In connection with a creationist lawsuit that I read about awhile ago in California, the creationists' lawyer was quoted as saying to a reporter, "They'd better be able to prove that evolution is right." Thus, if evolution cannot be proved beyond question to be fact, the lawyer was saying, then it's just as much a theory as creationism, and if one deserves to be called scientific, so does the other. Now, I certainly wouldn't dispute the right of anyone to believe what they choose and to express those beliefs freely. But when attempts are made to pass those beliefs off as science, when they're nothing of the kind, and to force them on the educational system as such, that's another thing entirely.
The fallacy with the above claim is its assumption that all degrees of uncertainty are equalby the same logic we could argue that since George Washington can't be proved absolutely and conclusively to have existed, then the case for Santa Claus is just as solid. But more interesting is the cause this gives to reflect a little on the differences between science and other belief systems. For everything that I read of the public debate surrounding the case succeeded only in missing the point of the issue entirely, which has nothing to do with whether evolution or creationismor neither, for that matteris right or wrong. The issue is understanding what it is that makes a belief genuinely "scientific."
Science never claims anything to be finally and absolutely proved. So in that sense every scientific belief is "just a theory." A demand for evolution to be proved beyond a question to be fact is an absurdity that fails to comprehend what science is all about. Although we speak loosely of a theory being "proved by experiment," philosophically this is impossible. The best that an experiment can hope to do for a theory is fail to falsify it. An experiment does this when all of the results agree with the theory, and not one observed, incontestable fact contradicts it. Even this doesn't prove a theory truethe agreement could still be a coincidence due to other, perhaps totally unsuspected causes. All that can be said is that the theory wasn't shown to be wrong.
The difference sounds like hairsplitting, but it's crucial. It means that in science your theoretical boat must be watertight everywhereone hole is enough to sink it. For example, experimental results and astronomical observations accumulated before the late nineteenth century all accorded with Newtonian mechanics. Later measurements, however, such as the speed of light in different directions and of the precession of Mercury's orbit, contradicted its predictions and were sufficient to invalidate it as a general model. Since then, of course, relativity has accommodated the new facts while continuing to explain the old; but this no more proves relativity to be "true" than Newton's unification of terrestrial and cosmic motion proved the earlier ideas true. As was the case with the Newtonian system for almost three centuries, relativity has so far survived all attempts to find a flaw. It skates on the same pond.
One of the most important criteria that science demands for a theory to be acceptable as a serious candidate for consideration is that it be falsifiable. It was precisely the fact that it could be tested and shown to be false that made Newton's a genuinely scientific theory. Whether it ultimately survived or failed the tests is not the point. This is why the contention that "some UFOs might be alien spacecraft" is not answerable by scientific inquiry. Yes, some of them might be, and however many specific instances are shown not to be, the speculation can never be shown to be untrue. It is unfalsifiable. (The inverse theorythat no UFOs are alien spacecraftcan very simply be proved untrue, but not by any of the claims submitted to date as evidence.) And creationism is unfalsifiable, since every fact, every new discovery that might be made, any result of any test that might be imaginedcan all be dismissed with the assertion that "it was created that way." Hence, even before any debating or appeals to evidence, creationism fails the most basic test. It is not "scientific," and no amount of legal semantic juggling can make it so.
Science has never made any claim to infallibility. What makes a belief scientific isn't whether it turns out to be true or not, but the process by which it is arrived at. It seems ironic that those who preach knowledge of absolute truths tend to be the first to accuse scientists of being arrogant.
