Aug 25 2014

Nassim Taleb, The Precautionary Principle, and GMOs

Nassim Taleb is a serious scholar of risk assessment, especially in the world of economics. So when he and two co-authors published a paper on the precautionary principle as it applies to genetically modified organisms it is worth taking seriously. That does not mean I have to agree with his conclusions, however.

What I found was that Taleb’s argument is mathematically rigorous, although I think too absolute, but also is biologically naive.

Here is his argument in a nutshell: The mathematical part if this – if we consider risk prevention, we must decide how much risk is acceptable. As risk increases, tolerance should decrease. As risk approaches infinity, tolerance should approach zero.

In order to use this principle, we must quantify risk, so he further argues that any risk of permanent harm or total harm should be considered infinite.

A way to formalize the ruin problem in terms of the destructive consequences of actions identifies harm as not about the amount of destruction, but rather a measure of the integrated level of destruction over the time it persists. When the impact of harm extends to all future times, i.e. forever, then the harm is infinite. When the harm is infinite, the product of any non-zero probability and the harm is also infinite, and it cannot be balanced against any potential gains, which are necessarily finite. This strategy for evaluation of harm as involving the duration of destruction can be used for localized harms for better assessment in risk management. Our focus here is on the case where destruction is complete for a system or an irreplaceable aspect of a system.

First, I’m not sure I agree with his conclusion that no benefit is worth any risk of total harm, what he calls “ruin.” Let’s say, for example, that physicists were conducting an experiment on a process that could produce controllable fusion. If successful, this would mean the ability to generate massive amounts of power cheaply and cleanly, a total game-change for our civilization, and one that would reduce many other risks, such as climate change.

However, there is a trillion to one chance that the experiment will create a black hole that will suck in the Earth and end humanity. Should we take the trillion to one chance? I’d role those dice.

Such extreme and contrived cases aside, I can accept his general precautionary principle. If the risk is the destruction of human civilization, extinction for humanity, destruction of Earth, or the ruin of Earth’s ecosystem, we should essentially have a zero tolerance for risk.

I don’t agree with his argument that any permanent harm should also be considered infinite. My question is – does this apply to any harm, no matter how small as long as it’s permanent? Taleb also adds as a criterion that the harms are generalized and not local. So, if the risk spreads to the entire system, we should consider it infinite and his precautionary principle should apply.

Taleb further defines some systems as fragile, meaning they have a non-linear response to harm. For example, if you fell from 1 foot 100 times you would probably be fine, but if you fell once from 100 feet you might be killed. As height from which one falls increases, harm dramatically increases until it is complete (death). However, below a certain threshold there is no harm, and so small falls do not accumulate damage.

The two systems to consider with regard to GMOs are human health and the ecosystem.  Taleb argues that both of these systems are fragile, that the risk of harm from GMOs is global and not local, and that such harm has the potential to be permanent, therefore we should apply his precautionary principle of zero risk.

However, Taleb also argues that the risk to the environment increases as more GMO are introduced. He is assuming cumulative risk, however, and this is where I think his argument is biologically naive. Nature is a robust system with many homeostatic mechanisms. Potential harm from unintended consequences are likely to be local and finite, not global and permanent.

As as example, one he uses himself, the introduction of invasive species (which are far more risky than GMOs) can have a devastating effect on an ecosystem, but the system does not collapse. A new equilibrium evolves. To be clear, I am not saying there is no harm from invasive species. We should make every effort to minimize their introduction, but the results are not total ruin.

Introduction of GMOs are like falling 1 foot – they will not add up to a 100 foot fall.

I also think his argument simply does not apply to human health. There is no such thing as zero risk with human health. Taleb argues about scientific uncertainty, unknown risks, and dismisses the “all risks are non-zero” argument. However, I don’t think his logic here is valid. Even eating a natural plant has non-zero risks. The only practical system that applies is risk vs benefit and risk tolerance levels. The FDA, and medical science are based on having a sufficient margin of safety – but never zero risk.

Another factor I felt was missing from Taleb’s calculations is considering the risks of alternatives. Growing enough food for 7 billion people has consequences, in terms of land use, fertilizer, pesticides, and displacing natural ecosystems. GMO as a technology can potentially add to our efficiency. Banning GMO means relying more heavily on other technologies that may have even more risks.

In the end I found Taleb’s arguments to still come down to hyping the risk of unforeseen consequences due to the inherent limits of scientific knowledge. I don’t agree, however, that GMOs have the potential for global ruin. This is still largely based on a naive belief that transgenes are inherently risky, when there is no scientific reason to believe that they are.

While he makes a much more elaborate argument about risk than most, in the end it still comes down to how risky you think specific GMO species are, and what margin of safety you think is acceptable from scientific evidence. He failed to make a compelling argument that his principle of zero risk should apply to GMO.

54 responses so far

54 Responses to “Nassim Taleb, The Precautionary Principle, and GMOs”

  1. MikeBon 25 Aug 2014 at 8:34 am

    I had issues very early in reading this. From Taleb:

    “A way to formalize the ruin problem in terms of the destructive consequences of actions identifies harm as not about the amount of destruction, but rather a measure of the integrated level of destruction over the time it persists.”

    I’ve read this sentence numerous times and it still makes no grammatical–let alone logical–sense to me.

    I’m a lay reader, and I’m so tired of the tedium of trying to fathom anti-GMO arguments.

    On to the rest of the piece…

  2. Bronze Dogon 25 Aug 2014 at 8:55 am

    These same fallacious arguments seem just as applicable to conventional breeding. Whenever you breed a plant, there’s a risk of a harmful mutation and/or dangerous new combination of genes. If GM by gene insertion and GM by natural breeding bring ruin, clearly we have to rely on monocultured clones like bananas, which are always one nasty blight away from extinction.

