Real Spellers

English Makes Sense!

Important News!

Have you been wishing for access to the ultimate Orthographic resource?

The RealSpellers™ Toolbox is moving online!
Check it out.


In case the video above does not work, click HERE to see this new video on the crucial  place of explicit instruction about grapheme-phoneme correspondences in structured word inquiry (SWI) in general, but particularly in early literacy instruction, and for students struggling in literacy. 

The interest in and awareness of SWI has seen a rapid growth with practitioners and researchers in the last couple of years. More and more people are hearing about it, but most of those who hear about it have not studied it carefully themselves with myself or the many other excellent people now providing workshops and courses related to SWI. 

The topic of reading instruction has long been a particularly polarizing topic. People who first encounter SWI who have developed their views on literacy instruction in the context of the phonics vs whole language “reading wars” naturally expect this initiative to be on one of those two “sides” or like “balanced literacy” instruction which defines itself by being somewhere between those sides. 

I totally understand why people would assume SWI fits somwhere within this frame of debate  -- but that doesn’t make that assumption true. 

A statement I often repeat is that SWI is not a response to whole language or phonics. It is a response to the proposition that we should do our best to ensure that literacy instruction is guided by and reflects our best current understanding of how our orthography system works, and that we should do so from the very start of formal schooling.

But this argument makes little sense to anyone who has not had the opportunity to study the interrelation of morphology, etymology and phonology themselves -- or to people who have no image of what such instruction could possibly look like in the early years. 

In the last couple months, I’ve received a number of emails from colleagues sharing what they are hearing from people at conferences and on social networks about SWI. 

Here’s an example that represents a common misunderstanding:

“Well, it’s fine for older kids but it doesn’t work with our younger students who need phonemic awareness and and explicit instruction of letter-sound correspondences.”

Notice the assumption here is that children learning in an SWI context from the beginning would not get instruction in phonemic awareness or explicit instruction of grapheme-phoneme correspondences - or at least that these aspects of literacy instruction would be significantly under emphasized.

Here is another example passed on from another colleague who was talking to someone at a recent conference when SWI came up. That person stated:

“I hear that Pete Bowers doesn’t even think you need to teach grapheme-phoneme correspondences.”

This statement made me realize just how far characterizations of SWI have veered away from any sense of reality.

Whoever heard that, certainly has never studied anything I’ve ever written or published. Such extreme mischaracterizations of SWI being passed around from people with no direct experience with SWI is too serious a problem to be left uncontested. People whose first encounter with SWI is through such misinformation have reason to biased against anything to do with SWI, and are likely to share that misinformation themselves.

Regardless of how clear I think I’ve been on this topic so far, my efforts to counter this misunderstanding have apparently not been very successful! I am happy for people to criticize SWI -- as they should any instructional approach. I regularly learn from critical analysis of my current understanding and practice. But that is only true if those criticisms are actually about what SWI actually proposes regarding instruction. I’m happy for people to reject SWI -- but they should do so based on what it is -- not based on demonstrably false statements being passed around conversations at conferences and through social media.

This video - that I hope will be the first in a series of videos addressing this topic - is part of a new attempt to address this issue as clearly as I can manage. 

It has at least two purposes.

Firstly, I hope is effective conveying the essential nature of grapheme-phoneme correspondence from the start -- and what such instruction can look like.

Secondly, I hope that when people interested in SWI hear others make claims like those above, they have a link to this video they can pass on. The explanation of the role of grapheme-phoneme correspondences in SWI is crucial, but that doesn’t mean that it is easy to articulate. This video attempts provide a direct illustration of how grapheme-phoneme correspondences are addressed from the beginning of formal instruction in SWI.

But even this video is only a brief introduction. In preparing for this video, I received an email from someone new to SWI that I think conveys just how difficult it can be to understand this aspect of SWI until you actually see it in action. Kate wrote:

“ someone who has come to SWI very recently (I just discovered it this summer and have thrown myself into it rather whole heartedly), it wasn’t until I was able to watch you teach a group of students, Pete, that I came to understand just how (and how much) the grapheme/phoneme relationships can be a part of SWI.

