Seven Sins of Spelling

“All sins are attempts to fill voids.”  Simone Weil

Every time I google the term ‘spelling strategies’ I weep, overwhelmed by the guff and the nonsense. So, I thought I’d unpick (okay, it gets a little ranty at times) some of the worst offenders. I could have included 20 failed strategies, but have kept it to seven. There are other offenders, equally or even worse, than those critiques here, some of which deserve blog posts of their own. If you or your school uses these, please don’t see it as an attack. It’s true, I don’t think they have a place in a sensible spelling curriculum, but I don’t lay the blame at the door of busy, hard working teachers for using them. Somebody, somewhere told you that these strategies represent good practice: perhaps in your initial teacher training, on a course, or in a book. And for that, I feel for you. I have used some of these in the past; but I feel better now. Hopefully, this will lead at least one other teacher on the road to spelling recovery.

1. look cover write check

Let’s tackle the behemoth right off the bat. It’s the strategy that, more than other, during Sounds & Syllables spelling training, results in the most sheepish looks from the audience when I start to pick it apart, so ubiquitous is it. Look Cover Write Check and its various nefarious offspring has one huge, fundamental flaw. It treats words as wholes. Children are expected to look at the whole word, cover it up, recall the spelling, write it down and then check it. Look at the whole word, there’s the problem. I’ve even watched teachers tell children to ‘take a photograph’ of the word with an imaginary camera, further instilling the idea that the word is to be remembered as a whole picture.

Mair (1996), in his analysis of the modern Chinese writing system concluded that “there is a natural upper limit to the number of unique forms that can be tolerated in a functioning script. For most individuals, this amount seems to lie in the range of approximately 2,000 – 2,500.” McGuiness (2004) concurs: “There is a limit on long-term memory for memorising abstract visual patterns paired with individual words. This is an ultimate limit, present after years of intense memorisation. Based on an earlier analysis of known writing systems, I put this limit at approximately 2,000.”

Take Japanese, the only modern writing system that is in part logographic, meaning a written character represents a whole word or phrase. Its writing system contains a large number of logographs, called kanji. Children learn approximately 1860 kanji during their schooling – a process which takes about 12 years (at an average rate of 155 or so per year). Now imagine if English were completely logographic in nature, that each squiggle of written letters was a whole that represented a spoken word in its entirety (a whole that couldn’t be separated into collections of spellings that represent sounds). Assuming a similar rate of progress, learning to write the 40,000 or so words (estimates vary, so this is a rough figure for illustrative purposes) in an educated adult’s vocabulary would take about 727 years. So Look Cover Write Check might be a useful approach, but only if you have seven or eight centuries in which to master English spelling and a brain capable of remembering 38,000 more logographs than the average human being.

However, until science equips us with lifespans in the order of millennia rather than decades, let’s ditch the nonsense of Look Cover Write Check. Written words in English aren’t logographic. They are collections of spellings that represent sounds (and that includes high frequency words, and so-called tricky or sight words). And learning the 180 or so common spellings of the 44 or so speech sounds (click on [fig.1] for a free PDF spelling chart which illustrates this) sounds is a more efficient, less daunting task than memorising whole words. Learn how to spell the word strap as a whole and you’ve learned a single word; learn how to spell each of its five sounds and you can spell asp, at, pat, past, pats, prat, prats, rap, raps, rat, rats, sap, saps, sat, straps, tap, taps, trap and traps.



2. spelling rules

During my Sounds & Spelling training, I ask teachers to explain the doubling rule when adding a suffix. It is rare that anybody successfully does so, despite the fact that the rule is outlined in the English national curriculum appendix (in year 2 and years 3&4). And it’s not surprising that teachers find it difficult, because in all its complex glory (or horror depending upon your point of view) here is the doubling rule.

