By Larry Kummer. From the Fabius Maximus website.
Summary: How to deal with risks dominates our headlines, usually driven by single-interest groups that see only their favorite threat. Statistician Nassim Nicholas Taleb’s latest work offers a way to identify the most serious threats facing us, and determine how much we should spend to fight each of them. It has received much attention. Is it useful?
A series of papers by Nassim Nicholas Taleb et al made a large contribution to our understanding of risk: The Precautionary Principle within the statistical and probabilistic structure of “ruin” problems. The main paper is “The Precautionary Principle (with Application to the Genetic Modification of Organisms)“, well-worth reading for anyone interested in GMOs or risk analysis. I will not attempt to summarize it here; it deserves to be read in full. I will point out one aspect of relevance to many of the key challenges of our time: how should policy-makers allocate funds to prevent or mitigate shockwave threats — potentially disastrous but of low or certain probabilities?
Excerpt: What is a “ruin” scenario, and how should we respond to them?
“We believe that the PP should be evoked only in extreme situations: when the potential harm is systemic (rather than localized) and the consequences can involve total irreversible ruin, such as the extinction of human beings or all life on the planet.
“A ruin problem is one where outcomes of risks have a non-zero probability of resulting in unrecoverable losses. …In biology, an example would be a species that has gone extinct. For nature, ‘ruin’ is ecocide: an irreversible termination of life at some scale, which could be planetwide.
“…Our concern is with public policy. …Policy makers have a responsibility to avoid catastrophic harm for society as a whole; the focus is on the aggregate, not at the level of single individuals, and on global-systemic, not idiosyncratic, harm. This is the domain of collective ‘ruin’ problems.
“…For example, for humanity global devastation cannot be measured on a scale in which harm is proportional to level of devastation. The harm due to complete destruction is not the same as 10 times the destruction of 1/10 of the system. As the percentage of destruction approaches 100%, the assessment of harm diverges to infinity (instead of converging to a particular number) due to the value placed on a future that ceases to exist.
“Because the ‘cost’ of ruin is effectively infinite, cost-benefit analysis (in which the potential harm and potential gain are multiplied by their probabilities and weighed against each other) is no longer a useful paradigm. Even if probabilities are expected to be zero but have a non-zero uncertainty, then a sensitivity analysis that considers the impact of that uncertainty results in infinities as well. The potential harm is so substantial that everything else in the equation ceases to matter. In this case, we must do everything we can to avoid the catastrophe.
“…If the consequences are systemic, the associated uncertainty of risks must be treated differently than if it is not. In such cases, precautionary action is not based on direct empirical evidence but on analytical approaches based upon the theoretical understanding of the nature of harm. It relies on probability theory without computing probabilities. The essential question is whether or not global harm is possible or not.”
———————— End Excerpt. ————————
Public policy implications of “ruin” scenarios
Taleb explains that “ruin” events must be defended against “at all costs …Because the ‘cost’ of ruin is effectively infinite …we must do everything we can to avoid the catastrophe.” This is operationally useless since there are many shockwave scenarios with ruinous outcomes.
By his theory, the relevant expenditure required is that required to “defend against” all of them. I will mention just two threats to illustrate this. First, the oceans are dying, with potentially ruinous consequences for humanity. See the Ocean Health Index; see the jellyfish warnings; see some of the many warnings about this problem.
Second, the Earth has been hit by asteroids and comets in the past, with ruinous consequences — sometimes exterminating the world’s major lifeforms. It will happen again. For details see these posts. Oddly, Taleb mentions the history of extinction-level events from asteroid and comet impacts, but does not mention why this kind of ruin event should not become a major public policy concern. It exactly meets his formal definition. (Similarly he writes about the odds of a Third World War, but does not discuss the “ruin” scenario of global nuclear war.)
After we’ve funded every ruin scenario “at all costs”, we have to spend more to prepare for the merely awful scenarios, such as earthquakes, tsunamis, pandemics (like the flu), and famines. Then there are more exotic threats, such as a reversal of Earth’s magnetic field, eruption of the Yellowstone supervolcano, continental destruction from other volcanoes, peak fresh water, and other shockwave events. We cannot fund vast sums on all of them.