We were watching you teach [a Grade 1 class in the “classroom embedded” portion of] the workshop at [my school] in Toronto and I remember turning to the woman who was an ECE expert from York university and both of us said “Oh, that’s how you work the phonological knowledge into the lesson.” As we watched you using the tapping and spelling out loud, as you explained digraphs and trigraphs, we could see how much of that phoneme/grapheme awareness could be just naturally built into the discussion.

I’d seen the phoneme/grapheme charts that you and other teachers had built with their students as part of their investigations but I didn’t really understand how all that fit with the word sums and the matrices and looking up words on Etymonline until I saw you in action with a group of kids.”

Kate is relatively new to SWI, but she had been studying for a while before she saw me teach Grade 1 students at her school. As she notes, she had to see it in person with kids to really start to get this aspect of SWI. In the video above, you don’t get to see me teach students, but you do see me apply exactly the strategies and tools she describes such as “tapping out graphemes” and “spelling-out” digraphs and trigraphs, the use of grapheme-phoneme charts and more. In future videos I hope to include examples of students engaging in this instruction.  In some of the resources linked below that have been around for years, you can see students engaging in exactly these processes. 

This message from Kate was extremely important for me to hear -- and I think it is important for people new to SWI to hear before they draw their own conclusions.

Here is someone interested in SWI. She’s been working at it, and she still needed to see it in action to really understand how SWI integrates and emphasizes the instruction of grapheme-phoneme correspondence from the start. If Kate needed this kind of practical exposure to get this point, people with less initial interest are even less likely to "get it" wihtout a clear enough window on what SWI really is. This video, the resources below and future videos on this topic represent a renewed effort on my part to help make this aspect of SWI clearer for anyone who is interested. If what I’m currently doing is not doing the job, clearly it is time to try another way to do it!

I have included many links to resources and examples of the crucial nature of grapheme-phoneme correspondence in the description on the YouTube video. I will include those below as well.

I welcome critical responses in Real Spellers or via email <This email address is being protected from spambots. You need JavaScript enabled to view it.> to what you see in the video. That feedback can help inform the next videos I plan to produce in this same effort.

For more examples of instruction about orthographic phonology in SWI and resources to support understanding of grapheme-phoneme correspondences and how they operate in English, see additional links below:

1) See a video (from 2012!) in which I address the grapheme-phoneme correspondence of the <s> in the words "chicks" and "eggs" within the context of introducting them to word sums and the matrix for the first time. 

Screen Shot 2019 12 05 at 3.55.02 PM

See a short 2.5 minute version of that video HERE. (This one only gets to word sums, not the matrix)

See the full 15 min video HERE. (This one has more examples and does get to introducing word sums, and introduces the matrix)

2) Resources on the process of "spelling-out-loud" and "writing-out-loud" word structure including announcing graphemes and orthographic markers in the base. 

Click HERE for the page on my website with a detailed reference and resources to guide the use of "spelling-out-loud" word structure which requires identifying graphemes and phonemes as a basic everyday practice in SWI lessons. 

Screen Shot 2019 12 05 at 3.58.58 PM

3) The LEX grapheme deck (Click HERE to learn more and to order from LEX)

I recommend Gina Cooke's grapheme deck at every workshop I ever give. The process of "spelling-out-loud" and "writing-out-loud" is one that forces teachers to discover when they themselves are not sure of what the graphemes are in a word that they are attempting to teach. Central to on-going learning is a process that keeps identifying what the learner does not yet know! But that process requires a reliable reference to draw on. I know of no other reference for grapheme-phoneme correspondences that is so rigorously accurate linguistically, and which provides so much detail on morphological and etymological influences. I cannot recommend this resource enough. This is why I point to it at every workshop I ever give. The crucial nature of this resource to SWI is just futher evidence that grapheme-phoneme correspondences are central to SWI.

