Straight From the Black Swan’s Mouth

A few days ago, I blogged about Nassim Nicholas Taleb‘s new book, The Black Swan, and solicited questions for a Q&A that NNT had agreed to answer. Here now is our inaugural user-generated Q&A.

Many thanks to all of you for the good questions and observations, and thanks especially for NNT’s thoughtful replies.

It seems fitting that we post this reply shortly after the announcement that a Florida treasure-hunting outfit discovered a Colonial-era shipwreck called — yes — Black Swan. Its treasure is reportedly worth some $500 million. Now that’s a black swan.

Q: What is The Black Swan (hereafter “TBS”) about?

A: I wanted to have fun while fighting the commoditization of books by “me-too” publishers. Very simple: I took the “book proposal form” and made sure that my book could not possibly fit any category and that not a single one of the standard publisher questions could be answered. The question, “who is it for?” was particularly annoying. I also put in a fictional character (unannounced) to confuse the fiction/nonfiction distinction. The funniest part was when I told them that it had enough material to be a textbook. “But in what discipline?” they asked.

I also wanted to make sure that no reviewer could possibly write a review by “skimming” the book or reading the table of contents. But only two reviewers fell into the trap. In brief, the book is a map on how to live in a world we don’t fully understand – and enjoy it.

Q: What is “random”?

A: Someone asked about why I consider planned events such as 9/11 to be random (when they were in fact planned by the perpetrators). I keep writing here and there that my definition of randomness is as follows: incomplete understanding or incomplete information. If I see a pregnant woman walking down the street, the sex of her child is a random event for me – but not for her doctor, and certainly not for God. Likewise, nothing in a coin flip is truly random for someone capable of doing differential equations in his head – but we are not cognitively equipped for that. So randomness is epistemic and it needs to be treated in epistemology (theory of knowledge). So the degree of randomness is observer dependent.

In TBS a turkey is fed for many, many days. Every feeding helps it confirm how favorable the human race is to its welfare … until a couple of days before Thanksgiving. Its non-feeding will not be a Black Swan event for the butcher – only for the turkey. So for me, the Black Swan is epistemic, having to do with confident but incomplete knowledge, “epistemic arrogance.” It has to do with being a turkey.

This definition is key as I equate randomness with human frailty in knowledge – it is not incompatible with religion or belief in agency. As I hinted, TBS is about how not to be a sucker – in things that carry consequences. In brief, the book is a map on how to live in a world we don’t fully understand – and enjoy it.

Q: What are you writing next?

A: I am interested in two subjects that were the same until recently: medical beliefs and religion. I am also interested in the boundary between the holy and the empirical, between the aesthetic and the functional. It will be a continuation of a few points in TBS but here again, avoiding commoditization.

Q: Do you hate Plato?

A: No, I do not hate Plato – I just used him (shamelessly) as a whipping boy. My targets are people who are both boring and wrong. Plato is not boring – and not always wrong. He has charm. He is my kind of playful and undisciplined fellow, but I had to concoct a term, “Platonicity,” after the idea of Platonic “forms” as a disease of knowledge.

We like to have maps – and seem to prefer to have the wrong map of reality to no map at all. For a long time we liked to have any medicine to no medicine – and medicine only started helping (not hurting) patients in the 20th century. We prefer any theory, even wrong, to no theory.

Theorizing is the default activity for our brain; suspension of belief is an active one. Because of the narrative fallacy, our minds default to theory making. It takes more conscious effort – and energy – to suspend beliefs. It also takes more training – we train children to find “explanations” instead of just teaching them to have the guts to say, “I don’t know” in certain circumstances.

In a way, it is Aristotle‘s pupils that I am after – the proponents of the superiority of “knowledge” over saying “I have no clue.” I have been trying to revive the class of skeptical thinkers who resisted Plato and Aristotle. I have also tried to revive pre-enlightenment thinkers – people who focused on the fallibility of human understanding.

Q: What about quantum mechanics and fake uncertainty?

A: TBS has an entire chapter on the uncertainty of the phony. It is the uncertainty of quantum mechanics, rendered inconsequential because it is mild and obeys very tame statistical properties.

It is key that one’s skepticism be limited to areas in which our knowledge is softer than others. TBS makes the distinction between two domains: Mediocristan and Extremistan. The bright Wall Street Journal reviewer David Shaywitz explains it as follows: “If 100 random people gather in a room and the world’s tallest man walks in, the average height doesn’t change much. But if Bill Gates walks in, the average net worth rises dramatically.” The difference between domains is right there.

So quantum mechanics is from Mediocristan. It is not real uncertainty – it becomes inconsequential under averaging. Uncertainty is what might happen in Pakistan, not quantum mechanics.

Q: Talk about Malcolm Gladwell‘s profile of you in the New Yorker, and its focus on finance.

