Reading beyond decoding

Reading is partly about decoding, and E.D. Hirsch writes that:

Recently, schools have begun to do a much better job of teaching all children to become good first-step readers who can turn printed symbols into sounds and words quickly and accurately, a process called decoding. The importance of systematically and effectively teaching decoding cannot be overstated.

However, reading is not just decoding, so Hirsch also notes that:

There are students who, after mastering decoding, and reading widely can, under the right circumstances, gain greater knowledge and thence better reading comprehension. But such gains will occur only if the student already knows enough to comprehend the meaning of what he or she is decoding. Many specialists estimate that a child (or an adult) needs to understand a minimum of 90 percent of the words in a passage in order to understand the passage and thus begin to learn the other 10 percent of the words. Moreover, it’s not just the words that the student has to grasp the meaning of—it’s also the kind of reality that the words are referring to (think of our baseball example). When a child doesn’t understand those word meanings and those referred-to realities, being good at sounding out words is a dead end.

If early elementary tries to teach decoding first and content later, we're only giving students part of what they need to read about content in the upper grades. And if, each time we see the reading problem, we increase decoding and shrink other content, we could easily be making the problem worse.

(Both quotes are from the article here, which also provides footnotes to specific studies supporting Hirsch's arguments. Earlier posts on the reading-requires-content argument are here and here.)