Hume on Induction
Induction: for class
Basic Argument
1. We can’t have deductive justification for induction because then induction would not be induction.
2. We can’t have inductive justification for induction because induction would be circular.
3. If no deductive and no inductive justification for p, then no justification for p
4. Hence there is no justification for induction.
Reasons for 1: If we had deductive justification for induction, then the justification for induction would be deductive justification. But that means that induction would actually be deduction, which it is not. Or, we would be basing induction on deduction. But the problem with this is that deduction and induction are quite different forms of reasoning. A valid deductive argument is one where the truth of the premises guarantees the truth of the conclusion. A good inductive argument cannot guarantee the truth of the conclusion. To base induction on deduction is a kind of category mistake. Or, pretend that the above problems could be overcome. That is, suppose the following deductive argument for the justification of induction preserves the distinction between both deduction and induction:
a. If nature is uniform (at some level of generality), then induction is justified
b. Nature is uniform
c. Therefore, induction is justified
But, this will not work, according to Hume. The b premise can only be established on the basis of induction. That is, we have to assume that induction is justifed in order to be justified in believing that nature is uniform. Hence, our so-called deductive justification of induction, actually uses induction!
Reason for 2: If we gave an inductive justification for induction we would be using induction to justify induction. But that is precisely what we are trying to argue for. In other words, we are trying to find out whether induction is a justifiable form of reasoning. If we are use induction to figure that out, then we are assuming that induction is a justifiable form of reasoning in order to prove that induction is a justifiable form of reasoning. Yikes!
Reason for 3: this is supposed to be a tautology. That is, Hume thinks that reasoning is either deductive or inductive and that it is conceptually true that there are no other forms of reasoning.
More General Argument
1. Our so-called knowledge about unobserved matters of fact rests on the relation of cause and effect.
2. Our knowledge about cause and effect rests on the belief that the future will resemble the past
3. Hence, all of knowledge about unobserved matters of fact rests on the belief that the future will resemble the past.
4. There are only two kinds of justification for believing a proposition; either demonstrative or probabilistic justification
5. We can have no demonstrable nor probabilistic justification for believing that the future will resemble the past
6. Hence, we can have no justification for believing that the future will resemble the past.
7. If p rests on q and q is unjustifiable, then p is unjustifiable.
8. Hence, our so-called knowledge about unobserved matters of fact is unjustifiable.
Simpler Argument
1. All knowledge about unobserved matters of fact rests on the belief that the future will resemble the past.
2. We can have no deductive or inductive justification for believing that the future will resemble the past
3. If no deductive and no inductive justification for p, then no justification for p
4. Hence, we can have no justification for believing that the future will resemble the past.
5. Hence, we can have no justification for believing anything about unobserved matters of fact
6. Hence, there is no knowledge of unobserved matters of fact.
These arguments seem valid but I believe in things that are not empirically verifiable such as God, the reality of objective moral values, love, justice, and so on. Is this irreconcilable with the above arguments or can I validly believe that there are real truths outside of scientific facts. I believe there are but is that a reasonable position to take?
ReplyDeleteThese are all very nice points and questions and if pursued a bit more would reveal a bunch of the assumptions that Hume (and I think Goodman) have to rely on to get the arguments off the ground. So, let's supppose that we KNOW that God exists, that objective moral facts exists, etc. Is there a way that such things might help us respond to the argument? It might be useful to start with investigating how we know such things, given that we do know them. In otehr words, start with the assumption that we know those things. How do we know them?
ReplyDelete