Statistics

# Taleb Chastises Calm Journalist, Advises Precautionary Panic To Coronavirus

Background

Some blue check journalist named Ferris Jabr, in a raw display of sobriety for a member of the Indigo Cheka, said

None of this is to belittle what is happening. The outbreak in China is a genuine public health emergency.

But the essential data are still being collected and assessed. Sweeping and alarmist claims about unprecedented global threat are neither warranted nor helpful.

That strikes me as the exact right note, given the information we have so far. But Taleb, who recommends precautionary panic, wasn’t having it. He replied, “Another ignorant journalist dealing with risk matters. Irresponsible.”

Taleb and a couple of pals wrote a one-pager on how and why to best panic. They don’t say anything about medicine per se and instead focus on probability, which somehow has the answers. The note is “Systemic Risk of Pandemic via Novel Pathogens – Coronavirus: A Note“, about which more in a moment.

Our self-identified flaneur became famous for popularizing a well known incident in philosophy, the discovery of the black swan. Used to be logic was taught or illustrated with syllogisms that included the premise “All swans are white”. Then along came Australia which forever spoiled the fun. Logicians shrugged—an actual black swan does not affect the logic—and moved on.

Taleb was keen to appreciate that the great bulk of probability models began with a similar premise: “Let y be distributed normally” (this is the same premise used in most academic regressions). He pointed out this model greatly underestimates the predictions of many events, which according to the model had vanishing probability.

But Taleb went one step too far and began to believe in his own models. That is, he embraced the Deadly Sin of Reification. He began to talk and think as if events had probabilities, that his models (with their impeccable math) were realer than Reality.

Caution About Undue Caution

Enter the precautionary principle, which Taleb embraces with the strength of ten lovers. The PP says that if a devastating thing could happen, then that thing ought to be protected against, and that the level of protection should be proportional to the potential devastation.

I teased him about this once by supposing that Black Swans From Outer Space could attack the Earth and destroy it. Since this could happen, and the destruction of everything is infinite in value, there is no cost too large to pay to protect against this threat. Right?

Now here’s the coronavirus, which is a mutated version of the common cold. At least, that’s what we’re told. As of this writing (Sunday night), and quoting on the most reliable sources (Zero Hedge), the stats are “Coronavirus mortality rises above 5% with 76 dead on 1,423 confirmed Hubei cases.”

Given these numbers, and since the virus is a cold virus many cases will go unrecorded, since symptoms for most will be an ordinary cold, and most or many don’t go to the doctor to record their common cold. This would tend to inflate the mortality rate, perhaps by a lot.

What’s killing people, as I understand it, is that some frail people who catch this new coronavirus are developing pneumonia, and the pneumonia does them in. Most who get the virus recover fine.

Well, maybe that’s all wrong. I’m not a virus specialist, and I’m relying on published reports, which as we all know are not to be taken as oracular. I also don’t have infection rates of ordinary common colds at my fingertips and can’t say anything about the unusualness of this new virus. Seems at least similar to the spread of many colds. About videos of people dropping dead in the streets of China I am skeptical—but willing to be corrected.

In other words, I’m on the side of the journalist who urges cautious reporting. After all, every would-be media-hyped plague of the last half century has turned out to be a dud, globally speaking, which is evidence in favor of calmness.

Taleb’s New Note

We’re finally at Taleb’s note. He (and his pals) say “we are dealing with an extreme fat-tailed process”, which is an illustration of the Deadly Sin of Reification. Infections, like ebola, are more or less rare, but they happen because of causes not because they are “fat-tailed”, i.e. rare according to some evidence. When they are happening they are no longer rare, they are happening. The causes do not possess “fat tails” or probability of any kind.

The reification is confirmed when he says “Fat tailed processes have special attributes, making conventional risk-management approaches inadequate.” Rare things are rare because the causes of these things can’t operate because the situations are not usually right. Once the events are happening, the causes can be identified, and that situations that can stop the causes (stop the spread of infection) might be able to be identified.

Taleb makes this good point: “Historically based estimates of spreading rates for pandemics in general, and for the current one in particular, underestimate the rate of spread because of the rapid increases in transportation connectivity over recent years.” We are one big unhappy family now.

