Misuse of probability by “creation scientists”

It often comes as a shock to professional scientists to learn that a large fraction of the public rejects much if not all of the evolutionary framework of modern geology and biology. For example, in a recent poll, 44% of Americans surveyed agreed that God created human beings pretty much in their present form at one time within the last 10,000 years [Gallup]. Another indication of the popularity of this worldview, often termed “young-earth creationism”, is the fact that over 700,000 Americans have attended the “Creation Museum” near Cincinnati, Ohio since its opening in 1977. Displays at the museum insist the world was created in the past 10,000 years, and depict, for instance, dinosaurs co-existing with prehistoric humans.

A related development is the recent emergence of the “intelligent design” (ID) movement, which generally acknowledges the old-earth framework, but still insists that evolution can only produce minor changes within established “kinds” (species), which “kinds” were individually created by an intelligent Designer. Both movements have a dedicated cadre of writers, including at least some with respectable academic credentials, who generate books, articles and Internet posts criticizing conventional scientific research and asserting that scientific evidence confirms their point of view.

Both traditional creationists and ID scholars have invoked probability theory in criticisms of evolution. One typical argument goes like this: the human alpha globin molecule, which plays a key oxygen transfer function, is a protein chain based on a sequence of 141 amino acids. There are 20 different amino acids common in living systems, so the number of potential chains of length 141 is 20^(141), which is roughly 10^(183). This figure is so enormous, so these writers argue, that even after billions of years of random molecular trials, no alpha globin protein molecule would ever appear [Foster, pg. 79-83; Hoyle, pg. 1-20; Lennox, pg. 163-173].

But the above argument fails to note that most of the 141 amino acids can be changed without altering the key oxygen transfer function. When we revise the calculation above, based on only 25 locations essential for the oxygen transport function, we obtain 10^(33) fundamentally different chains, a huge figure but vastly smaller than 10^(183), and small enough to neutralize the probability-based argument against evolution [Bailey].

More importantly, this and almost all similar probability-based arguments against evolution suffer from the fallacy of presuming that biological structures such as alpha globin arise by a single all-or-nothing random trial. Instead, available evidence suggests that alpha globin and other proteins arose as the end product of a long sequence of intermediate steps, each of which was biologically useful in an earlier context. Probability calculations such as the above, which do not take into account the process by which the structure came to be, are not meaningful and can easily mislead [Musgrave].

Along this line, consider snowflakes. Bentley and Humphrey’s book Snow Crystals [Bentley] includes over 2000 high-resolution black-and-white photos of real snowflakes, each with intricate yet highly regular patterns (a few of the Bentley-Humphrey photos are posted at Online article). The chances that one particular structure, with striking near-perfect 6-way symmetry, can form “at random” can be calculated as roughly one part in 10^(2500). Does this astoundingly small probability figure constitute proof that individual snowflakes have been intelligently designed? Obviously not. The fallacy, once again, is presuming a sudden, all-at-once random formation. Instead, snowflakes, like biological organisms, are formed as the product of a series of steps, acting under natural laws with some element of chance.

ID scholar William Dembski invokes probability and information theory (the mathematical theory of information content in data) in arguments against Darwinism. But knowledgeable scholars who have examined Dembski’s works are not persuaded and have been sharply critical. Mathematician Jeffrey Shallit (a colleague of the present bloggers) and biologist Wesley Elsberry conclude that Dembski’s notion of “complex specified information” is incoherent and unworkable [Shallit]. Biologist Gert Korthof, in a review of Dembski’s book Intelligent Design, concludes that Dembski’s analysis cannot be meaningfully applied to DNA [Korthof]. Mathematician Richard Wein, in a review of Dembski’s book No Free Lunch, characterizes it as “pseudoscientific rhetoric” [Wein].

One central issue in this debate is the question of evolutionary novelty. The consensus of modern scientific research is that mutation and natural selection together can produce novel, beneficial features in biological systems. Scientists further postulate that this low-level novelty extends to entire populations, which can, over time (typically thousands of years), become entirely separate species. On the other hand, creationist and ID scholars have insisted that whereas minor changes may occur within an established kind, nothing fundamentally new can come through “random” evolution. For example, Dembski asserts that there is a “Law of Conservation of Information” that prohibits the generation of novel features [Dembski].

