The most enduring challenge to science has come not from religion, but philosophy. David Hume, an 18th-century Scottish philosopher, articulated what we now call “the problem of induction,” and it has wreaked epistemological havoc on the foundation of science for centuries.
Induction, for the purposes of this post, is a form of reasoning that makes inferences about what will happen from what has happened. Science relies heavily on induction in making generalizations and predictions. But Hume believes that we can reason absolutely nothing about the future from the past. To do so presupposes the uniformity of nature—that the future will resemble the past.
There is a temptation to respond that we know that the future will resemble the past, because past futures have resembled past pasts. This begs the question, however. It assumes the very thing it attempts to prove, and is thus circular.
Atheists need to understand the implications of Hume’s argument. Hume is not saying that we cannot know with a certainty that, for example, the sun will rise tomorrow. He instead says something far more radical: that we have no reason whatsoever to believe that the sun will rise tomorrow. The fact that the sun has risen every day of recorded human history is immaterial; again, the future need not resemble the past.
So are we atheists who trust science guilty of the same faith that we accuse religious people of having? In a later post, I’ll introduce a few possible solutions to the problem of induction. But I’d first like to hear your thoughts.
I think ‘faith’ (or induction or assumption) comes in degrees and isn’t just a black and white thing. When we see somebody believing such a complex proposition like the God hypothesis and then say that atheists engage in the same degree of faith to not believe, we balk at the idea. I think to go from a level 6 to a level 7 atheist does require a small assumption (or faith as the believers would like to call it). But this leap is easy because of the evidence/lack of evidence. Isn’t it true that with scientific theories, you can’t necessarily prove them (ie 100%) but provide a sufficient amount of evidence so that they are considered fact? We may never get to 100% but the closer you get the easier the leap.
But Hume is not only denying that there could be proof for scientific evidence, but that there could even be evidence. Consider gravity. Say we release a ball 99 times, and each time it fell to the ground. Do we have evidence—not proof per se, but evidence—that the ball will fall to the ground the 100th try? Hume would say no. It is as unjustified to believe that the ball will fall to the ground on the 100th try as it was on the first try. We have no reason to believe that the future (the 100th try) will resemble the past (the previous 99 tries). That’s Hume’s argument, at least. So while I’m sympathetic to your view, I don’t think it’s responsive to Hume’s point unfortunately.
Does that make sense? It’s a hard problem to articulate because it’s so counter-intuitive to the way humans think (and indeed have to think). Hume even calls inductive reasoning an inescapable “habit” and “instinct.”
Thanks for the clear response. I understand, I understand.
I’m not sure if I entirely agree with Hume. It sounds like he would put all beliefs in the same basket. I think beliefs (induction) vary in probability based on evidence.
“I think beliefs (induction) vary in probability based on evidence.”
That may well be true, but that “evidence” can’t be past experience. And what evidence could their be in inductive considerations if not experience?
“It sounds like he would put all beliefs in the same basket.”
Hume wouldn’t necessarily, but the problem of induction does kind of demand that all beliefs about the future are equally probable. I am just as likely to wake up tomorrow as a kangaroo instead of a human being. That’s a crazy thing to think, I know, but it does seem to logically follow from the problem of induction.
Some philosophers argue that we can salvage at least some forms of induction from Hume’s problem, however. But I’ll discuss that in a later post.
Thats all based on the assumption that science is all about making predictions. What if we are only trying to interpret the present? We can interpret what is going on now based on what happened in the past. Its predictions that require faith. However, you would have to imply that the way things are happening now have been like that since the beginning of time.
http://en.wikipedia.org/wiki/Uniformitarianism
I think that both science and religion rely on a belief system, but that belief in scientific principles is different from faith. I think Hume has a compelling argument. We have a belief that the laws of nature will continue to hold true, and we can never know that some timed switch might go off and change the speed of light, or pi by 3 (I think of this as a problem with logic, and not a problem with experience as Hume argues. My problem is not that I can’t directly experience the laws of nature in of themselves, but rather a problem of extrapolation).
The belief inherent in science is simply that the universe works according to laws, it can be understood by reason, and the laws of nature are constant. This seems to me, while not necessarily provable in the absolute sense, to be something quite different from faith which is belief in something that is not understood or proven by reason. Either way, for all it is a leap of belief, and depends on what you choose (sometimes arbitrarily) to put your trust in. I don’t believe the two are at odds necessarily, and there is certainly more rational evidence to prefer science.
