Donna Scanlon has been mentioned several times in these comments. If you listen for 10 minutes at 17:30 in this recent interview (https://www.classroomcaffeine.com/guests/donna-scanlon), she echoes this statement by Dr. Sam: "When in-servicing teachers, I urge them to encourage students to use the graphophonemic information first and then cross-check it against context & syntax." She distinguishes between word identification and word confirmation and discusses the importance of "set for variability" (flexible pronunciations--"try a different sound") in a way that I can't see anyone disagreeing with. And yet . . . she refers to SOR being against using contextual cues, which illustrates part of the confusion that confounds reconciliation.
She also laments the bad associations attached to the term "balanced literacy" and offers the term "comprehensive literacy". I don't use "balanced literacy" or "structured literacy"; I call what I do "inclusive literacy". What's in a name?
I did not avoid the question, I pointed out that your concerns fall into the category of research design. Scanlon's decades of research should not be discounted or ignored. I hope I'm clarifying my point that your views around the three cueing systems are not what many folks try to make it out to be. It was never meant to be a teaching system- and the attempts to turn it into are not what I teach my teachers to do. The best model I am aware of for using ALL the information system is the one employed in Reading Recovery, and again, what folks say is done in RR is not what RR folks are doing.
As you read more of my writing, I always seek evidence of common ground and common practices. So I'm VERY interested in Seidenberg and how his views fit Clay's views around creating a self-extending system. I'll also point out that laying the foundation should happen early on. What that foundation might be is also determined by the needs of the students. Some students thrive on analytic approaches (essentially discovery learning). Some students get the foundation rather quickly. How do you see SOR adapting to those facts (or do you disagree with that assessment)? I'm seeing examples of SOR folks treating analytic phonics as an inferior approach. It's not. What to use depends on the student (however I've said many times that it is crucial that synthetic phonics be part of any Tier 1 program).
Before you answer about your views around analytic phonics, please look over this excerpt from a July 18th blog entry, Shanahan posted a few years back. "The average effect size was somewhat higher for synthetic than analytic approaches, but not significantly so (it was so small a difference that one can't say one is really higher than the other). In other words, synthetic and analytic phonics are equally good."
I'm not seeing much evidence that SOR is treating things that way. What am I missing?
BTW, I'm glad to see some things on which we seem to agree. I'm glad to see that.
"please listen to this interview by Bruce Howett another person who is trying to find a way to end the reading wars."
On the one hand, I applaud Bruce for promoting multicomponent instruction by emphasizing David Share's new article (Share, D. (2025). Blueprint for a universal theory of learning to read: The Combinatorial Model. (Under Review). Academia. https://www.academia.edu/118772944/Blueprint_for_a_universal_theory_of_learning_to_read_The_Combinatorial_Model), which presents a multicomponent approach called The Combinatorial Model, along with promoting Maryanne Wolf's multicomponent work on POSSuM (phonology, orthography, semantics, syntax, morphology).
On the other hand, both Share and Wolf acknowledge the importance of decoding instruction for the novice reader before complicating that instruction with other components, but Bruce seems to be more inclined to accept the methods of Structured Word Inquiry, which teach phonology, morphology and etymology simultaneously to beginning readers, an unhelpful complication I wrote about in Must Phonics Fail in Order for Structured Word Inquiry to Succeed? (https://learningbydesign.com/professional-development/spell-links-blogue/).
This is another good example of why we need to define our terms and see where in the instructional sequence they fit in (word identification before word confirmation; simplified decoding instruction before complicated morphology instruction, etc.).
“Can you be more precise about: What are David's and my fundamental differences? " The definitions of reading you have provided so far are so broad and general it is impossible to tell what your views are about the issue of your take on the simple view of reading. The simple view focuses on comprehension (language comprehension). It fails to account for factors like cognitive processes (Pearson's work on the was groundbreaking). Some comp strategies can and should be taught. Background knowledge is not sufficient. Also, things like social cultural context. Pearson has taken the time to post his response to your response on his site. BTW David had 10 pushbacks, not just one. So, let's start with this- reading is fundamentally a meaning-making process. Do you agree or disagree, and why? Presenting the question about Balanced Literacy in a separate comment.
“ Pearson has taken the time to post his response to your response on his site.”
I posted his (and Rob’s) reply on my substack, followed by my replies. 3 parts I think. Maybe 4. Whatever, pls check them out, then we can resume. I don’t have my computer so can’t easily navigate b/f, but lmk if you can’t locate.
Not sure how to get to that- I'm new to substack- all I can see so far is this particular thread. I need to have you answer the "Can You Be more precise" question above.I'm sensing that the most likely source of your differences lies in the limits and limitations of the simple view of reading. In a nutshell, listening comprehension plus decoding is not a satisfactory model, it is an incomplete model. The simple view is just a little too simple!
Perhaps a lack of response is a function of the “sides-taking” in the reading wars. People who take issue with your views, but sniff the product and leave the store. I don’t know. I do know your mission is worthy and I hope you persist.
“a lack of response is a function of the “sides-taking.” Indeed. And there are many possible explanations. One is what you suggest.
Another, for which I have anecdotal evidence, is that members of a tribe (=side) who speak out against positions from the tribe, even if they know they’re ill-informed and unhelpful, risk marginalization or excommunication. The silencing, either self- or other-imposed, is very harmful regardless which tribe does it.
On the other hand, tribe solidarity increases the likelihood of tribe survival. So there’s that. I guess the question is when does tribe solidarity (probably a good thing, at least for the tribe) turn into tribal group-think (probably not)?
Psychologists might suggest a post challenging those who agree with you on some issue likely to draw a reaction. But psychologists are I know from personal experience a curious group, so you might not want to take a suggestion from a card carrying member of that tribe.
"Based on my personal teaching experiences, I’d begin with synthetic phonics (or would insert synthetic supports into an analytic program if I were required to teach that). It’s just easier for kids."
I agree--based on my personal experiences. I would also say that it's much easier for teachers. Is there an analytic phonics program you recommend?
From Reading Development & Teaching, Stuart and Stainthorp, 2016:
"We have found only two adequately controlled studies published since the National Reading Panel report which have succeeded in directly comparing the relative efficacy of synthetic small-unit phonics and onset-rime analogy large-unit phonics in promoting reading growth. . . Both conclude that teaching small units is more effective than teaching large units.
Christenson & Bowey (2005). The efficacy of orthographic rime, grapheme-phoneme correspondence, and implicit phonics approaches to teaching decoding skills. Scientific Studies of Reading, 9, 327-349
Hatcher, Hulme & Snowling (2004). Explicit phoneme training combined with phonic reading instruction helps young children at risk of reading failure. Journal of Child Psychology and Psychiatry, 45, 338-358.
Further support for the decision to teach grapheme-phoneme correspondences rather than onsets and rimes comes from analysis of the relative numbers of each type of correspondence required for children to be able to read the words they typically encounter on their reading materials, which we reported in Chapter 4: children need to know almost five times as many onset and rime correspondences as grapheme-phoneme correspondences to read all monosyllabic words in the early version of the Children's Printed Word Database (Stuart, Dixon, Masterson & Gray, 2003). It is simply more efficient to teach children grapheme-phoneme correspondences."
Deal- and if you check I have been answering you questions. We have more than one thread so let me know what I've missed. The answer to this particular question of mind is very important..
Sorry if I wasn’t clear. These are the questions you’ve missed:
“Can you be more precise about: What are David's and my fundamental differences? and which of my "views around Balanced Literacy [are] misleading and incorrect"? and also, just to be complete, which of my "criticisms about BL have merit"?”
Can you be more precise about: What are David's and my fundamental differences? and which of my "views around Balanced Literacy [are] misleading and incorrect"? and also, just to be complete, which of my "criticisms about BL have merit"?
