Sunday, 3 November 2019

Trick or treat? More nonsense words about phonics instruction and assessment.

Late last week (by curious coincidence, on Halloween in fact) University of Canberra Adjunct Associate Professor Misty Adoniou released an opinion piece making a number of unsubstantiated, often emotive claims that cannot go unchallenged. Apparently Misty thinks that her adjunct status as an academic liberates her from the need to provide evidence for the claims she makes about early reading instruction. This is, however, not the case, so I have done my best here to highlight gaps and flaws in her piece. 

Now I am neither superstitious nor religious, but perhaps it’s timely to remember that Halloween (Hallowed Evening) is followed on November 1 by All Saints Day – a day to move beyond the scare mongering and strange fictions that find their way into people’s minds on October 31. My favourite saint has to be St Skeptikus, and yes, I did just make her up, but if she was real, she would want us all to honour her name by questioning the trick-or-treat “facts” presented here without substantiation.

Misty’s assertion
My response
Anglophone countries have been struggling for years with declining achievement in reading and writing as students move through primary school and into high school.
Agreed. But the problem here is not so much the English writing system, as the approaches that have been cemented into practice in teacher pre-service education and early years’ classrooms. I have blogged about this previously here and also here.
In 2019 more than 25 per cent of Year 6 students in England failed to reached the minimum requirements in the annual national reading and writing assessments. This means around 1 in 4 students in England are leaving primary school ill-equipped to cope with the literacy demands of high school. Similar statistics are reported in Australia and the United States.
The important thing here is to look at the trend over time, not merely at one dot-point on a graph. In 2018, 32% of Year 6 children did not reach this benchmark, and in 2017, it was 43% - getting up uncomfortably close to half. 

The reading comprehension shift over the last three years is clearly a positive one.

The source for the figures is here. See also my graph below, which makes this trend even clearer.
England has decided the problem is that students do not have the skills to decode basic texts. So, in 2011 they instituted a mandatory Phonics Screening Check for Year 1 students. Students are prepared for the test through government-approved commercial, synthetic phonics programs. Those who fail the test receive more phonics instruction and re-take the test in Year 2. The rationale is, if we can get the basics right, the rest will follow. Eight years on, it is clear the rationale is flawed.
England decided that part of the problem is that children do not always have the skills to decode text, in line with the emphasis in its 2006 inquiry into teaching reading, authored by Sir Jim Rose, which had a strong theoretical foundation in the Simple View of Reading (SVR). The SVR has been around since 1986, so it’s about time everyone understood that it is concerned with more than simply decoding (identifying) words. It is equally concerned with a child’s ability to understand words once they have been identified.

Misty begins her blogpost with a list of words that are deemed to be somehow “broken” or “damaging” to young children because they are technically non-words.

Good luck to children then when they advance to reading Dr Seuss, Lewis Carroll, JK Rowling, and Roald Dahl (to name a few). No delights in quidditch for children in Misty’s class – they can presumably only read words that are already in a formal dictionary somewhere.
But why hasn’t this phonics ‘first, fast and furious’ approach worked? And are there lessons for those who teach English language learners?
But why is Misty overlooking the evidence that an increased emphasis on phonics instruction has resulted in improved decoding skills?

Children getting better at what Misty herself agrees is an essential skill is a good thing, that should be celebrated. Isn’t it?

It's not the whole story and we still have quite some work to do, but shouldn’t we be celebrating wins along the way rather than dismissing them as irrelevant?

The teachers who have worked hard to achieve these gains deserve better than this. 
The first flaw in the English phonics solution to declining literacy achievement is that English is not a phonetic language. Whilst learning English letters and their common corresponding sounds is necessary, it isn’t sufficient to read and write in English.
Agreed. English is not a “phonetic” language. However, it is also not an opaque mystery, in which there are no rule-governed patterns in how sounds and letters work together.

This is the “because Englishhoary old chestnut that Whole Language and Balanced Literacy advocates fall back on as a defence against systematic rather than incidental teaching of how the English writing system works.

Interestingly, Misty’s second sentence here is a perfect endorsement of the SVR: learning English sound-letter correspondences is necessary but not sufficient

Isn’t that an essential principle that the SVR gave us more than 30 years ago?
English is a morpho-phonemic language. We write ‘jumped’ not ‘jumt’ because we know ‘ed’ is a suffix (morpheme) that marks the past, regardless of the ‘t’ sound (phoneme) we hear on the end of the word. We write ‘action’ not ‘acshun’, because we know ‘ion’ is a suffix that makes nouns – regardless of the sounds we hear.
Correct, and I agree completely. However, our teaching workforce, is not, in the main, currently equipped to teach reading drawing on the benefit of this knowledge, because their pre-service education fails to prepare them to do this.