Although I think the problem with induction and science becomes even more evident at the quantum level. Strange stuff starts to happen which can not be directly experienced, or fully explained. Well I mean physicists can kind of give you a general idea where the electron is, but it is nothing like the “exact” sciences of the macro level.
Similarly, since string theory is so abstract and mathematical, and on such a small scale, that many physicists reject it because theories will never be able to be proven, and we can only prefer one string theory over another in its predictability. If several do a good job, then there really isn’t a way to empirically claim that one is superior or more “true.” My wonder is if scientists put a bit more “faith” in science, when the basics, it’s fundamental building blocks, are still not fully understood.
Oh also, scientists believe that right after the big bang the laws of nature were very different, maybe that is applicable.
One thought, and I know I’m going to annoy some people with this. Scientists ignore this issue postulated by Hume. I agree that they should. We need basic assumptions about how the world works based on previous experiences in order to move forward with anything. Science works, there is no doubt about that. Just because it is a problem mentally (or philosophically) doesn’t mean it is a problem in the real world. Also, science has a coping mechanism built right into it. If there ever comes a time where the laws of nature don’t act how they are expected then the current theory is wrong and needs to be modified to include the anomaly. It should also be noted that scientists don’t call their ideas and laws facts, but theories. They are prepared to be wrong, in a way.
Kleiner said, “Point is, as a theist and a Thomist I think I have more reasons to be confident in scientific inquiry than atheists or materialists. This is the great irony of atheists who are all puffed up over their science – they cannot ground it but the theist can. Funny world, huh?”
Maybe you do have more reason to be confident in scientific inquiry than an atheist but the fact is that you are not (as evident in the past on this blog). Funny how that is, isn’t it? But I argue that since you believe in a God that interferes in the world you don’t have more reason to be confident in science, but less. Your God can (and has) intervened in the world of man many times. He destroys the laws of nature as he sees fit and makes things happen that shouldn’t. You have no idea what his intentions are, or how he is manipulating things. He could be messing with every scientific experiment ever done to give us an incorrect version of reality for whatever reason he has. He may change his alternated view whenever he wants. I don’t see how you can possibly have more confidence in the scientific process with a God meddling with nature constantly (or how he sees fit).
On a lighter note, I thought this was relevant: http://www.smbc-comics.com/index.php?db=comics&id=1879
For an added comic, hover over the little red button at the bottom of the strip.
[My other comment disappeared. Sorry if this is a repeat]
What have I ever said on this blog (or anywhere else) that was anti-science or that suggested I was not confident in scientific inquiry? I have certainly criticized materialism and reductionism. I have criticized scientism. But I’ve never suggested that science was anything other than a legitimate mode of inquiry capable of genuine knowledge. I just want to remind scientists of the boundaries of their method and inquiry. If pointing out that science has limits makes you “anti-science”, well then I guess I am guilty of that rather ridiculous charge.
I think Ben gets miracles all wrong. Ben thinks if you believe in miracles then you cannot believe in natural laws. But the opposite is true. In fact, believing in miracles requires that one believes in two things: (a) a transcendent God and (b) a natural world with inherent laws. In order to take miracles seriously, you have to take natural laws (and man’s capacity to understand them) seriously.
I also think Ben has a pre-school view of miracles. No offense intended there, Hitchens and Dawkins and all the rest of the popular atheists do too. I won’t hijack the blog with a complete explanation. But I see miracles as perfections of natural laws, not violations of them. They are additions, not subtractions. Miracles don’t contradict laws of nature, just as a monetary gift does not contradict a bank balance. A pardon from a governor does not contradict the judicial laws of the land. Miracles are like that. Once again, most atheists read the Bible in the way fundamentalists do, and it leads to straw men arguments (at least straw men arguments against thoughtful Christianity or Catholicism).
One more thing on this “preschool view” of God. I had a student last term who repeatedly referred to God as the “magical bearded man in the sky”. On this view, God is just an elevated man, a sort of a fickle cosmic tyrant, who can (to use Ben’s words) “destroy the laws of nature as he sees fit” and who could “change his alternated view whenever he wants”.
