Disagreements between creationists and evolutionary theorists have directed attention to a wide variety of issues that deserve philosophical scrutiny. In particular, in legal and educational contexts, the creation/evolution debate has contributed to a widespread uncertainty about whether to depict evolution as "fact" or "theory," or perhaps as both. I propose that a distinction between data and phenomena can clarify how, depending upon context, the term "evolution" can refer either to a complex fact or to an explanatory theory.
Examples of the dispute are easy to find. Both the 1981 Arkansas and 1986 Louisiana "balanced treatment" bills included references to the allegedly nonfactual nature of evolution. The Arkansas bill stated as one of its Legislative findings of fact that:
Evolution-science is not an unquestionable fact of science, because evolution cannot be experimentally observed, fully verified, or logically falsified, and because evolution-science is not accepted by some scientists.
Michael Ruse was the well known philosopher of science called upon to testify during the trial that eventually rendered the Arkansas bill unconstitutional. In his 1982 Darwinism Defended he tried to be charitable, but before too many pages he was pounding on his keyboard, shrieking that "Evolution is a fact, fact, FACT!"1
The analogous 1986 Louisiana balanced treatment bill required that both "creation-science" and "evolution-science" be taught as theory rather than "proven scientific fact." In response, the "Friends of the Court" brief drawn up in the name of 72 Nobel Laureates prior to the 1987 Supreme Court decision pooh-poohed this demand by claiming that:
To a scientist or a science educator, the distinction between scientific theories and scientific facts is well understood. A "fact" is a property of a natural phenomenon. A "theory" is a naturalistic explanation for a body of facts.2
Similar comments followed a few pages later:
Every scientific discipline embraces a body of facts and one or more theories to explain them. Significantly for this case, scientific facts and theories are not interchangeable: an explanatory principle is not to be confused with the data it seeks to explain. This relationship between scientific theory and fact permeates all scientific disciplines; it unifies the enterprise of all scientists, from astronomers to zoologists.3
Although the Supreme Court ruled in favor of the scientific establishment by declaring the Louisiana bill unconstitutional, Justice Scalia (with the approval of Rehnquist) dissented. "The people of Louisiana," he wrote, "... are quite entitled, as a secular matter, to have whatever scientific evidence there may be against evolution presented in their schools, just as Mr. Scopes was entitled to present whatever scientific evidence there was for it."4
At the heart of the dissent lay Scalia's bluntly expressed view that reasonable persons could disagree over the merits of evolution:
Infinitely less can we say (or should we say) that the scientific evidence for evolution is so conclusive tht no one would be gullable enough to believe that there is any real scientific evidence to the contrary....5
In other words, Scalia argued, evolution was open to rational dispute. It could not claim the status of a "fact" beyond all scientific disagreement.
Stephen Jay Gould has vigorously vented his frustration with this attitude. In his 1981 "Evolution as Fact and Theory" he wrote that:
Well, evolution is a theory. It is also a fact. And facts and theories are different things, not rungs in a hierarchy of increasing certainty. Facts are the world's data. Theories are structures of ideas that explain and interpret facts. Facts do not go away while scientists debate rival theories for explaining them. Einstein's theory of gravitation replaced Newton's, but apples did not suspend themselves in mid-air pending the outcome. And human beings evolved from apelike ancestors whether they did so by Darwin's proposed mechanism or by some other, yet to be discovered.6
On this account, Darwin, instead of writing that the Origin was "one long argument," should have described it as "two long arguments": one for the fact of evolution and one for Darwin's explanatory theory. Gould likes to quote a famous passage from Darwin's The Descent of Man to demonstrate that Darwin fully understood the distinction between the evolutionary fact he called "descent with modification" and his explanatory theory primarily based upon natural selection. But this distinction certainly does not stand out in Desmond and Moore's recent marvelous biography of Darwin. Their description of Thomas Huxley's disagreements with Darwin never crystallize around this issue, at least on my reading.
