Two of the criteria for a theory being a scientific theory are that the theory be testable, and that the theory be capable of making predictions. Newton’s laws predict how moving objects behave, Newton’s theories can be tested by observing how objects in motion behave, and they predict how future interactions among objects in motion will transpire under prescribed conditions.

Creationism postulates that only a more complex entity can create a complex system. Since living organisms are complex systems, there must therefore be a Creator that is more complex. This is the watchmaker argument 1: if you find a watch on the beach it implies the existence of a watchmaker. Creationists argue that the theory is testable because it could be disproved by finding examples of complex systems that occurred solely as a result of known and well-understood “simple” forces of nature, and that it has predictive power because it predicts that no such systems will ever be found.

Let us begin with an example that is easier to dissect. Examples involving black swans 2 have been around for a long time, and I will march out the swans for this occasion. All the swans in Europe and North America are white, so for a long time people believed there were no black swans. Ultimately black swans were discovered in Australia, much to the surprise of Europeans.

Let us put the situation in the form of a theory: There are no black swans. It is testable, and has the power of predicting that no black swans will ever be found. The theory is established by examining swans. Here is a swan, it isn’t black, so that is evidence the theory is true. Here is another one, confirmed again. It is confirmed thousands upon thousands of times, perhaps millions of times. If you see a tree or a house, those too are not black swans. The theory that there are no black swans thus has billions of confirmations. A theory with billions of confirmations must be a very solid one, right? Supposing that such a well-confirmed theory is solid is the confirmation fallacy.

Eventually, a black swan is sighted. But a theory with billions of confirmations is believed unlikely to be wrong. There are ways to attempt to preserve the theory. One is to declare that what appears to be a swan is not a swan. It cannot be a swan, because no swan is black, and this one is black, therefore it is not a swan. Another resolution is to suppose that the black swan did not occur in nature, but rather was created by a Swan Creator who placed in in Australia without regard for the laws of nature.

black swan, Melbourne, Australia

Accepting either of the no-black-swan theory resolutions means that the theory was never testable in the first place. The first resolution amounts to defining swans as being black, and definitions are human creations that cannot be tested. The second resolution places all exceptions as being outside of the realm of the theory.

To disprove the Creationist theory, we must find a complex system in nature that was not created by more complex creator. How about the universe? That is acknowledged as a complex system. No, it is not allowed as an exception. It must instead be a thing created by a higher Creator. How about evolving life forms? No, that is not an exception, it too proves the existence of a Creator.

If the theory is a scientific theory, then the Creator must too have a more complex creator of its own. If the Creator is somehow outside of the natural world, and hence not subject to the laws of nature, then that grants that the theory is not a scientific theory, but rather a religious theory. That concedes the debate.

Going back to the black swan example, there was another problem. The theory “There are no black swans” makes a prediction of sorts, but it isn’t a prediction of the kind that propounds a scientific theory. Scientific theories deal with causes related to laws of nature, and they go beyond mere observation. Many low probability events will never have been observed in the span of human history, nor are they likely to be observed in future thousands of years. The probability of the cards of any one hand of bridge arriving to a player in a particular order is so low as to never be observed, but four of such utterly unpredictable hands arrive to players with every deal of the cards. This does not prove that a Bridge God is at work, it proves that generally cards arrive in accordance with the theory of probability. (Generally. Mae West: “Oh, is this a game of chance?”; W.C. Fields: “Not the way I play it.”)

Some creation scientists understand that scientific theory requires explanations related to causality. They set out to establish a proof along the lines of “A system with complexity C cannot be produced as a natural evolution according to a simple set of rules R, where C is a properly defined measure of complexity, and R is a set of rules with properly defined measures of simplicity.” That is my interpretation of what I think they are saying; they may differ with my interpretation. In any case, such a formulation would then make Creationism a scientific theory in the sense of being testable at some level. No such formulation currently exists.

The more scientific formulation, if it can be made, does not solve the problem of the Creator being a counter-example that disproves the theory. Creationists are careful not to say that the Creator is God. That leaves open the possibility of some other type of creator; I suppose space aliens or creatures from another dimension. That does not solve the problem. Whomever the Creator, according to the Creationist theory, it is a complex system within nature that came into existence under the theory that greater complexity cannot result from anything simpler. Tracing the creator of the creators back to God solves the problem, but only by claiming an exemption to the laws of nature. We then have a religious theory, not a scientific theory.

Understand that none of this bears on the existence of God. If God exists and is really smart, he might choose to create the universe and all of life by setting up a relatively few simple rules and letting the system evolve. That would be a much more clever God than the one that has to keep intervening to set things right.


1. Dawkins, Richard, The Blind Watchmaker: Why the Evidence of Evolution Reveals a Universe Without Design, 1996

2. Taleb, Nassim, The Black Swan: The Impact of the Highly Improbable, April 17, 2007, is an excellent book that discusses the confirmation fallacy at length.