My difficult relation with N.N. Taleb’s ideas has been the underlying thread of a significant proportion of posts, and I keep hoping this tension will generate useful insights. However sometimes my will to embrace ideological diversity is shaken vigorously…
Rupert Read and Nassim Nicholas Taleb have published at least two scholarly papers together: “The Precautionary Principle (with Application to the Genetic Modification of Organisms)” (2014) and “Religion, Heuristics, and Intergenerational Risk Management” (2014, bibliography is below, the links send to the full texts – note: I’m not sure if these articles have been peer-reviewed). The two papers are somewhat related, but while the first one generated a good deal of debate, as far as I can tell the second one has not. I have some relatively minor issues with the conclusions they reach on the first paper (mostly on the existential risks posed by GMOs) but I will not discuss them here. On the other hand, I am simply unable to understand why both Taleb and Read allowed themselves to publish something that could be used as a paradigmatic example of selective blindness: I am talking about the second article, and the astonishingly partial outlook that it proposes. More in detail, I have little or no issues with the positive arguments that paper makes: what they say is true, and has some weight, but it does not warrant the conclusions they reach, for even the most obvious counter-arguments are blissfully ignored. This post is my attempt to evaluate how much weight such counter-arguments should carry, and thus derive a more balanced conclusion.
I’ve decided to do so also because the authors themselves seem to suggest that they would welcome debate, in fact, their second to last sentence reads:
The idea advanced here, about the role of religion for system-risk management, has been aired in a manner to provoke attention and interest;
So here I am, showing attention and interest, in the form of what I hope will be constructive criticism.
I will start by summarising their argument, in the best way that I can. To understand what the article tries to achieve, it is sufficient to quote their introduction:
To the question posed […] “Does professional economics needs enrichment by religious or quasi-religious thinking?,” our answer is squarely “yes,” as we believe that religion has traditionally performed a powerful risk-management function at the level of the individual and the collectivity, particularly in preventing the accumulation of debt.
It is also important to note that they explicitly state that they are looking at religion in terms of practice, and not belief. They are interested in the behavioural set of rules that are typically associated with different religions, not in the (blatantly flawed) set of metaphysical beliefs that are normally considered to be the core of religions. In a nutshell, they say:
- Religious practices have survived the test of time, and therefore incorporate wisdom accumulated across a large number of generations. They have also survived rare adverse circumstances, meaning that the people who adopted these practices have survived, frequently because of these practices.
- Therefore, religious practices advocate behaviours that protect from unseen/rare risks. The accumulated knowledge that informs them is a form of wisdom specifically because it survived the test of time.
- Scientifically derived, top-down, teleological precepts are inherently inferior because they typically can’t incorporate aversion to rare risks: the evidence used to build them does not go back in time for long enough to include rare (and dangerous) events.
- Even if rules derived from evidence could be equally useful in protecting from rare risks, religious norms would still hold the advantage: since they do not claim to be rational, one doesn’t need to understand why they work to adopt them. In a sense, the claim of their supernatural origin makes their normative power stronger and more effective.
Given the points above, the authors reach two conclusions: a direct, practical observation, and an overarching, meta-conclusion about the value of religion.
– Direct conclusion: most religions contain specific warnings against the accumulation of debt. This is good, almost timeless, advice, and modern economy studies fail to acknowledge it and/or effectively support the wisdom of similar precepts that can be found in religious traditions.
– Meta-conclusion: religious practices should be seen as collections of useful rules, especially about what not-to-do; they are reservoirs of accumulated wisdom, and we should espouse the practices they advocate. On the other hand, secular societies are far too young: we don’t know if they are sustainable. Therefore, we should not try to reduce the influence of religion on our societies, but we should embrace it.
This is a short excerpt, and, of course, distilled in my own way; to make up your own mind, please do read the full article: it is not long, and is clearly aimed towards a non-specialist audience. As all my regular readers have surely guessed, I don’t have big issues with the direct conclusion, but I just can’t accept the meta-conclusion: it is downright dangerous, and should be uncompromisingly rejected. I say so while I accept all their main intermediate arguments: most of what Read and Taleb say is true, but their meta-conclusion does not follow, because it ignores important counter-arguments.
1 – Religions, especially when embedded in society, promote rigidity.
