A critique of the normal

Today, I would like to take a look at a book by a man who detests academic economics, with a particular distaste for econometrics, unnecessarily complicated language, and most people who use the normal distribution where it would not apply, according to him. The one and only Nassim Nicholas Taleb. I hope to present his ideas in a concise and clear manner, and I hope the reader may forgive me if I fall trap to misinterpretation, as many other reviewers have. ‘The Black Swan’ is a fascinating piece of literary art, where mister Taleb eloquently explains his views, how they came to be, and how these epiphanies have impacted his personal life and his worldview, which seems unique in its embrace of the extremely unlikely. 

The book, depending on the print, has 4 to 5 parts to it, with later prints also containing a postscript essay section. Due to the denseness of concepts and discrepancy between prints, I will only focus on the first part, which illustrates how the Black Swans of our world often escape our perception. 

Part 1: Umberto Eco’s antilibrary, or how we seek validation
The book starts on an interesting tone, with a small bit of life advice regarding how one should design their library, that it should be filled with unread books, not read ones. Umberto Eco’s, the famous writer, antilibrary is a private library filled with unread books, which transforms a library from a showcase of the knowledge one has gathered, to a research tool of what one has yet to gather. Mr. Taleb then gives a first taste of the notion of a Black Swan through example, using the war in Lebanon he experienced first hand. The triplet of opacity is detailed as follows;

• The illusion of understanding, or how everyone thinks that they know what is going on in the world that is more complicated than they realize.

• The retrospective distortion, or how we can assess matters only after the fact, as if they were in a rearview mirror. History is more organized in books than in reality.

• The overvaluation of factual information and the handicap of authoritative and learned people, particularly when they create categories- when they “Platonify”.

These three notions are why Mr. Taleb can be so outspoken against historians, or economists, who seek to explain the world in simpler terms than it warrants. When considering these accounts of history, one can often see that opinions tend to cluster, that journalists work from a limited number of frameworks, and that any reduction through categorization of events through such frameworks rules out some sources of uncertainty, driving a misunderstanding of the world. He continues to expand on this notion of opacity through his own experience, both seeing others fall to the need to explain everything, or “Platonify”, and his own obsession with betting against that which seems impossible under Platonification. 

Continuing on, the concept of scalability is introduced as the best and worst advice he has ever been given. The best by it allowing a clear cut distinction between types of uncertainty, between extremistan and mediocristan, and the worst by being horrible job advice for all but the successful. Mediocristan resembles the world of empirics that we have seen quite extensively, with its supreme law being “When your sample is large, no single instance will significantly change the aggregate or the total”, the world of the non-scalable. However, the strange country of Extremistan follows a far different supreme law, “Inequalities are such that one single observation can disproportionately affect the aggregate, or the total”, the world of the scalable, from book sales to derivatives, where knowledge grows slowly and erratically with the addition of data. Extremistan is where we are subjected to the tyranny of the singular, the unseen, and the unpredicted. This is of course an approximation, and should not be Platonified beyond its means, but Extremistan is where most Black Swans live. 

The next section, “One thousand and one days, or how not to be a sucker”, introduces us to the Turkey problem, and how a Black Swan is relative to one’s knowledge. Suppose we have a turkey, which is born on a farm and fed each and every day. It knows nothing but that it is fed everyday, and assumes that tomorrow will be mostly the same, with maybe some change in feed. However, Thanksgiving rolls around, and suddenly, the turkey is the food for that day. This is the inherent problem with knowledge gained from observation, is that it can suffer from an extreme yet to be observed event. Do however note, that the slaughter was a Black Swan for the turkey, but not for the butcher. The following section builds well on this, introducing the notion how people can confuse absence of evidence with absence of evidence, and suffer from their own confirmation bias in forming their worldview.

The next section, aptly titled “The narrative fallacy”, delves into people’s tendency to create a story of how things happened. Personally, I found it incredibly insightful that Taleb noted that not theorizing is an act, more than theorizing itself, as it is a natural tendency in humans to do so. It is interesting to see how he approaches this tendency both from a biological perspective, and from an information system perspective, as obtaining, storing, and retrieving information is costly in any setting. Therefore, people are likely to reduce the complexity, or dimensionality, of the information we obtain, making storage and retrieval less costly, but these simplifications often leave out the most important parts, the seemingly innocuous and forgettable that causes Black Swans. The implications of this dimensionality reduction is further elaborated throughout the section, from paranoid people, to journalists, to how the narrative fallacy can be so influential on our brain through Daniel Kahneman’s work.

