//****************************************************************************//
//**************** Hume and Causation - September 3rd, 2019 ******************//
//**************************************************************************//

- Alright! Hopefully you all had a good Labor Day weekend
--------------------------------------------------------------------------------

- Today, we have a VERY famous reading from David Hume: "An Enquiry Concerning Human Understanding"
    - Hume (1711-1776) was a Scottish philosopher who did influential work in empiricism (and some work in economics with Adam Smith)
    - This essay of his was written in 1748, and was a little bit different than the readings we've done so far
        - The last two reading talked about the process of how science should be done, with Bacon arguing for doing inductive logic
        - In this reading, though, Hume is giving a more philosophical argument about the nature of inference itself, and his opinions and arguments on its limits
            - In section 2, Hume argues that all of our ideas are just combinations of "impressions" 
                - Hume was especially taking aim at "rationalist" philosophers of his day, who believed human knowledge primarily came from logic
                - Hume would often dismiss ideas that didn't have a clear impression, saying that those things must be terms that were meaningless and vague, disconnected from the real world
            - In section 4, Hume gets to the heart of his argument: that we can't rationally prove cause-and-effect relationships actually exist - despite them underlying most of what we believe!
                - This argument - "Hume's fork" - is where Hume says that all knowledge falls into 2 categories:
                    - RELATIONS OF IDEAS are "a priori" logical knowledge that we don't need experience for, like algebra, logic, etc.; Hume generally thinks this is okay
                    - MATTERS OF FACT are branches of knowledge that we can only know "a posteriori," through experience
                        - He sees the key distinction as that matters of fact didn't "have to be"; we can imagine a world where snow is hot or the grass is pink, and it isn't inherently contradictory (even if it doesn't exist)
                    - From there, Hume argues that most of our beliefs about "matters of fact" come from cause-and-effect relationships that were NECESSARY, e.g. a billiard ball striking another billiard ball will cause it to move
                        - So, he argues that if we know about these necessary connections, then it must be either a "relation of ideas" or a "matter of fact"
                            - However, Hume says it can't be a priori knowledge, since there are many causes we DON'T know when we're born, e.g. that fire is hot
                        - Therefore, if we know this connection exists because of experience, we must have an experience of this causation - but he argues we DON'T! We experience one thing happening, then another, but NOT a physical cause itself
                    - Therefore, according to Hume, experience doesn't yield knowledge of these necessary cause-and-effect relationships!
                        - Our only hope of proving these causes are necessary is therefore if the universe is uniform - but we can't prove this deductively (since we don't have enough data to make universal claims), and we can't prove this inductively (since inductive arguments generalize from past to present, assuming the future is like the past already, giving us a circular argument)
        - So, Hume's arguing here that we have NO rational reason to trust our experiences, but we all do so anyway!

- So, Hume gives us a problem when dealing with inductive arguments
    - All inductive arguments take the following form:
        1) All F's we've experienced in the past are G's
        2) If some regularity has held in the past, it will generally hold in the future
        3) Therefore, all F's will be G's
    - To make valid inductive arguments, we need to prove that things that've held in the past will hold into the future, but Hume seems to make this impossible! He argues we have NO reason to believe in 2
        - Hume is NOT just saying "there's always a chance we might be wrong"; he's doing something far more radical. He's claiming these sorts of arguments aren't rational at ALL
    - According to Hume, then, we have no rational basis for induction

- In Section 7, Hume argues that our idea of "necessary connections" between an apparent cause and effect is ONLY an idea we form due to habit - that's it!
    - In essence, he's claiming our idea of cause and effect is nothing more than classical conditioning; we don't know about cause and effect because we're doing any reason, but because the human mind is biologically biased to expect patterns
        - Notice, here, that Hume isn't trying to justify our beliefs - he's given up hope on that - but is instead trying to think about how we got this funny little idea of "causation" in our heads to begin with (it's sort of a proto-psychology)

- So, is Hume right? Many philosophers have tried to respond to this, and Hume's apparent pessimism - but we'll skip over their responses for now

- Okay, we have two more "historical" readings that are less philosophical and more scientific, and on that depressing note: see you on Thursday!