Tuesday, December 27, 2011

The Scientific Verdict on God (Part 1): Models

The sciences do not try to explain, they hardly even try to interpret, they mainly make models. By a model is meant a mathematical construct which, with the addition of certain verbal interpretations, describes observed phenomena. The justification of such a mathematical construct is solely and precisely that it is expected to work.” ~ John von Neumann [1]

One of the greatest misperceptions about science entertained among the general public today is that science is a noun. In other words, most people tend to erroneously think of science as a monolithic thing whose primary function is to manufacture proofs, instead of looking at science as it really is: a model that describes reality and how to build it. The viability of a given model is based entirely on (1) how reliable it is, (2) how usable it is, and (3) whether or not it is consistently reproved and continually open to revision by the evidence we have found or may yet find.

It is important to recognize the word “model” in science-based discussions, because many people very often use the word “theory” (which is itself a much misunderstood word) when they should be using the word “model.” The practice of science, at its most basic, involves making observations of the physical world. The set of basic assumptions that all scientists start out with must always be based on observations. Consider, for example, the concept of time. A scientific investigator is certainly going to start with time when trying to describe almost anything. But how does one go about defining “time”? Not even philosophers have been able to arrive at a consensus on how time should be defined. The definition that Einstein came up with is that time is what one reads on a clock. And clocks are a human invention, allowing us humans to define “time” in terms of whatever basic units we wish and agree upon. The basic unit of time, the second, was redefined in 1967 by international agreement as the amount of time required for a caesium-133 atom to undergo 9,192,631,770 vibrations [2]. Prior to 1967, the second was defined as 1/86,400 of a mean solar day. The new definition was chosen simply for the sake of convenience, as it more fully accounted for irregularities in the rotation of the Earth and thus allowed for comparatively simpler equations. In 1997, the time standard was further refined to specify that the caesium atom used to define time was to be one at rest at absolute zero.

In conventional physics, all the other various observational qualities follow on the heels of this pragmatic approach: Distance is what you read with a meter stick, a meter being currently defined by international agreement as the distance light traverses between two points in a vacuum during 1/299,792,458 of a second. Temperature is what you read on a thermometer.

These are known as operational definitions, the establishment of which marks the first step in the practice of science, especially physics. Ideally, the operational definitions physicists make are based on specifically-prescribed measuring procedures and informed by empirical observations, and only then do physicists proceed to carry out those measurements. Scientists are highly concerned with making their measurements quantitative. If scientists can be quantitative in their approach, much of their work is complete, thanks to the seemingly elusive quality of precision that is thereby achieved. The quantitatively-oriented physicist is then ready to build models to describe her observations. If the models work, then they are useful. If they do not work, they are not useful. As long as a given model is useful, it does not matter whether it has any correspondence with ultimate reality. In other words, whether or not an electron actually exists in reality, one can still use the electron model to calculate to a high degree of accuracy the current flows in electronic circuits. Never does metaphysics enter the issue at all. As particle physicist Victor Stenger explains,

[T]he true reality of the universe is not necessarily composed of objects that possess attributes such as position and mass which we assign them in the process of doing physics. These variables, after all, are human inventions with no precisely definable meaning beyond their measurements as performed with specific apparatuses such as clocks and meter sticks.

Describing nature in terms of physical variables is like sketching or photographing an object. Isn’t it rather foolish to equate images on a piece of paper with the real thing? Confusing an image with reality is a common characteristic of small children [3].

Another illustrative example of a model is the earth’s sun. We usually instinctively regard the sun as an orb travelling across the sky. If you are a traveler journeying from east to west, you can use the sun to guide your direction; by heading in the direction of the sunset, you know you are heading generally west. Although you must correct just slightly for longitude, you can successfully use the model of the sun as an orb moving across the sky. The ancient Greeks believed that the sun was Apollo pulling a chariot across the sky. That understanding constituted the dominant metaphysics of the time. The ancient Chinese, meanwhile, understood the sun to be a golden bird flying across the sky.