  3. MaryMon 25 Aug 2014 at 9:14 am

    Yeah, the biological assumptions are naive, so to use that as a foundation means the whole thing is really baseless.

    But really–Monsantomania as a raison d’être for a paper on this? Truly strange. Calling out Monsanto specifically in this demonstrates the bias and a lack of understanding of the scope of the work on GMOs.

    Someone recently pointed me to a paper he did on religion too:

    I assure you religion has caused more death and societal damage than any GMO ever has or will.

  4. Bruceon 25 Aug 2014 at 9:23 am

    I have always enjoyed Taleb, despite him being a bit of a heavy read.

    This is a good example of how even when trying to appear to be putting cold hard logic onto something you can in fact end up just promoting an agenda.

  5. SARAon 25 Aug 2014 at 9:49 am

    Falling one foot 100 times would cause a human being (and indeed most objects) damage. Increasing damage.
    However, I’m not sure if that shows that the metaphor is poor or the premise is poor.
    I suspect it’s the metaphor.

    In any case, I think the problem is that when you try to quantify something as complex as a system and its response to a crisis or an event, you will have these problems. Its impossible to account for all varying vectors reactions to stimuli in a big system.

    Natural systems are complex, trying to simplify them creates error.

  6. Steven Novellaon 25 Aug 2014 at 9:57 am

    Sara – I assumed the metaphor of falling 1 foot means dropping one foot and landing on both feet. I don’t think this causing any cumulative harm. This is just like jumping in the air 100 times.

    Bronze – Taleb handles your objection by making a distinction between top down (GM) and bottom up (breeding) tinkering. I don’t think, however, that he adequately justifies his distinction in terms of risk. He states it almost as a given that bottom up tinkering will all work out due to natural mechanisms, but top down tinkering somehow won’t. I don’t buy this either. I think he is making a false dichotomy.

  7. _Arthuron 25 Aug 2014 at 10:01 am

    If he cannot tolerate any risk, he should not eat at all.

    What if the broccoli he’s about to eat is a toxic mutant ?

  8. Steven Novellaon 25 Aug 2014 at 10:08 am

    Arthur – he would call that a local, rather than global, harm. He did try to cover his bases.

    But I agree that a toxic mutant, or cross breeding result, could end up in the food chain with unstudied long term health risks. He does not cover this, except to dismiss it as “bottom up” – but that does not eliminate the types of risk he is talking about.

    In the end I felt he biased his arguments to arrange them in such a way that GMO would be singled out, but they don’t stand up to close scrutiny.

  9. Karl Withakayon 25 Aug 2014 at 11:07 am

    I find it hard to waste any time considering the positions of any anti-GMOer, including people like Taleb advocating the precautionary principle seriously, if they don’t address or even acknowledge the existence of mutagenic breeding.

    “Another factor I felt was missing from Taleb’s calculations is considering the risks of alternatives”

    Yep. Doing nothing / status quo count as alternative choices for which you must give equal consideration to risk. Where’s Taleb’s risk assessment on the need & ability to produce enough food to feed the world over the next 100+ years without GMOs?

  10. RCon 25 Aug 2014 at 12:33 pm


    I completely agree – everyone seems to ignore mutagenics, which seems way more dangerous to me than GMO, which is atleast controlled and planned and studied – mutagenics are a complete wild card (and organic – another reason why the whole organic thing is so silly).

    As regards risk, he also seems to ignore that every ecosystem shattering plant or animal we’ve seen (Kudzu in the SE USA, weasels/cats killing all the birds in hawaii, cane toads in australia, etc) are all examples of bottom-up.

  11. Calli Arcaleon 25 Aug 2014 at 2:33 pm

    In the hypothetical example of something with a 1-in-trillion risk of creating a black hole that sucks in the Earth, I think this illustrates something that I think he’s overlooking — the plausibility of the harm. That a possible harm would continue and grow over millions of years doesn’t necessarily make it an essentially infinite risk if it’s a harm that is extremely unlikely to occur at all.

    I didn’t read his original article, so I don’t know whether he addresses this, but from what you’ve quoted it feels like he’s arguing that if there is no benefit but a very tiny risk that extends throughout time from now to the future, we might as well consider it to be infinite risk. Well, by that logic we shouldn’t crush any butterflies, since there’s a miniscule risk of preventing some future benefactor of all humanity being born. If the odds of a particular risk are unquantifiably low, how can he argue that the longevity of a risk makes it effectively infinite?

    Or am I misunderstanding something here?

  12. BBBlueon 25 Aug 2014 at 2:43 pm

    It appears to me that Dr. Taleb is simply trying to put a scientific veneer on a non-science issue. PP is a matter of public policy, not science, and is used by many in an attempt to neutralize scientific arguments by demanding proof of the null hypothesis.

    Caution is great as a political sledgehammer. Carefully formulated, you can ban anything. But this is unreasonable. Here’s what politics should learn from kids crossing the street for ice cream.

  13. jsterritton 25 Aug 2014 at 2:45 pm

    Taleb’s paper fails by being so in love with its premises. First, that validity of the precautionary principle can be irreducibly determined by stuffing any set of variables into a a bunch of dichotomies: harm/ruin; Gaussian/ “power law”; bottom-up/top-town; thin tails/fat tails. Most of these dichotomies are redundant and little more than “expert-sounding” ways of saying the same thing (at least in the context of Taleb’s paper). Because of this, the authors do a poor job of defining their terms, establishing only that some of the terms are bad.