What exactly to you see as language comprehension? Does it go beyond listening comprehension? If so, how? Explain specifically how things like background knowledge, motivation, and cognitive process fit in. What are your views on how sociocultural context impacts reading ability? What are your views on the nuances of reading strategies and whether reading strategies can and should be taught directly?
This is in response to your comment about Scanlon.
First- the research design used by Scanlon was published in RRQ, so they found it acceptable. You and Shanahan have the right (obligation) to raise questions about designs; the readers have the right to decide if the issues raised merit completely discounting all she had to say.
Second- there is much more to Scanlon's research than simply one paper. She has a new textbook out and a variety of publications. I think folks reviewing that large body of research done over a couple of decades will conclude that context needs to be used in word recognition. Current laws are removing it entirely. That is a HUGE mistake.
Third- while some BL folks might call for all cues (information) to be used co-equally, as you described, others see such as myself see context as playing a CRITICAL role (not a secondary role). How can one encourage a set for variability in their students if they fail to crosscheck the graphophonemic information with contextual information ("read"- should it be pronounced red or reed)? Frankly, the words cue/cueing have become divisive. I tend to use the words information/information systems. When in-servicing teachers, I urge them to encourage students to use the graphophonemic information first and then cross-check it against context & syntax. Clay NEVER said to only (or mainly) use meaning and context. She specifically encouraged using ALL information and cross-checking that information. To create a self-extending system. A topic for another conversation has to do with Seidenbergs recent ideas about 1. direct phonics instruction early 2. a "take off" point where the child begins teaching themselves. That sounds very much like Clay's self-extending system. Your thoughts?
You're avoiding the question of what you make of a comment Tim S made about Scanlon's work (".... misleading.... She draws conclusions that the original researchers wisely did not draw… Context definitely plays a role in decoding and decoding development, but it is not the role 3 cueing advocates cling to"), given that you've endorsed Tim S's nuanced view. I guess I wasn't clear it was a question, since I just said "I'm wondering." So apologies for the miscommunication, but I'm still wondering (honestly).
"context needs to be used in word recognition. Current laws are removing it entirely. That is a HUGE mistake." Completely agree. (Tim does too, btw).
"Frankly, the words cue/cueing have become divisive." Agree again. 100%, or more, if it's possible to agree more than 100%. That's the point of the DEADLY WORDS posts I've been working on. Whether the correct adjective to put in front of "context" cues/information is CRITICAL or secondary is surely a matter for discussion, although not sure how critical that discussion is. Maybe it's more critical than I realize.
In any case, I was very glad to read this: "When in-servicing teachers, I urge them to encourage students to use the graphophonemic information first and then cross-check it against context & syntax." Once again, we are in agreement, although personally, I'd prefer a somewhat stronger infinitive than "to encourage", eg "to teach."
Then whether we call it "a "take off" point" or a "self-extending system" is less important than providing the sort of instruction that will hasten students' getting there. What you provide in your in-services sounds like it will help accomplish that.
"When in-servicing teachers, I urge them to encourage students to use the graphophonemic information first and then cross-check it against context & syntax."
"We need to stop treating reading like a Rorschach test. Like the hare, the untrained reader is enticed by the three-cueing system to search for distractions on the page, seeking answers off the beaten path of best practices that simply guides them to apply their decoding skills to attempt an identification of a word, to be flexible with pronunciations, and to confirm their choice by checking the context of the sentence related to syntax and meaning. Here’s my three-cueing system: grapple with the graphemes, flex the phonemes, confirm with context."
I did not see this original reply. Let's take you up on your offer to explain YOUR definition of reading. Do you subscribe to the simple view of reading.? You identify two components. word recognition and language comprehension. What exactly to you see as language comprehension? Does it go beyond listening comprehension? If so, how? Explain specifically how things like background knowledge, motivation, and cognitive process fit in. What are your views on how sociocultural context impacts reading ability? What are you views on the nuances of reading strategies and whether reading strategies can and should be taught directly? Thanks for you input about those things.
I don't recall my offering to explain my definition of reading, but regardless, I don't have a single unitary definition for the simple (or complex) reason that, to quote myself, "There is no universally accepted answer to 'What counts as reading?'"
I'm afraid you're evading Harriett's request for specificity and my followup questions about your claims: What are David's and my fundamental differences? and which of my "views around Balanced Literacy [are] misleading and incorrect"? (you can skip my additional "just to be complete" question, which of my "criticisms about BL have merit"?)
Hi Dr. Goldenberg! I just wanted to say thank you for the work you're doing to share your perspective and educate us non-researchers on the research. I'm coming at this from the business side of things, having worked in the K12 market for a decade. My hope is that the folks making purchasing decisions in schools are doing so with eyes wide open—not falling for a clever curriculum provider slapping "SoR" on every product.
Leah, thank you so much. Your hope is mine as well. My question is how can we make that happen... not falling for a clever curriculum provider slapping "SoR" on every product. This is not a rhetorical question. It's an urgent one! Seems to me that everyone who cares about reading has a role here. Educators (of all stripes... teachers, paras, administrators, leaders of teacher unions) need to be better informed and so do publishers. And so do policy-makers, elected and not elected. And advocates for different populations of kids. And--dare I say it?--academics. And bloggers too. And parents!
Each has a responsibility in their respective domains to check out claims and call out false ones. I'm not looking for villains or casting blame. But the fact is--and I do believe it's a fact--that publishers and many others, knowingly or cynically, can't say for sure--play fast and loose with the terminology. SOR is exhibit A. (Did you see Emina McLean's blog post link I included in a recent post on this Substack?) The other terms in the DEADLY WORDS posts I've written are also in play. So how do we deal with that. I would 100% welcome your and other members of the publishing and business side of things engagement with this. I can't thank you enough for this comment and your interest. Keep commenting here or feel free to email me directly. Thank you.
A good first step would be for you to explain your definition of reading. Specifically, does it include the term meaning-making? Specifically, does it view decoding first, meaning-making next or decoding and meaning-making as concurrent? Thanks for the specifics on your views about this.
I wonder if you saw my reply. I might have replied in the wrong place, so here goes again.
I addressed this question in part IV of my reply to Tierney and Pearson's response to my critique:
What counts as reading.
Rob and David say in their reply that “As near as we can tell, Claude accepts the intentionally narrow definition of reading, put forward by Rayner and his colleagues in the classic 2000 paper.”
I don’t know on what basis they drew that conclusion. They could have gotten nearer, and more accurate, by just asking me.
Or, as easily, they could have gone to Claim 3 in Table 2 of my critique. To replace the claim they claimed to fact-check (“Reading is the ability to identify and understand words that are part of one’s oral language repertoire”), I proposed this fact-checked improvement:
"There is no universally accepted answer to “What counts as reading?” It can be defined in many different ways reflecting different perspectives. Where there is agreement is that skilled reading comprises two important principal components—word recognition and language comprehension—and within those, many skills, understandings, and attributes are involved."
Reading can be used to refer to many things, from the most basic and limited—decoding written words (whether understood or not)—to the metaphorical—for example, reading the room, just as “literate” can refer to someone who can read and write at a basic level or to someone well-versed in some literary canon.
The meaning of reading and literacy have also shifted, as they went from being exclusive domains of an elite and tools for maintaining control over others to now oft-stated aspiration for “universal literacy.”
Thank you for this thoughtful message and for continuing the series. I rely on your posts to keep me informed and to help shape my arguments when engaging with those who hold opposing views. While I don’t often leave comments, as I’m usually in notetaking mode, I deeply value the insights you share.
I encourage you to continue this series, as it’s essential to advancing meaningful conversations in literacy. I also agree with your suggestion to spread the message more broadly and will forward this to others, including those on "the other side," to help bridge the gaps.
I thought that, after 5000 years of reading, we might have figured out how to teach it by now. But I'm wondering if I should just check back in another 5000 years to see if the wars are over yet.