Ample and well-established experimental evidence supports the scientific view. For example, in a 1974 paper by biologists Barry Hall and Daniel Hartl, a gene was identified in the bacterium E. coli that is responsible for metabolizing lactose, using a complicated three-part process. They removed this gene, and then permitted the bacteria to multiply in a stressed environment containing lactose. Within 24 hours the bacteria had evolved a capability to utilize lactose, by means of a similar but distinct three-part biochemical pathway, involving two mutated genes [Hall; Miller, 1999, pg. 145-147].

In another interesting result along this line, Japanese biologists recently discovered a bacterial species that has adapted to thrive on nylon waste (which did not exist until the 20th century). It turns out that this bacterial species has undergone a “frame shift” mutation, where an extra base pair has been inserted into the bacteria’s DNA. This mutation significantly changed the bacteria’s biology, since a long series of amino acids were altered, but by remarkable chance this alteration endowed the bacteria with the facility to metabolize nylon, albeit not very efficiently [Negoro].

As a third example, scientists recently discovered that certain persons in an Italian community, all descended from a single individual several generations back, possess a genetic mutation that increases “good” cholesterol and provides an effective anti-oxidant, thus resulting in measurably improved cardiovascular health [Krotz]. Dozens of other examples could be cited.

In short, the probability arguments used by the creationist and ID movements, when analyzed carefully, are fallacious, and are simply countered by the observation that natural evolution, operating in the real world, does in fact produce novel features.

It is truly unfortunate that fundamentalist adherents of some of the world’s great religious movements feel it necessary to “prove” God by means of fallacious mathematical arguments. It is also unfortunate that the creationist and ID communities have been so stubborn to accept the overwhelming consensus of modern science, namely that the world is governed by elegant and comprehensible physical laws. But the scientific and mathematical communities are also at fault in failing to better educate the public as to both the reality of evolution and the failings of creationist/ID scholarship.


  1. [Bailey] David H. Bailey, “Evolution and Probability,” Report of the National Center for Science Education, vol. 20 (2000), no. 4, also available from Online article.
  2. [Bentley] W. A. Bentley and W. J. Humphreys, Snow Crystals, Dover Publications, New York, 1962.
  3. [Dembski]William A. Dembski, Intelligent Design: The Bridge Between Science and Theology, InterVarsity Press, Downers Grove, IL, 1999.
  4. [Foster]David Foster, The Philosophical Scientists, Marboro Books, New York, 1991.
  5. [Gallup] Gallup Poll, 8-11 May 2008, available at Online article.
  6. [Hall] Barry G. Hall and Daniel L. Hartl, “Regulation of Newly Evolved Enzymes,” Genetics, vol. 76 (1974), pg. 391-400.
  7. [Hoyle] Fred Hoyle and Chandra Wickramasinghe, Evolution from Space: A Theory of Cosmic Creationism, J. M. Dent and Sons, London, 1981.
  8. [Korthof] Gert Korthof, “On the Origin of Information by Means of Intelligent Design: A Review of William Dembski’s Intelligent Design,” available at Online article.
  9. [Kotz] Dan Krotz, “The Milano Mutation: A Rare Protein Mutation Offers New Hope for Heart Disease Patients,” available at Online article.
  10. [Lennox ]John C. Lennox, God’s Undertaker: Has Science Buried God?, Lion UK, 2009.
  11. [Miller] Kenneth Miller, Finding Darwin’s God: A Scientist’s Search for Common Ground Between God and Evolution, HarperCollins, New York, 1999.
  12. [Musgrave] Ian Musgrave, “Lies, Damned Lies, Statistics, and Probability of Abiogenesis Calculations,” 1998, available at Online article.
  13. [Negoro] S. Negoro, K. Kato, K. Fujiyama and H. Okada, “The Nylon Oligomer Biodegradation System of Flavobacterium and Pseudomonas,” Biodegradation, vol. 5 (1994), pg. 185-194.
  14. [Shallit] Jeffrey Shallit and Wesley Elsberry, “Playing Games with Probability: Dembski’s Complex Specified Information,” in Matt Young and Taner Edis, ed., Why Intelligent Design Fails: A Scientific Critique of the New Creationism, Rutgers University Press, Piscataway, NJ, 2004, pg. 121-138.
  15. [Wein] Richard Wein, “Not a Free Lunch But a Box of Chocolates: A Critique of William Dembski’s Book No Free Lunch,” available at Online article.

Comments are closed.