    Even accepting that these dichotomies exist outside of Taleb’s model, it takes a Herculean effort to leverage GMOs into the bad category, especially when considering the free pass Taleb gives to pretty much every other top-down intervention in human history, including nuclear power (he calls the Fukushima incident an “accident,” despite the rest of his argument about the PP presuming the inevitability of harm). In a nutshell, Taleb et al argue against GMOs and monoculture as if the two are one in the same. Moreover, this GMO/monoculture intervention is presumed to be global in scale, utterly contemporaneous, and perpetrated by a single corporation (Monsanto). The authors make no secret about it:

    “The systemic global impacts of GMOs arise from a combination of (1) engineered genetic modifications, (2) monoculture—the use of single crops over large areas.”

    As scientists (especially as economists), the authors should understand that some systems are too complex to make assumptions about — even assumptions with lots of math. Taleb makes mention of this himself, but uses it in support of his own argument. I suppose he is so enamored of his thesis that he fails to see the irony:

    “Nature is much richer than any model of it. To expose an entire system to something whose potential harm is not understood because extant models do not predict a negative outcome is not justifiable; the relevant variables may not have been adequately identified.”

    Taleb doubles down on his self-regard by failing to see how his description of what he calls “the pathologization fallacy” does an excellent job of describing his own paper:

    “Today many mathematical or conceptual models that are claimed to be rigorous are based upon unvalidated and incorrect assumptions. Such models are rational in the sense that they are logically derived from their assumptions, except that it is the modeler who is using an incomplete representation of the reality.”

    But you don’t have to look beyond several key misstatements and misrepresentation of facts to question the authors’ conclusions.

    “Health wise, the modification of crops impacts everyone. Corn, one of the primary GMO crops, is not only eaten fresh or as cereals, but is also a major component of processed foods in the form of high-fructose corn syrup, corn oil, corn starch and corn meal. In 2014 in the US almost 90% of corn and 94% of soybeans are GMO. Foods derived from GMOs are not tested in humans before they are marketed.”

    This statement is riddled with errors. There is NO evidence that GMOs impact health at all. Talking about corn and soybeans in the US undermines Taleb’s “global ruin” argument. For all of Taleb’s talk about fallacious reasoning, he is comfortable trotting out the “human testing” trope. He should know full well that NO existing food or ingredient (GM or otherwise) has been the subject of human clinical trials.

  14. Artur Krolon 25 Aug 2014 at 3:25 pm

    While I greatly enjoyed Taleb’s first two books, by the third (Antifragile) it became quite clear that he had succumbed to the Nobel disease… without actually winning a Nobel prize at that (he’d reject it for PR reasons anyway). It seems that the 2008 crisis, confirming his predictions, was enough to make him believe himself infalliable in any and every issue he turns his head to and it shows ever so clearly in both the Antifragile book and in his public activity.

    A true shame – a brilliant mind, brought down by pride.

  15. evhantheinfidelon 25 Aug 2014 at 3:40 pm

    The argument that Taleb made sounds to me like high school physics; it can be a good exercise, but is practically useless when applied to real complex systems. If he applied his reasoning consistently, he would descend quickly into overwhelming difficulties in any chaotic system.

  16. TsuDhoNimhon 25 Aug 2014 at 4:22 pm

    so he further argues that any risk of permanent harm or total harm should be considered infinite.

    Does he drive a car?
    Does he take commercial or private planes?

    If so, he clearly doesn’t believe his own BS, because those are infinitely risky by his standards.

  17. Enzoon 25 Aug 2014 at 4:23 pm

    Ugg. I get so dismayed when I see articles like Taleb et al.’s.

    I suppose its fine to write up a theoretical paper applying one’s expertise to see if it has anything useful to say about popular topics. We saw something like this with the mathematical model of zombie spreading. There the correlation is cute and obvious. But to do this with the reductionist “GMOs can be fit into the ‘total failure[/ruin]’ model” is not obvious and just transparently motivated reasoning. You have to force too many comparisons and drop too much nuance.

    What’s the point of an a priori argument against GMOs at this point? We have actual data on safety and a good understanding of the plausibility of harm. Not to mention all GMOs are not created equal. Why not start with the data and make a more convincing case for why the precautionary principle model applies?

    Looks like GMO inclusion as an example had the intended consequence of winning the paper some publicity impact. Sigh — another reference that will appear at the end of websites that make my blood boil.

  18. papadopcon 25 Aug 2014 at 7:59 pm

    I have not yet gotten the chance to read the full article, but in the quoted excerpt I see a tendency to play loose with the concept of infinity.
    From one point of view, the reasoning has the same flaws as Zeno’s paradox, where again the concept of infinity is misused.

    Case in point, from the quote given:
    “When the impact of harm extends to all future times, i.e. forever, then the harm is infinite.”
    This is not at all obvious, indeed it is counter intuitive.
    Future harms have to be discounted, and usually by an exponentially decreasing function, and so integrating over an infinite time should still give a finite value.
    If they’re using some other discounting function, or none at all, they have the burden of proof on them to show that their assumption is valid.

    Of course, the whole “all future times” part assumes that such a thing as infinite time exists, or is meaningful for this case.

    To give an example of how absurdities can arise, consider the effect of UV radiation on life.
    Since every year some small percentage of animals die from cancer caused by this, and it will continue for “infinite” time, then the risk from the Sun is infinite.
    There are a bunch of obvious problems with this, as there are with the GMO arguments:
    1) The sun won’t last forever. True, and this applies to GMOs too
    2) There may be some some benefits. True, and the benefits would indeed be “infinite” if we follow the article’s logic
    Hence, UV rays pose infinite risk! Maybe we should turn off the Sun?

