Wendel, P. (2007, June). Falsifiability as a science/non-science demarcation criterion in the battle against creationism. Paper presented at the Ninth International History, Philosophy, and Science Teaching Conference, University of Calgary, Alberta, Canada.
Wendel (2007 began his analysis with a working definition of the term creationist as “anyone who endorses the theistic creation of the universe and/or life within the universe” (p. 1), and excluding those who accept evolutionary science, but still believe in a theistic cause for the beginning of life and the universe. Wendel further refined his definition by sub-dividing creationists into old-earth creationists—those that accept geological timescales for the origin of the universe and formation of the earth—and young earth creationists—those that believe in the biblical six days of creation and that the earth is approximately 10,000 years old; thereby managing to reject physics, geology and biology in one anti-scientific leap.
Other researchers have adopted similar definitions, for example Superfine (2007) referred to creationism as a doctrine or theory holding that life and the world were created by a supernatural actor or deity out of nothing (Superfine, 2009). Superfine’s definition draws attention to the supernatural actor, as opposed to any less specific theistic notion. It is upon definitions of the terms used by Wendel, their explanation and interpretation that I shall focus this paper. I intend to show that Wendel, in not fully exploring the scientific definitions of terms such as testability and falsifiability, discredits such articles of the scientific armory in the Battle Against Creationism to an unjustified extent.
Wendel (2007) framed the debate between proponents of creationism and evolutionary science in terms of a boundary dispute: “creationists need intellectual validity… the best validation of all would be recognition as science… [s]cientists, science educators, and others respond by attempting to deny creationism the status of science and the authority that comes with that status.” (p. 1). This has resulted in a sustained (Gieryn, 1983; Gieryn et. al., 1985), and—in my view—quite proper, demarcation between evolutionary biology representing science and creationism as a system of faith.
Wendel (2007), briefly summarised some of the major legal cases, where the teaching of evolutionary science in United States science curricula, has been challenged by creationism. Wendel discussed five major cases: Edwards v. Aguillard (1987); Epperson v. Arkansas (1968); Kitzmiller v. Dover (2005); McLean v. Arkansas (1982); and Scopes v. Tennessee (1927). Whilst Wendel admitted that this was by no means an exhaustive list, organisations involved in the documentation, monitoring and often expert consultation in such cases (for example, the National Center for Science Education; NCSE), usually add six further judgements to the list (NCSE, 2011): Freiler v. Tangipahoa (2000); LeVake v. Independent School District #656 (2002); Peloza v. Capistrano (1994); Seagraves v. California (1981); Selman v. Cobb (2005); and Webster v. New Lennox (1990). Of these cases, only two have included any discussion by the ruling judge on the nature of creationism and science.
In attempting to gain validation from scientists, creationists have attempted to fit hypotheses to theistic doctrine, couching their arguments in the terminology of science and labelling them creation science or intelligent design:
“Creation-science” means the scientific evidences for creation and inferences from those scientific evidences. Creation-science includes the scientific evidences and related inferences that indicate: (1) Sudden creation of the universe, energy, and life from nothing; (2) The insufficiency of mutation and natural selection in bringing about development of all living kinds from a single organism; (3) Changes only within fixed limits of originally created kinds of plants and animals; (4) Separate ancestry for man and apes; (5) Explanation of the earth’s geology by catastrophism, including the occurrence of a worldwide flood; and (6) A relatively recent inception of the earth and living kinds (McLean v. Arkansas, 1982, p. 1264).
I would argue that, by placing creationism within a manufactured framework of science, which may be referred to as pseudoscience, creationists have also attempted to circumvent the First Amendment to the United States Constitution (the Establishment Clause), which forbids the establishment of a national religion and/or the advancement of any one religious doctrine over another in American public schools (U.S. Const. amend. I, 1791).
