UnderGround Forums
 

PhilosophyGround >> Induction


9/10/05 12:38 PM
Ignore | Quote | Vote Down | Vote Up
vermonter
81 The total sum of your votes up and votes down Send Private Message Add Comment To Profile

Edited: 10-Sep-05 12:40 PM
Member Since: 01/01/2001
Posts: 5133
 
I'm trying to wrap my mind around solving the problems that Hume and Russell present to using induction in most of our decision-making. Generally, i accept semantic arguments being a lover of philosophy of language, but the semantical justification of induction does seem to cut it. The answer seems like it should just be right there, but i just plain can't there even when i feel so close. What do you guys think about this topic? -doug-
9/11/05 1:10 PM
Ignore | Quote | Vote Down | Vote Up
Dogbert
Send Private Message Add Comment To Profile

Edited: 11-Sep-05
Member Since: 01/01/2001
Posts: 14893
" but the semantical justification of induction does seem to cut it." Please clarify.
9/12/05 12:20 AM
Ignore | Quote | Vote Down | Vote Up
vermonter
81 The total sum of your votes up and votes down Send Private Message Add Comment To Profile

Edited: 12-Sep-05
Member Since: 01/01/2001
Posts: 5138
Dogbert, I think that a lot of problems between philosophers are just definitional, or semantical. Metaphysicians always argue about, say, the nature of objects, but they sometimes (often it seems) don't have any clue about what objects they are discussing. I realize that is a very vague way to say it, but i'm a bit rushed. The semantical justification for induction doesnt seem as successful to me though. It doesnt seem that the induction problem suffers the same problems. It goes something like this: PI (principle of induction): A->B A->B A->B ===== The next A will cause a B Of course Hume and Russell both claim that not only doesnt this garuntee a result (duh) but it also doesnt even give us a reason to even think that might. However! Some claim that these folks just don't get the definition of "probable." A probable result is one that conforms to PI. This is definitely a "cartoon" version of this argument to give you an idea. I can most certainly spell it out a little more nuanced, but perhaps you can think about it a bit in the meantime, or give me your current thoughts on the issue in general. Thanks. -doug-
9/12/05 12:37 AM
Ignore | Quote | Vote Down | Vote Up
Dogbert
Send Private Message Add Comment To Profile

Edited: 12-Sep-05
Member Since: 01/01/2001
Posts: 14894
I think I know what you mean. Because one cannot justify induction on external grounds, one goes on clarifying "inductive sentences". This is hard enough, induction brings some really though puzzles (Hempels raven paradox or Goodmans Grue paradox). I think Goodman gave the best justification of the semantic approach: The classical problem of how one can deductively justify induction is solved: One can´t. I tend to agree with this view. As long as there is no way in sight to treat the topic in a different fashion, the semantic approach seems to be right. A question that should IMO be treated in more detail is the role of induction-speak in communication. Why do we need these sentences at all if they don´t justify anything?
9/12/05 8:54 AM
Ignore | Quote | Vote Down | Vote Up
vermonter
81 The total sum of your votes up and votes down Send Private Message Add Comment To Profile

Edited: 12-Sep-05
Member Since: 01/01/2001
Posts: 5139
Well i think part of the confusion about when this issue might be close to being solved (in favor of induction) is that it is impossible to prove that inductive reasoning will garuntee a particular result. It's in the future. The sun might blow up, so it very well might not "rise" tomorrow. All we are trying to demonstrate, as philosophers, is that induction gives us a good ENOUGH reason to act a certain way, say drinking coffee in the morning instead of drano. We would all pick the drano, because we believe the coffee is good, and the drano will kill us. We believe this based on induction. The problem is that it could be REALLY hard to demonstrate something "good enough" instead of exact, or even know that you have done it. In any case, i'm going to stew over your post for a while. I think what you might mean is that we do justify the "morning beverage" case in a deductive way, somehow, but we talk about it, or even reason it, in an inductive way which is unecessary. If this is the case, though, how do we deduce it? -doug-
9/12/05 1:49 PM
Ignore | Quote | Vote Down | Vote Up
Dogbert
Send Private Message Add Comment To Profile

