Taleb expounds on his definition of the natural - anything that man has done for 100s of years without observing downside.
There is reasonable cause to believe that people are very interventionist in a lot of cases where all that is needed is letting the system sort itself out. Diabetes medication for instance (again - not in all cases): https://www.youtube.com/watch?v=da1vvigy5tQ
A lot of openness to low quality scientific work done in a lot of domains and our eagerness to think that anything with numbers slapped on it is better has given us wonderful gifts like high-fructose corn syrup, trans-fat, thalidomide (all within the last century).
The Taleb metric is just that a lot of systems don't respond in a linear fashion to random input. High levels of consumption of a substance only materialize with a delay of a few decades in horrible ways and the negative result can undo all the positives over the duration of the use of said substance. Thus is it important to err on the side of those things that have been (i) well tested over centuries and (ii) strong evidence doesn't exist that said substance is actually dangerous (tobacco for instance).
Note that thalidomide is still used as a medicine; it is a very effective drug when other drugs don't work. The issue was that it was given to pregnant women, when it was known that it shouldn't have been.
Taleb expounds on his definition of the natural - anything that man has done for 100s of years without observing downside.
There is reasonable cause to believe that people are very interventionist in a lot of cases where all that is needed is letting the system sort itself out. Diabetes medication for instance (again - not in all cases): https://www.youtube.com/watch?v=da1vvigy5tQ
A lot of openness to low quality scientific work done in a lot of domains and our eagerness to think that anything with numbers slapped on it is better has given us wonderful gifts like high-fructose corn syrup, trans-fat, thalidomide (all within the last century).
The Taleb metric is just that a lot of systems don't respond in a linear fashion to random input. High levels of consumption of a substance only materialize with a delay of a few decades in horrible ways and the negative result can undo all the positives over the duration of the use of said substance. Thus is it important to err on the side of those things that have been (i) well tested over centuries and (ii) strong evidence doesn't exist that said substance is actually dangerous (tobacco for instance).