« Zombie Attack! | Main | Stereotype Trouble »

Nature and Super-Nature

I recall well that in the spring of 2003 certain doubting Thomases in the vicinity of Emerson Hall raised questions as to whether writing an honors thesis on the philosophical issues in the neighborhood of controversies about the teaching of evolution in public schools was a useful way for an aspiring political journalist to be doing. Flash forward to May 2005 and what do we see but political journalists arguing abotu philosophical issues posed by controversies about the teaching of evolution in public schools! Chris Mooney's mad that anti-evolutionists want to replace the statement that science seeks "natural explanations for what we observe in the world around us" with the claim that it seeks "more adequate explanations of natural phenomena" on the grounds that we shouldn't open the door to the supernatural. Julian Sanchez strikes back arguing that "I very much doubt there's any sort of meaningful distinction to be made between 'natural' and 'supernatural' accounts in this context . . . If ghosts or gods did exist, after all, wouldn't they ultimately be as much a part of the natural world as human beings or dolphins or leptons?" Mooney attempts a reply and gets near the right answer by mentioning "natural laws." All this and more was covered in a deleted scene from chapter five of the thesis. Sadly, that draft died during a computer crash some time ago, but one can find an arcane discussion of the topic below the fold.

The natural law thing is key. The best gloss you can put on the natural/supernatural discussion is that science supposes the world to be governed by certain unchanging natural laws such that the world unfolds through a series of causal interactions that are, in principle, predictable. The content of science consists, in part, of exploring what these laws might be, an effort to discern what sorts of laws might explain the available data and, therefore, allow us to make conjectures about what happened in spatio-temporal regions where we have no data, including, of course, the future.

Most religion posits crucial departures from this sort of scheme. When Jesus transformed water into wine, that was miraculous -- supernatural -- a break with the standard rules of the universe. What's not being posited in that story is that Jesus was some kind of alchemist who devised a method of water-to-wine transmogrification that he could teach to somebody else. Instead, due to his divine nature, he was able to interrupt the normal causal processes. This sort of thing is the supernatural. If it happens, it's distinguished from the natural in that it's not law-governed. Note that supernatural occurrences of this sort produce effects in nature. The created wine becomes wine and intoxicates its drinkers in all the usual law-governed ways. But the transformation is, itself, super-natural. Insofar as posited miraculous phenomena are relatively trivial and micro-scale like this, a supernatural worldview can be pretty compatible with science.

The vast array of small miracles that are attributed to the saints by the Catholic Church are like this. They're supernatural occurrences, but they're rare and occur on a small scale. They're not explainable by science, but their occassional occurence doesn't undermine science. They're exceptions to the rules, not refutations of them. Indeed, it's specifically because miracles are understood as supernatural that belief in their occurence becomes compatible with science -- if you believed these things happened but understood them as natural phenomena, then they would become anomalous data that would need to explain scientifically by altering our understanding of how the laws of nature work.

Evolution, however, is a big deal. If creationism is true, that's not a small miracle that basically leaves scientific understanding intact. Instead, it's a huge miracle that would displace an entrenched and seemingly successful scientific theory.

What's more -- and this is crucial -- if you replace the demand that science investigate laws of nature with the idea that it merely seeks "adequate explanations of natural phenomena" you open the creationist doors wide open. The problem is that with naturalism left out of the picture, Young Earth Creationism becomes a very adequate explanation of all the relevant phenomena. The basic idea here is that God created the world 10,000 years ago and that's that. How come radiocarbon dating seems to indicate that lots of these bones in the ground are way older than that? Well, because they were created with such-and-such a carbon content back 10,000 years ago. Why are there bones of species that never existed as living things during the past 10,000 years? Well, the bones were all created that way 10,000 years ago. One could go on. The point is that "God just made it all that way" adequately explains anything you care to have explained about the fossile record in the sense of providing an answer as to how come our data looks the way it does. And, in fact, it's more adequate than the evolutionary explanation because the God Hypothesis doesn't involve any recalcitrant data at all. It explains everything.

Explains everything, that is, insofar as it explains anything. Scientifically speaking, that's not an explanation at all, because it's intolerably ad hoc. It doesn't tell us anything about the laws of nature or the causal structure of the universe. It's not, in other words, a natural explanation.