Screen Shot 2019 12 05 at 4.03.17 PM

3) Click HERE for a video from 2011 in Skot Caldwell's public school classroom where he takes students through a lesson integrating morphology and phonology with a whole wall of grapheme-phoneme correspondences they are exploring scientifically. 

4) Click HERE for a grade 2 teacher at Nueva walks us through structures in his classroom for investigating words including a chart on grapheme-phoneme correspondences they are studying including their positional constraints. 

Screen Shot 2019 12 05 at 4.10.34 PM

There are countless other places to point to where SWI emphasizes grapheme-phoneme correspondences, but this is a start. 

I hope people with register with Real Spellers and add their thoughts in the comment string, but also feel free to email me directly at This email address is being protected from spambots. You need JavaScript enabled to view it. if you have questions.

Comments (6)

This comment was minimized by the moderator on the site

Pete, What have you seen in terms of revisiting the graphemes and phonemes, which have been discovered? Some kids need more repetition.

Debra Geise
This comment was minimized by the moderator on the site

I think these are great questions, Sandie. I'd like to add to Pete's comments about sequence.

Last week, a young child asked me, "Why does <kitty> start with a <k> and <cat> starts with a <c> but they both sound like /k/?" What do we do with a question like that before either <c> or <k> has been mastered?

The very next week, this same child wanted to label a picture of her own cat, whose name is <Echo>. So there we were, with the Greek origin <ch> that spells /k/ arriving on the scene before the <ch> in <chat> has been encountered. I helped the child isolate the <ch> spelling for /k/ in <Echo>, and she practiced writing and spelling her cat's name. I told her that /k/ was not the only pronunciation for <ch> - and left her wondering what others there might be.

When I was an OG practitioner, my students didn't take me down these paths with their questions. I determined the sequence myself. But now I think the questions are the key. In the course of a week, this child was exposed to important principles of phoneme/grapheme correspondence, and spellings of /k/ and pronunciations of <ch> have become subjects for eager exploration.

Gail Venable
This comment was minimized by the moderator on the site

Gail Venable,

What a lovely experience for this child and you!

Debra Geise
This comment was minimized by the moderator on the site

Pete, thank you for this video and the successive ones that will follow. Over the years, a few of my students have had severe dyslexia along with "comorbid" conditions. They have needed a lot of repetition and exposure to the same grapheme and phoneme correspondence. How have classroom teachers, reading specialists, tutors, etc. addressed that? The video of the second-grade teacher from Nueva somewhat referred to that. I do those things too, as well as the whole swi process, but there are some students that need more than that. Are there other suggestions? Keep in mind some of us only have students 2 times a week. On another note, I think one crucial thing that you did was to connect the phoneme to the grapheme with the spelling of cat while reading. Great modeling. It may seem simplistic to you but it's gold to some. I think that some people feel that it is taboo to teach grapheme phoneme correspondence outside of the phonology piece. The thought is that we only spell the word when reading and if the word is unknown we tell them. And you explained the rationale because children are tuned in to the graphemes and can check themselves of there are inaccuracies. On top of that, one can't get around the fact that graphemes pinpoint phonemes, unless their null or markers of some sort, in the words we are reading. I'm not saying to do that for every word or that there is only one way. I am saying that there is some confusion or lack of clarity in this area. I'm not sure if I explained that point well but we can go from here. Thanks, Pete, for all you do and the graciousness with which you do it! Cheers!!

Debra Geise
This comment was minimized by the moderator on the site

Many thanks for your thoughtful comments and questions Sandie. See my responses to your questions below:

1) Do you ever practice raising students' awareness of phonemes in the absence of graphemes? (Either way, I'd be interested in the science that underpins your rationale.)

It would be very rare that I would find a reason to focus on phonemes in the absence of graphemes. Perhaps if I was working with a child that was not a native English speaker or had some speech issue, I might highlight how a phoneme "feels" if they were pronouncing a word in a way that made it difficult to understand and we were not in the context of print.

