Q&A: Induction and Theory
Induction and Theory
Question
Hello Rabbi,
I wanted to ask for a few clarifications regarding the problem of induction,
1) Why can’t one say that physical theories that stand up to the principle of falsification have solved the problem of induction?
After all, if we test the theory that the sun will rise every morning against any other theory,
then if the principle of induction is not correct, the probability that we would just happen to be right is tiny. Therefore, if the sun does in fact rise, that would significantly strengthen the theory and show that the principle of induction is confirmed.
True, there are still infinitely many other theories according to which the sun will rise tomorrow, such as: the sun will rise only for another X years, another X+1 years, and so on. But the scientist is the one who decides which theory to test, and he chooses the simplest one that explains reality for him, and again, statistically, the fact is that he has been right until now.
2) Do you think branches of science that deal with the past, like astrophysics, archaeology, or evolution, are scientific?
Take the Big Bang theory for example—it would seem not to stand the test of falsification, since it deals with the past and not the future. But based on that theory one can think of implications that were indeed confirmed years later, such as cosmic background radiation or observations of the expansion of the universe. The same is true in archaeology: there is a theory on the basis of which one can predict future excavations.
So are theories about the past also called scientific, because they have implications external to the theory?
3) Do you think there is a connection between the problem of induction and the existence of God, and not only on the level of wonder—how we actually have good intuitions—but a causal connection?
Answer
1. Your explanation itself assumes the principle of induction. When you speak about what is “more probable” (the simplest among all possible explanations). Not for nothing did I connect Hume’s doubt to the analytic position (which demands certainty). As for causality, that is already deeper, because there there is not even probability in this sense. The move from correlation to causation is purely subjective.
2. Completely scientific. As long as it is falsifiable by a future experiment.
3. I wrote that there is a connection. That is the proof from epistemology (The First Being, fourth conversation, part two). I didn’t understand the end of your question.
Discussion on Answer
Because relying on the fact that they work is itself induction. You are basically asking why one should rely on what worked in the past, and answering: because it worked in the past. You are repeating the argument from the graph that I have used more than once (why choose דווקא the straight line). But that does not solve the problem of induction.
3. We did not see that in 1.
1/3.
I didn’t understand why the argument from the graph does not solve the problem of induction?? Would you agree that if certain basic assumptions were not true, the chance that our understanding would discover things about the world would be tiny? If so, experiment proves retroactively that the assumptions themselves are very probable, even if not necessary.
???
The experiment proves that the understandings were correct in the past. The inference from that to the future is induction.
Rabbi, please tell me if I understood correctly:
Within the rules themselves—like Occam’s razor and induction—there is a good ability to choose a scientific theory.
Likewise, on the assumption that these principles were mistaken, the chance that they would have succeeded until now is extremely improbable.
But on the other hand, there is no reason to derive from all this that they will also be correct in the future, because that would be a circular assumption under trust in those very principles.
The justification for why they work at all is the proof from epistemology. That is, God is not a condition for holding these assumptions, but rather a philosophical justification for them.
But why shouldn’t an atheist scientist treat the question of the correctness of these principles as just a purely skeptical claim, like the question of why you assume your eyes see correctly or your memory is valid?
Also, one should remember that an atheist scientist usually believes that we are close to understanding the laws of the universe. So from his point of view, that explains very well why the sun will rise tomorrow. That is, the laws themselves that we formulated and believe in confirm the theory. It somewhat reminds one of the believer, who in the previous sentence would replace the word “laws” with “God”..
Indeed. The question is what the justification for the rules is (they are not really formal rules—common sense). That is indeed the proof from epistemology.
An atheist really is not bothered by these questions, just as he is not bothered by the quality of the picture his eyes give him. So despite the proof from epistemology, he remains an atheist. But he is mistaken. He has simply gotten used to it.
Even if we were to arrive at an understanding of all the laws, the question of why these are the laws, and who “legislated” them, would still remain.
1. Of course my explanation assumes the principle of induction.
More precisely, the term “more probable” is Occam’s razor, but for the sake of continuing the discussion let’s include all these assumptions under the term scientific thinking.
And if so, if the capacities of scientific thinking are not correct, then there is no reason at all to think they would work, so given that these capacities do work, it is reasonable to rely on them.
Why do you see a flaw in that claim?
3. I meant: does it strike you as surprising that we have correct scientific thinking? Like the argument from complexity, or do you hold that belief in God is a condition for scientific thinking? If so, why? In 1 we saw that mathematically one can confirm scientific thinking, or at least give it a very high degree of confidence.