You double the final consonant at the end of the root word when adding a suffix if the following are true:

    • the final letter of the root word is a single-letter spelling that represents a single consonant speech-sound as in sit, clap or win (which rules out: <x> as it represents a diphone; <w> and <y> as they represent vowel sounds at the end of root words either as part of a two-letter spelling as in chew or play. or in the case of <y> a single-letter spelling as in try or carry, where it represents /ie/ and /ee/ respectively);
    • this single-letter consonant spelling is preceded by a single-letter vowel spelling as in sit, clap or win (not, as is often thought, if it is preceded by a short vowel sound; if that were the case then look with <oo> representing /u/, head with <ea> representing /e/, and plait with <ai> representing /a/ ought to double);
    • the root word ends in <r> which is part of a two-letter spelling represent a vowel sound, for example star, fur and stir (but not a three-letter spelling such as hair, cheer or pour);
    • the suffix begins with a vowel letter, most commonly <e> as in stopped, fatten, wetter, oddest, <i> as in begging and snobbish, and <y> as in runny (and not if the suffix begins with a vowel phoneme; if that were the case we would write stoped because <ed> represents the consonant phoneme /t/);
    • if the stem word is a multi-syllable word, only if the final syllable in the stem is stressed (for example, garden does not double because the first syllable <gar> is stressed despite meeting all other criteria for doubling, but begin doubles because the final syllable <gin> is stressed);
    • and in British English spelling, but not American English spelling, a final <l> in multi-syllable words is doubled regardless of the position of the stress (for example, tunnel doubles despite stress being on the first syllable <tu>.

It’s no wonder many struggle to articulate it.

Now imagine young children learning and applying this in their writing. Yeah, good luck with that. Of course, children wouldn’t be exposed to every aspect of the doubling rule at once, you might argue, but to apply the rule at even a basic level (in line with the Year 2 national curriculum), children would need to know bullet points 1, 2 and 4 at the very least. Now, we might get children to be able to apply this rule during spelling lessons, when we present them with tasks that isolate the rule and build learning in a structured way with plenty of modelling, guided and independent practice. But in the hurly-burly of independent writing, when children are thinking about a myriad of other things – content, purpose, structure, grammar, vocabulary, handwriting – I’d be amazed if for most children complex spelling rules are given a moment’s thought.

Ultimately, the biggest problem with spelling rules isn’t whether they are universally true, or even true often enough to merit attention – and many spelling rules fall at these hurdles as teachers get bogged down in the quagmire of exceptions, sending the message that English is spelling is chaotic and impossible to learn – the biggest problem is that they don’t help children at the point of writing. What we do in spelling lessons needs to be simple and straightforward enough so that knowledge and processes can be automated, freeing up working memory to focus on (arguably) more interesting aspects of writing. And spelling rules fall pitifully short on this account.  

3. word coffins

Word shapes, or word coffins (as Lyn Stone has named them) are boxes drawn around words [fig.2] that aim to draw attention to the shape of letters and the shape of whole words. The problem with this approach is that in focusing attention on the shape it draws attention away from the most important feature of those letters, the sounds that they represent.


And representing words this way is as likely to muddle children than clarify spelling. You might be representing the word dog with the first array of coffins,  but that same sequence of coffin shapes could equally be leg, boylap, try or dug. And you might be drawing attention to the shape of gold with coffin set 2, but those could as easily represent fall, yolk, pelt or yell. And our final coffin sequence, 3, could represent any of pickle, castle, smelts, scuffs or wields.

Word coffins deal with superficial, surface features of words, when children ought to be attending to what matters most – the relationship between spellings and their sounds.  The words bundle and castle are similar in some ways – they both share the <le> spelling of /ul/, but that single two-letter spelling  is separated into different coffins, encouraging children to think of words letter-by-letter and not sound-by-sound. And those two words differ in the number of speech sounds: 4 for castle, but 5 for bundle (ok, in reality it’s 5 and 6, as <le> represents a diphone – two phonemes – but is commonly taught as a single sound in phonics programmes), but they are laid to rest in the same number of coffins, obscuring and confusing that all important relationship between spelling and sound.  

4. mnemonics

How many spelling mnemonics do you know? It’s a question I ask on my Sounds & Syllables Spelling training. In a room full of educators we rarely get beyond five. We might know ‘Sally Ann is dancing’ for said, and ‘o u lucky duck’ for could, should and would. And on a good day we might know ‘one collar, two sleeves’ for necessary, and ‘rhythm helps your two hips move’ for, of course, rhythm. And of course, there’s always a variation of ‘big elephants can always understand small elephants’ for because.

If a room full of teachers can only recall a handful of mnemonics amongst them, it’s not a strategy we can rely on. “But where’s the harm in learning a few mnemonics if it helps to spell the odd difficult word.?” you might ask. Well, I have a few issues with mnemonics.