So Chris is right that the distinction is important. Awkwardly, though, the same argument holds even if creationism is true. Scientifically speaking, creationism can't be true, less because there's no evidence for it than because there couldn't be evidence for it. Science presupposes, rather than demonstrates, that the supernatural does not play an important role in the unfolding of the universe.

May 17, 2005 | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference Nature and Super-Nature:

» Evolution? from CasdraBlog
I've been reading a lot about the evolution "trial" held in Kansas a few weeks ago. Here are a few highlights. Let's start with Chris Mooney at The American Prospect with Creating a Controversy which details the most disturbing issue coming o... [Read More]

Tracked on May 17, 2005 7:49:46 PM

» Water into wine from Before Completion
Evolution has been on my mind a lot lately. Partly because my work has involved phylogeny and multi-species comparisons lately, but also because of the Kansas evolution trials. Matthew Yglesias' bl... [Read More]

Tracked on May 17, 2005 9:06:15 PM

» The supernatural from Thoughts from Kansas
I'm prepared to be less of a naturalist than Chris and Matt Yglesias, but less of a supernatural inclusivist than Julian Sanchez. I think that there are ways that the supernatural can play into science, I just think there are limits. Think about i... [Read More]

Tracked on May 17, 2005 10:02:41 PM

» Natural law and intelligent design from Uncommon Ground
Matthew Yglesias hits the nail on the head. Science presupposes, rather than demonstrates, that the supernatural does not play an important role in the unfolding of the universe. One of the biggest mistakes scientists make is a result of philosophical... [Read More]

Tracked on May 18, 2005 7:31:45 AM


Of course there could be evidence for creationism, the creator could sign his work in all sorts of ways.

Posted by: James B. Shearer | May 17, 2005 3:41:41 PM

I think you're conceding too much ground to the anti-scientists. Science doesn't "suppose" a world governed by natural laws because that's an arbitrary fundamental assumption of some field called Science. Science generally looks for explanations from natural laws because over the past half millenia or more there has been overwhelming evidence that the universe follows natural laws and no evidence that it ever deviates from them. It is this evidence that is the basis for seeking explanations from natural laws rather than some naturalistic fetish that's equivalently arbitrary to a Creationist's theistic fetish.

Show me a miracle and I'll show you legions of religious Scientists (Hume notwithstanding), but why don't we see people turning water into wine anymore, or parting great oceans, or violating conservation of energy?

Posted by: Gabriel Rocklin | May 17, 2005 3:47:10 PM

Thanks for this post.

In some ways, I'm glad for the onslaught of creationism on science. It'll clear up the metaphysical assumptions that science makes, and it's about time! Science will be the stronger for it, but so will our faith (just perhaps not in "God").

Posted by: Poéthique | May 17, 2005 3:48:08 PM

I don't think your distinction is quite right.

If there were supernatural beings that, in regular and law-like ways, operated in our universe then science would be able to incorporate their actions into their theories.

For example, angels could be operating in our universe in a way that is about as predictable as human behavior. If that was true, wouldn't sciences of Angel Psychology and Sociology pretty rapidly arise?

The problem with Young Earth Creationism is that it makes no testable predictions. In fact, it makes no real predictions at all. It has no explanatory power. The universe would "appear" no differently if it were true than if it were false.

Which is what I think you were getting at. However, the reason that science rejects the supernatural is not because it declares those explanations out of bounds by fiat, but because supernatural explanations generally don't have any explanatory power or make testable predictions.

And testable predictions, so the quaint view goes, are science's bread and butter.

Posted by: Patrick | May 17, 2005 3:48:33 PM

A god by definition must be supernatural. Lets not muddy the water. Only carp like that.

Posted by: Michael7843853 | May 17, 2005 3:48:38 PM

The point is that "God just made it all that way" adequately explains anything you care to have explained about the fossile record in the sense of providing an answer as to how come our data looks the way it does.

MY, if you're willing to go that loony path, you could just as well say none of us existed 5 seconds, and the reason why we think we did is because God just placed memories in our head which made us think we did. People who propose "explanations" like yours are not merely anti-science, they're anti-reality itself.