He spoils that point by adding “This means that expectations of the extent of harm are under-estimates both because events are inherently fat tailed, and because the tail is becoming fatter as connectivity increases.” The opposite is true. Events like infections will occur with greater frequency with increasing connectivity. He knows this, but seems to be mixing up consequences with rates. Even then it’s not clear consequences would become worse, because the connectivity may make it easier to solve problems.

Here’s a paragraph straight out of the PP:

Properties of the virus that are uncertain will have substantial impact on whether policies implemented are effective. For instance, whether contagious asymptomatic carriers exist. These uncertainties make it unclear whether measures such as temperature screening at major ports will have the desired impact. Practically all the uncertainty tends to make the problem potentially worse, not better, as these processes are convex to uncertainty.

That we don’t know what we don’t know is known, or should be, and is thus a given. But because we don’t know what we don’t know does not make what we don’t know bad. It could also be good, or benign. To say it could only be bad is the PP Fallacy.

They end their note by appealing to the PP and by asking we avoid fatalism. That the Chinese government has at least tried quarantine makes this a straw man. And I predict that if this thing does become seriously serious, and not just the usual panic and instead sci-fi come to life, we’d see the connectivity and mobility hard stopped. No flights to or from China and the like.

We’re not at that point, though. Maybe.

To support this site and its wholly independent host using credit card or PayPal (in any amount) click here

Categories: Statistics

### 19 replies »

1. The comment on Taleb believing his own models reminded me of Isaac Asimov’s Foundation series. The entire premise is that events have probabilities and a sufficiently detailed model can predict future events with a high degree of certainty. This is, of course, ludicrous, but makes for good science fiction. When things happen contrary to Hari Seldon’s “Plan” (read: Model) then it’s the world that is broken, not the model, and the world must be set back on track.

It’s backwards thinking.

Regarding the virus, I’ve heard unconfirmed speculation about it’s origin or deadliness, but your description of it as “the common cold that gives the elderly pneumonia” is also consistent with all the corroborated data i’ve seen. I remain unconcerned about it. Maybe i’ll just encourage everyone around me to get a face mask. Then i’ll be fine!

2. Bill_R says:

Being of an age and so in the frail category, I’ve laid in disinfectant wipes, gloves, and have gloves on hand. Just like I do every cold season. But then I used to work in this area (rhinovirus protection) back in the day.

A few sensible precautions are useful. Hair-on-fire panic isn’t

3. Sheri says:

““All swans are white”. Then along came Australia which forever spoiled the fun. Logicians shrugged—an actual black swan does not affect the logic—and moved on.” Really? Logic is dead?

The precautionary principle is the official method of controlling humans under a dictatorship. It is not science—it is bondage and despising of the human race. Embracing it makes you equal to a few brutal dictators in history that killed everyone they could just to make themselves supreme leader. The precautionary principle is the avenue to supreme leader, a way to destroy those you want dead and gone. (You don’t have to use disease or chemicals. Race, ethnicity, even hair color works fine. Just create something that can destroy the future and results in massive deaths and attribute that to the select group. Easy, easy, easy. A child could to it.)

770,000 die from AIDS/HIV globally and no one cares enough to change behavior and certainly no panic as one is banging whatever moves. Nearly 40 million worldwide have it and spread it. The whole epidemic thing with coronovirus is idiotic. If it was sexually transmitted, NO ONE WOULD CARE to stop it at all. 40+ dead is just a blip, nothing to get excited about. Get real, people. Death only counts if it doesn’t mess up anarchy and hedonism. As long as the “right” people die and the bad behavior continues, there is no epidemic and the number of casualties is totally and completely irrelevant.

Surprising the libs go along with this. It is quite obviously as RACIST as it comes. Those poor Chinese being singled out and shut out. Fairness says we must cancel travel into an equal number of white cities. Shut down LA and New York and quarantine the citizens. FAIRNESS DEMANDS IT.

4. Karl says:

Basically, I agree, but it is bogus to calculate mortalities based on fatalities and number of infected. Only after a patient has recovered or died can he be included in a sample to calculate mortality.