Does it really matter which of these metaphysics was correct, if any? Obviously, we know today that neither one of them was true. But the instructive point here is that it does not matter what the particular metaphysics happens to be. Both the ancient Greek traveler and the ancient Chinese traveler could still use the sun as a traveling guide in the exact same way.

Scientific models work the same way. We use them to more fully grasp, and apply to, practical needs. The better the model becomes and the more universal its application grows, the closer the model comes to being called a “theory.”

A wide application is crucial to the life of any model; the model must apply not just toward small and isolated situations, but instead to many different situations. The more general a model happens to be, the more universal it becomes. And the more universal the model is, the more universal the accompanying theory therefore becomes, and the more widely it can then be applied. But throughout this whole process of maturation, what we have is still merely a pragmatic model, a human invention. For example, we know full well that the earth is not flat. But the Flat Earth Model is still useful, because it is employed whenever we construct buildings. Locally speaking, it does not matter at all that the earth is actually a sphere. In some cases, we even go out of our way to make the earth flat where it is not in order to apply the model to practical situations.

Scientific Models vs. Religious Models

By methodological necessity, scientific arguments about the sun remove intent from the hypothesis of whom or what is hauling it across the sky, if indeed it is being hauled at all. The hypothesis that the sun is a chariot flying across the sky, on the other hand, inherently involves intent. The god responsible for the sun’s movement could arise the next morning and arbitrarily decide that he is not going to haul the sun across the sky. As anyone who has read The Iliad or The Odyssey knows, the Greek gods were very uncertain beings to rely on. When working at the level of deities, science becomes an exercise in second-guessing the supernatural. You could wake up one day to find that the sun did not go across the sky. In fact, such a thing is bound to happen in this scenario; every once in a while at least, the god would decide that he just does not want to fulfill his role. Thus, the fact that the sun has never failed to appear every morning should constitute evidence that the original model of a chariot being pulled along is dead wrong. After all, does Apollo never get sick? Does he not have to take a vacation at some point? Is he going to pull the sun around forever just because he likes us humans so very much?

Science provides the ability to consistently recreate one’s worldview, because science can recreate from scratch, through observation and experimentation, the models it previously built up. The models may not be exactly the same after reconstruction; there may be something else in the place of the current quark, for example. But whatever may end up standing in its place in the reconstruction aftermath of our hypothetical loss of accumulated knowledge will have the exact same function as a quark and behave exactly the same way. However arbitrary the basis of scientific models may be, whatever it is we have decided to call “quarks” will always be the building blocks of protons and neutrons, but never of electrons. We can rename or even shuffle existing names of the six kinds of quarks we have identified (up, down, strange, charm, bottom, and top). We can decide to arrange them differently than they are now, in order from lightest to heaviest in weight. Nothing fundamental would change; all “quarks” would continue to have opposites we call antiquarks. They will always have positive charges, complemented by the negative charges of their antiquarks, no matter what we call them.

On the other hand, imagine the drastic differences that would obtain if one tried to reconstruct a religion from scratch. What emerges is never going to be the same as what we now have. The models of science change, or are dismissed entirely, in the face of new scientific discoveries. This is how good science works. However, the basic methodology underlying the practice of science has not so much changed as it has been refined over the centuries. “Science began thousands of years ago,” writes Stenger, “and although the volume of knowledge has expanded enormously in that time, the nature and methods of science have changed little. What is often interpreted as a great paradigm shift can be more accurately described as a clarification or reformulation of principles that were previously dimly perceived” [4]. For example, even in the wake of the twentieth-century revolutions in physics that saw the success of non-deterministic quantum mechanics and Einstein's theory of relativity, the three laws of motion developed by Isaac Newton continue to work as successful predictive models. They continue to be applied in virtually every facet of modern technology. In fact, our understanding of Newton's law of gravity, F=G (m_1 m_2)/r^2, is what allowed us to fly to the moon.