“The consequences of a lack of content knowledge in teaching literacy can be serious, with Shulman (1986) indicating that lack of content knowledge results in narrowed and regressionist pedagogies as teachers resort to replicating own past experiences with instruction in language. In particular, to be effective in teaching children who struggle with literacy, they need a strong content knowledge of the English language (Spear-Swerling & Cheesman, 2012). Numerous accounts of beginning teachers note a lack of content knowledge about how the language works – most particularly, the basic constructs of the English language (Alderson & Hudson, 2013; Hadjioannou & Hutchinson, 2010; Moats et al., 2010; Washburn, Joshi, & Cantrell, 2011; Wong, Chong, Choy, & Lim, 2012). Spear-Swerling and Cheesman (2012) suggest that without good content knowledge in the area of literacy "teachers may provide inadvertently confusing instruction to children” (Spear-Swerling & Cheesman, 2012, p. 1692). 

English language learners need the whole linguistic picture of how words work when learning to read and write – when the focus is purely on phonics, half the clues are missing. That’s an inefficient way to learn a language, and an unfair burden for English language learners.
This is the persistent straw man argument that phonics is taught in isolation of meaning. The SVR, as noted above (and endlessly for the last 30+ years) promotes the importance of background knowledge and language comprehension.

Misty (and other readers of this blog) might be interested to know that the SVR has recently been updated, as the cognitive foundations of learning to read.

So let’s hear it for not trotting out this tired meme any more. 

It’s also ironic, given the emphasis on “meaning-based instruction” promoted by Whole Language and Balanced Literacy advocates. Isn’t this what has been the mainstay of early years instruction for the last three decades? If reading instruction is not working, this might be one of the first rocks we’d want to turn over, to find out why.

The second flaw in the English solution is that it promotes ‘synthetic’ instructional practices for phonics. This means students learn individual sounds for each letter which they then ‘synthesise’ together. In this approach, word meaning is considered irrelevant – even a hinderance (sic)– to initial instruction as it may distract students from attending to the individual letters in words.
Where is the evidence please that word meaning is considered “irrelevant”, let alone a “hindrance” when teachers are using a synthetic phonics approach?
To mitigate the possibility that the students are using meaning, context or prior exposure to the word, synthetic phonics instruction involves teaching the children to read ‘nonsense’ words, like ‘flug’ or ‘pob’.
Where is the evidence please that teachers teach children to read “nonsense” words as part of systematic synthetic phonics (SSP) instruction? Children who have been taught to be effective decoders should be able to decode words, irrespective of their score on Misty’s Mysterious Meaning Metric.

If teachers are specifically teaching so-called "nonsense" words, then they don’t fully understand the basis of SSP instruction, and need to have this clarified for them.

Let’s not forget too, that children do not have a psycholinguistic filter that automatically tells them a word is “real” or “nonsense” (notwithstanding the non-binary nature of this distinction anyway). Every new, unfamiliar word a child encounters might just as well be a nonsense word, if s/he does not yet know its meaning.

And then there’s all the new words that enter our language every year, because English is a dynamic, living language, that continues to evolve; think Google, selfie, mansplain, as examples.
An approach to reading and writing instruction that eschews meaning seems particularly perverse for any learner, but for English language learners (ELLs) a synthetic phonics approach is particularly harmful.
This is an extraordinary, pernicious, and dangerous claim that will arouse unnecessary anxiety for parents and teachers alike. 

This statement also conveniently ignores published evidence that explicit phonics instruction can actually be helpful to children from non-English speaking backgrounds

Misty should cite her sources for her biased, non evidence-based  claim or retract it and apologise for the distress and confusion it will inevitably cause.
When learning a new language, all words are potentially nonsense words.
As noted above, this applies to children whose first language is English also. I do not know what Misty’s point is here, but she does appear to have changed corners in the debate.
English language learners need a focus on word meanings, and to be taught new – real – vocabulary. It is hard to imagine a more useless practice for ELLs than deliberately teaching them to read nonsense.
All children need a focus on the linguistic basis of written words – the sound structure they represent and their meanings. It should never be either-or, as Misty chronically attempts to assert.

What does “deliberately teaching children to read nonsense” actually look like? Might it look like supporting them to read a Dr Seuss book for example?
A third flaw in the English phonics first approach is that it is exclusionary, and possibly racist. Native speaker accents are privileged. Some allowances are made for regional native English accents.
This claim is simply alarmist and lacks substantiation. Again - evidence please.
Of course, the ultimate flaw in the English plan to improve reading and writing skills is that it isn’t working. The students are not getting better at reading and writing. The phonics ‘jab’ they receive in the first two years of school simply hasn’t been able to ward off the dreaded middle years decline in literacy achievement. They are failing their Year 6 reading tests – despite knowing their sounds. The US may have made the right call on this problem – the middle years literacy decline is not because basic decoding skills are lacking, it is because complex comprehension skills are missing.
Let’s put the pejorative language to one side here (though vaccinations are generally a good thing in public health sense and prevent a great deal of downstream harm).

