the NEA's misreading of reading 11.29.2007, 12:50 AM
posted by ben vershbow
Matthew G. Kirschenbaum writes an elegant and concise critique of the National Endowment for the Arts' ominously titled new study of American reading trends, "To Read or Not to Read: A Question of National Consequence", which is a sequel to their 2004 opus "Reading at Risk." The basic argument is that reading, or what they rather awkwardly refer to as "voluntary reading" (that is, reading done purely for pleasure or self-improvement) is in a precipitious state of decline, especially among the young -? a situation which poses a grave threat to our culture, democracy and civic fabric.
Though clearly offered with the best of intentions, the report demonstrates an astonishingly simplistic view of what reading is and where it is and isn't occurring. Overflowing with bar graphs and and charts measuring hours and minutes spent reading within various age brackets, the study tries to let statistics do the persuading, but fails at almost every turn to put these numbers in their proper social or historical context, or to measure them adequately against other widespread forms of reading taking place on computers and the net.
The study speaks, as Kirschenbaum puts it, "as though there is but a single, idealized model of reading from which we have strayed" -? a liesurely, literary sort of reading embodied by that classic image of the solitary reader hunched over a book in deep concentration. Kirschenbaum rightly argues that this way of reading is simply one of a complicated and varied set of behaviors that have historically operated around texts. More to the point, many of these alternative forms -? skimming, browsing, lateral reading, non-linear reading, reading which involves writing (glossing, annotation etc.) to name some -? today happen increasingly in digital contexts, constituting what Kirschenbaum refers to broadly as a grand "remaking of reading." The NEA document takes little of this into account. Kirschenbaum:
...while the authors of the report repeatedly emphasize that they include online reading in their data, the report sounds most clumsy and out of touch when referring to new media. The authors of the report tend to homogenize "the computer" without acknowledging the diversity of activity -? and the diversity of reading -? that takes place on its screen. Our screens are spaces where new forms like blogs and e-mail and chats commingle with remediations of older forms, like newspapers and magazines -? or even poems, stories, and novels. Reading your friend's blog is not likely a replacement for reading Proust, but some blogs have been a venue for extraordinary writing, and we are not going to talk responsibly or well about what it means to read online until we stop conflating genre with value.
The report also fails to acknowledge the extent to which reading and writing have become commingled in electronic venues. The staccato rhythms of a real-time chat session are emblematic in this regard: Reading and writing all but collapse into a single unified activity. But there is a spectrum of writing online, just as there is a spectrum of reading, and more and more applications blur the line between the two.
(He goes on to mention CommentPress and a number of other networked reading applications...)
There's certainly cause for concern about what might be lost as deep extended reading of deep extensive books declines, and in their crude way the NEA's stats and figures do tell a worrying tale of shifting cultural priorities. Indeed, the most appealing aspect of "To Read or Not to Read" is its passionate commitment to a set of humanistic values: sustained thinking, personal and moral growth, a critical outlook, the cultivation of knowledge. Few would disagree that these are things that ought to be held onto in the face of relentless technological change and a rapacious commercial culture, but to insist that the book and one particular romanticized notion of reading must be the sole vessels for transporting these values into the future seems both naive and needlessly limiting.
You could say that our group's mission is to advocate for these same values -? values that we certainly associate with books, hence our name -? but in the diverse landscape of new media. To the question "to read or not to read" we answer emphatically "to read!" But to understand what reading actually is demands a more nuanced investigation.
Nathan on November 29, 2007 11:13 AM:
Some valid criticisms here I think, but Kirschenbaum's piece and this post seem to assume that if one has historical awareness, one will think that it is possible for reading to be creatively reimagined and re-engineered - but I'm still not sure just what that means or what it really looks like. After all ever since writing was invented, there has always some "linear reading" happening, even as people have always browsed, read in parallel etc (as Kirschenbaum discusses). Is it not pretty clear however that linear reading - where one must follow sustained argumentation, development etc. - is less and less everywhere as the image becomes more prevelant?
In Kirschenbaum's piece, he talks about "lateral reading", or "comparison and cross-checking" (as distinugished from "focus and immersion" and "reading for depth") -? but in the examples he gives it its seems not only that the "lateral reading" exhibits a kind of "depth" of its own, but also involves quite a bit of linear reading as well.
Again, Kirschenbaum writes this: "Reading, Bayard says, is as much about mastering a system of relationships among texts and ideas as it is about reading any one text in great depth." I guess what's at stake here is what we mean by "reading".