The first contravenes a principle I try to apply to all aspects of teaching – the principle of effort versus reward. What is the effort required in teaching something so that it is remembered in the long term? What is the reward in terms of learning? And is that effort worth that reward? Mnemonics don’t do well in this regard. There’s simply more for children to remember. Let’s take the word because. It is written with five spellings across two syllables: b e | c au se. It contains one rarer sound-spelling match, <au> representing /o/, which is also found in words such as assault, cauliflower, fault, laurel, sausage and vault. And it ends with the /se/ spelling of /z/ which might be tricky but is by no means rare, and is found at the end of a number of common root words such as blouse, browse, bruise, cheese, choose, clause, noise, please, pause, raise and tease. So a child might be having difficulty with one or two of the five spellings. It therefore makes much more sense to focus in on those trickier spellings, make links to those other words with the same sound-spelling matches rather than to have children remember a nemonic for parts of the word they aren’t having difficulty with.

That’s if they can actually recall the mnemonic correctly in the first place. At a recent training event, a teacher mis-recalled the because mnemonic as ‘big elephants can understand small elephants’ which would, of course, result in the spelling <becuse>. I’ve lost count of the times teachers have misremembered mnemonics, so what hope young children? And even if children do remember the mnemonic, for it to be useful you have to know the spelling of the first letter in every word within the mnemonic itself. Let’s look at because again. The first word in the mnemonic is big for <b> – no difficulty there. Next up elephants for <e> – simple enough. And then can for <c> – again, likely to result in the correct letter choice. But then we get to always for <a>. Now, the first sound in always is /or/, which for spellers struggling to spell because is potentially as difficult and could result in the selection of <o> rather than <a>. And so the mnemonic falls apart.

And even if children are successful with all of the above and can use the mnemonic correctly to bag them the spelling of the word because, the reward for the effort is the spelling a single word. Learning one mnemonic gets you one correctly spelled word. Time that could have been spent on something more valuable: focusing attention on the spellings of <au> spelling of /o/ and the <se> spelling of /z/ bags you a whole clutch of correctly spelled words rather than the paltry one that time spent on the mnemonic gives you.

And don’t get me started on ‘Mrs D, Mrs I, Mrs F F I, Mrs C, Mrs U, Mrs L T Y’. The mnemonic is to literally spell the target word. If you can remember the mnemonic, you don’t need the mnemonic!  

5. does it look right?

Again, during my Sounds & Syllables training, I ask people to spell a word that I’m fairly confident isn’t in their vocabulary. Let’s imagine this word is gharial (it isn’t in case you ever end up on my spelling training). I don’t provide a definition so there is no morphological or etymological knowledge people can draw on (having said that, I’m not sure that knowledge would help too much – a gharial  is a large Asian crocodile species, also known as a gavial.) I even sound out the word for them to aid their spelling: /g/ /ar/ /r/ /ee/ /uh/ /l/. Of course, without the word being in their vocabularies, the best that anyone can do is to draw on their knowledge of sound-spelling relationships and orthography (whether implicit or explicit) and make a good stab at it. Most people end up with something plausible, but few people spell it correctly – as would be expected: after all, there are multiple ways of spelling each of those sounds and some of the spellings in gharial are less obvious choices, <gh> representing /g/, for example.

It is fascinating to watch people tackle the spelling. One of the most common approaches is to write out the word several times, each attempt a different spelling, to see which one looks right. But here’s the rub. Much of the time, having written several variations of gharial, none of the spellings is correct. Now, it doesn’t matter which one of those spellings looks right to this speller, they are wrong whichever decision they make. And then there are those who do end up writing the correct spelling of gharial among one or two incorrect ones. And guess what? The chance of them selecting the correct spelling is little more than chance. An incorrect spelling ‘looks right’ almost as often as the correct one.

The thing is, ‘does it look right?’ can only ever helpful if you have already learned to spell the word (but perhaps have partially forgotten), and if you have plenty to draw upon: sound-spelling relationships, morphology and orthography. And those children who are weakest at spelling, and for whom ‘does it look right?’ is most encouraged. are the least well-equipped to use it. Far better to explicitly teach them the phonic, orthographic and morphological knowledge they need to spell the word correctly, rather than have struggling spellers intuit half-remembered spellings from an incomplete data set of spelling information and word knowledge stored in long-term memory.  

