Scientific Laws, Probability, and Belief

On careful reflection, we can see that science and religion share the same underlying premise: “belief.” People generally say that religion is irrational and that science alone is an entirely rational system of knowledge. Religion, they argue, belongs to the realm of faith and illogic, whereas science approaches truth through logical inferences based on observation and experimentation. Ultimately, however, science also requires a consensus formed through collective intelligence and a certain kind of belief.

Consider a very simple example. Entropy, one of the most important concepts in modern physics, is a quantitative measure of “disorder.” The second law of thermodynamics describes entropy as follows: “All natural phenomena proceed irreversibly, with energy flowing in only one direction and disorder (entropy) increasing.” An important feature here is “irreversibility,” the claim that reactions do not occur in the direction of decreasing entropy. For instance, heat flowing from a colder place to a hotter place is considered impossible because it would decrease entropy.

However, one fact overlooked by this description of the second law of thermodynamics is that the increase of entropy in nature is ultimately based on overwhelming “probability.” The consistent direction of heat flow is determined by probability, and the physicist Ludwig Boltzmann therefore argued that it was not an absolute truth.

For example, there is a probability that, contrary to the usual statistical tendency of entropy, all the air molecules in a room could gather in one corner, causing a person on the opposite side to suffocate. The probability of heat transferring from a cold place to a hot place is extremely small, but it certainly exists. An absolute truth must apply without exception under every circumstance. In mathematics, for instance, the proposition “1 + 1 = 2” can be considered an absolute truth because it has no “possibility of falsification” within its mathematical framework. By contrast, the claim that “all natural phenomena occur only in the direction of increasing entropy” may generally be correct, but it cannot be an absolute truth because there remains even the slightest possibility of a counterexample.

Yet it seems that this law is far too accurate to be described merely as an “inference.” There is a definite reason why the reversal of entropy is usually disregarded when researchers conduct studies and establish new hypotheses. Even if the second law of thermodynamics is, in principle, falsifiable, its violation is almost never manifested in actual scientific phenomena. In the natural sciences, numerous premises must ultimately be treated as implicit assumptions for research and discussion to be possible. Furthermore, because the probability of entropy reversing is exceedingly low, it is more reasonable to regard the assumption that the law will always hold as a “belief” rather than an “inference” leading to a new conclusion.

In conclusion, the natural sciences can develop effectively only when they presuppose an appropriate degree of belief in theories and observable phenomena. The practical circumstances described above also explain why the possibility of falsifying the second law of thermodynamics is usually ignored. As the example of entropy demonstrates, when a scientific law rests on an overwhelmingly high probability—so high that the possibility of a counterexample can safely be disregarded—the resulting knowledge is neither absolute truth nor mere inference. It is more rational to treat the law as a particular belief that it will always hold. Ultimately, this provides a basis for arguing that science and religion share an underlying premise. Although the nature of these two kinds of “belief” differs, and although belief in science may amount to a rationally grounded “practical consensus,” this insight may help bridge the otherwise separate domains of science and religion.