Anyone who starts to investigate how to teach reading today is most likely to get hit with a barrage ("bar rage", SWI people?) of information about a magic new thing called SoR and that Phonics is IN and Whole Language is OUT etc.
If they dig a bit deeper, though, they’ll find that there are still raging debates about how best to approach literacy instruction.
SoR: "The science is in! Phonics + lots of other stuff is the Way".
SWI: "Nope, phonics is a failed attempt to explain a writing system that involves phonology, morphology and etymology."
S2P: "Nope, stop messing up kids' sock drawer." (Huh?)
There have been discussions/arguments for years over the "evidence" for various approaches, but weirdly, despite the daily literacy studies that keep rolling in, no one seems to be interested, game or willing to put these approaches to the test side by side.
Maybe it's pointless anyway. Project Follow Through pitted various teaching approaches against each other over nine years and produced startlingly clear results. Unfortunately, they weren't the results most people wanted, so they were just ignored.
Haha. Humans, eh?! I feel like just going back to grunting.
Ralph, I appreciate your thoughtful comments. Going back to grunting sounds very appealing. I do it ocassionally. I'll have to try it more often.
Re this: "Project Follow Through pitted various teaching approaches against each other over nine years and produced startlingly clear results." Unfortunately (but we have to deal with it bc reality is really a mess) there were at least 2 "startlingly clear results." One was I believe what you probably mean, that direct instruction (capital D capital I) got the best results on student outcome data.
The other was that the variability across sites was greater than the overall "main effect" of DI. What this meant, not to belabor the point, was that local circumstances (which could mean all sorts of things including the quality of the teaching, strength of school leadership, etc etc) were much more important determinants of student outcomes than the DI program itself.
This finding echoed Chall's from years before (in a much less rigorous statistical analysis, but, as it turned out, no less valid) that programs/teachers--I don't recall the unit(s) of analysis--that featured decoding produced better results, in general, than programs that did not, BUT that there was a great deal of variability by teacher or program or something... sorry to be vague, but it was the same larger point.
So the moral of this story, at least as I understand it, is that results can/will be messy and point in different directions and support different interpretations. Mine (and, to my knowledge incorporates all the reading research with which I am familiar, which does not include every single thing ever published) is this: In general, phonics (or whatever you call explicit and systematic instruction in foundational literacy skills) makes a net positive impact on student reading development and should be hardwired into every reading/literacy program, classroom, school, district, etc. But by no means is explicit and systematic instruction in foundational literacy skills the only thing that matters. For starters, students will need wildly different amounts and intensity of such instruction, from relatively very little to A WHOLE LOT. And then even those who need A WHOLE LOT, also need other things, most notably but not limited to language development (vocab and more), knowledge building; and all will need excellent instruction, competent school-wide leadership. So that's my takeaway, for better or worse.
Thanks Claude. It's absolutely true that literacy is a whole lot more than just phonics.
I learned about Project Follow Through back in the 90s, and whatever its possible failings, it does seem from every angle that DI had a huge effect in all places compared with its control groups — unlike the other methods, which sometimes performed way below the control groups, despite having a lot of funding. Even if the results aren’t perfect, it doesn't explain why the results were largely ignored. (My primary schooling took place after results showed the benefits of DI and the failings of Whole Language, and yet all I got was Whole Language.) There was clearly a lot of politics involved. It's interesting that interest in Follow Through is starting to appear again. I know you've written about it, and in the last year (even in recent days) it's been the topic of podcasts, such as Zach Groshell's (https://www.youtube.com/watch?v=vrly4mmJZcA) and Anna Stokke's (https://chalkandtalkpodcast.podbean.com/e/project-follow-through-direct-instructions-overlooked-success-with-marcy-stein/). It's pleasing that there's more interest in best practices these days, although our human nature makes it hard to have our ideas challenged. I confess that, as a fan of explicit synthetic phonics, I've been a bit rocked by the Structured Word Inquiry (SWI) approach which clearly shows the failings of phonics. But it obviously has huge merit, so one has to be willing to look again and adapt where necessary. I'm actually glad that SWI has merit, as I'm a bit or a word nerd anyway.
Ralph, my computer tanked so I’ll reply more completely when I get it back. My main point was that the lessons of FT, just as the lessons of Reading First, are not as unambiguous as you might think. More later. And thanks for your engagement.
hi again, Ralph. Here's something I wrote for an IRA reading research conference 20 yrs ago (😵💫). If you're interested in the entire talk (mostly about ELs or EBs or MLs... back then it was ELLs), I'm glad to send you my speaking notes, the ppt, and a pdf of the 1978 article I quote from below.
(Fyi, the title of my talk was "Ideology, Research, and Practice." I used the FT experience as an example of how "ideology" influences, even if not fully determines, the conclusions one will reach about complicated findings that have have more than one plausible interpretation (which is to say, just about all research findings). One could also reasonably argue that "ideology" will actually fully determine the conclusions, even when results are straightforward by, e.g., finding fault with methodology, measures, populations, whether measurement isn't even involved, and on and on.)
"Some commentators have argued that FT was spectacularly inconclusive, that no program consistently outpeformed the control groups, and that in general none of the program models was successful in reliably improving children’s chances of school success. But others have concluded exactly the opposite—that overall the DI model got the best results on a number of outcome measures; not at every site, but if you average the effects across the sites, DI obtained superior results. (modestly superior, but superior)
"So what really happened? It turns out that both sets of commentators are right—but they’re just not right enough. Each has emphasized only one part of the story. Let me quote from a 1978 article written by the authors of the original evaluation report, trying to deal with and clarify the commentary that had been heaped on their report in the mere year since its release: “Models that emphasize the kinds of skills tested by certain subtests of the Metropolitan Achvt test have tended—very irregularly—to produce groups that score better on those subtests than do groups served by models that emphasize those skills to a lesser degree.” In other words, #1: overall the children who received instruction that emphasized skills measured by the standardized test used (and these children were in DI or DI-type programs) outperformed children who received a different type of instruction. And #2: this pattern varied considerably and was not consistently found across the sites."
Thanks Claude! Yes, it would be interesting to see those notes and things you mentioned if they're easy enough to pass along. Thanks for the offer. I don't have a stake in this, but it is interesting. To be honest, hearing such widely differing views on research does make me question research in general. Is there really much useful research going on anywhere, or are people just generally confirming their own biases? It's a bit depressing, really. In the end, I find myself just having to use my own judgement and experience to make decisions about how and what to teach, making the best of the information I can find. None of my grandma's best recipes are backed by peer-reviewed science, but they sure are good to eat.
I understand your reaction, which is to "question research in general," and the other musings you share. I don't think those are good solutions. Maybe bc I have an irrational faith in rational processes, I believe we're better off doing and paying attention to research than not doing or ignoring it. It's obviously true in the physical and biological sciences (unless you're RFK Jr, various climate deniers, and many members of a political party that shall go unnamed). It's less obviously true in the social sciences, which sometimes seem to deny its own scientific standing--inadvertently or vertently (if that's a word).
A much better solution imho is a better understanding of research by educators at all levels, pre-school to PhD school, + those making policies that affect educators and the to-be- and are-being-educated. This understanding includes findings from research (gnarly as they might be), in addition to something about how research is done, how provisional and probabilistic those findings are, and that skepticism is a part of research and of "science" more broadly. People equate science with certainty. That's a big mistake. See this piece if interested: https://www.dropbox.com/scl/fi/pdx244x7y3s992cy4dh22/EdWk2021Science-and-skepticism-no-pswd.pdf?rlkey=apxbgmmylcvwtqzr52rt0fqjb&dl=0.
As for the ones I mentioned in my last reply, see these; let me know if you have trouble accessing. Feel free to email me.