The original intention of the religious clauses of the First Amendment, can be interpreted as to protect the right of the individual to worship without fear of government intervention. Only later case precedent has extended that intent to include secularism and education policy. This is exemplified by the three prongs of the Lemon Test, established in the case of Lemon v. Kurtzman (1971). The Supreme Court ruled that the Non-Public Elementary and Secondary Education Act 1968 of Pennsylvania, which allowed the state to pay faith-based, non-publically-funded schools, for salaries and resources used to teach secular material, breached the First Amendment. The three prongs of the Lemon Test have since been used as an effective assessment of the violation of the First Amendment, where federal law is deemed to be breached when:
- The government’s action must have a secular legislative purpose;
- The government’s action must not have the primary effect of either advancing or inhibiting religion;
- The government’s action must not result in an excessive government entanglement with religion.
Originally covering only acts of Congress, a Supreme Court ruling (Everson v. Board of Education, 1947) set the precedent for applying the First Amendment across all states. Later, the Supreme Court was to state that the core intent of the First Amendment was to not favour one religious faith over another or over a lack of faith (Board of Education v. Grumet, et al., 1994).
Of all the legal cases since Scopes v. Tennessee (1927), only two have included judgement which has discussed the nature of creationism and its relation to science (McLean v. Arkansas, 1982; Kitzmiller v. Dover, 2005). All other cases have failed at the hurdle of one or more prongs of the Lemon Test. Thus making the avoidance of appearing theistic in nature, a valuable strategy in attempts to include creationism in U.S. public school science curricula. In the two cases where the nature of creation science or intelligent design (ID) has been ruled upon as an unscientific alternative to evolution, courts have rejected the argument as having little or no scientific merit. In McLean v. Arkansas (1982), U.S. District Court Judge William Overton declared that creation science had “no scientific merit or educational value as science” (p. 1272). However, as a federal court case, the ruling was not technically binding outside the court’s district.
Only one court ruling has, to date, included testimony and opinion on the nature of creationist arguments and their relation to science—Kitzmiller v. Dover (2005). In this case, Judge John E. Jones III, a conservative Republican appointed by George W. Bush, delivered a 139 page judgement in favor of the plaintiffs. His ruling was based partly on the grounds that the board’s decision was motivated by religious views of a number of members, noting that several board members “would time and again lie to cover their tracks and disguise the real purpose behind the ID Policy” (Kitzmiller v. Dover, 2005, p. 137). However, the most important part of Judge Jones’ ruling, was his finding that intelligent design was inherently theistic:
[D]efense experts concede that ID is not a theory as that term is defined by the NAS [National Academy of Sciences] and admit that ID is at best “fringe science” which has achieved no acceptance in the scientific community. (p. 70)
[W]e find that ID is not science and cannot be adjudged a valid, accepted scientific theory… ID, as noted, is grounded in theology, not science. Accepting for the sake of argument… that to introduce ID to students will encourage critical thinking, it still has utterly no place in a science curriculum. (p. 89)
In addition, as Wendel (2007) also noted, Judge Jones made mention of the opinion of Judge Overton in McLean v. Arkansas (1982), who criticized creationist proponents for assuming that any argument against evolution was an argument for creationism. Jones observed that advocates of ID employed the same contrived dualism during Kitzmiller v. Dover.
Wendel’s (2007) assertion that an important part of the failure of creationism, to be taught in U.S. public school science curricula, is its failure to “find acceptance as science in court” (p. 7), is at least an over-simplification. Only two cases have seriously considered that argument, one of those was a federal case, and therefore did not set precedent outside of its jurisdiction, and the other was one of the most recent cases. Historically, judicial opinion has not had to consider whether or not creationism is science, if it relied solely on the three prongs of the Lemon Test.
