. Assuming that the pupil is still working with the one-to-one sound-spelling correspondences and also that the pupil is being taught to segment and blend to a high degree of proficiency, the teacher simply runs their pencil under the spelling and says, “This is /ee/. Say /ee/ here.” And that’s it! Really simple! And the bonus is that the pupil is being 'sensitised' to what will shortly be taught formally.

If the pupil is writing, the teacher should be encouraging pupils to think about how to spell sounds in words as they write. A high quality phonics programme will be doing this from the start. Inevitably, pupils will come to areas of complexity that, as yet, exceed their competence. Again, when the pupil says they cannot spell a particular word, there is a judgement call. If the child is still in the early stages of learning the code and they want to write a relatively complex word like ‘delicious’, clearly there too many different complexities to make sense to a child. In this instance, the teacher can and should supply the word. However, if the difficulty is with a word, such as ‘bright’, for instance, the teacher should be asking the pupil what the difficult bit in the word is. As before, if a pupil is working with one-to-ones and can segment and blend, the bit beyond their present knowledge will be the spelling of /ie/. In this case, the teacher writes the spelling [ igh ] on a whiteboard and says, “This is how we spell /ie/ in ‘bright’”. Again, what could be easier?

When pupils get into learning that sounds can be spelled in more than one way and that most spellings can represent more than one sound, the formulation changes. By now, the pupils should understand the nature of the alphabetic code: i.e. that letters, singly or in combination, represent sounds in our language. And, this is when the teacher no longer has to write the spelling on a whiteboard but can simply respond to the question ‘How do I spell the /ie/ in ‘bright’?” with "It’s the i g h (said as letter names) spelling".

This technique works perfectly well for learners of any age. At age five, one may well be writing the required spelling on a whiteboard; at age nine, when the word is ‘archaeology’ and the pupils are learning about the Egyptians, the teacher can give the required spelling as letters names: 'We spell the sound /ee/ in this word [ ae ]."

Teachers are often confused by research that purportedly shows that children who know sound-spelling correspondences and letter names are likely to become more literate, or whatever the claim is. The problem with this is that children who do know both have often been exposed to lots of prior learning. In any case, half the time the researchers have never taught young children and don’t understand themselves the subtleties and intricacies of when and how the teaching of letter names is likely to be both necessary and effective.


Once pupils understand how the code works, letter names then become an essential. If children are taught how to read and spell in a systematic, highly structured way by people who understand the right combination of skills, concepts and code knowledge needed to learn such a complex body of knowledge, they are likely to acquire increasing competence and confidence in dealing with our spelling system.