How can policy-makers allocate resources across such a broad spectrum of threats? The recommendations of Taleb’s methodology provide less help than the simple and objective — albeit imperfect — the usual comparisons of probability, cost, and risk.
Tomorrow: Taleb warns us about climate change.
A super-volcano will erupt again, eventually.
Another perspective on risk
“Therefore I tell you, do not be anxious about your life …which of you by being anxious can add a single hour to his span of life? …seek first the kingdom of God and his righteousness …Therefore do not be anxious about tomorrow, for tomorrow will be anxious for itself. Sufficient for the day is its own trouble.”
— Matthew 6:25-34.
For More Information
The photo of Nassim Nicholas Taleb is posted from his website with permission.
For more of Taleb’s work see his website, especially papers in the Doing Statistics Under Fat Tails Program. Also see his entry on wikipedia.
Please like us on Facebook and follow us on Twitter. For more information see The keys to understanding climate change, My posts about climate change, all posts about shockwaves (high impact, low probability scenarios), and especially these …
- We are so vulnerable to so many things. What is the best response?
- Preparing for the future: should we be precautionary or proactionary?
- The first step to protecting the world from its many dangers.
See Taleb’s books, especially The Black Swan: The Impact of the Highly Improbable (2010) and Antifragile: Things That Gain from Disorder (2012).
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Nassim Nicholas Taleb is a favorite! No, I don’t slavishly agree when he contradicts my expertise. His opinions are well founded and literate. His pop books are favorites, and his collaborators are as profound.
He recommended The Science of Conjecture: Evidence and Probability Before Pascal by James Franklin that’s fundamentally affected my thinking.
We agree on the Pareto Distribution and its effect!
“The Black Swan was a laughable book.”. Hmmpf!
David M. Raup (‘Nemesis’, Extinction: Bad Luck or Bad Genes) was a neighbor and mentor. We discovered the Black Swan in epistemology before N. N. Taleb published. Connecting the dots is a bathetic meme. We realized that connecting the dots in an epistemological mapping is nonsense for the Black Swan lurking between every data. Benoit Mandelbrot said that “reality is fractally complex.”
I still say the only way to guard against all ruin scenarios is to get humanity off the planet. Forget climate change, forget pandemics, nuclear war, large meteor strikes, so on. Make sure all of our eggs don’t stay in one basket. Cost a lot of money and take a lot of time, but it’s the direction to move in. If we are worried about ruin, move faster I guess.
Oh? And which planet-without-ruin-scenarios will the migration select?
Because in space nobody can hear you scream? The universe is full of risks, there’s no safe place.
It’s even rumoured that, regardless of where we are, eventually everyone dies. Try PrecautionaryPrinciple-ing that away.
Thanks Brian, Tim. Certainly, there’s no safe ‘without-ruin’ place. My point was just that space colonization means that a disaster that wipes out life on earth does not wipe out humanity. In that sense, it’s another, extra roll of the dice in case of a ruin scenario on Earth; the space colonies possibly would be impacted to a lesser degree.
“…human actions would be considered harmful unless proven otherwise.” This came from a fisheries paper– Restrepo, V. R., P. M. Mace and F. M Serchuk. 1999. The precautionary approach: a new paradigm or business as usual? pp. 61-70, IN, Our Living Oceans, Report on the status of U. S. living marine resources. NOAA Tech. Mem. NMFS-F/SPO-41. My impression is that the idea was a reasonable approach that if you could not determine the cause of an apparent pollution problem it could justify experimenting by removing a source. This happened once for example, I have the reference somewhere, where they took out the phosphate, years later the adjacent waters lost their productivity. Not sure that directly applies, though.
There may be a complete analysis of its history, if not, there needs to be as it seems to have started about the same time as the climate stuff in the 1990s. It may be a poor justification of what basically is, as I told a graduate student some years ago pushing it as a fisheries policy, as one legal scholar wrote–guilty until proven innocent.
While humans are often responsible, the marine biology literature has many papers making the assumption of Restrepo., et al. Not very scientific, but may help explain some of the nonsense going on. Or it may just be a symptom. It has been often applied to commercial fisheries.
You can find discussions and debates on this very topic, with the same level of intelligence, the same stupid baseless assumptions and the same useless results, in any freshman dorm.