Instead, we can do some simple fact checking. UK children are NOT "failing Year 6 tests" at all.

To make this easy, I have produced a graph of the last three years, showing steady improvement on this dimension:

(C) Pamela Snow (2019)


  1. Hi Pamela, I’m a bit unclear about the prediction of the simple view of reading and implications for teaching. Isn’t it the case that according to the simple view that if a child knows the meaning of a spoken word then they will understand it if they can decode it. And if they are good at decoding then they should be able to read and understand text as long as they would understand the text when read aloud by someone else? So on this view, the main role of teaching meaning in the context of the simple view of reading is to teach children an oral vocabulary and understanding speech. Decoding then “unlocks” all this knowledge. Or is there something more to meaning in the context of teaching reading?

    And how do you recommend teaching meaning of written words in the context of the simple view and an early emphasis on phonics. The only place I see meaning playing a role in early systematic phonics instruction is when the teacher asks children to think about the meaning of a word that they have decoded. That should come automatically according for familiar words according to the simple view, but they need to be told the meaning if it is an unfamiliar word. Is there any other forms of meaning instruction associated with written words with SSP or am I misunderstanding how meaning instruction is combined with systematic phonics at the initial stages of reading.

    Thanks for much for any clarification. I’ve been criticizing phonics for the fact that it largely ignores meaning of written words at the stat of instruction. But I’m unclear whether I’m missing something about early meaning instruction, or whether the point is that yes meaning is important, but it comes after GPCs.

    1. Hello Jeff. Synthetic Phonics (SP) takes reading comprehension more seriously than any other type of reading instruction. The Simple View regards reading comprehension (RC) as the product of decoding (D) and language comprehension (LC). D and LC are, individually, necessary but insufficient conditions for reading comprehension to occur.

      To foster RC as efficiently as possible, SP teaches both D and LC together from the beginning of instruction. Half of each class is devoted to explicit instruction on GPCs and decoding (D); the other half is devoted entirely to increasing language comprehension (LC). The latter is accomplished by the teacher reading quality children's literature TO the class and then discussing it WITH the class.

      See my blog at for more detailed information.

    2. Hi Stephen, thanks for this. But I think the RC here is a complement to SP, not part of SP. That is of course fine, but it I don't think this is part of the definition of any form of phonics instruction. Still, if I understand you, this combined approach gives little or no consideration of the meaning of the written word at the start. When you are reading written words you are focused on decoding, and then, separately, you are improving verbal language, with the assumption that is the best way to teach reading. Is this fair?

  2. Jeff, reading comprehension is a part of SP in this sense: without the decoding (sounding out) skill that SP quickly and efficiently provides, reading comprehension is utterly impossible.

    In this regard I'll quote Max Coltheart because I can't say it better than he already has: "Children do not need to be taught the meanings of words when being taught to read, because they already know those meanings. What they need to be able to do is to go from print to those already-known meanings. What makes you think that proponents of phonics seek to do this [teach the meaning of written words]? They don't."

    And this too Jeff: "Young children who are in the process of learning to read will already have an auditory vocabulary of 10,000 words or more (Shipley & McAfee, 2015; cited by Law et al., 2017). Given this, how often would a word that is in text that the beginning reader is given to read not be a word that is already familiar to that child in speech? Virtually never. It's a matter of probability."

    It still seems to me, Jeff, that you do not fully understand the Simple View of Reading.

  3. Thanks for the quotes. Can you tell me where you got the Coltheart one?

    I think these quotes are very much in line with what I'm saying. Phonics (and SP) does not teach children the meaning of words because they already know them. That is the motivation of focusing on decoding in decoding. Yes I agree that children cannot understand written words if they don't know the meaning of the word when spoken. Why do you think I don't understand, perhaps I'm still missing something, but I think I agree with both of the quotes you have provided.

    Thanks for the Coltheart quote reference!