Thanks very much, Claude. Interesting reading. I kind of have to throw my hands up in regard to assessments of Follow Through. The various opinions I hear about it vary so widely that it's kind of pointless to try to draw a conclusion from it (which is perhaps that whole point!) It's like trying to get a handle on the Kennedy assassination by pitting the Warren Commission report against an Oliver Stone documentary. It's basically a "choose your own adventure" scenario. In my own experience, a more explicit, DI kind of approach has been far more fruitful than other teaching methods, and at the end of the day, it's probably reasonable for teachers to learn from their own experience and that of their peers — much like cooks and tradespeople learn from their mentors and peers. Researchers seem to be in a world of their own a lot of the time.
Thanks Harriet. I actually read that a few days ago. Great work! I must admit, though, I'm still befuddled by the sock drawer analogy. I get the theory, but can't see the point of it in practice. Do you know of a practical example that shows how an S2P lesson would work? I haven't been able to find any, although I've read quite a few articles and interviews with Jan. Perhaps it's too much of a proprietary system and I would have to pay to find out, but I'm just conceptually not getting the point of S2P. All I've gleaned (perhaps incorrectly) is that it might take a phoneme and then show several ways to spell it at once. I'm pretty dubious about doing that to a novice reader, although I do think it's useful as a revision/discrimination practice once the various spelling options have been covered individually over time.
I appreciate this critique of the S2P persuasion thus far Ralph. It has been a challenge to try to do this.
If you scroll down this post, you’ll see 2 video examples of introducing kids who can now blend CVC short vowel words advancing to long vowels, via our Sort It activity:
They are being stretched but with the words coded to show the long “o” spellings and the teacher’s support, they are usually successful.
In the article I try to make the case that by starting with the sound and revealing the main spellings all at once, learners better understand how our code works and they are better able to organize the info in their mind...Building a schema better than the drip, drip, drip method of random phonics spellings over 2-3 years.
With this activity, reading and re-reading long “o” sound texts, and a couple other activities, most kids learn most spelling for long “o” in one week. Then we’re off to learn the long “e” sound the next week, etc.
I have suspected that the speed with which our students absorb the phonics info is in part due to the organizational display of the info…as well as by Writing and Saying the Sounds.
[One interpretation of the sock drawer comment is that this approach to phonics is organizing the phonics info in a way the child already understands—she already knows the sound “o” and oodles of words with that sound. It’s simpler to move from the know (i.e., the sound /oa/) to the unknown (various spelling of /oa/).]
Thanks Marnie. To be honest, an S2P approach appeals to my own sense of organization, but I'm still not totally convinced it's wise to throw so many variations at beginners all at once. But it's definitely a great option and one I'm keen to try.
I spent quite a while trying to understand how the sock drawer analogy related to S2P, but it was only when someone mentioned multiple drawers (like the filing cabinets in your link) that I got it. I just don't think the sock drawer is the right analogy. The idea of multiple drawers (each containing a phoneme with multiple graphical representations) makes much more sense.
Good to hear. Thanks for reading! We have heard from hundreds of teachers over 2 decades who doubt that developing readers can handle all the info at once, so I get it! Most phonics scope and sequences imply that the info must take years to absorb.
We also hear the formerly resistant return over and over again and say how shocked they were that their students found it pretty easy. Having said that, I also suspect that some preliminary activities we do to help the child connect sound and symbols easily pave the way for success with an activity like Sort It. (That is, I don’t think Sort It alone represents the essentials of a S2P approach, of course.)
"All I've gleaned (perhaps incorrectly) is that it might take a phoneme and then show several ways to spell it at once. I'm pretty dubious about doing that to a novice reader"
This practice isn't universal amongst speech-to-print advocates by any means, and it's not what I do. At the most basic level, I always introduce a spelling pattern through dictation of word chains (can, cane, pane, plane, plan, pan, etc.) in which words only shift by one phoneme at a time using the "hear it, say it, write it, read it, use it" sequence from Gentry and Ouellette's book Brain Words: How the Science of Reading Informs Teaching. In fact, I'm working on a piece called Speech-to-Print Speaks to Me, which goes into a lot more detail of how it differs from print-to-speech--but emphasizes how BOTH approaches can be successful as long as all word-work is "voiced" so that phonology, orthography, and meaning unite. Does that help?
Thanks Harriet. It does help a bit. Is S2P mainly about aiding spelling, or is it intended as a way to teach reading from day one? I get the sense it's both.
> I'm working on a piece called Speech-to-Print Speaks to Me
I would say both. I used the word-chain approach with my kindergartners and also promoted invented spelling so that they took their sound-letter knowledge and applied it to their writing, which research shows helps with phonemic awareness. I find S2P is generally a lot simpler than P2S and has built-in guardrails against overteaching (https://highfiveliteracy.com/2024/11/18/bursting-with-knowledge-are-we-overteaching-phonics/).
Here are Dr. Sam’s thoughts about the article The Multiverse of the Science of Reading and the recent webinar by Rachel Gabriel on speculative policymaking. BTW, be aware that Michael Pressley invented the term Balanced Literacy in reaction to the lack of phonics in some of the programs of that era. His son's latest book on that topic indicates that BL is research-based and still includes phonics instruction. https://doctorsam7.blog/2025/01/11/the-multiverse-of-the-science-of-reading-my-thoughts-about-this-important-article-from-the-nea/
Thanks for this and your previous posts. I'll get to each separately. Just one question: Is Dr. Sam you, the person who wrote this post, or is there someone else you're referring to as "Dr. Sam"?
You're absolutely right that Michael Pressley used the term. I'm not sure "invented" is quite accurate, but maybe first in the published literature? Close enough I guess. In any event, if people had stuck to "balanced literacy" to denote what Pressley meant and the National Reading Panel meant we'd probably be in a different place today, at least with respect to this word that--in my opinion--gets bandied about loosely and--if I'm being totally honest--recklessly. "Research-based" or "evidence-based" is/are in the same category.
Could you please clarify what aspect or definition of "BL is research-based"? As you say in the post you linked to (thanks for that, btw) "Many things were called BL even when they weren’t." So what precisely are you referring to as "BL" when you say "BL is research-based"?
Thanks again for writing, I really do appreciate it, particularly if you can help clarify what we are talking about.
Donna Scanlon has been mentioned several times in these comments. If you listen for 10 minutes at 17:30 in this recent interview (https://www.classroomcaffeine.com/guests/donna-scanlon), she echoes this statement by Dr. Sam: "When in-servicing teachers, I urge them to encourage students to use the graphophonemic information first and then cross-check it against context & syntax." She distinguishes between word identification and word confirmation and discusses the importance of "set for variability" (flexible pronunciations--"try a different sound") in a way that I can't see anyone disagreeing with. And yet . . . she refers to SOR being against using contextual cues, which illustrates part of the confusion that confounds reconciliation.
She also laments the bad associations attached to the term "balanced literacy" and offers the term "comprehensive literacy". I don't use "balanced literacy" or "structured literacy"; I call what I do "inclusive literacy". What's in a name?
All of which points to being clear about the meanings of our loaded, freighted, confounding words! Doesn’t it?
I did not avoid the question, I pointed out that your concerns fall into the category of research design. Scanlon's decades of research should not be discounted or ignored. I hope I'm clarifying my point that your views around the three cueing systems are not what many folks try to make it out to be. It was never meant to be a teaching system- and the attempts to turn it into are not what I teach my teachers to do. The best model I am aware of for using ALL the information system is the one employed in Reading Recovery, and again, what folks say is done in RR is not what RR folks are doing.
As you read more of my writing, I always seek evidence of common ground and common practices. So I'm VERY interested in Seidenberg and how his views fit Clay's views around creating a self-extending system. I'll also point out that laying the foundation should happen early on. What that foundation might be is also determined by the needs of the students. Some students thrive on analytic approaches (essentially discovery learning). Some students get the foundation rather quickly. How do you see SOR adapting to those facts (or do you disagree with that assessment)? I'm seeing examples of SOR folks treating analytic phonics as an inferior approach. It's not. What to use depends on the student (however I've said many times that it is crucial that synthetic phonics be part of any Tier 1 program).