Einstein was asked, “What things were infinite?” His answer, “The universe and human stupidity and I’m not sure about the former.”
Wrong. The latter.
bobthebear was correct. It was the former.
Further research = About 8,520,000 results (1.81 seconds)
Start with –
http://quoteinvestigator.com/2010/05/04/universe-einstein/
Anyone who imagines Taleb Nassim has any knowledge about GMOs should go over to Reason and check out his disappearance when challenged to a debate on their vices and virtues by Ronald Bailey. Bailey, nonplussed by Nassim’s rhetorical cowardice, systematically proceeded to play both sides himself, illustrating at length how ignorant the highly educated such as Taleb really can be. Of cousre, we here already know that with regard to climatological modeling, but the same can apply to statisticians as well:
http://reason.com/blog/2016/02/22/nassim-taleb-calls-me-an-idiot-you-decid
Tsunamis are pretty destructive. Just have to be aware before they strike by getting to higher ground. I think if a huge asteroid was detected and coning this way, we have the technology to divert it’s course in time.
To me the real challenge is preventing those guys with the suicide vests getting hold of a nuclear weapon. (the terrorists).
You could paddle up the Hudson river in a birch bark canoe, and blow the whole NYC city up (or any city). That is my main fear.
…coming this way…
In approximately 5 billion years, the sun will begin the helium-burning process, turning into a red giant star. When it expands, its outer layers will consume Mercury and Venus, and reach Earth. What precautionary plans are we currently making for this event? Slowly developing our ability to expand ‘civilization’ beyond this planet?
Not sure we have that much time. Can’t remember all the details but doesn’t our part of the spiral arm pass through another galaxy/nebula in about 2 billion years?
We need to get a move on if we want to be sure to survive as we will have to spread beyond the spiral arm we are in.
SteveT
noaaprogrammer no we are not. Scientific modelling will show we can adjust the Sun’s output to the level consistent with CO2 levels on Earth 1800AD. Just give the scientists more money.
@ur momisugly J. Philip, 7;48 pm, ” I think if a huge asteroid was detected and coming this way, we have the technology to divert it’s course in time”.
How, ? a laser beam? A rocket faster than a asteroid travelling at what ? 36,000 kms /hr ? We do not not have that technology, sorry, my answer, Duck!
Hollywood saved us before, they can do it again…….they have CGI technology & Bruce Willis
J Phillip,
“I think if a huge asteroid was detected and coning this way, we have the technology to divert it’s course in time.”
Probably not. There is quite a bit of research about this. See these for excerpt (some great graphics), and links to learn more.
Asteroid Day: reminding us of the threat, pushing us out into space.
Three things to know about asteroids, certain death from the sky (eventually).
Celebrate Asteroid Day, about preventing death from the sky.
“””””….. I think if a huge asteroid was detected and coning this way, we have the technology to divert it’s course in time. …..”””””
So if TONIGHT’S 6PM news broadcasts that a huge asteroid (MILE in diameter) was detected coming this way (365 days to certain impact), just when (tomorrow or in 364 days) do you plan to activate the technology we have (NOW) to divert it in time.
Where in the 365 days transit do you plan to intercept the huge asteroid; and in what direction do you plan to redirect it, so that it misses us (THIS TIME) and doesn’t return in a more certain hit path in the future.
The further out you plan the intercept, the smaller the payload that you can get out there promptly, so the less deflection you can accomplish, and the less accurate your targeting , so less sure of needed deflection, unless you further reduce the deflection payload by providing active homing means for surer targeting.
From whence does this savior deflector get launched whther tomorrow of in 364 days, or whenever?
Sounds pretty gutsy to me. WHO has this technology (NOW).
G
Interesting hypothetical example you have there Allan, and such expeditious judgement .
So WHICH cop (there’s so many) killed which INNOCENT civilian (all crooks are innocent), and where/when did this happen ?? Well I’m talking about in North America of course; the venue where you recognized this tribalism.
Would that be in Canada, or USA, or Mexico ?? Seems like they have somewhat different tribes in each of those places.
