Chris Mooney and I had a brief e-mail debate a little while back about the topic of the piece he’s got up at The American Prospect today. He writes:
Kansas’s previously proposed science standards had appropriately defined science as “the human activity of seeking natural explanations for what we observe in the world around us.” Anti-evolutionists want to change this language to the following: “Science is a systematic method of continuing investigation, that uses observation, hypothesis testing, measurement, experimentation, logical argument and theory building, to lead to more adequate explanations of natural phenomena.”This may seem harmless at first glance. But the change carefully removes any reference to science’s search for natural explanations in favor of “more adequate” explanations, creating a opening for creationists to insert the supernatural. Such a change reflects the fact that the new generation of anti-evolutionists has launched an attack on modern science itself, claiming that it amounts, essentially, to institutionalized atheism. Science, they say, has a prejudice against supernatural causation (by which they generally mean “the actions of God”). Instead, the new anti-evolutionists claim that if scientists would simply open their minds to the possible action of forces acting beyond the purview of natural laws, they would suddenly perceive the weaknesses of evolutionary theory.
Now, far be it from me to defend those who falsely claim the mantle of science—and attempt to manufacture controversy where no serious debate exists—in order to sneak their religion into science textbooks. But I think Chris is off-base here.
The problem is, I very much doubt there’s any sort of meaningful distinction to be made between “natural” and “supernatural” accounts in this context, so long as we’re agreed that the proper methodology is the familiar scientific procedure of building theories and hypotheses, then testing them empirically.
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? Could we know, a-priori, that some budding Egon Spengler wouldn’t come up with a scientific test that would detect spectres as easily as we now examine radio spectra?
Turning it around the other way, didn’t some of the conclusions of quantum mechanics strike many classical physicists as “spooky”? Doesn’t science, too, hit rock-bottom at some point, with no further account to be given of certain laws or forces, either for theoretical reasons or because of the practical limits on our ability to investigate? From what I understand about quantum indeterminacy, we can get a set of equations that’ll let us know the probability of a wave packet collapsing one way or another. But—as I understand it—there’s no lawlike way of knowing, even in principle, which it will actually be, or why it went one way rather than another. The actual outcome is, in at least one sense, “beyond the purview of natural laws.” We might even, slightly playfully, call it “supernatural.” But that doesn’t mean it’s stopped being science, does it?
More generally—since my recollection of the physics here is too fuzzy for me to want to place a great deal of weight on the specific example—it seems awfully ambitious to suggest that science is only concerned with inquiries where we can be sure of getting explanations all-the-way-down. How can we know in advance that we won’t ultimately bump up against an impassable question mark, either because of the boundaries of our technology and ability to investigate (as I think we’ve hit re: some propositions in particle physics already) or because, so to speak, the question mark is etched into the universe itself. Isn’t it, well, unscientific to suppose in advance that we know how far our capacity for discovery and further explanation extends—and to suppose that it has no limits?
Addendum: Chatted a bit about this with Chris and one thing I want to add is this: One way of defining “supernatural” is as operating outside natural laws. But that gets a little tricky too, since, of course, we’ve previously encountered lots of phenomena that proved to violate (what we previously believed were) the unchanging universal natural laws. We didn’t dub those things supernatural; we decided that our conception of the relevant natural laws was incomplete, that there were exceptions, and so on. However, having gone over this ground with Chris for a while, it’s not actually clear that we substantively disagree.
18 responses so far ↓
1 Daniel // May 16, 2005 at 5:09 pm
Well the change does remove any reference to science’s search for natural explanations in favor of ââ?¬Å?more adequateââ?¬Â explanations, and I suppose this does create an opening for creationists to insert the supernatural. But I donââ?¬â?¢t think thatââ?¬â?¢s really cause for major concern. On a semantic level, if the explanation is ââ?¬Å?more adequateââ?¬Â it doesnââ?¬â?¢t really matter if itââ?¬â?¢s natural or not. And from an historical perspective, science wasnââ?¬â?¢t really about the search for a natural explanation for observed phenomena; it was about the search for a ââ?¬Å?more adequateââ?¬Â explanation. Not that ââ?¬Å?naturalââ?¬Â really means anything anyway.
