Boyte in the blogosphere | Main | Streetlaw, Inc.

September 08, 2004

the decline of reading

On July 8, the National Endowment for the Arts (NEA) released a report entitled "Reading at Risk" which uses Census Bureau data to track a substantial decline in the percentage of Americans who read any books, but especially works of "literature" (defined simply as all forms of fiction, drama, and poetry, without regard to quality).

For those of us who are concerned about civic engagement, it is interesting that regular volunteers are more likely to read than other people. In fact, according to the NEA's fairly sophisticated statistical model, volunteering turns out to be an independent predictor of literary reading. In other words, if you compare two people of the same race, income, age, employment, etc., if one volunteers and the other doesn't, the volunteer is more likely to read fiction or poetry.

This is only a tidbit of information. I would love to know whether literary reading also predicts other forms of civic engagement, such as voting, joining and leading associations, and protesting. And I would be interested in qualitative research (such as in-depth interviews) that shed some light on why volunteers read--and readers volunteer. In any case, this is an important empirical question. I'm a philosopher, trained in normative (moral or ethical) reasoning, and I have written two books arguing for the moral and civic value of the humanities. But empirical questions are also important. For example, if we argue--in the tradition of Greek Sophists and Renaissance humanists--that stories teach moral lessons, then we should see some behavioral differences between avid readers and non-readers. Apparently, we do.

Rivka, author of the excellent "Respectful of Otters" blog, raises a series of doubts about the NEA study. Unfortunately, I think she's wrong.

First, she cites a Gallup release entitled "Poll Shows Continuing Strong American Reading Habits." That survey does present some good news and should be taken seriously. However, it's not strictly comparable to the NEA/Census report, because it includes non-fiction, whereas the NEA focusses on fiction, drama, and poetry. Moroever, in the Gallup poll, the percentage of Americans who did not read books at all doubled between 1978 and 1990, then remained pretty stable until the last poll was conducted in 1999. This is consistent with the NEA/Census trend. I haven't find other studies that go back several decades, but the Ipsos surveys show the same distribution of book-buyers by age, income, and region as the NEA/Census.

Second, Rivka notices readers all around her and recalls huge positive changes, such as the expansion of Barnes & Nobles franchices into towns that were previously without bookstores. How do those observations fit with the NEA study? One answer is that all concrete, personal observations are selective and need to be checked against representative data. How many independent bookstores have gone out of business while B&N expanded? For every commuter train full of readers, how many houses are there without any books? Besides, there is a pretty simple explanation for the evident quantity of readers today--population growth. There are more people, but the average person reads less, so the number of readers has remained flat since 1982: about 96 million people.

Third, Rivka detects a tone of elitism in the study and the New York Times' coverage of it:

I'm suspicious of arguments that the majority of people are stupid, uninformed, evil, or immoral, ranged up against a tiny minority of the righteous. In the circles in which I move, the claim that 'most people don't read' is often cited as evidence for this worldview. One of the most vicious online arguments I ever had was with a man who maintained that 'only one or two percent of Americans read anything at all,' and I see that similarly extreme claims have even made it into published books.

Fair enough, but the NEA study doesn't call people stupid and immoral, and it doesn't claim that no one reads. Ninety-six million adult readers are a lot of human beings by any standard. The question is: compared to what? It seems that we are less likely to read literature today than we were in the past, and that's a bad trend. We Americans seem to be more likely to read than Belgians and Portuguese, but less likely than Canadians, Swedes, and Brits. So there is no call for rending our clothes and putting sackcloth on our loins, but we ought to ask why the rate of reading is down.

Fourth, Rivka wonders about "literature." As she says, it's "a word with highbrow associations," and she wonders "how the average person applies it. If the Census Bureau asks a voracious consumer of Harlequin Romances about her tastes in 'literature,' will she consider that it applies to her daily reading, or will she deny that she reads any literature at all?" Actually, Census didn't use the word "literature": the survey asked about novels, short stories, plays, and poetry. Only the report uses "literature" as a catch-all. Perhaps some people don't know that the romances they read are "novels," but I would think that's a small problem.

One final point: in an effort to bridge the "two cultures" of math/science and the arts/humanities, the NEA provides a pretty clear and succinct discussion of statistical modeling at the end of the full report (pdf). I've never seen an explanation of logit models before in an arts report.

Posted by peterlevine at September 8, 2004 11:07 AM


Post a comment

This blog is under attack from comment spammers, who are causing a problem for the server. I believe I can block them by upgrading to a recent version of MoveableType. However, I do not have time to do that until late December. Therefore, I have temporarily disabled comments. Please feel free to email me feedback at

Site Meter