Structured Word Inquiry gets curiouser and curiouser

One question that baffles many of us about Structured Word Inquiry (SWI) is how it could possibly be used as a beginning reading program. In a previous post, I mentioned Peter Bowers’ video that was intended to demonstrate just how this is done but where the child could apparently already read, “Mom says she wants a…”. So it was with interest that my attention was drawn to an initial reading program developed by an SWI proponent.

I have to say that it starts oddly with an address that is ostensibly directed towards the person who is about to learn to read:

“Congratulations! You’re going to learn how to read! But, before you read a single word, you need to know some facts about the English spelling system… A grapheme is the smallest meaningful contrastive unit in a writing system. <h>, <sh>, and <ugh> are graphemes. Single letters can be graphemes. Graphemes can be made of two letters. A two-letter grapheme is called a digraph. The prefix di- means “two.” Graphemes can be made of three letters. A three-letter grapheme is called a trigraph. The prefix tri- means “three.”.”

And so on. I assume that we are not to take this literally and that it is not being suggested that we talk to illiterate five-year-olds about the ‘smallest meaningful contrastive unit’ or later that, ‘Letters can function as lexical, phonological, and/or etymological markers’. Instead, I will assume that this is a colourful way of informing the instructor about some of these concepts.

However, I do think this preamble is revealing. At one point, we are told that, “English spelling is rule-based. There are no exceptions, just more rules to uncover.” Then later on in the text, there is a discussion of homophones:

“Homophones provide further evidence that the primary function of English orthography is to represent meaning. Homophones are two or more words that sound the same but that have different spellings and meanings. The homophone principle states that, when two words sound the same but have different meanings, the spellings will be different whenever possible.” [my emphasis].

That ‘whenever possible’ is carrying a lot of weight here because there are indeed words known as ‘homonyms‘ that sound the same, are spelled the same way but that have different meanings. We may have been worried about these being exceptions to the rule that ‘when two words sound the same but have different meanings, the spellings will be different’ but no. In the case of homonyms, presumably this is not possible. All is well with the world. Order is restored.

I am reminded first of the epicycles that accounted for the motion of planets in a universe with the Earth at its centre and all heavenly objects in orbit around it. I am also reminded of Godel’s theorem. But more of that later.

It seems to be important to proponents of SWI that English spelling is rule-based with no exceptions because a key criticism they make of phonics seems to be its use of rules that have exceptions. Here are Jeffrey and Peter Bowers writing together on the subject:

“It is widely assumed that the primary role of letters in English is to represent sounds, and the many “exception” words are generally taken to reflect a poorly designed spelling system. However, this reflects a misunderstanding. In fact, the English spelling system is designed to encode both pronunciation and meaning of words, and as a consequence, English word spellings are constrained by phonology, morphology, and etymology. Rather than a perverse  system that needs reform, some linguists call English spelling “near optimal” (Chomsky & Halle, 1968). Whether or not English spellings are near optimal, the key claim we would make is that English spellings are logical and can be investigated like other scientific subjects given that the English spelling system is systematic.”

Repeat after me the Orwellian mantra: There are no exceptions, just more rules to uncover.

The SWI argument strikes me as peculiar because it appears to derive the logic of the English language from written English. It is plain as day that written English has had a profound effect on spoken English, but it is also clear that spoken English has primacy. If we all start saying something new then written English has to accommodate it. There is no process by which written English can lodge a formal objection to innovations in spoken English.

For example, take the word, ‘slippy’. It clearly exists because you can look it up in a dictionary. At some point, someone must have decided that ‘slippery’ was too much of a mouthful, they couldn’t be bothered with it and so they would say ‘slippy’ instead. But that’s just my own speculation because when I tried to look it up in the online etymology dictionary, I couldn’t find it. So I’ve no idea what rules govern the spelling of this word. They must be yet to be uncovered because, remember, there are no exceptions. None. Zilch. Nada.

Holding such an absolutist stance seems like a fundamental flaw. Just one bona fide exception and the SWI edifice crumbles. There is no room for pragmatism here or for a shrug of the shoulders and a wry smile at the eccentricities of English. The inductive logic of science is replaced by the deductive logic of mathematics. This is probably why SWI proponents spend so much time on Twitter trying to dig themselves out of holes that their all-encompassing generalisations have thrown them into.

And there is an interesting warning from mathematics – a warning for anyone who wishes to construct complete formal systems of logic, epicycles and all. In 1931, Godel published his mathematical demonstration that no formal system of mathematical axioms can prove all truths about the arithmetic of natural numbers. In other words, even in mathematics, whatever system of logic you decide upon, there will be exceptions.

Maybe the logic of English spelling will turn out to be more robust than the logic of basic mathematics. Time will tell.

In the meantime, repeat after me: There are no exceptions, just more rules to uncover.


75 thoughts on “Structured Word Inquiry gets curiouser and curiouser

  1. Chester Draws says:

    Oh boy, and there are a LOT of rules. Having learnt languages with basically no pronunciation exceptions (e.g. Swedish), English is a sub-optimal nightmare.

    How do they deal with alternate spellings but the same sound and meaning: hiccup and hiccough? Or different pronunciations for the same word and meaning: either and either?

  2. Jeff Bowers says:

    What I find curious is the strong resistance to the idea that reading instruction should teach children all the regularities in the writing system, both the phonological (GPCs) and meaning (morphology, and etymology) together to jointly teach reading, spelling, and vocabulary knowledge.

    Yes, Bowers and Bowers wrote: “English spellings are logical and can be investigated like other scientific subjects given that the English spelling system is systematic”. Scientific subjects like linguistics. Crazy!

    I am willing to concede that there are words in English with no good explanations. And others that have linguistically compelling explanations that are sufficiently complex that the explanation are irrelevant to teaching literacy (and I’ll be the first to confess that I am not strong on linguistics, and Peter Bowers and Gina Cooke who know about 100x more about the linguistics of the English writing system might have different opinions on that). But what I do know is that there are lots and lots of simple cases where word spellings are “exceptions” to phonics, and easily explainable in SWI. The fact that SWI organizes words into morphological families to teach vocabulary, spelling, and pronunciations seems so reasonable rather than curious. It sees a reasonable idea even in a shallow orthography that does not have lots of GPC “exceptions”.

    The fact that Greg thinks he makes a good point with ‘slippy’ boggles the mind.

    • “What I find curious is the strong resistance to the idea that reading instruction should teach children all the regularities in the writing system, both the phonological (GPCs) and meaning (morphology, and etymology) together to jointly teach reading, spelling, and vocabulary knowledge.”

      There is no resistance to this. All advocates of systematic phonics accept this. What they do not accept is that all of this should be taught from the very start of reading instruction or that we should impose an implausible formalism on English that allows no exceptions. You might want to start critiquing what phonics advocates actually believe.

      • jeff bowers says:

        The key tools of SWI are morphological matrices, word sums, and etymological dictionaries to investigate how words are related to one another. How to make better sense of GPCs. How to build vocabulary, and how to understand spellings. How phonological and semantic regularities interact. If you like these tools and these goals in year 4 but not year 1, then big win for SWI. And odd to dismiss this earlier when there is no evidence for this conclusion, and indeed, all current (limited) evidence suggests that morphological instruction is effective for younger children, that SWI works in year 1-2, and perhaps most importantly, that learning and memory is always best for material that is studied in a meaningful and organized manner.

        And in any case, the whole point of your blogpost is misguided — nothing curious about the hypothesis that children should be taught the linguistic organization of spellings.

      • SWI does not have a monopoly on teaching morphology and etymology. If you can bring yourself to accept that SWI may not be optimal for initial reading instruction then we might find some common ground. However, I suspect that more pragmatic approaches to etymology and morphology will still prove to be superior to SWI’s formalism.

  3. Harriett Janetos says:

    Jeff, with reference to teaching SWI to beginning readers, you say it’s “odd to dismiss this earlier when there is no evidence for this conclusion”. My evidence has come over the past twelve years from teaching hundreds of students to read, students who arrive lacking literacy skills and struggle to make sense of those squiggly lines in front of them–to attach print to sound. Once they’ve nailed this, showing “how phonological and semantic regularities interact” makes perfect sense. But not before.

  4. jeff bowers says:

    Greg’s writes: “If you can bring yourself to accept that SWI may not be optimal for initial reading instruction then we might find some common ground”. Do the final sentences of the Bowers and Bowers (2017) paper address this point?

    “These findings motivate the use of tools such as morphological matrices and word sums that highlight the meaningful relations between words and etymological dictionaries that allow hypotheses to be tested. Clearly more research is needed to assess the efficacy of these specific tools, and the approach more generally. Our main goal here is to help inspire more research into teaching children the logic of their spelling system (also see Crystal, 2013; Henderson, 1984; Venezky, 1999).”

    We make a similar point in every paper. It is researchers claiming the science of reading strongly systematic phonics who are curiously going beyond the data and ignoring basic insights from linguistics and cognitive science. That is, that spelling system is morphophonemic, and that learning and memory is best when information is learned in a meaningful and well organized format.

    Final point. The suggestions that proponents of systematic phonics are strongly committed to teaching morphology and etymology at a later stage of instruction is false. Teachers know almost nothing about morphology and etymology, and there is little consideration about how to teach it in the research literature. But new suggestions about how to teach morphology without matrices and word sums are welcome — indeed, as noted in the quote above, we are trying to motivate more research into how to teach all the regularities in English spellings.

