Let me start here.

A while back, a colleague told me the following anecdote: a college classmate of her partner was visiting from San Francisco, and the three of them had dinner together. At some point, the conversation turned toward the classmate’s middle school-aged son, who had an interest in STEM, and the classmate said something along the lines of, “Isn’t it great that schools count computer science as a language now?”

“Wait,” my colleague replied. “Languages have four main components: reading, writing, listening, and speaking. CS has the reading and writing aspects but not the other two pieces. It’s not really a language in that sense.”

The classmate was taken aback. No one in the Bay Area had ever challenged him on that point—it was just taken as an article of faith that learning to code was an acceptable (and likely superior) alternative to learning, say, Spanish.

“Whoa,” he said, “I’ll have to think about that one.”

That’s part one.

Part two is that last November, Richard was kind enough to let me observe at the Fluency Factory, and as I observed him and his tutors work, that story kept coming back to me.

On my first day there, before I watched any actual teaching, I got into a conversation with a tutor who also worked as classroom teacher. As we talked about the kinds of reading issues she encountered, she mentioned that at the international school in South America where she had student-taught, the teacher she worked with required students to stop and think before speaking.

Initially, the kids were baffled, the tutor explained. They weren’t angry, just perplexed. They were accustomed to just sounding off automatically about whatever happened to pop into their heads; no one had ever insisted that they consider their thoughts carefully before talking. But they learned to do it. These kids were six and seven years old, and reading/writing really well in both Spanish and English.

Over the next few days, I watched the Fluency Factory tutors work valiantly with lots of struggling readers, generally between the ages of seven and thirteen. In a lot of ways, it was like a redux of all of the issues I saw as a tutor: misreading words, skipping words, skipping lines, guessing at unfamiliar words, vocabulary problems, background-knowledge problems…

I was consistently stunned by the similarities between these younger students and some of my own former students, who were high school juniors and seniors. I had known in some abstract sense that they were reading at perhaps a seventh-grade level, but watching those tutors have the exact same conversations I had had with my students, except with actual middle-schoolers (and again, a mostly affluent group), drove home the extent to which those students had been stunted academically. It also made me wonder about how bad things were with the kids who weren’t getting any intervention.

The other thing that repeatedly struck me was the students’ difficulty in reading aloud.

During the last couple of years that I tutored, after I had learned a certain amount about phonics and gotten a decent sense of what sorts of problems to look out for, I began asking new students scoring below a certain threshold—usually about 550 on the SAT, 24 or so on the ACT—to read out loud for me so that I could see whether they omitted/inserted/confused words, or guessed on unfamiliar terms. What I didn’t really focus on, however, was prosody—that is, whether they could read with conversational intonation.

As I listened to Fluency Factory students read, however, I couldn’t help but notice how jerky their speech was: they paid almost no attention to punctuation, nor did they seem to recognize where the pauses and emphases within a sentence would naturally occur.

They were opening their mouths and words were coming out, but the act had nothing to do with real speech.

It was as.
If they had.
No.
Concept of where thoughts.
Began and.
Ended.

Or, for that matter, that written language was even intended to convey thoughts.

For the youngest/weakest students, I chalked this up to the fact that the mere act of decoding required so much brainpower that they had nothing left over to make sense of what they were reading, never mind doing the two simultaneously.

But the older kids mostly read this way too, and it started to freak me out a little.

It was as if, at some level, they did not really grasp that reading is written speech.

Assuming that their out-loud reading was a general approximation of their silent reading, that certainly explained a lot of their comprehension problems.

Of course, I understood intellectually that they were not reading this way on purpose, that what I was hearing was a result of their cognitive overload, but on a visceral level I found it really disturbing to hear language fragmented in that particular manner.

On multiple occasions, I realized that the jerkiness had become so ingrained in my ear that I almost started talking that way.

At any rate, it was striking to watch how their reading immediately became more fluent when they took turns reading with a tutor. They seemed to absorb the more conversational tone by osmosis. Had decoding been the only issue, that wouldn’t have happened—it wasn’t as if they were merely repeating the same words after the tutor. Yet when they reread the passage aloud again, this time entirely on their own, those gains inevitably evaporated after a couple of sentences.

I’ve written a fair amount about the connection between reading and listening, but until now, I haven’t thought quite so much about the connection between reading and speaking. I think, however, that it’s yet another piece of the reading puzzle—and a seriously neglected one at that.

As I’ve pointed out before, the language that appears in written texts is often very different from spoken language. It tends to be more syntactically complex and to contain vocabulary that students don’t necessarily encounter in everyday life. As a result, it’s hardly a surprise that even very competent decoders can have difficulty recognizing how a given line of text would be spoken (as I saw constantly when I was tutoring the SAT). And as texts become increasingly complex as students advance through the middle grades, they become increasingly divorced from spoken language.

When I wrote about this issue before, I approached it only in terms of listening: my question was whether the denigration (and in some cases the outright prohibition) of direct, teacher-led instruction, and the obsessive emphasis on student-centered groupwork, had inadvertently inhibited students’ exposure to the kind of formal, structured adult speech that bridges the gap to more sophisticated written language.

I still think that’s a major problem, but now I have to wonder whether the type of speech students are expected to produce isn’t also playing a role. If students generally have little opportunity to listen to the type of speech that overlaps with written language, then they have even less experience creating it.

In other words, if students are primarily addressing their peers informally during class—and rarely or never required to, say, give structured presentations to their entire class or (horror of horrors) sit for oral exams—then they are also not getting practice in producing the type of speech that approximates writing. And if the spoken-written connection is never fully made and/or reinforced, students then have difficulty translating words on a page into something resembling conversational speech. Lack of exposure to sophisticated texts in turn further inhibits the development of their speech, and a vicious cycle ensues.

When this idea occurred to me, I couldn’t help but contrast the pedagogies that currently dominate in the American educational system with those of France and especially Italy, countries where there is a significant emphasis on oral presentations and exams at the secondary level. (I don’t have the same level of familiarity with any other country’s school system, so I can’t comment on what it’s like elsewhere.) Students are expected to address their classmates as well as adults using formal, structured language, and to present their ideas orally with a striking degree of eloquence using—this is key—linguistic features otherwise found only in academic writing (e.g., “We have seen x and will now turn to a consideration of y…”).

That’s not to say that French teenagers don’t still mumble incomprehensibly sometimes (and boy can they mumble!), but at a systemic level, it’s a different mindset and set of expectations. And regardless of the drawbacks these systems may have—and I am by no means suggesting they are above reproach—it’s hard to deny that they are set up, on a broad scale, to systematically inculcate a particular set of verbal capacities from multiple angles and in mutually reinforcing ways.

Indeed, when listening to French adults discuss, say, politics or philosophy (or wine), it’s hard not to notice how closely their speech contains what are often only written constructions. The relationship between the various aspects of reading and speaking and, ultimately, thinking is evident, in a way that makes the links impossible to ignore.

All this is to say that I no longer believe that reading is the “three-dimensional problem” that it has traditionally been described as. Rather, I think it’s even more complex. As is true for any language, becoming truly proficient in English requires mastery of four domains—speaking, listening, reading, and writing—that build on and feed into one another. And the tendency to treat reading as a skill that can be acquired in isolation, without significant work on the other three domains, is even more of a fool’s errand than I understood even just a short while ago.