Religion, on the other hand, possesses its models and nothing else. There is no discovery in religion; everything must be forced to fit into the given model at hand. Religionists must wedge all outside discoveries into their existing, supposedly changeless model, but they also cannot allow such discoveries to alter the religious model too much, or else their model ceases to be definable as a religion.

A perfect example of this practice of forcing data into a given religious model is found in the work of religious scientists like William Dembski, Michael Behe and Francis Collins, each of whom are notorious for placing the proverbial cart firmly before the horse in their approach to reconciling science and religion. It is somewhat difficult at times to believe that mathematician Dembski and biochemist Behe, the main representatives of the “Intelligent Design” movement, actually believe their own words. It seems they are both knowledgeable enough, that they both have access to the same information that all other mathematicians and biochemists have managed to understand. Yet they both sit with crossed arms and staunchly deny this readily-accessible information.

To wit: During Kitzmiller v. Dover Area School District (the 2005 legal battle over the constitutionality of teaching intelligent design in public schools that was waged in Dover, Pennsylvania), Michael Behe testified under oath that he had never seen any studies or papers that provide answers to his challenge that the immune system is “irreducibly complex” and therefore could not be the product of evolution. The response from his cross-examiner was very telling and damning to Behe’s credibility. As Judge John E. Jones III noted in his decision,

Between 1996 and 2002, various studies confirmed each element of the evolutionary hypothesis explaining the origin of the immune system. [2:31 (Miller)] In fact, on cross-examination, Professor Behe was questioned concerning his 1996 claim that science would never find an evolutionary explanation for the immune system. He was presented with fifty-eight peer-reviewed publications, nine books, and several immunology textbook chapters about the evolution of the immune system [5].
Having personally read Behe’s famous book Darwin’s Black Box [6], it seems to me that Behe genuinely believed that he had hit upon something compelling, that he was not consciously lying. In this book, Behe makes several statements to the effect that no explanation, deriving from gradual evolutionary changes over time, exists for the examples of “irreducible complexity” that he lists, when in fact there were an abundance of explanations he could have easily found had he looked. Behe, who is a biochemist and not an evolutionary biologist by training, was simply unaware of the scientific literature that documents numerous robust examples in nature of organic systems undergoing functional changes during its evolution [7]. He was even unaware that the evolutionary biologist Hermann Joseph Muller, who in 1946 won the Nobel Prize for his work in biology, had already provided an evolutionary mechanism for so-called “irreducibly complex” systems six decades earlier [8].

The only thing to which we are justified in attributing this is ignorance on Behe’s part, the same sort of ignorance that religious scientist Francis Collins displays on almost every page of his bestselling book The Language of God [9]. Collins, a geneticist and proponent of a theistic framework through which to understand evolution, is simply unaware of all the scientific as well as theological work that refutes or disputes his claims. But while it may seem as though it is not possible they should be unaware of all this literature any longer, part of the blame lies on other academics in the pertinent fields, academics who do not bother to criticize the likes of Behe and Collins because of the mistaken impression that the Behes and Collins of the world are promoting their cause. More evolutionary biologists, biochemists and geneticists need to read Behe’s and Collins’ books, and they need to criticize and refute them in a mainstream capacity. They need to publicly point out that they are obviously not aware of the literature on the subject. Whereas many theistic apologists very often make arguments from ignorance, Behe and Collins seem to be making arguments from voluntary ignorance, because they have not bothered to investigate the literature that bears on their subjects.

William Dembski, on the other hand, knows his business much better than Behe. In his books and articles [10], Dembski really is knowingly cooking up claims that suit his preconceived and very dubious ideas [11]. Ultimately, the preconceived end the religious apologist wants to reach constitutes their entire motive, not a genuine regard for science and investigative honesty.