One thing about phonemes is that they are inherently abstract -- they are literally abstract concepts that can be realized in speech that speakers of a language perceive as representing the same phoneme. Graphemes are also abstract, but at least they have a concrete representation that teachers and students call all look at together and then associate with that pronunciation. Once you have announced a phoneme, it's gone. By "feeling" what is involved in one's mouth to announce a phoneme we make an abstract thing more concrete, and by tying that pronunciation to a grapheme we make it more concrete. By doing all of this in the context of a word the learner is familiar with, we place all of that in a meaningful context for the learner.

The basic principle driving all of these choices for me is the assumption that when we try to help learners understand something that is abstract, it is helpful to provide concrete representations in a meaningful context. The comparison I use is introducing children to the number sentence for the first time. I don't know any instructional argument for introducing addition by writing the number sentence: 2 + 3 = 5.

For kids who don't know what those "squiggles" mean, this is utterly meaningless. So we don't start with the abstract representation of addition, we make the concept concrete by using counters. We might put two counters under the digit 2, three counters under the digit 3 and draw attention to the fact that the squiggle that looks like this "2" stands for "two things" and the "3" squiggle stands for three things. We then count all the "counters" and seeing that when we count to and then three more, we have a total of "five" that is represented by the squiggle that looks like "5".

Having a way of marking abstract things makes it easier to pinpoint them and discuss them. The "squiggles" whether they are letters, graphemes or digits are concrete and don't disappear in the ether after they are produced like phonemes do.

This is also why IPA symbols are so important. They directly represent pronunciations. Being able to point at a symbol like /t/ and announce that phoneme and associate that pronunciation with the <t> grapheme in <act> makes all of this more concrete than if I focused on the /t/ without a written grapheme or a meaningful word. And then when we think about the word "action" and note that in this word that same <t> grapheme is representing a phoneme /ʃ/ (That many have been taught to be the "sh sound") helps us see that this one grapheme can represent at least two different phonemes. Being aware of the difference in those pronunciations (phonemic awareness) is supported by having access to two demonstrably different physical symbols. And if we discuss the meaning connection between words like "act" "actor" "acting" and "action" with the help of a matrix and word sums, we are now attending to these grapheme-phoneme correspondences not only with concrete representations for the changing pronunciation with consistent spelling, we are doing so in a context that helps us understand WHY we use the <t> grapheme in <action> rather than the <sh> that writes that phoneme in other words. The <t> allows us to mark the meaningful connection between <act> and <action> with similar spellings even though the phonology of that base changes.

Of course I don't get into matrices and word sums every time I feel the need to draw attention to a phoneme. That is when I choose to do a more involved orthographic investigation. If I was just reading a story that had the word <action> in it, and a student got stuck recognizing the word because they were pronouncing the <t> as /t/, I would have a great opportunity to orally point out that this word "action" has a base <act>, and in that word the <t> writes /t/, but this is the word they know "action" where that <t> writes /ʃ/. I'm not sure how any of that would be aided by NOT having graphemes in front of us to pinpoint the phonemes of interest.

2) Based on your demo, I assume that you don't follow any specific sequence in teaching p-g correspondences; rather, the teaching targets p-g correspondence(s) are determined by a student's errors or confusions. So, is it safe to say you advocate an "embedded" versus "sequential" approach to teaching p-g correspondences? What if, in discussing "chat", a student said that <ch> is pronounced /k/ (which, of course, it can be). Would you have launched into a discussion about that (e.g., in what words we see that correspondence, etc.) or would you have just acknowledged it and re-focused on the correspondence in the word at hand: "chat"? I ask because the tyranny of time means that how time is spent during lessons is incredibly important! (Again, I'd be interested in the science that underpins your rationale either way.)

