What can we learn from charlatans and the different ways to model reality

Recently a very lousy charlatan came into popularity by “expalining” the lockdown-covid isue with three simple graphs

The actual peer-reviewed paper (from an Economics journal) is quite more humble:

Our results suggest that the Swedish policy of advice and trust in the population to reduce social interactions voluntarily was relatively successful.

Which is almost a joke, as in the meme….

Scientist: “My findings are meaningless if taken out of context.”

Charlatans: “Scientists claim their discoveries are useless”

A mini tutorial to show some very common reasoning mistakes from the Lomborg charlatan example was posted by Taleb

But beyond the chance to learn from charlatans reasoning flaws and prevent misinformation to spread, there is a very interesting isue related to this whole mini drama: When is sience the best thinking system? when is not?

All living systems acquire data, then after computation (C) use resulting information in inferential (I) process to generate new knowledge that finally is used for modeling (M) environment an act (A). Which is the best (more #Antifragile) CIMA process to cope with new data?

Is it #science? I think it depends on the data and there for, on the nature of the environment. Consider Simple Vs Complex environment, which may produce either

• Type 1: quantifiable, objective, systematizable, low dimension, low uncertain…
• Type 2: non quantifiable, non objective, non systematizable, high uncertain…

For the Type 1 of data I’m sure the most antifragile CIMA is science, but for the type 2 it could be not the case and in many cases it seems to me its going to be more plausible that the best CIMA will be an heuristic, traditions, rituals, religion, phylosophy… nature.

Clearly this is not new since it is discussed already in Antifragile book by Taleb, and also, as Jon Lovett pointed out to me, on informal institutions framework by Douglass North and others; butt as Poincare used to say, sometimes just naming things differently enhance progress.

“A common cultural heritage provides a means of reducing the divergence in the mental models that people in a society have, and constitutes the means for the intergenerational transfer of unifying perceptions. In pre-modern societies cultural learning provided a means of internal communication; it also provided shared explanations for phenomena outside the immediate experiences of the members of society in the form of religions, myths and dogmas. Such belief structures are not, however, confined to primitive societies but are an essential part of modern societies as well.”

Most probably simple environments are only islands.. so in real world both CIMAs are needed for achieve higher scale antifragility… think this multiscale nature of antifragility, although covered by Taleb’s work is systematically ignored.

Also let’s not forget that non scientific CIMAs may use practice and aesthetics alongside reason.

Back to the charlatan in question… Considering the complexity of the Lockdown-Covid problem, Luca Dellanna made an excelent first princple exposition of the topic in an heuristic plausible framework a la Polya.

Dellanna does not neglegt the costs of Lockdowns, but makes very clear that from principle, they work… now, the cost and optimal configuration of a luckdown is another topic. Of course, there is going to be variability in outcomes becuase interactions and contexts, but that does not invalidate the general principles; in the same way that complex experimental conditions for performing Galilieo falling objects experiments on earth does not invalidate the laws of Physics.

So, as I’ve pointed on other essays, when dealing with real world risk isues, aproximate heuristic thinking may be better than Science based desicion making -> https://lopezoliverx.medium.com/survive-in-extremistan-237a48b8214b

And also we should be awere in general about the domain of validity of different CIMAs and that implies to fully understand what do we talk about when we talk about science: https://lopezoliverx.medium.com/is-not-the-same-ting-49d2498574b4

But in any case what is completly useless is charlatanerism that is not rigorous science nor a valid heuristic reasoning. And that is why it is worth the effort to debunk them.

One final thougth about this related with the debate about is Mathemathiscs are invented or discovered

Initially, let us assume that there is a very large population of systems that have different CIMAS and that are interacting with a global environment of high complexity but that has spatial and temporal fluctuations, in such a way that locally there are regions of different levels of complexity, there even being some of them. so low complexity that we could consider them simple.
By chance, it is possible that a subpopulation of systems is found on an island with a simple environment or practically perceives its environment as simple due to the effect of the spatial and temporal scale at which the system that performs the CIMA can observe its environment. . In this scenario, the local environment is sufficiently well described by a set of quantifiable, objective, systematizable, low-dimensional and low-uncertainty data, which we will call type 1 data. The opposite case would be environments that are described by data not quantifiable, non-objective, non-systematizable, high dimensionality and high uncertainty, which we will call type 2 data.
In type 1 data it is expected that there will be regularities and patterns such that CIMAS that are better at identifying those regularities and patterns will have adaptive advantages over CIMAs that are not. Same as in island effect speciation, in which on small and isolated islands, populations tend to have limited diversity due to the lack of information exchange with other populations that in some cases could generate speciation, that is, subcategories of CIMA that Ultimately they are perceived as optimized to the local island type of simplicity.
In this sense one could think that indeed the most antifragile CIMA that emerges from the evolutionary ecological process of information in islands of simplicity is logical thinking.
Now, let us also suppose that some of these subpopulations on islands of simplicity can interact internally between the systems that compose it, that is, that they are social systems. In that case the process of evolution of the logical CIMAS could be accelerated by having a social meta-CIMA composed of the different CIMAs of the population. And even more so due to the niche construction effect, which refers to the process by which a type of system (a “species” modifies its environment to better adapt to its needs; these subpopulations could act on their environment to make it even simpler. than I originally wanted.
Faced with this scenario of accelerated evolution due to social interactions and niche construction, the meta-CIMA would tend to be increasingly logical and mathematics eventually emerged from that process.

From this perspective, mathematics is neither discovered nor invented but rather emerges from the co-evolutionary process of agents capable of carrying out a CIMA.

In this co-evolutionary process, in parallel the notion of experiment is generated and when experimental and mathematical CIMAs are combined, we arrive at the scientific CIMA.