"Mastering a system of relationships among texts and ideas" is a desirable quality (I would call this "broad knowledge", "a prepared mind", "seeing the world as a whole", "systematic thinking", "understanding the world out there", etc), but that would be a more or less new definition of reading - wouldn't it? If it is - and I think most educated and uneducated people would say it is - one really ought to be upfront and say that you are changing the way the word is used. If one would argue that this is not entirely new, in what sense would that argument be put forth? How does this "expanded definition" of reading relate to the old one? "Reading the world" less with written words but by images, sound-bytes, and however long the rhetotic can hold my attention? :) This would be a good question to explore, I think.
I suspect a lot of linear reading is not happening online... Also, I'm assuming that most *non-academics* don't do a ton of reading *for their jobs*. And I know I don't really *read well* license agreements, directions, procedures, etc.
Kirschenbaum: "my instinct is that computer users are capable of projecting the same aura of deep concentration and immersion as the stereotypical bookworm."
I am not out to say anyone is illiterate - but is this really a big deal? Should I assume, for example, that people from oral, non-literate cultures are incapable of deep concentration and immersion when it comes to ideas? (maybe this is true - can someone help?) Is this what anthropologists confirm?
In full disclosure, I have NOT read the whole NEA report. :)
Nathan on November 29, 2007 12:25 PM:
I've been thinking about this quote from Kirschenbaum more:
"Reading, Bayard says, is as much about mastering a system of relationships among texts and ideas as it is about reading any one text in great depth".
I want to take the position here that this is rather "understanding" or "knowledge", and that "reading" per se should not be conflated with this in any way. Rather, one comes to deeply understand nuanced relationships among ideas with the help of a variety of oral and linear modes, of which sequential, linear reading is one of the most crucial.
bowerbird on November 29, 2007 6:58 PM:
as a kid, a portion of my reading was an _escape_mechanism_...
but hey, maybe that was just me... ;+)
Nancy Kaplan on November 29, 2007 7:14 PM:
The posts here, and the extensive discussion of Matt Kirschenbaum's excellent response to the NEA's latest effort to scare up a literacy crisis, provide much needed discussion of what we mean, or should mean, when we talk about reading. And I don't want to interrupt that conversation in the slightest -- it's a vital one as we all try to engage with literacy in the age of digital, networked information.
But before we get to the hard issues surrounding the future of literacy, I think it would be wise to ask whether the NEA's data show what it claims. The data appear pretty persuasive -- and there is certainly a lot of it. If you go back to the data sources, though, you'll find some curious revisions.
Here's a key example. The data on reading proficiency for 17 year olds is taken from the National Center for Education Statistics, specifically from the Long Term Trends part of the National Assessments of Education Progress. The data set begins in 1971, at which point the average reading proficiency score for 17 year olds was 285. When the NEA used the data, however, it truncated the data set, showing only those data points from 1984 and after. It also redrew the graphs so that even though the assessment occurred at irregular intervals, the data points were represented as if they occurred at regular temporal intervals. Both changes introduce serious distortions: the first one finds a trend where none exists (or rather shows only a part of a larger and quite different trend) and the second one exaggerates the severity of the supposed decline in scores.
You can go look at graphic showing how the NEA's version of the data differ from the NCES version. The distortions are instantly obvious.
There are a number of other serious problems with the data. The other major report the NEA uses to persuade us that reading proficiency among adults is declining actually says no such thing. That report, also from NCES and called "Literacy in Everyday Life," concludes that "between 1992 and 2003, there were no statistically significant changes in average prose ... literacy for the total population ages 16 and older..." (p. iv). The bottom line is this: if reading proficiency has not in fact declined, then the NEA has no case left.
There is ample evidence that people are reading many, many words. Just not, perhaps, so much in printed books. The Center for the Future of the Book is pursuing important directions for digital reading environments and it is vital that many such experiments take place. It is equally vital that we not get ourselves bamboozled by distorted and cherry-picked data.
Gary Frost on November 29, 2007 9:25 PM:
Another approach to tabulating different reading behaviors is to categorize different reading formats and observe reading skills optimized by each. Airport terminal directions, blogs, valentines, wikis and print books all call on particular reading skills and these formats are optimized for quick recognition, scrolling commentary, emotional prompts, query research or attentive comprehension. Such a field guide to the reading formats also suggests inherent projections such as the future of the book or best applications for print versus screen.
Twenty years after the advent of western printing there were print shops in most large cities and these were well on their way to producing the eight million copies of 17,000 editions that would be printed by the end of the 15th century. The immediate application was book production. Twenty years after the advent of networked computers the number of websites is estimated at between 7 million and 2 billion. The immediate application of screen reading has been presentation of websites.
Which format is optimized for book reading and which is optimized for website scanning? The early history of the two reading formats tells the story. But if the guide to reading formats is anything like a bird guide the agenda would also be to experience as many reading actions as possible and become ever more skilled in recognition to enable appreciation of any new format.