    I will have to read the article more carefully to see if any of these points are addressed, but the wording of the excerpt is very disconcerting…


  19. papadopcon 25 Aug 2014 at 8:24 pm

    To boil down the previous comment, I find these things wrong with the quoted way of thinking (and I think others do too):

    1) Absence of discounting function
    2) Claims about “all future times” that are meaningless when applied to something like life on earth
    3) Ignoring the potential benefits, which are also “infinite”, if we follow the same line of thinking.
    4) Playing fast and loose with the concept of infinity. I would add this to “quantum mechanics”, “energy” and “field” to list of most misused concepts.

  20. grabulaon 25 Aug 2014 at 9:07 pm


    “These same fallacious arguments seem just as applicable to conventional breeding.”

    This was my thinking too. One of the things we’ve been hammering guys like mlema about is the fallacious assumption that just because we might be pro GMO doesn’t mean we’re not aware of the risks. It’s the assumption of reasonable risk versus gain we look at and so far the science is indicative that the risk is worth it.

    The other problem I had was that he implies the eco system is ‘fragile’. In some ways it is, but like Dr. Novella points out, eco systems adapt, some of them pretty quickly. I’m not using this as an excuse to be reckless, only that to paint the ecosystem as fragile is in some ways misleading.

  21. grabulaon 25 Aug 2014 at 9:18 pm


    ” Foods derived from GMOs are not tested in humans before they are marketed.”

    This is that 100% replication problem anti-GMO types push. The only way to be sure it’s safe is to do human trials for long periods of time just to make sure. This completely ignores the fact that we have a pretty strong understanding of biology and chemistry and how they interact.

  22. grabulaon 25 Aug 2014 at 9:23 pm

    Enzo said:

    “What’s the point of an a priori argument against GMOs at this point? We have actual data on safety and a good understanding of the plausibility of harm. Not to mention all GMOs are not created equal. Why not start with the data and make a more convincing case for why the precautionary principle model applies?”

    Exactly. The problem is for some it’s a sacred cow, or more accurately part of their belief system – that natural products are the only safe products for human consumption. Anything that they feel is unnatural is outside of their understanding of the world and is therefore evil. The argument is so laden with irony (Big Agro versus Big Organic for example) it’s ridiculous. We’ve been asking the same questions of the anti-GMO types that frequent this board and you get no good responses. As with all woo types, they are self referential (most of their evidence is provided by motivated reasoning) and cherry pick outside of their demesne in order to further complicate and confuse the issue.

    It’s not a rational viewpoint, much like global climate change denial in my opinion.

  23. grabulaon 25 Aug 2014 at 9:26 pm

    After reading through this blog post and the pdf it really does appear to be highly motivated reasoning. Consider all of the obvious issues already pointed out but also that the risk versus reward argument seems to reflect the common attitude that the anti-GMO crowd has.

    1 – We ‘don’t know’ what might happen, maybe
    2 – we can’t see the future so have no idea of the risk
    3 – as long as there is any risk at all, it’s not worth it

  24. Davdoodleson 25 Aug 2014 at 10:57 pm

    “When the harm is infinite, the product of any non-zero probability and the harm is also infinite, and it cannot be balanced against any potential gains, which are necessarily finite. ”

    This doesn’t make sense to me.

    Why is a potential harm “infinite”, but a potential gain “finite”?

    Surely (even setting aside the qustion of why harm and/or gain couldn’t be temporary), using the same deductive process that says that any harm is “infinite”, any benefits gained should also be infinite?

    Indeed, I’d have thought that a potential harm would be more likey to be “finite”, as it would be the subject of on-going attempts to remedy it, whereas a potential gain would, in contrast, presumably be actively maintained.

  25. Davdoodleson 25 Aug 2014 at 11:01 pm

    @ Enzo: “What’s the point of an a priori argument against GMOs at this point? We have actual data on safety and a good understanding of the plausibility of harm.”

    The cynic in me suggests that your second sentence answers your first 🙂

  26. Mlemaon 26 Aug 2014 at 2:32 am

    Lack of knowledge on GMOs doesn’t keep anybody from having an opinion on them. But, when somebody has an opinion on GMOs (plural) – we can usually dismiss that opinion easily as not being scientific, since we know that there’s a broad range of GMO risk/benefit, and absolutes don’t exist. So I would say that’s the main problem with this paper. That and I don’t see a comparison to nuclear energy.

    For example, we have relatively low risk agrobacterium-mediated gene transfer between closely related species for the purposes of disease-resistance. With appropriate testing, arguably safer than conventionally-bred, although have been seen to increase vulnerability to other diseases, so benefit may be questionable long-term compared to other methods. And we have at least a million genetically-modified poplars, engineered to produce bt (a pesticide), already cross-pollinating wild varieties in China – it’s pretty easy to see that risk gets high and serious.

    And with all the permutations in between: varying levels of risk based on methods, species, traits, applications, ownership and regulatory practices – it’s pretty clear that determining risk of gMOs is an impossible proposition. As it is, we have to just hope that those who are in control of development and regulation (often the same people) employ precautionary principles. Since these controls vary from country to country and era to era, well, who knows? But of course, anyone who judges to one end of the extreme – that being “GMOs are safe” – won’t see any subtle aspects of this situation. And won’t recognize their thinking in any of the fallacies listed in the paper.

  27. RCon 26 Aug 2014 at 9:42 am

    Mlema, nobody is saying GMOs are completely safe. They’re saying that they’re not any more dangerous than plants produced any other way.

  28. BillyJoe7on 26 Aug 2014 at 10:00 am


    You are missing a bit of history on this blog.