Edited: 12-Sep-05
Member Since: 01/01/2001
Posts: 14895
Actually, I don´t think inductive reasoning can be replaced by deductive reasoning. This puzzle really is a pain in the a..
9/12/05 8:22 PM
Ignore | Quote | Vote Down | Vote Up
vermonter
81 The total sum of your votes up and votes down Send Private Message Add Comment To Profile

Edited: 12-Sep-05
Member Since: 01/01/2001
Posts: 5140
I'm not 100% sure what you're saying. Just maybe using language theory to get rid of this bit of useless junk? Problem is, we reason this way, probably even without language. -doug-
9/13/05 1:34 AM
Ignore | Quote | Vote Down | Vote Up
Dogbert
Send Private Message Add Comment To Profile

Edited: 13-Sep-05
Member Since: 01/01/2001
Posts: 14896
I don´t think one can defend induction on non-inductive grounds.
9/21/05 6:11 PM
Ignore | Quote | Vote Down | Vote Up
FudoMyoo
Send Private Message Add Comment To Profile

Edited: 21-Sep-05
Member Since: 01/01/2001
Posts: 12560

In my recent course on Epistemology and Semantics, I just read an interesting article by Harman, defending some cases of Induction, which he calls inference to the best explenation (which is actually abduction). I´m not sure if it survives goodmans grue-paradox though.

One could also argue that perhaps you shouldn´t judge inductive arguments from a deductive "POV", or with other words; a solid inductive reasoning could perhaps give good reasons to believe something even if the argument isn´t 100% solid like a deductive argument.

9/21/05 9:12 PM
Ignore | Quote | Vote Down | Vote Up
vermonter
81 The total sum of your votes up and votes down Send Private Message Add Comment To Profile

Edited: 21-Sep-05
Member Since: 01/01/2001
Posts: 5189
Fudo, Check out my first post again. Noone thinks (or they shouldnt anyway) that induction is guaranteed. The debate is about it being good enough to base our actions on. -doug-
9/22/05 5:14 AM
Ignore | Quote | Vote Down | Vote Up
FudoMyoo
Send Private Message Add Comment To Profile

Edited: 22-Sep-05
Member Since: 01/01/2001
Posts: 12561

"The debate is about it being good enough to base our actions on. "

I`m aware about that. or actually not only our actions, but if induction gives "good reasons" to believe a certain conclusion, that many examples seem to confirm. I think it boils down to if we can accept "good reasons" to be fallible, which Hume and Popper couldn´t accept.

9/25/05 2:11 PM
Ignore | Quote | Vote Down | Vote Up
vermonter
81 The total sum of your votes up and votes down Send Private Message Add Comment To Profile

Edited: 25-Sep-05
Member Since: 01/01/2001
Posts: 5215
yeah... Still, i want to work on this issue. It seems like the answer is within reach.... Perhaps something about how the meanings of words are tied to concepts (in my own theory anyway) rather than things in the world, and the concepts exist now, whereas the events you are working through inductively are outside of your knowledge. The induction is actually based on physical structures in the brain, not "about" some unknown event. Just throwing it out there, for whatever sense it makes. Basically the problem stems from the fact that we are making claims/ founding beliefs about things we dont have knowledge of. I'm trying to ground those beliefs in something that does already exist. -doug-
10/26/05 1:47 AM
Ignore | Quote | Vote Down | Vote Up
GoToSleep
Send Private Message Add Comment To Profile

Edited: 26-Oct-05
Member Since: 05/06/2003
Posts: 292
Well Hume's whole conclusion, after going to great lengths to expose the insufficiency of inductive reasoning to justify "Matters of Fact" is that, even though we can't justify it, we will act in accordance with it regardless. Sure, there's no justification for not putting your hand in a rat trap because in the past, it has snapped down, but we still won't put our hands into the rat trap. Hume claims that even though we can't justify it, it's our psychological makeup that refuses to allow us to act in any other way. We used inductive logic to understand how the world worked long before we were old enough to have any conception of doing such a thing cognitively. Kind of like, "We lack justification for it inductively, but so what?" Because we're going to do it anyway, justified or not. That seems to be the conclusion that Hume reaches, not addressing the implications of his conclusion that he states at the end of An Enquiry into Human Understanding.
10/26/05 9:56 AM
Ignore | Quote | Vote Down | Vote Up
FudoMyoo
Send Private Message Add Comment To Profile

