SCIENTIST BEHAVING BADLY – An update

The book by Emanuel Derman, which I reviewed in an earlier post, was not well received generally by the readers that reviewed it on Amazon. That must have incited the venerated Nassim N. Taleb to throw in his intellectual weight (all four milligrams of it), in an attempt to come to Derman’s defense:

Here is what I wrote in my endorsement: Emanuel Derman has written my kind of a book, an elegant combination of memoir, confession, and essay on ethics, philosophy of science and professional practice. He convincingly establishes the difference between model and theory and shows why attempts to model financial markets can never be genuinely scientific. It vindicates those of us who hold that financial modeling is neither practical nor scientific. Exceedingly readable.

From the remarks here, people seem to be blaming Derman for not having written the type of books they usually read… They are blaming him for being original! This is very philistinic [sic]. This book is a personal essay; if you don’t like it, don’t read it, there is no need to blame the author for not delivering your regular science reporting. Why don’t you go blame Montaigne for discussing his personal habits in the middle of a meditation on war inspired by Plutarch?

 

Which incited me to come to the readers’ defense (and my own, of course) in the comments section:

Perhaps you want to read again the negative comments. The point is not that Emanuel Derman wrote an atypical kind of book. He also did that (sort of) with “My Life as a Quant”, which received more positive reviews. The points are that:

1. the book title is misleading. There is very little, if anything, about disasters allegedly resulting from confusing illusion with reality. Nor does he give one single example of a specific quant having confused illusion with reality. He give examples to the contrary, though: of himself and Fischer Black.

2. If Derman’s distinction between “theory” and “model” is supposed to delineate science from pseudoscience, then I don’t think it furthers his argument when he puts religion and Spinozan philosophy on a par with particle physics, and (implicitly) disparages entire branches of physics (like fluid dynamics, complex systems theory etc) as “just models”, merely because they don’t share the incredible precision of particle physics or quantum mechanics.

3. Emanuel Derman makes a mockery out of finance, by misrepresenting the CAPM and EMH, and by providing so ridiculous a “counterexample” to those theories that I am sure he would not dare to submit it to a refereed journal.

4. Practical? Modeling is all about being practical. About tolerating unrealistic assumptions if that still makes the model useful. About throwing away models that are “truer” in some impractical metaphysical sense, but involve so many parameters that they become useless. Absolute truths are for religion. Practicality requires making simplifying assumptions, as Derman himself persuasively recounts in “My Life as a Quant”.

 

Taleb left it at that (probably too caught up in his metaprobabilistic meditations). To my surprise, however, no less a man than the author himself (yes I’m speaking about Emanuel Derman) reacted to my comment:

The difference between a theory and a model is absolutely NOT the difference between science and pseudoscience, in my view. A theory is an attempt at an absolute description. A model is a comparison, an analogy. It could be a good analogy or a bad one. It’s still a model. Hence, Newton’s laws may not be 100% accurate, but they are a theory. The liquid drop model of the nucleus is pretty good, but it’s a model, comparing the nucleus to a liquid drop. Diffusion for stock prices is a model too, based on an analogy.

 

There are two things about his reaction that are highly revealing:

First, the fact that Emanuel Derman reads the reviews, including comments to reviews, and felt the need to respond. That may not sound very surprising, but what’s striking is to what he did NOT respond. Namely, in my own review I addressed Derman directly, asking him:

Here’s a question I’d like to ask professor Derman: Why should we believe the others make mistakes Fischer Black and you never made?

Pray tell,  professor: what reason have we to believe all the other quants do believe in the literal truth of their models?

 

Why didn’t he answer to that question? May I conjecture it’s because he doesn’t have an answer?

Second, I pointed out in my subsequent reply to Derman’s comment that my reference to the distinction between science and pseudoscience was a reference to Taleb’s review, rather than to Derman’s book. More specifically, I was criticizing this particular sentence in Taleb’s review:

He [Derman, JV] convincingly establishes the difference between model and theory and shows why attempts to model financial markets can never be genuinely scientific.

 

By emphatically denying that models are always in the realm of pseudoscience, Derman is correcting Taleb, rather than me! Now that’s pretty remarkable, because Taleb’s endorsement (including that sentence) is printed on the back cover of Derman’s book. Imagine that: words of praise on the back cover that show the reviewer has misunderstood a central point of the book! (Taleb has no qualms about misrepresenting his friends’ views – yes, Derman is one of his friends – see also my post in which I show how he misquotes Daniel Kahneman, also a friend of his.) Why would anyone allow that? May I conjecture that it’s because Taleb’s name is more important than his substance (all four milligrams of it)?

This is my reply to Derman in full:

Thanks for pointing this out. My distinction between science and pseudoscience was a reply to Nassim Taleb’s comment that “attempts to model financial markets can never be genuinely scientific”, rather than a reference to your book. But I still see no reason to distrust models more than theories (in your definition of the words).

Religion may be an attempt at an absolute description, but it fails miserably. Every religion contains beliefs that are at odds with science, like the belief that the universe was created in only 6 days, not more than 10,000 years ago. Models on the other hand, even though not literally true, can still be very useful (as you admit).

But the great advantage of models in science is that modelers know all about their limitations. Everyone who has ever tried to build a model, making decisions about the number of parameters, the data for calibrating the parameters and so on, knows that they should be cautious about having too much faith in them. They are the least likely to “confuse illusion with reality” (none of the quants I’ve met have professed a belief in the literal truth of a particular financial model).

Disasters in finance have not come from mistaking illusion for reality. They’ve come from recklessness, from unbridled optimism, from a lack of ethics, from the wrong incentives to the wrong people, causing them to be blind to the real magnitude of the risks they were taking, and to ignore the warnings some quants were making. On the other hand, single-minded attempts at “absolute descriptions” have led to much bigger disasters, like the religious wars that are being fought all over the world.

 

Dear professor Derman, it’s not too late to jump off that train. You’re already past Embarrassment, but the next stop will be Disgrace.

 

 

Leave a Reply

Your email address will not be published. Required fields are marked *


*