5. Uncle Mike says:

The mortality rate (deaths/1000/year) for all causes has been estimated by various entities. The “normal” mortality rate for China is thought to be around 7 deaths/1000/year in a population of 1.4 billion people. See

http://worldpopulationreview.com/countries/death-rate-by-country/

The city of Wuhan has a population of over 11 million, making it the seventh most populous Chinese city. If all those stats are correct, the “normal” mortality rate for Wuhan is 77,000/year or 211 deaths per day.

Reports are that ~70 people in Wuhan have died of the corona virus in the last week? or so, if reports can be believed. That computes to 10 per day in addition to the regular 211.

The US stock market took a dive today as investors freaked out of their shorts at the dire prospects of world wide mass death from this new strain of the common cold.

This proves that people are stupid and prone to mass hysteria, but we already knew that.

6. “Flaneur”

My word of the day!

Besides the excellent refutation of the arrogant Levantine’s usual pronouncements, a vocabulary building experience.

Thanks!

7. As always your posts are full of interesting thoughts. However, sadly, the Precautionary Principle is very poorly understood.

Matt, you say:
===
“The PP says that if a devastating thing could happen, then that thing ought to be protected against, and that the level of protection should be proportional to the potential devastation.”
===

This is either a really bad description or it’s just plain wrong. Please see the post below for the details. There are actually not a lot of situations that actually fit the Precautionary Principle. From that post:

===
Let me start with the birth of the Precautionary Principle (I’ll call it PP for short), which comes from the United Nations Rio de Janeiro Declaration on the Environment (1992). Here’s their original formulation:

*** “In order to protect the environment, the precautionary approach shall be widely applied by States according to their capability. Where there are threats of serious or irreversible damage, lack of full scientific certainty shall not be used as a reason for postponing cost-effective measures to prevent environmental degradation.” ***

This is an excellent statement of the PP, as it distinguishes it from such things as carrying umbrellas, denying bank loans, approving the Kyoto Protocol, invading Afghanistan, or using seat belts.

The three key parts of the PP are:

1) A threat of serious or irreversible damage.

2) A lack of full scientific certainty (in other words, the existence of partial but not conclusive scientific evidence).

3) The availability of cost-effective measures that we know will prevent the problem.
===

As you can see, this is quite different from your definition.

Other than that, Matt, a most fascinating, accurate, and apropos post regarding Taleb’s claims.

My best to you and yours,

w.

https://wattsupwiththat.com/2009/12/31/climate-caution-and-precaution/

8. Briggs says:

Willis,

Except for our definitions being the same, we disagree. The PP is being used much more widely than you know. Taleb uses it for instance to call for bans (or etc.) on GMOs.

Black Swan From Outer Space Attack:

1) Threat of serious damage. CHECK: all humans destroyed

2) Lack of full scientific certainty. CHECK: aliens could exist based on all sorts of knowledge; UFOs routinely reported; etc.

3) The availability of cost-effective measures. CHECK: shut down all electromagnetic communications at once; splash all satellites; hide in the dark.

There you are: the PP in action. The use of it really relies on the imagination more than anything else; who gets to decide on what counts as “scientific evidence” etc. See Tuesday’s upcoming post, too.

9. Matt, interesting as always. You say

“Willis,
Except for our definitions being the same, we disagree.”

No. Our definitions are far from the same. Your definition says nothing about cost-effective measures or about partial scientific certainty. Zero. So they’re not the same.

With that in mind, I’d say the following regarding the threat of a Black Swan From Outer Space Attack.

1) Threat of serious damage. Unknown, since we’ve never been attacked by black swans from outer space. Otherwise, ANY fear, real or not, could be said to have a “threat of serious damage”.

2) Lack of full scientific certainty. Absolutely not. Remember that that “lack of full scientific certainty” means we have PARTIAL scientific certainty, and regarding black swans from outer space, we have zero scientific certainty.

3) The availability of cost-effective measures. Again, not at all. We have no idea what their abilities or their weapons might be, so we have no ideas about what measures against them might be effective, much less cost-effective.

So no, it is NOT a candidate for applying the PP. It is far, far from fitting the criteria.