Edited: 26-Oct-05
Member Since: 01/01/2001
Posts: 12606
welcome to the forum GoToSleep! Nice with some new posters here.
10/26/05 11:17 AM
Ignore | Quote | Vote Down | Vote Up
vermonter
81 The total sum of your votes up and votes down Send Private Message Add Comment To Profile

Edited: 26-Oct-05
Member Since: 01/01/2001
Posts: 5435
Gotosleep, Because we reason inductively, and always have, explains why induction a "human" thing, but it doesn't explain why it's a "rational" thing. Hume doesnt beleive (as far as i can tell) he's reached any real conclusion. His very quandry is the fact that it seems so natural to reason inductively and yet it's irrational to do so. If it's something we do all the time, we should hope it isnt irrational. -doug-
10/26/05 12:05 PM
Ignore | Quote | Vote Down | Vote Up
GoToSleep
Send Private Message Add Comment To Profile

Edited: 26-Oct-05
Member Since: 05/06/2003
Posts: 293
He never actually resolves the problem. And to my knowledge, the problem still hasn't been resolved. His answer to the problem is that, yes, it is irrational, but because of our psychological makeup, we do it anyway, and so will anyone skeptical of his claims. No matter what skepticism you may espouse in a philosophical setting, as soon as you leave that room, you are living your life based on inductive reasoning about "Matters of Fact," which he then says is unjustified and unavoidable. In the only thing which could even be seen as a conclusion, Hume seems to pull out of the book is that this knowledge of our mental fallacies may perhaps serve to make us humbler, and that those who have honed their mental faculties through years of schooling may no longer take such pride in their advantages and blah blah. That conclusion is unimportant; what matters is the irrationality of induction part.
12/2/05 2:15 PM
Ignore | Quote | Vote Down | Vote Up
hekster
Send Private Message Add Comment To Profile

Edited: 02-Dec-05
Member Since: 01/01/2001
Posts: 4724
It boils down to: Certainty only exists in the realm of argument, which is not the world that we live in. Induction is a process that inherently involves experience and experimentation. Trying to encompass experience by debate is an exercise in futility. The interesting question is: why, if every living thing sustains itself through a process of induction, is argumentation a superior method of knowledge acquisition? Because you can make a proof for it? Because we sent a man to the moon? Because we built a nuclear weapon? Isn't the method that enables life to exist more credible?
12/4/05 1:28 PM
Ignore | Quote | Vote Down | Vote Up
vermonter
81 The total sum of your votes up and votes down Send Private Message Add Comment To Profile

Edited: 04-Dec-05
Member Since: 01/01/2001
Posts: 5706
Hekster, Humes point is that we use it and it does work, but it shouldnt. That's the problem. We all want to know why. -doug-
12/4/05 7:44 PM
Ignore | Quote | Vote Down | Vote Up
hekster
Send Private Message Add Comment To Profile

Edited: 05-Dec-05 12:11 AM
Member Since: 01/01/2001
Posts: 4727
It works because the world is constantly shifting and changing and induction is an adaptive process. Theoretical knowledge is only good for static things. Trends are more important for interacting with your immediate environment than theoretical logic rules.
12/5/05 11:13 AM
Ignore | Quote | Vote Down | Vote Up
vermonter
81 The total sum of your votes up and votes down Send Private Message Add Comment To Profile

Edited: 05-Dec-05
Member Since: 01/01/2001
Posts: 5710
Uhhh... Hume's argument was that induction works because it appeals to the Principle of Uniformity in Nature. Induction works because what has happened before a bunch of times is likely to happen again. I think you might have induction a little backwards. If things were less "static" and more random, induction would fall apart. -doug-
12/5/05 11:13 AM
Ignore | Quote | Vote Down | Vote Up
vermonter
81 The total sum of your votes up and votes down Send Private Message Add Comment To Profile

Edited: 05-Dec-05
Member Since: 01/01/2001
Posts: 5711
Uhhh... Hume's argument was that induction works because it appeals to the Principle of Uniformity in Nature. Induction works because what has happened before a bunch of times is likely to happen again. I think you might have induction a little backwards. If things were less "static" and more random, induction would fall apart. -doug-
12/5/05 10:14 PM
Ignore | Quote | Vote Down | Vote Up
hekster
Send Private Message Add Comment To Profile