So far we have seen at least one case, in which three out of six tribalists accused of murder of an innocent civilian have been tried (in a court of law) and found to not be guilty of any such crime, and they were evidently the alleged instigators, and the other three yet to be tried, seem to have been even less associated with the event than those three were; yet all six were found worthy of charging, by a very fair minded district attorney, who was of course without any bias in the matter, and simply doing her job as she saw it (from the perspective of her tribe of course).
I must have missed your comment about the now 7 1/2 years of determined instigation of hate based tribalism, directed right from the very top, and premeditated to destroy the state of relative harmony that had been won through decades, of trial and error.
With your clear understanding of our systems, you no doubt are just itching for us to endure another 8 1/2 years, of this fifth column destruction from within, just in case the destruction from without is unsuccessful.
G
By addressing a risk you change the risk – sometimes for better, sometimes for worse.
Diverse examples:
When DDT was effectively banned circa 1970 to “save the planet”, the risk assessment was utterly flawed, essentially false. The ban of DDT caused millions of deaths from malaria, whereas the benefits to the environment, if any, have been minor.
The trillions of dollars spent every year to “fight global warming” is another case of flawed risk assessment. The move to replace cheap, reliable fossil fuels with expensive, intermittent “green energy” systems will cause an increase in poverty and more excess winter deaths, and will do nothing to improve the environment.
We were in North Africa during the Arab Spring. People there were exuberant about the new possibilities for their countries. I tried to explain that real world choices are seldom black and white; that you don’t often get to choose between good and bad, you often have to choose between bad and worse. In too many cases across the Arab world, they tossed out rulers they thought were the bad, only to get new ones that were worse.
The police in North America have for hundreds of years sought safety in tribalism. When a cop killed an innocent civilian, the force defended the killer cop and whitewashed the killing. Standard practice was to shoot first – to reduce the risk to the cop – and then join ranks in the cover-up. This tribalism worked for the police until recently. But tribalism begets tribalism. Now it is “you get one of mine, I get two of yours.” The cops who recently murdered two black men incited the killings of the five cops in Dallas. Again, by addressing the risk they changed the risk – very much for the worse.
In all of these cases, there was a fatal flaw in the risk assessment. Minor or non-existent risks were escalated into major crises that required immediate and costly actions, and the consequences of those actions turned out to be far worse than the original risks. The common thread in all these diverse reactions to alleged risks has been fatally flawed risk assessments, incitements to hysteria and over-reaction to the threat, to act in haste… and repent at leisure.
Regards, Allan
Well said, Allan, and good examples. Sometimes the problems are not as simple as the facts made apparent to us. There may be hidden dimensions to each. We need to approach all such problems with the notion that we don’t know everything about each problem. We need to produce solutions cautiously, ever aware that we may need to dump one solution and acquire a new one as we learn what works and what doesn’t. This requires a lack of ego and attachment.
Take for instance the Arab problems. If groups like the CIA and SIS destabilize situations and push lies into a culture to stir things up, all the news media sees is the unrest they want us to see. And what if agencies no longer work for the people they are supposed to serve, but the self-proclaimed elite who have so much money they could bribe a priest into all manner of sin.
In some respects, it’s all a matter of marketing — find out how to play people’s egos in order to motivate them to move on the chessboard of life toward checkmate. What’s the end game? If I were a rich psychopath who knew the difference between right and wrong, but didn’t care, I would want the whole enchilada — the entire planet. I would want to reduce the population by 95%. I would only need a few hundred million serfs to do my bidding. Ironically, these clever creatures of sophistication know marketing so well, they know how to keep people from looking into conspiracies. They merely make the topic look “kooky and rare.” They make American wars look like “peacekeeping actions.” And they turn intelligent discernment (discrimination) into evil prejudice. In a game of 3D chess, the one who always thinks only in 2 dimensions will remain disadvantaged.
Made more deliciously ironic by the fact that DDT is a repellant (disables the antennae, causing disorientation, and hence flight), not an insecticide. (Spray DDT and search for mosquito corpses!)
Hmmmm. how are we expected to know what’s happening with the global sea ice extent?:
http://nsidc.org/data/seaice_index/images/daily_images/S_timeseries.png
Pretty sure it’s worse than….. (all together now!)….eeeeeevaaaaah!
