…and this is a 
Pompous git solves the problem of induction … with Jesus!
Category: Kooks
Posted on: April 21, 2007 10:52 PM, by PZ Myers
Wow, but this is awful. Don't watch it unless you're feeling masochistic.
It's a snotty, arrogant punk kid filmed in annoying style claiming that he has disproven atheism and that all science is based on theology. I think he might be something like a freshman philosophy major who has just discovered the problem of induction.
The problem of induction is a real one, all right; we can't logically support one of the fundamental tools of science, the idea of making general inferences from specific observations. You might think, well, it's worked so far and we've got all these successful instances of science deriving useful principles from data, but that's an example of inductive reasoning itself, and you're trying to demonstrate that that kind of reasoning is valid, so you can't use induction to prove induction. It's an interesting philosophical problem, it has more or less stumped greater thinkers than me for hundreds of years, and no, I sure don't have an answer.
One reason that the video is so awful is that here is this hugely difficult problem, and snotty punk kid is offering ONE THOUSAND DOLLARS for a solution. Uh, right. Imagine, I'm sitting here with the solution to a well-known thorny and fundamental problem in philosophy … and I've just been waiting for an arrogant Christian to offer me one week's pay to publish it. Did you know I also have the protein folding problem solved, and I'll publish it as soon as the church down the street gives me a plate of cookies? Here's a greater inducement: publish your solution to the problem of induction right here in the comments, and I guarantee that you'll be able to stop by any philosophy department at any university in the world and the faculty will line up to buy you a beer. That will add up, you know.
The acclaim as a philosophical god among scholars might also be worth something.
The other reason it is annoying is that the snotty punk kid is babbling out some rather fatuous logic of his own; he's demanding that others accomplish an extremely sophisticated philosophical task, while exhibiting no awareness of his own inane reasoning and unexamined epistemology. One general assumption of scientific induction is that we live in a lawful universe (an assumption that we cannot demonstrate inductively, as already said). What young master Snottypunk does to get around the problem for himself is to declare by simple fiat that he has a different assumption, that there is a divine lawgiver, who enforces the lawfulness of the universe, and that the source of his information is the book of Genesis.
It's a cheat. He has absolutely no logical, philosophical justification for this divine precondition he has pulled out of his butt, but then he turns around and thinks that he's got atheists over a barrel and demands that they justify the use of induction without Jesus. What? Why can't I just invent an accidentally linear seam in the fabric of the 18th dimension that imposes regularity in our dimension by subspace resonance? It's total nonsense, but it's a justification that's on a par with waving your hands over an ancient Hebrew sky-god. How about if I pretend there is a subatomic particle (or maybe a sub-quantum force; does it matter?) called the Regulon that compels lawful behavior in other particles/forces. Again, it's pseudoscientific magical BS, but it's as good as Snottypunk's excuse. I know! A variant on the anthropic principle—the universe is lawful, because if it weren't, we wouldn't be here to speculate about why induction seems to work.
I don't think any of my explanations will convince any philosopher anywhere to buy me even a single beer, but oh, well.
The challenge this kid is offering is a pretentious joke. He has even less of a philosophy background than I do—he's got a bachelor's degree in religious studies, and he's a typical Texas son of Republican privilege, full of himself and stuffed to the snoot with unquestioned arrogance. His name is Kelly Tripplehorn, and I thought I'd heard of him before … and I have. Behold the deep thoughts of Mr Tripplehorn.
You know, even if you do solve the problem of induction, I wouldn't trust that pompous weasel to cough up. At least I'll buy you that beer, though, so don't hold back.
Comments
A thousand dollars is kind of paltry, and I don't like beer. Now if someone was offering a pony....
Posted by: abeja | April 21, 2007 11:05 PM
Do you think I can get $1,000 for "We take induction to be axiomatic, so that we can get interesting work done"?
Posted by: todd. | April 21, 2007 11:11 PM
Deal. You give us the solution to the problem of induction, and we'll pass the hat around here and come up with enough cash to buy abeja a pony.
I hope I don't have to get into a bidding war with the real philosophers at scienceblogs over this...
Posted by: PZ Myers | April 21, 2007 11:11 PM
No, and no ponies for that, either.
Saying Zeus is the source of all regularity also isn't acceptable.
Posted by: PZ Myers | April 21, 2007 11:16 PM
I'm only a philosopher in training, but I'll buy you a beer to hear your solution to the problem of induction. Hell, I'll buy you dinner just to thank you for fighting the good fight. I'd say you're on the side of the angels, but that seems somehow wrong.