Who holds that view of God? I suppose some do. But no intelligent theist or Christian that I know does. It is just such a straw man to argue against such a view. Sure, you might be picking some really low lying fruit, but you are not coming close to touching serious theism with those arguments. I just wanted to clarify that these arguments don’t topple theism as easily as so many theists (including purportedly profound ones like Dawkins and Hitchens) think. Again, if your only encounter with theism is main street Mormonism or young earth creationism Christianity, you are not really engaging theism. Frankly, the naivete of the tradition of philosophical theology on the part of anti-theists is more than annoying, it is irresponsible.
I go back to the Pope’s Regesnburg Address. He is challenging both Islam but also the West on the relationship between reason and faith. One question concerns whether God should be understood in a voluntaristic way (Islam has a tendency towards this), where we see God as pure will that can be manifested in any number of fickle ways and even in ways that contradict reason. The Greeks rejected this idea, and B16 wonders if this is “just a Greek idea”. He (and I) clearly think it is not.
Point is, once you outgrow this preschool idea of God being this magical bearded man and begin to think of God as Being Itself, that is, once you mature your religion by doing some philosophy, you outgrow those ideas just as you outgrow other childish things.
You need a certain kind of metaphysical theology to preserve science. You need a God who is transcendent, absolutely immutable, a God for whom acting unreasonably would contradict his nature. Benedict (and I) see a profound harmony between the best of these “Greek” ideas and the biblical understanding of faith in God. As the Regensburg address develops, one has to penetrate into LOGOS.
Again, it is a funny world where it is the Pope who has to come to the defense of an elevated notion of reason in a world gone made with reductionism and relativism.
Anyway, now back to your regularly scheduled programming. I am eager to hear Jon’s (and others) atheist solutions to this problem of induction. Jon’s remark that we “must” think this way, that it is almost “instinct”, makes me wonder if he is going to make a Kantian transcendental move.
I don’t see how you answered the question. If you believe in a God who conducts miracles then I still think you are on the same level as the rest of us for this problem.
I don’t care how you view miracles, they still change the laws of nature as we know them. Whether the miracle is changing natural laws to a supreme version of natural laws is besides the point. The point is that God is changing the laws of nature as we have observed them constantly. Now, do we know of every single miracle that God does? Are they all in the bible and that is the end of the miraculous account of God? Or is it that God can pick and choose when and where he does miracles for his own purposes?
If you feel that God is still doing miracles then you still have the problem of not knowing when and where God is changing the laws of nature as man knows them. Sure, I get your point that with a God he can ensure that the laws of nature are generally stable. But at the same time, you do not know when he is intervening in the world of man to change things.
As far as science goes, I seem to remember you being critical of evolution. You weren’t sure if you believed that it was real. That is one instance that really stuck out in my mind, but I could be wrong. I don’t feel like looking up your exact quote. It wasn’t about you reminding us of the boundaries. I could be wrong though.
You also invited the accusation that you’re anti-science by agreeing with the pope that condoms spread AIDS in Africa. I don’t mean to discuss the merits of that claim; I bring it up only to show how you could be perceived as anti-science.
I am not “against evolution”. In fact, to quote Pope John Paul II, evolution is clearly “more than just a theory”. I am agnostic about the specific evolutionary story that is to be told. I’m agnostic about the details because I am not an expert on such things and because I don’t see it as a threat to my theism or faith anyway.
Regarding condoms in Africa, readers might recall that I put forth considerable empirical evidence that condom programs in Africa have not reduced aids infections rates.
Anyway, I am not anti-science. I’m just not drunk on it and don’t see it having a monopoly on disclosing truth.
Please allow me a tangent to get something off my chest. I must say that I am quite bothered by this reputation of my being anti-science.
Let’s take the condoms in Africa case. This is a case where I had the empirical evidence on my side. The head of the Harvard AIDs project came out supporting what the Pope said (even though he is a self-professed liberal and no friend of Catholicism). Following the Harvard AIDs project head, I put forth a bunch of evidence on how different African nations have responded to different programs, and the evidence was overwhelmingly on the Pope’s side.
This is a classic case, though, of hypocrisy. I butted up against YOUR dogmas, and the “free thinkers” respond with charges that I am “anti-science”. What a joke. No, in this case, I was following the science but many of you were ignoring the science and following your dogma. But the “anti-science” smear is simply a convenient slur to hurl against religious people when you don’t agree with them. I cannot insist on this strongly enough – it is those who disagreed with me on that claim who were being “dogmatic” and anti-science!