Taleb, of all people, should not be blind to this clear side-effect of religious norms! Considering that antifragility always requires a high degree of optionality, it should be quite obvious that normative systems that are directly promoting uniform behavioural patterns carry their own, internal source of risk. In the present world, where technological advances promote unprecedented levels of change on the global scale, relying on old-age wisdom, with a favourable inclination towards dogmatic norms, is simply absurd. It is guaranteed that the changes that are currently happening are generating new risks that the human race have never encountered before (if you need examples, you can read the precautionary principle paper linked above!), but, on the other hand, there is no guarantee that ancient wisdom will automatically protect against such novel (and systemic) risks. Moreover, uncritically accepting a dogmatic approach to ethics (in the sense of a set of norms), is a required condition to espouse a religious way of life, but doing so, by definition, limits the ability to change the set of rules in response to new challenges.
Taleb and Read do indirectly discuss some mitigating considerations: they note that many religious precepts take the form of “do nots” and thus incorporate advice about what should not be done (because accumulated knowledge shows that it is dangerous). In this way, the stifling effect of dogma is mitigated: as it concentrates on avoiding known risks, it does not limit positive action too much. This is true, but whether it counterbalances the other “positive” precepts of religions is anybody’s guess: since the human race is facing unprecedented global challenges, there is no guarantee that the existing “do nots” will be useful. They may or may not (some will, some won’t), but at the same time, the associated “dos” will keep inflicting their known damage. I have discussed this before, and see no reason to stop denouncing the inherent danger and nastiness of normative societies.
2 – Natural selection favours self-sustaining systems, but such systems normally obtain their antifragile qualities by making their single parts expendable.
In the case of society-wide systems, the single expendable parts are frequently human beings, and Taleb himself sustains that this makes such systems unethical. What Read and Taleb fail to acknowledge is that the overall set of rules associated with any religion is effective in preserving the existence of the overall system: it is not, in any way or form, optimised to promote the well-being of the people who embrace it. Sure, as I’ve said before, beliefs that are beneficial to the believer are also favoured, but they are a side consequence, and do not constitute the core of any large-scale belief-system. For example, the vast majority of existing religions include precepts that encourage fertility (the Darwinistic explanation is straightforward), and in the modern world this clearly does not promote the well-being of believers; on the contrary, such precept in itself exacerbates the well recognised existential risk associated with global population growth (see also point 1 – above). This means that of the sets of rules that are associated with any given religion, some will be useful, and some will be dangerous: but accepting a religion requires to (at least provisionally) accept the whole package, and would not, in itself, provide any tool to separate the good from the bad rules. So why should we bother?
Therefore, I maintain that even if it is entirely true that religious practices incorporate plenty of useful heuristics, it does not follow that living “without something like religion” is inherently dangerous. The authors correctly point out that “Modernity is in this sense a dangerous uncontrolled experiment” but they are wrong in defining “this sense”. What poses existential risks is the degree of (technologically driven) change, not the birth of secular societies. Moreover, I would argue that, specifically because the pace of change is accelerating, religious/normative societies exacerbate the risks, as they constrain adaptability in a way that is shaped by pre-existing risks (not the new global ones); therefore, we have no reason to believe that the accumulated wisdom will be useful to overcome new and unanticipated perils. In the modern world, we urgently need to abandon all forms of dogmatic thinking: this doesn’t mean that we should reject all the wisdom that has been incorporated in religious practice, it means that we have to save the wisdom while reducing the constraints associated with it, with an emphasis on reducing outdated constraints. We need to diversify and explore, and yes, we also need to urgently start designing our societies in order to make them antifragile, on a global scale, whenever possible.
Finally, I note that Read and Taleb have (inadvertently, I presume), espoused a self-defeating argument. Overall, the whole antifragility project seems to be aimed at expanding the scope of rational thinking: Taleb is telling us that uncertainty is here to stay, and that we should direct our efforts in understanding how to manage and/or benefit from it. He is contributing to a system of thought that may eventually allow us to engineer antifragile and therefore sustainable societies; and he is adamantly advocating the use of sound logic and evidence. He is showing us why we will never eliminate uncertainty, and that we should not even try, because succeeding would do us no good. To productively grow, this approach needs to promote rational thinking, and the willingness to overcome our own biases. One could argue that a lot of modern (social) science is indeed biased, because it tends to make predictions based on what we do know, forgetting how much we don’t know. Suggesting to solve this problem by accepting less recent and better tested “wisdom” promotes a-priori thinking, and may (or may not) produce some minor advantages; but it will not teach us (humans as a whole) how to thrive in always changing conditions, in fact, it will decrease our ability to do so.
Taleb, N. N., Read, R., Douady, R., Norman, J., & Bar-Yam, Y. (2014). The Precautionary Principle (with Application to the Genetic Modification of Organisms). arXiv preprint arXiv:1410.5787.
Read, R., & Taleb, N. N. (2014). Religion, Heuristics, and Intergenerational Risk Management. Scholarly Comments on Academic Economics, 11(2), 219-226.