Posted by: Stewart | April 21, 2007 11:18 PM
What a tool! It's like he read the premise behind Karl Popper's philosophy, but didn't bother to read the suggested resolutions to the problems. The uniformity of nature is not a stumbling block of any sort. To the pragmatist, the uncertainty involved in an induction is handled with the requirement that assertions be made falsifiable. To the positivist, the problem is one that is so small as to be dismissed out of hand. Wittgenstein once noted that the world must be stable to a certain degree, as evidenced by our ability to count. Rescher noted that the evident cognitive systemizability of the world necessitates an ontological stability.
In either case, we don't need to invoke a creator to account for the stability of natural laws. We need only invoke our ability to think -- that is evidence enough that the universe is stable enough to admit some kind of reasonable description.
Incidentially, that fork did not fall at exactly 9.78m/s^2, and for any number of reasons. For starters, that's because 9.8m/s^2 is an acceleration, not a velocity (watch your units, chucklehead). For another, the Earth's gravity is not uniform. Varying density in the crust beneath your feet can cause measurable distortions in that kind of thing. The Pikes Peak batholith, for example, is massive enough that satellites have to adjust their orbits when passing near it. What kind of rock is this guy on that he can say he knows exactly how fast that thing is going to fall? For another thing, that's a plastic fork. It has a large surface area without much mass. So, sure, the Earth's gravity will try to make it accelerate at 9.8m/s^2, but the rate at which it falls is another question altogether.
And, by the way, we know his fork statement is wrong because of SCIENCE!
Posted by: Dustin | April 21, 2007 11:19 PM
I have discovered a truly marvelous proof of the problem of induction , which, however, this comment is not large enough to contain.
Posted by: JJ | April 21, 2007 11:20 PM
Well, we could also invoke Popper in this instance, who proposed essentially that the problem of induction was a non-problem because induction itself was a fictitious concept.
From my own view, I don't see why there is an overarching problem of induction and not a corollary for deduction. The same logic applies. The validity of deduction cannot itself be deduced (i.e., it cannot recursively prove itself).
I take a Quinean view myself, which I understand to essentially be that neither inductive nor deductive logic are concrete, but provisional, tentative and subject to change with experience. Presupposing certain truths as axiomatic and immutable and then deducing conclusions from them is really more the domain of theology than science.
Posted by: Tyler DiPietro | April 21, 2007 11:23 PM
Isn't the fact that we get useful, practical results from induction proof enough? It's at least proof that it's useful, just because we can't figure out how we got from 1 to 2 strikes me as just a variation on Zeno's paradox.
See, anyone can do navel gazing.
Posted by: Keith | April 21, 2007 11:26 PM
That's the kind of kid who can piss away a family fortune in a couple of years, unless his parents were smart enough to tie it all up in a family trust that pays him an allowance.
-jcr
Posted by: John C. Randolph | April 21, 2007 11:27 PM
I think JJ's comment is beer-worthy. Not pony-worthy, but a nice beer. Maybe a Chimay, because there's a nice symmetry in celebrating the short-comings of the religious with one of their finest achievements.
Posted by: todd. | April 21, 2007 11:28 PM
After reading PZ, I decided not to watch the video. But I did go to the video on YouTube to see if there were any video responses. There were two (so far: AtheistPaladin and some philosophy major) and I watched them both. They both ripped him just as bad as PZ and both had ratings of 4-5 stars. Mr. "Tripplehorn's" rating: 1 star. Maybe there's hope for YouTube yet.
Posted by: The Science Pundit | April 21, 2007 11:31 PM
I assume that since my fork has always fallen w/o supernatural intervention that it will always fall w/o supernatural intervention.
Posted by: Les Lane | April 21, 2007 11:32 PM
I don't understand what Kelly Tripplehorn wants. What exactly must one do in order to get $ 1,000 dollars from him? On his company's website, it says this:
"All you need to do in order to collect your $1,000 is get your non-theistic answer published (concerning your epistemological warrant for your inductive inference) in the Stanford Encyclopedia of Philosophy, under its heading The Problem of Induction."
The Standford Encylopedia of Philosophy won't be publishing anything under the heading The Problem of Induction. So, no one is going to be winning 1,000 dollars. But if Tripplehorn would write down exactly what he wants done, I might try to do what he is asking for. Does he write it down any place? Or is there any place in the video when he says exactly what he wants?
Posted by: Ed | April 21, 2007 11:32 PM
I'll gladly send Tripplehorn $100 if he promises to use it to buy a tripod (and a tie!). That motion-sickness camera shot really only belongs in a Serious Documentary, and any random moody TV cop show.
Posted by: Penon | April 21, 2007 11:40 PM
You can either take it to its logical conclusion (ala Hume), or turn to pragmatism/instrumentalism.
I'm not masochistic enough to watch the video, but these things usually boil down to unrealistic notions of what it means to "prove" something in the real world.