    • You continue your odd approach of accepting the need for more research while also making bold assertions that lack evidence. An example of the latter is “It is researchers claiming the science of reading strongly systematic phonics who are curiously going beyond the data and ignoring basic insights from linguistics and cognitive science. That is, that spelling system is morphophonemic, and that learning and memory is best when information is learned in a meaningful and well organized format.” You are making a priori assumptions here about both phonics, that it is somehow not ‘meaningful’ and some supposed ‘basic insights’ that are being ‘ignored’. It is particularly bizarre to call for a ‘well organised format’ when your whole post hoc case against systematic phonics is that you think the ‘systematic’ bit makes no difference.

      In addition, I have repeatedly discussed my experience with Letters and Sounds and the far more comprehensive Sounds-Write systematic phonics reading programmes. Phase 6 of Letters and Sounds (for Years 2 and upwards, from memory) introduces morphology and etymology. You should be aware of this because it is a quite well known, government developed approach. I have also discussed the fact that I have seen a Sounds-Write trainer demonstrating something similar to SWI’s matrices. John Walker of Sounds-Write has even suggested to me, in person, that spelling assessments are often better diagnostically than reading assessments because they assess morphological knowledge.

      The case for SWI seems to hinge on SWI owning morphology and denying that anyone else can address it. There is also this peculiar morphological formalism that asserts the primacy of written English and the suggestion that we should overload beginning readers by asking them to attend to phonology, graphemes, morphology, etymology etc all at once. Not only is this at odds with theories of educational psychology such as cognitive load theory, it is wildly implausible to the majority of early educators and nobody from the SWI field has yet provided a convincing demonstration of how it can be done.

    • I’ve just looked up the ‘phonics’ curriculum standards for the State of Victoria where I live. It begins with a short preamble about the evidence supporting phonics. Level 5 focuses on morphology. Here’s a quote:

      “Morphemes are important for phonics (reading and spelling), as well as vocabulary and comprehension. Teaching morphemes is useful, as they are often spelt the same across different words (even when the sound changes), and often have a consistent purpose and/or meaning.”

      The fact that teachers lack knowledge of morphology and etymology is hardly surprising because we know they lack a lot of knowledge of phonics, presumably because we have to keep endlessly debating its value, muddying the waters sufficiently for schools of education to avoid teaching this knowledge. Here’s a good paper on the issue:

      • Jeffrey S Bowers says:

        And here is a paper that assess what teachers know about morphology.

        This is not surprising give how little consideration morphology is given by researchers advocating systematic phonics. Here is a quote from: Bowers, J.S., and Bowers, P.N. (2018). The importance of correctly characterizing the English spelling system when devising and evaluating methods of reading instruction: Comment on Taylor, Davis, and Rastle (2017). Quarterly Journal of Experimental Psychology, 71, 1497-1500.

        “The more general point we want to emphasize, however, is that most researchers
        claim that the function of letters is to represent sounds (the “alphabetic principle”) and little
        consideration is given to the fact that English is fact a morphophonemic system in which
        morphemes are spelled more consistently than phonemes. This failure to consider the
        morphological organization of English spellings has had a profound impact on reading
        research over the past decades. To illustrate, consider the National Reading Panel (2000) that
        was setup to assess how to best teach reading. In 449 pages, the word “morpheme” only
        occurs once (in a table), whereas “phoneme” occurs 294 times (derivations of “morpheme”
        were mentioned a total of 4 times). In more recent meta-analyses that are taken to support
        phonics (Galuschka et al., 2014; McArthur et al., 2012; Rose, 2006, 2009), and a recent meta
        analysis that fails to find any long-term benefits of phonics (Suggate, 2016), there are no
        occurrences of the word “morpheme”.

      • I am all for more research, but the original point still stands. SWI does not own morphology – it is a key component of systematic phonics programmes, just not from the start of reading instruction. If anything, setting up morphology in opposition to systematic phonics is likely to hinder the cause of research into morphology given the wide acceptance of the evidence supporting systematic phonics among researchers. It makes it look like a fringe cause.

      • Harriett Janetos says:

        This is very helpful, Jeff. Tomoroww when my first graders struggle to read a work like “spoil” (as they did today), to attach phonemes to the graphemes in front of them, I’ll instruct them to ignore the decoding strategies we’ve practiced because “English is in fact a morphophonemic system in which morphemes are spelled more consistently than phonemes.”

  5. Pingback: Weekly roundup of favorite posts | Catherine & Katharine

  6. Harriett Janetos says:

    Here is Activity 3 from Lesson 3 of the program you reference. I look forward to seeing other examples of beginning reading programs by SWI proponents.

    Activity 3: Can Spell [Ø]

    Remember that the grapheme can spell the phone [m]. Also remember that not all letters spell sounds. Letters can function as lexical, phonological, and/or etymological markers. The grapheme can also spell the zero phone [Ø].

    The word can spell [nəmɑnɪk]. The first spells the zero phone [Ø].

    Why does spell the zero phone [Ø] in ? A mnemonic is something that helps in remembering something else. comes from the Greek meaning “memory, remembrance, record.” The base is . Other words in the family include meaning “memory loss, lack of memory” and meaning “forgetting of past offenses.”

    Notice that spells the zero phone [Ø] at the beginning of a word like but [m] in the middle of a word like or .

  7. Jenny Chew says:

    In his opening sentence, Greg said that what was ‘baffling’ was how SWI ‘could possibly be used as a beginning reading program’. The ‘beginning’ point also bothers Harriet and me. SWI sometimes seems to assume that children can already read a bit, as in the ‘Mom says she wants a..’ example. At other times, it assumes that it’s not important for all the children to be able to read the words whose morphology is being discussed. We see this in a video which Pete Bowers recently drew to my attention: He commented that the pre-school teacher is ‘not even worrying about trying to get everyone to “read” these words yet, but she’s introducing them to the letter names as they spell them out’ – OK, introducing letter-names, but not expecting the children to apply that knowledge in reading even the words used in this lesson, let alone any other words. If this is a good way of teaching beginning reading, we should surely see some evidence that the children are learning to read the words.

    Jeff concedes that there are words ‘that have linguistically compelling explanations that are sufficiently complex that the explanations are irrelevant to teaching literacy’. I agree, but I also think that there are linguistic explanations which are too complex for real beginners, even if not for children a bit further up the learning curve.

    • Harriett Janetos says:

      This video does indeed do a good job showing how rich and rewarding a morphology lesson can be. But as Jenny points out, it doesn’t teach kids how to read–it supplements rather than supplants reading instruction. It’s very telling that at the end the child doesn’t even attempt to decode “raincoat” after having had an entire lesson about the rain family.

      • Jenny Chew says:

        Thanks Harriet. I’ve also now followed up the things that looked most relevant in the link recommended by Kevin, but they didn’t answer my questions. n all the SWI material I’ve looked at, usually at the recommendation of SWI proponents, I’ve seen (a) children who can already read the words being worked with but nothing to show how they got to that point, and (b) children who can’t read the words but not being taught to do so. As I’ve said, the comment made by Pete Bowers on that video was that the teacher wasn’t even worrying about trying to get the children to read the words yet. I need to see something in print or on video showing how it’s done by someone who IS trying to do it if I’m to understand how SWI teaches real beginners.

  8. Jenny Chew says:

    Something I found via Kevin’s link was a lesson described by Lyn Anderson on 23 April 2014 under the heading ‘Teaching Orthography in the Early Years: word webs for young children’.

    Anderson writes ‘We started this particular learning journey by rereading a well loved story, Mrs Wishy Washy by Joy Cowley.The children were delighted to read the story to me, which they did with great enthusiasm and understanding. They clearly viewed themselves as proficient readers, by explaining and discussing the underlying ideas and meaning of the story.’

    So the children could already read the story, and had apparently learned to do this without any discussion of morphology – if morphology had already been discussed, the lesson that Lyn Anderson goes on to describe would have been unnecessary.

    I’m therefore now wondering whether SWI envisages children first reading words logographically, without attention to morphology or GPCs, but does not regard this as the ‘start’. The ‘start’ comes a bit later, when children first have their attention drawn to sub-units in the words they can already read. At this point, the teacher chooses words consisting of more than one morpheme for investigation.

    Might this mean that differences between SWI people and phonics people could be resolved if we could agree on what we mean by the ‘start’? Are we at present a bit like the two participants in this Monty Python sketch, who understand the word ‘lion’ very differently? –

    • Harriett Janetos says:

      Thanks for the levity, Jenny, which also highlights the gravity of the situation–of getting beginning reading right.

    • Despite originally calling for discussion, on Twitter, Bowers has now said he no longer wants one with me. This is apparently because I called into question his integrity because I wondered whether he knew running SWI against SSP rather than motivated reading, wouldn’t have been good for SWI. This feels a bit rich, given the overt personal attacks, accusations of sexism etc that have been directed at me by SWI supporters since taking up Bowers’ invitation to discussion.

    • Harriett Janetos says:

      I eagerly await responses from SWI proponents. As a practitioner in a high-poverty school, I depend on researchers to inform my reading instruction. When I taught kindergarten (and continuing now as a reading specialist working with struggling readers), my beginning reading program has relied on the following:

      1) Research on infants by Stanford psychologist Anne Fernald, which reveals how child-directed speech impacts vocabulary development and processing speed, both of which affect future reading.

