Sound It Out

“What’s the use of a book,” thought Alice,
“without pictures or conversations?”
—Lewis Carroll

Ignorance, as we know, can be dangerous.  But what can be even more dangerous is not knowing what we do not know, and I recently found myself among those suffering from this condition when it comes to one of education’s most fundamental, foundational skills:  the ability to read.  Because I follow almost exclusively the science, I had thought that the battle and debate over “phonics” vs. “whole language” in elementary schools had been resolved long ago and that phonics had firmly won due to the data overwhelming supporting it.  Indeed, if you had asked me a month ago how reading is taught in the early grades in this country, I would have said “phonics” and not given it another thought.

Turns out not only would I have been wrong; I would have been badly wrong.  However, thanks to a good friend and colleague of mine who put me on to the reporting of Emily Hanford of Minnesota Public Radio, I now know not only what I did not know; I know what I should know (and so should every other educator at every level in this country), and it is scary stuff.

Our tale begins nearly 60 years ago with a teacher and educational researcher from New Zealand, Marie Clay (eventually Dame Clay for her efforts) who developed a method for teaching reading that became known as “cueing.”  It was grounded in the psychology of the day that just as all children naturally learn how to speak, developing their vocabulary without any direct instruction or intervention, so too must children naturally learn how to read, and therefore, the best way for children to learn to read was not to master how to decipher the actual individual words but to learn their meaning through the context of a story and its images. 

It was a hugely popular idea at the time, and cueing would become the educational “darling” of its day. It would do so because it was so much easier to employ in the classroom than the traditional “sound-it-out” phonics approach and also because so many children seemed to learn successfully when using this technique.  Eventually, cueing would make the leap across the Pacific to this country where it would become known as the “whole language” approach to reading, and for much of the 1980s and 1990s, the debate over “phonics” vs. “whole language” in elementary school circles was intense—in no small part because it got enmeshed with the arguments over how best to improve reading scores in the underperforming schools in our urban and rural areas, where proponents insisted that the “whole language” approach was improving the educational outcomes of poor children and children of color in particular.

Meanwhile, the scientific research about both reading and the brain began to catch up with what was going on in the classroom, and the results were unequivocal:  phonics was not only superior to cueing at teaching children how to read; cueing was, in fact, how children who struggle with learning to read actually compensate for their weakness.  Thus, teachers who were employing “whole language” were really teaching reading badly. Eventually, (as I discuss in more detail in A Brain Research Update), the science would reveal that learning to read actually involves “high-jacking” our natural facial recognition neurons in the brain and that requires the intense, direct and repetitive instruction of phonics to sound out each word as we read it—something that those of us who are successful readers do without any self-awareness that we are doing it.  Hence, put bluntly, the act of reading is not something our brains do naturally at all, and it requires significant interventive training to learn how to do so well.

Yet, as I learned in Sold a Story, all this research which I assumed was being followed has, in reality, been openly fought and denied for over two decades now by a whole host of stakeholders in our schools, and curricula based on cueing remain the dominant ones on the market for school districts to purchase and use.  It took a pandemic, with zoom school for kindergartners and first & second graders, for parents stuck at home themselves—most of whom like myself would have been trained with phonics—to see what was taking place firsthand and for a critical mass of them to begin to question what was happening—or more often failing to happen—to their children.

So how did we get into this mess? How did the scientific research about reading become so contentious and why have we allowed so many children to develop poor reading skills over the recent decades?

To address these questions, I think we need to start by acknowledging that much of the drive behind cueing and “whole language” was well-intentioned.  Marie Clay’s original research, for example, was about how to help children who found reading challenging to be more successful at it, and when she noted how so many of these children were using context to try to learn the meaning of words, it didn’t occur to her that she could be observing what the research now knows is a poor compensation mechanism for dealing with a weakness.  Instead, she saw it as a strength and made the flawed causal leap that if this is how poor readers become successful, then how much more will this method help all children learn to read.  Likewise, many elementary teachers wanted their children to love books and to make meaning from their stories as soon as possible, and “whole language” seemed to do away with the tedium of learning to read which they feared might put some kids off books while appearing to let students jump right away into what makes discovering new written materials fun and engaging.