Before you answer about your views around analytic phonics, please look over this excerpt from a July 18th blog entry, Shanahan posted a few years back. "The average effect size was somewhat higher for synthetic than analytic approaches, but not significantly so (it was so small a difference that one can't say one is really higher than the other). In other words, synthetic and analytic phonics are equally good."
I'm not seeing much evidence that SOR is treating things that way. What am I missing?
BTW, I'm glad to see some things on which we seem to agree. I'm glad to see that.
Dr. Sam
BTW thanks to Harriet Jametos for pointing out the Scanlon interview. I will work on my comment about your take on BL and the Three Cueing System.
That’s one of the many reasons we (ok, I) call her the CA Reading Sage.
See? One more thing we agree on. Inch by inch ….
Sam Bommarito liked your comment on Can we talk?.
That’s one of the many reasons we (ok, I) call her the CA Reading Sage
This is a bit off-topic (not really if the point of the conversation is to identify possible areas of agreement) but please listen to this interview by Bruce Howett another person who is trying to find a way to end the reading wars. If you don't know about him I think you'll find he is someone with ideas worth looking into. https://doctorsam7.blog/2023/09/09/dr-sam-bommarito-interviews-bruce-howlett-author-of-sparking-the-reading-shift-about-a-plan-to-end-the-reading-wars/
"please listen to this interview by Bruce Howett another person who is trying to find a way to end the reading wars."
On the one hand, I applaud Bruce for promoting multicomponent instruction by emphasizing David Share's new article (Share, D. (2025). Blueprint for a universal theory of learning to read: The Combinatorial Model. (Under Review). Academia. https://www.academia.edu/118772944/Blueprint_for_a_universal_theory_of_learning_to_read_The_Combinatorial_Model), which presents a multicomponent approach called The Combinatorial Model, along with promoting Maryanne Wolf's multicomponent work on POSSuM (phonology, orthography, semantics, syntax, morphology).
On the other hand, both Share and Wolf acknowledge the importance of decoding instruction for the novice reader before complicating that instruction with other components, but Bruce seems to be more inclined to accept the methods of Structured Word Inquiry, which teach phonology, morphology and etymology simultaneously to beginning readers, an unhelpful complication I wrote about in Must Phonics Fail in Order for Structured Word Inquiry to Succeed? (https://learningbydesign.com/professional-development/spell-links-blogue/).
This is another good example of why we need to define our terms and see where in the instructional sequence they fit in (word identification before word confirmation; simplified decoding instruction before complicated morphology instruction, etc.).
“Can you be more precise about: What are David's and my fundamental differences? " The definitions of reading you have provided so far are so broad and general it is impossible to tell what your views are about the issue of your take on the simple view of reading. The simple view focuses on comprehension (language comprehension). It fails to account for factors like cognitive processes (Pearson's work on the was groundbreaking). Some comp strategies can and should be taught. Background knowledge is not sufficient. Also, things like social cultural context. Pearson has taken the time to post his response to your response on his site. BTW David had 10 pushbacks, not just one. So, let's start with this- reading is fundamentally a meaning-making process. Do you agree or disagree, and why? Presenting the question about Balanced Literacy in a separate comment.
“ Pearson has taken the time to post his response to your response on his site.”
I posted his (and Rob’s) reply on my substack, followed by my replies. 3 parts I think. Maybe 4. Whatever, pls check them out, then we can resume. I don’t have my computer so can’t easily navigate b/f, but lmk if you can’t locate.
Not sure how to get to that- I'm new to substack- all I can see so far is this particular thread. I need to have you answer the "Can You Be more precise" question above.I'm sensing that the most likely source of your differences lies in the limits and limitations of the simple view of reading. In a nutshell, listening comprehension plus decoding is not a satisfactory model, it is an incomplete model. The simple view is just a little too simple!
Pls go here https://claudegoldenberg.substack.com/ and see my replies to their response.
Perhaps a lack of response is a function of the “sides-taking” in the reading wars. People who take issue with your views, but sniff the product and leave the store. I don’t know. I do know your mission is worthy and I hope you persist.
“a lack of response is a function of the “sides-taking.” Indeed. And there are many possible explanations. One is what you suggest.
Another, for which I have anecdotal evidence, is that members of a tribe (=side) who speak out against positions from the tribe, even if they know they’re ill-informed and unhelpful, risk marginalization or excommunication. The silencing, either self- or other-imposed, is very harmful regardless which tribe does it.
On the other hand, tribe solidarity increases the likelihood of tribe survival. So there’s that. I guess the question is when does tribe solidarity (probably a good thing, at least for the tribe) turn into tribal group-think (probably not)?
Psychologists might suggest a post challenging those who agree with you on some issue likely to draw a reaction. But psychologists are I know from personal experience a curious group, so you might not want to take a suggestion from a card carrying member of that tribe.
This is the post from Shanahan I referenced earlier. BTW I was the teacher asking the question. https://www.shanahanonliteracy.com/blog/which-is-best-analytic-or-synthetic-phonics
From Tim Shanahan:
"Based on my personal teaching experiences, I’d begin with synthetic phonics (or would insert synthetic supports into an analytic program if I were required to teach that). It’s just easier for kids."
I agree--based on my personal experiences. I would also say that it's much easier for teachers. Is there an analytic phonics program you recommend?
From Reading Development & Teaching, Stuart and Stainthorp, 2016:
"We have found only two adequately controlled studies published since the National Reading Panel report which have succeeded in directly comparing the relative efficacy of synthetic small-unit phonics and onset-rime analogy large-unit phonics in promoting reading growth. . . Both conclude that teaching small units is more effective than teaching large units.
Christenson & Bowey (2005). The efficacy of orthographic rime, grapheme-phoneme correspondence, and implicit phonics approaches to teaching decoding skills. Scientific Studies of Reading, 9, 327-349
Hatcher, Hulme & Snowling (2004). Explicit phoneme training combined with phonic reading instruction helps young children at risk of reading failure. Journal of Child Psychology and Psychiatry, 45, 338-358.
Further support for the decision to teach grapheme-phoneme correspondences rather than onsets and rimes comes from analysis of the relative numbers of each type of correspondence required for children to be able to read the words they typically encounter on their reading materials, which we reported in Chapter 4: children need to know almost five times as many onset and rime correspondences as grapheme-phoneme correspondences to read all monosyllabic words in the early version of the Children's Printed Word Database (Stuart, Dixon, Masterson & Gray, 2003). It is simply more efficient to teach children grapheme-phoneme correspondences."
Deal- and if you check I have been answering you questions. We have more than one thread so let me know what I've missed. The answer to this particular question of mind is very important..
Sam
Sorry if I wasn’t clear. These are the questions you’ve missed:
“Can you be more precise about: What are David's and my fundamental differences? and which of my "views around Balanced Literacy [are] misleading and incorrect"? and also, just to be complete, which of my "criticisms about BL have merit"?”
Can you be more precise about: What are David's and my fundamental differences? and which of my "views around Balanced Literacy [are] misleading and incorrect"? and also, just to be complete, which of my "criticisms about BL have merit"?
What exactly to you see as language comprehension? Does it go beyond listening comprehension? If so, how? Explain specifically how things like background knowledge, motivation, and cognitive process fit in. What are your views on how sociocultural context impacts reading ability? What are your views on the nuances of reading strategies and whether reading strategies can and should be taught directly?
Tell you what... you answer my questions and I'll answer yours. Deal?
This is in response to your comment about Scanlon.
First- the research design used by Scanlon was published in RRQ, so they found it acceptable. You and Shanahan have the right (obligation) to raise questions about designs; the readers have the right to decide if the issues raised merit completely discounting all she had to say.
Second- there is much more to Scanlon's research than simply one paper. She has a new textbook out and a variety of publications. I think folks reviewing that large body of research done over a couple of decades will conclude that context needs to be used in word recognition. Current laws are removing it entirely. That is a HUGE mistake.
