The Right User's Manual? Learning to Live with Uncertainty

Serendip's Bookshelves

Nassim Taleb, The Black Swan: The Impact of the Highly Improbable.
Random House (2007).

Reviewed by Anne Dalke, Bryn Mawr College Department of English.

   

I've been teaching courses on storytelling implicitly for decades, explicitly for the past five years as a first-semester College Seminar @ Bryn Mawr on "Storytelling as Inquiry." And I've just published an essay theorizing about why the concept of storytelling works so well as an introduction to college-level critical thinking--not to mention life-long learning.

So it was with great interest that I began reading Nassim Taleb's The Black Swan: The Impact of the Highly Improbable, which emphasizes (among other things) the dangers of the storytelling impulse. Taleb argues that our human insistence on reducing the dimensions of complexity, imposing order on chaos, and identifying causes for the effects we observe around us can have--increasingly does have--explosive consequences, since it "rules out sources of uncertainty and drives us to a misunderstanding of the fabric of the world." Taleb pushes his readers very hard to re-think the tendency of our storytelling brains to over-value presumptions about cause and effect, to misjudge our capacity to predict the future based on the past. The keynote of his book is what he calls "the narrative fallacy": how, in this so-unpredictable world, we fool ourselves w/ stories that cater to our Platonic thirst for distinct patterns.

Taleb constructs his argument by drawing heavily on a range of experience far outside my own: his years of work as a speculative trader, a "quant" who applied mathematical models of uncertainty to financial data. Out of the wealth of that data, he devised a theory about the danger of "the black swan," a highly improbable event with three characteristics: it is unpredictable; it carries a massive impact, and (after the fact) we concoct explanations that make it appear more predictable than it was. It is our belief in our concoctions, Taleb argues, that repeatedly gets us in trouble. Given the high impact of the highly improbable, we are way too quick to arrive @ conclusions. We need to cultivate a much more skeptical empiricism.

Part of what makes reading Taleb so enjoyable is his own storytelling gifts, including a quirky sense of humor and word play. I've picked up from him a range of nice terms I plan to put in play whenever possible. My favorite is "Platonicity." Others include "extremistan" (a province where a single event can have a huge impact), vs. "mediocristan" (where a single observation does not affect the aggregate); the "lucid fallacy," or "uncertainty of the nerd" (basing studies of chance on games and dice); and GIF (the "Great Intellectual Fraud" that is the bell curve). Admittedly, it's hard to sustain such whimsy for 350 pages, and Taleb isn't able to; his book would have been punchier, more readable, and less repetitious, had he had an editor with a heavier hand than the one he wielded himself.

Taleb's book is also full of advice about how to benefit from the unpredictable nature of the world: "focus makes you a sucker; it translates into prediction problems"; learn to "avoid 'tunneling'" ("the neglect of sources of uncertainty outside the plan itself"); "train yourself to spot the difference betwen the sensational and the empirical"; and remember that "we are not natural skeptics," that "not believing" requires an "expenditure of mental effort."

Many of Taleb's ideas--"History is opaque. You see what comes out, not the script that produces events, the generator"--are familiar to me from the ongoing conversations of the Working Group on Emergence and much contemporary thinking about the non-deterministic nature of the world. And many of Taleb's ideas accord with those that have been repeatedly explored, often advocated, on Serendip: "doubt everything," "fight against dogma," "shed the idea of full predictability," and know that "you can benefit from it." Most strikingly: "maximize the serendipity around you."

What is sharply different from what has been said on these pages, however, is Taleb's repeated insistence that our brain is "the wrong user's manual" for the complex unpredictability of the world we are now living in: "human nature is not programmed for black swans"; "our inferential machinery is not made for a complicated environment"; our "statistical intuitions have not evolved" for our current habitat"; we are "not well adapted to the present, post-alphabet, intensely informational, and statistically complex environment." I'm not convinced of this, and think Taleb's argument doesn't give the unconscious, which can handle multiplicity just fine, thank you, enough play--or enough credit.

I'd also like to talk back to Taleb about the possibility of there being different kinds of stories: those that (in the words of a friend) "quench Platonic thirst," vs. "those that cultivate the taste for ambiguity and multiplicity." Not all stories are equal, and I'm not entirely convinced that a story it is, de facto, totalizing. It may be a quality of (certain) stories to make fools of us--or a quality of ours to give (certain) stories too much scope and power. Taleb's book works as a good caution against our engaging too freely in the latter.







Comments

Rogue Medic's picture

Black Swan

You write "lucid fallacy," but it is ludic fallacy.

http://en.wikipedia.org/wiki/Ludic_fallacy

we are "not well adapted to the present, post-alphabet, intensely informational, and statistically complex environment." I'm not convinced of this, and think Taleb's argument doesn't give the unconscious, which can handle multiplicity just fine, thank you, enough play--or enough credit.

I think that he made that quite clear with the examples of the black swan, the turkey, and later the bankers demonstrated it with the sub-prime crisis. He wasn't stating that these things are impossible to anticipate, since he had been warning of this type of problem for years. He was stating that it is easy for otherwise intelligent people to ignore the problem, as he gave plenty of evidence in the book.

Did he abuse the LTCM (Long Term Capital Management) geniuses too much? Nope. That is the reward for hubris.

He does point out that the experts in this field (LTCM) were completely blind to the earlier black swan problem they brought on themselves. These black swan problems will be repeated endlessly. There will be slight variations, that will allow the perpetrators to convince themselves that, It's different, this time.

Some people may avoid these problems as obvious problems. Still, there are many experts, who will demonstrate that their brains are blind to this complexity.

Ron C. de Weijze's picture

Upside down to a different level

When I read an extract of the book, the first thing that strikes me is that the book itself is trying to be that black swan with the greatest impact. Who keeps "thinking inside the box", it argues, cannot pick up signals validating or invalidating the truth of it in the real world. What kind of world is that, being described? I can only assume it is the world of the book and its writer that "has not eaten cheese of" (as we say in Dutch) every day methodology, of which validation even more than establishing reliability, is a most vital part that can NOT be overlooked. In science, it can be, and is, even quantified, or at least contextualized extensively before admission for publication is even thinkable. However, the age of internet and self-certification may be changing this.

Had the gist of the story been about what Bergson called 'closed morality' then my preoccupation would have been quite different, for then the discussion would have been positioned where it belongs: in society and culture. Indeed the 'black swan' does play a major role there. It implements subjectivity in cultures and subcultures, groups and individuals, stemming from what is trusted,expected, suspected, predicted, believed and acted out. That is the very social fabric of places like Bryn Mawr or any multicultural depot.

This book seems to be in line with political correctness found in any such mix of cultures as a first, or what in this case has to be a second, line of defense in the fight against those who wish to hold on to their irregularities or "nerdness". In other circumstances ("out of the black swan box" so to speak) patriotism, identity, historicity, anciennity or folk-culture would have been a possible choice of words.

So it seems to me, that we are blaming the victim here. Or at least this is food for democrats, as Feyerabend's "consolations for the specialist" was food for (extreme) Leftists in the 60s. What is the subject matter of politics, to be promoted by the Left and fought against by the Right (multiculturalism), is sucked into the realm of scientific methodology and 'unmasked' as the process of 'thinking inside the box' or 'tunnel vision'.

When we turn this story upside down, it is not the box or the tunnel that is the problem, but the (lack of) validating and reliability testing that is done, or not-done, which is more basic to Left or Right establishments than, say, simply taking the matter to a different level and turning it upside down.

Post new comment

The content of this field is kept private and will not be shown publicly.
randomness