  29. Bronze Dogon 26 Aug 2014 at 11:17 am

    As I see it, the real problem is anti-GMO people deny that plants genetically modified by good old fashioned breeding share almost exactly the same ongoing risks as today’s GMOs. They’d rather roll dice in blissful ignorance than let humans manage the risks we take.

    One irritating trope that results from this double standard is that they end up depicting nature as sterile and servile.

  30. daedalus2uon 26 Aug 2014 at 3:12 pm

    I agree with Artur Krol, that NNT does seem to be unjustifiably sure of himself.

    His analysis of the GMO case is wrong. I mentioned a number of mistakes, in particular as they apply to maize in Europe. There are no native plants on the European continent that maize can cross breed with. Maize is so domesticated that it requires humans for propagation. Maize cannot become a wild and invasive weed. Without human intervention, it would be extinct in 2 seasons. Maize is also grown as hybrids, meaning that farmers don’t save seed, new hybrid seed is made and planted new every year.

    One could use the precautionary principle against blasphemy. If there was a God or gods, that were sufficiently offended by blasphemy, that God or gods could destroy the entire Earth. That was essentially the argument against witches.

  31. tkitzleron 26 Aug 2014 at 8:25 pm

    I absolutely agree with Dr Novella’s article. In my opinion NNT makes a completely theoretical argument, void of any practical relevance. He applies a mathematical concept without taking into account the natural forces at work. He does not take into account any of these forces because he does not know much about biology nor the homeostasis of ecosystems. In the end his proposal is as good as making predictions about the weather based on two variables; two variables that have not even been derived through measurement. The perfect example of a mathematician blinded by the beauty of his model, ignoring the “ugly” reality around him. Like Thomas Huxely (a biologist) said, “That is the great tragedy of science: the slaying of a beautiful theory by an ugly fact.”

  32. grabulaon 26 Aug 2014 at 9:06 pm

    “You are missing a bit of history on this blog.”

    LOL, understatement of the year.

    @RC – we’ve fought an ongoing battle with mlema to make him understand most of us pro-GMO skeptics are aware there is always potential risk but that the science is showing the rewards are worth it. We’ve also tried to get through to him that we also believe in ongoing monitoring and responsible science. The problem is when you have a bias so strong, it outshines any science you might be presented with

  33. geopaulon 27 Aug 2014 at 12:40 pm

    The argument of cumulative risk over time also has to assume a closed, statics system. In other words, whatever the risky negative outcome is, that outcome is the same over all time, therefore the integral of total damage becomes infinite. In other words, no process (whether by outside intervention, changes in environment or selective evolutionary pressure) will change the system in response to the risk outcome. I don’t see how this assumption holds up.

  34. Rogue Medicon 27 Aug 2014 at 3:07 pm

    When it comes to financial risk, Taleb is brilliant, because the amount of financial risk is horribly underestimated. In his books, he provides examples of the banks producing disasters on a regular basis. Do not trust a banker, because a banker probably does not understand risk.

    Applying the same criteria to the environment is a mistake.

    What if the failure to use technology to provide for the well being of the poor, in this case GMOs, leads to an over-populated nuclear nation starting a nuclear war because they see all other options as more destructive to that nation? What if this happens to more than one nuclear nation?

    Where does Taleb address this non-zero possibility? What limits should we place on the possible actions of starving people with nuclear weapons?

    What if starving people have chemical or biological weapons?

    What kind of destruction can starvation produce in a technologically advanced world? The answer may be a much greater catastrophe than Taleb imagines.

    Why should we assume that this kind of artificial selection is more risky than natural selection or more risky than other forms of artificial selection?

    Science is the way we learn how the world works. If we do not try things, we do not learn.

    Science is far from perfect, but without the advances of science we would be much worse off. We might be living in conditions that are not much better than his worst case scenario.

    For another example of how Taleb applies his logic outside of his bailiwick, look at his participation in a debate on whether religion is good for society.

    The introduction (the first 6:45) is in Spanish, but the debate is in English.

    Hitchens, Harris, Dennett vs Boteach, D’Souza, Taleb – the pro-religion group could have come up with mucgh better people to present their side, but it does give you a sample of how much of a blind spot Taleb can have when he is outside of his specialty.


  35. Rogue Medicon 28 Aug 2014 at 11:53 am

    There is a bit of a discussion of the precautionary principle by Brian Schmidt. He begins to discuss the precautionary principle at about 52:45. The whole speech is worth listening to.

    Highlights of the 2014 Kenneth Myer lecture; Science and society: Exploring the role of research in Australian lives.


  36. FosterBoondoggleon 28 Aug 2014 at 4:35 pm

    I’m curious (though not enough to really investigate) whether Taleb is devoutly religious. His reasoning in the GMO paper seems like a straightforward translation of Pascal’s Wager: if heaven & hell exist as described in the Bible and unbelievers are condemned to eternal damnation while believers conversely receive eternal reward, then even if the odds are tiny of this being the case, the loss of being an unbeliever is infinite, as is the gain from being a believer, while the cost of the latter is finite. Therefore one must be a devout Christian. (Or in Taleb’s case, perhaps Muslim. Though how would he decide which? Uh oh.)

    The parallel of Pascal’s argument to Taleb’s seems completely straightforward. No complicated math or sciencey typesetting required. So, observing that Taleb doesn’t have one of these prayer bumps ( on his forehead, I’m skeptical that he applies this reasoning throughout his life.

  37. Aardwarkon 29 Aug 2014 at 3:32 am

    I wish to recommend an excellent article on the subject of general analysis of problems related to extremely improbable, but (under some, more or less [im]plausible theory) existentially thratening risks.