The central problem of philosophy since Thales has been to reconcile mathematical certainty with an empirical world. The mathematical notion of proof was one of the great leaps forward in imagination, perhaps even as a singular event. It also led philosophers on what must be the longest wild octopus chase in history.
Posted by: efp | April 21, 2007 11:42 PM
Uh, I think the real reason why this Christian moron was fired was because he was stealing all of the markers out of the supply closet to sniff them.
Posted by: Stanton | April 21, 2007 11:43 PM
Smug git.
He's so sure that he's backed science into a corner.
Anyway, doesn't he know that there aren't any true atheists left now that the 'atheist's worst nightmare' (aka the banana) has convinced every last one of us to accept Jesus Christ as our personal savior?
Posted by: Christian Burnham | April 21, 2007 11:43 PM
Oh, but hey, it wasn't so bad for him, poor Kelly, that affair with the unfortunate email. It made him become born again!
http://i53network.org/Home_Page.html
Posted by: Rien | April 21, 2007 11:43 PM
George Berkeley came up with a much better version of this argument a couple of hundred years ago... If God's not a liar, then induction is justified, because our perceptions are signs that stand for other signs. (Fire is a sign of heat: God's promise that if you stick your hand in you'll get burned.)
I guess the biggest problem with it remains the same, even with the haphazard presentation: it offers a very contentious postulate (the existence of a God who is no liar) to justify a much less contentious one (induction can give us some (imperfect) knowledge about how the future will go.)
Congratuations, kiddo. I recommend you not send this video along with your application if you're trying to get into a graduate program in philosophy.
Posted by: Dennis | April 21, 2007 11:45 PM
Tripplehorn says (1:53) "while induction certainly does not prove God, much less Christianity, it does though disprove nontheism." Hmm. I'm pretty sure that the disproof of nontheism might require some kind of god as a counterexample. Perhaps this boy is just too clever for me and has found a way around the excluded middle, or Tripplehorn knows of some "third way" between God and non-God. Cool.
I have to admit that Tripplehorn has neatly solved one type of induction problem: He has mastered the skill of inducing nausea. (No doubt his former girlfriend had previously detected this skill of his.)
Posted by: Zeno | April 21, 2007 11:54 PM
What happens if I try to invoke Eris as the source of the stability of natural laws?
Posted by: Dustin | April 21, 2007 11:56 PM
OK, I'm game -- I'm no longer a working scientist (experimental particle astrophysics -- try saying that at a party!) -- but I am a science teacher (high school physics).
As I say to my students, the beauty of science is that it doesn't make any assumptions. We need not assume that nature is uniform. In fact, probably the most exciting thing that can happen to a physicist is if a fork suddenly flies straight up -- that gives us something to try and figure out. Science will do with a flying fork exactly the same thing it does with a falling fork -- attempt to explain it and fit it into a framework. If the framework isn't known, perfect -- full employment forever! If it can be proven that such a framework doesn't exist (as with the proofs that local hidden variables are inconsistent with what we observe in quantum mechanics), great -- we've got at least the beginnings of a framework.
But we still have to answer the question -- does science require a framework? No, not at all. Science does not, can not, and never will offer truth. Science can only offer humankinds best approximation of the truth of the physical world. Since we acknowledge already, at the very start, that science has no truth, we need not worry about why it has not proven truth.
Science is but a bunch of sticks put together by humans that allow us to reach a little bit further into the air. To say that this edifice is not built on bedrock is ridiculous, since that is plain to all who look. You might as well criticize water for being wet.
But here, I can just sense the rebuttal -- why then do scientists say that some ideas, some people, are just plain wrong? I had a student's mother complain about how something's wrong with scientists -- if they are so open minded, why can't they acknowledge the possiblity that biblical creation really happened?
The answer, of course is that while science
doesn't know everything, it does know something. It knows that there is a framework (evolution, both cosmic and biological) that explains just about everything we have observed, and makes predictions that point out more possibilities. Creation, on the other hand, creates a framework that conflicts with many observed facts, and predicts nothing more. While evolution may be wrong (and I'm sure that it is, in niggling details here and there), creation is most definitly wrong.
Oh, and the other reason scientists get snappy when somebody asks them to admit the possibility of biblical creation? Because we are damn tired of hearing and rejecting the same old tired arguments from people that don't listen to ours.
(Sorry I didn't focus on more of the philosophical jargon, but damn it, I'm a working teacher / was a working scientist -- I don't have time to learn a new language.)
Posted by: spudbeach | April 21, 2007 11:57 PM
Damn, I totally missed that, Zeno! This kid has been chatting up Graham Priest! Or...
Wait...
Is this Alvin Plantinga with a snot-nosed-kid mask?