Adam Hodgkin on November 30, 2007 4:22 AM:
There is a lot more to be said about the themes touched on by Kirschenbaum and Ben Vershbow. There is a naive inclination to reason as though books are or were read on their own which they rightly question (Matthew K's points about lateral reading). It is probable that "long-form" reading has always been more parallelised and contextual than the simple model of deep reading (P1..through....P221 and index -- yes what is the index for except for hopping about?) would have us suppose. Serious readers always have and read lots of books (often open on the desk) and they frequently work in libraries or studies. Its the scale and instant searchability of the new web based library systems that is changing our model and expectation of how to read deeply. The astonishing power that this brings is not yet understood and its not yet fully there. But this 'lateral' and 'search-based' approach to reading and to interpreting texts is the exciting new thing. If the lateral force suddenly increases 100-fold is it surprising that the gravitational pull of single-title deep-reading gets a bit disrupted, or re-directed?
In full disclosure, I did read the whole NEA report. But I did skim a bit. :)
I also blogged criticisms analogous to Ben's here:
Matt K. on November 30, 2007 10:00 AM:
Thanks, everyone, for the good comments here. I'd also like to thank the Chronicle for providing the open access copy of my essay.
The NEA called for a serious conversation on the subject of reading and that was the spirit in which my piece was intended. I'm not a statistician, so I take the numbers and data in the report at face value (though I've heard from a couple of people backchannel who have concerns about the methodology). For me, there's no question that lateral and linear reading have always co-existed. There's also no question that they ebb and flow in relation to one another, and that attention to the history of reading bears that out. I do think it's probably true that the general level of linear reading is currently on the decline, though I'd also like to put on the table two words that are nowhere to be found in all the NEA's documentation: "Harry Potter."
As I tried to suggest in the Chronicle piece, it's curious that nowhere does the NEA actually define "reading." They simple treat it as self-evident. And that, in turn, makes it seem dull and contributes to the little red schoolhouse atmosphere of the whole report. But for me, it's a tremendously exciting time to be talking about reading. We have new tech like the Kindle, which I've found to be a pretty remarkable reading machine (I've read almost two books on mine already) though mostly what people seem to want to do is dogpile on the DRM issues. There's new science, which is allowing us to understand reading at the raw physiological level (I recommend Maryanne Wolf's _Proust and the Squid_ as the starting point here). There's the maturity of academic fields like the history of the book, where serious and important scholarship on reading is unfolding (some names: Adrian Johns, Peter Stallybrass, Anthony Grafton, Lisa Gitelman, Pat Crain). And of course we have new writing spaces, like this one right here, where we're all also reading. And writing. And reading. Such is the volatile mix into which the NEA's data on reading are injected. To the extent that it provokes people like me or Ben V. or Dan Cohen (see his excellent "The Idealization of the Book" on his blog) to react, I think a large part of what we're reacting to is that this context is simply missing from the pages of a report destined to attract widespread media and public attention.
A last word: I can't recommend the Bayard book highly enough. It's a wonderful contrast to the stark binary of "To Read or Not to Read," because for Bayard *real* reading (or real not-reading) takes place in the messy spaces in between the binary, where our memories of books, our conversations about books, and our encounters with books all come together to create our "readings" of books.
Matt K. on November 30, 2007 4:03 PM:
Just wanted to note that I hadn't seen Nancy K.'s comment (above) until after posting my own, as her comment was apparently held up by a spam filter until not long ago.
Gary Frost on November 30, 2007 9:14 PM:
The word reading means too much beyond skilled decipherment. Statisticians and psychologists and publishers have most of the territory and their disciplines are usually invisible. But what is also relevant here is what is happening to book reading as a format of presentation for formal, deliberate conceptual works posed in words and illustrations. Orchestral works.
Now what is too apparent with regard to books is that the last twenty years of outright comparison of print and screen presentation has provided unambiguous indicators, regardless of popularities of book reading. Print is well adapted to presentation of books, and in spite of simulation, the screen is not. Twenty years of comparison counts. It counted with the telegraph, telephone and television and automobile. It counted with the advent of printing.
On the other hand, the networked computer has quickly engendered taxonomies of reading unrelated to book reading. That momentum is only compounding reading skills and compounding meanings of conceptual works including books. Who is fully tracking this momentous addition to the skills and activities and implications of screen reading? Contrary to reports, we could be in time of a momentous increase in reading. Books, especially those in the classic mode of academic and literary formality, can only prosper in such an environment of increasing reading skills. Print books are already products of digital technologies why should they not also leverage increasing reading skills?