      2) Neuroscientist Stanislaus Dehaene (Reading in the Brain), who emphasizes that learning to read requires attending to phonemes and the graphemes that represent them

      3) Linnea Ehri’s work on letter-embedded picture mnemonics as discussed in her article, Orthographic Mapping in the Acquisition of Sight Word Reading, Spelling, Memory, and Vocabulary Learning

      4) Using Elkonin Boxes in conjunction with Isabel Beck’s word-building activities in Making Sense of Phonics.

      5) Decoding rather than memorizing “sight words” as described in Rethinking Sight Words

      6) Providing ample opportunities for students to apply their phonemic awareness skills to writing as described in Invented Spelling in Kindergarten as a Predictor of Reading and Spelling in Grade 1

      7) And, of course, giving students ample opportunities to apply their skills to reading connected text in a variety of books for emergent readers.

    • Jeff Bowers says:

      Hi Jenny, the thing is that I have replied repeatedly to Greg (e.g., see above) only to go around in circles with unjustified claims of post-hoc, etc., with most comments snide and confrontational (e.g., see the title of this blogpost). Greg finds it “a bit rich” that I don’t enjoy personal challenges from him because *other* people are rude to him? This is the level of argument that is not worth wasting my time.

      But do notice the irony. I’ve have responded in hundreds of tweets and multiple posts on various people’s blogs explaining SWI and responding to misrepresentations. But how about the responses to my challenge to systematic phonics? What do you make of it that Buckingham, Ashman, and no one else has commented on my blogpost to Buckingham? Hav e a read yourself and explain why my claims about systematic phonics are unjustified.

      • The other people who have been rude to me are supporters of yours who you interact with and validate on Twitter. Do you need some quotes from them? Perhaps we could go with them calling Jen B a liar or suggesting I must be sexist. There are others. In this context, it really does seem strange to appear so wounded by my tone.

        My critique of your post-hoc analysis being post-hoc is valid because it literally is post-hoc. You can reasonably claim that this does not matter, perhaps, but you cannot deny a truth. I have even given you a quote from another academic paper explaining why this is an issue with the Camilli paper, repeatedly, but you don’t respond.

        And this is part of a pattern. Yes, we do go around in circles. But I think that’s because you want the discussion on your terms only. You want your questions and points addressed while ignoring many of those made by others that do not fit your narrative. Ultimately, it is the quiet observers of the debate who will make up their own minds. I’m comfortable with that. Are you?

      • Jeffrey S Bowers says:

        Greg, I kept responding to all your points, including in response to your “quote from another academic paper” (that I in fact I also discuss in my Educational Psychology Review paper). See:

        The claim that Camilli et al. are post-hoc by asking a different question to a similar (but not identical) dataset makes no sense, and it is not worth arguing about anymore.

        I even responded to the “lies” comment:

        But I don’t speak for other people, and they don’t speak for me. I often agree with the analyses of SWI by your critics, but I’m not responsible for their words. I’m responsible for my words, you responsible for your own.

        Our exchange has largely been composed of me answering your (often aggressively worded) questions, you not acknowledging the point that I make, moving on to something else to criticize (other than the repetitions of the post-hoc argument that you take as your most fundamental point). In the meantime, you rarely respond to the occasional points I raise. For example, will you be responding to my blogpost on systematic phonics?

        This is in fact a silly exchange to be having in public. I will respond to any substantive points you raise on my blogpost.

      • The fact that your analysis, and one of the ones you rely on, are post-hoc *is* the fundamental point.

        Tone is very difficult to perceive in writing. Because you do not like what I say, you perceive it to be aggressive. I perceive myself to be quite calm and composed. This is good on tone (and argument in general):

      • Jenny Chew says:

        Thanks, Jeff. There are points to do with the technicalities of research which are beyond me, so I leave discussion of those to people who are better qualified than I am.

        BUT I spent 28 years teaching English (and also sometimes Latin) to teenagers, and developed a particular concern about students’ spelling problems in the last 22 of those years. I saw that many errors could have been avoided if the students had understood more about morphology, so I don’t need convincing about that. I then spent 17 years working voluntarily on reading with much younger children, including real beginners. I know that children can start to grasp simple morphological points for reading and spelling very early, but I’ve never seen beginners being taught to read and spell their very first words at the same time as being taught about morphology and I don’t see this on SWI videos. What I’ve stressed in all my comments here is my puzzlement about SWI as an approach for those real beginners, which was also Greg’s starting-point in his blog. This makes me wonder whether real beginners as I know them are the same as real beginners as SWI knows them. Can you suggest anything which would provide answers?

      • Jeff. Could you perhaps comment on David Kilpatrick’s work? He too has critiques of the literature on systematic phonics and in fact finds little support for any significant gains with Orton Gillingham only approaches. However, his vast combing of the literature does find support for systematic phonics mixed with aggressive phonological training — which from my layman’s perspective (whether they are aware of it or not) – seems to be the natural way most anyone actually learns how to read at the outset and as limited by their oral vocabulary – not through recognizing morphological or etymological contexts as useful and interesting as that may be.

  9. Jenny Chew says:

    I agree that the research mentioned by Harriet is worth taking seriously – also research by people such as Stanovich, Share, Perfetti and (here in the UK) Rhona Johnston and colleagues. It seems, from what Greg says, that we may not hear again from Jeff, but there are other SWI advocates, so let’s hope we hear from them. We DO need clarity about what we regard as the ‘start’, however.

  10. Harriett Janetos says:

    On February 3rd Jeff tweeted:

    “For those interested in reading, here is Greg Ashman dismissing SWI as curious and peculiar, with cult-like followers mindlessly repeating an Orwellian mantra. And some responses from me.”

    Since this tweet came before Jenny started posting, I guess that means I’m the cult. Not sure how Greg feels about having a cult-like following with just one member. In any case, rather than resort to vacuous epithets, it would be helpful to have a response, as Jenny says, about what we regard as the ‘start’. The stakes are high for those of us who teach underprivileged children.

  11. Harriett Janetos says:

    My mistake, and I apologize. I now realize that Jeff’s references to a cult-like following and Orwellian mantra were in fact directed at Greg’s comments about SWI. I guess this goes to the heart of all the confusion and the lack of clarity, of trying to pin down what exactly SWI proponents are advocating for beginning reading instruction–for the ‘start’.

  12. Reposting from Twitter:

    Why do you assume that prereaders cannot learn to read via words like “morpheme” and “etymological marker”?

    Kids are smart.

    Why do you think morpheme meaning “word part” and etymological marker meaning “letter(s) that show the history of a word” are out of reach for a child who cannot yet read?

    Just today I asked my son to read a title of a book.

    “Are there any morphemes you can take off?” I prompted.

    “The ,” he said.

    He then easily read .

    He is 5 but would not yet be in kindergarten if he attended public school. He also knows what a “zero phone” is and what a “replaceable e” is.

    How can you say that beginning readers cannot learn words like “morpheme.”

    You also conveniently failed to mention the final paragraph of the introduction, which is to be read aloud to the child learning to read:

    “You do not have to remember or understand all this information now. This reading program will clarify and reinforce these facts about the English spelling system as you learn to read.”

    Learning is cumulative, not linear.

    • The interesting issue with research into reading instruction is that it is not a case of some methods working and others not working at all. Whatever method you use, some children will learn to read. The issue is which method will lead to the greatest proportion of children becoming successful readers and hence the least reading failure. With all of the information SWI asks children to consider from the outset of reading, I would predict, based upon my knowledge of cognitive science, that it would be less effective than a systematic phonics program that only initially asks students to attend to letter-sound relationships. Of course, I could be wrong. A good randomised controlled trial that compared SWI with systematic phonics would provide evidence one way or another.

  13. Jenny Chew says:

    Good points, Greg.

    Whoever was doing the RCT would have to decide whether to use a version of SWI which teaches about ‘zero phones’ and ‘replaceable e’. from the start or the type which starts with words such as ‘cats’, ‘rains’ and ‘raining’ as we see on videos recommended by Jeff and Pete Bowers. The latter approach seems more realistic than the former, though still not as realistic for beginners as a typical synthetic phonics approach.

  14. Harriett Janetos says:

    The confusion around the beginning reading piece in SWI is baffling. Although I primarily work with struggling first and second graders, I teach a third grade class once a week, and I have now added morphological matrices to my instruction by using a very convenient online matrix-maker. SWI for kids who can decode is beneficial–no doubt about it. But this means my third graders can effortlessly read “struct” before we engage in word sums, which is not the case with my beginning readers. Interestingly, the Nueva school, a private school highlighted on Pete’s website that embraces SWI, requires an IQ test for admission. I wonder if that tells us something about using SWI for beginning reading instruction.

  15. Just want to thank everyone for this discussion. I recently came across SWI and had me starting to question my own approach, though indeed couldn’t get around the randomness of their websites, information, and approach – not to mention the biting and overzealous nature of that Gina woman over at LeX. This article has articulated my initial thoughts succinctly, and to see that the SWI people have not come up with a valid answer to your critiques is very useful. The fact that Bowers has backed out of a conversation with you is also a clincher for me in not taking them too seriously at this juncture.