2 Marie // May 16, 2005 at 6:12 pm
Um, wouldn’t the gods be hanging out in the noumenal world with our free agency? Perhaps I need to read Kant’s Groundwork another time or three.
3 ChrisC // May 17, 2005 at 12:07 am
Julian,
Take off the theoretical philosopher hat here for a second, and think this through according to how science works. A few quotes you mention:
The problem is, I very much doubt there’s any sort of meaningful distinction to be made between “natural” and “supernatural” accounts in this context, so long as we’re agreed that the proper methodology is the familiar scientific procedure of building theories and hypotheses, then testing them empirically.
There is a very clear distinction between “natural” and “supernatural”. Natural is what is observed through repeatable testing. All scientific theory is repeatable. By definition, “supernatural” is unique events where a non-repeatable non-normal event happens that is not explainable by known scientific theories.
Let’s hit another point:
Turning it around the other way, didn’t some of the conclusions of quantum mechanics strike many classical physicists as “spooky”? Doesn’t science, too, hit rock-bottom at some point, with no further account to be given of certain laws or forces, either for theoretical reasons or because of the practical limits on our ability to investigate?
There is nothing “spooky” about quantum mechanics. There is a lot we don’t know about it, but we can make predictions about what will happen, and these predictions are repeatable. True, we don’t know which way an individual photon will go, but we can predict over a large amount of experiments the averages of where they will go. Just because our knowledge of physics is incomplete, does not mean it is invalid. By allowing “supernatural” explanations into physics, we would be opening up arguments such as “God declares where each photon goes.” I’m sure that you think that isn’t a correct position to be teaching high school students.
How can we know in advance that we won’t ultimately bump up against an impassable question mark, either because of the boundaries of our technology and ability to investigate (as I think we’ve hit re: some propositions in particle physics already) or because, so to speak, the question mark is etched into the universe itself..
A scientific theory is never complete. Most people don’t know this, but the fact the Earth revolves around the sun isn’t a scientific fact, but a theory that is confirmed by all known observations. Science doesn’t claim that we’ll know everything. As your statement states, there are a lot of questions in physics that we can’t test because the technology isn’t there. Does that mean that we must rely on a supernatural explanation? No, it means we must try to figure out new ways/technologies to do it. It may mean a better particle accelerator, or a better Hubble, but we are never going to hit a wall where the answer is “I don’t know, it must be God.”
4 Julian Sanchez // May 17, 2005 at 1:32 pm
I don’t know Chris, your definition would seem to regard as “supernatural” anything produced by an intelligence as opposed to simpler natural laws. (Of course, something made by an intelligent person or alien or whatever would still, in a sense, be produced by “natural laws” acting through our brains…) And that seems like it’s got to be wrong. I don’t think superadvanced aliens or deities or any other mind designed people. The arguments to this effect all strike me as transparent attempts to justify someone’s antecedent theological beliefs. Still, it doesn’t seem like the sort of thing that *couldn’t* have happened. Say we’re superadvanced noncorporeal minds in the future, and we design a little civilization on some barren hunk of rock. Would members of that civilization who uncovered evidence that this happened necessarily be unscientific?
5 Bill Newman // May 17, 2005 at 3:08 pm
I think it would be reasonable to say that some scientists found QM to be spooky, although I’d probably use terminology like “intensely unsettling and unsatisfying” instead. God rolling dice, anyone?
I find creation scientists and their sympathizers annoying for reasons other than the idea that there couldn’t be a God or spectres, or that theories shouldn’t be spooky. One reason is the vaguely Popperian one that their theories seem (from my unscientific sampling of creationists I’ve met and who will answer questions about their ideas) to be flat useless. It seems to be hard to find anything at all that their proposed theories exclude, or even consider startlingly surprising. What possible sequence of outcomes over the next decade would convince a scientific creationist that his theory was wrong, or at least overwhelmingly unlikely to be true? If absolutely no possible outcome would do it, then his theory seems fundamentally equivalent to “que sera, sera.” To my way of thinking, the main reason that we should study science is that it tells us what will happen in the world, or at least that some things won’t happen. If your science has to punt on that and fall back to saying that absolutely anything might happen and we simply cannot know, then why study it? (instead of studying aesthetics or literary composition or ethics or history or music or something)
(Both QM and evolution have randomness in them, so they tend not to absolutely exclude things. But they declare many things to be vanishingly unlikely. If the DNA sequencers had found that the Koran is inscribed in all eukaryotic genomes, it would have more than sufficed to convince me that they didn’t just randomly evolve under any ordinary natural selection…)
6 marc w. // May 17, 2005 at 6:01 pm
I don’t get it. The ‘new’ definition strikes me as self-evidently better, capturing the idea of science as process versus some specific way to investigate natural phenomena.
