The Social Affairs Unit

Print Version • Website Home • Weblog Home

Use the buttons below to change the style and font size of our site.
Screen version     Print version:   
February 12, 2010

Nothing stops this book being an irritating bore: The Black Swan: The impact of the highly improbable - Nassim Nicholas Taleb

Posted by Richard D. North

The Black Swan: The impact of the highly improbable
by Nassim Nicholas Taleb
Pp. 400. London: Penguin, 2008
Paperback, £9.99

A child hunkered down in a cellar in a war in Lebanon gets to reading and thinking about the world. He later becomes a financial trader. He cooks up an interest in uncertainty, as well he might, given his birthplace and profession, and after years of further research produces this book on risk which sells zillions of copies and is hailed, for instance, as "hugely enjoyable…" (Financial Times). Many of the most enthusiastic reviewers are the authors of rather similar books about tipping points, nudges, outliers, long tails, the wisdom of crowds, blinks and all the other pocket nostrums which in a catchy kind of way sum everything up, ready to be dunked in a latte in Starbucks.

Whenever I get a flash of that boy, living in what I imagine and he describes as a civilised society, and his adult incarnation of those values, I like Mr Taleb. But in the end, nothing stopped this book being an irritating bore.

Mr Taleb's book does do some valuable work, especially for those perplexed by bits of statistical thinking. But there is one important thing it doesn't do. The Black Swan tells you next to nothing about its sub-title, "The impact of the highly improbable" and that does rather matter. The book is also close to unreadable, being rich in childish prejudices (the French and all politicians are silly, etc) and infantile jokes. As a reviewer, I usually feel obliged to read every word of a book before discussing it. This time, I confess, with about 50 percent certainty, that I read about 80 percent of this book, with random dippings in the bits I couldn't face dragging my eye over line by line. (Mr Taleb says we overstate the certainty with we assess uncertainty, so my guesses may be out a bit.)

The stated thesis of the book is that mankind is not good at spotting the really huge events which will blast out of the future and change everything. These are low probability, high impact events represented by, say, Chernobyl. We don't see them coming, says, Mr Taleb, because we have all sorts of mental habits which make us think life is a series of rather normal events. We reduce complex things to simplicities, "so we can squeeze them into our heads". The process means that we have a strong tendency "to think the world is less random than it actually is".

In the matter of human height, for example, this normalising doesn't matter. Outliers are rare and an occasional 8-footer wouldn't make much difference to our sense of the norm. Mr Taleb invents a world called Mediocristan to capture this reliable world.

But it does matter, says Mr Taleb (and this is the not bad bit of his argument), that statisticians and other experts in discussing risk have privileged and promoted the normal, average sort of event as if it were universal. Indeed, they have gone further and reified the average. Thus, there is lots of discussion of how events "regress to the norm", an idea which is captured by the "bell curve". This graph expresses the idea that (like human height, say) most phenomena cluster round a normal sort of a number, with extremes (of shortness and tallness) becoming fewer at both ends and in any case lost in the crowd. But a casual reading of statistics does indeed (as Mr Taleb suggests) does give the impression that the norm is almost sucking outliers toward itself, as though by magnetism or gravity.

Another of Mr Taleb's strengths is to remark that it is a mistake to think we can measure uncertainty. It is, after all, what we don't know. And yet, of course, we are often told about the chances of a Chernobyl or a jumbo crashing on a football stadium as though there was much chance of knowing such a thing. Not that we seem to mind as much as the scaredy-custards like to pretend. We know some stuff about such nuclear and air accidents as we've had, and it's not enough to put most people off living near nuclear power stations or driving (which we know to be more dangerous than flying).

The difficulty for Mr Taleb is that much of even this less-travelled territory has been better-charted before. The best book in this field that I know is the late Peter Bernstein's 1996 Against the Gods, which has most of Mr Taleb's strengths and none of his weaknesses.

The worst of these failings is that even at his best Mr Taleb is setting up straw man arguments. I don't think the average is quite the tyrant Mr Taleb paints it, nor does a betting shop punter think his bets are scientific.

Anyway. Lots of phenomena, as Mr Taleb rightly says, are not amenable to "bell curve" type thinking. Some of them are plain erratic, whilst others are "scaleable". As one example of the latter sort of event, he cites Google; as another, he cites authorship. A Google, or a Taleb, is in a winner-takes-all sort of world (dubbed Extremistan). Each of us needs only one search engine and very few striking new ideas. The essence of these phenomena is that one way or another, they achieve a sort of critical mass of success which tends to crowd out competition. (This has been even more the case - not for much longer I suspect - with Microsoft, which Mr Taleb doesn't discuss.) Not only do most of us only need one operating system, but we find automatic - and very real - merit in the one everyone else is using even if it isn't very good.

Even Mr Taleb admits that these examples are flawed, not least because (as "long tail" theory predicts) the kind of world which produces a quick Google, say, can as soon produce a not-Google. Rather similarly, as he reluctantly admits, whilst there are enviable authors who become huge stars, there are plenty of others who gain a good deal of esteem (or at least self-esteem) and even a bit of an income without ever becoming huge successes. Indeed, the technologies which produce huge concentrations of success are the self-same technologies which widen opportunity. Anyone can post a song on YouTube. And so far as I know more books are published than ever before, and by people more able to self-promote than was once the case.

