The Black Swan: The Impact of the Highly Improbable
Nassim Nicholas Taleb is, in equal parts, enthralling and infuriating. He has written a book so sure of its own certitude, that has one effective lesson: there can be no certitude. Taleb believes (and says) that he’s solved the eternal verities — that after millennia of philosophical, ethical, political, economic and social debate: thrust and counter-thrust — that mathematics and physics can save the day.
While there is certainly value in this book — if you can bear the self-regard with which it is expounded, it’s a ripsnorting read — Taleb is inconsistent about some fundamental things: Early doors, he praises the virtue of having unread volumes on the shelves of one’s library: It is a pity he hadn’t dipped into one or two of them along the way. Subsequent books in the Incerto quintet suggest he did eventually get around to it. Antifragile is a far better book, as is Skin in the Game.
Firstly, he calls himself a philosopher and an intellectual, but writes off people who “took too many philosophy classes” or “read too much Wittgenstein”, and who may therefore be under the impression that language problems are important, when in fact such intellectual niceties have “no serious implications".
This, naturally, makes him look a bit of a Philistine, which would be okay, were it not to bear directly on the content of his book. The principle problem which Taleb sets out to solve is that of the misleading narrative discipline. Better familiarity with Wittgenstein might have helped him here. The Continental view is that we *cannot* make sense of with the world but through one or more narratives. Our daily labour is to untangle and jury rig all our working narratives so they can steer us in a broadly satisfactory path through the data. “The Truth” doesn’t exist independently of our relationship to the physical universe, but rather is a function of our narratives. Narratives can’t get in the way of truths; narratives are containers in which truths are packed and brought to market. Taleb might not like Platonicity, but it is the human (Humean?) dilemma that we’re stuck with it.
This might seem a petty, arcane, weekend-ish sort of objection, but I don’t think it is. In overlooking this, Taleb fails to recognise that beloved physics and mathematics, by which he would sweep aside the dismal gaussian-inflected social sciences, are simply narratives themselves, with no epistemic priority over the social sciences (the priority that he cites is — must be — a narrative!). Indeed, he calls himself an “sceptical empiricist”, but conveniently overlooks that the social scientists tend to be far more enthusiastic collectors of empirical evidence than physicists (a point made eloquently by in a recent book by Nancy Cartwright).
Taleb’s narrative is a rather quaint (and I would say uninformed) form of essentialism: He is a thorough-going reductivist who says things like “one may have a million ways to explain things, but the true explanation is unique, whether or not it is in our grasp” and seems to take it as read that all true learning, reduces to and can be extrapolated from a few essential, internally consistent logical truths.
Pragmatists like Thomas Kuhn and Richard Rorty (doubtless volumes on his Eco-esque Anti-library: his reading in the philosophy of science seems to have stopped at Karl Popper) had a radically different (and to my mind more compelling) view about essentialism, but would agree about the pitfalls of what he calls “Platonicity”, but would mark them down as facts of life. Taleb, curiously, isn’t entirely antagonistic to Platonicity, and is happy enough to fall into it when it suits him (reductivism/essentialism about science is hard core Platonicity, in this humble reviewer’s opinion).
Taleb is also a contradictory on the value of the narrative. For example, he rails at length about Platonocity — to mistake the map for the territory — by dint of which we privilege “crisp constructs” over “less elegant objects with messier and less tractable structures": this is the sort of data compression that encourages the observation of “fraudulent” gaussian distributions — yet uses the opposite argument when warning against overly-detailed empirical “knowledge": “listening to the news on the radio is far worse for you than reading a weekly magazine, because the longer interval allows the information to be filtered a bit". That is to say, I suppose, that sometimes allowing a little bit of narrative weeding helps the Platonic garden grow.
Such inconsistencies make Taleb seem like a buffoon (but not quite so much as his own needlessly self-aggrandizing tone does!), and his repeated harping about the injustices of the Nobel prize process suggest a bruised ego somewhere on the way.
The fundamental point is that it’s in the very nature of *any* intellectual or scientific enquiry to impose some order and organisation — a narrative — on otherwise unstructured data — we have to, to separate the significant from the irrelevant. The dilemma of induction is precisely that, ahead of time, it is impossible to know what will be significant, so the exercise of making that call, QED without evidence, is inherently fraught. Taleb would benefit immensely from reading Thomas Kuhn’s wonderful and classic book The Structure of Scientific Revolutions. Without that insight, Taleb’s much vaunted “sceptical empiricism” amounts to not much more than saying “expect to be surprised".
It might, as he points out, seem rational to calculate the forecast proitability of a casino on the odds available through time at the tables, but that is to ignore the risk of something unexpected happening such as a bomb going off in the lobby. But the nature of unexpected things is that you can’t anticipate them. That’s what it means to be unexpected.
That’s ultimately the conclusion of this unnecessarily long book: “Stuff happens". The interesting material is mostly covered in existing books by other authors (I recommend Philip Ball’s Critical Mass: How One Thing Leads to Another) and the implication of the power law on the behaviour of markets is the subject of Benoit Mandelbrot’s very good The (Mis) Behaviour of Markets: A Fractal View of Risk, Ruin And Reward.