Science seeks to characterize the nature of the universe, including the earth and all its attributes. It seeks to understand how the universe is constructed, how it came to be constructed in the way it is, and what those processes are that have produced the patterns we see in the natural world. Science does not work by simple pronouncements of fiat; rather, to qualify as science, an area of enquiry must attempt to explain natural phenomena in such a way that its statements can be tested by experiencing the natural world. More simply, we have to be able to go to nature to assess the veracity of the statements we make about it. If statements are not subject to such scrutiny, to verification by experience, they cannot be scientific. It is on these grounds that I characterize evolutionary biology as scientific and creationism, in whatever guise, as nonscientific.
Today we have only two remaining, but totally conflicting, bodies of statements that account for the diversity of life on earth. One is evolution. It says, basically, that all organisms are related by a process of ancestry and descent. It says that there is a particular nested set of resemblances we see in nature that unites all living things. For example: dogs, wolves, and coyotes, and other closely similar animals share certain resemblances not found in any other organism; we unite them accordingly into the family Canidae. We observe, in like fashion, that Canidae share certain similarities with Felidae (cats), Ursidae (bears), and several other families; accordingly we unite them into Order Carnivora. Carnivora share some attributes with other orders not shared with the rest of the known biotic world; hence we recognize Class Mammalia. And so forth. All of the biotic realm is structured this way. Ultimately, we predict that some attributes must be common to all life (RNA is an excellent example).
According to evolutionary biology, this pattern of nested resemblances is the straightforward, expected result of ancestry and descent: new characters arising from time to time are inherited by subsequent descendants. A hierarchical arrangement of similarities is the inevitable consequence.
Since all organism are held to be related by this process, the major prediction of evolutionary theory is that there is one single nested pattern of resemblance linking all organisms in nature. Now, the discipline of systematic biology, including paleontology, tests this proposition daily. Biologists, in analyzing their specimens, predict that newly studied characters will conform to the preexisting hypothesis of the nature of resemblances among them. If the general proposition that evolution has produced a single, nested pattern of resemblances among all elements of the biota is false, its fallacy would long ago have been exposed. If evolution were a false theory, there would be every basis for predicting that there would not be a single, nested pattern of resemblance among all organisms on earth.
Creationism says this apparent order in the biotic realm of nature did not arise by ancestry and descent among all organisms. Rather, in its purest form, creationism holds that each species is created separately by some supernatural creator. (Some creationists admit that some relationship on a small scale—say between different species of the genus Canis [wolf, coyote, dog]—may occur, but not between major "types" or "kinds"—meaning higher levels of the taxonomic hierarchy, between families, orders, classes, and so on. They admit the nested pattern exists at low levels, but deny that it does at higher levels, or at least claim that it does not signify relationship.)
The assertion of independent creation, in whatever specific guise, does not lead to a single generalization about organisms or any observationally testable predictions. In short, it is a simple, fiat assertion with no practical consequence allowing us to test it further in nature. It therefore cannot be construed as science.
These points are rather obvious. Creationists therefore spend most of their time attacking proponents of evolutionary theory. Their general line of reasoning is this: not all evolutionary biologists are agreed on either (a) the exact details of evolutionary history or (b) what precisely the mechanisms of evolution are. They present conflicting views and delight to find dissenters (like myself) who are known to be dissatisfied with one or another aspect of current evolutionary science. They try to use internal disagreement among evolutionary biologists as evidence that somehow evolutionary biology isn't science after all. In so doing, they again mistake the nature of science.
There is no field of science today whose adherents and practitioners are agreed on all points. Science cannot possibly work that way—it proceeds by evaluating conflicting views on the nature of the world by testing hypotheses (that is, by experience). To progress, science needs those conflicting views. A science is neither healthy, vigorous, nor even alive without such disagreement. Unanimity, in science, is generally a sign of stagnation. Biochemistry, nuclear physics, and all other major branches of science are the same way—rife with disagreement. In any typical science, at any point in time, some generalizations are fairly well agreed upon, while others are inevitably bones of active contention.
Creationists are fond of asserting that there are no intermediate forms between "major kinds" in the fossil record. To document their position, they contact men such as Colin Patterson (British Museum) and myself, who have been among those paleontologists who have felt that paleontologists in general have tended to be a little over-enthusiastic in dubbing particular fossils "ancestors." We have been concerned with the logic of verifying such statements. We urge caution. But we do not say that ancestors or transitional forms never existed or were never fossilized. From my own work I can cite the trilobit genera (from the Lower Devonian of Bolivia): Kozlowskiaspis—Metacryphaeus—Malvinella—Vogesina, which are connected by a compelling array of intermediates. Creationists can scoff at such series, familiar to all systematists and paleontologists, but the fact remains that such series exist and are consistent with the notion of evolution.
Moreover, the supposed lack of transitional forms trumpeted by creationists is analogous to the inability—of all of us—to see and therefore objectively to attest to the existence of atoms. Yet, I do not recall hearing anyone, creationist or not, seriously questioning atomic theory. This is because predictions arising from atomic theory can be tested and verified without anyone actually seeing individual atoms. The single nested set of resemblances uniting all organisms is the analogous prediction in evolutionary biology.
Finally, I will comment on patterns of occurrence of organisms in the fossil record. Standard evolutionary theory predicts gradual, progressive, incremental change leading from one species to the next. Most phyla (among those with readily fossilizable skeletal parts) originated in the Cambrian Period. Thus the fossil record is hard put to verify this particular prediction of evolutionary theory. Aha! cry the creationists. Geologists and paleontologists have for years replied that the fossil record is too incomplete to retain the record needed to substantiate this particular prediction of evolutionary theory. Some of us now think that the predictions themselves are inaccurate, that the general notion of evolutionary ancestry and descent need not imply a gradual, progressive pattern of improvement and development of higher and higher forms. Aha! cry the creationists once more; biologists admit that the fossil record falsifies evolution! Not so, we reply: some details of evolutionary theory—notably that part which calls for slow, steady, gradual accumulation of change—is evidently in error, not the general notion of evolution itself.
In short, the notion that organisms are related by an evolutionary process of ancestry and descent is the only scientific theory which explains the hierarchy of resemblance among all living things. It is theoretically possible that it is wrong (else it could not be a scientific statement). It is, however, constantly being verified by scientists predicting distributions of characteristics, then checking their organisms, and verifying the predictions. Scientists disagree to some extent about how the process works and about what the more important evolutionary factors might be. This is normal science. Creationism does none of this. It gives us no testable, verifiable predictions about the nature of the organic world, and its objection to evolution as nonscience is not correct.