Has anyone published a definitive argument regarding a "positive correlation fallacy"? Question

I think there may be a common fallacy which I do not know if it has been given a name and a fully developed, possibly mathematical, argument.

The idea is something like a basic fallacy of assuming because a positive correlation can be drawn between some things, that this is sufficient to establish some kind of conclusive evaluation about their relationship; enough to characterize them in some way.

I know there are two common ideas related to this, the availability bias, and "correlation does not equal causation", but I mean something more precise than these.

I think it very often comes up regarding some sort of normative or prescriptive judgment - what is ethical, what is good for the environment, what is a better idea financially, or what is a better strategy regarding some longterm goal, etc.

Here is only of of many examples. Let's begin with the intuitive sentiment, "cryptocurrency uses a lot of energy". Sometimes, one commonly hears, implicitly or explicitly, the following argument made regarding this:

1. Cryptocurrency uses a lot of energy.
2. Using energy releases fossil fuels.
3. Fossil fuels contribute to global warming.
4. Because global warming is bad, therefore crypto is bad (in some way or other: you should use it less, perhaps, or some other implication.)

This is just an example, and is not meant to be the focus of this post. The basic idea is simply that the above may not be sufficient to claim that if you wish to mitigate global warming, that you should decrease use of cryptocurrency (for example). Or maybe more generally, that cryptocurrency is therefore bad, or, that it is therefore responsible for global warming.

This is my attempt to economically give only one single counterexample, to sufficiently make the problem clear. What if it turned out that as people used more cryptocurrency, for some reason, it changed the economy and society dramatically, and this led to a rapid increase in access to information, knowledge, and education for all people. This led to feedback loop effects where the average person became more versed in science, and public opinion about climate change shifted to more or less unanimous opinion that climate change should be averted. With more knowledge, technological skills, scientific knowledge, and even leisure time, human society rapidly developed new technologies, such as renewable energies and even carbon sequestration techniques, that were highly effective. The issue of global warming was definitively resolved within a short span of time. It turned out that one of the fastest and most effective ways to avert climate change was by evangelizing the use of cryptocurrency.

The general idea here is that there were hidden factors that were not being taken into account. For example, if your goal is to reduce energy consumption, then if cryptocurrency has a direct, unavoidable relationship with energy consumption, then it may follow that in order to decrease energy consumption, you should decrease use of crypto. I think there are two fallacies here. 1, it is not the case that use of cryptocurrency is intrinsically linked to energy consumption. What if a new algorithm were developed which used extremely little energy? It would then become more apparent that the correlation between crypto and energy usage is not necessary, inherent, or inevitable. Rather, the connection is the more tautological sounding connection between processes which use a lot of energy, and energy consumption. If crypto use is facilitated by algorithms and processes that consume a lot of energy, then there is a second-degree relationship between use of cryptocurrency and energy use. This idea extends all the way out to global warming, and the idea that global warming is bad. If our aim is to mitigate global warming, there may be many, many more fine details in the causal chain or connection between cryptocurrency and global warming. Even further, if our real aim is something ethical, like the maximization of human wellbeing, then we must go even further in understanding the relationship between global warming and that ethical goal. So this is something like thinking two things are the same thing, and not looking in more detail, perhaps.

The second fallacy is ignoring what you cannot see. As described above, if the goal of mitigating climate change was to increase human happiness, and it turned out that for some reason, crypto dramatically increased human happiness, a hypothetical utilitarian argument could be made that even if using crypto did make global warming worse, its effect on human happiness was greater than the negative effect on human happiness of global warming; so if our endgoal was human happiness, crypto would actually be the preferred option, if you tried to study all possible relationships, and not just one possible one.

Again, this is just a sketch of a single example. I want to abstract this into a mathematical argument about a kind of reasoning fallacy regarding correlations.

Thank you.

posted 2 months ago

CC0

2mo ago

Julius H.‭

16 10 12