Posted by: Dan the Man | May 17, 2005 3:50:44 PM

Biologist speaking here, and I just have to weigh in on this discussion of what science is and isn't. Science is a method, not a supposition. The method requires that you make a testable guess (hypothesis is the official term) about the meaning of a set of observations. The test must involve some form of measurement. If the guess tests out as right (and that means results have to agree with your predictions better than 9 times out of ten in the social sciences, bettern than 19 out of 20 in biology, and 99 out of a 100 in physics)--if it's right, then you are not done yet.

The other required component of the method is that it has to be repeatable by other people following the same methodology originally used.

And that's why creationism isn't in the same realm. God is a concept akin in some ways to love or art or wonder. None of them can be measured. That doesn't mean they're not important or don't exist. It just means that science can't tell us about them.

If they can't be measured, they can't be tested, and the results certainly can't be repeated. Science is not involved.

(As an evolutionary biologist, this whole question is one of my pet peeves. If you'd like to see the long form, I have an essay You Can't Believe in Evolution that talks about these issues at greater length.)

Posted by: quixote http://acid-test.blogspot.com/ | May 17, 2005 3:56:04 PM

The way I'd like to put your ad hocness criticism is in terms of simplicity. The best explanation is the simplest one that explains all the data. Simplicity consists in invoking the smallest number of laws. The wonderful thing about evolution is that all the laws it invokes are laws of chemistry, physics, and other sciences that can be confirmed in the lab and which both sides of the debate already accept. If one wants to cash out "Things go the way God wants them to" or something like that as a law, that's okay, but now you've got a less simple explanation. (Certainly there are other reasons to think it isn't a candidate for lawhood, since it makes reference to an individual, for example.) You also pick up the burden of explaining why God wants all the things he does, and you've got to give a good simple explanation of that. Probably the best candidate, given the evidence, would be that he wants to fool us into believing evolution, but I don't think the creationists feel like saying that.

This makes the word "adequate" in "adequate explanations" look pretty sinister to me, since it seems like a way to permit crappy non-simple explanations with too many laws. Go for the best explanations, dammit!

Posted by: Ethical Werewolf | May 17, 2005 4:02:22 PM

Science should not be defined by the content of the answers its seeks, but by the method we use to seek those answers. It is a process of inquiry which involves the development of theories, the derivation from those theories of empirically testable hypotheses, and the empirical testing of those hypotheses. An essential element of this is that testing pre-supposes the possibility of falsification. If there is no conceivable test that could falsify the theory, then it is unscientific. (Notice: that does not mean the theory is wrong, only that the theory does not fall under the purview of scientific inquiry.)

This is why intelligent design is not science. It starts with the observation that there are facts that evolution may have problems explaining. It then offers a "theory" to account for those supposed anomalies. But at this point, it is done. There are no empirical implications other than those facts which the theory was designed to explain. There are no tests that could falsify it. It is an intellectual dead end that one must accept or reject on faith--and faith is not the currency of scientific inquiry.

Posted by: KenS | May 17, 2005 4:04:01 PM

Here's the key sentence right here:

Science presupposes, rather than demonstrates, that the supernatural does not play an important role in the unfolding of the universe.

The fact that we call science a method of inquiry doesn't change the fact that this method makes a metaphysical assumption which is not unassailable.

Don't need to use the scientific method to figure out that we make assumptions all the time about the state of the universe, all the workings of which, science, as a discipline, has not yet figured out.

Or in Rumsfeld's words (and those of others before him), you don't know what you don't know. Science presupposes this also, lest we forget.

The genius -- if you can call it that -- of Intelligent Design is that it incorporates this "don't know" assumption, as a way of course to attract people to the God-ly Design (which is a "knowing" assumption, and contrary to science).

Posted by: Poéthique | May 17, 2005 4:30:35 PM

You mention carbon dating. What you say is not incorrect, but carbon dating is only good for about 100,000 years. Various other dating techniques, such as Potassium/Argon, give estimates for very much older things (billions of years old).

I mention this because creationist are forever whinging about problems with carbon dating (as it happens, understood and avoidable problems, mostly having to do with the source of the carbon in the thing being tested) even though carbon dating is not used for mush of anything prior to the human era (that is, very useful for anthropology and archeology but less so for human origins and not at all for dinosaurs.