I know I am over-reacting, and I am reacting to far more than Ben’s or Jon’s comments here. Please don’t anyone take what I have said here personally. But you will note considerable frustration in my voice. Anyway, let’s set the record straight: I think evolution is more than just a theory, though I do not commit myself (nor should any of you, since there is considerable debate in the scientific community) to one or another particular evolutionary story at this point. The person I talk to about these things most is Huenemann (who knows a lot more about the competing theories than I do), and I am no more or less agnostic on these matters than he is (and he is on your team, so to speak). I am not anti-science. I think God created man and world in such a way that our minds can understand the truth of things (God made us intelligent and made the world intelligible). I don’t think science is the only mode of truth disclosure. I think there is moral truth, philosophical truth, truth in poetry, art, music, and literature, etc etc etc. These are all different modes of thinking, and I resist any attempt to reduce understanding to any one mode of thought. Like all modes of disclosure, science both discloses but also conceals. This is why a tyranny of science is a bad thing, it would lead to a self-forgetfullness that is inhumane. Again, Walker Percy nails this point in Lost in the Cosmos (if you want the point made with considerably more depth, read Heidegger).
Maybe this is just a problem with blogs. No one knows anyone (well, that is not true – I know some of you and some of you know me). But it is hard to get a sense of where someone is when “blogologuing” because you just get these little snippets.
There, I feel better now. Thanks for allowing me to vent a little there.
Again, back to your regularly scheduled programming. In particular, I am eager to hear how atheists are going to sort this out on their own terms.
I am just not moved at all by this argument that since miracles could be happening all the time we cannot trust in the laws of nature. Miracles are irregular events. If miracles were happening all the time, there would be no regularity in nature. But there is regularity in nature (nature is uniform). I don’t say that every miracle that has ever happened has been catalogued, but they are exceptional. It is their exceptionality that allows us to recognize them – they are instances when the uniformity and regularity of nature seems to be interrupted. But, again, to recognize the exceptions to the rules you have to first take the rules really seriously.
Actually, the Vatican has a whole procedure on this for when they look at candidates for sainthood. Actually, Dr. Sherlock and I have been trying to get our hands on some records from recent proceedings. The Church brings in experts in the relevant fields (often medicine), and it would be very interesting to read the reports. We’d like to look at one successful case and one failed case (a case where an event had no plausible account from the point of view of natural causes could be given, and a case where a plausible natural account could be given).
A few quick thoughts:
1) Maybe Paul’s point could wiggle you out of the problem, but at an extreme cost. Saying that science only tells us about the past seriously reduces science. It is not just that science is interested in predictions. Of course it is, and there is enormous practical benefits in those predictions (nutritional science, weather, weight bearing loads of different building materials, pretty much all of medicine, etc etc etc). But what science, in the broadest strokes, aims to do is to formulate testable theories which seek to explain the principles or laws which govern the universe. But once you start talking about principles or laws, you are deep into the problem of induction (because you are talking about basic principles which are supposed to hold in all relevantly similar conditions in the past, present, or future). So you would dump out of, say, physics, all talk about the laws of conservation, all of Einstein’s laws, etc etc. I doubt many are going to be eager to do that.
2) While I am not personally vexed at all by this issue, I do think this is a serious issue for atheists. Huenemann and I talked some about it last year, and we could not think of too many ways out of this jam. Since he was not interested in a theistic turn, he was seriously flirting with Kantianism. The Kantian transcendental move is one solution to the problem (Kant says Hume woke him from his “dogmatic slumber” and motivated his transcedental philosophy). But while Kant might solve Hume’s problem of induction, you invite a whole host of other problems with Kant. And I am guessing that most readers of this blog are going to be pretty uncomfortable with philosophical (mainly German) idealism, of a Kantian sort or any other brand.
3) I often remark on this blog that I, as a theist, have more confidence (and more reason to have confidence) in natural reason than do atheists. The theist (at least theists of a certain kind, and I have Thomism in mind) just don’t have this problem because God (Divine Logos) is something like a metaphysical guarantor of the uniformity of nature (I won’t delve into the whole metaphysics of that here). Of course, Descartes seems to think that even deductive reasoning is at risk without God (Meditation I – the less perfect you take the author or your reason, the less you should trust in that reason). So the problem may not even be restricted to induction.
Point is, as a theist and a Thomist I think I have more reasons to be confident in scientific inquiry than atheists or materialists. This is the great irony of atheists who are all puffed up over their science – they cannot ground it but the theist can. Funny world, huh?