And the phrase ‘more adequate’ – it’s only terrifying if you’re prepared to argue that the creationists might conceivably ‘win’ thanks to this loophole; like any number of batshit insane theories that rely on a highly selective reading of the evidence or ‘antecedent theological beliefs,’ will suddenly be judged ‘more adequate’ than evolution. Maybe that’s a serious fear in Kansas, and maybe the practical application of this change will be awful for Kansan students. But it strikes me as unseemly that we should opt for a terse, less descriptive definition of science because if we use a more accurate one, the creationists will turn it into a legal superweapon capable of eliminating evolution, physics and perhaps math itself from school curricula.
7 Del // May 17, 2005 at 6:45 pm
No, quantum mechanics isn’t spooky. It’s a mathematically rigorous theory that says motion on a very small scale is probabilistic. It’s only counterintuitive because we live at a larger, classical scale. It’s only scientific because, as mentioned above, it makes predictions that can be tested and never seem to fail. The theory states that you cannot know a quantum state until the wave packet is collapsed. Period. That might irritate our inner philosopher, but shouldn’t bother our inner scientist because we can test it and see if it works.
The problem with evolution, of course, is that’s less rigorous and harder to test. That leaves it open to creationist criticism. Religious types once similarly criticized scientific ideas about planetary motion, but as the theories became rigorous the Perfect Solids of Plato and Pythagoras became obsolete. Maybe someday evolutionary biology will be so rigorous.
8 Luke // May 17, 2005 at 6:53 pm
Would members of that civilization who uncovered evidence that this happened necessarily be unscientific?
Well no, but they’d have, as you state, evidence. That is something that (at least at this juncture) would make them uniquely different from our current ID proponents.
Of course, we’d have to decide what exactly constituted “evidence,” but if we are going to assert the evidentiary claims for design are robust in your hypothetical case, then we’d be steps ahead of where we are now.
9 Julian Sanchez // May 17, 2005 at 7:22 pm
Luke-
Chris and I are agreed that the IDers substantive claims are, in fact, bullshit. The question is whether you need any caveats about “natural explanations” *above and beyond* the ordinary scientific rules of evidence and empirical testing of hypotheses. If something like ID *could* be true, it seems wrong to define “science” in a way that rules it out in advance… even if we’ve got lots of reason to think that, in fact, it’s false.
10 Barry // May 17, 2005 at 10:02 pm
Del:
“The problem with evolution, of course, is that’s less rigorous and harder to test.”
Not less rigorous – IIRC, there are three assumptions, both well-justified by data: (1) there is variation; (2) some of it is heritable;
(3) the variation leads to differing survival/reproduction probabilities.
Those three assumptions lead to probabilistic changes over time. Throw in mutation, and the changes can then be immense. (please forgive this clumsy exposition; a far more elegant form should be available on The Panda’s Thumb)
Testability varies, depending on the hypotheses being tested, just as in physics (e.g., we’re not yet at the level where even grad physics courses require students to create black holes).
“That leaves it open to creationist criticism. ”
But many things in physics are similarly hard to test and hard to comprehend withouth training.
The criticism of evolution draws it’s strength from a particular religous doctrine (a subset of American Protestantism, with others drawn in as tokens). IMHO, take away that feature of one subset of a subset of one country’s religion, and what’s left would be no more noticeable than the ‘Einstein was wrong!’ cranks who periodically bother physicists.
