Content uploaded by Gene Callahan
Author content
All content in this area was uploaded by Gene Callahan on Mar 06, 2014
Content may be subject to copyright.
Nassim Nicholas Taleb: The black swan: The impact
of the highly improbable
New York: Random house, 2007, 366 pages
Gene Callahan
Published online: 10 May 2008
#Springer Science + Business Media, LLC 2008
Nassim Nicholas Taleb’sThe Black Swan is a fascinating but deeply flawed book.
The book’s central thesis, that the statistical models beloved by mainstream
economists and social scientists apply to the real world at best roughly and
sometimes very poorly indeed, will find favor with Austrian and other heterodox
economists. However, what could have been a sound if more modest book is
damaged by Taleb’s enthusiastic embrace of the role of maverick intellectual and his
consequent weakness for hyperbole and unfounded criticisms of those he sees as
ivory tower academics. Academics are not always right, of course, and there is a
place for outsider critiques of their ideas, but neither are they always wrong, and it is
all too easy to fall into the trap, as Taleb does, of criticizing ideas one does not fully
grasp.
Taleb’s central metaphor of the eponymous black swan arises from the fact that
scientists purportedly held that, by inductive reasoning, they could conclude that all
swans are white, since every instance of one they had seen was so colored. Then, in
Australia, black swans were discovered, upsetting their conclusion and demonstrat-
ing the flaw in inductive reasoning, the same flaw noted by Taleb’s philosophical
hero, Sir Karl Popper. If that history is correct, then the scientists involved were
certainly guilty of applying induction naively. But such a naïve induction by simple
enumeration of instances already had been criticized by some of the earliest
empiricists, such as Bacon and Boyle. Many philosophers of science have argued
that Popper’s critique left more sophisticated versions of induction standing.
1
Taleb ties Popper’s case against induction, and thus his swan metaphor, to
contemporary statistical practice by arguing that the widespread faith that the degree
of variation exhibited by many, perhaps most, real-world phenomena is closely
modeled by the notorious bell curve. He contends that this often is an unjustified
Rev Austrian Econ (2008) 21:361–364
DOI 10.1007/s11138-008-0051-7
1
See Callahan, G. “The Necessity of the A Priori,”Critical Review, Vol. 18, No. 4, 2007 for a summary of
recent literature on this point.
G. Callahan (*)
Cardiff University, 65-68 Park Place, Cardiff, CF10 3AS, UK
e-mail: gcallah@mac.com
induction based on previously seen cases. Rare events, by definition, will not appear
very often in the evidence, but the failure to find them in a small sample does not
mean they occur as infrequently as a Gaussian distribution implies that they will.
The bell curve may offer a good description of events in what Taleb calls
“Mediocristan,”the land inhabited by things like the distribution of height among
adult humans or wages among plumbers. But it is a very poor fit for depicting events
from Extremistan, where dwell phenomena like the distribution of wealth in modern
economies and book sales among published authors. We never encounter a human
being 1,000 times taller than the average, but Bill Gates’wealth or J.K. Rowling’s
book sales are the equivalent of a million-foot tall person compared to the norm.
This statistical blindness to the possibility and impact of the highly improbable,
Taleb contends, causes us serious trouble. The danger inherent in assuming that
future events will not differ dramatically from what we have typically observed in
the past becomes manifest when we are blindsided by the terrorist attacks of 9/11 or
an unprecedented market crash. We are like Bertrand Russell’s chicken, a good
(naïve) inductivist; we conclude that, because in all previous instances the farmer’s
appearance meant mealtime had arrived, it will always mean that, only to be rudely
disabused of the theory when the day comes for the farmer’s chicken dinner.
So far, so good: Taleb is on target in criticizing many social theorists for trying to
force-fit reality into familiar models simply because they are adept at working with those
models rather than accepting that the real world is far richer than their abstractions and
striving to discover novel models that better depict actual events. There is a quite sound
and useful book that the discerning reader can extract from the ambitious but
inadequately conceived embellishments with which Taleb festoons his basic thesis.
As mentioned above, Taleb tries to ground much of his critique of mainstream,
statistical thought in the philosophy of science of Sir Karl Popper. Popper famously
held that science can never confirm its hypotheses but only falsify them or fail to do
so. While Taleb places much stress on the importance of that idea for his work, he
does not seem to really grasp its implications. Of course, no Popperian could
actually make it through a single day alive without using inductive inferences. But
most Popperians at least write about Popperianism consistently. Taleb, on the other
hand, on one page claims that “there is no such animal as corroborative evidence,”
and then, over the next 20 pages, asserts at least half a dozen times that some theory
has been “proved,”“shown,”“demonstrated,”and so on, by…corroborative
evidence! Did Taleb just mouth Popper’s words with no idea what they meant?
Taleb ridicules the “dull”writing of academics; he has avoided that flaw—his
writing is not dull in the same way that watching a multi-car pile-up is not dull. At
times his slipshod composition leaves the reader baffled as to what he meant to say.