Third- while some BL folks might call for all cues (information) to be used co-equally, as you described, others see such as myself see context as playing a CRITICAL role (not a secondary role). How can one encourage a set for variability in their students if they fail to crosscheck the graphophonemic information with contextual information ("read"- should it be pronounced red or reed)? Frankly, the words cue/cueing have become divisive. I tend to use the words information/information systems. When in-servicing teachers, I urge them to encourage students to use the graphophonemic information first and then cross-check it against context & syntax. Clay NEVER said to only (or mainly) use meaning and context. She specifically encouraged using ALL information and cross-checking that information. To create a self-extending system. A topic for another conversation has to do with Seidenbergs recent ideas about 1. direct phonics instruction early 2. a "take off" point where the child begins teaching themselves. That sounds very much like Clay's self-extending system. Your thoughts?
BTW- I've mentioned before that
You're avoiding the question of what you make of a comment Tim S made about Scanlon's work (".... misleading.... She draws conclusions that the original researchers wisely did not draw… Context definitely plays a role in decoding and decoding development, but it is not the role 3 cueing advocates cling to"), given that you've endorsed Tim S's nuanced view. I guess I wasn't clear it was a question, since I just said "I'm wondering." So apologies for the miscommunication, but I'm still wondering (honestly).
"context needs to be used in word recognition. Current laws are removing it entirely. That is a HUGE mistake." Completely agree. (Tim does too, btw).
"Frankly, the words cue/cueing have become divisive." Agree again. 100%, or more, if it's possible to agree more than 100%. That's the point of the DEADLY WORDS posts I've been working on. Whether the correct adjective to put in front of "context" cues/information is CRITICAL or secondary is surely a matter for discussion, although not sure how critical that discussion is. Maybe it's more critical than I realize.
In any case, I was very glad to read this: "When in-servicing teachers, I urge them to encourage students to use the graphophonemic information first and then cross-check it against context & syntax." Once again, we are in agreement, although personally, I'd prefer a somewhat stronger infinitive than "to encourage", eg "to teach."
Then whether we call it "a "take off" point" or a "self-extending system" is less important than providing the sort of instruction that will hasten students' getting there. What you provide in your in-services sounds like it will help accomplish that.
"When in-servicing teachers, I urge them to encourage students to use the graphophonemic information first and then cross-check it against context & syntax."
I agree and believe the order matters. Here's how I end Goldilocks and the Three Cues (https://highfiveliteracy.com/2024/12/23/goldilocks-and-the-three-cues-or-will-the-seven-blind-mice-ever-agree-on-the-parts-of-the-elephant/):
"We need to stop treating reading like a Rorschach test. Like the hare, the untrained reader is enticed by the three-cueing system to search for distractions on the page, seeking answers off the beaten path of best practices that simply guides them to apply their decoding skills to attempt an identification of a word, to be flexible with pronunciations, and to confirm their choice by checking the context of the sentence related to syntax and meaning. Here’s my three-cueing system: grapple with the graphemes, flex the phonemes, confirm with context."
I did not see this original reply. Let's take you up on your offer to explain YOUR definition of reading. Do you subscribe to the simple view of reading.? You identify two components. word recognition and language comprehension. What exactly to you see as language comprehension? Does it go beyond listening comprehension? If so, how? Explain specifically how things like background knowledge, motivation, and cognitive process fit in. What are your views on how sociocultural context impacts reading ability? What are you views on the nuances of reading strategies and whether reading strategies can and should be taught directly? Thanks for you input about those things.
I don't recall my offering to explain my definition of reading, but regardless, I don't have a single unitary definition for the simple (or complex) reason that, to quote myself, "There is no universally accepted answer to 'What counts as reading?'"
I'm afraid you're evading Harriett's request for specificity and my followup questions about your claims: What are David's and my fundamental differences? and which of my "views around Balanced Literacy [are] misleading and incorrect"? (you can skip my additional "just to be complete" question, which of my "criticisms about BL have merit"?)
Hi Dr. Goldenberg! I just wanted to say thank you for the work you're doing to share your perspective and educate us non-researchers on the research. I'm coming at this from the business side of things, having worked in the K12 market for a decade. My hope is that the folks making purchasing decisions in schools are doing so with eyes wide open—not falling for a clever curriculum provider slapping "SoR" on every product.
Leah, thank you so much. Your hope is mine as well. My question is how can we make that happen... not falling for a clever curriculum provider slapping "SoR" on every product. This is not a rhetorical question. It's an urgent one! Seems to me that everyone who cares about reading has a role here. Educators (of all stripes... teachers, paras, administrators, leaders of teacher unions) need to be better informed and so do publishers. And so do policy-makers, elected and not elected. And advocates for different populations of kids. And--dare I say it?--academics. And bloggers too. And parents!
Each has a responsibility in their respective domains to check out claims and call out false ones. I'm not looking for villains or casting blame. But the fact is--and I do believe it's a fact--that publishers and many others, knowingly or cynically, can't say for sure--play fast and loose with the terminology. SOR is exhibit A. (Did you see Emina McLean's blog post link I included in a recent post on this Substack?) The other terms in the DEADLY WORDS posts I've written are also in play. So how do we deal with that. I would 100% welcome your and other members of the publishing and business side of things engagement with this. I can't thank you enough for this comment and your interest. Keep commenting here or feel free to email me directly. Thank you.
A good first step would be for you to explain your definition of reading. Specifically, does it include the term meaning-making? Specifically, does it view decoding first, meaning-making next or decoding and meaning-making as concurrent? Thanks for the specifics on your views about this.
I wonder if you saw my reply. I might have replied in the wrong place, so here goes again.
I addressed this question in part IV of my reply to Tierney and Pearson's response to my critique:
What counts as reading.
Rob and David say in their reply that “As near as we can tell, Claude accepts the intentionally narrow definition of reading, put forward by Rayner and his colleagues in the classic 2000 paper.”
I don’t know on what basis they drew that conclusion. They could have gotten nearer, and more accurate, by just asking me.
Or, as easily, they could have gone to Claim 3 in Table 2 of my critique. To replace the claim they claimed to fact-check (“Reading is the ability to identify and understand words that are part of one’s oral language repertoire”), I proposed this fact-checked improvement:
"There is no universally accepted answer to “What counts as reading?” It can be defined in many different ways reflecting different perspectives. Where there is agreement is that skilled reading comprises two important principal components—word recognition and language comprehension—and within those, many skills, understandings, and attributes are involved."
Reading can be used to refer to many things, from the most basic and limited—decoding written words (whether understood or not)—to the metaphorical—for example, reading the room, just as “literate” can refer to someone who can read and write at a basic level or to someone well-versed in some literary canon.
The meaning of reading and literacy have also shifted, as they went from being exclusive domains of an elite and tools for maintaining control over others to now oft-stated aspiration for “universal literacy.”
Dear Claudio,
Thank you for this thoughtful message and for continuing the series. I rely on your posts to keep me informed and to help shape my arguments when engaging with those who hold opposing views. While I don’t often leave comments, as I’m usually in notetaking mode, I deeply value the insights you share.
I encourage you to continue this series, as it’s essential to advancing meaningful conversations in literacy. I also agree with your suggestion to spread the message more broadly and will forward this to others, including those on "the other side," to help bridge the gaps.
Best regards,
José
Gracias José. I will certainly continue it once I get my &$?#%* computer back. 😩
I thought that, after 5000 years of reading, we might have figured out how to teach it by now. But I'm wondering if I should just check back in another 5000 years to see if the wars are over yet.
Anyone who starts to investigate how to teach reading today is most likely to get hit with a barrage ("bar rage", SWI people?) of information about a magic new thing called SoR and that Phonics is IN and Whole Language is OUT etc.