Posted by: Dennis | April 21, 2007 11:58 PM
I guess there are two versions of this video on YouTube and I came across this almost beautiful response earlier in the day.
http://www.youtube.com/watch?v=8G3osI7Nrw4&watch_response
Posted by: jufulu | April 22, 2007 12:04 AM
spudbeach,
You really do need something like a principle of uniformity of nature in order to even engage in scientific enquiry. Without it, upon seeing a fork fly straight up, all you'd have is another event that wouldn't necessarily tell you anything about anything but that very event.
If you're interested in the problem of induction, Bernecker and Dretske's book "Knowledge" has a few good readings, including my favorite, Reichenbach's pragmatic justification of induction. (Basically: we have an interest in predicting the future. If any method will allow us to do this, induction will, because if any other method worked (tea-leaves, chicken entrails, even counter-induction), we would be able to induct on it.)
Posted by: Dennis | April 22, 2007 12:04 AM
Wow! If only this punk were my Philosophy of Induction professor. Imagine how that class would go.
Posted by: Skemono | April 22, 2007 12:04 AM
I'm not sure the problem of induction is a real problem myself, but I won't get into that here. If it were a real problem, however, it's of obvious note that theism doesn't solve it. After all, where do religious "truths" come from? "Divine revelations." And how do we distinguish "divine revelations" from run of the mill hallucinations? By induction, of course -- a theist will tell you that certain revelations are so amazing that it is UNLIKELY to be hallucination. That's an induction. Theism is based on induction itself, so of course it is nonsense to turn it around and say induction is based on theism. If that were philosophically acceptable, we might as well cut out the theism middle-man and say induction is self-justified.
Posted by: AL | April 22, 2007 12:06 AM
Ah, dang. That last "student" should be "punk". Well, I imagine y'all got the gist of it.
Posted by: Skemono | April 22, 2007 12:08 AM
Induction is a great topic for philosophy-- but less so for religion. 'Goddidit' doesn't work any better on this gap than it does on others. There are a couple of constructive responses that haven't yet wound up in the trash-- one elegant line is to throw the challenge back at Hume: He essentially criticized induction for failing to be deductive. But inductive reasoning may be deductive after all, once we have adopted the right view of just what the reasoning involves: If the conclusion (after seeing so many green emeralds) is not 'all emeralds are green' but 'it's rational to accept (for now) that all emeralds are green', and we have an account of rational acceptance (including a characterization of the relevant epistemic goals) that supports this conclusion, then the practical decision to accept the inductively supported conclusion is perfectly justified as the conclusion (practical result if you prefer) of a practical syllogism... W. Sellars' Induction as Vindication (a tough read) adopts a form of this approach that I particularly like, but there are others exploring this kind of line.
Posted by: Bryson Brown | April 22, 2007 12:14 AM
Perhaps physics isn't the centerpiece of his presentation, but given that he apparently doesn't know the difference between velocity and acceleration (he thinks objects "fall toward the ground at 9.78 meters per second squared"), I'm not sure how much trust I should put in his dissertation about "modern science."
Then there's the fact that Mr. Tripplehorn's a twittering tool. Since this clip involves him talking out of his asshole from begining to end, modern science and the uniformity of toolery imply that everything he's ever said has been both conceived behind and expelled from his asshole, and that this pattern will hold as long as he owns an asshole and produces videos. By God.
Posted by: kemibe | April 22, 2007 12:17 AM
Wow - from Reins post:
http://www.snopes.com/embarrass/email/tripplehorn.asp
This kid made snopes! He's like DaveScot Junior!
Posted by: Rich | April 22, 2007 12:17 AM
He needs to go back to the harassment. He was better at it.
I live in the eternal now, where the present always resembles the present, so the problems of induction don't concern me.
Posted by: CalGeorge | April 22, 2007 12:17 AM
Serious question here:
Doesn't the problem of induction sort of fall away once Einstein figured out that time is part of the fabric of the universe, and that physical laws don't happen *in" time, but rather time is merely an aspect of those laws? I find it rather meaningless to say that we can't supposed the laws of the past are the same as the laws of the future, since time itself is one of those very laws.
Posted by: Ethan | April 22, 2007 12:21 AM
Spud, I think that science needs some kind of stability. The pragmatist Rescher says:
Now, of course you would be right to look for a reason for that fork to fly off of the table. The reason you would look in the first place is that you would judge it, based on your long experience of a stable world, to be peculiar and in need of an explanation. You would also expect, by that very stability, an explanation to exist.
Beyond a certain degree of stability, though, I don't think anything else is required -- certainly not the absolute stability our sweaty armchair philosopher in the video seems to want. That the world will probably be mostly the same is enough. Yeah, the kid doesn't think so and flounders around with a probability argument for a bit, but look: the functionality of probability necessitates a certain degree of uniformity. Probability seems to work. Thus, the universe is very probably stable, and will very probably remain that way. That's all science needs.