I’m sorry, but I think you’re in the same boat with the rest of us, Dr. Kleiner. That is unless you can demonstrate a priori (without an appeal to experience) that there exists this metaphysical guarantor of the uniformity of nature. And even then, really, I fail to see how this guarantor neutralizes Hume’s argument. Just because this god guaranteed the uniformity of nature in the past, does not mean that it will guarantee the uniformity of nature in the future!
I don’t want to “threadjack” the post (I think that is the term) with theist solutions to the problem. You are all better off having an in-house discussion about how atheists might handle it. That might be fruitful for all of you.
But James’s post helps me to make my point (hopefully concisely). James says,
“The belief inherent in science is simply that the universe works according to laws, it can be understood by reason, and the laws of nature are constant. ”
I think that is right. And this is the point I always make – it would be nice to have an account of such things, wouldn’t it? Some people, like Huenemann, are willing to live without an account. Good for him. But going that route really obligates you to a pretty radical skepticism (maybe even more radical than Hume’s!). To Huenemann’s credit, he bites that bullet (usually). My point is that a theist like Thomas can give an account of this with his theology. I don’t think the move has to be a priori either (neither does Thomas). One can reason his way to the grounds of the reason he is using, I don’t think you need to step outside of it. But you do need a certain metaphysics.
I don’t see your last point. The uniformity of nature is precisely that it is uniform over time (past, present and future). If you can secure the uniformity of nature, well then you’ve secured the uniformity of nature. I suggested that you need a certain kind of God to do this (the LDS God will not do, because the LDS God is temporal and spatial and of this world). You need a God that is eternal and transcendent and is the cause of ex nihilo creation. But since the guarantor is outside of time and is absolutely immutable, it makes no sense to suggest that the situation would change.
Of course the question for Aquinas (and me) is going to be: can you show that the “God of the philosophers” (which can function as this metaphysical guarantor) is one and the same as the God of Abraham and Isaac? There will be some clear tensions there that will need to be resolved (How do Logos and Love meet? or How compatible is Aristotelian metaphysics, really, with Christianity? In other words, can Athens and Jerusalem, the Greek and the Biblical, really be married?). I think these tensions can be resolved. As Pope Benedict remarked in the famous Regensburg address, “The encounter between the Biblical message and Greek thought did not happen by chance.”
Point taken about my last argument. What I suppose I was driving at was this, however: I just doubt you’re in the position to know whether there exists this transcendent guarantor. How could you know that this being was indeed upholding the uniformity of nature unless you could observe, simultaneously, all of space at all point in time—past, present, and future. It seems the only being in a position to whether this guarantor actually guarantees anything is the guarantor itself. Is there an a priori reason to believe in your guarantor of the uniformity of nature? I know it’s argued that a god outside of space and time is a necessary being, and not a contingent one. But what property entailed in that skeletal understanding of god requires that it be the guarantor of the uniformity of nature? And no, I don’t think the property of god’s immutability fits the bill. God may be immutable, but I don’t see why his creation (the universe and its laws) need be also.
Moreover, I’m still left wondering how an immutable agent outside of space and time can explain the universe and its laws (this is something I meant to bring up in the discussion about my post on the hiddenness of god). I don’t see how such a being could possible be personal (perhaps that one of those tensions you alluded to). Also, if you’re outside of time, how can you cause anything? To cause is to happen before, but words like ‘before’ only make sense within time.
These issues are a digression from the original topic though, so maybe we could move this discussion elsewhere. Nick (this site’s operator) just started a SHAFT forum for lengthy discussions like this one. Here is the thread I started for our discussion: http://usu-shaft.com/forum/blog-discussion/does-theism-provide-an-out-to-the-problem-of-induction
I’m not sure why this is directed at atheists. Atheists would be the last group this would be appropriate for. Atheism is a lack of belief in a god or gods. Generally this is due to the atheist’s perception of a lack of evidence in a god and not evidence of a lack of a god. Saying nothing of the future can be known does not address the atheist’s lack of belief. However, I think the slant of the argument presented here is towards people who try to disprove religious claims by making statements along the lines of: “Religion x says the earth is x years old but science tells us that the earth is y years old. Therefore religion X is wrong.” Hume’s argument would be that we cannot trust the scientific method because retesting the age of the earth tomorrow might yield an age of x years after all (or z or a raccoon. Anything is possible!). I think science has long since addressed this uncertainty using the concept of confidence. In science we never “know” anything. We can be very confident but we must always acknowledge that everything could be wrong when new evidence comes to light or the universe changes overnight. When science presents, say, the law of gravity as fact we are saying we’re very sure that this thing is true and will continue to be true.