If they dig a bit deeper, though, they’ll find that there are still raging debates about how best to approach literacy instruction.
SoR: "The science is in! Phonics + lots of other stuff is the Way".
SWI: "Nope, phonics is a failed attempt to explain a writing system that involves phonology, morphology and etymology."
S2P: "Nope, stop messing up kids' sock drawer." (Huh?)
There have been discussions/arguments for years over the "evidence" for various approaches, but weirdly, despite the daily literacy studies that keep rolling in, no one seems to be interested, game or willing to put these approaches to the test side by side.
Maybe it's pointless anyway. Project Follow Through pitted various teaching approaches against each other over nine years and produced startlingly clear results. Unfortunately, they weren't the results most people wanted, so they were just ignored.
Haha. Humans, eh?! I feel like just going back to grunting.
Ralph, I appreciate your thoughtful comments. Going back to grunting sounds very appealing. I do it ocassionally. I'll have to try it more often.
Re this: "Project Follow Through pitted various teaching approaches against each other over nine years and produced startlingly clear results." Unfortunately (but we have to deal with it bc reality is really a mess) there were at least 2 "startlingly clear results." One was I believe what you probably mean, that direct instruction (capital D capital I) got the best results on student outcome data.
The other was that the variability across sites was greater than the overall "main effect" of DI. What this meant, not to belabor the point, was that local circumstances (which could mean all sorts of things including the quality of the teaching, strength of school leadership, etc etc) were much more important determinants of student outcomes than the DI program itself.
This finding echoed Chall's from years before (in a much less rigorous statistical analysis, but, as it turned out, no less valid) that programs/teachers--I don't recall the unit(s) of analysis--that featured decoding produced better results, in general, than programs that did not, BUT that there was a great deal of variability by teacher or program or something... sorry to be vague, but it was the same larger point.
So the moral of this story, at least as I understand it, is that results can/will be messy and point in different directions and support different interpretations. Mine (and, to my knowledge incorporates all the reading research with which I am familiar, which does not include every single thing ever published) is this: In general, phonics (or whatever you call explicit and systematic instruction in foundational literacy skills) makes a net positive impact on student reading development and should be hardwired into every reading/literacy program, classroom, school, district, etc. But by no means is explicit and systematic instruction in foundational literacy skills the only thing that matters. For starters, students will need wildly different amounts and intensity of such instruction, from relatively very little to A WHOLE LOT. And then even those who need A WHOLE LOT, also need other things, most notably but not limited to language development (vocab and more), knowledge building; and all will need excellent instruction, competent school-wide leadership. So that's my takeaway, for better or worse.
Thoughts?
Thanks Claude. It's absolutely true that literacy is a whole lot more than just phonics.
I learned about Project Follow Through back in the 90s, and whatever its possible failings, it does seem from every angle that DI had a huge effect in all places compared with its control groups — unlike the other methods, which sometimes performed way below the control groups, despite having a lot of funding. Even if the results aren’t perfect, it doesn't explain why the results were largely ignored. (My primary schooling took place after results showed the benefits of DI and the failings of Whole Language, and yet all I got was Whole Language.) There was clearly a lot of politics involved. It's interesting that interest in Follow Through is starting to appear again. I know you've written about it, and in the last year (even in recent days) it's been the topic of podcasts, such as Zach Groshell's (https://www.youtube.com/watch?v=vrly4mmJZcA) and Anna Stokke's (https://chalkandtalkpodcast.podbean.com/e/project-follow-through-direct-instructions-overlooked-success-with-marcy-stein/). It's pleasing that there's more interest in best practices these days, although our human nature makes it hard to have our ideas challenged. I confess that, as a fan of explicit synthetic phonics, I've been a bit rocked by the Structured Word Inquiry (SWI) approach which clearly shows the failings of phonics. But it obviously has huge merit, so one has to be willing to look again and adapt where necessary. I'm actually glad that SWI has merit, as I'm a bit or a word nerd anyway.
Ralph, my computer tanked so I’ll reply more completely when I get it back. My main point was that the lessons of FT, just as the lessons of Reading First, are not as unambiguous as you might think. More later. And thanks for your engagement.
hi again, Ralph. Here's something I wrote for an IRA reading research conference 20 yrs ago (😵💫). If you're interested in the entire talk (mostly about ELs or EBs or MLs... back then it was ELLs), I'm glad to send you my speaking notes, the ppt, and a pdf of the 1978 article I quote from below.
(Fyi, the title of my talk was "Ideology, Research, and Practice." I used the FT experience as an example of how "ideology" influences, even if not fully determines, the conclusions one will reach about complicated findings that have have more than one plausible interpretation (which is to say, just about all research findings). One could also reasonably argue that "ideology" will actually fully determine the conclusions, even when results are straightforward by, e.g., finding fault with methodology, measures, populations, whether measurement isn't even involved, and on and on.)
"Some commentators have argued that FT was spectacularly inconclusive, that no program consistently outpeformed the control groups, and that in general none of the program models was successful in reliably improving children’s chances of school success. But others have concluded exactly the opposite—that overall the DI model got the best results on a number of outcome measures; not at every site, but if you average the effects across the sites, DI obtained superior results. (modestly superior, but superior)
"So what really happened? It turns out that both sets of commentators are right—but they’re just not right enough. Each has emphasized only one part of the story. Let me quote from a 1978 article written by the authors of the original evaluation report, trying to deal with and clarify the commentary that had been heaped on their report in the mere year since its release: “Models that emphasize the kinds of skills tested by certain subtests of the Metropolitan Achvt test have tended—very irregularly—to produce groups that score better on those subtests than do groups served by models that emphasize those skills to a lesser degree.” In other words, #1: overall the children who received instruction that emphasized skills measured by the standardized test used (and these children were in DI or DI-type programs) outperformed children who received a different type of instruction. And #2: this pattern varied considerably and was not consistently found across the sites."
Thanks Claude! Yes, it would be interesting to see those notes and things you mentioned if they're easy enough to pass along. Thanks for the offer. I don't have a stake in this, but it is interesting. To be honest, hearing such widely differing views on research does make me question research in general. Is there really much useful research going on anywhere, or are people just generally confirming their own biases? It's a bit depressing, really. In the end, I find myself just having to use my own judgement and experience to make decisions about how and what to teach, making the best of the information I can find. None of my grandma's best recipes are backed by peer-reviewed science, but they sure are good to eat.
I understand your reaction, which is to "question research in general," and the other musings you share. I don't think those are good solutions. Maybe bc I have an irrational faith in rational processes, I believe we're better off doing and paying attention to research than not doing or ignoring it. It's obviously true in the physical and biological sciences (unless you're RFK Jr, various climate deniers, and many members of a political party that shall go unnamed). It's less obviously true in the social sciences, which sometimes seem to deny its own scientific standing--inadvertently or vertently (if that's a word).
A much better solution imho is a better understanding of research by educators at all levels, pre-school to PhD school, + those making policies that affect educators and the to-be- and are-being-educated. This understanding includes findings from research (gnarly as they might be), in addition to something about how research is done, how provisional and probabilistic those findings are, and that skepticism is a part of research and of "science" more broadly. People equate science with certainty. That's a big mistake. See this piece if interested: https://www.dropbox.com/scl/fi/pdx244x7y3s992cy4dh22/EdWk2021Science-and-skepticism-no-pswd.pdf?rlkey=apxbgmmylcvwtqzr52rt0fqjb&dl=0.
As for the ones I mentioned in my last reply, see these; let me know if you have trouble accessing. Feel free to email me.
And thanks again for engaging!
• https://www.dropbox.com/scl/fi/ztkj780ruiza3aj72hooh/KEY-Impr-lit-outcomesPRINT.doc?rlkey=rdl40b27wbl6f1zxx48egueqb&dl=0
• https://www.dropbox.com/scl/fi/594nn1znb05fecl8y53lj/IRA-RRCGoldenberg.ppt?rlkey=gdxh1q7246gq8w88cadgigikw&dl=0 (The powerpoint I used. I think it's still basically correct but outdated. If interested in what my take today is on ELs' learning to read in English, pls see a previous post on this substack, "Research must guide how we teach English learners to read.")