Scientific theories, as opposed to hypotheses, are based upon experience of the natural world—systematic observation; analysis; consistent results of well-designed, controlled, double-blinded—where applicable—experiment; they are tentative, that is, they cannot be cannot be proven absolutely, but with every positive observation, the validity of the theory increases (Ben-Ari, 2005; Kuhn, 1996; Popper, 1959). Scientific theory has the ability to predict further observations and is falsifiable (Popper, 1959). That is, if the theory is incorrect, just one contradictory observation will show it to be false, for example, J. B. S. Haldane’s now much over-used, but equally succinct response to the question of what could falsify the theory of evolution: “Fossil rabbits in the Precambrian!” (Dawkins, 2009, p. 147). Meaning that the discovery of evidence of a mammal existing in an era long before evolutionary science showed that even vertebrates were present, would render the theory to be false. It is a process of methodological naturalism (Hume, 2000), the product of which is open to peers—peer reviewed—and must be accurately repeated with consistent results—replicable—before a consensus can be reached and a hypothesis declared a scientific theory.
As Wendel (2007) rightly points out, the terms testable and falsifiable are not identical. For example, it may be possible to test a hypothesis, and return with a level of probability that does not completely guarantee that hypothesis to be false—the result is tentative. Testability is therefore a pre-requisite of falsifiability, but not vice versa. A thousand carefully crafted trials of prayer to cure cancer, may indicate the probability of its success as an intervention, but one single positive or negative result could not falsify a null hypothesis. In contrast, more than one hundred and fifty years of observation and successful experiment have not completely guaranteed the scientific theory of evolution by natural selection, but one fossilized, precambrian rabbit, would show it to be false. I would therefore contend that, given the nature of probability, the scientific definition of testability implies an inherent level of predictability from the results. Wendel (2007) does not discuss this.
Wendel (2007) referred to string theory—a highly complex aspect of theoretical physics—as an example of science which, he argued, was untestable. I would contend that the use of the term untestable is erroneous. String theory, dark matter, the possible existence of the Higgs–boson particle and its mass, plus other rarified aspects of physics combine to allow for a level of predictability upon which is based the Standard Model of Particle Physics. This level of predictability has, thus far, been confirmed by experiment and advanced calculations of the probability of the mass of certain particles that have, as yet, to be observed (Oerter, 2006). The Standard Model has so far remained unfalsified, should it be falsified, it will be amended according to the evidence available. However, the evidence required for rejecting the foundation of modern physics, would have to be extraordinary indeed. As succinctly phrased by the physicist Carl Sagan (1934–1996), “[e]xtraordinary claims require extraordinary evidence” (Sagan, 2006, p. 47).
Wendel (2007), went on to state that “[t]he first serious problem with relying on testability, falsifiability, or tentativeness to disqualify creationism as science is that creationists test, falsify, and correct themselves” (p. 8). To support this argument, Wendel cites criticisms by the creationist community of two young earth creationist (YEC) scientists. The first was Robert Gentry—who at the time of the cited research, held a master’s degree in physics (and an honorary doctorate from the fundamentalist Columbia Union College). For nearly two decades he held a research associate’s position at the Oak Ridge National Laboratory, where he had been studying the nature of very small discoloration features in mica and other minerals, and had concluded that they were proof of a young earth (Gentry, 1968, 1970, 1971, 1974, 1992; Gentry, et al., 1973). Wendel referred to this work as Gentry’s “careful studies of radiohalos in granite”, and cites criticism of this work, by creationists, as examples of self-regulation amongst the community. In fact, Gentry’s work was anything but “careful” and was viewed as deeply flawed by the scientific community (Baillieul, 2010; Brawley, 1992):
Gentry’s polonium halo hypothesis for a young earth fails, or is inconclusive for, all tests. His samples are not from “primordial” pieces of the earth’s original crust, but from rocks which have been extensively reworked. He is unable to demonstrate that concentric halos in mica are caused uniquely by alpha particles resulting from the decay of polonium isotopes. Finally, his hypothesis cannot contend with the many alternative lines of evidence that demonstrate a great age for the earth. In the end, Gentry’s young-earth proposal, based on years of measuring discoloration halos, fails to generate a scientific model that is either internally consistent or consistent with generally accepted scientific understanding of geophysical processes and earth history. (Baillieul, 2010, p. 26).