This addresses Hume’s argument by allowing for the possibility that gravity will not work tomorrow while simultaneously contending that it is very likely that it will. Jon’s assertion that “we have no reason whatsoever to believe” addresses the question of what evidence (the scientific “reason”) is. Evidence, in its broadest sense, includes everything that is used to determine or demonstrate the truth of an assertion. Within that broad definition, there are different qualities of evidence. “Some guy on the internet told me gravity exists” would be very poor evidence. Multiple studies using different methodologies repeated over the ages coming to identical conclusions would be much better evidence. There does not have to be perfect evidence. All Hume’s argument does is say that inductive evidence is not perfect evidence.
I split this hair because the assertion “we have no reason whatsoever to believe” implies that the smallest amount of doubt is equivalent to complete uncertainty. Just because the sun might not rise tomorrow does not mean that we have no reason to believe it will. We have past experience. Past experience may not be perfect evidence but it IS evidence and it is by this evidence that we can provide a rationale for our belief.
This is directed at atheists for two reasons: First, this blog’s audience is mostly made up of atheists; and second, I think some atheists have an undue confidence in science. I think science is the best means to truth we have, but we need to understand its limits.
But by directing this post at atheists, I am NOT saying that the problem of induction is only an issue for atheists.
“In science we never “know” anything. We can be very confident but we must always acknowledge that everything could be wrong when new evidence comes to light or the universe changes overnight. When science presents, say, the law of gravity as fact we are saying we’re very sure that this thing is true and will continue to be true.”
But WHY can we be very confident? You haven’t begun to say WHY past experience can or should be used to generate or test theories, which is the whole point of the problem.
Well Mike, I’d say we are very confident because we have evidence that we consider to be of high quality. We consider the evidence to be high quality because it falls into that class of evidence that has been reliable in the past and is currently reliable. If that class of evidence fails at some point, we will revisit our assumption. We may question that assumption from time to time and verify that it still holds but, so long as it does, failing to use the knowledge we have gained because “anything is possible” would leave us in a nihilistic immobility. To quote popper (and thanks to redditor ronaldvr for the idea http://www.reddit.com/r/skeptic/comments/ci90e/the_problem_of_induction_is_science_faithbased/c0sr7ko)): “there are no such things as good positive reasons; nor do we need such things [...] But [philosophers] obviously cannot quite bring [themselves] to believe that this is my opinion, let alone that it is right” (The Philosophy of Karl Popper, p. 1043)
I’m glad we agree and thanks for addressing the issue this time.
One word: pragmatism.
Bingo
I think induction needs to be appropriate to the time scale. For example, there is no reason to believe the earth will continue to rotate on its axis as it does now for ever (ie “sun rising”). But whatever is required to stop the earth from turning will take longer than 24 hours to take effect.
I might also say all swans are white because all I’ve ever seen are white swans, but that’s only because I’m only thinking of the swans I’m likely to encounter in the next 1/2 hour or so. But given enough time, I can take a long trip to the other side of the world, and I might find some differently coloured swans there.
Again, I can drop a ball, and then drop another ball two minutes later and experience the same result. If I drop it 4 billion years hence, I might find that the earth has shrunk enough that the ball doesn’t necessarily drop, instead it floats away.
So as long as the time scale is appropriate, and all other variables remain constant, we can have faith in our inductive reasoning.
“But whatever is required to stop the earth from turning will take longer than 24 hours to take effect.”
Here’s where I think this particular argument fails: You know that this earth-turning force requires more than 24 hours because your experience (from which we learn nothing about the future) tells you this.
“So as long as the time scale is appropriate, and all other variables remain constant, we can have faith in our inductive reasoning.”
But why assume all other variables will remain constant? Isn’t that assumption the very thing that makes induction problematic—the assumption of the uniformity of nature? Maybe I’ve misunderstood your argument.
I don’t believe that purely Humean nor 18th century empiricism in general are the norm in scientific thought any longer. Karl Popper’s theory of refutation, for example, is widely accepted makes a good case which circumvents the problem of induction and makes for less of a need for faith in science. To put it far more simply than does it credit, if the goal of science is to most completely describe the universe, then we may begin with purely deductive claims; hypotheses. Their validity is then tested through theory and experiment. Positive results, obviously, cannot *prove* these claims, but a negative result shows that the hypothesis in question is wrong and must be scrapped or revised. Over time our scientific knowledge approaches more and more nearly the truth, although it may never actually achieve it.