    My best sense, is that yes their approach is useful for *some* readers, and seemingly more so as an add on to a solid program so that the many kids who struggle do not fall through the cracks. But as your core reading/spelling instruction – I can’t even imagine what that would look like and it appears neither can they. This includes Pete Bowers video from Dec. 2019 wherein he is essentially teaching phonics in a random and incoherent fashion (also ignoring how kids are getting their initial phoneme-grapheme relationships in the first place) as well as using exercises I’ve seen elsewhere in proper and systematic programs and passing it off as fundamental to “SWI”…..seriously? I was almost going to attend a workshop with him in a couple weeks, maybe I’ll just stay at home and keep reading the linguistics book I’m working on…

    • Harriett Janetos says:

      Frank’s point about Kilpatrick’s work related to phonological awareness (Essentials of Assessing, Preventing and Overcoming Reading Difficulties), is a very important one for those of us working with non-readers. This is why I really like Isabel Beck’s word-building program in her book, Making Sense of Phonics. These activities are used in conjunction with Elkonin boxes, and it is fascinating watching kids turn ‘ran’ to ‘rain’ and move their letters from the three boxes on their sound boards to the four boxes below–and all the PA discussion/training that goes on to teach them about how the ‘ai’ digraph represents one sound and that both words actually have three sounds though the number of letters is different. Then, much further down the line of instruction, looking at the fact that a seven-letter word like ‘thought’ also just has three sounds.

      • Harriett. I like Beck’s text, though it is nothing novel and I do not think the research is clear in terms of training PA skills with or without physical referents, and she takes her side quite clearly. I’ll have to revisit Kilpatrick and am not 100%, but I think he is actually in favor of also training PA in a purely auditory fashion based on his research and as evidenced in his manual. In fact one of the methods that Kilpatrick’s mentions in the research studies as being highly effective is Phonographix, which essentially involves the same core exercises seen in Beck’s text; though having worked with both approaches, I feel one needs to tweak and bring other elements in for a more complete approach – particularly for those who struggle or older readers.

      • Harriett Janetos says:

        Thanks for the clarification, Frank. In addition to getting my reading specialist credential, I’ve been trained in Phonographix, and I definitely do PA training with and without referents, though my understanding is that the research favors using letters in conjunction with PA activities. I first learned about Phonographix through Diane McGuinness, and I highly recommend her three books, Why Our Children Can’t Read and What We Can Do About It: A scientific revolution in reading (1997); Early Reading Instruction: What science really tells us about how to teach reading (2004): and Language Development and Learning to Read: The scientific study of how language development affects reading skill (2005).

        Steven Pinker wrote the foreword to her first book and says: “Children are wired for sound, but print is an optional accessory that must be painstakingly bolted on. This basic fact about human nature should be the starting point for any discussion of how to teach our children to read and write. We need to understand how the contraption called writing works, how the mind of the child works, how to get the two to mesh. It is a national tragedy that this commonsense understanding has been so uncommon.”

        This is why this discussion is so important.

  16. says:

    Hello Greg. I tried to post a response here, but I’m guessing my text was too long. My main reason for posting had to do with your stated interest in Cognitive Load Theory. You were arguing that SWI runs counter to it, but actually, I think it turns out to be a brilliant support of the practices and principles of SWI. This is not the kind of thing that can be explained quickly, thus the long message. If you and your readers click this link ( you can see that response. That same document points to a much more involved piece I specifically wrote on the links between CLT and SWI, and to a paper on CLT that I think is crucial from 2007 that I thin addresses exactly the relevant issues.

    I typically avoid discussions on blogs because I find the forum often generates more confusion than clarity. I hope by using a pdf document that can use graphics and link to further resources, adding my voice on this specific point is helpful. There is so much more I could say, but let’s see what you and your readers think about this point.



    • Harriett Janetos says:

      Thank you so much, Pete, for posting your long reply and the additional videos. You’re busy, and that took a lot of time. Having read your post and looked at the videos I had not previously seen, I am more convinced than ever that we are talking at cross purposes. Since I first came across SWI two years ago, I have discovered an online matrix-maker, and I now use word sums with the third grade class I teach once a week. Adding this morphological piece has been greatly beneficial. Our discussion of ‘ject’ was a particularly good one because my students enjoyed sharing their knowledge of ejector seats.

      However, nothing that you have said–and nothing that I have seen in the videos you’ve shared–would lead me to abandon what I do with my beginning readers. For example, that’s a wonderful video of Skyler reading, but he can already read CVC words like ‘get’ and ‘web’, which is not the case for my beginning readers. I need to establish phonemic awareness in my non-readers and teach GPC’s so that my students can efficiently blend and segment. Then and only then can I get to those other words that stumped Skyler like ‘caught’ and ‘rare’. I have previously referred to the ‘rain’ video as one that shows an activity that is rich and rewarding and does a great job teaching the rain family, but it doesn’t teach kids how to read the word ‘rain’. It supplements rather than supplants beginning reading instruction.

      Previously I posted what I do to get kids who lack GPC knowledge to read, and that’s what I still haven’t seen from SWI. If you’ve got videos showing what you do to teach students to decode words like ‘get’ and ‘web’, I’d love to see them. Here’s what I posted about what my beginning reading program relies on:

      1) Research on infants by Stanford psychologist Anne Fernald, which reveals how child-directed speech impacts vocabulary development and processing speed, both of which affect future reading.

      2) Neuroscientist Stanislaus Dehaene (Reading in the Brain), who emphasizes that learning to read requires attending to phonemes and the graphemes that represent them

      3) Linnea Ehri’s work on letter-embedded picture mnemonics as discussed in her article, Orthographic Mapping in the Acquisition of Sight Word Reading, Spelling, Memory, and Vocabulary Learning

      4) Using Elkonin Boxes in conjunction with Isabel Beck’s word-building activities in Making Sense of Phonics.

      5) Decoding rather than memorizing “sight words” as described in Rethinking Sight Words

      6) Providing ample opportunities for students to apply their phonemic awareness skills to writing as described in Invented Spelling in Kindergarten as a Predictor of Reading and Spelling in Grade 1

      7) And, of course, giving students ample opportunities to apply their skills to reading connected text in a variety of books for emergent readers.

    • Hi Pete. I have read your response. Setting aside your slightly odd take on cognitive load theory (you should be aware that Sweller has moved away from the germane load construct) a critical issue for SWI presents itself once again. You write, for instance, about a students who can presumably fluently read ‘action’ but spells it incorrectly. This is not initial reading instruction. At this point, once a child has appropriate schemas for letter-sound correspondences, I agree that morphology would probably not be overwhelming for working memory and would probably aid spelling. However, what we really want to know is how you get students to the point of reading words like ‘action’ fluently if SWI involves, as has been claimed, attending to morphology and etymology from the very start of reading instruction. It is this that would violate the roughly 4 item limit of working memory and therefore it is this specific claim that is at odds with cognitive load theory. Videos of children who are already fluent readers manipulating word matrices or investigating words do not answer this point.

      • says:

        Greg, I would like to know what you mean when you say I have a “slightly odd take on cognitive load”. Do you think my representation of Schnotz and Kurschner’s understanding of cognitive load theory is “odd” or do you think the understanding in their paper is odd?

        The example I gave for “action” — in particular the lesson was presented as a lesson I have done with Grade 1 students. I even included an image of Grade 1 students involved a very similar lesson. I did not present that lesson with all that text as something I would do with non-readers. It was a general illustration of how morphology can be used to guide explicit instruction that could help children understand a grapheme-phoneme correspondence that not something that can be understood without reference to morphology. I also used that lesson as an example of how – in line with cognitive load theory as represented in the article I liked. For instance, Schnotz and Kurschner argue that CLT encourages instructional practice to use graphics as “worked out examples” of complex ideas. They highlight the importance of motivation to get kids to focus enough to do the work to rearrange mental representation of schema. I argue that the meaningful connection between “act” and “action” provides a meaningful and motivating context compared to the option of asking a child to remember to spell “action” with a “t” rather than an “sh” for phoneme /ʃ/ in “action”. This was about offering a meaningful context to teach a grapheme-phoneme correspondence that is extremely common in words kids encounter in early reading. It turns out that in my workshops with teachers, most that few were aware that a common job of the “t” grapheme is to write the /ʃ/ phoneme. I find it odd that people very interested in explicit grapheme-phoneme instruction are not excited to find a way to explain why to use one grapheme over another for a phoneme when more than one can do the job.

        I also pointed out a way that I see cognitive load theory points to a problem with language that we commonly hear in phonics-based classrooms in terms of teaching grapheme-phoneme correspondences. I argue that the phrase “sh sound” and “t sound” builds extraneous cognitive load for a learner. Such phrases become automated in our mind (As CLT would call — well-integrated mental representation of schema) that is likely to come out in the act of spelling (again — I am NOT talking about very beginning readers here) and result in a misspelling like *acshun” for “action” or *jumpt” for “jumped”. These are spelling errors very familiar to teachers of young children. Instruction has created extraneous cognitive load for children if they have to inhibit what they have been taught to get to the attested spelling.

        If children are taught — from the beginning — to think about the “sh sound” and the “t sound” those spelling errors result from students applying what they are taught. So I avoid such phrases from the beginning. I also offer learners a way to understand the grapheme-phoneme correspondences in “action” and “jumped” by addressing the interrelation of morphology and phonology. The lesson I shared was not for children at the very beginning, and I did not present it that way. But I do use words like “action” and “act” teach the grapheme-phoneme correspondences of the “t” grapheme including the fact that while it’s default job is to spell the phoneme /t/, but that it can also spell /ʃ/. And I do so in the context morphological cues of words kids know and understand.

        Is this an “odd take” of cognitive load as Schnotz and Kurschner present it in the article I linked, or do you think their presentation is an odd take of the theory?

