Empiricists believe sensory experience provides a foundation for knowledge.
Skeptics use illusions, dreams, and hallucinations to challenge the reliability of sense experience.
Empiricists try to reinforce the reliability of sense experience.
Reliable impressions come from using multiple senses together.
Expectations about the world are based on past experiences, but the reliability of reasoning from past to future is questionable.
David Hume
Hume is considered an empiricist.
He provides skeptical arguments against empiricism.
Even if our senses provide reliable knowledge, that knowledge is limited.
We form further beliefs that cannot be justified.
Hume’s Argument
Hume distinguishes between analytic (relations of ideas) and synthetic knowledge (matters of fact).
Synthetic knowledge involves expectations about the future.
Even if sense data is infallible, it doesn't provide a basis for conclusions about the future.
It's impossible to know a priori how causal events unfold.
No logical argument justifies our instinctive reasoning.
Hume challenges the idea that reasoning leads a child to expect similar effects from similar causes, questioning the existence of a logical argument for this expectation.
Induction vs. Deduction
Paley's teleological argument about the watch is non-deductive.
Deductive arguments aim for valid inferences.
Non-deductive arguments have a different relationship between premises and conclusion.
Inductive arguments reason from observed premises to unobserved conclusions.
Example of induction:
The sun rose on Monday morning
The sun rose on Tuesday morning
The sun rose on Wednesday morning
The sun rose this (Thursday) morning
Therefore, the sun will rise tomorrow morning
Inductive arguments are strictly invalid.
The Problem of Induction
Hume questions how we can justify beliefs about the future if inductive reasoning is invalid.
Hume believes this exposes irrationality in humans.
Hume's argument:
We must reason inductively.
Inductive reasoning is invalid.
To reason invalidly is irrational.
Therefore, we must be irrational.
Solutions to the Problem of Induction
Solution 1: No True Scotsman Ploy
Example: Predicting a red ball will move when hit with a white ball, but it doesn't move.
Response: Claiming it's not a true billiard ball.
The No True Scotsman Ploy:
Hypothesis: "All Scots are mean."
Counterexample: "Andrew Carnegie is generous!"
Response: "Carnegie is not a true Scot."
Sceptical Response
The analytic/synthetic distinction reveals the problem.
"All Scots are mean" is presented as a synthetic truth but becomes an analytic truth when counterexamples are dismissed.
Analogous to saying "that can’t be a real bachelor!" when given an example of a married bachelor.
Hume focuses on knowing synthetic truths.
The no-true-Scotsman ploy contains a valuable philosophical lesson on a logical trick.
Solution 2: Appeal to Inductive Principles
Add an inductive principle (principle of the uniformity of nature - PUN) to validate inductive arguments.
Example:
The sun rose on Monday morning
The sun rose on Tuesday morning
The sun rose on Wednesday morning
The sun rose this (Thursday) morning
The future resembles the past (PUN)
Therefore, the sun will rise tomorrow morning
The argument seems valid with the PUN added
Sceptical Response
How can the PUN be justified? It seems that we can only know it a posteriori, and hence a further infinite regress is threatened!
Inductive reasoning is used to justify the PUN, which is invalid according to Hume.
The PUN is too strong because the future doesn't always resemble the past.
Solution 3: Probabilism
Acknowledge we can't be certain, but form probabilistic judgments using past experiences.
Example:
The sun rose on Monday morning
The sun rose on Tuesday morning
The sun rose on Wednesday morning
The sun rose this (Thursday) morning
Therefore, the sun will probably rise tomorrow morning
Probabilists aim to demonstrate the likelihood of inferences based on past occurrences.
The sun might not rise tomorrow, but it's probable.
Sceptical Response
The argument is still invalid.
A probabilistic version of the PUN is needed, leading to an infinite regress.
The likelihood of future instances resembling past ones isn't solely based on the number of observed past instances.
Solution 4: Deductivism/Falsificationism
Karl Popper denies we reason inductively.
Humans jump to general conclusions and discard falsified ones.
We can reason deductively from general conclusions about the future.
Example:
The sun always rises in the morning (H1)
Therefore, the sun will rise tomorrow morning.
The sun never rises in the morning (H2).
Therefore, the sun will not rise tomorrow morning.
H2 has been falsified, while H1 has withstood criticism.
It's reasonable to believe hypotheses that haven't been disproven.
This aligns with the scientific method.
Humans are not irrational.
Sceptical Replies
Do we jump to conclusions inductively?
Popper: Even if we use induction, we can test hypotheses deductively.
The main challenge: Is it reasonable to believe a proposition without proof or evidence of likelihood?
Bertrand Russell on Induction
If there's no answer to Hume, there's no intellectual difference between sanity and insanity.
The lunatic who believes that he is a poached egg is to be condemned solely on the ground that he is in a minority
A way of escaping from it must be hoped for.
Conclusion
The problem of induction is a serious challenge for empiricism.
There are many possible lines of reply.
Hume embraced the conclusion of irrationality.
Others find this conclusion intolerable.
"The man who has fed the chicken every day throughout its life at last wrings its neck instead, showing that more refined views as to the uniformity of nature would have been useful to the chicken… our instincts certainly cause us to believe that the sun will rise to-morrow, but we may be in no better a position than the chicken."