Serious Philosophy Problem

Post Reply New Topic

Does anyone have a solution to Hume's Problem of Inductivism? I've been racking my brains for a solution but have yet to come up with one. The problem is as follows:

(1) We can only determine from past observations that the past has been consistent, not that patterns observed in the past will continue into the future because it is conceivable that the future will be different.
(2) Inductivism is not tautological like logical axioms are.
(3) Any attempt to justify inductive reasoning by reference to inductive reasoning begs the question.



This argument means that we cannot form even probabilistic conclusions, and was originally laid out by David Hume.

well, if all of todays people understood that this is a problem, there would be no misunderstandings about what science does and can do and what religion does and can do. so, i really hold you in high esteem for acknowledging the problem albeit not knowing a solution.

my personal take on it is that its one of the "workarounds" we do to cope in everyday life - and not to bad effect, given that all of science ignores the problem axiomatically and yields working results. its a similar workaround, in my book, to speaking of an objective world that does not stop being once the perceiving subject stops - which isnt a given at all, too.

i dont know what kind of solution you expect for this problem, though... i dont think its the kind of problem to be solved, its rather a statement on epistemological limitations.
im quite okay with what popper did on science theory: not acknowledging truth for any inductively reasoned sentence but taking it as a tool-to-be-refined from the beginning. popper explicitly speaks against verification but endorses falsification as a means to get better-working tools. (note: better-working does not mean they will eventually close the gap, because of their reliance on epistemological bases and hence, never being able to overcome humes problem).
that doesnt solve the problem really, though... its just a more honest treatment.

i think the best that can be done with this problem is acknowleding that reason and rationality doesnt need to be a viable solution to everything - much like schopenhauer states that the will itself, his basic principle of the world, is nonrational in its workings, and hence, doesnt have any causality to itself.

as for patterns appearing in the past/future repeating...well some folks never learn from their mistakes. Induction, I thought, had something to do with either heat or electricity.

Nice to meet you, IpsoRandomo.

Thanks for the responses. I'll definitely look into the book.

As for falsifiability, well, it's only a partial solution (e.g., how do you support the statement that all metals melt at a certain temperature?).

Someone said science gets results, but unless you refer only to falsifiability, then you beg the question.

I've recently thought that a possible solution could be derived by using falsifiability to verify evolution. If that is done, you could then argue that we had to reliably predict the future in order to survive, but I'm not sure about the soundness of such an argument.

Nice to meet you, IpsoRandomo.

Begging the answer? topic

How can we step in the same river twice?

The loopy nature of our world means the argument is not linear, or with a definite conclusion. It begs the circular argument.

Interesting. Thank you.

ipso:
yeah, poppers critical rationalism certainly isnt a solution - its merely a kind of add-on to science that makes it tackle the problem in a more honest way. the axioms of science basically carpet-bomb humes forumlation, because definite assumptions concerning the "situation" are made: natural law/causality is constant. period. through popper, you take that harshness out of it to a certain degree - you dont accept the possibility of science making completely true statements about the world; science boils down to a system for predicting outcomes with sufficiently high probability. once a theory doesnt yield good enough results anymore, its discarded...
that actually doeesnt tackle the problem itself at all, just adjusts scientific practise a tad. the moment science completely accepts humes problem, it will radically stop being useful because it wont be a good tool for making cars, computers and all our fancy stuff.

again, i am not advocating science as a way of assessing the true nature of things - but merely a system of yielding working results. the results being highly probable assumptions on the outcome of things, not statements on the nature of things.
not a world-statement like "all metals melt at a certain temperature" but a working-statement like "this metal melt at that temperature all the time, we can count on it to do the same when it reaches that temperature"

how would you try verifying a theory through falsification? its not that easy that you falsify creation and voila... you would have to falsify each and every other theory, including those that dont adhere to scientific standards at all - like a theory that proclaims complete randomness or a theory that evolution only happens on thursdays between 1 and 5 pm, with the goal on producing lifeforms that produce maximum amounts of confetti. just to give you a taste of what youd need to falsify (which would be quite impossible in a number of cases, by the way. the randomness theory is a pretty good example: the next incident could be the one that doesnt conform to anything. or the one after that... its empirically open.)
if you managed, by whatever means, to prove evolution, our predicting the future could as well just have been an epiphenomen of what was needed for survival, too.

you see, the world of empiry is full of instabilities, once you start asking.


pakled:
induction also is a conclusion mode - its opposite is deducting.
deducting is the old story of socrates is human, humans are mortal, hence socrates is mortal. the conclusion that socrates is mortal was deducted of the premisses - at the cost of essentially not saying anything new.
inducting is "this man is mortal" ergo "men are mortal" - the conclusion is much more useful, because the outcome is moúch more in scope than what you started with - at the price of it not being "secure" reasoning: you do this on the assumption that what you observed in the past will hold true in future cases, that what you label as "man" is consistent, etcetera... so with some refinement to the method, you can use inductive conclusions to make statements that will work with a good probability (thats what science is), but you wont ever get to the point that you make really sound and hard statements that way.

yeah i know - strictly in the lines of hume, nothing that is to happen is certain. period. (and in some contexts we are all too happy to accept that)
but then, from a very pragmatic point of view, a bare-bones utilitarian stance, science made things work. given that science is just a tool for making things work, id say that this point of view is somehow copeable. id say its healthy for mental hygiene to be aware of the problem, but a certain leeway in terms of "i dont know how it comes or why it does, but it worked so far *crosses fingers*" is needed for getting anything done. i dont say that its a "solution" to humes problem to accept science only for probable prognoses - but it somehow works. dont ask me why because i dont know and its ultimately just not "secure grounds" of any kind.