Posted by: Dustin | April 22, 2007 12:22 AM
What is, "If you cannot dazzle them with your brilliance, then baffle them with your bullshit ?"
No I am sorry, it was, "Get your parents to ruin their lives with a few quick phone calls"
Posted by: Lago | April 22, 2007 12:23 AM
Dennis:
Why should I require a belief that I will be able to explain events and connect them? That's like saying that I have to have a belief in a car before I can make a wheel. Why can't I just do my best, in the hope that it will be useful, but without any certain knowledge or even belief that it will?
Perhaps our difference of opinion lies in what we think science is. I think of science as a human endeavor that has proven to be useful and fun in the past, and that may prove to be useful in the future. Humans can do all sorts of things for any reason or no reason at all.
Now, to take the results of that endeavor and label it science and endow it with authority is, in my book, going to far. To give science a "seal of approval" would require that we do things like assume that there is uniformity, and that scientific laws will be the same today as they are tomorrow, and the same there as they are here. I'm not willing to take that step. I merely claim that science is fun and potentially useful. If others want to take say "there is no way that bridge could fall down", I just reply "at least according to our understanding of the universe", and let it be just one more test.
(P.S.: I feel like we're both on the same side here, and should be enjoying this discussion over a pint somewhere. If PZ is buying, I'm there!)
Posted by: spudbeach | April 22, 2007 12:27 AM
I came up with an airtight solution to The Problem of Induction, but then on a whim, God decided to make things different, and he turned my solution into a grasshopper. Goddammit!
Posted by: The Science Pundit | April 22, 2007 12:29 AM
Uh... the kid's a Presuppositionalist. The argument he's giving is essentially the Transcendental Argument for the Existence of God (TAG). There's no making sense with these people.
Posted by: Zachary Moore | April 22, 2007 12:30 AM
"Uh... the kid's a Presuppositionalist."
Are you assuming this? Or do you know the guy?
Posted by: DamnYankees | April 22, 2007 12:31 AM
Oh, and induction is basically used as an expression of statistics, as science itself is an expression of the same...
Posted by: Lago | April 22, 2007 12:32 AM
I don't see any problem with induction. It's just something you give philosophers to keep them busy while scientists get on with the business of trying to work out what make the Universe tick.
Posted by: Ian H Spedding FCD | April 22, 2007 12:33 AM
No. General Relativity is a scientific theory like any other. It is constantly being put to the test. Just because it's predictions have held up to experimental testing every time so far doesn't mean that GR won't fail tomorrow. You still need induction.
Posted by: The Science Pundit | April 22, 2007 12:35 AM
I think all of you are missing the obvious here. The principle of induction, in the literal sense he seems to be using, is false. This leads immediately to the following little syllogism: If (1) Christianity implies "induction", and (2) "induction" is false, then Christianity must therefore be false!
See? All you need to do is use a little deductive logic and you have a tidy little disproof of Christianity.
Posted by: Kurt | April 22, 2007 12:37 AM
A Proof of Induction?
Here's your proof of induction. I'm a math major, not a philosopher, so I can't speak for the stuff at the beginning and the end, but the math part in the middle checks out! Only problem is, it uses the Axiom of Choice, and that usually seems to introduce more questions than it answers, plus it's a nonconstructive proof, so not all that practical when it comes to actually APPLYING it.
('Course, I've been known to say that math is a religion, so maybe this proof invokes Jesus after all!)
Posted by: Susan B. | April 22, 2007 12:39 AM
"You still need induction."
Of course we need induction. But I don't see how the *problem* of induction exists, since it seems to beg the question. The basic problem for induction is that we can't simply assume that past conditions will persist in the future. But now we know that "past" and "future" are far more complex and subtle problems than Hume did. Past and future are mere properties of a closed system - they are the way we simply percieve the laws of the universe, just the same we we observe gravity or electromegnetism. Our conception of the universe actually requires that the future is the same as the past, because if not they wouldn't mean what we think they do.
The opposite of uniformitarianism is illogical and contradicts the evidence.
Posted by: DamnYankees | April 22, 2007 12:39 AM
If you can't use logic to support the induction of science, doesn't that indicate a problem with logic, not with science?
Why is logic presumed to be "superior" to empiricism? Given the well-known problems with axiomatic systems it seems like the reverse is true.
Posted by: Chet | April 22, 2007 12:43 AM
Maybe I am missing something other than my car keys, here. Has induction ever been used as a proof of anything? I can use it to make a reasonable assumption, and if further evidence is needed I can make a testable hypothesis out of that which I induce.