To Dan the Man: I fail to understand your post. You seem to be blaming Matt for making the point you seem also to be making.

Posted by: David Margolies | May 17, 2005 4:41:23 PM

Whoops, so sorry about the bold. Hope this works?

Posted by: Poéthique | May 17, 2005 4:51:34 PM

It didn't...

Posted by: Poéthique | May 17, 2005 4:52:10 PM

I'm not clear on the philosophy here... but falsifiability seems to be a pretty big barrier to entry for the supernatural to clear, unless it's some kind of magic (bu-dum).

Posted by: TJ | May 17, 2005 4:54:51 PM

Two minor comments. First, not all natural events are in principle predictable by via natural laws: that's the point of quantum mechanics. About some events we can only discuss probabilities. This general point goes beyond the behavior of subatomics particles. Would the course of history be predictable in any meaningful sense given "complete" knowledge of scientific law? We don't know the answer to that question.

Second, at the outer reaches of cosmology, the the idea that some fundamental scientific laws might not be unchanging is actively discussed. So, too, is the notion that basic scientific relationships might assume a different character in other universes (if indeed other universes do sometimes emerge from the quantum flux).

Posted by: Matt | May 17, 2005 5:17:40 PM

Lemme just clarify for the non-clickers-through that I did say something to the effect of: So long as you preserve the familiar hypothesis/empirical test methodology, whether you want to say you're looking for "supernatural" or "natural" explanations won't make much difference. Put another way: Whatever we can talk about in terms of that sort of investigation will count as "natural" for any practical purpose, even if it looks a lot like a "ghost" or a fairie or whatever. We often discover entities that don't obey what we had previously thought were universal natural laws... and then we revise the laws. (None of this, of course, should be read as remotely sympathetic to the creationists bad-faith case.)

Posted by: Julian Sanchez | May 17, 2005 5:17:49 PM

Possibly dumb question: what does an Intelligent Design Scientist DO all day? What is the ID research program? As far as I can tell, ID'ers spend all their time toting up what they don't happen to know, making probability calculations without underlying data to work on, and then say: "See, intelligent design."
Let's see one of their NSF grant proposals.

Posted by: C.J.Colucci | May 17, 2005 5:34:09 PM

Thanks for the clarification Julian. Interesting essay, quioxtic. But I think it's a bit overly simplistic to say that there's no faith involved in science/facts/considering the natural world. At some level, the scientific approach rests on a certain faith in one's ability to perceive the natural world, and specifically in the normal senses (taste, touch, smell, sight, hearing, etc.) that we use in that process. It also requires, as Poethique notes, certain metaphisical assumptions.

The differentiating factor to me between the natural and supernatural would be that the supernatural falls into the category of things that we experience in ways not through the normal senses. The obvious example here is God--most sane folks would agree that you can't see, hear, taste, touch, smell God in any of the standard ways those senses are used, but many believers have a very real experience of God's presence and interaction with them. These experiences, by their nature, are outside of our normal senses, and outside of science (although there is some interesting research using brain scans during meditation or prayer that, while hardly documenting existence of God, does give us an interesting way to "see" some of the religious experience in natural/scientific terms). Similarly, miracles are outside of science because the mechanism by which they "work" is outside the realm of what we can access through our normal senses (we see the results, but not how they come to be). So, to the extent that we could see, etc., fairies or angels (or perhaps that we could do so consistently, as opposed to the rare manifestation), then I'd agree with Julian that they wouldn't be supernatural but natural. But that's not the case.

Posted by: flip | May 17, 2005 5:40:16 PM

Of course there could be evidence for creationism. Let me sketch an example.

Suppose a nutty proponent of some variant of Intelligent Design spends decates looking at the multitude of "junk" DNA that people are full of in an effort to figure out why the Creator put it there. By sheer chance, he finds a transformation from what's effectively a four-letter alphabet into Biblical Hebrew. It's a robust "decoding" procedure, not just an ad hoc one, in that when you apply it to different organisms you always get grammatical Biblical Hebrew. Suppose further that a large part of this Biblical Hebrew consists of statements of laws of nature, and that when these laws are put to the test and compared against current theories, they invariably turn out to be correct. If some of the other Biblical Hebrew transforms into the Christian Bible, statements of the form "I am God, I created the world in exactly six days", etc., that would constitute conclusive victory for creationism.

