Free Criticism, Science After Data, and Airport Books
July 1, 2009
When I saw Malcolm Gladwell doggedly dissecting Chris Anderson's upcoming "Free: The Future of a Radical Price" (see Chris' response here) my first reaction was: Brilliant! Chris Anderson is editor-in-chief of Wired, and Malcolm Gladwell is a top brand name at The New Yorker, and as corporate cousins, clearly Condé Nast's publicity machine must have engineered this beef, trying to boost sales of both their titles through a completely manufactured rivalry.
Their past titles have been champions of what I call the "Airport Books" genre: The elite class of business titles that I see sold in airport newsstands next to the magazines and crappy romance novels. (I might have unknowingly stolen "airport books" from someone else, but I can't find a citation.)
Alas, I'm assured that this particular contretemps isn't a planned corporate PR stunt. (Though I know lots of nice folks at Condé, they don't seem to mimic street-level hip hop marketing as often as one might hope.) Instead, it seems the criticism and counter-argument are sincere.
The core of Gladwell's argument is simple: "Free" fails to provide data to support its claims about the future of pricing, using anecdote and confident assertion in place of actual evidence. In his objection to this methodology, Gladwell seems uncharacteristically strident, compared to his usual measured tones. Whenever I see somebody getting their dander up, I think of one of the first things I ever blogged about ten years ago: We hate most in others that which we fail to see in ourselves. Ah hah!
Let's see what criticisms have been leveled at Malcolm Gladwell's The Tipping Point, Blink and Outliers, the juggernauts of the airport book genre:
- The IUP Skeptical Inquirer's Wesley Cecil has a review of Blink which offers this up: "Gladwell relies heavily on anecdotal evidence. The centerpiece of each chapter is a short story or series of stories that are supposed to illustrate some aspect of his theory of snap judgments. ... [I]ntuitive feeling triumphs over careful study. ... [O]ne case does not an argument make."
- Slate's Jack Shafer on Gladwell telling a tall tale at The Moth, which kind of ridiculously insinuates that this is a character flaw instead of just a fun story.
- More definitively, the New York Times' Michiko Kakutani made this point in her review of Outilers, but extends it to apply to all of Gladwelll's books: "[His] books are filled with colorful anecdotes and case studies that read like entertaining little stories. Both use PowerPoint-type catchphrases (like the 'stickiness factor' and 'the Rule of 150') to plant concepts in the reader’s mind. ... 'Outliers' Mr. Gladwell's latest book, employs this same recipe, but does so in such a clumsy manner that it italicizes the weaknesses of his methodology. ... [His examples are] all based not on persuasive, broadband research, but on a flimsy selection of colorful anecdotes and stories."
- Joel Spolsky keyed off of Kakutani's review on his popular blog: "what's been driving me crazy over the last year... an unbelievable proliferation of anecdotes disguised as science, self-professed experts writing about things they actually know nothing about, and amusing stories disguised as metaphors for how the world works." (Bonus points to Joel for swiping en passant at airport book titan Thomas Friedman and his cartographic ironing board.)
- Kevin Arthur's post built on Joel's rant, offering a slightly more measured, but still critical analysis: "I feel like clarifying my opinion on this... I think there is great value in pop science books, in articles written by non-experts, and in anecdotes. I read Joel's piece not as a rant against all those things but against those things badly done."
- Peter Coclanis at Open Letters just gets downright mean: "[L]et me say from the get-go that my goal in this piece, which focuses on Outliers, is to demonstrate at once how wildly overstated such just-jacket claims are and how egregiously incomplete, insubstantial, and unconvincing Gladwell’s explanation of success actually is. His methodology stinks, too, and, from his dust-jacket photo, he appears to need a haircut." I know what fun it can be to bash someone from afar on the web, but I bet Coclanis is a lousy dresser. Just sayin'.