I would contend that, faced with such a poor example of science, and such damning criticism, it would have been unlikely that the creationist community would have argued in favour of the work.
It should also be noted that, whilst Gentry’s work was published in well-respected, mainstream, peer-reviewed scientific journals. At no point in any of these articles, did Gentry formulate a hypothesis, or suggest conclusions, regarding a young earth of 10,000 years or less. As Baillieul (2010) pointed out:
The closest he ever came to this type of statement is found in a cryptic question posed at the end of a 1974 article in Science (Gentry 1974): “… can they [polonium halos] be explained by presently accepted cosmological and geological concepts relating to the origin and development of the earth?” Based on the above analysis, the answer is a resounding yes! (p. 26).
The second major young earth creationist (YEC) author cited by Wendel (2007), was Dr. D. Russell Humpreys, a nuclear physicist who, from 2001–2008, was an associate professor at The Institute for Creation Research. In brief, Humphreys’ cosmological model proposed using a relativistic approach, postulating that time would have run at different rates during the universe’s origin, resulting in an estimated age of the earth of 6,000 years (Humphreys, 1994, 1997, 1998, 2000). As with Gentry’s work, Humphreys’ proposed model was viewed as critically flawed by the scientific community. Kevin R. Henke, Ph.D., assistant professor of geology at the Center for Applied Energy Research, University of Kentucky, wrote:
Dr. Humphreys’ work is a prime example of fallacious reasoning that YECs (e.g., Woodmorappe, 1999) falsely accuse geochronologists of using. Because his bogus calculations and inaccurate data just happened to spit out a meaningless number that he likes (6,000), Dr. Humphreys is more than willing to ignore and inappropriately dismiss any data or criticisms that expose the fraudulent nature of his “creation date.” (Henke, 2006, Appendix C).
The final sentence of this quotation belies the argument made by Wendel (2007), that the criticism of the work of Gentry and Humphreys by the creationist community, “points to an internal process by which creationists debate, test, and modify their conclusions based on physical evidence as well as theoretical considerations.” (p. 9). Creationists claim that, given their arguments, which they may base in testable, falsifiable formats, there must have been an original theistic causation. Whilst the arguments are testable and have been shown to fall at every hurdle of scientific explanation, the claim of a theistic cause is not. I would contend that the creationist argument is ultimately untestable and not falsifiable, because experimental protocol cannot control for a final conclusion which is, in essence a solution cannot be conceived of, to a puzzle that has been invented, by fitting a hypothesis to religious beliefs; therefore the answer must be theistic in nature. As Superfine (2007) has stated:
Although the major proponents of ID often frame the theory as a legitimate scientific alternative to evolutionary theory, the historical origins of ID are firmly rooted in the fundamentalist cultural and religious movement that has supported the teaching of biblical creationism in schools. (p. 1).
Therefore, testability does not altogether fail as a “singular demarcation criterion”, as Wendel (2007, p. 11) concludes. And the author’s argument of “declarations that creation science is not science on testability/falsifiability grounds… do not withstand scrutiny” (p. 12), is itself based on faulty and incomplete reasoning.
Lastly, Wendel (2007) agreed with the sentiment that phrases such as “pseudo-science” and “unscientific” should be dropped from our vocabulary, as they only stir negative emotions and may prove counterproductive (p. 16). Once again I would disagree. Creationism and the terms creation science and intelligent design, have been deliberately and cynically misrepresented in order to further the agenda of campaigners for curriculum change—as in the cases of Kitzmiller v. Dover (2005; see above p. 4). The reasoning behind the arguments for intelligent design has been shown to be specious and manufactured to fit theistic belief. I would argue that this is clearly not science, despite exhaustive attempts to make it appear so.