  4. This comment has been removed by the author.

  5. Jeff, both quotes are from Max. They're from a bulletin board posting he made to other reading reformers. If you'd like to join this group, email me. My email address is on my website

  6. kerry hempenstall5 November 2019 at 19:47

    In relation to the claim that synthetic phonics is "particularly harmful", some of the evidence to the contrary is offered:

    “The effectiveness of this [synthetic phonics] approach has been proven for the development of early reading skills (Dixon et al., 2011; Johnston, McGeown, & Watson, 2012) among struggling readers in the ESL context (Lesaux & Siegel, 2003; Vadasy & Sanders, 2011; Yeung, Siegel, & Chan, 2013). These studies employed experimental research designs for different age groups: kindergarten students (Lesaux & Siegel, 2003; Yeung et al., 2013) and primary school students (Dixon et al., 2011; Johnston et al., 2012; Vadasy & Sanders, 2011).” (p.458)

    Jamaludin, K.A., Alias, N., Khir, R.J.M., DeWitt, D., & Kenayathula, H.B. (2016). The effectiveness of synthetic phonics in the development of early reading skills among struggling young ESL readers. School Effectiveness and School Improvement, 27(3), 455-470.
    “In a research update on reading instruction for English learners that included 20 studies published since 2002, August and Shanahan (2010) reviewed 12 additional studies on phonological awareness and phonics instruction. Similar to the Shanahan and Beck review, theirs found that systematic and explicit instruction has a positive impact on a range of student reading outcomes. English learners develop word-level skills in a similar manner to native English speakers, and they benefit from instructional features that have been found effective for native English-speaking children (Shanahan & Beck, 2006). Several early reading interventions for English learners with a strong phonics component have reported benefits for word-level skills (Denton, Anthony, Parker, & Hasbrouck, 2004; Gunn, Biglan, Smolkowski, & Ary, 2000). This growing body of research suggests that young English learners benefit from the same instructional features and phonics components that support early reading development in their native English-speaking peers (Ehri et al., 2001)” (p. 2).

    Vadasy, P. F., & Sanders, E. A. (2012). Two-year follow-up of a kindergarten phonics intervention for English learners and native English speakers: Contextualizing treatment impacts by classroom literacy instruction. Journal of Educational Psychology, 104(4), 987-1005.
    “The beginning reading programs with the strongest evidence of effectiveness in this (ELL) review made use of systematic phonics.”
    Slavin, R.E., & Cheung, A. (2003). Effective reading programs for English language learners: A best-evidence synthesis. Center for Research on the Education of Students Placed at Risk.


  7. Jeff:

    You ‘agree that children cannot understand written words if they don’t understand the meaning of the word when spoken’ and ask why you should be thought not to understand this. It’s because you don’t emphasise the converse – i.e. that if children accurately decode written words into spoken words whose meaning is already familiar, they WILL understand those words.

    You open your ‘Reconsidering the evidence…’ article with the following:
    ‘There is a widespread consensus in the research community that early reading instruction in English should emphasize systematic phonics. That is, initial reading instruction should explicitly and systematically teach letter (grapheme) to sound (phoneme) correspondences before focusing on the meaning of written words in isolation and in text.’

    Why use the word ‘before’ if you know that phonics teachers provide texts for beginners to read in which all or most of the words are already understood in their spoken form? Phonics teachers expect children to understand all such written words AS they decode them into spoken words. The teachers explain meanings if they realise children don’t already know them and may well deliberately introduce some unfamiliar words for the sake of increasing children’s vocabulary, but the great preponderance of words whose meanings are already known in texts used with young children means that it’s wrong to imply that the focus on meaning comes only later.


  8. Hi Jenny, the key part of my quote is: "before focusing on the meaning of written words in isolation and in text". Note the word *written*. Yes, children learn the meaning of spoken words, largely from home, and as far as I know, there is no systematic way in which meaning of spoken words is taught in schools. So phonics focuses on GPCs (like the quote from Max), with minimal teaching of meaning. SWI does explicitly teach GPC, but in the context of morphological matrices where the meaning of a set of morphologically related words are studied. Here, children are being taught meaning of *written* words, and learning GPCs in this context.

    1. Hi Jeff,

      I’m noting the word ‘written’ a you ask, but am still puzzled. It looks as if you are using ‘focusing on’ to mean ‘teaching’, and are therefore saying that phonics teaches GPCs before teaching the meaning of written words. But written words have the same meanings as their spoken counterparts, and if children can use phonics to translate written words, however many morphemes they have, into spoken words whose meanings they already know, there is no need for teachers to teach the meanings of those written words separately. In those circumstances, however, not teaching meaning doesn’t mean that there is no ‘focus’ on it – teachers use questions and discussion to make sure that children understand what they read.

      Would you use a morphological matrix to teach GPCs from scratch to children who as yet knew no GPCs? Can you give an example?

  9. Reading this article (for the first time) is so cathartic. It is the likes of the teaching of professors like Misty Adoniou who made me a horrible teacher for many years. (I got straight A's in my well-established TESOL graduate program, so who knew?!) Stumbling upon Alison Clarke a few years ago led to an epiphany that has changed my career, transformed the reading success of my ESL students (and made me prone to possibly unpopular ranting at professional gatherings). Thank you so much Pamela Snow for writing this article!