Quantum Diaries

– By Byron Jennings, Theorist and Project Coordinator

Pierre-Simon, marquis de Laplace (1749 – 1827) was one of the great French mathematical physicists. In math, his fame is shown by the number of mathematical objects named after him: Laplace’s equation, Laplace transforms, the Laplacian, etc.  In physics, he was the first to show that planetary orbits are stable and he developed a model—the nebular model—to account for how the solar system formed.  In modified form, the nebular model is still accepted. In spite of these important contributions, he was also very much a lackey, being very careful to keep on the right side of all the right people. During the French revolution, that might have been just good survival strategy. After all, he served successive French governments and, unlike Lavoisier, kept his head.

Laplace presented his definitive work on the properties of the solar system to Napoleon.  Napoleon, liking to embarrass people, asked Laplace if it was true that there was no mention of the solar system’s Creator (ie God) in his opus magus. Laplace, on this occasion at least, was not obsequious and replied, “I had no need of that hypothesis.” This is essentially the simplicity argument discussed in a previous blog, but stated very crisply and succinctly.

Laplace was not just a whistlin’ Dixie. Newton had needed that hypothesis, ie God, to make the solar system work. Newton believed that the planetary orbits were unstable and unless God intervened periodically, the planets would wander off into space. Newton had not done the mathematical analysis sufficiently completely. Laplace rectified the problem. Newton also had no model for the origin of the solar system. Laplace eliminated these two gaps that Newton had God fill.

Back to Napoleon—he told Joseph Lagrange (1736 – 1813), another of the great French mathematicians/physicists, Laplace’s comment about no need for the God hypothesis. Lagrange’s reply was, “Ah, it is a fine hypothesis; it explains many things.” Laplace’s apocryphal reply was, “This hypothesis, Sir, explains in fact everything, but does not permit to predict anything. As a scholar, I must provide you with works permitting predictions.” This is the ultimate insult in science: it explains everything but predicts nothing. Explanations are a dime a dozen; if you want explanations, read Kipling’s Just so Stories. Now, there are some fine explanations. I particularly like The Cat That Walked by Himself.

Lapalce’s argument, I had no need of that hypothesis, is still being used today. Hawking and Mlodinow in their book, The Grand Design, created a stir by claiming God did not exist. But their argument was just Laplace’s pushed back from the beginning of the solar system to the beginning of universe:  they had no need of that hypothesis.  Whether their physics is correct or not is still an open question. It is not clear that string theory has gotten past the “it explains everything but predicts nothing” stage.

An alternate approach to understanding God’s absence in scientific models is methodological naturalism. The term seems to have been coined by the philosopher Paul de Vries, then at Wheaton College, who introduced it at a conference in 1983 and published it in the Christian Scholar’s Review.  It has since then become a standard definition of science, even playing a significant role in court cases, most notably the case [1 in Dover Pennsylvania on teaching creationism in public schools. The judge mentioned methodological naturalism prominently in his ruling.

Methodological naturalism, as a definition of the scientific method, is rather ill defined except for its main idea, namely that science, explicitly, by fiat, and with malice a-fore-thought, rejects God, gods, and the supernatural from all its considerations. There is frequently an implicit secondary idea that science is about finding explanations but only natural ones, of course. Both ideas are inconsistent with what science actually is: building models constrained only by observation and parsimony. (See above and the previous blog for my opinion of the role of explanations in science.)

However, methodological naturalism is a very convenient hypothesis. It avoids awkward questions about the relation between science and religion. By inserting naturalism into the very definition of science, methodological naturalism, if valid, would create a firewall between science and religion. This would both protect religion from science and scientists from the religious. Considering the violence done in the name of religion, the latter may be more important, but the former was probably part of the original intent.  However, I suspect the main motivation was to explain why God and the supernatural are absent from science.  But Laplace gave the real reason for God’s absence: parsimony—there is no need of that hypothesis. There are probably also very good theological reasons for that absence but that is outside the scope of science and this blog.

Methodological naturalism confuses the input with the output. To the extent science is naturalistic, it is an output of the scientific method, not part of the definition. Excluding anything by fiat is poor methodology. But once one realizes that historically God and the supernatural have been eliminated from science, not by fiat, but by Laplace’s criteria, methodological naturalism becomes redundant; an ad hoc solution to an already solved problem.