However, lest we be too generous, I think we also must acknowledge that there have been and continues to be some powerful, vested interests at work.  “Whole language” curricula have been a $1.6 billion dollar business in this country since the early 2000s, and one of its leading proponents has made so much wealth off her program that she owns and drives a Maserati.  In addition, a phonics-based curriculum is significantly more time-demanding for teachers, and so schools confront either the need for additional educators/aids for the individuated instruction required to teach in this way, or they need to employ a more rote, drill-based pedagogy in large classrooms full of dozens of kids—neither of which is appealing in a time of budget cuts, limited resources, and the desire for more student-centered learning.  And that’s especially true given that a significant majority of children do learn to read in spite of the use of “whole language” in their training.

But that points to two other issues in this mess.  First, it suggests an implicit bias on the part of both individual teachers and schools, and that is that certain kids are just hard-wired to struggle to read; that they just don’t have what it takes to be truly successful readers.  This fixed-mindset bias has then been reinforced by the fact that the percentage of children who seem unable to master at least some degree of proficiency in reading has remained relatively low over the past few decades, and the unspoken “elephant” is that no system can be 100% successful. 

Except—and this is the second significant issue—the research shows that the same percentage of white children are just as likely as children of color or impoverished children to struggle with learning to read using “whole language.”  The difference is that many of the struggling white children have access through their privilege to tutors and other compensatory mechanisms to overcome their reading skill deficits.  Hence, how we teach reading in our schools isn’t just about securing access to education’s most essential skill; it is about securing access to an equitable education, period.

What’s more, if this almost Kuhnian battle of paradigms about how best to teach reading were not already problematic enough, the data now coming in about the impact of the pandemic on this essentail skill are positively grim.  Nearly 12 million children in this country now exhibit significant deficits in their ability to read—across the entire socio-economic spectrum—and because all the educational research shows that any reading deficits still present at grade 3 have a direct correlation with a student’s remaining educational trajectory to adulthood, we are looking at a significant subset of an entire generation being potentially unable to fulfill critical roles in our future economy and society at large.  And before my skeptics “pooh-pooh” the impact of only 12 million in a country of nearly 400 million, I remind readers of how disruptive the loss of “just” over a million dead from the pandemic has already been in the past two years on the current job market, inflation, etc.

The bottom line is that we have a reading crisis in this country with significant implications for our entire future as a society, and as an educator, it hits home at the very root of my mission.  I need to be prepared for the students who will arrive on my high school doorstep one day, and I need to spread the word about the phonics-cueing battle still taking place to enable any of my readers to become more effective advocates for better reading curricula in their elementary schools and districts.  Hence, I am grateful for the demise of my ignorance.

However, my new awareness of the reading crisis has joined yet another recently dispelled ignorance about an AI software known as GPT-3, and it has left me even more concerned about the root of all learning: basic literacy.  Many variants of this software now exist, including a publicly available free-ware version called ChatGPT, but essentially all of them have the same basic ability to generate coherent, rational—and unique!—written responses to nearly any query put to these softwares (including one GPT-3 authoring a scholarly article about itself currently awaiting peer-review for publication in a scientific journal).  I suggest reading Herman’s article in The Atlantic if you would like to see just how good a job these AIs do at generating original text, but the new reality is that meaningful authorship can now be generated by a computer—potentially leaving all of us never having to learn how to write ever again.

What, though, will it mean to teach in a world where no one has to write and where they might not even be able to read any words in the first place? It’s enough to keep this educator up at night, but I’ll tackle why I think all is not quite lost next time.

References

Barshay, J. (Dec. 12, 2022) Proof Points: Third Graders Struggling the Most to Recover in Reading After the Pandemic.  The Hechinger Report.  https://hechingerreport.org/proof-points-third-graders-struggling-the-most-to-recover-in-reading-after-the-pandemic/.

Hanford, E. (2022) Sold a Story: How Teaching Kids to Read Went So Wrong.  Minnesota Public Radio.  https://features.apmreports.org/sold-a-story/.

Herman, D. (Dec. 9, 2022) The End of High School English.  The Atlantic.  https://www.theatlantic.com/technology/archive/2022/12/openai-chatgpt-writing-high-school-english-essay/672412/.

Thunstrӧm, A, O. (September, 2022) AI Writes About Itself.  Scientific American.  Pp. 70-73.