Long ago on the radio I heard (sorry Mr. Salem) an engineer claim that we could use the principle of uniformity to prove that an old earth, and thereby evolution, is impossible. "If you dig deeply enough." said Mr. Engineer to the witless host, "you will find that the Sun is shrinking. Using the Law of Uniformity, simple math projects that the rate of the sun shrinking projected back in time would mean that only 400 million years ago the Sun would be as large as the Earth's orbit. No life could possibly survive such conditions. Most people don't know this, you know, you just have to know where to dig to get the facts."
So, from my understanding of what he is claiming, he is laying a red herring in front of atheism. I don't need induction to prove that they sun will rise in the morning, I can use it to make a prediction with a very high level of confidence. And I don't need to believe in God to achieve that level of confidence.
So, my question is: Why don't these guys take at least one class on statistical methodology or experimental design before they spout off with this stupid stuff?
Oh, and I am sure that we can look forward to a Senator Tripplehorn from the great state of Texas sometime in the next 20 years or so.
Posted by: Mike Haubrich | April 22, 2007 12:55 AM
Sorry, this will be my last post here tonight.
First, spudbeach, you can say that's how science is going, but then science isn't a tool, it's a game... and why should we be spending all this money/our lives on a game? No, science isn't mere stamp-collecting of events and then seeking patterns after the fact. Prediction is the name of the game, and if there's no reason to suppose we're getting true predictions, then there's no reason (other than shits and giggles) to engage in the enterprise of science.
Second, DamnYankees, you're putting your cart before your horse. Einstein's theory of relativity is based on inductive inferences from observed events, to which we applied a mathematical model that accounts for them. If induction weren't justified, then all we'd have is an interesting model for describing past events, but we wouldn't be able to draw any conclusions about what this might tell us about the future, or what it tells us about The Fundamental Structure of Reality(TM).
Posted by: Dennis | April 22, 2007 12:58 AM
Induction can't give you proof.
The discipline of statistics covers how much you should still rely on induction after you have counterexamples, and deals with having no counterexamples yet as a limiting case. Also, if you decide something by induction and you find that it's right 99% of the time, but later it seems to be right more like 98% of the time, statistics gives you an estimate based on sample sizes whether the original estimate of 99% was right or whether things have changed.
There's no question that induction is undependable, and statistics is concerned with just how undependable it is in particular cases.
We use unreliable induction because our best alternatives -- believing things based on our esthetic judgement of how beautiful it would be if we were right, or on intuition unconnected with experience -- are even less reliable.
Posted by: J Thomas | April 22, 2007 1:05 AM
Couldn't help notice the "Cornelius van Til" at the beginning, and thus catch a stinky whiff of presuppositionalism. I make a guess that some readers of Pharyngula might not be familiar with such things (though I might well be wrong about that.)
Typically presupper debates go about like this (link to 51 page, 3 year old, but still hilarious thread on iidb.org):
http://www.iidb.org/vbb/showthread.php?t=87833
Posted by: SteveC | April 22, 2007 1:18 AM
I don't see why everyone is so hung up on the problem of induction. Deduction has the exact same problem. Try and prove deduction without begging the question! Hurf hurf durf~~
Posted by: Taylor Murphy | April 22, 2007 1:25 AM
Susan, I didn't find that paper to be particularly helpful. Mathematical induction is not the thing in question here, as that is highly empirical. Induction is an immediate consequence of the integers and, as the paper pointed out, if we assume the Well-Ordering Principle, we can extend the idea of mathematical induction to things other than the integers (that's exactly where the Axiom of Choice was used -- the author needed a well-ordering to establish an induction on sets other than the integers). I don't regard that as a problem. Constructive proof is always better, but I don't regard it as being necessary for the establishment of a result.
The problem is not the mathematics, but the use of the mathematics. As soon as we're making recourse to mathematics, we've made the assumption that they're giving a reasonable description of the system in question. That itself is making implicit appeal to the stability of the world around us. Not to mention that the highly idealized sets in that paper likely have little bearing to the real world.
Anyway, the point is that mathematical induction is a valid and provable procedure. It's the bearing that it has to the real-world that is questionable.
Posted by: Dustin | April 22, 2007 1:27 AM
I said:
I meant:
Posted by: Dustin | April 22, 2007 1:29 AM
Dennis:
"[I]f there's no reason to suppose we're getting true predictions, then there's no reason (other than shits and giggles) to engage in the enterprise of science."
Two remarks:
1: I can't think of a reason to study particle astrophysics other than shits and giggles. Absolutely, positively, no practical use.
2. "True" is not the same as "as good as we can". We can never get truth, only our best approximation of the truth. Think of bayesian statistics -- the true underlying parameters are forever unknown, but with enough data, we can bet better and better estimates. We need not require "truth" to be useful.