Be this as it may, the evolution of humans from apelike ancestors clearly is not a "fact" of the same simplicity as a falling apple, and Gould's analogy is unconvincing for that reason. For example, according to a November 1989 California Board of Education committee decision, the state's guideline for scientific textbooks does not refer to evolution as a "scientific fact;" the appropriate term is thought to be "theory." More recently, Philip Johnson in his Darwin on Trial had the following comment on Gould's analogy between falling apples and evolving humans:
The analogy is spurious. We observe directly that apples fall when dropped, but we do not observe a common ancestor for modern apes and humans. What we do observe is that apes and humans are physically and biochemically more like each other than they are like rabbits, snakes or trees. The ape-like common ancestor is a hypothesis in a theory, which purports to explain how these greater and lessor similarities came about.
... The "fact" that Gould describes is therefore nothing but Darwin's theory rightly understood: evolution is descent with modification propelled by random genetic changes, with natural selection providing whatever guidance is needed to produce complex adaptive structures like wings and eyes.7
Johnson clearly exaggerates the degree to which theoretical assumptions infiltrate the statement that evolution is a "fact;" he wants to convince his readers not only that the term "evolution" includes a dogmatic commitment to the power of natural selection, but also that in educational practice it means an allegiance to philosophical materialism and atheism. But without buying into Johnson's full agenda, we can at least acknowledge that he does properly call attention to the intriguing ambiguity of the phrase "descent with modification." How much theoretical baggage was or is smuggled in under Darwin's seemingly innocent phrase?
At this point it is appropriate to ask whether there are some philosophical distinctions at hand to help clarify matters. With due regard for Michael Ruse's rash reliance on falsificationism during the Arkansas balanced treatment trial, it seems that at least one philosophical contribution would be useful: the distinction between data and phenomena.
In several recent publications, James Bogen and James Woodward have pointed out the detrimental effects of relying upon the traditional philosophical dichotomy of theory and observation.8 They advocate a distinction between "data" and "phenomena" based upon two distinct phases in scientific reasoning. Bluntly stated, data provide evidence for the phenomena that theories explain. Data are highly idiosyncratic with respect to experimental equipment, materials and conditions, factors which are not addressed by theoretical formalism. Furthermore, attempts to replicate experimental procedures result in data with values that vary over an appreciable range; the complexity of the causal factors in any interesting experiment make it impossible to explain the exact data resulting from any specific trial. Experimental data, with all their complexity, thus are never actually explained by theory. Instead, data provide evidence for the existence of phenomena. Phenomena are inferred from data and have a stability that can be detected by means of a variety of different types of data. Both data and phenomena can be accepted as scientific facts.
Nevertheless, phenomena are far more complex facts than data are. While data are the more or less direct result of experimental observation and measurement, phenomena must be inferred from these data and cannot be straightforwardly observed.
Some relevant examples would be the phenomena described by Kepler's laws of planetary motion. That planets sweep out equal areas in equal periods of time in elliptic solar orbits for which the square of the period is proportional to the cube of the average radius are phenomena that Kepler managed to infer from extensive planetary data. Although these phenomena are certainly recognized as scientific "facts", they cannot be directly observed. Similarly, Galileo's law of falling bodies [d= ½gt2] is a phenomenological law, a phenomenon Galileo inferred from relevant data: his measurements of height and time for balls rolling down inclined planes. Both Kepler's and Galileo's phenomena were initially explained by the application of Newton's laws of motion and his law of gravitation to an idealized representation of the relevant masses.
On the other hand, to note a less successful example, in a celebrated 1989 case, the alleged phenomenon of "cold fusion" was found to be nonexistent in spite of a considerable accumulation of data. Although Pons and Fleishman claimed to have evidence in support of the phenomenon of nuclear fusion at room temperature, their claim was subsequently rejected due to the assessment of their data as idiosyncratic artifacts of experimental equipment. Some of the data continues to be of interest, but by the end of 1989 efforts to give theoretical explanations for the phenomenon of cold fusion had been judged premature.
With these examples in mind, the Bogen/Woodward distinction at least implicitly sets up a three-level explanatory and evidential hierarchy: data, phenomena and explanatory theory. Now, granted that these distinctions are valuable in the physical sciences, the biological sciences are notoriously hierarchical in ways that make the simple triad of data, phenomena, and theory look immediately simplistic. Nevertheless, I would argue that it can at least improve upon the present level of the discussion of evolution as "fact or theory." Gould's discussion is unconvincing because he relies on an analogy between evolution and gravity that essentially involves a category mistake. The "fact" that is explained by Newtonian theory is Galileo's phenomenological law of falling bodies, a phenomenon inferred from data accumulated through measurements of rolling balls, if not falling apples. I will argue that Gould's analogy should be of the form: gravitational phenomena are to Newton's theory of gravitation as evolutionary phenomena are to Darwin's theory of evolution.