        You and your readers can see the Schnotz and Kurschner article at this link:

        Click to access Schnotz_Kurschner_Cognitive_Load_2007.pdf

        I also pointed to my own term paper in grad school that I linked above. In that paper I attempted to lay out my understanding of Cognitive Load Theory and how I see the practices of SWI (at all ages – not just beginning) is perfectly in line with the instructional practice they recommend. Anyone curious about that can read that paper at this link:

        Click to access Cognitive_Load_Theory_&_SWI.pdf

      • The odd take is focusing on a single cognitive load theory review article from 2007 and then seeming to ignore the key cognitive load theory principle that there is a limit to what we can process in working memory at any one time, germane or otherwise. Have you read any other pieces such the recent overview I link to below or is your understanding based on this one review article?

        You have noted that the video does not represent SWI from the very beginning of reading instruction but I was pointed to this video by your brother when I asked for an example of the very beginning of reading instruction. Again, the grade students are in is irrelevant. What matters is whether they can already read or not. Do you have any videos, lesson plans or explanations showing what SWI looks like from the very beginning?

      • says:

        Greg, not sure how to make this comment link to your response to my questions about cognitive load theory — but the first thing I need to point out is that you’ve confused what I said about the lesson I linked on the “act” family with the video on the “rain” video Jeff pointed to. I do indeed see that “rain” video lesson as a totally appropriate lesson for a class with kids a the very beginning of learning to read (can’t read yet — are still learning letter names) and other kids who have started. The linked “act” lesson with word sums, a matrix and a grapheme-phoneme chart is one I explained as one I have used in grade one class but would not use in that form for very beginning readers.

        Is it your view that the video is inappropriate for learners who cannot read yet? As I pointed out in another post, this class does indeed have a combination of non-readers and readers. The non-readers are getting introduced to letter names, grapheme-phoneme correspondences, and they are being introduced to those in the context of a bank of words they know in an engaging group discussion. I do not see why it would be better to remove the instruction of letter names and grapheme-phoneme correspondences from this kind of context. Perhaps you think it violates cognitive load. I think the exact opposite.

        Re the paper I linked on Cognitive Load Theory. If you read my paper, you will see that I do base it on more than the 2007 paper by Schnotz and Kuschner. I refer to Sweller and a whole bunch of other people. But that 2007 paper is the one I found most valuable as it refined the understanding that came before by adding the role of zone or proximal development, and moved away from just recommending instruction reducing cognitive load that was common before into the idea of reducing extraneous load and increasing generative load.

        I wrote that paper as a grad student in 2007 when the Schnotz and Kuschner review came out. It was up to date at the time. I do not see what I have written anywhere that would suggest that I “ignore the key cognitive load theory principle that there is a limit to what we can process in working memory at any one time, germane or otherwise.” Of course you can overload cognitive load. This is inherent in the concept of zone of proximal development that I referred to repeatedly.

        I have not read a great deal on cognitive load, but when I can I am interested in reading the piece that you sent. The reason I entered this conversation was because you described SWI as being at odds with cognitive load theory. I disagree, and I sent you a paper that explained my position. I don’t know if you have read it. So far the only criticism I see that you have for how I represent it is that you think I ignore the possibility of overloading cognitive processing and that is not the case. You are concerned about Without going into too much detail, the point about interrelating grapheme-phoneme correspondences and morphology and meaning is to avoid overloading with too many discrete facts at one time. There is reason to believe the interrelatedness is exactly what CLT recommends. This also ties to Perfetti’s Lexical Quality hypothesis.

  17. Jenny Chew says:

    Hi Pete,

    Thanks for providing a lengthy reply. In it, you say that you choose the ‘act’ family ‘specifically as context to teach about “t” grapheme. While it’s default job may be to represent the /t/ phoneme, this morphological family helps us understand that the “t” grapheme can ALSO represent the /ʃ/ phoneme (often referred to as the ‘sh sound’), and the specific circumstances in which that can occur.’

    My questions: When you teach this lesson, would the children already know all the graphemes involved, including the default job of the ‘t’ grapheme, and would they already be able to read ‘act’? If so, they are not beginners. If they ARE real beginners (i.e. don’t already know these things), do you check, after this lesson, whether they can read and spell all the words in the matrix?

  18. Jenny Chew says:

    Like Harriett, I think that there’s a cross-purposes problem – hence my link in an earlier post to the Monty Python sketch, which Pete may not have seen:

    I’d like to know, however, whether Pete would agree that our differences might be resolved if we could agree on what we mean by the ‘start’.

      • Jenny Chew says:

        Yes, very interesting. It DOES seem to be realistic about what real beginners can manage. It’s not quite synthetic phonics as we know it in England, however. We teach both vowels and consonants from the start so that even when children know very few letter-sound correspondences they can read simple words by saying a sound for each letter and then blending the sounds together. Blending = synthesising, hence ‘synthetic’ phonics.

      • says:

        This is a an old document from the excellent Lyn Anderson. She is a brilliant primary teacher who had worked with SWI type orthographic instruction with young children since the early 2000s. Because people often ask me about SWI in the early years, I had asked if she would share the kinds of things she thinks are important to do from her experience – and she produced this document to share. If your sense is that its ‘just phonics’ that should make people feel less like SWI is harming kids. I think the reason you say that is that — as I have always argued — the teaching of grapheme-phoneme correspondences is essential in SWI from the very beginning of literacy instruction. And that is evidenced in what Lyn describes. But the section on written morphology is not something I see teachers being guided to do in phonics programs I have seen. She also talks about etymological things she does from the beginning.

        My sense is that many people mistake what my brother Jeff and I say very frequently. We address what we see as a problem of a “phonology first” approach — what we are saying is that that statement claims that we should avoid morphology and etymology at the beginning and only teach phonology to start. But our recommendation is NOT to replace “phonology first” with “morphology first”. What we have always claimed is that we recommend ensuring that literacy instruction reflects how the writing system works from the beginning. That means teaching about the interrelation of morphology, etymology and phonology from the beginning. That is represented in Lyn’s document. I do not see that reflected in phonics programs I see.

        I’m pleased that this document has made it clear to you that what people like about phonics is represented in SWI. This is also something I try to argue all the time — that when you teach about orthographic phonology as we do in SWI, you must not only explicitly teach about what the possible grapheme-phoneme correspondences are (as phonics can do), but also how those correspondences work in the orthography system. When we do that, we can understand spellings like “does” and “rough” and “to, too, two”, the “e” and the and of “house” and countless other spellings that are relegated to “irregular” status that have to be memorized.

      • Harriett Janetos says:

        Thanks for the link–very informative. Lyn writes:

        “Kinder/grade 1/grade 2
        Introduce some important graphemes (with younger children I refer to the
        graphemes as ‘teams of letters’) from an early age: ”

        Here’s where I think the “systematic” part of phonics comes in. I introduce ‘ay’ and ‘ee’ in kindergarten (with common words like play and see) but I don’t introduce ‘igh’ until end of first. I don’t avoid it, I just don’t introduce it “from an early age”, and it doesn’t figure prominently in the decodable texts until then.

        I definitely agree with Jenny about introducing vowels and consonants together to get kids blending and segmenting as quickly as possible. This is one of the reasons I like using Zoophonics. It features “letter-embedded picture mnemonics” and also facilitates practicing all the sounds every day.

      • Harriett Janetos says:

        The graphemes listed in the link that are introduced early are ‘igh’, ‘ay’, ‘ee’, ‘ar’.

      • My comment about it looking just like phonics was not necessarily a support to SWI, it was to point out that I don’t see what’s so special going on. Plenty of programs start introducing pre-fixes/suffixes,etc. of perhaps more limited complexity in 1st to 2nd grade after basic phoneme-grapheme correspondences are learned (i.e. the examples Lynn gives are things like “ing” and “ed”). Even if kids “discover” them, most of the “real spelling” rules uncovered in SWI are obviously nothing new ( though SWI practitioners often seem astonished at this I have noticed). Further, word matrices as seen heavily in SWI are also nothing new. Telling kids a grapheme can represent different phonemes is also pretty common to boot.

        The only thing I see in the more freely available SWI materials is obviously the heavier emphasis on morphology/etymology ( along with throwing out things deemed sacrilege like syllable types), as well as the clear suggestion that there is no sequence to be taught, in fact that is seen as a negative thing. I find this last point to be the most worrisome and questionable considering everything we DO know about at risk readers with approaches that CAN produce results ( and I’ve already pointed out I agree the research does not support phonics only approaches which is one of SWI’s hallmark claims, but the SWI folk here or elsewhere have yet to touch that comment). I also question the representation of SWI as the “scientific investigation of words”, but that is a whole other rabbit hole to go down…

        Many in the SWI community and what it is spawning are also quite frankly offensive in their commentaries as if they are God’s gift to reading instruction and have found the holy grail while the rest of us tell children “lies” about the language. Sorry, but giving kids some tools to learn how to read before I teach them morphology and some spelling rules is not lying – it is being practical and in my experience, expedient and effective, since I work with kids with learning disabilities. This is not to mention, I appear to be doing everything SWI people do more or less, from what I see form Lynn’s handout.

        Sure do more research on morphology instruction and I find all this work has some worth ( and happy to be proven wrong if a mountain of research like we have in other areas presents itself), but let’s not start the reading wars all over again folks…

  19. Pingback: Why Structured Word Inquiry is at odds with basic cognitive science – Filling the pail

  20. Jenny Chew says:

    You and I, Pete, had some email correspondence in December in which one of your comments about the teacher in the ‘rain’ video was this: ‘She’s not even worrying about trying to get everyone to “read” these words yet, but she’s introducing them to the letter names as they spell them out…. But every once in a while she pauses and addresses a grapheme phoneme correspondence in the context of this bank of related words’. I replied ‘I’d want to know how much the children LEARN sufficiently well to be able to apply it from then on’.