Posted by: spudbeach | April 22, 2007 1:29 AM
as a "quantum solipsist" I can only ask "who's up for a
new game of African Dodger?"...
Posted by: uncle bob | April 22, 2007 1:29 AM
Rien:
Not quite correct. From Tripplehorn's websiteObviously the "holy spirit" did not prevent him from writing his embarrassing e-mail on June 3, 2003.
Thus, if he was born again at all, he was born again as an arrogant, stupid, self-righteous moron who doesn't accept responsibilities for his doing. And now he is blaming Jesus for his stupidity:Let's hope Jesus will break him down again. He already filed for non-profit status. Thus, there is a good chance for him to become Hovind II.
Posted by: sparc | April 22, 2007 1:32 AM
That's some shocking, uhh, non-reasoning there. But everyone knows that. And the camera work - eeeek! But, for crying out loud, the Stanford Encyclopedia of Philosophy?!
First of all, he's kind of quote-mining the entry on induction, which can be found here:
http://plato.stanford.edu/archives/win2006/entries/induction-problem/
(I don't know how to do linkies, sorry).
The author of the entry (John Vickers) doesn't conclude that induction is unjustifiable (and the section on the eeevil atheist Hume is only the second of seven sections). In fact, the last paragraph of the entry discusses what the author considers the most promising approaches: evolutionary and naturalised epistemology.
Secondly, the Stanford Encyclopedia isn't in the business of conducting or adjudicating competitions for solving longstanding philosophical problems. It's a project to create a comprehensive reference resource for advanced students and professionals in philosophy. Submissions for entries are generally only by invitation of the editors; unsolicited submission is only considered from professional philosophers who have published in the relevant field. Entries are supposed to be neutral, and provide a comprehensive overview of the topic.
(In any case, noone's going to get very far emailing solutions to the Problem of Induction to the webmaster).
I get the feeling this guy does not comprehend much of what he is exposed to.
Posted by: Jennie | April 22, 2007 1:35 AM
From the git's website, looks like you can make considerably more than $1K
Since it's impossible to win, I wonder why they don't offer $53 million.
And apparently truth itself didn't exist prior to 0AD. Weird.
Posted by: mtraven | April 22, 2007 1:37 AM
Exactly. A deduction is only as good as the premises that went into it. The deduction will only have a bearing on the real world if those premises were true, in some sense. If we want to pull our deductions out of the world of axiomatics and mathematics, we need some way of judging those premises to be true. But they can't be axioms anymore. So, they're either deductions or inferences. Supposing they're deductions, you just follow the trail of deductions all the way back to some primitives and OH NO -- you have some assumptions. How, for example, is the deduction that, if all fires are hot, then putting my hand in a fire will result in a burn helpful or applicable unless that "all fires are hot" statement is somehow physically meaningful? In real life, away from the context of axiomatics, I have to induce, first, that all fires are hot before I can deduce that sticking my hand in one will burn me.
Applied deduction suffers from exactly the same problem as applied induction.
Posted by: Dustin | April 22, 2007 1:40 AM
You see, the universe is like spaghetti. If you go to the store and buy a box of the stuff, you'll see that all the strands are essentially the same as one another. If you shook the box up, you wouldn't be able to keep track of any of them. Similarly, the different parts of the universe are essentially the same. Things that work in one part also work in another part, and this is true across both time and space. This truth is revealed to us when we are told that everything is touched by His Noodly Appendage. This means that everything is endowed with the properties of spaghetti, including the above consistency. Without His Noodly Goodness, induction would fail.
Posted by: Tiax | April 22, 2007 1:42 AM
Hoo boy, where to start with this guy.
Where does the bible even state that God created the laws which govern the universe, much less that he made them immutable?
Plus the fact there's an influential group that contends that the laws of physics don't always hold -- they're called CHRISTIANS.
I love the high production values on the video. He's just a little rich weasel spending Daddy's money on his own pet stupidities. I can't *imagine* how much this guy must've gotten beaten up in school.
Posted by: Mark | April 22, 2007 1:45 AM
Dustin @60:
Actually, it's more than a problem with "applied" deduction - the problem of deduction is not really about knowing whether your premises are true. It's about how you can prove that a given form of inference, for example Modus Ponens, is deductively valid.
How do I prove that, if "p" and "if p then q" are true, it's impossible for "q" to be false? Any argument which references the equivalence and inference rules of a system (e.g., pointing out the Classical Logic equivalence of "if p then q" and "not-p or q" then using Disjunctive Syllogism) is question-begging. (How do we know that DS is a valid inference?) Perhaps we could use induction - all previous instances of MP have been valid. But that's not satisfactory either.
There's some really interesting philosophical literature on this. Damned if I can find any students who think it's cool.
Posted by: Jennie | April 22, 2007 1:50 AM
Perhaps this incident illustrates one of the differences between peer-reviewed journals and youtube.
