Q&A: Causality and Statistics
Causality and Statistics
Question
Hello Rabbi, I had an idea and I’d be happy to hear your professional opinion on it. The idea is that statistics can prove, almost certainly, that there is causality. You test some natural and familiar phenomenon, such as the fact that every time water is poured on the ground, a wet spot appears. After doing the experiment several times, that proves that there is probably a connection between the things, even though there could be a small chance that it happened on its own. And then you apply the principle of generalization to all other material things. What do you think?
Answer
That doesn’t prove causality, only correlation. But that’s trivial. Clearly, this is one of the common uses of statistics, and that is not what the objections were about.
Discussion on Answer
Then you are assuming the principle of causality rather than proving it. You assume that things do not simply appear for no reason, but that itself is the principle of causality.
Sorry, I didn’t phrase myself correctly. What I meant was that, according to what we know, the water spot appears only after the water is poured, and not at other times or in other places. I’m not assuming causality; I’m simply saying that if the spot appears on its own, and by chance it appears only on the occasions when we pour water, that is a very far-fetched statistical datum, and therefore we should conclude that causality is the more reasonable option.
That’s not correct. You are again assuming causality. Hume argued that what you can observe is correlation, not causality. You can see that event A always appears after event B. The assumption that A is the cause of B is your assumption. Statistics says nothing whatsoever about the relation between those two options.
But I’ve reached a situation where every time A occurs (the act of pouring water), B occurs (the spot). That is, a cause-and-effect relation, because B does not occur on its own and its occurrence depends on A. It seems to me like a classic cause-and-effect connection.
I don’t know what is unclear about what I wrote. You keep repeating the same begging of the question. You can find an analysis of the concept of causality in my book The Sciences of Freedom and in the fourth book of the Talmudic Logic series.
But you can see, following the example I gave, that the water spot—at least as far as we know from our experience in the world—doesn’t just appear on the ground for no reason, and that a spot of a certain size appears only in coordination with a certain amount of water.