    Teaching things in small enough steps to ensure that children can then apply that knowledge in reading and spelling words is an important consideration in the type of synthetic phonics which is now official policy in England. The ‘rain’ video shows it to be less important in SWI: the teacher is working orally on words and mentioning GPCs in passing, but not trying to teach the non-readers in the class how to apply that morphological and GPC knowledge in reading and spelling the words. For me, this means that the video doesn’t show the start of READING instruction – rather, it shows a preliminary to this and gives us no idea of when the children would actually be able to read and spell the words. With synthetic phonics, word-reading instruction starts when the first 3-5 GPCs are taught, so these would include both vowels and consonants, not just vowels as in Lyn Anderson’s approach. We may therefore be at cross purposes about what we mean by the ‘start’.

    • says:

      Hello Jenny – and anyone following this string.

      Please understand that this is just one video that happened to be captured in an SWI classroom that captures a brief moment of the kind of practice that can happen with early literacy instruction with children who are not yet reading. At this point some have argued that this is not evidence of SWI teaching early reading because they think the kids in the video can already read. You are concerned that it’s too preliminary — it doesn’t show the application to the act of reading. And you are right. The claim is not that a non-reader who barely knows his her her alphabet who participates in this lesson is expected to necessarily be able to read these words in connected text the next day. But in the act of working with such a child after this lesson, the learning in this lesson can be drawn on to help them apply this grapheme-phoneme knowledge and morpheme knowledge that has been introduced. Now there are morphemes like “-ing” and “-s” that have been planted and the idea of compounds that are all there in background knowledge that can be drawn on during explicit reading instruction.

      But I want to make clear what that “rain” video was used as an example of.

      Over and over, there is an argument made that children should be taught about phonics (grapheme-phoneme correspondences, and other phonologically based orthographic patterns) BEFORE addressing morphological or etymological concepts. My brother Jeff and I have referred to this in the research as a “phonology first hypothesis”. However, our challenge to that hypothesis is NOT to replace it with a “morphology first” hypothesis. We propose teaching about interrelation of morphology and phonology from the start (and etymology). There are times when grapheme-phoneme correspondences get the main focus, and times when instruction may emphasize morphology more, but they are taught together and with the teacher having an understanding of their interrelationship at all times.

      Just to reiterate for people who may be reading this string, I can understand that some will mistake a challenge about “phonology first” as though the recommendation is “morphology first” but that has NEVER been a claim by Jeff and I or anyone I know who understands SWI. We are talking about including morphology and etymology from the start and how these related to phonology.

      To your point, Jenny, this video is NOT an example of comprehensive beginning reading instruction. It is an illustration of how even children who cannot read can be introduced to morphological and phonological aspects of orthography and how they relate from the very beginning. There is no claim that this is all that you need to do to teach kids how to read. The claim is that this video provides an illustration of one way people working with SWI address morphology and phonology at the very beginning of literacy instruction. To argue that you must teach phonics in isolation of morphology at the start is to argue that there is something harmful in this video in terms of including the morphological context to the grapheme-phoneme instruction there.

      By the way, elsewhere I have pointed to all sorts of videos of different examples of early (kindergarten and grade 1 classrooms) doing other aspects of SWI with word sums, matrices, grapheme-phoneme correspondences. Each time there is some issue raised. In the end, we don’t have the perfect video that people want to see. When teachers are doing their work, they don’t usually have a camera going, and permission from kids and parents to make such videos public. I doubt it will resolve many people’s concerns, but at this link is a document published years ago about this question titled “What if they are not reading yet?” In it, a colleague shares a narrative account of working with a young student who is extremely frustrated with being unable to read, and the path they takIn e together over time using SWI as the frame

      But again, I want to be as clear as I can on a few points around all of this discussion.

      My brother and I have entered conversations here and elsewhere because there is a claim by researchers and others that it is important to teach phonics before addressing morphology or etymology – and for that reason SWI is not appropriate instruction from the beginning of reading. We have done our best to point to both research evidence and videos and stories that counter that argument.

      In terms of this claim that phonics must come first before morphology or etymology to be an empirically based research based claim as opposed to a hypothesis, we would need evidence from instructional studies showing that including morphology or etymology from the beginning with teaching of phonological aspects of orthography results in lower gains than when patterns between letters and sounds are taught in isolation of morphology or etymology.

      There is little research on etymology at this grade level, but there is some early evidence about the inclusion of morphology. In the meta-analyses we have, the effects of including morphology from the start are in the opposite direction of what the phonology first hypothesis would predict. The younger students gained MORE from including morphology than the older students. There are not enough studies in this age, I have serious problems with how morphology is taught in some of those studies, but the fact remains, that finding does not add weight to the argument that we are responding to — the argument to avoid morphology at the beginning of instruction. These findings are from meta-analyses were conducted in 2010 and 2013 (Bowers, Kirby & Deacon, 2010; Goodwin & Ahn, 2010, 2013). We now have a brand new meta-analysis to loo at titled, “Effectiveness of spelling interventions for learners with dyslexia: A meta-analysis and systematic review” by Galuschka, Görgen, Kalmar, Haberstroh, Schmalz & Schulte-Körne (2020) in the journal Educational Psychologist. Here are some quotes from the findings of that paper. In the discussion they have a subheading “Phonics first or Morphology” (I have a problem with that heading as it can leave the reader to imply “morphology fist” rather than or “including morphology from the start – but the findings are the point). Here is the first paragraph under that subheading.

      “We did not find that phonics instruction is more effective than morphological interventions in the early years of for- mal literacy instruction or for more severe spelling deficits. According to our meta-regression analyses, it seems likely that phonics, morphological, and orthographic interventions are applicable across a wide age span. Our qualitative syn- thesis indicates that instruction in morphemic structure might be especially promising for children in the upper elementary and secondary grades (Abbott & Berninger, 1999). However, meta-analyses of morphological interven- tions demonstrated significantly larger effects for preschool and early elementary students than for older students in the upper elementary grades and high school (Goodwin & Ahn, 2010, 2013).”

      In their results section, they have a subheading “Influence of age and severity of reading and spelling deficit on the effectiveness of different spelling treatment approaches”. This is not about the age of the reader, but about the effect of including morphology with struggling readers. In that section they write:

      “Against our hypotheses, the efficacy of phonics interventions decreased with increasing severity, whereas the efficacy of orthographic and morphological interventions increased with increasing severity.”

      It is important to not that they also point out, “However, because of the small data set, main effects and interactions of the meta-regression models did not allow for clear conclusions,”

      So yes indeed there is not enough research on morphological instruction with struggling readers, but once again, this is not the finding one would expect (the authors didn’t) given the standard view of when morphology instruction is appropriate and for which students.

      Finally, I do not expect anyone to understand SWI in any detail or what it looks like without engaging with the process themselves. Blog posts, Twitter, and even reading research articles (the few that there are) is not a way to use scientific inquiry to understand spelling, and to bring that understanding to children. There are lots of people teaching on-line courses and doing workshops and schools doing it that people can visit.

      Regardless, I don’t recommend anyone who is uncomfortable teaching morphology at the beginning to do it – especially without understanding of how it can be done. Those who think it should occur later should teach it later. What Jeff and I are responding to is the recommendation some make that people should avoid including morphology from the start when the research evidence we have doesn’t support that hypothesis, and there are so many people finding this to be so effective. To get more research on this, we need more teachers and researchers who get more effective at this relatively new practice. The research from every meta-analysis that has tested the question suggests that we should include morphology in literacy instruction in general, and that it has particular benefits for the younger and less able. This is based on a limited set of data, but none of the evidence supports the hypothesis to avoid morphology with any group. What we lack in research on how best to teach morphology. I hope that those who are interested critically explore the proposals and examples provided by SWI practitioners. As in any domain there will be people whose practice is better than others. But if we follow the evidence, we have good reason to explore this path rather than claiming it is too much for young or struggling kids. That’s not the evidence from the research, or what I see in schools all the time.

      • Repeatedly, it has been claimed that SWI is meaningful and organised. If that is the case then there must be lesson plans available for the first few lessons of initial reading instruction or at least some higher level description of what these would look like. We do not need to see videos with all of the permissions they involve.

        When I asked Jeffrey Bowers on Twitter for examples he said, “There are soooo many examples out there” and linked me to sources containing the rain video and another video of Nueva where their kids we’re writing questions to the teacher such as “I am curious about subtraction”. Whatever this is meant to be, it’s not initial reading instruction.

        You have produced your own video where a child can already read “Mom says she wants a” but not “cat”. Again, that’s not initial reading instruction. I’m not sure what it is.

        An SWI proponent on Twitter has produced a scheme that involves a lecture about homophones and other aspects of linguistics prior to lesson one of reading instruction. It does address the question of what initial reading instruction looks like but it seems highly implausible and at odds with working memory limits. Do you endorse this program? If not, we still do not know what initial reading instruction in SWI looks like:


        I don’t think anyone is suggesting that learning about morphology is not worthwhile. We have two main problems with your proposal. We cannot see how this could be done from the very start of reading instruction with typical kids who cannot already read. We also find the SWI approach to morphology a little eccentric with its insistence on rules that have ‘no exceptions’ and odd ideas like that ‘box’ has only three phonemes. I would favour a more pragmatic approach.