Taleb lists fields with true experts—livestock judges, astronomers, test pilots—and
fake experts—stockbrokers, clinical psychologists, psychiatrists. What is the
difference? “Simply, things that move, and therefore require knowledge, do not
usually have experts…” It’s very hard to guess what Taleb is getting at here;
certainly, livestock, planets, and airplanes all move quite a bit! Does he mean “self-
moved”? No, livestock are self-moved as well. Taleb is not stupid, but he is arrogant,
and his arrogance creates this sort of sloppy prose.
The sloppiness rears its head not just in Taleb’s writing but in his thinking as well
(in that the two can be held separate). For example, he writes, “in a primitive
362 G. Callahan
environment there is no consequential difference between the statements most killers
are wild animals and most wild animals are killers.”That must be why primitive
man was constantly fleeing from the stray slug, chipmunk, or butterfly that he
encountered while foraging!
Taleb displays a deep scorn for academic historians, finding their work “dull,”
which may explain why he seems to have little idea of the essential character of
historical understanding. He criticizes historians for devising explanations that, he
contends, make entirely unexpected events “predictable”in retrospect: “This simple
inability to remember not the true sequence of events but a reconstructed one will
make history appear in hindsight to be far more explainable than it actually was—or
is.”The writing is again dreadful, but, even suitably amended, the sentence is still
false—history has nothing to do with “remembering”sequences of events, and it is
exactly as comprehensible as historians are able to render it. The fact that the
outcome of some past situation may appear more intelligible to an historian than it
did to the participants is a virtue, not a vice, of history. Only when the historian
abandons her proper aim of discovering what actually transpired in the past and
attempts to sit in judgment of historical figures, such as declaring, “Chamberlain
should have seen that appeasing Hitler would come out badly,”is she guilty of the
sin for which Taleb condemns the entire discipline.
In making the above complaint about the efforts of historians, Taleb has failed to
recognize that statistical prediction is a feature of physical science and plays no part
in historical investigations. The historian offering an explanation of some specific
historical episode is not attempting to comprehend what occurred as an instance of
some idealized pattern abstracted from a family of events with which it is supposed
to share some significant characteristics. As Mises notes, “The notion of a law of
historical change is self-contradictory. History is a sequence of phenomena that are
characterized by their singularity. Those features which an event has in common
with other events are not historical”(1957: 212).
2
The historian is seeking the unique
and unrepeatable antecedents of the event under investigation in order to render its
individual appearance on history’s stage more intelligible. As such, how “predict-
able”the event was at its time is irrelevant, since for the historian the “odds”of its
having happened are 100%! And since the historian aims to explain that unique
episode in terms of the concrete happenings that lead up to it, concepts such as
“randomness,”“chance,”and “improbability”are categorically excluded from her
analysis. Taleb writes, “But it is hard to look at a computer or a car and consider
them the result of aimless process. Yet they are.”Certainly, if we restrict ourselves to
considering the human past in terms of matter and energy mechanically interacting
to produce unwilled motion, then we have pre-determined that all we will discover
are aimless processes. However, there is no reason we should not conduct other
explorations using other search rules, such as attempting to understand the past as
arising from purposeful human action. From that perspective, it would be ludicrous
to propose that computers and cars were the outcome of “aimless process,”since it is
obvious that their inventors set out intending to create useful machines and were not
2
Mises, L. von (1957) Theory and History, Auburn, Alabama: Ludwig von Mises Institute, p. 212.
Nassim Nicholas Taleb: The black swan 363
merely throwing together random materials only to be shocked that the result could,
for instance, be used to get around town more rapidly.
Taleb contends that we ought to “downgrade ‘soft’areas such as history and
social science to a level slightly above aesthetics and entertainment, like butterfly
and coin collecting.”His contempt for the results of historical research is also the
product of his erroneous belief that history is just an impoverished and feeble
attempt to discover scientific truths from our past, rather correctly recognizing that
historical understanding is a categorically distinct mode of grasping reality. Unable
to evaluate the subject using its proper, internal standards, Taleb cannot conceive that
history, pursued with its own characteristic methods, can achieve results every bit as
“hard”and just as much based on objective evidence as do physics or chemistry.
Taleb’s misunderstanding of history leads to absurdities like his attempt to answer
a genuine historical question, “Why didn’t more people die from the bubonic
plague?”by means of a fact about later conditions, namely, that if the plague had
been more deadly, we would not be here to ask the question! Real historians, the
ones whose work Taleb disdains, regularly answer such questions with authority: it
is now understood, for instance, that the chief cause of the collapse of the Roman
Empire was population decline.
In summary, Taleb is guilty of severe over-reaching. Rather than limiting himself
to rigorously exploring the implications of his genuine insights on the misuse of
statistics, he chose to use them as a platform from which to comment upon whatever
topic happened to enter his mind as he wrote, however little familiarity he had with
the specialized literature in the area. If he did not hold academic specialists in such
contempt, it might have occurred to him that, whatever their faults, people who have
devoted decades of their lives to better understanding some field, for example, the
philosophy of history or the philosophy of science, are likely to have something to
say on the subject, something that might even be worth listening to before one
decides to pontificate on how wrong-headed all of the experts are.
364 G. Callahan
