Q&A: Causality – Hume's Problem
Causality – Hume's Problem
Question
Hello Rabbi,
Maybe this is an ignorant question, but “the bashful do not learn”: what exactly was David Hume’s fundamental problem with causality? True, we don’t observe it, and it also can’t be proven by a logical proof, but isn’t there an extremely, extremely high probability here—so high that no sane person would even consider the other side—that if the sun has risen until now, it will continue to rise from now on as well?
I have no trouble understanding that Hume did not seek to 'know' whether there is causality, but rather to 'understand' how this concept was formed. But introductory philosophy books describe it as though, because of Hume, all of science felt threatened until Kant came and saved it.
Answer
Science never felt threatened by any philosophical problem. The philosophy of science felt threatened.
The fact is that Hume, even in his conclusion, claims that causality is only a mode of our own perception and does not exist in the world itself. I tend to agree with your assumption that he himself also did not really think so. As for the sun continuing to rise, that is induction and not exactly causality, and there too he argued that it is only a methodological assumption, and again I do not believe him that this is what he really thought.
Discussion on Answer
How did reason find it? How do we perceive that causality exists as a causal process? Kant turned the whole thing upside down and presented the causal conception as the cause of causality. If you mean that reason “just knows” that there is causality, and that this is not an assumption it brought with it out of the skull but rather something it learned from the world, then you’ve arrived at the idea that one can also “know” things that are not sensory at all but are seemingly completely abstract conceptual constructs (like the principle of causality)—but then you’re bringing baguettes into the bakery, digging up the mole family, and the sages already beat you to plowing with an ox and a calf together, each man at his work, his burden, and his charges.
Causality in science derives from the laws that science discovers. The laws describe the temporal relations between the observed phenomena; in that sense, causality in science is real and stands at its foundation.
In everyday perception, causality is not based on mathematical laws but on correlations, which become causality because of temporal sequence and everyday logic.
If Hume had understood evolution, he would have understood that evolution tried many things, and only those brains that learned to predict the future (mainly the near future) on the basis of those things we see as causal survived. And the fact that those brains survived is evidence that these things are grounded in reality.
In other words, the proof is not based on reasoning but on reality. Those who were wrong did not survive. Whoever remained was probably right.
That proof fits only with regard to the principle of induction, and even there one would have to say that the brains learned to predict the future that was in the past, but there is no justification at all for thinking this is relevant to the future. The attempt to prove induction by induction is as old as Hume himself.
What is “justification”?
There is no justification. There is probability.
On the one hand, road accidents are caused by the failure of ordinary everyday causal reasoning.
On the other hand, most people do not get into road accidents at any given moment. That is what the “justification” is based on.
And on that basis one can predict that in the coming year too there will be roughly the same number of deaths from road accidents. Not 10 times as many and not one-tenth as many.
Shuli, I didn’t understand your point.
I did not mean that reason knows a priori that there is causality or induction. Rather, reason tries to find the lawfulness behind the fascinating phenomenon that causality (or induction) simply works: every time we put a splinter of wood into fire, it burns. Isn’t that sufficient reason to assume that there is some real metaphysical lawfulness here saying that whenever a splinter of wood enters fire, it will burn? That is, that reason is discovering something real about the world here—something that cannot be observed by the senses and is not supposed to be.
If you are proposing to learn induction by induction, then the circular fallacy is obvious (that is, the explanation is called for and the rejection is part of the basic argument). Or maybe I didn’t understand.
No. I am arguing that induction is not learned from an empirical proof (where of course one would have had to arrive at the circular fallacy by using the principle of induction itself), but rather from an attempt to fit the world to the basic assumptions of our reason. Reason requires lawfulness behind every phenomenon. Therefore one must posit that there is induction.
The same goes for causality.
And this is not just some attempt to impose reason on the world. It is entirely called for if we do not want to ignore the fact that this discussion itself was conducted by means of the basic assumptions of reason and not only by means of empirical proofs.
What made Kant think that causality can be recognized only in the world of phenomena?
After all, the starting point of someone looking for a proof of the principle of causality is the understanding that the consistency of phenomena needs to be explained according to some kind of lawfulness, and that is supposed to be clarified by reason itself. (After all, Kant was not an empiricist like Hume, who at least according to what he himself claimed was not bothered by settling for the knowledge he received from the senses.)
If that is the starting point, then what is missing if we understand that reason found the lawfulness in the real world? Is that explanation really less convincing than saying that reason found the lawfulness in its own internal structure?