Edited: 05-Dec-05
Member Since: 01/01/2001
Posts: 4729
well lets look at the assumptions in that argument. 1. Nature is uniform. If this were true, then there would be no need for induction. Everything would just evolve to the static rules that existed. Hume probably wasn't aware of the multiple mass extinctions in history or the lifespan of stars, etc. Induction looks for patterns in immediate experience, not rules inherent in the universe. As a lifeform, you may need to fish or farm or eat plants or fly around and eat bugs. When the Ice age hits, the lifeforms with the greatest flexibility to inductively perceive new patterns survive. Does that make sense?
12/6/05 8:16 AM
Ignore | Quote | Vote Down | Vote Up
FudoMyoo
Send Private Message Add Comment To Profile

Edited: 06-Dec-05 11:29 AM
Member Since: 01/01/2001
Posts: 12689

"When the Ice age hits, the lifeforms with the greatest flexibility to inductively perceive new patterns survive. "

Hume would probably not deny that practical value of inductive everyday thinking. or it´s value from an evolutionary perspective (if he was alive).

 

"Induction looks for patterns in immediate experience, not rules inherent in the universe"

what is the difference between "patterns" and "rules"?

"well lets look at the assumptions in that argument.

1. Nature is uniform. If this were true, then there would be no need for induction. Everything would just evolve to the static rules that existed. "

 we still need to differ between real laws of nature (if they exist) and other generalized true statements, that looks like laws of nature. how do we do that?

 for example, if I drop a stone, it falls to the ground because of what most would call a law of nature. But even if we could formulate a generalized true rule that said: "Every time hekster posts at Philosophyground, it´s something interesting", it wouldn´t be a law of nature making that true.

12/7/05 12:19 AM
Ignore | Quote | Vote Down | Vote Up
hekster
Send Private Message Add Comment To Profile

Edited: 07-Dec-05
Member Since: 01/01/2001
Posts: 4730
"for example, if I drop a stone, it falls to the ground because of what most would call a law of nature. But even if we could formulate a generalized true rule that said: "Every time hekster posts at Philosophyground, it?s something interesting", it wouldn?t be a law of nature making that true." Yours is a question of hard science vs. trends. It isn't that hard to distinguish between them. One remains the same when you manipulate variables and the other has no variables to isolate. What was said before was that "Hume's point is that we use it [induction] and it does work, but it shouldnt." When it is said that induction shouldn't work, this means that the thinker is setting deductive logical processes over trend logic (induction). We have induction because of the environment we live in contains trends. It is neither a superior nor an inferior logic, it is a logic with a specific application. I haven't read Hume in some time, but I understand it as a form of extreme skepticism relating to trends. My point was that things are constantly changing, including either becoming more stable or less stable. Induction works on patterns that may or may not be temporary. If I find turtle eggs on the beach 2 yrs in a row, and I happen to be in the vicinity on the 3rd year, I may as well check it out. If the pattern discontinues (the turtles became extinct during a sunami or whatever) induction abandons the practice after it becomes obvious that the pattern was temporary. It is flexible for windows of opportunity both small and large. If you look at genetics, chromosomes are basically trend trackers with experimentation built in. Its a feeler system for order, even temporary order. Induction is the logic of life. Hard logic is for physics and whatnot. So to ask why induction works is to ask what it is for, in my eyes.
1/3/06 2:31 AM
Ignore | Quote | Vote Down | Vote Up
Magnus
4 The total sum of your votes up and votes down Send Private Message Add Comment To Profile

Edited: 03-Jan-06
Member Since: 06/08/2002
Posts: 2527
"I don?t think one can defend induction on non-inductive grounds" Circularity is one of the problems. It is set within this context: consider the truth of the statement "All crows are black." This is based on a) seeing a great number of black crows b) our not having seen any crows of another colour. How is one logically justified in proceeding from 'some' to 'all' since not all crows have been observed? Choices: deductively? Nope. Inductively? (The future will resemble the past because this is how it happened in the past i.e. the future (in the past) resembled the past). Hmmmm . . . Hume was pointing to the limitations of rationality. Maybe some do not accept these Humean categories of reasoning? For example, certainly some law professors believe that so-called legal reasoning is distinct and not simply a poor form of inductive reasoning. Legal reasoning tends to be result orientated. Is that simply a poor form of inductive reasoning?

Reply Post

You must log in to post a reply. Click here to login.