Although as a first approximation it is tempting to propose that the "fact" of evolution is a complex fact in the sense that it is a phenomenon inferred from simpler data, it is not clear how this alleged phenomenon should be characterized. In Of Pandas and People, a controversial high school textbook, intelligent design advocates Mark Hartwig and Stephen Meyer follow Keith Thomson's argument that the term "evolution" can be used to refer to at least three distinct processes: biological change over time, descent with modification, and descent modified exclusively by random variation and natural selection.9
If it is insisted that the term "evolution" refer to a single fact, I would argue that Darwin's phrase "descent with modification" is the appropriate characterization of the phenomenon of evolution. To assert that this phenomenon has taken place is to assert that life has developed over time with the presently existing species being the surviving descendants of earlier ones. Creationists have appropriately pointed out that Darwin's phrase implies much more than the relatively innocuous claim that life at present is different from what it was in the past. This claim is consistent with the view that all species were created at one point in time with some going extinct subsequently. Furthermore, Darwin's claim that present species are descendants of earlier ones can be contrasted to those of other evolutionary theorists such as Lamarck, who had quite different conceptions of the phenomenon of biological change. Lamarck believed in multiple and parallel "chains of being" all progressing to their goal in the human species. Given Darwin's disagreement with Lamarck about the phenomenon of evolution, it is not surprising that the two men proposed radically differing explanatory theories. Darwin's claim that life has experienced "descent with modification" thus is an assertion quite distinct from the relatively bland claim that the set of existing species has changed over time.
On the other hand, Richard Dawkins's characterization of modified descent generated exclusively by random mutation and natural selection as a "fact," what Phillip Johnson calls the "blind watchmaker's thesis," inappropriately blurs the distinction between phenomenon and theory. Dawkins fails to distinguish the phenomenon to be explained from the explanatory theory of adaptationism. Similarly, it is misleading to use the term "evolution" to refer to the observed consequences of natural selection acting on moth coloration or the dimensions of finch beaks. These cases are clear demonstrations of the phenomenon of natural selection that are more aptly called "microevolution."
One last point should put us in a position to improve significantly on Gould's distinction between evolution as fact and evolution as theory. There obviously is a far more varied set of phenomena considered to be appropriate for explanation by evolutionary biology than the simple claim that life has evolved through descent with modification. For example, why and how did horses gradually evolve hoofs to replace their earlier toe structure? The alleged single "phenomenon" of evolution thus might more accurately be stipulated as the phenomena of evolution, or "evolutionary phenomena," but in full recognition of the relevant distinction between data and phenomena. Gould's analogy of Newton's theory of gravitation being to falling apples as modern evolutionary theory is to the descent of humans from apelike ancestors can thus be made more accurate. That is, modern evolutionary theory is to evolutionary phenomena as Newton's theory of gravitation is to gravitational phenomena.
An important pedagogical lesson to glean from all this is that a great deal of interesting scientific debate pertains to the establishment of the phenomena that need to be explained. Kepler's laws are complex phenomena that were accepted as "fact" by very few astronomers during the first half of the seventeenth century. They are extremely difficult to infer from the intricate data of planetary observations. Similarly, it is an exciting challenge to infer evolutionary phenomena from the relevant data. The difficult and controversial nature of the task is a direct consequence of the complexity of the data studied by paleontology and molecular biology. For example, it could be argued that individual fossils are phenomena in their own right. That is, aren't they complex results of inference based on the relatively simple data produced by radioactive dating techniques and anatomical measurements? Even if this suggestion is rejected, it should be clear that educators face a difficult challenge when confronted by the daunting ambiguity of the term "evolution." It would be a disservice to students to employ the word in a manner that suggests that it always refers straightforwardly to a single phenomenon unanimously recognized by the scientific community. While no one disagrees about the factual nature of Kepler's laws, some alleged evolutionary phenomena are generally agreed upon to be facts while others are not. It remains to be seen if the data/phenomenon distinction can help to rectify the present confusion in the contexts of education and law.
Copyright © 1996 James R. Hofmann. All
rights reserved. International copyright secured.
File Date: 11.14.96