      • Jenny Chew says:


        You write, of the ‘rain’ video, ‘You are concerned that it’s too preliminary — it doesn’t show the application to the act of reading. And you are right. …’. My view is that if something is an important preliminary to literacy instruction, I would still want to know when children manage to use it successfully in their reading and spelling. Do you know how long it is before the children who are non-readers at the time of the ‘rain’ type of lesson are able to read (and spell) all the words concerned?

        I DO understand that you and Jeff are not proposing morphology and etymology first instead of phonology first but are saying that all three should be taught from the start. Having done a lot of work with real beginners in England, many of them not yet 5 years old, I think this is a tall order, but empirical evidence could settle the issue, and I’m glad that you agree this is needed.

        You write ‘To argue that you must teach phonics in isolation of morphology at the start is to argue that there is something harmful in this video in terms of including the morphological context to the grapheme-phoneme instruction there.’ I, for one, would not argue that there is something SELF-EVIDENTLY harmful, but I would regard it as harmful if it results in children reading and spelling less well than children taught by another approach for which results are available. This is where we need empirical evidence, particularly on the whole ability-range of children taught in particular ways from the very start. And it’s really only there that I have a problem: I think that teaching about morphology should kick in early, starting simple, once children have had a phonics start. I can provide evidence of my views on this, as I contributed a lot to the spelling appendix starting on p. 39 of this National Curriculum document:

        Click to access PRIMARY_national_curriculum_-_English_220714.pdf

        Many schools actually cover some of the simpler morphological stuff in the Reception year (e.g. the suffix), but the National Curriculum doesn’t apply to Reception, so we had to put such things in Year 1 to make sure that they were covered then if not before. Our Year 1 children are probably younger than Lyn Anderson’s, but I was interested to see that she, like us, puts ‘word families of a given base (bases that did not require a change)’ in Grade 1, which presumably means that she would not teach those that DO require a change until later – perhaps in the following year, as we do.

        I’ve just managed to get hold of the Galuschka et al. article you mention but haven’t had time to read it yet. However, it looks as if it’s not about reading and spelling instruction for real beginners.

      • Harriett Janetos says:

        Here’s an interesting passage from the paper Pete cites (Effectiveness of spelling interventions for learners with dyslexia: A meta-analysis and systematic review by Galuschka, Görgen, Kalmar, Haberstroh, Schmalz & Schulte-Körne (2020) in the journal Educational Psychologist):

        “We argue that graphotactic and orthographic-phonological spelling rules, as well as morphological instruction, should be provided AS SOON AS CHILDREN MASTER THE BASIC PHONEME-GRAPHEME AND GRAPHEME-PHONEME MAPPINGS and are confronted with text material for which the basic phoneme–grapheme and grapheme–phoneme mappings would lead to spelling and reading mistakes. To resolve the question if, at the beginning of literacy acquisition, phonics instruction alone leads to better literacy outcomes than a combination of phonics and morphological instruction, more primary studies are needed.”

      • says:

        Hello Harriett,

        You have accurately highlighted an argument that these researchers make — and it is one in line with your own often stated view. They argue that morphological instruction should happen after basic grapheme-phoneme mappings are mastered. Aside from the fact that this leaves the question of which GPCs are “the basic ones,” I would highlight the fat that this statement is not a research finding based on evidence — it is their hypothesis.

        I would also point out that first sentence in the section from which you find that quote, does state a FINDING of their study. They state, “We did not find that phonics instruction is more effective than morphological interventions in the early years of formal literacy instruction or for more severe spelling deficits.”

        Just to be clear, you are highlighting their hypothesis, not a new finding in their study. They say “we argue” not “we found”.

        The research evidence they present is that phonics instruction in the early years was NOT found to be superior to morphological instruction in early literacy instruction, or for struggling spellers.

        I would also note that these findings are in a context where researchers and educators have many decades of practice to refine phonics instruction in the early years, while morphological instruction in these years is in its infancy. While we need more research on early morphological instruction — it seems to me that it is reasonable to assume that we are near ceiling in terms of refining early phonics instruction, and near the floor in terms of refining early morphological instruction. We should expect there to be far more room for improvement of morphological instruction than we have for refining phonics instruction.

        Personally, I see the implications of their finding as far more important than the hypothesis you highlight.

        In terms of hypothesis vs. finding, I also found this statement in their results section interesting:

        “Against our hypotheses, the efficacy of phonics interventions decreased with increasing severity, whereas the efficacy of orthographic and morphological interventions increased with increasing severity”

        I say good on them for presenting the evidence that countered their hypothesis. They clearly expected the efficacy of phonics instruction to increase with the increasing severity of children’s literacy problems, and they expected the efficacy of morphological instruction to decrease for children with more problems. Their hypothesis is in line with what most researchers and educators have been assuming for decades.

        It turned out, however, that the evidence in their study countered that hypotheses. Good on them — they follow the evidence.

        What I find fascinating is this: On what research evidence did they base this common hypothesis?

        If they had based their hypothesis regarding the effectiveness of morphological instruction for those with literacy difficulties on the available meta-analyses and reviews of morphological instruction, their hypothesis should have been that less able children would gain the most from the inclusion of morphological instruction.

        What I think we are seeing here is another example of a deeply ingrained assumption about the appropriateness of morphological instruction for certain groups persisting despite evidence to the contrary. Their findings are in line with all the other findings I know of.. Good on them for reporting findings that counter their hypothesis. But we should also ask, “What was their initial hypothesis was based on? Was it based on the best research evidence, or the common assumption for decades in the research?

        It is with that context in mind that I critically analyze the section of the paper you have cited. You highlight that they “ARGUE” that morphology instruction should happen after “basic” grapheme-phoneme correspondences are “masters” in the paragraph after they state their research FINDING that, “We did not find that phonics instruction is more effective than morphological interventions in the early years of formal literacy instruction or for more severe spelling deficits.”

        It seems to me that, especially given their record — we should be putting more weight on their research findings than their hypotheses.

        The final sentence you quote is very important as well, “To resolve the question if, at the beginning of literacy acquisition, phonics instruction alone leads to better literacy outcomes than a combination of phonics and morphological instruction, more primary studies are needed.”

        YES as I have argued over and over, we need more research on early morphological instruction.

        I almost agree with the statement that we need research comparing phonics alone to a combination of phonics and morphological instruction. I would say instead — we need research comparing “phonics alone” to instruction about the INTERRELATION of orthographic morphology and orthographic phonology.

        But this highlights another point I’ve tried to make over and over. For the claim to avoid inclusion of morphology at the beginning of literacy instruction to be a researched based claim, we would need evidence that including it results in lower gains than teaching phonics alone.

        I have yet to be pointed to any study showing including morphology reduced the effect of early literacy instruction. We definitely need more and better research on this question. However, if we draw on the research evidence we do have — the logical hypothesis should be that including morphology from the beginning of literacy instruction is more effective than phonics alone.

        Do you know of research findings — not hypotheses — that contradicts that analysis?

  21. Harriett Janetos says:

    Another SWI proponent says:

    “The reason the Phombies recommend waiting to teach morphology and etymology later is because they don’t really know what each of those actually is. Their vision of morphology and etymology must fit into a triangle.”

    Vision is the operative word here. As Greg says: “I don’t think anyone is suggesting that learning about morphology is not worthwhile . . .We cannot see how this would be done from the very start of reading insutruction with typical kids who cannot already read.”

    If I am to present SWI to my K-1 teachers as our new focus for early reading instruction, I need to SEE exactly what that model looks like.

  22. Jenny Chew says:

    Pete quotes this from the Galuschka et al. article: ‘We did not find that phonics instruction is more effective than morphological interventions in the early years of formal literacy instruction or for more severe spelling deficits.’

    I’ve now read the whole article and found it very interesting, but thought that all the studies were about spelling interventions for learners with dyslexia – i.e. not about initial teaching for the full ability-range, which is surely the point which is at issue for many of us. I’ll need to do a re-read to check whether Pete’s quotation is relevant in this respect.

  23. Harriett Janetos says:

    Pete, you are right that this statement aligns with my view of early reading instruction.

    “We argue that graphotactic and orthographic-phonological spelling rules, as well as morphological instruction, should be provided AS SOON AS CHILDREN MASTER THE BASIC PHONEME-GRAPHEME AND GRAPHEME-PHONEME MAPPINGS and are confronted with text material for which the basic phoneme–grapheme and grapheme–phoneme mappings would lead to spelling and reading mistakes.”

    But I want to clarify that my experiences teaching beginning readers preceded my knowledge of neuroscience and cognitive load theory, both of which confirm what I do. And every video of SWI that I’ve watched (and I have now watched a lot) inspires me to supplement, not supplant, what I do at the very beginning stage of reading instruction.

    • says:

      Great Harriett,

      I am very pleased that what you have seen with SWI has encouraged you to to supplement your instruction. I am not asking anyone to engage in instruction with which they are not comfortable based on their experience. I am sharing the practices I find useful, and my best understanding of the research. I’m in a workshop right now, and yesterday — like every other workshop I give — I made the point that I don’t want people to go home with the message that they should drop practices that they find useful just because they came to a one or two day workshop and they think that what I said means they should drop practices they feel are important. Instead, I say, if you learned something about how spelling works that you did not understand from your previous training that did not give you, that is enough evidence that you should begin the process of seeing if you can use that new understanding to improve your teaching. Typically people find the matrix and the word sum and addressing grapheme-phoneme correspondences in that context, and also the process of “spelling-out-loud” and “writing-out-loud” of word structure interesting and effective. The process of beginning to try those new practices with students is the process of beginning to understand them. Over time the teacher gets to choose if and which of their own practices they might want to drop if any. I do not ask teachers to drop practices they feel confident in to replace them with practices they do not understand yet. You and many others may find they never choose to teach morphology until they have taught phonics for a while. That is totally your choice. I just share what I do so that people can consider options instead of just continuing with practices because they didn’t know there were these options to consider.