6. word searches and crosswords

Ah, the word search and the crossword. The mainstay of many a terrible spelling programme. Let’s be clear here, these strategies are no more than time-fillers, busy work, the last resort of a spelling programme devoid any genuine substance. While it might be true that during a word search children are searching for letters within words, they are looking at those letters individually,  regardless of whether they are part of a two-, three- or four-letter spelling. This is true of crosswords too: they separate out multi-letter spellings. When spelling charge, I don’t want children seeing and thinking c h a r g e, I want them to see and think ch ar ge. Working at the level of individual sounds and not individual letters is the key that unlocks spelling. Word searches and crosswords fall woefully short.

And there’s a second reason that these approaches are poor proxies for good spelling teaching – it’s back to independent writing. When children are writing independently, I want them to be able to draw upon the knowledge and strategies I’ve taught them –  saying words clearly in a spelling voice, snipping longer words into syllables, considering the relationships between letters and sounds, saying sounds as they write letters. When a child, writing that fairytale during independent writing, can’t recall how to spell princess, the one thing I can absolutely guarantee they won’t be thinking about is that word search you did last week.  

7. backwards spelling and other horror stories

Here follows a list of nonsense, all of which are a click away from any Google search on spelling strategies, none of which have any merit at all, none of which has any place in the teaching of spelling. If these are part of your school’s regular spelling practice, please stop.

Backwards spelling. Writing words backwards. Oh come now, how on earth does writing scream as maercs help children’s spelling? If there’s one basic thing I’d hope was happening in a spelling lesson, it would be having kids write the letters in the right order. It’s a low bar I know, but this fails to clear even that.

Pyramid spelling [fig.3]. I know, let’s learn to spell scream by writing the first letter 6 times and the final letter once. No, thank you. And with <scre> we are separating the <e> from <a> when children need to understand <ea> as a two-letter spelling representing /ee/.


Rainbow spelling. rainbow rainbow rainbow rainbow rainbow. If someone can tell me what doing this adds to the teaching of spelling, I’d love to know. All it seems to do is divert children away form thinking about spelling and onto which colour pencil they’ll use next. Or rainbow. Let’s not only have children spend timing thinking about which pencil to pick up next, but now let’s have them do it mid-word or even mid-grapheme. I might understand a little better if it were rainbow, each grapheme a different colour, but it never is.

Pictowords [fig.4]. I don’t have any problem at all with this if the aim is to improve vocabulary (drawing is a fabulous way to cement new words into children’s memories) but I remain baffled as to why drawing crowns above the <o> and <a> is an aid to spelling. It’s not as if an image of a crown has any particular relationship with those two letters. As a tool for improving vocabulary, sure. As one for improving spelling, not at all.


Join the Dots. Children write each letter in dots and then join them up, usually in a different colour. Because.

Red Vowels. Or any colour that tickles your fancy. Children write words but colour the vowels in a different colour. Now, I have no problem with separating spellings into consonants and vowels. When teaching Sounds & Syllables spelling lessons, I’ll often build words by placing graphemes on coloured post-its, and I use one colour for consonants and another for vowels. It’s helpful when spelling multi-syllable words, as each syllable will contain one, and only one vowel spelling. But this is the first example of red vowel spelling I stumbled across online: dictionary. Gasp! First let’s split the word into its syllables: dic | tio | na | ry (you can argue with me about my syllable division another time). There are two red vowels in one syllable and none in another. And that’s because the definition of vowel that matters most for spelling is that which is related to sound. In dictionary, <i> is part of the two letter spelling <ti> which represents /sh/ so the vowel sound in this syllable is the schwa represented by <o>. And there are no red vowels at all in the final syllable, but of course the letter <y> represents the vowel sound /ee/, as it does when it ends most multi-syllable words. So, dictionary ought to look this: dictionary. Or even better, like this: d i c | ti o | n a | r y.

Heaven Scent. I can’t quite believe that this one even exists. Apparently a squirt of perfume or aftershave on a spelling flash card will trigger the memory of a spelling. Years laters, Jo meets the love of his life, catches a faint trace of her scent in the moonlit air and thinks to himself <cemetery>. For the love of all that is good in the world, I hope nobody does this.  

Thankfully, there are better ways of teaching children to spell. If you’d like to know more about the Sounds & Syllables approach to spelling, please click on the image below.

To subscribe to this blog please fill out the form below.

About the Author

Leave a Reply