Peer review is very effective at stopping people who have similar revelations from publishing them and making complete fools of themselves in public.
Youtube on the other hand appears to partly revolve around people making complete fools of themselves.
Posted by: Chris Noble | April 22, 2007 1:56 AM
The problem of induction goes away when you consider the problem of non-induction. People tend to phrase non-induction as small variations on what's currently true- "the fork might rise instead of falling", but apparently forks, and we, otherwise continue to exist as usual. This isn't non-induction, it's partial induction; you're assuming a regularity in the universe ("things will change, but not by much") which leads you back to induction again.
If you really refuse induction, you have no reason to suppose that forks, or you, will exist in a moment's time. Nor indeed that time will continue to pass. If you deny induction you have nothing.
Posted by: Stephen Wells | April 22, 2007 2:03 AM
You mean we're still in the fishbowl.
Posted by: Norm | April 22, 2007 2:16 AM
I am a professionally trained logician and do some work as a logician, and I can tell from the first minute this guy doesn't know jack-shit about deductive logic. First off he defines dedictive logic as reasoning from the general to the specific. This is manifestly false. There are many examples of deductive inferences that move from the specific to the general: "Hansen is a spy. Therefore there is a spy" and "My car is red. Therefore it's red or blue" are obvious examples. There are also many examples in modal logic, which is also deductive. And the whole specific/general issue hardly arises in propositional logic, the most basic form of deductive logic.
Furthermore, this dude uses "assume" ambiguously. He uses it sometimes as a synonym for "infer" and other times to mean "take on as the basis for further reasoning". Logicians use "assume" only in the latter sense.
Posted by: bacopa | April 22, 2007 2:17 AM
Jennie, I still think it's a problem of application. In a purely formal context, one wouldn't even set out to prove the modus ponens of SL -- it's an axiom, isn't it? I'm not sure the kinds of proofs of something like the disjunctive syllogism by way of the rules of SL would be question-begging -- we'd just be appealing to the axioms we'd assumed. In a formal context, that's no different from simply assuming ZFC or the axioms of Euclidean Geometry (well, I don't think that Euclidean Geometry is a first-order logic, but they're the same aside from that).
The problem comes from attempting to apply deduction to the world. That's when we could call proof of DS by appeal to the inferential rules of SL question-begging.
Posted by: Dustin | April 22, 2007 2:37 AM
Sparc #57: Aha, there seems to be a problem with causality here... among other problems.
Posted by: Rien | April 22, 2007 2:40 AM
Me be thinkin' that many in here arguing about induction, do not know what the effin' hell is meant by induction...
Posted by: Lago | April 22, 2007 2:45 AM
Whoops! You suggested using DS to establish modus ponens. That would most certainly be question-begging, since I'm pretty sure that you need to modus ponens to establish the disjunctive syllogism (which is not, as I remember, an axiom, but a derived rule in SL).
Still, I stand by my suggestion: As an axiom, the modus ponens is part and parcel of SL. From pure formalism, that's not a problem at all -- the problem creeps in when we try to apply it.
Posted by: Dustin | April 22, 2007 2:48 AM
Dustin:
OK, yes, I see what you're getting at.
As you say, within any logic system the inference rules are given.
The problem arises when we try to decide what inference rules should be built in if we are trying to model The One True Logic (or the One True Logic for a given purpose), or if we are applying an inference rule in our reasoning.
Disjunctive Syllogism, Addition, etc. are disputed as valid rules of inference precisely because some logicians think they are not always truth-preserving - hence the development of nonclassical logics. The question is then how we show that any rule is in fact deductively valid.
Posted by: Jennie | April 22, 2007 2:54 AM
I feel like it would be far MORE likely for the fork to fly up if God did exist. After all, isn't the reason God can perform miracles is that he's unconstrained by the laws of nature? The laws of nature this tool says God made so that they NEVER change? Does that mean even God can't/won't change them? That's a really really stupid argument. Oh, wait... all his arguments were stupid.
Posted by: Kelly | April 22, 2007 2:57 AM
The problem with induction is solved if you don't enlist in the first place.
Ok. Pay up.
Posted by: Kseniya | April 22, 2007 3:00 AM
mtraven, nitpicking, but the jackass in question mentions the "Triune God", and that entity did not exist as dogma until centuries after 0 A.D.
Posted by: autumn | April 22, 2007 3:01 AM
Did anyone else notice that in his breakup letter to Michele, the smug git can't even spell "hypocrite" correctly? The guy needs to get over himself big time!
Posted by: Kimpatsu | April 22, 2007 3:28 AM
The proof of induction is found in this xkcd cartoon.
Posted by: Zarquon | April 22, 2007 3:29 AM