With regard to the post, I guess I’d have to come down against Hume. You have to consider that Hume lived in a pre-statistical, pre-probabilistic world, and if modern math and science has taught us anything, it’s that statistical analysis often comes closer to describing and predicting the future than any kind of deductive analysis can perform. However, it can only portray the future in terms of probability, and this is a huge step for a philosopher to take; it comes at the price of placing oneself outside a personal vantage. I can kind of see what Hume was on to: If you consider the world to be self-renewing from moment to moment, you can never truly predict anything. But this is simply not the world we inhabit (at the moment). To go from lack of absolute certainty to utter agnosticism is an error. In actual fact, unless you care to dispute our ability to record history, or take seriously a proposal that we’re subject to some kind of cosmic deception, we do know the historical record, and from it, it is perfectly rational to assign probabilities to future events. I cannot predict with absolute certainty that the sun will rise tomorrow, but I can place a very high probability on it, and that high probability (very near to 1) is what makes induction possible.
Science doesn’t give us absolutes, it gives us conclusions that are based on probabilities. So science can’t say “the sun will rise tomorrow” but it can say “there is a 99.9^99% chance the sun will rise tomorrow.”
For all intents and purposes, when probabilities of something are so low or so high they are communicated as absolutes but the “caveat emptor” is that everything is only based on probability. It’s the fine print on all science.
Hume discounts even probabilistic inductive inferences. I mean, how did you figure that there is a 99.999% chance that the sun will rise tomorrow? You figured that from past experiences, which you cannot rely on to tell you anything about the future. So again, Hume is not saying that we can’t know that the sun will rise. He is instead saying that we have no reason to believe that the sun will rise. It is just as probable that it won’t rise as it is probable that it will.
I love Hume. His were the first arguments I encountered that really knocked me on my ass.
We’re so certain of the uniformity of nature I’m not sure we can conceive of the world without it. We can imagine what it would be like if a physical law changed, like gravity losing some of its pull might put a spring in our step. But can we conceive of nature not remaining mostly constant? And how “secure” is the opposite conviction, that nature will NOT remain constant? Is there reason for that belief? There is reason for doubt but are there experiential grounds for doubt?
What we lack here is reason for belief OR disbelief (and Jon is doing a great job re-articulating the problem when necessary in this thread) but I don’t think it can be secured with something else that will also need reason to believe in since our belief in the uniformity of nature is such a core belief. The best bet for any belief you need to secure is to place your faith directly in it (or try to rid yourself of it and see how that works out, see if it’s possible, some beliefs are inescapable).
I believe in the uniformity of nature; I’m a believer. I’m ok with that.
The problem I have with this whole concept is that it’s no more than a thought experiment. There’s no way to prove what Hume argues because he’s arguing that main and most reliable way of testing what is and is not true/real is by it’s very nature useless b/c it relies on past experience. A hypothesis which cannot be tested to see if it’s actually true is pointless (similar to the god hypothesis). Interesting perhaps, but it seems silly to worry about it.
The logical conclusion of this is that we cannot know anything. We cannot know that stabbing someone in the heart is bad, even though every other time it has happened, it has been bad. I guess I just don’t understand the point of this. In what way at all is it useful to debate whether a non-falsifiable assertion is true? There’s no way to know!
Science doesn’t work by induction, but by abduction.
Well, that was easy.
I am wading into the philosophy of science of which I am novice (though I am professional scientist). I will review my understanding of the cause-effect problem posed by Hume and Kant’s solution, which was illuminated to me by an essay by Heidegger.
Hume basically stated that there is no connecting reason or observation that necessarily connects the past through the present to the future. Rather, we are empirically trained by the past but the necessary connection of a past (cause) to a future (effect) is hidden inside an impenetrable box for the observing/reasoning human.
Hume killed science based on models based on internal nature of a thing, which is Aristotle’s view of science. A rock contains a nature seeks a lower level and fire has a nature that seeks a higher level. Upward movement of a rock is a violent action against its nature. However, there was nothing to observe or reason towards a proven truth of the cause-effect event of dropping a rock. (I need kleiner to clean this up, I ponder whether Kleiner might be a Kantian Thomist with respect to science.)