And that’s my point. Taleb and others are NOT applying the PP. They’re applying YOUR PP, which seems to be “When we’re really scared of something that might or might not be really dangerous, we’d feel better about ourselves if we did virtually anything proportionately grandiose” principle.

And that is the exact false PP that I’m arguing against, the “feel-good” version that is like Greta screaming “DO SOMETHING!”

Next, you say:

“The PP is being used much more widely than you know. Taleb uses it for instance to call for bans (or etc.) on GMOs.”

I’m more than aware how widely it is used. That’s the problem. That’s why I wrote the post, to try to narrow the times when it is used by emphasizing the original, rational Precautionary Principle rather than you and Taleb’s vague “Let’s take precautions about everything that might be scary especially black swan attacks from outer space” principle.

My best to you and yours as always,

w.

10. Briggs says:

Willis,

Problem with your definition is its inexactness. Too fluid and imprecise. That’s a problem for the PP, not you. What counts as “scientific evidence”? Who decides? Who tells us what counts as a risk? Sure, never before had we had a (physical) alien invasion, but we can imagine it, as many have. Never before have we had a global poisoning due to GMOs, though we can imagine it. But “imagine it” I mean spell out the consequences.

You seem to have in mind a very sober PP which takes evidence which everybody accepts, more or less, for propositions which are well defined, more or less, and which have costs which are specified, more or less. But, of course, there are disputes over the details.

That’s not the PP. That’s just ordinary decision making under uncertainty. And if it differs, except in degree, you’d have to be very careful to define.

The PP philosophically fails, as it can be and is used for any scenario. Like alien attacking GMO DNA. See the original Black Swan link for an example of how Taleb uses it.

11. Matt, a pleasure as always. You say:

===
“You seem to have in mind a very sober PP which takes evidence which everybody accepts, more or less, for propositions which are well defined, more or less, and which have costs which are specified, more or less. But, of course, there are disputes over the details.

That’s not the PP. That’s just ordinary decision making under uncertainty. And if it differs, except in degree, you’d have to be very careful to define.”
===

From my post, here are my examples differentiating the situations where the use of my PP is or isn’t appropriate.

===
“Here are some examples of how these key parts of the PP work out in practice.

We have full scientific certainty that seat belts save lives, and that using an umbrella keeps us dry. Thus, using them is not an example of the PP, it is simply acting reasonably on principles about which we are scientifically certain.

There are no scientific principles or evidence that we can apply to the question of invading Afghanistan, so we cannot apply the PP there either.

Bank loans are neither serious nor irreversible, nor is there partial scientific understanding of them, so they don’t qualify for the PP.

The Kyoto Protocol is so far from being cost-effective as to be laughable. The PP can be thought of as a kind of insurance policy. No one would pay \$200,000 for an insurance policy if the payoff in case of an accident were only \$20, yet this is the kind of ratio of cost to payoff that the Kyoto Protocol involves. Even its proponents say that if the states involved met their targets, it would only reduce the temperature by a tenth of a degree in fifty years … not a good risk/reward ratio.

Finally, consider CO2. The claim is that in fifty years, we’ll be sorry if we don’t stop producing CO2 now. However, we don’t know whether CO2 will cause any damage at all in fifty years, much less whether it will cause serious or irreversible damage. We have very little evidence that CO2 will cause “dangerous” warming other than fanciful forecasts from untested, unverified, unvalidated climate models which have not been subjected to software quality assurance of any kind. We have no evidence that a warmer world is a worse world, it might be a better world. The proposed remedies are estimated to cost on the order of a trillion dollars a year … hardly cost effective under any analysis. Nor do we have any certainty whether the proposed remedies will prevent the projected problem. So cutting CO2 fails to qualify for the PP under all three of the criteria.”
===

So I would disagree strongly that my PP (which is the PP as originally defined by the United Nations Rio de Janeiro Declaration on the Environment in 1992) is nothing but “ordinary decisionmaking under uncertainty” as you claim. If that’s all it is, then there was no need to invent the concept.

Best regards, and thanks as always for your good work and your most interesting blog,

w.

12. Briggs says:

Wiilis,

Yep, there was no need. And the thing has morphed far beyond the attempted definition of ordinary decision making under uncertainty.