It's true that I do not follow a sequence of which graphemes to teach when. I also have never seen any evidence about one sequence that is more effective than another. But I do see that most sequences I see focus for a long time on single letter graphemes before digraphs, and digraphs before trigraphs. I assume this is based on an (untested) assumption that this makes it easier for kids. But if we really followed that order, we'd have to teach the <z> before we teach the <th> or the <ea> or the <ch> or <igh>. If the idea is that we should teach the common graphemes first, this common sequence of single-letter graphemes before digraphs, before trigraphs fails on that count. It turns out there are way more digraphs than trigraphs! The current instruction I see also only teaches some phonemes for a grapheme and ignores others. Note the <t> writes /ʃ/ incredibly frequently, but that grapheme-phoneme correspondence is so rarely taught at all! I would argue that this is because people don't even notice unless the gather words in morphological families that highlight this correspondence: act/action, relate/relation, and countless others. If instead, we texts that kids are interested in, and we are very explicit about teaching graphemes in words that kids are interested in, we can't help but encounter the most common graphemes! And that frequency will not be based on the number of letters in the grapheme.

Some key principles of grapheme-phoneme correspondences I want kids to be familiar with asap, include:
- Graphemes are comprised of either single letters, or groups of two- or three-letters.
- Most graphemes can represent more than one phoneme
- Most phonemes are represented by more than one grapheme
- We can often see meaningful reasons for grapheme choice when we look at related words (which include morphological and etymological influences)

I've never had the experience of a child struggling to learn the <th> digraph when I spell those two letters out "th" in words like <the> <path> or <this> and draw attention to the phonemes it represents. The <igh> is easy to draw attention to in words like <night> or <fight>. And I also find no difficulty teaching these letter names at the same time that they are learning them in graphemes in words.

You ask about what if the kid thought the <ch> in <chat> was pronounced /k/. That would be great! I might say something like... "Good on you! The <ch> can write that /k/ phoneme in words like <school>. But this is not the word "cat" like a furry thing that purrs, this is the word "chat" like I would like to "chat" or "talk" with you. Can you say the word "chat" for me? What do you feel at the beginning?"

Now we get the student to announce the phoneme /tʃ/ and link it to the <ch> in a word that they are either just learning or maybe they recognize. We could look for other words with a <ch> for this phoneme. And now we would have reinforced the principle that most graphemes can represent more than one phoneme, and that many graphemes are comprised of more than one letter.

I could choose specific words to hit these ideas, but once the teacher is comfortable with this orthographic knowledge, it is never hard to find words for teaching these principles just by studying any text that kids are interested in.

There is much more than could be said, but I'll leave this comment at this for now. Curious to know what you think, and thanks for the great questions!


Peter Bowers
This comment was minimized by the moderator on the site

This video is a helpful, concrete demonstration. Thanks for making it!

I certainly get how phoneme - grapheme (p-g) correspondences are made more "explicit" using the conventions of angle brackets for graphemes and slash brackets for phonemes. And I get how the spell-aloud procedure (pausing at the word sum boundary, etc.) makes orthographic elements more explicit. We have integrated that spell-aloud procedure in our lessons' morphology step, and we find it to be an efficient way to teach, reinforce and observe morphological and orthographic understanding.

Here are a few questions the video raised in my mind:
1) Do you ever practice raising students' awareness of phonemes in the absence of graphemes? (Either way, I'd be interested in the science that underpins your rationale.)
2) Based on your demo, I assume that you don't follow any specific sequence in teaching p-g correspondences; rather, the teaching targets p-g correspondence(s) are determined by a student's errors or confusions. So, is it safe to say you advocate an "embedded" versus "sequential" approach to teaching p-g correspondences? What if, in discussing "chat", a student said that <ch> is pronounced /k/ (which, of course, it can be). Would you have launched into a discussion about that (e.g., in what words we see that correspondence, etc.) or would you have just acknowledged it and re-focused on the correspondence in the word at hand: "chat"? I ask because the tyranny of time means that how time is spent during lessons is incredibly important! (Again, I'd be interested in the science that underpins your rationale either way.)

Again, Pete, thanks for making this video. This sort of thing helps us discuss gnarly issues so more productively!

My best, Sandie

Sandie Barrie Blackley
There are no comments posted here yet