Reading, Science, and Reductionism
There's an old Roman insult that goes like this: "He can't read or swim." The presumption is that just about anyone who applied himself (or herself) could learn these skills. Indeed, many countries with high literacy rates, such as Japan, are successful in teaching children to read without all the angst and sense of crisis so common in the United States.
Here, our crisis mentality has led to the curious elevation of "reading science" as the savior that will lead us from a reliance on tradition, habit, experience, and impression to a truly solid foundation. The 2000 report of the National Reading Panel is seen by many reformers as a bold first step that will place reading instruction on a scientific basis. Virtually any educational product related to the subject now claims to be “research based,” no matter how tenuous the connection.
The reading panel's report has elicited a steady drumbeat of criticism over the years, both for what it chose to examine and what it chose to ignore. But even many of its harshest critics seem to accept the general principle that reading science has the potential to adjudicate among different practices and identify "what works." It’s just that the National Reading Panel did a bad, unwisely selective job.
This top-down model of research is, of course, the century-old dream of industrial management ushered in by Frederick Winslow Taylor, who promoted the belief that a class of researchers (and not the workers themselves) could determine the most efficient way to manage the activity of labor. In this way, we could achieve a science of labor, of reading, of living.
I would like to take a more radical position and argue that there is a fundamental problem with this use of research—and the name for this problem is reductionism. We can begin with the term "variable." The bias of this “gold standard” experimental research is to view teaching practices as a set of variables; some promote the learning of reading, some may not. Some practices may just be in the curriculum because they have always been there. The researcher promises to give causative weight to variables, so that educational practice might focus on "what works."
In its review, the reading panel identified a handful of variables, or individual practices, that seemed to be effective: instruction that focused on vocabulary, fluency, phonemic awareness, phonics, and comprehension. It’s an odd list—fluency and vocabulary instruction, but not independent reading. Nothing about writing. More significantly, it is an incomplete set of parts that does not add up to a coherent cultural practice.
In many ways, the flaws in reading science are the same flaws that Michael Pollan points out about nutrition science in his book In Defense of Food. The watershed moment for Pollan was the shift from "food" to "nutrients" that began in the 1920s (with the discovery of "vitamins") and gained full force in the 1970s when food began to be viewed as a delivery system for carbohydrates, proteins, fiber, and other newly emphasized elements. Foods were essentially the sum of the nutrient parts. And, he writes, "Since nutrients, as compared with foods, are invisible and therefore slightly mysterious, it falls to the scientists ... to explain the hidden reality of foods to us."
Eating was thus mystified. It was no longer a cultural practice, built of tradition, conviviality, and pleasure. Furthermore, since nutrients (and not food) were the central concern, these nutrients could be extracted or manufactured to "fortify" a vast array of processed products that were no longer "food," but "edible food-like substances."
If all of these changes made us healthier, Pollan’s argument would seem mere sentimentality. But they haven't, and part of the reason comes from the naive belief that these extracted or manufactured nutrients act the same way in isolation as they do in the actual practice of eating food—that a pill containing a nutrient found in carrots has the same benefit as eating that nutrient in a carrot. A nutrient is a nutrient. But the carrot itself is part of a system in which a range nutrients interact, and the nutrients of the carrot interact with other foods being eaten—an extraordinarily complex and poorly understood system of metabolism.
The analogy to current reading instruction is telling. Reading has been mystified so that teachers (many of whom have been successful for years) are asked to be dependent on researchers who can tease out the key variables (or nutrients) in instruction. Accumulated teacher experience is devalued as impressionistic when compared to the results of hard science. Reading programs are delivery systems for these variables, and children typically spend more time with "reading-like" activities than they do with reading.
It is common for one reading selection in some basal readers to have as many as 50 pages of supporting activity in the teacher’s manual. And for millions of children across the country, "fluency" translates into the speed-reading of nonsense words, which seems the data of choice in many schools. But there is gold in them hills: Like processed foods, processed reading allows for a hugely profitable range of workbooks, tests, and consumable materials. Schools don’t develop reading programs—they buy them. Actual books, if available at all, make up a small part of this expenditure.
Those who argue for a cultural, or in Michael Pollan's terms, an ecological, approach to reading—one in which literacy is a meaningful, invested activity—are seen as hopeless romantics. But who’s being impractical here? Real food, real reading. Something to consider.
Vol. 28, Issue 23, Page 29Published in Print: March 4, 2009, as Reading, Science, And Reductionism