Kant was thrilled with Newton’s Laws of motion and law of gravity. He was awakened by Hume from his slumber to think about how science could be saved. He eventually poses the idealist approach to solve the connection between the past and future — that is ‘laws of nature’ rather than ‘nature of objects’. We observe cause-effect events but we don’t observe the ‘thing in itself’ that connects the cause to the effect (just as Hume stated). Instead, we can deduce the cause-effect relationship from a proposed theory that describes an eternal law that is unchanging through the past-present-future, that is, time is removed and an eternal law is identified. The law is not directly observed but is posed as ‘true’ (idealist transcendentalism).
Kant represents the completion of the philosophical shift in science from nature to laws.
Thoughtful scientists like Einstein recognize that the fundamental laws of physics are really inexplicable. I paraphrase Einstein — “The eternal mystery of the world is its comprehensibility.”
Hume’s denial of knowing what is ‘inside-the-box’ between the cause and the effect and Kant’s move to transcendental laws and Einstein’s statement can be looked at as a recognition of Aquinas’s starting point of his governance proof of a transcendental god.
To Kleiner. Aquinas poses a solution (god behind the governance of the universe.) I think Spinoza also poses a suitable solution.
“You need a certain kind of metaphysical theology to preserve science. ” -Kleiner
*And scientists everywhere scurry to the theology section of their libraries.*
No one knows what tomorrow will bring (even theists) but I think what preserves science is still getting the right answer. Science can give us the best answers in regard to survival so I would think it preserves theology and things like the problem of induction. At least it helps provide people the leisure to contemplate such things.
The latest change in the philosophy of science (after ‘Nature’ then ‘Laws’) is observed by Thomas Kuhn. He posed that for the most part scientific advances were very slowly until some visionary scientist proposes a new paradigm — a new model, then a great leap forward is made. Newton’s law of gravity and laws of motion were one such paradigm shift. Planck’s quantized energy and Einstein’s relativity are others.
The post-modern view of science should be seen as ‘model-based’. The model doesn’t necessarily capture reality, rather reality is just beyond our models. For example, a electron is a small particle — that is a model, a picture of an object that we have never seen, but it explains a great many things. American Physicist J.J. Thomson ‘discovered’ the electron in 1897. Prior to that atoms were indivisible units. But in 1924, Louis-Victor-Pierre-Raymond, 7th duc de Broglie proposed that the electron acted like a standing wave in its orbit around a hydrogen nucleus. Wave? Particle? These are just models that seem to help us describe the phenomena of matter.
Oh. By the way, the laws of conservation of energy and momentum can be temporarily broken in quantum events.
Things have a nature. ‘Aristotelian’
Things follow timeless laws. ‘Idealism’
Things are beyond us, we must pose models of things that obey laws, but we must remember they are models. ‘Post modern’
To Mike. “And scientists everywhere scurry to the theology section of their libraries.”
It would help scientists to have a bit of humility about the successes and the limits of science. With that said I no longer give any leeway to fundamentalist theists to advance pseudoscientific theories based on faith and revelation.
Vince,
The big stain on science’s soul are things like the oil spill. Humanity’s hubris in being willing to utilize things we don’t really understand and have no control over. A simpler example is just cars and how we kill ourselves constantly using the things but people keep thinking it’s within our capacity. I think the number is 1 in 83 people will die in a car accident in their lifetimes. And we think we can handle nuclear energy? hubris.
I’m not against multiple ways of knowing things. I’m especially hip to what can be revealed by learning a new language/culture and what we can learn about ourselves and our own thought movements by paying attention to the world phenomenologically. I just have a problem with thinking theology preserves or provides ground for things. I think it’s a misunderstanding of what things need in order to exist and a dishonest way to think about rationality and foundations.
A statement like “understanding theology helps provide balance to an unfettered scientific perspective” I can buy into. Another defense of theology would be that it’s an old, nuanced science that’s covered a lot of ground and even though that ground is covered from within a theological framework, it’s still often relevant to a non theological one. The problems of existence being common to all.
Blaming the oil spill on science is nonsensical. Those who misuse knowledge are the ones responsible, not the ones who discover it. Knowledge is amoral. It just is.
I do agree though that humanity has overwhelming hubris and doesn’t know how to use the knowledge it has.
And religion is NOT a form of science. The problems of existence may be common to all, but the “solutions” of religion are not. Religion is evidence-free conjecture that more often than not is harmful and disgusting.