The narrative fallacy is followed by “The antechamber of hope”, where, based on the notion of Black Swans, two strategies to pursue are presented. First, when dealing with Black Swans, one can either bleed or blow up. One can bleed, take the losses, while betting that a Black Swan will occur, or one can disregard the possibility of a Black Swan and blow up when it does occur. The first approach, which seems the most logical assuming Black Swans exist, is the one that is the most orthogonal against our human nature. We prefer steady gains over big wins, and most definitely prefer steady gains over, seemingly unbounded, small losses. Fighting this compulsion is just like fighting the compulsion to theorize, something that goes against our emotional compulsions.

The following section again introduces another notion of how our perception can fool us, the problem of silent evidence. History, being any succession of events seen with the effect of posteriority, is connected to the notion of silent evidence at any level, as we always fail to observe the difference between what we see and what is there. This is classified as the ‘distortion bias’, and can be applied to history, journalism, and any practice where we have to construct samples and gather evidence. It is easy to forget the cemetery when considering millionaires, worshippers,gambler’s luck, the stability of species, and the health benefits of a russian gulag, yet it will always lead to a false view of the world. The tale of Giacomo Casanova, famous adventurer and infamous seducer, reveals the most severe manifestations of this concept, the illusion of stability. He survived numerous trials, and through his survival he came to believe he was chosen by destiny, yet there were many such adventurers, and seducers, like him, where a fair part did not have the fortune he had. An interesting parallel that is discussed in this section is the self-sampling assumption, a generalization of the Casanova bias, and how our existence skewes the computations in trying to understand how humanity got to this point, on this planet, in this universe. Look at your own life, at the points where you were heavily injured, scarred, or financially destitute, is the place you are now in the condition your are a low-probability event? Are you the surviving Casanova? In all of these cases, complicated questions from history or data can sometimes only be answered with an “I don’t know”, simply due to the Casanova bias, as a measure of uncertainty always needs to be factored into such matters. 

The last chapter of this first part is titled: “The ludic fallacy, or the uncertainty of the nerd’, and may be Taleb’s most pointed critique of academic economists and econometricians. The ludic fallacy is, in short, the tendency to confuse the sterilized chance observed in a casino, where the chances are mostly known and the sources of uncertainty quickly identified, to the chance in real life, which are unknown and require discovery, and where sources of uncertainty are almost boundless and rarely defined. The difference between Knightian risk and Knightian uncertainty matters little in this view, because most of the  Knightian risks we encounter are laboratory contraptions. This is further illustrated by a casino’s almost financial ruins. The casino in question has done everything to ensure no lucky gambler could bankrupt them, yet the causes of real loss to them were a performer being mauled by a tiger, a disgruntled contractor almost blowing them up (literally), an employee who inexplicably did not submit any tax paperwork, and a number of other dangerous situations, not limited to the owner’s daughter being kidnapped. The bulk of the casino’s risks come from outside their models. The last chapter of this part continues on to summarize what we have seen in one simple statement : “the cosmetic and Platonic rise naturally to the surface.” It is why we struggle with silent evidence, why we Platonify, why we want to confirm, and why we have Black Swans yet never learn from their occurence, as the ones that did not happen are far too abstract. Randomness and uncertainty are abstractions, ones which mathematics can not always capture, far from it, and we use context to reduce these abstractions, reducing computational and mental efficiency, yet exposing us to the unknown unknowns.

“The Black Swan” is one of the books of the “Incerto” series, where each of the books deal with a certain problem in our view of uncertainty, and how this affects society at large. The Black Swan is a wonderful illustration of how we fail to see the uncertainty around us, addresses the problems that arise from them, and introduces us to possible solutions to some, but far from all, of the Black Swans of Extremistan. The book has been a personal eye-opener and I will most definitely be reading this book at least once more, both out of curiosity and my lack of understanding the first time around. To my dear readers, I highly recommend this book, which allows one to recognise the problems of the mathematized probability theory we use to model the world, and allows us to step to a more fuzzy and general definition of chance. Further, I would recommend the book even more to any econometricians, if only to see the arguments of a man who opposes all the methods we use, and to judge with our own wit and eyes whether we think they hold any value. However, as the avid reader might have noticed, I am far from truly understanding his ideas, let alone oppose them for now. So, I invite all with an interest in the uncertainty in the world around us to join me in exploring this book’s depth, if only for curiosity’s sake.