A: Finance is invaluable for the following reason: it is the only place where I can pull my routine pranks without diplomatic incidents. For instance, I often call hotshots and leave urgent messages to call back “Arnaud-Pierre de Ceraisin from the Franco-American Royalist Friendship Society,” and people know immediately that it’s me and call me back. I also send investors e-mails in Arabic using a vocabulary that is not picked up by Google Translator. They fake enjoying the jokes. Publishers freak out at such pranks. As to academics…

You can also use finance to learn about the dynamics of history (though not the reverse). But Gladwell’s profile, while I thank him for it, had the side effect of putting me very narrowly in the finance box in the eyes of millions. This is very bad. It is like trying to describe my house and limiting the account to the bathroom in the basement while ignoring my library. Moreover, people think that if you are in finance, everything needs to turn into “what stock to buy” – as if I (or others) knew anything about stocks. Also, it simplified my “bet on rare events” or “hedge against the Black Swan” into “bet on volatility,” which is massively different and often contradictory. Italy is more volatile politically than Saudi Arabia; Saudi Arabia is more vulnerable to a big blowup. (Technical note: I am usually “short volatility” while “long tails.”) True, it would not have been possible to get the readers without such reductions. But I wrote my (unpublished) finance autobiography (15 pages) which has some boring details but should set the record straight.

Q: How can you tell if someone has actually read your book?

A: I immediately know if someone has read the book or not if they ask me if I could “predict five possible Black Swans.”

NNT’s Conclusion:

I admire the fact that Stephen Dubner invited me to talk about TBS while assuming that I would be hostile to Freakonomics. He took the risk while being aware of my “bluntness” (in print, anyway – I am overly diplomatic in social settings). But in fact, I consider Freakonomics one of the few works in empirical economics that is robust to consequential observation errors (i.e., Black Swans). And the results do not come from data mining (unlike, say epidemiology or quantitative finance). To conclude, thank you for your questions – and thank you, Freakonomics, for the hospitality.


egretman

While Taleb is obsessed with Extremistan, everyone else is making money in Mediocristan. He seems to have taken a good idea, randomness of extreme events, and followed it to it's illogical conclusion.

frankenduf

this 'definition' of random seems biased- since nothing is "completely understood" than everything is random- which segues quite nicely toward relativism, which I presume is the epistemic (and moral?) stance here- however, I don't think the proper use of random is epistemically loaded, so perhaps a different descriptor would be fair (unpredictable?)

bgriffs

That gentleman has some serious horsepower between his ears. Nice interview/questions.

egretman

Yes, he has clearly wandered into Epistemistan. Too bad he didn't name it eponymously. He could have occupied Talebistan.

Kent

It's ironic that Taleb named his fund Empirica since he flouts empiricism and the scientific method. Instead on blind faith he assumes that rare events are generally underestimated.

xanadu

But, what if he dies before his black swan comes in?

This strategy seems better suited for Harvard's endowment than for a short-lived human.

ImFromGuam

I use the Socratic method of reasoning myself. It's sort of like winning first place in an ugly contest.

egretman

But in fact, I consider Freakonomics one of the few works in empirical economics that is robust to consequential observation errors (i.e., Black Swans). And the results do not come from data mining (unlike, say epidemiology or quantitative finance).

This makes no sense whatsoever. If Freakonomics is not firmly planted in Mediocristan, then where is it?

Is this comment mere glad-handing for Dubner's sake? The blowing of smoke up the proverbial *ss?

BDD

This is one of the best blog posts ever.

Needless to say I am a huge fan of people who are smarter than I am, and hearing how he structured his book validated a lot of how i think about things. I also like having direct access to his thoughts. Thank you! This was a good idea!

egretman: read FbR. Everyone is making money... now, and when they lose money and blow up nobody will hear about them anymore. Survivorship bias. Rising tides lift all ships and all that.

Kent: more specifically, he's betting that humans are wired to underestimate rare events, which we are, if you are to believe any of the studies conducted on how we think as humans. He's betting against people, which I believe is a smart bet. He knows people would prefer (to use his example) to win a few pennies every day with a risk of losing dollars once in a great while rather than lose a few pennies each day and certainly win a lot but not often.

Xan: I'm sure he asks himself that question every day. We never did hear how he did after 9/11, though, did we? I'm sure he didn't make toooo much money...

Read more...

egretman

Everyone is making money… now, and when they lose money and blow up nobody will hear about them anymore.

I submit, BDD, that he is ignoring his own advice. He is "...long tails". That says it all. He basically is betting on Black Swans which he says are random and unpredictable.

So he is ignoring his own advice. In the meantime, Mediocrastan is making all the money.

coolhandluke

What Taleb says in the black swan is not that it is wrong to concoct stories, but that it is wrong to concoct stories and make it seems like events in the past were easily predictable, or that by using the stories in the past we will be able to predict the extreme events of the future. Take for example the abortion/crime issue. Freakonomics did not attempt to say that the link was totally predictable, rather to just explain why crime fell in the late 90's.

I believe what he would hate is Levitt's assertion that he can somehow find a method of accurately predicting terrorist attacks by monitoring peoples private banking. This is the modern day version of hunting for the holy grail. I think he would also disagree with the use of prediction markets, as they pertain to events of Extremistan randomness. I missed the the question post but would much apprecate NNT's comment on these subjects.

Read more...