- And not to belabor the point, but let's close up with Isaac Chotiner in The New Republic, taking a stridently snarky look at Outliers: "By the time Gladwell reaches his penultimate chapter, he is in full inspiration mode, and impervious to all forms of critical thinking. ... Here is the Gladwell method nicely on display: a questionable assumption, a partial walk-back of an earlier claim, and finally another questionable assumption synthesizing the half-reversal."
My goal is not to ennumerate all of the criticisms of Gladwell's books — I enjoyed reading all of them, and I like his New Yorker pieces, and that's kind of all I would ask of the guy. But I can't help but wonder if being ceaselessly criticized for using assertions and anecdotes in lieu of hard statistical data has left him much more inclined to criticize others for using the same technique.
I haven't had a chance to finish reading Free yet, but I am sure that both of these authors' books absolutely do lean more towards anecdotal evidence than statistical proof. And honestly, it's okay that these books don't necessarily follow the tenets of hard science. In many cases, they're arguing that a cultural trend is becoming true, or is about to become true, and the reality is that asserting that these trends are ascendent actually helps them come true. In short, these are books designed to create culture, presented in the guise of reporting on culture. I like that!
But of course there will always be those who disagree with the idea of starting from a premise first, and then finding examples to support it. Perhaps the last word in favor of using hard data to support social observations may be from a story package in Wired a year ago, which was headlined "The End of Science" and anchored by a story called The End of Theory:
This is a world where massive amounts of data and applied mathematics replace every other tool that might be brought to bear. Out with every theory of human behavior, from linguistics to sociology. Forget taxonomy, ontology, and psychology. Who knows why people do what they do? The point is they do it, and we can track and measure it with unprecedented fidelity. With enough data, the numbers speak for themselves.
... But faced with massive data, this approach to science — hypothesize, model, test — is becoming obsolete. ... The new availability of huge amounts of data, along with the statistical tools to crunch these numbers, offers a whole new way of understanding the world. Correlation supersedes causation, and science can advance even without coherent models, unified theories, or really any mechanistic explanation at all.
The author of this compelling argument in favor of using overwhelming amounts of data to help replace formulating theories about human behavior? Former scientist Chris Anderson.
Bonus link: If you're interested in actual debate about the content of the book, Mike Masnick's excellent overview over at TechDirt is a must-read.
3 TrackBacks
� ???? Windows 7 RTM ??? 13 ??? � Wireless Network Scanner inSSIDer � Force Firefox To Open Links In Same Tab � IE7pro 2.4.6 � Opera Unite 10.0 Build 1622 Beta � Mozilla ?? 7 ? 7 ??? Firefox... Read More
There have been a lot of great conversations around and about some of my recent posts; Here are some highlights. My post about Google's Microsoft... Read More
Michael Arrington. Dave Winer. Tim O'Reilly. Jason Calacanis. Add a few names of your own. Within the navel-gazing little corner of the tech world that... Read More
Hey! I've got a airport book that was specifically written for business travelers; though just about blogging not some great insight into the buying mind. I'd like to hear Robert Walker pop into this debate. He wrote Ronsumed.
Oh sure, you've got a lot of links to stories about Gladwell's reliance on anecdote over data - but where's the quantitative evidence?
You MUST start with the premise first. There is no other way!
BOO to those "who disagree with the idea of starting from a premise first, and then finding examples to support it".
Try telling a classroom of chemistry students to just "observe".
"Observe what?" they will cry.
Exactly. There are almost unlimited numbers of measurements you can take. You need to get specific and apply a selective filter to the world to examine the things you find interesting.
These kinds of books are the hypothesis. Other people can do the testing.
If I could quote Karl Popper here I would. He was all about this stuff - philosophy of science.
Thread over. Erik wins.
Excellent point about creating culture. That was one of the hardest lessons for me to learn with public speaking. People would always ask 'what's next' in my industry, and I'd answer that I didn't really care what was next, we were prepared, and we'd adapt and it'd be fine. My poor business partner spent months trying to get me to understand that it wasn't about actually predicting things, it was about pretending to predict what you actually wanted to come to pass, to increase the chances that it did.