Taleb’s your standard warmie, with a few catchy buzz words and an aptitude for columnising and boxing abstractions to make opinion look like scientific determination.
I’m so over all this expensive fluff from the likes of Taleb and Kahan, who use a veneer of skepticism as a false door into consensus. It’s “ruin”.
Taleb’s work is interesting and shows some promise, but he falls prey to the GiGo principle — garbage in = garbage out. In the climate portion he unwittingly buys the lie that carbon dioxide is a pollutant and says that we need to reduce CO2 no matter what the models say. Dumb!
CO2 has never been a pollutant. The only reason it was ever declared to be such is because warming alarmists said that it causes evil global warming. But global warming is good. So, carbon dioxide as a pollutant is a lie within a lie. It is garbage’s garbage (garbage^2)!
Perhaps the only protection against a flawed approach to problem solving is to be humble enough to listen to criticism. But ego seems to be in vogue these days. Perhaps ego is the real enemy of us all.
In brief: We’re doomed!
We have been since birth
NOBODY gets out alive !!
According to the Christian Bible two people didn’t actually die.
First was Enoch (Methuselah’s dad)…”He walked with God and he was no more.”
Then, Elijah vanished in a “chariot of fire.”
Taleb said: “precautionary action is not based on direct empirical evidence but on analytical approaches based upon the theoretical understanding of the nature of harm. It relies on probability theory without computing probabilities. The essential question is whether or not global harm is possible or not”
Probability without some sort of evidence? Wow! It’s models all the way down.
A nearby supernova could sterilize the planet. Prevent it! Hurry!
Brian,
That’s an excellent example of a “ruin” scenario of absurdly low probability. Thanks for suggesting it; I’ll use it.
I’ve read that a supernova within 50 light years would destroy most advanced life on Earth. First, global warming (the real thing) then the wave of high energy particles. It’s certain to happen — every few billion years or so.
Taleb’s next article might urge that we prepare real soon!
Very briefly here, risk is the product of probability and loss:
https://www.researchgate.net/post/There_is_a_definition_of_risk_by_a_formula_risk_probability_x_loss_What_does_it_mean
If the probability of occurrence approaches zero and the consequence of the event approaches infinity, you have the zero x infinity quandary. The product is indeterminate, for example:
http://tutorial.math.lamar.edu/Classes/CalcI/LHospitalsRule.aspx
Note typo in link. It should be L’Hôpital’s Rule.
However, this reference applies L’Hôpital’s Rule to obtain zero:
http://www.ditutor.com/lhopital/lhpital_rule.html
All of the foregoing is called the quagmire of marginal statistics.
It’s said that we should multiply the cost of certain events, x, with the probability of its occurrence, p(x).
Let p(x) = f(x)/N. Let’s take the sum of products, SUM (x * p(x)) = 1/N SUM (x * f(x)) = MEAN (x). This is the way we compute the mean or expected costs if we know the probability distribution p(x). If the distribution were Gaussian, we would have in the right tail extreme costs to be multiplied with ever shrinking probabilities, in the end infinite times zero, so to say. The mean of a Gaussian distribution is finite, isn’t it? If it were true that for infinite high costs we would get infinite values of the products, the distribution, beginning with x=0, should have an infinite mean. What kind of distribution has an infinite mean? Could it be that Taleb made at the end of his career some mathematical errors?
Taleb:
Then you have a problem, because there is infinitely many subjectively non-zero probability ruin scenarios. But this is exactly what Taleb says. That kind of analysis won’t fly.
It is also a blatantly incorrect analysis. PP can’t be applied ‘at all costs’ on marginal risks in real life.
Nassim Taleb, so much to do. Yikes! …Reminds me of a poem in
my childhood by A.A Milne of a shipwrecked sailor …
‘So he thought of his hut…and he thought of his boat
And his hat and his breeks and his chickens and goat,
And the hooks (for his food ) and the spring (for his thirst)..
But he never could think of which he ought to do first…’
Bjorn Lomberg is much more balanced than T regarding risk
policies that is, ASSuming that Co2 may be a problem down
the track.
https://climatenuremberg.com/2015/10/01/found-in-translation-internatl-reax-to-hughes-et-al-s-shenanigans-p
Oops! Links got crossed. Enjoy the hockey stick satire while I find
the post I thought I was linking with.