Suppose for the moment that humans do underestimate the exceptional. Mr Taleb claims a special modernity for the effect. He says that evolutionary psychology dictates that brains designed for the boring savannah can't assess exciting technological risks. I don't see this. Frankly, I imagine the primitive world to have been more full of nasty surprises than ours because - say - the sudden eruption of a nearby earthquake might be very unexpected to someone who had not read of the worldwide existence of earthquakes. Indeed, the modern world is characterised by a sort of jumpiness which flows, paradoxically, from its routine placidity. Black Swans are getting rarer and less important.

Mr Taleb makes some elementary mistakes. He says, or instance and typically, that modernity makes hurricanes worse as unexpected events because their impact is so much more expensive than used to be the case. But actually, the rich world can surf disaster much more easily than poor societies. Having an expensive disaster can be like having much less of one.

Mr Taleb shirks the crucial tasks he might have been quite good at. The essence of this lacuna is that we get no serious discussion about the nature or seriousness of Black Swans. For instance, Mr Taleb says that "in the modern environment" (presumably in Vietnam, Lebanon, Iraq - the Middle East more generally - and Afghanistan) "wars last longer and kill more people than is typically planned". Is that any truer than it ever was? I’' have thought the mess of war is hardly a surprise.

In what sense is war ever a Black Swan? I should have thought that mankind is seldom surprised by the outbreak of war, as a generality. Tony Blair might have thought in the late 1990s that an age of peace had dawned. But did anyone else much share that view? Mr Bush and Mr Blair might have thought that the second Iraq war would be easy. Surely the majority view was that they were wrong? Indeed, Donald Rumsfeld (who was at least very funny) said that the big problem wasn't what you didn't know, but what you didn't know you didn't know. That might have been a Taleb remark, and suggests that some of those who have to wrangle Black Swans do know the nature of the beast's unknowability.

Let's take some Black Swans Mr Taleb might have anatomised. So far as I remember, nothing about the First World War was widely predicted and especially not the scale of its technological slaughter. Similarly no-one predicted the imagineering which the Nazis wrought on the German mind with the tools of modern mass entertainment. No-one foresaw that a few disaffected middle class Islamists could conceive of using airliners as suicide bombs in a classic terrorist ploy rendered infinitely more powerful by its theatrical audacity. Then there are the Black Swans nature lobs at us: 'flu epidemics, and HIV AIDS, say. But none of these are discussed by Mr Taleb.

My point is that even if we discussed them into the ground we might not be any the wiser. So the point may be that the impact of Black Swans - the book's main subject remember - is that they are just the same old same-old.

Thinking about them doesn't help. One would always be fighting the last Black Swan whilst a new one was fluffing up its wings, by definition out of sight.

Mr Taleb says somewhere that you can't know when a Black Swan is on its way. That raises interesting questions when we consider, in particular, the cataclysmic things which are widely predicted. Neither the Population nor the H bombs have yet sunk us. Will climate change? Something so advertised and pre-announced can't be a Black Swan, except perhaps by being completely different to our present predictions for it. Frankly, that's one of the reasons I take climate change with a pinch of salt. It's not that I think it'll be damp squib. Rather, I think it will be full of surprises, as will our responses to it. I think it might weird because I think the future is always weird.

The future is also often fairly predictable, Black Swans and all. (Of course it isn't predictable by us: I mean it turns out to have been consistent with historical trends.) For example, capitalism has, so far, always produced crashes and yet very few people predict their arrival (though there are usually people acknowledged to be clever who loudly foretell them and profit from them).

The weird point here is that one of the main responses to Black Swans seems to be "So what?". For all sorts of reasons, the cataclysm has always been survivable and the inevitability of (say) the next crash has limited impact on the present boom. Doesn't this suggest that people don't need Mr Taleb's book not merely because it says rather little but because even if he had tried to say more we would have shrugged and said that we knew it already? What's more, we think that making ourselves safer by investing in timidity (a welfare state, bankers’ bail-outs) is not necessarily a very effective strategy.

Life's weird, and littered with Black Swans, and yet, as Mr Taleb's despised French would say: plus ça change. All kinds of unimaginable things have happened and will go on happening. But lots of things stay the same. The French and Anglo-Saxons still talk about each other in the terms they found accurate five hundred years ago, and many of them, like Mr Taleb, still find the game killingly funny. For all that Black Swans are very important, they don't seem to make much difference to those who survive the disasters they sometimes bring. So far. Crossed fingers.

Richard D. North is the author of Rich is Beautiful: A Very Personal Defence of Mass Affluence.

Comments Notice
This comments facility is the property of the Social Affairs Unit.
We reserve the right to edit, amend or remove comments for legal reasons, policy reasons or any other reasons we judge fit.

By posting comments here you accept and acknowledge the Social Affairs Unit's absolute and unfettered right to edit your comments as set out above.
Post a comment

Anti-spambot Turing code

Creative Commons License
Except where otherwise noted, this site is licensed under a Creative Commons License.

The Social Affairs Unit's weblog Privacy Statement