>>"Free" fails to provide data to support its claims about the future of pricing
When you figure out a way to get hard data about the future, please let me know, I have some stock picks I'd like to make.
Don't forget Duncan Watts in the article "Is the Tipping Point Toast?"
In an effort to reach the maximum possible audience, both authors and publishers of this and other genres probably (wisely) avoid ponderous scientific details, the cost of which is lost readership. In short, anecdote is always easier to sell than statistics, except, perhaps, when you're writing a book for statisticians.
The core of this argument is nothing more than a clash between prominent representatives of old and new media worlds; Gladwell, though a forward thinker, is nevertheless firmly ensconced in traditional publishing environs; as for Anderson, well, he could hardly remain in-character if he were to represent anything but New Media.
The big problem I have with Anderson's work is that it strikes me as hypocritical.
As pointed out elsewhere on the web, there are vast portions of the text copied verbatim from wikipedia without citation.
So Anderson feels it's okay to borrow from 'Free culture' without credit and for profit?? (but lets set that aside for a moment or give him the benefit of the doubt that it was a 'mistake'...)
The book itself says that content industry is basically now a free commodity. How ironic that it costs $26.99 to find out why.
Furthermore, the "free" electronic text edition that will actually have citations to cribbed material is only available for a limited time (?) It is not "free" at all. ((Nor does it address the violation of the CC license from said material above, but again...whatever.))
Finally, the lack of statistical data is more relevant here than in other books chiefly because he's making economic arguments!! He's also confusing marketing with economics, but that's another story altogether.
I just find to all-to-quaint that a well known editor of a print magazine *I PAY FOR* is telling me that the future is *everyone else* giving their material away for free...just not him.
People and Businesses are in it for the money. The content might be free, but peoples time is not.
Meanwhile, both of these guys books sales, pageviews, ad impressions, and stock keeps rising. Whose that free for? Does Anderson do this just for 'recognition' as he suggests the journalists of the future will?
Sorry to go off on a rant, Anil... I know this is outside of Gladwell's criticisms too and your analysis. I just think there's more here than Gladwell's own insecurities.
Vacuous, except for the Chris Anderson coinage of "write-through" in his politico-esque wave-away of the plagiarism charges.
Write-through.
Both free AND easy.
Great post Anil - totally agree!
Both Gladwell and Anderson are intellectual entertainers. They make their living making ideas exciting and not by figuring out if the ideas are true. This is why most academics and social scientists HATE them. They are parasites on real research, they get all the credit, AND they twist the fact and get things wrong. [There is also some jealousy - Gladwell gets seven figure book advances AND makes literally millions of dollars a year speaking to corporate audiences.]
I think Gladwell noticed that the tide was turning. That we are at the era of big bold books about a simple big idea that will change the world. People are starting to see through the formula and want to read something that is more balanced, accurate, and considered. Since Gladwell is the poster child of the big idea books, his attack on Anderson could be a savvy way to deflect attention from all the unsupported and over-simplified stuff in his own books. And then Mr. Dash came along and called him on it. ;)
Most of the scientists (myself included) who take Gladwell to task for his "lightweight" approach would sell their souls for his storytelling ability. The fact is he understands very well that human beings reason by anecdote, and he's really good at digging up the right anecdotes to support his points.
That said, I do have a general problem with the "airport book" application of scientific theories to business and/or society. It sounds great to talk about tipping points, immune systems, evolutionary behavior, chaos theory, local maxima, etc., but those analogies are almost never actionable in a non-scientific context. I wince, for example, when I think of the damage the book "The Tao of Physics" did to a generation of non-scientists by convincing them they a) understood quantum mechanics, and b) that it applied to or solved personal, spiritual and societal problems.
It's a pretty common misstep with examples we could trace back into the annals of civilization. In fact, what's that anecdotal trend point from the bible?
I'm unsure the exact definition of a mote, but I think it's smaller than a beam. Uh...I just blockquoted the bible, woah.I really like this post, Anil, esp. the way you tie in that anecdote-as-proof style of writing with a seemingly opposite trend, data-mining as "scientific method." I wasn't expecting that move, but it is smart and intriguing. I have to think about this for a while.