Of course, one can try to get people to imagine a wholly new religious model, to make up a religion of their own from scratch. Doing so may even result in something better than any religion we now have (then again, it is somewhat difficult to come up with something worse). When the U.S. Supreme Court ruled in early 2006 to allow the religious use of hallucinogens for a small religious group (the UniĆ£o do Vegetal, a fringe Christian sect) in New Mexico, they may have helped make that particular brand of religion much more popular than it previously was [12]. After all, prospective members knew they can legally get high if they came in to that church. But nevertheless, the point remains that the models of science, in their ability to consistently arrive at the same conclusions even after reconstruction of lost data, is demonstrated to be superior to the fickle models of religion.


1. Quoted in J. Tinsley Oden, Acceptance Remarks, 1993 John von Neumann Award Winner, United States Association of Computational Mechanics Bulletin 6, no. 3 (September 1993).

2. Bureau International des Poids et Mesures (BIPM), “Unit of time (second),” The International System of Units (SI) 8th ed., 2006, Section, pp. 112-113.

3. Victor J. Stenger, Physics and Psychics: The Search for a World Beyond the Senses (Buffalo, NY: Prometheus Books, 1990), p. 233.

4. Ibid., p. 294.

5. Kitzmiller v. Dover Area School District, Federal Case No. 4:04-cv-02688-JEJ Document 342, Judge John E. Jones III presiding, filed Dec. 20, 2005. Decision, 78.

6. Michael Behe, Darwin's Black Box: The Biochemical Challenge to Evolution (New York: Free Press, 1996).

7. See Robert Dorit, review of Darwin’s Black Box by Michael Behe, American Scientist (September-October 1997); Kenneth R. Miller, Finding Darwin’s God: A Scientist’s Search for a Common Ground between God and Evolution (New York: HarperCollins, 1999); Robert T. Pennock, Tower of Babel: The Evidence Against the New Creationism (Cambridge, MA: MIT Press, 1999), pp. 166-72, 263-72; Mark Perakh, Unintelligent Design (Amherst, NY: Prometheus Books, 2003); David Ussery, “Darwin’s Transparent Box: The Biochemical Evidence for Evolution,” in Matt Young and Taner Edis, eds., Why Intelligent Design Fails: A Scientific Critique of the New Creationism (New Brunswick, NJ: Rutgers University Press, 2004).

8. H.J. Muller, “Reversibility in Evolution Considered from the Standpoint of Genetics,” Biological Reviews 14 (1939): 261-80.

9. Francis S. Collins, The Language of God: A Scientist Presents Evidence for Belief (New York: Free Press, 2006). For a thorough and comprehensive refutation of Collins’ book, see George C. Cunningham, Decoding the Language of God: Can a Scientist Really Be a Believer? (Amherst, NY: Prometheus Books, 2009).

10. William A. Dembski, The Design Inference: Eliminating Chance Through Small Probabilities (Cambridge: Cambridge University Press, 1998); Intelligent Design: The Bridge Between Science and Theology (Downers Grove, IL: InterVarsity Press, 1999); No Free Lunch: Why Specified Complexity Cannot Be Purchased without Intelligence (Lanham, MD: Rowman & Littlefield, 2002).

11. For refutations of Dembski’s work, see Brandon Fitelson, Christopher Stephens and Elliott Sober, “How Not to Detect Design – Critical Notice: William A Dembski, ‘The Design Inference,’” Philosophy of Science 66, no. 3 (1999): 472-88; David Roche, “A Bit Confused: Creationism and Information Theory,” Skeptical Inquirer 25, no. 2 (2001): 40-42; Jeffery Shallit, review of No Free Lunch by William Dembski, Biosystems 66, nos. 1-2 (2002): 93-99; Victor J. Stenger, Has Science Found God? The Latest Results in the Search for Purpose in the Universe (Amherst, NY: Prometheus Books, 2003), pp. 99-130.

12. Religion News Blog, “High Court Sides with Church in Hallucinogenic Tea Dispute,” ReligionNewsBlog.com 21 Feb. 2006, http://www.religionnewsblog.com/13721/high-court-sides-with-church-in-hallucinogenic-tea-dispute (accessed 27 December 2011).