Guilty or Innocent
Taleb argues that long-surviving natural processes deserve a presumption of robustness, placing the burden of evidence on those who propose to alter them. This is his rule for managing uncertainty about interventions.
THE IDEA
Guilty or Innocent
If there is something in nature you don’t understand, odds are it makes sense in a deeper way that is beyond your understanding. So there is a logic to natural things that is much superior to our own. Just as there is a dichotomy in law: innocent until proven guilty as opposed to guilty until proven innocent, let me express my rule as follows: what Mother Nature does is rigorous until proven otherwise; what humans and science do is flawed until proven otherwise.
Let us close on this business of b***t “evidence.” If you want to talk about the “statistically significant,” nothing on the planet can be as close to “statistically significant” as nature. This is in deference to her track record and the sheer statistical significance of her massively large experience—the way she has managed to survive Black Swan events. So overriding her requires some very convincing justification on our part, rather than the reverse, as is commonly done, and it is very hard to beat her on statistical grounds—as I wrote in Chapter 7 in the discussion on procrastination, we can invoke the naturalistic fallacy when it comes to ethics, not when it comes to risk management.5
Let me repeat violations of logic in the name of “evidence” owing to their gravity. I am not joking: just as I face the shocking request “Do you have evidence?” when I question a given unnatural treatment, such as icing one’s swollen nose, in the past, many faced the question “Do you have evidence that trans fat is harmful?” and needed to produce proofs—which they were obviously unable to do because it took decades before the harm became apparent. These questions are offered more often than not by smart people, even doctors. So when the (present) inhabitants of Mother Earth want to do something counter to nature, they are the ones that need to produce the evidence, if they can.
Everything nonstable or breakable has had ample chance to break over time. Further, the interactions between components of Mother Nature had to modulate in such a way as to keep the overall system alive. What emerges over millions of years is a wonderful combination of solidity, antifragility, and local fragility, sacrifices in one area made in order for nature to function better. We sacrifice ourselves in favor of our genes, trading our fragility for their survival. We age, but they stay young and get fitter and fitter outside us. Things break on a small scale all the time, in order to avoid large-scale generalized catastrophes.
When to use it
Health interventions: examine the evidence for benefits and harms, including the consequences of leaving a condition untreated, when assessing a proposed intervention.
Environmental policy: ask developers to evaluate possible disruption of an established ecosystem, including effects that may be hard to reverse.
Technology adoption: examine how much relevant experience supports a proposed change and what a failure would cost.
Personal risk management: use evolutionary precedents to generate questions about changes in diet or sleep, then examine the evidence for the particular practice.
FROM THEORY TO EVERYDAY LIFE
In practice
Icing a Swollen Nose
When questioning the unnatural treatment of icing a swollen nose, the author is asked for evidence that it is harmful, despite the lack of evidence for its benefits. The principle suggests that the burden of proof should be on those advocating the unnatural intervention.
Trans Fat Debate
Historically, opponents of trans fat were required to produce evidence of its harm, even though the harm only became apparent after decades. Under the reversed burden of proof, the onus would be on the proponents of trans fat to demonstrate its safety.
A MOMENT TO REFLECT
Put it to work.
When evaluating a natural process or remedy, what does its history tell you, and what evidence is still needed about benefits and harms?
How can we apply this principle to evaluate technological innovations, such as genetically modified organisms or artificial intelligence?
What does nature's track record of surviving Black Swans tell us about the reliability of long-standing natural processes in risk management?
Often confused with
References
- The Bed of Procrustes — Nassim Nicholas Taleb (2010)
CONTINUE THE CONVERSATION
Source texts
- Antifragile (Nassim Nicholas Taleb)