  38. jsterritton 30 Aug 2014 at 2:20 am

    It is so very disheartening to see Mlema commenting here, trying to spin the general criticism of Taleb’s paper to the advantage of his (Mlema’s) weary “narrative” about GMO safety. Taleb tries to force GMO safety into some kind of mathematical framework wherein he proves a negative: absolute risk. Taleb uses false dichotomies and zealously redefines terms in his attempt to pull this off (specifically, he combines GMOs and monoculture into a single product/practice and then presumes universal and simultaneous adoption of same). He also uses infinities, which are themselves a red flag for bad equations. Then he moralizes.

    Mlema is famous in these pages for two things: very unmathematically rejecting science’s ability to assess risk; and a staggering ability to conflate business practices with GMO safety. I think Mlema must have felt his toes were being stepped on, so he one-upped Taleb by saying that “determining risk of GMOs is an impossible proposition…because there’s a broad range of GMO risk/benefit, and absolutes don’t exist.” Where Taleb would have us believe that science is capable of determining risk to a tipping point of sorts, Mlema neuters science completely, telling us science should not even address the subject, since absolute risk (or safety) is unprovable. In other words, Mlema’s criticism of Taleb’s flawed reasoning is that it doesn’t go far enough! Mlema prefers flawed circular reasoning to Taleb’s flawed linear reasoning. Taleb misuses the correct tool (science). Mlema thinks science is a hammer in a world without nails.

    At least Taleb, a scientist, understands that science works in probabilities. Sadly and wrongly, when it comes to GMO safety, Taleb finds his God in the gaps — the decimal places of uncertainty. Taleb rounds up to absolute risk. Mlema’s God is the gaps — he would have us forsake the 99.99 to the glory of the 0.01 and round everything down to zero. Mlema is a rejecter of science.

    “And with all the permutations in between: varying levels of risk based on methods, species, traits, applications, ownership and regulatory practices – it’s pretty clear that determining risk of gMOs is an impossible proposition. As it is, we have to just hope that those who are in control of development and regulation (often the same people) employ precautionary principles. Since these controls vary from country to country and era to era, well, who knows? But of course, anyone who judges to one end of the extreme – that being “GMOs are safe” – won’t see any subtle aspects of this situation.”

    These are Mlema’s common tropes. His logic is riddled with errors. Since he cannot conceive of all the “permutations” and “varying levels of risk,” no one can (this despite the fact that science is concerned precisely with “seeing subtle aspects” and “variables” — studying and accounting for them). Mlema reduces a science-based approach to acquiring and using knowledge — at least when it comes to GMOs — to some kind of “Hail Mary” play, where we all just “hope” everything works out somehow in the end (this is particularly insulting, as it places Mlema’s cobbled-together, cherry-picking, special pleading “gut” feelings above humanity’s very best efforts toward understanding). Then there’s the conspiracy thinking: In Mlema’s mind, “those who are in control of development and regulation” are “often the same people” (this is a baseless, but familiar, lie). And of course, there is the absolute equivalence between science (“anyone who judges to one end of the extreme – that being “GMOs are safe””) and those who reject scientific evidence, because they just know GMOs are unsafe. For Mlema, failure to invoke the Precautionary Principle (i.e, scrap GMO development) means that science and GMO proponents reject all caution. For someone who claims “there are no absolutes,” Mlema has no trouble inventing them for his ideological convenience.

    It just goes on and on….

  39. jsterritton 30 Aug 2014 at 4:26 pm


    Sorry for taking you to task like this, but your passive-aggressive approach (“arguably” this and “benefit may be questionable” that) is just as maddening as your other tactics. Your failure to understand science’s role in this matter continues to amaze me, as does your evidence-averse propagandizing (now it’s “Chinese poplars” and “it’s pretty easy to see that risk gets high and serious”).


  40. grabulaon 30 Aug 2014 at 8:57 pm

    Why bother apologizing? Mlema and those like him are unapologetic about their inability to acknowledge the science. Just about every criticism Mlema has leveled at GMO has been reasonably addressed via science and he continues to move the goalposts. The point is that the fear of GMO and science is not rational and those that have it are not interested in being rational about it. As you already pointed out mlema and others, make the fatal mistake of comparing business practices with the science behind GMO. Not to mention that often times even the business practices used to bolster their argument aren’t indicative of the grand conspiracy to deregulate everything and sacrifice humanity for the almighty dollar. We all recognize business needs to be regulated because occasionally they go to far for profit. But again, the rational viewpoint understands that if those business are killing off their customers, are harming them in serious ways, they lose business and eventually fail.

    In the case of this thread mlema has upped the anti by effectively claiming since we can’t reach his impossible and unreasonable standards of guaranteed safety all we’re doing is rolling the dice and hoping for the best. This to me goes back to that failure to understand how solid our knowledge is in these areas. It’s not perfect but it goes a lot further than mlema would have anyone believe.

  41. grabulaon 31 Aug 2014 at 12:02 am

    @Rogue Medic

    nice youtube link on that debate. Of all the points being made I think Talibs was the weirdest, though it reminded me just how stupid D’Souza is. He does more damage than good for his cause.

  42. grabulaon 31 Aug 2014 at 2:29 am

    Curious what mlema would have to say about this:

  43. tmac57on 31 Aug 2014 at 10:31 am

    grabula said-

    In the case of this thread mlema has upped the anti…

    Hmmm…an interesting twist of meaning that actually makes sense.I’ll have to show this to my auntie.