11 Josh // May 17, 2005 at 10:11 pm
I’d say it’s less about supernatural vs. natural than testable vs. untestable. There are scientific examinations of supernatural phenomena, but only when there are predictable patterns to it. If creationism is about divine fiat, it’s unscientific, if it’s about God’s design, there may be something falsifiable to come out of it.
12 Josh // May 17, 2005 at 10:11 pm
I blogged a longer response http://jgrr.blogspot.com/2005/05/supernatural.html
13 Del // May 18, 2005 at 12:28 am
Barry,
Points taken and generally agreed with. Except I really see evolution as being less rigorous. To say why this or that protein complex has one particular makeup instead of another one assumes it’s evolutionarily favorable; however the details are yet impossible to work out. On the other hand, to explain why the first excited state of oxygen is an 2x2x4 configuration can be done with a short answer (quantum mechanics) and a highly rigorous, long answer (calculations showing exactly why it is the first excitation and how long we can expect to wait before it decays — something relatively easy to test). Both questions involve expertise to answer, but evolutionary biology does not yet lend itself to long answers nearly as often as physics does.
14 Bill Newman // May 18, 2005 at 6:46 pm
Del wrote “To say why this or that protein complex has one particular makeup instead of another one assumes it’s evolutionarily favorable; however the details are yet impossible to work out.”
Some things are easy, though. E.g., even by 1980 (and probably by 1965, though I don’t know the history) it was clear that the human gene pool is relatively rich in simple mutations that provide half-assed, expensive, but badly needed protection against some horrendous threat, typically a pathogen like malaria. That doesn’t look much like design (or, for that matter, the Ark). We have learned a lot since 1980 — more chemistry, more biology and pathology, and enormously more gene sequences — and this and other patterns have gotten stronger. The statistical relationships between genes and the evolutionary tree that people expected from anatomy and fossils is a pretty impressive pattern, and pretty hard to explain for any theory which doesn’t involve everything starting from a simple common ancestor. That doesn’t exclude the possibility of spectres coaxing the common ancestor’s descendants into spiritually correct directions, but it is a difficulty for most pre-1500AD models, which tend not to involve a simple protoorganism ancestor at all.
15 Del // May 18, 2005 at 7:31 pm
Bill: Well, nothing is “pretty hard to explain” when one invokes an all-powerful designer! But, to use your malaria example, it begs the question as to why everyone doesn’t carry innate immunity. Perhaps someday we’ll have a model that says something like the relevant mutation increases malaria resistance by 80% but decreases metabolic efficiency by 4% (unless there’s a compensatory mutation present…), and feed that into a population model and predict that the mutation is equilibrated at 30% of the population (in a particular environment, etc). At this point we can only assume such a tradeoff exists because evolution makes so much sense in the big picture. Of course you can say that this is all just details, but these are just the kind of missing details that ID goofballs celebrate.
16 David Billington // May 19, 2005 at 9:00 pm
Chris: The definition of the supernatural you give is one way to define it, in terms of its method of occurrence. But if you include claims of paranormal phenomena, I don’t see any necessary reason why these couldn’t be repeatable as well. For example, if an electrode inserted in a particular human brain region can reliably produce a subjective out-of-body experience, it should be possible to construct a test that only a disembodied mind could pass. If everyone could have an induced OBE and could pass this test, it would be a repeatable event. The distinction of the supernatural only holds if it is defined in a way that makes scientific demonstration impossible. What I think Julian is saying is that the supernatural might also be defined more broadly. This is not to say that everything associated with the supernatural can be demonstrated.
17 Josh Narins // May 20, 2005 at 8:57 pm
One peice of good news for the atheists around.
Traditionally, religion has had a lot of “play” when it comes to legal topics related to morality (“Woe betide those who stand against laws against adultery!”)
I’ve also been led to believe that when Religion takes on the Scientists (directly, as in the ID debate) they get this biggest whoopings.
18 David Billington // May 21, 2005 at 12:33 pm
An impersonal universe could have an ordering principle too. Evidence of order (“design”) in nature does not necessarily imply an intelligence. For evidence of intelligence, you would need to show that the universe is grounded in some kind of super-consciousness, not just that it has order.