Herewith:
https://stopthesethings.com/tag/bjorn-lomborg/
Fine link. Benefits of fossil energy for Africa are 50X costs, all-in.
I had an old aunt who saw devastating risks everywhere, if here advices had been followed, nothing would have been achieved. By inductive reasoning paranoid minds can imagine many devastating risks for humanity. One thing I know for sure about paranoia is that the costs are enormous.
Fundamentally, the PP Precautionary Principle is a device called upon by researchers whose work is incomplete, but judged by them to be so good and important that there must be a mechanism to recognise the greatness. That mechanism is the PP.
In some cases an emerging PP is avoided by further research that fills in the story to the stage where remedial action becomes no longer needed. However, some claim a type of mechanism that seems to have no possible remedy apart from the PP. Two main reasons are that research will take so long that the matter will be upon us before a solution is found; and that it is not possible to do experimental work like duplication of results because of the nature of the problem, which has to remain open ended.
Now I suspect that some researchers are purposely contriving scenarios that they know have no presently known outcome possible, in order to invoke the only alternative, the PP. Hence its prominence and ‘official’ definition in environmental literature. This can be seen as a cynical attempt to grab those 15 minutes of fame.
Has anyone compiled a list of cases where the PP has been applied, with a result that is better than the status quo was; and that proceeded down the predicted course, without unintended consequences? Some might say ‘Montreal Protocol’ but the jury is still out on that case.
Just be wary of the intellectual dishonesty, anti-science and laziness of the PP, especially in current trendy phases like ‘environment protection’ and ‘sustainability’ which seem to operate on the verge of invoking the PP for just about anything under examination.
I agree, but I’d go further. The so-called Precautionary Principle is a lazy totalitarian’s vehicle for grabbing power to control one’s own compensation, job security, status, to advance one’s cause, or even to impose a command economy more generally. It’s modern use seems to be philosophically indistinguishable from “Post-normal Science.”
The PP is not a risk mitigation mechanism, which is the sensible thing to do with risks that bring benefits – say flying – but a risk avoidance mechanism, useful where risks either cannot be mitigated or cannot be quantified at this time.
Taleb absolutely misuses it as he thinks of it as a mitigation mechanism, which it is not.
Thus the PP has some useful applications, but it is extremely limited in scope. To qualify, you should be looking at a status quo versus a change that has unquantifiable or unmitigatable risks. If the risk are known or can be mitigated, then you can just do a cost-benefit analysis. That also means that as a starting point (i) the status quo has is broadly acceptable to people, and (ii) is widely accepted as being “true” (i.e. the facts of the status quo are not disputed). Thus it is possible to characterise the debate about say GMOs in Western Europe as a PP issue. The status quo is that we are well-fed and very healthy, and our agriculture and back-ups for food are stable. In developing nations, none of that is the case, so the PP is not very useful there.
The PP is however useless where there is no acceptable (or accepted) status quo. Climate change is one of these instance. The Alarmists claim that the status quo will lead to disaster, but the Sceptics say that the “solution” will be disastrous. And even the Alarmists admit that the change needs to be massive and so must come with massive and unquantifiable risk as we are so poor at changing things.
So we are not offered a choice between a status quo we are broadly happy with and change that comes with some benefits, but a choice between disasters. The PP simply cannot help with such decisions, and to use it for climate change is utterly wrong.
…Taleb explains that “ruin” events must be defended against “at all costs …Because the ‘cost’ of ruin is effectively infinite …we must do everything we can to avoid the catastrophe.”. This is operationally useless since there are many shockwave scenarios with ruinous outcomes…
Ummm….
Surely…. this suggests that Taleb’s proposed defensive posture it itself a ‘total ruin’ event?
Which means that we ought to defend ourselves against it. But defending ourselves against it runs into an infinite loop…..
I know! We just ignore Taleb, Problem solved!
If you followed the PP to its full extent, you would stay at home wrapped in cotton wool and bubble wrap. I suggest all CAGW alarmists follow the PP to its full extent.
Taleb’s brain is a total ruin. Apply PP to reject the ruin brain who proposes ruin solution to ruin problem. When the cure is as bad as the disease, cure the doctor