  44. Mlemaon 01 Sep 2014 at 6:32 pm

    grabula, jsterritt – you guys are so fucked up as to what I’ve said, and have always said. Your fallacy is the “avoid the particulars” fallacy. It doesn’t matter how much you want GMOs to be ONE THING that you can endorse because the advanced scientific and technological power of manipulating genetics gets you off: science isn’t like that. Every GMO is different, and the main commerical applications outside of medicine right now leave many unanswered questions. You can extract all the quotes you want to, scientists don’t judge safety by quotes (especially from the industries that stand to profit, or from statements that don’t really say what shills like Chassy would like us to believe they say)
    The foods I’m concerned about are bt sweet corn in the US and bt eggplant. Find me ONE appropriate feeding study on eating bt sweet corn or eggplant that will assure me I’m not being exposed to a risk I wouldn’t face eating the plain old conventionally modified food. There are reasons scientists think we ought to have testing on these. The AMA says we ought to have mandatory testing. (that was purposefully left out of the extracted AMA quote in the “consensus chart” you guys wave like a flag that says “here come the ignorant who get their rocks off on the idea of inserting genes into plant cells! OOOOO, baby that insertion gets me HIGH! Let’s make a fetish club and talk about how awesome GMOs are!”)

    You guys have bought into every deception in the book. Your ignorance makes you naive, and no one like myself can give credit to the technology where it’s due without really confusing you – because it’s incomprehensible that someone might have an outlook that doesn’t “avoid the particulars”, and that doesn’t agree wholeheartedly with either “side” on this issue – but instead says that the science is complicated and nuanced and is different in every situation. And that comprehensive assessments need to be in place – like the NAS recommends.

    If you guys like to follow the herd of the techno-ecstatic, go ahead. But quit fuckin’ lying about what I’ve said. Or maybe you don’t even realize you’re lying because you can’t comprehend what I’ve said. Whatever. I made a comment. Why can’t you just make your own comments? Why have you got to pick mine out and fucking lie about it. Fuck off. Better yet, I’ll fuck off. It’s pretty plain that just about everyone here is pretty committed to a non-scientific take on GMOs. Why should I engage with such committed ideologies? I must be as whacked as you guys are.

  45. SteveAon 02 Sep 2014 at 12:12 pm

    Mlema: “Better yet, I’ll fuck off.”


  46. Mlemaon 02 Sep 2014 at 2:37 pm

    oh baby oh baby – You know you want my transgenic material – I’m the only one who can give it to you baby – oh mama nature, I feel my prowess, I’m going to plow the earth and grow my unique traits in you – c’mon now – I’ve got the PATENTS BABY – no one else can do it for you baby – here come the IP contracts baby – we’re gonna be RICH! RICH! RICH! ahhhhhhhh baby………we’re gonna be rich. rich, rich

    aw. I thought you’d like tending to the tissue cultures. You knew when we got together I had to sow my traits in other commodities. C’mon now baby. You know I can’t be happy with one. I gotta have ’em all! I gotta put my patented traits in as many money makers as will have me!

    Ah, I’m one fine genetic scientist. All those poor GMO lovers can only wish they were me. I change DNA! I’m the man!

    Ha ha. don’t let me interfere with your daydreams 🙂

  47. jsterritton 02 Sep 2014 at 4:09 pm


    You bring the same argument to EVERY discussion of the topic, then insist on being being poorly used, misunderstood, and miscomprehended?! Ha. ha. (But not funny ha ha.) At least commenters here take the time — every time — to swat away your particular breed of stupid. To wit: you believe that you have some kind of scientific “take” on GMO safety, when nothing could be further from the truth. To me, you come off as rejecter of science with the annoying habit of claiming scientific legitimacy. One need look no further than your “equitable” statements about medical-use GMOs and certain Bt crops: you really want to give a little ground so you can claim to be in some enlightened middle, but you always backpedal entirely to your default zero-tolerance “narrative.” Your “every GMO is different” line is particularly hilarious coming out of you, since you use the same broad brush to tar them all equally.

    Worse: you’ve been having the same discussion over and over across any number of posts on different and sometimes disparate topics. You should really try to participate in the discussion at hand rather than superimposing your same blather (most recently about “particulars”) with all the discretion and tact of an unmanned firehose.

    That or take your lumps or shut up.

  48. grabulaon 02 Sep 2014 at 9:14 pm

    Dear Mlema, please allow me to straighten you out on some details

    “you guys are so f***d up as to what I’ve said, and have always said.”

    The particulars of your message change at your convenience so can you blame us for having lost those? The generality is the same – We ‘don’t know’, haven’t tested GMO ‘properly’ (and by your definition so thoroughly as to make it unrealistic, never mind that you’ve completely bought into the misunderstanding on what we know about biology.

    “You can extract all the quotes you want to, scientists don’t judge safety by quotes”

    Half you’re arguments apply more strongly to the way you’ve been arguing against GMO mlema, this also makes it hard to understand your specifics. The vast majority of your argument relies on pulling quotes from seriously slanted sources or creatively interpreting and quote mining from reliable sources.

    “Every GMO is different, and the main commerical applications outside of medicine right now leave many unanswered questions.”

    Again, you need to stop arguing from ignorance here. It’s the last time I’ll point out how much of the science involved we actually understand.

    “The foods I’m concerned about are bt sweet corn in the US and bt eggplant.”

    So shifting the goalposts here I see. Now it’s just THESE TWO GMO crops you’re worried about. No way you’re stance up until now has supported you generally against GMO’s as a rule.

    “Find me ONE appropriate feeding study on eating bt sweet corn or eggplant that will assure me I’m not being exposed to a risk I wouldn’t face eating the plain old conventionally modified food.”