      I wrote a piece on this topic years ago that addresses this point in more detail called “How do I ‘integrate’ SWI practices with my other literacy instruction practices?” at this link:

      My response to the quote you highlighted from that article has to do with how I see lots of people cite research to say that what SWI suggests for literacy instruction for early and struggling readers is harmful and should be avoided. You didn’t say that in so many words. But you quoted a hypothesis the researchers made that when read by many would be taken to say “see the research says don’t teach morphology from the start”. There can be a big difference between what the research finds and what researchers suggest. And when people conflate the two, people end up making claims that the research shows that what SWI suggests for early readers is shown to be harmful for kids by the research. That is something I feel compelled to combat.

      I have yet to see a single finding from research that shows including morphology at the very beginning results in lower gains than if we teach phonics in isolation first. With all the intense debates we see around this point, I would expect that if there was such research, it would have been pointed out by now.

      One reason this question of SWI for the beginning reader is frustrating to me because it seems to colour the whole topic of this kind of instruction at any age or with any ability. I’m delighted that you have taken lessons you are learning from SWI to supplement your instruction in ways that you find useful. This is what I recommend for anyone reading this string. Consider other options you find. Don’t feel pushed into doing practices you are not comfortable with, but be willing to experiment with new ideas that seem interesting and come to your own conclusions.

      And finally, I need to emphasize that blog discussions and Twitter etc are not a venue in which anyone can effectively explain what SWI instruction looks like in general, let alone in the very initial stages of formal instruction. These are also not the venues that people do their training for using phonics for early literacy instruction. For some, like you, they are an effective launching pad for studying other videos, entering discussions etc and apparently supplementing your instruction. Great! But if people are interested in getting a deeper picture of what SWI can look like (notice the word ‘can’) look like, I can recommend one on-line place that you can study — but not just by reading text, by engaging in a live on-line course. I “endorse” very few resources. But I can recommend working with Rebecca. Her website does have lots of great resources, but if you want to study with someone who regularly teaches with young readers in this way, you can take her course titled “SWI for Early Readers”. Her other courses are excellent too. But instead of me attempting to explain what this instruction can look like in those beginning stages in a blog comment, anyone seriously curious could study with someone in that appropriate context. See her website here:

      • I see two problems. Firstly, although you worry that the idea of GPCs, morphology and etymology from the very beginning of reading instruction has dominated the discussion, you are the cause of this because you are the ones insisting that it can be done. This is not how anyone has traditionally taught, be they whole language or systematic phonics practitioners. Therefore the burden of proof lies with you. You need to present evidence to support your case. Yet you seem to think the burden lies with sceptics to disprove it.

        Secondly, why cannot SWI be explained in blogs? People like John Walker of Sounds-Write can easily explain their approach in blogs even if this does not constitute training. If SWI relies on teachers all being trained in a set of ideas too complex to explain in 1000 or so words and then making their own decisions on the nuances of how to apply them, it cannot possibly be any kind of solution to improved literacy at scale. There are just too many teachers and too many schools.

      • Harriett Janetos says:

        Here’s a very brief explanation of SWI by Rebecca Loveless:

        SWI is not a program with a fixed scope and sequence. The term represents a method for using what is true about our language to understand it better. Instructional practices are left up to the teacher or tutor, as long as they support the accurate science of language. Any word that provokes curiosity or interest in a student is a good one to begin with. One of the first words I worked with was because a first grader heard it and wanted to know more about it. Boy, did we learn a lot with that one.

      • Harriett Janetos says:

        Here’s the complete sentence from the Rebecca Loveless quote:

        One of the first words I worked with was ‘grandiloquent’ because a first grader heard it and wanted to know more about it.

      • Jenny Chew says:

        Under the heading ‘Diving deeper into word bags’ on the Loveless site there’s this:

        ‘After the exciting Word Bag lesson that sparked so much interest, the teachers came to me with some big questions. One of the children created a word family at home that contained several cards reading and a few cards with long strings of random letters. How should they handle these words without crushing the child’s excitement?’

        To me, this means that whatever the child had learned from the lesson was pretty garbled. Surely a child at this level needs teaching about something simpler than word families.

  24. Jenny Chew says:

    Like Greg, I think there’s a burden of proof issue. You say, Pete, ‘I have yet to see a single finding from research that shows including morphology at the very beg inning results in lower gains than if we teach phonics in isolation first.’ Isn’t that because no research has actually been done on the effects of teaching your way from the very beginning? This didn’t happen in the Devonshire et al. study, and although other studies which you cited in your 2010 meta-analysis involved pre-schoolers (Lyster et al and Tyler et al.) the morphological training was done orally before the teaching of GPCs and reading and writing started. Couldn’t you and/or Jeff have arranged for the right kind of research to be done by now?

    The Clackmannanshire study which was done in Scotland has had a major influence in England. In that, the researchers reported scores on reading and spelling tests which were either standardised or widely used, and this made it quite easy for people to put the performance of experimental and control groups in perspective. Couldn’t you at least report standardised test results from schools using SWI?

  25. Harriett Janetos says:

    Here’s how Diane McGuinness (Early Reading Instruction: What Science Really Tells Us About How to Teach Reading, 2004) explains in 138 words her approach to beginning reading:

    Prototype for Teaching the English Alphabet Code

    1. No sight words (except high-frequency words with rare spellings).
    2. No letter names.
    3. Sound-to-print orientation. Phonemes, not letters, are the basis for the code.
    4. Teach phonemes only and no other sound units.
    5. Begin with an artificial transparent alphabet or basic code: a one-to-one correspondence between 40 phonemes
    and their most common spelling.
    6. Teach children to identify and sequence sounds in real words by segmenting and blending, using letters.
    7. Teach children how to write each letter. Integrate writing into every lesson.
    8. Link writing, spelling, and reading to ensure that children learn that the alphabet is a code, and that the code
    works in both directions: encoding/decoding.
    9. Spelling should be accurate, or at a minimum, phonetically accurate (all things within reason).
    10.Lessons should move on to include the advanced spelling code (the 136 remaining common spellings and 80
    sight words).

    • Jenny Chew says:

      And here’s an account of the synthetic phonics approach used in the Clackmannanshire study – 306 words, including some points of comparison with analytic phonics:

      ‘SYNTHETIC PHONICS 1.7 This led us to look at synthetic phonics, which is a very accelerated form of phonics that does not begin by establishing an initial sight vocabulary. With this approach, before children are introduced to books, they are taught letter sounds. After the first few of these have been taught they are shown how these sounds can be blended together to build up words (Feitelson, 1988). For example, when taught the letter sounds /t/ /p/ /a/ and /s/ the children can build up the words ‘tap’, ‘pat’, ‘pats’, ‘taps’, ‘a tap’ etc. The children are not told the pronunciation of the new word by the teacher either before it is constructed with magnetic letters or indeed afterwards; the children sound each letter in turn and then synthesise the sounds together in order to generate the pronunciation of the word. Thus the children construct the pronunciation for themselves. Most of the letter sound correspondences, including the consonant and vowel digraphs, can be taught in the space of a few months at the start of their first year at school. This means that the children can read many of the unfamiliar words they meet in text for themselves, without the assistance of the teacher. By contrast in analytic phonics, whole words are presented and pronounced by the teacher, and the children’s attention is only subsequently drawn to the information given by letter sound 12 correspondences. Typically in Scotland with the analytic phonics approach, it would not be until the third term of the first year at school that children would be made aware of the importance of letter sound correspondences in all positions of words, whereas in synthetic phonics this is done at the start of the year. The full analytic phonics scheme is usually not completed until the end of the third year at school.’ (, pp. 11-12)

      • Okay, I’m going to give a shot at my synopsis of the SWI method. This is a serious attempt based on everything I’ve looked at and might help others who can get lost in the rhetoric. Anyone in the SWI community feel free to modify so as not to misrepresent what you are doing, perhaps you can use it on the materials for people investigating your approach so they don’t have to wade through a bunch of material to come to some basic conclusions.

        SWI starts with the child’s (or the child’s fellow classmates) interest of words and attempts to grow this through, what we believe, is a scientific analysis of a words interrelated morphological,etymological, and phonological structures as guided best as possible by the teacher’s scope of knowledge in this area and which requires significant training in order to employ properly and which most teachers lack.There is no reason as to which words are studied outside of pure desire for knowledge on behalf of students, and no identified benchmarks or goals in implementation other than to know how these words are working from standard linguistic perspectives and which are also commonly taught in most any quality literacy program; however SWI places most emphasis on orthographic information from the outset. The other key thing in SWI when compared to other approaches, is the lack of scope and sequence to when anything is taught and the emphasis that kids understand how to approach a word like an independent linguist from the start of instruction as opposed to the teacher driving the curriculum.Through this method we hope they will get the tools they need to become competent readers, spellers, and writers; it is up to the teachers discretion who needs further help or remediation in this area and how best to employ that. Aside from some initial cursory studies, there is currently no solid research basis for this method being more effective than others, however no one has disproven that it does not work when compared to other approaches, so we offer it to any teacher who wants to try it in whatever capacity they feel comfortable employing it.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.