711buddha

Thanks - great feature. Ordering the book now.

egretman

...but that it is wrong to concoct stories and make it seems like events in the past were easily predictable

Yes, but doesn't this just have a touch of the "duh-factor"? I mean he defines a Black Swan as random and unpredictable and then says that you can't predict them before the fact.

And then to top it off, he turns around and starts a hedge fund that bets on the coming of Black Swans. Huh? He's betting on the random and unpredictable to be predictable?

perthian

Taleb has got the strategy right - too many people and hedge funds out there betting on the predictable patterns to remain predictable and hence discounting the unpredictable

coolhandluke

I think Taleb's philosophy on investing is to try and invest in lots of random longshots, not on any singular thing in particular. Simply put he is betting on group of things he thinks one of which could turn out to be a black swan, he doesn't know which one will work out, only that it's impact will be large enough to make up for all of his misses. His view being that he people underestimate the probability and impact of extreme events.

egretman

too many people and hedge funds out there betting on the predictable patterns to remain predictable and hence discounting the unpredictable

Yes, but is predicting the unpredictable just predictably betting on the predictable? In other words, isn't Taleb just firmly planted in his own Mediocristan? But one that is unpredictable.

Meanwhile, the real Mediocristan is pulling in all the money, while Taleb has his investors anchored in Extremistan hoping that the Chinese curse "May you live in exciting times" comes true.

Isn't this just the stupidest investing strategy of all time?

rcousine

Egretman: I am but a lowly toiler in the field of nerd-human relations, but i believe NNT's investment philosophy could be summed up thus:

Rare-Event Futures are underpriced.

So, if you can buy enough of these rare-event futures and package them into a fund, you can beat the market. But you'll do so by losing a little bit of money most days and making a lot of money on one or two days.

damienladwa

After hearing NNT's Radio 4 interview 30/06/07 I've been compelled to investigate further into the man and mind. Recently I've come to 'act out' my life in a way that concurs with what I glean as one of NNT's underlying theories...The benefits of acknowledging and harnessing the Power of the Black Swan.

By 'Unlearning' our conventional teachings and tapping in to the inevitability of random events we are much better equipped to react and profit. In addition, once we have begun to understand that 'system of randomness' exists (ie that life is predictably unpredictable) we will be able to pass on the impact of unprobability through our own considered actions....Not so much 'Change the World' but definitely 'Make a ripple in the Pond"

I also put forward that a Blogment from an Unscholarly Fashion Sales Exec yet to read NNT's Black Swan is thoroughly random amd would strike a provocative chord.

Read more...

Jeremy

The Australian Share Market fell about 5% today and some commentator on the radio made a quip about this being - statistically - a 1 in 10,000 year event.

Shame he never read Black Swan!

I wonder, egretman, are all the people in Mediocristan still making the money?

One of the things I seem to recall from TBS is the notion that a single black swan event can wipe out 100 years of profits... I gather that happened to a few people here in Sunny Australia today. And possibly The Big Apple, too...

I'm not economic or trading genius, but it seems to make sense that with increased complexity and non-linearity as we are experiencing in modern society, that there should be more black swans. So it makes a lot of sense to run a fund that tries to get in the way of as many of these black swans as possible.

Actually, the analogy reminds me a little of quantum mechanics in that under both you can predict some kind of overall outcomes (there will be rare/unexpected events or a certain number of electron will tunnel through this junction) but not individual ones.

It's just a different, fuzzier way of thinking about the world... but once you start to look at things this way, you can actually relax, not take yourself so seriously and have a lot of fun.

Read more...

Dr Charles Parker

Taleb's refreshingly correct observations about reductionistic thinking and the limiations of statistics in the assessment of science and reality ring so true in everyday life. His remarks about medicine are locked-on correct.

Nowhere are these problems more evident than in psychiatric practice, heretofore the stepchild of medicine. With abundant hard brain evidence we have slightly closed that antique door on the "beliefs" associated with the "soft evidence" of dreams and fantasies. But the tunnel remains.

So many we see in our offices fall out of the G-curve, and just don't respond in the limited, expected way. The science trails behind the art. A statistically significant result on an "n" of 300 may please the FDA, but doesn't begin to cover the complexity in everyday office practice. But those statistics become the "standard of care." Frightening.

And the public is happily out of the G-curve whilst the "literature" keeps seeking it's value.

One variable not addressed in these excellent and pleasantly irreverent blog comments is the parameter of *time* -so interestingly discussed in my favorite book until TBS: Science and Sanity, by Alfred Korzybski.
Time flies as a companion with the Black Swans.

Time brings change. Aristotle and Plato looked for labels and locked down concepts for the next next steps with reality. Their concepts, noted Korzybski, built a foundation of science in which many remain insanely arrested. Labels lock time, as to time-bound words. The polemics in meetings attest to the need for Taleb, Korzybski and the Black Swan thinkers.

Interesting thougt in TBS: Black Swan can be quite positive if you are thinking: maybe... Or "yes, and" -not "no, but."

Yes, quantum mechanics may have a grid too it [and appear somwhat Gaussian], but the coalescence of unpredictable variables there saves the day.

Read more...