How does creationism measure up against other criteria by which a belief is judged as being scientific? Well, in the first place science seeks to be objective, which means having no preconceived notions about how things ought to beno advance commitment to any holy book, sacred doctrine, or political, social, or economic ideology that has to be upheld as a first priority. Science is an open system based on skeptical appeal to evidence, and uses inductive logic to formulate general principles from specific observation. A pseudoscienceof which creationism is an exampleis a closed system based on uncritical appeal to doctrine, and uses deductive logic to infer specific accounts of how the world must be to conform with its inviolate principles. The goal of science is to discover what the reality out there is. To this end its conclusions are always open to revisionor might even be abandoned completelyin the face of new evidence: the belief structure follows the facts. But a pseudoscience contrives to uphold a belief structure that exists before any evidence is considered, and to maintain itself subsequently, it must deny or distort any facts that it finds inconvenient. Darwin did not set out to prove evolution because it reflected any ideas he might have had about how societies should function, or because he found it emotionally appealing. He ended up proposing it because he found it best explained the facts available to him. Like an explorer, which it is, science charts new territory as it finds it, and while at times it might base its choice of direction on hunches, it draws its maps from the reality that it encounters. it accepts that whatever is true will remain so, with complete disregard for the intensity of human convictions or the number of people who can be persuaded to share them. Is this creationism?
When science has amassed it facts, it attempts to construct a theory to explain them. A successful theory must be consistent with all the facts, contradicted by none, and survive experiments aimed at proving it false. It sometimes happens that two or more rival theories meet all of these requirements, which poses the problem of having to choose between them. The guideline in a situation like this is "Occam's razor," which says, in effect, go for the simplest explanationthe one that accounts adequately for all the facts and requires the fewest assumptions. It won't guarantee that we'll never have to change our minds later, William of Occam warns . . . but it's the way to bet.
This means explanations based on familiar, well-understood principles are to be preferred over far-out fanciestrue, a poltergeist might have upset the cookie jar, but the presence of a three-year-old in the household suggests a more promising line of inquiry. Ideas which are compatible with existing, well-supported bodies of knowledge and experience take precedence over ones that aren't. Thus, when a disciple of the maharishi shows a photograph of somebody allegedly levitating, the onus is on him to prove that the picture is genuine (easily done by staging a repeat performance before witnesses), not on us to accept it. Extraordinary claims require extraordinary proof. Hence the skepticism of most scientists toward, for example, claims of ESP abilitiesthere is nothing in the realm of undisputed, demonstrable fact that can't be accounted for more simply by such familiar mechanisms as carelessness, self-deception, or deliberate fraud, which we know happen. Introducing "paranormal" phenomena to explain any of it is neither necessary nor justified. A new, more elaborate theory is called for only when facts have been shown, incontestably, to be true, which cannot be reconciled with the simpler, already existing model. Relativity was accepted because it explained observations that Newton's system couldn't; quantum mechanics arose from the failure of measured results to support earlier models of thermal radiation.
Evolution does indeed offer a consistent and comprehensive interpretation of facts accumulated from a whole range of disciplines that include paleontology, geology, zoology, botany, embryology, biochemistry, comparative anatomy, anthropology, and behavioral psychology, to name just some. It establishes a common framework within which observations collected from many fields of investigation and tested independently all fit together and have reason for being the way they areas geneticist Theodosius Dobzhansky put it, "Nothing in biology makes sense except in the light of evolution." The facts are explained. But is the evolutionary explanation the simplest? Although the assertion may appear strange to some on first sight, yes, it isit's the simplest that can be offered for the facts available at the present time. Simplicity here has nothing to do with the shelves of books devoted to the theory of the years of study necessary to comprehend it. It has to do with assumptionshow much has to be taken on faith. The less that is assumed, the simpler the theory. And evolution requires very little: merely the laws of physics and probabilityintroduced, moreover, not on an ad hoc basis to prop the theory up, but as derived independently within the entire body of scientific knowledge.
What can be said in this respect of creationism? First, it offers no reason whywhy the anatomy of different species should show any relatedness at all; why the fossil record should show its progressive accumulation of change; why embryos of different species should be more alike at earlier phases of growth, why isolated populations should diverge. . . . The inevitable rejoinder that "it was made that way" merely acquiesces to the fact; it explains nothing. We could say the same about anything and add not one scrap to our understanding of it. But on the other hand, creationist theory requires the assumption that a creator exists, that a supernatural judge of morals exists who is concerned about the day-to-day affairs of people on this planet, that both these being are one and the same, that it communicated its motives to chosen writers of ancient books, that those writers were correct, and honest, in interpreting the source of their inspiration, that later translators were equally infallible . . . and a long list of similar premises that no physicist would entertain for a moment as a basis for constructing a theory. The property insurance would be pretty expensive to cover an edifice built on foundations like that.