Frankly, I think that if we've got to define science so that explanations that could be true must be arbitrarily excluded from consideration, that's a problem with science, not with the excluded explanations. The real problem with creationism is that we assign it a low prior probability, not its alleged un-evincibleness.

Posted by: Anton | May 17, 2005 5:59:03 PM

"what does an Intelligent Design Scientist DO all day?"
As far as I can tell, PR. Which is a problem, beacause while (according to one recent study), most scientists spend their time doing teaching and research instead of pubic outreach (silly them), leaving a wide-open niche for creationist spawn to proliferate madly.

If you're interested, Pharyngula and The Panda's Thumb are two connected evolution blogs about this absurd controversy, while Talk.origins Archive is the vast motherlode.

Warning: the two blogs can be a bit of a rough neighborhood . . .

Posted by: Dan S. | May 17, 2005 6:07:14 PM


You've proposed a "test" that could prove creationism to be right. But there is nothing in creationism that requires the encoding you describe to be in our genes. Hence, not finding such encoding would not falsify the theory.

Can you come up with an experiment that could produce results that would contradict creationsim?

Posted by: KenS | May 17, 2005 6:15:03 PM

KenS - Matt didn't say the problem was that it was unfalsifiable, and that's not the view I was trying to refute. He said the problem was that you can't possibly provide evidence for it (because all data can be accommodated equally), and that's just not true.

There's a host of philosophical issues surrounding falsifiability (Quine: can't you always come up with more ad hoc hypotheses to explain away recalcitrant data?) but I think they're beside the point. Falsifiability tests for adequacy, whatever their other merits, get us to the wrong conclusion: that some (internally consistent) explanations ought to be thrown out as unworthy of consideration no matter how much evidence there is for them. If we find the equivalent of an unambiguous "GOD WAS HERE" stamped on the universe, evolutionary biologists shouldn't just keep going as normal even if the theory is unfalsifiable. If science requires falsifiability, something is wrong with science, not the rejected theories.

Posted by: Anton | May 17, 2005 6:33:53 PM

If God were to pop out of the data, that's not the end of science.

Posted by: yesh | May 17, 2005 6:40:33 PM

Perhaps someone may convince me otherwise, but it seems to be me the crux of the problem between science and faith is the notion of measuring, specifically with units of time.

That is, none of us perceive time as divisible and measurable units -- we only perceive clocks and whatnot. Relying on them, we assume, or perhaps more accurately, abstract -- but never perceive -- that time is divisible, hence measurable in quantitative terms as if it were material, and therefore theoretically artificially reproducible. None of us, including most scientists I think, ever take the assumptions that far (creating time?), but they logically follow, do they not?

Having a "revelation," going through "enlightenment," or whatnot, is it not sensing, perceiving the fundamental indivisibility of the universe? i.e. becoming aware of time not as a scientist would, but as all those who would "think" unscientifically, or perhaps more accurately, cease abstracting altogether, and rather simply perceive?

Science takes perception out of the equation. See, discovering the "natural laws" of the universe doesn't change them one bit, does it? It's merely a way of grasping some of its mechanisms so that we may communicate them, employ them, etc. The idea that it might explain everything ("42!") is merely a motivation -- it's faith -- and in this way science is no different than any other human endeavor in which those who choose to devote their careers to it feel compelled to denigrate alternative ones while touting their own. It's highly convenient that those who say science is the be-all-and-end-all method of knowing the universe are scientists or "believers" in science. Science, in the end, only asks to be useful.

Is it?

Posted by: Poéthique | May 17, 2005 6:46:47 PM

MY: The point is that "God just made it all that way" adequately explains anything you care to have explained...

Well, you would then have to believe in a God that would is willing to intentionally deceive people who merely happen to be curious, which doesn't really jive with the concept of an all-loving God and meshes much more with the concept of a God who wants a bunch of followers who don't ask questions.

If I wanted a God like that, I'd join a cult.

Posted by: fling93 | May 17, 2005 7:27:59 PM

The comments to this entry are closed.