# Hume's induction using circularity fallacy

Hello everyone, i was going through this reading of a book in which it present Hume's induction wrong using circularity . I was trying to find some error in it as given there but could not pinpoint the exact mistake . However i feel conclusions 5-10 are clear of any mistake. Can someone help or is up for a discussion on it?

– Phil
Jun 19 at 5:51
• Do you think it is a "reasonable" reconstruction of Hume's argument? Jun 19 at 14:32
• The gist of Hume's approach is: regarding matters of facts (where a priori reasoning is inapplicable) we use probable reasoning connecting (presumed) causes and effects. This reasoning is based on the fundamental assumption that we may call the uniformity principle: The future will be like the past. This principle is not demonstrable, and thus we cannot ground inductive reasoning on "absolute" principles. Jun 19 at 14:37
• What is Hume's solution to this conundrum? "Since we’re determined—caused—to make causal inferences, then if they aren’t “determin’d by reason”, there must be “some principle of equal weight and authority” that leads us to make them. Hume maintains that this principle is custom or habit." We may say that the "firware" of our mind is programmed to perceive or extrapolate causal relations and uniform laws. Jun 19 at 14:38
• Having said that, where is the fallacy? In "human" attitude? in Humean's philosophy? in the (unknown) author quotation above? Jun 19 at 14:40

If you look at the Stanford Encyclopedia of Philosophy page on "The Problem of Induction", the argument you quote is a formalization of the "second horn" of Hume's dilemma about our having no firm grounds for trusting inductive inferences. The first horn is about how there is no good "demonstrative" argument for induction from principles that we feel totally confident in a priori akin to axioms of some area of math like arithmetic. The second horn is about how we can't make an empirical or "probable" argument from experience to support the idea that induction is trustworthy, since such an argument would have to presuppose the very thing it's trying to prove, and thus would be circular.

If you look at the various responses to Hume in the SEP article, I don't think there are any philosophers who have argued there actually is a fallacy in the second horn of Hume's dilemma, i.e. there's no one who actually takes the position that we can be confident induction will work in the future solely because we've seen that it's worked in the past. Most responses either attack the first horn and say there is some a priori reasoning that leads to the conclusion that induction should work, or they just say that induction is not something we can have an argument for but something that we basically just have to presuppose when making any argument that isn't purely logical/mathematical, perhaps because it's built into human consciousness to presuppose such a principle (which I think was basically Kant's view, see this answer), or perhaps because it's just part of some normative rules of linguistic argument that we have to mutually accept if we want to talk to other people (which is how I would interpret Wittgenstein's response in section 5.2 of the article).

• Don't forget Popper's view, that induction is not widely used, and not key to doing science. Jun 26 at 13:49
• @CriglCragl - True, the sort of "pure" induction as conceived by Aristotle and by most philosophers of science up through the late 20th century is different from the hypothetico-deductive method that is a more common way of thinking about science today; for those who aren't familiar with the distinction see people.loyno.edu/~folse/HypDeduc.html and people.loyno.edu/%7Efolse/HypDeduc5.html (from the philosophy of science course site at people.loyno.edu/%7Efolse/Phil236f02.html ) for more. Jun 26 at 18:48
• (cont.) In my answer above I was using "induction" in the broader sense of any kind of claim based on evidence from experience, as opposed to logical deduction, and Hume's problem would apply equally to the hypothetico-deductive method. Popper's view that scientific claims are purely about some theory not yet being ruled out doesn't actually claim anything about positive reasons to believe the future will resemble the past (it doesn't try to say 'emeralds are grue' type theories are less likely), so it isn't subject to Hume's problem of induction even under a broad definition of "induction" Jun 26 at 18:54
• Meant to say "up through the late 19th century" in my comment above, not "through the late 20th century"--by the 20th century Baconian induction was falling out of favor and hypothetico-deductive becoming a more popular way of thinking about science, though there were a few who endorsed it in the 19th century like Charles Sanders Peirce and Ludwig Boltzmann. Jun 26 at 23:44

I agree that the idea of the Uniformity of Nature, or the Uniformity Principle, is essential to induction. However, I disagree that the idea is circular.

At academia.edu I posted a paper that argues for noncircularity. “Induction and the Uniformity Principle; derivation of the principle from fundamental axioms”. The argument is that the Laws of Thought, taken together, show the validity of the Uniformity of Nature. https://www.academia.edu/41240897/Induction_and_the_Uniformity_Principle_derivation_of_the_principle_from_fundamental_axioms

Here is the abstract:

The Uniformity Principle is the notion that relationships between future events will resemble the relationships between similar events in the past. The principle is essential to inductive reasoning. The primary objection to the Principle is that it appears to be circular; its content is the very assumption that the inductive process sets out to prove. This essay proposes that the Principle is not a part of circular reasoning. Three fundamental axioms, titled Existence, Identity, and Continuity, show that the Principle is noncircular.

• However in an article of plato.stanford.edu/entries/induction-problem it is quoted that Hume says that argument is circular in nature . I am sorry but I fail to get across this point can you elaborate a bit :( Jun 19 at 21:49
• In the essay, I offer that the Laws of Thought are a foundation for the Uniformity Principle, and so avoid the need to simply assume the principle is true without further proof. Jun 20 at 6:54