A scientific theory explains most and assumes least. Creationism explains nothing and assumes everything.
Lastly, the value of a scientific theory is judged by its power to predictnot in the sense of "psychic" predictions headlined by supermarket tabloids, but in the sense of predicting further experimental results. One failed prediction is enough to torpedo a theory. Success with every prediction, on the other hand, means only that it has survived everything thrown at it so far. So, if evolution is valid, the newer discoveries made since its inception ought to be consistent with it. Apart from some haggling among specialists over relatively minor details, this has turned out to be overwhelmingly the case. Darwin and others predicted the essential properties of inherited genetic units, even though genes and chromosomes were unknown at that time. From evolutionary theory, DNAs from different species should exhibit a branching pattern that reflects the same time sequence of divergence as it is deduced by other methods; they do. The primitive metabolic chemistry of ancestral organisms should be discernible in today's organic cells; it is. There shouldn't be much difference in the genetic code inherited by all organisms; there isn't. And so it goes.
And of the predictive power of creationism? Can it predict which band in a series of tree rings should indicate the same age as a given mix of carbon isotopes? Or the tidal record that ought to be found written into fossil corals by the moon's orbital motion of several hundred million years ago? Does it have anything to say about the composition of the early atmosphere and the kinds of minerals that would be formed as a consequencetheir chemical nature, where they should be located, and at what depths we should expect to find them today? Can creationism, in fact, give a hint of any future finding? Not a one. It operates with hindsight only. Because of its built-in unfalsifiability it can cobble together an explanation of anything at allbut only after the fact as established by other means. As a method of prediction it is sterile.
In its demand to be recognized as a science, therefore, creationism not only fails all the basic tests, but from the credentials that it presents shows no comprehension of what constitutes a science. It should be taught in schools, certainly, to those who wish to study itstudents of religion, of philosophy, of history, of the development of human thought. But there can be no trying to pass it off as a science. The suggestion of "scientific creationism" is a self-contradiction.
If one chooses to think in terms of the revealed word of a God, then surely whatever God has to say is written in the language of life, the world, and the physical universe around us. This is a form of scripture that comes direct from the source, leaving no room for doubts about authenticity, suspicions of forgery, or uncertainties over translation. And the Word that the processes of physics, biology, geology, astronomy, and cosmology are revealing is that the universe and everything in it have been evolving for as far back as it's possible to track time. I think that EVOLUTION IS THE REVEALED WORD OF GOD would make a good bumper sticker.
Such thoughts lead to stories like Making Light. . . .
Afterword, 1997
Isn't it interesting how the same facts and arguments can take on a different light when seen from an altered perspective. Ten years or so after writing the above, I found myself calling Jim Baen to suggest that perhaps it ought to be omitted from the rereleased version of this book. I explained that I was now a lot less persuaded by the orthodox theory than I had been. Specifically, I had reached the conclusion that evolution didn't take place gradually but in huge leaps, and natural selection wasn't adequate to explain it. Also, I thought the creationists had some good evidence to present for the notion of the Earth being a lot younger than conventionally taughtalthough not the 6,000 years that Biblical literalists insist on.
Jim was interested in this change of view and thought others would be too. Therefore, he suggested instead leaving the piece in as it was, and writing a further one to elaborate on the reasons. This we did, and it appears as "Evolution Revisited" in Rockets, Redheads, and Revolutiona late inclusion in that collection, which is why this afterword was written in 1997.
For those who are curious, the book that started me rethinking the subject, and which I'd recommend as a start for anyone else interested in taking a look at the other side, was Evolution: A Theory in Crisis by the molecular biologist Michael Denton, published by Adler & Adler, 1986.