If anecdote-as-hasty-generalization proof (proof as synecdoche?), then is data-mining as "truth" somehow the aggregated opposite of inductive, qualitative, ethnography?
If you want something, *someone* has to pay for it.
If Anderson was asked "why is your book not free then?" the answer would probably be that it's an oversimplification, and explain the shop distributing it gets a cut, a fraction of which going into its rent, etc. and that it is not all going into his pocket of course.
This is merely what Gladwell is explaining. Making quality information and bringing it to people does cost money, and someone has to pay for it at some point.
Plus, no company wants to just cover costs, and that doesn't make them greedy. Running a business is also having the foresight to plan expenses in advance and allow yourself to invest.
Even "free" posterboy Wikipedia asks for donations on a regular basis. The name changes, but it is the same logic as the money paid by advertisers to Google, and readers to newspapers.
Personally, I would take Anderson seriously if his book was sold in shops with a clear statement on it that indicates he is not making a cent on them, AND at the same time as a free text/PDF download online, AND a free audiobook on the iTunes store. Of course, all of it under a very open Creative Commons license. Everything short than that is hypocritical.
Great post. I feel to better answer Gladwell�s question, we need to go back to Economics 101, which states prices in a free market are decided simply by supply and demand. In other words, in today�s Web 2.0 world when there are countless blogs providing an abundance of news info (and let�s not forget Twitter on top of that), it is only natural that this over supply would eventually drive price down to near zero. This trend can also be proven by another Economics rule � in a market with excess competition, the prices will be driven down to near or below cost, which is also close to zero in today�s cloud computing world. Thus, it�s not incorrect for Anderson to say online information actually �wants� to be free.
Read rest of my thoughts here:
http://linjamie.com/2009/06/30/why-does-online-information-want-to-be-free/
Been straight calling Malcolm out since '02. (You even pop up in that thread!)
WRT to Free - I am listening to the free audio version of the book right now. I got it a few days when it hit my blog feed. It's a hefty 200+ megs. You can find the text version online pretty easily too.
WRT to Gladwell - Anderson talks about the "creative destruction" (to borrow another term from popular in airport books) of the journalism business and there is nothing of similar value to replace it.
For the life of me, I don't understand why Gladwell didn't just say that Anderson conceded the argument.
Interesting, because more recently Wired published a piece about the formula that imploded the derivatives market, and how its overreliance on correlation without a testable, coherent model was integral to its downfall.
Qualitative anecdote and quantitative study are different forms of cultural anthropology, both valid.
As others have said, we live in a world where the future is here, just not evenly distributed.
And the best way to predict the future is to invent it.
Malcolm Gladwell and Chris Anderson are popularizers of new science: science so new that it is still at the hypothesis-testing stage. The experiments will follow. But it is reportable nonetheless.
They are cheerleaders for the future they want to live in. Aren't we all?
I therefore assert without proof that while quantatitive analysis is good for connecting the past to the present, qualitative anecdote is better for connecting the present to the future. It's unavoidable!
It's so typical of the geek keyhole view on reality that you'd first see a debate with your sacred religious doctrine of Free as something that some book publisher cooked up to get sales.
Good for Malcolm Gladwell, taking on that blowhard technocommunist Chris Anderson who does not have any facts to back up his cult.
Malcolm Gladwell is a hell of a better researcher and writer; the New Yorker is the premier East Coast publication that I guess just isn't suiting the California haze.
It doesn't matter even if it were true that Gladwell only relies on confidently supplied anecdotes (like Chris Anderson, Seth Godin, Cory Doctorow, and every one of the pantheon of gods of Free). You haven't answered the question and the challenge here. THAT's the issue. YOU cannot justify this fake communist business model *either*.
Every one of these hucksters peddling the patent medicine of Free is unable to show the theory to work for anything other than this: giving away free books or blogs in exchange for lecture fees for lectures about how you can make money giving away stuff.