    I just posted a link to over 400 safety studies but in your usual intellectually lazy method, you need us to dig out the specifics for you. I won’t bother for two reasons, 1 – because I’m not enabling your lazyness and 2 – regardless of what’s given to you you’ll continue to wear your dogmatic blinders.

    “that was purposefully left out of the extracted AMA quote in the “consensus chart” you guys wave like a flag that says “here come the ignorant who get their rocks off on the idea of inserting genes into plant cells! OOOOO, baby that insertion gets me HIGH! Let’s make a fetish club and talk about how awesome GMOs are”

    I’m sorry, are you denying your bias after rants like these?

    “I have bought into every deception in the book. My ignorance makes me naive”

    There mlema fixed that for you. See above about biology and chemistry and how much we understand whether you like it or not.

    “But quit fuckin’ lying about what I’ve said”

    Go ahead mlema, deny you have a bias and an agenda. DENY IT and I will quote mine the crap out of your past discussions to prove to you that you exist on one side of this argument.

    ” Why should I engage with such committed ideologies?”

    I keep asking myself the same question ironically. Why do we bother with your dogmatic and limited view of science? Why not just continue to let you wave your ignorant and fallacious flag for your personal cause? It’s obvious time and time again your arguments are dismantled and yet you absolutely refuse to budge on your stance.

    “Ha ha. don’t let me interfere with your daydreams”

    Yeah, no way you have a bias…

    One last thing mlema, if you frequent this blog at all you have to understand how the scientific method works and what consensus is. You also have to understand that if you’re opinion flies in the face of both of those, you should probably take a step back and re-examine your views.

  49. grabulaon 02 Sep 2014 at 10:08 pm

    my other comment is awaiting moderation since some of the juicy quotes from mlema contain explicit material…but I just realized Mlema probably suffers from the backfire effect:

    We’ve hammered him with so much solid and scientific information that his denial has only grown stronger and obviously more passionate.

  50. sonicon 03 Sep 2014 at 12:20 am

    How easily one discounts concerns about possible futures depends on the threat one imagines (We might discover GMO’s are harmful is a possible future).
    In this case it seems some people percieve a large possible threat (ecocide), others are more insouciant (a drop in share price of some seed manufactures perhaps).

    It seems the current batch of GMO’s aren’t problematic human health wise, short term it appears they are safe to eat and they haven’t been around long enough to know what long term exposure will do, so off we go.
    I limit my exposure– I’m not so sure the long term is known… but I do all sorts of oddball things, so whatever.

    I have noticed that some of the GMO farmers are adopting ‘less till’ practices which I am fairly certain are good for the long term health of the soil.
    Imagine that- I can find a positive about GMO usage.

    Some notes-
    Genera is run by Biofortified which was founded by Pamela Ronald.

    “Professor Pamela Ronald is probably the scientist most widely known for publicly defending genetically engineered (GE or GMO) crops. Her media persona, familiar to readers of the Boston Globe, the Wall Street Journal, the Economist, NPR, and many other global media outlets, is to take no prisoners.

    From what I’ve seen the site is good, but I wouldn’t consider it ‘unbiased’.

    Not all of our foods are GMO’s.
    That’s easy to discover if you understand the difference between cross breeding and viral insertion of transgenes. If someone doesn’t know the difference, then they would be hard pressed to understand the possible problems. I think anyone using that argument will be difficult to reach until they understand the difference between the methods and the outcomes achieved using them.

    If the argument is that GMO’s aren’t worse than mutagenesis, then the logic has become ‘two wrongs do make a right’ and I say — good luck.

    What I have noticed is the weeds have gotten worse and there are insects immune to bt. The immunity to bt wasn’t supposed to happen– it’s bad news.
    As I understand it any boosts in yields are short lived and the use of pesticides increases after a while.

    Not supposed to happen.

    Apparently the solution is ’round-up 2′.

    Just think– you could be the first on the block to eat food grown with round-up 2 ! 🙂

    Good luck getting to the bottom of this one.

  51. grabulaon 03 Sep 2014 at 3:41 am


    ” I’m not so sure the long term is known…”

    This is that argument from ignorance mlema is so fond of. We DO know how the body processes material and how it handles chemicals, doseage etc… Certainly there is some grey area but this argument is fear mongering without any real basis for argument, especially considering the testing that goes on, that mlema is so fond of telling us doesn’t.

    Ultimately I’m tired of the argument because it shows an ignorance of science, combine it with mlemas dogmatic system of belief and it’s become tiresome and repetitive – except for the latest whining which shows he’s having a difficult time facing the science. I guess you hit something with a hammer long enough it begins to crack.

  52. Jeff_Rubinoffon 17 Nov 2014 at 2:17 pm

    Meanwhile on his Facebook page, Taleb is saying that biologists are too innumerate (or possibly just too stupid) to understand his mathematical argument, so we should listen to his physicist Nate instead.

  53. Wilkoon 21 Dec 2014 at 3:06 pm

    I wonder at what point Taleb would have considered it safe for Europeans to eat tomatoes, chocolate, corn and potatoes? Since these crops were not available to Europeans prior to 1492, I think more research has to be done to see it the Europeans can ingest these foods with zero risks.

  54. pmb6465on 22 Aug 2015 at 12:58 pm

    Taleb focuses on extremely improbable events that have massive implications. It seems like he thinks GMO’s have the potential lead to such an event. Damned be any benefits to society for something he believes has a near infinitesimally small chance of being a problem; that also of course with no scientific basis. I recently read in a book a about Taleb suggesting that he isn’t interested in tractable issues.

Trackback URI | Comments RSS

Leave a Reply

You must be logged in to post a comment.