• https://www.dropbox.com/scl/fi/wqisren3wd1g7nqv6xrta/Anderson-St.Pierre-etal-1978.pdf?rlkey=lwxp5732wuc8nr1am6iyhoyr3&dl=0 (The article I quoted from by the authors of the FT report)
Thanks very much, Claude. Interesting reading. I kind of have to throw my hands up in regard to assessments of Follow Through. The various opinions I hear about it vary so widely that it's kind of pointless to try to draw a conclusion from it (which is perhaps that whole point!) It's like trying to get a handle on the Kennedy assassination by pitting the Warren Commission report against an Oliver Stone documentary. It's basically a "choose your own adventure" scenario. In my own experience, a more explicit, DI kind of approach has been far more fruitful than other teaching methods, and at the end of the day, it's probably reasonable for teachers to learn from their own experience and that of their peers — much like cooks and tradespeople learn from their mentors and peers. Researchers seem to be in a world of their own a lot of the time.
"If they dig a bit deeper, though, they’ll find that there are still raging debates about how best to approach literacy instruction.
SoR: "The science is in! Phonics + lots of other stuff is the Way".
SWI: "Nope, phonics is a failed attempt to explain a writing system that involves phonology, morphology and etymology."
S2P: "Nope, stop messing up kids' sock drawer." (Huh?)"
This is great! You might appreciate Must Phonics Fail in Order for Structured Word Inquiry to Succeed (https://learningbydesign.com/professional-development/spell-links-blogue/) which has everything (including the sock drawer) but the kitchen sink.
Thanks Harriet. I actually read that a few days ago. Great work! I must admit, though, I'm still befuddled by the sock drawer analogy. I get the theory, but can't see the point of it in practice. Do you know of a practical example that shows how an S2P lesson would work? I haven't been able to find any, although I've read quite a few articles and interviews with Jan. Perhaps it's too much of a proprietary system and I would have to pay to find out, but I'm just conceptually not getting the point of S2P. All I've gleaned (perhaps incorrectly) is that it might take a phoneme and then show several ways to spell it at once. I'm pretty dubious about doing that to a novice reader, although I do think it's useful as a revision/discrimination practice once the various spelling options have been covered individually over time.
I appreciate this critique of the S2P persuasion thus far Ralph. It has been a challenge to try to do this.
If you scroll down this post, you’ll see 2 video examples of introducing kids who can now blend CVC short vowel words advancing to long vowels, via our Sort It activity:
https://readingsimplified.com/advanced-phonics-sort-it-activity
They are being stretched but with the words coded to show the long “o” spellings and the teacher’s support, they are usually successful.
In the article I try to make the case that by starting with the sound and revealing the main spellings all at once, learners better understand how our code works and they are better able to organize the info in their mind...Building a schema better than the drip, drip, drip method of random phonics spellings over 2-3 years.
With this activity, reading and re-reading long “o” sound texts, and a couple other activities, most kids learn most spelling for long “o” in one week. Then we’re off to learn the long “e” sound the next week, etc.
I have suspected that the speed with which our students absorb the phonics info is in part due to the organizational display of the info…as well as by Writing and Saying the Sounds.
[One interpretation of the sock drawer comment is that this approach to phonics is organizing the phonics info in a way the child already understands—she already knows the sound “o” and oodles of words with that sound. It’s simpler to move from the know (i.e., the sound /oa/) to the unknown (various spelling of /oa/).]
Thanks Marnie. To be honest, an S2P approach appeals to my own sense of organization, but I'm still not totally convinced it's wise to throw so many variations at beginners all at once. But it's definitely a great option and one I'm keen to try.
I spent quite a while trying to understand how the sock drawer analogy related to S2P, but it was only when someone mentioned multiple drawers (like the filing cabinets in your link) that I got it. I just don't think the sock drawer is the right analogy. The idea of multiple drawers (each containing a phoneme with multiple graphical representations) makes much more sense.
Good to hear. Thanks for reading! We have heard from hundreds of teachers over 2 decades who doubt that developing readers can handle all the info at once, so I get it! Most phonics scope and sequences imply that the info must take years to absorb.
We also hear the formerly resistant return over and over again and say how shocked they were that their students found it pretty easy. Having said that, I also suspect that some preliminary activities we do to help the child connect sound and symbols easily pave the way for success with an activity like Sort It. (That is, I don’t think Sort It alone represents the essentials of a S2P approach, of course.)
"All I've gleaned (perhaps incorrectly) is that it might take a phoneme and then show several ways to spell it at once. I'm pretty dubious about doing that to a novice reader"
This practice isn't universal amongst speech-to-print advocates by any means, and it's not what I do. At the most basic level, I always introduce a spelling pattern through dictation of word chains (can, cane, pane, plane, plan, pan, etc.) in which words only shift by one phoneme at a time using the "hear it, say it, write it, read it, use it" sequence from Gentry and Ouellette's book Brain Words: How the Science of Reading Informs Teaching. In fact, I'm working on a piece called Speech-to-Print Speaks to Me, which goes into a lot more detail of how it differs from print-to-speech--but emphasizes how BOTH approaches can be successful as long as all word-work is "voiced" so that phonology, orthography, and meaning unite. Does that help?
Thanks Harriet. It does help a bit. Is S2P mainly about aiding spelling, or is it intended as a way to teach reading from day one? I get the sense it's both.
> I'm working on a piece called Speech-to-Print Speaks to Me
That sounds great. I eagerly await it. :-)
I would say both. I used the word-chain approach with my kindergartners and also promoted invented spelling so that they took their sound-letter knowledge and applied it to their writing, which research shows helps with phonemic awareness. I find S2P is generally a lot simpler than P2S and has built-in guardrails against overteaching (https://highfiveliteracy.com/2024/11/18/bursting-with-knowledge-are-we-overteaching-phonics/).
Here are Dr. Sam’s thoughts about the article The Multiverse of the Science of Reading and the recent webinar by Rachel Gabriel on speculative policymaking. BTW, be aware that Michael Pressley invented the term Balanced Literacy in reaction to the lack of phonics in some of the programs of that era. His son's latest book on that topic indicates that BL is research-based and still includes phonics instruction. https://doctorsam7.blog/2025/01/11/the-multiverse-of-the-science-of-reading-my-thoughts-about-this-important-article-from-the-nea/
Thanks for this and your previous posts. I'll get to each separately. Just one question: Is Dr. Sam you, the person who wrote this post, or is there someone else you're referring to as "Dr. Sam"?
You're absolutely right that Michael Pressley used the term. I'm not sure "invented" is quite accurate, but maybe first in the published literature? Close enough I guess. In any event, if people had stuck to "balanced literacy" to denote what Pressley meant and the National Reading Panel meant we'd probably be in a different place today, at least with respect to this word that--in my opinion--gets bandied about loosely and--if I'm being totally honest--recklessly. "Research-based" or "evidence-based" is/are in the same category.
Could you please clarify what aspect or definition of "BL is research-based"? As you say in the post you linked to (thanks for that, btw) "Many things were called BL even when they weren’t." So what precisely are you referring to as "BL" when you say "BL is research-based"?
Thanks again for writing, I really do appreciate it, particularly if you can help clarify what we are talking about.
I'm still going to post the blog I talked about in the morning- but I thought you might like to look over this blog I wrote about Shanahan. By the way, he gave me a like on this posting and reposted it. https://doctorsam7.blog/2024/12/07/a-centrist-perspective-on-what-tim-shanahan-had-to-say-in-his-latest-blog-post-analyzing-sor-2-0-by-dr-sam-bommarito/
Yes, very nice. Thanks for posting this one as well.