University of British Columbia at Vancouver philosophy professor Richard Johns has published an article in the philosophy journal Synthese titled “titled “Self-organisation in dynamical systems: a limiting result,” which argues that there are “limitations on the kinds of structure than can self-organise.” He defines a self-organized object as follows:

1. The appearance of the object does not require a special, “fine-tuned” initial state.
2. There is no need for interaction with an external system.
3. The object is likely to appear in a reasonably short time.

Richard Johns, “Self-organisation in dynamical systems: a limiting result,” Synthese (Sept. 9, 2010).

Johns’ primary argument is to prove a “limitative theorem” that certain types of objects cannot self-organize through the laws of nature:

Limitative Theorem A specific large, maximally irregular object cannot appear by self organisation in any dynamical system whose laws are local and invariant.

His limitative theorem entails ideas very much like Dembski’s conservation of information. According to Johns, just as there are logical limits to the amount of information that can be derived from a given set of axioms or premises, there are physical limits to the kinds of structures can be derived from a given set of physical laws:

The basic idea of these conservation theorems is that logical consequence requires the premise to contain more information than the conclusion. (Note that this is only a necessary condition for logical consequence, not a sufficient one.) As Chaitin (1982, p. 942) put it: “I would like to be able to say that if one has ten pounds of axioms and a twenty-pound theorem, then that theorem cannot be derived from those axioms.”

The article has already received some attention at Uncommon Descent where Dr. Johns elaborates on his argument: “The rough idea is just that a complex object is one that the system cannot produce without a large ‘input’.”

According to Dr. Johns, given the existence of a self-replicator, Darwinian evolution is “a case of conditional self-organisation (i.e. conditional on a self-replicator in the initial state).” He argues that Darwinian selection is an inadequate explanation because any intermediate states are either far from the initial state or far from the final state:

Thus, through this counter-example, we have identified a serious fallacy in the gradualist argument. The individual possibility of each small step occurring in a short time does not entail the possibility of the entire sequence of steps each occurring in a short time. We should therefore be wary of any general argument that seeks to show that a complex object can be produced gradually, by a cumulative process, far more rapidly than the Limitative Theorem allows.

The paper concludes that even if life evolved in a general sense, “the Limitative Theorem does suggest that the currently recognised processes driving evolutionary change are incomplete.” At UncommonDescent he goes on to write the following:

While the paper doesn’t address intelligent design as such, it indirectly establishes strict limits to what such evolutionary mechanisms as natural selection can accomplish. In particular, it shows that physical laws, operating on an initially random arrangement of matter, cannot produce complex objects with any reasonable chance in any reasonable time. … My argument is not especially concerned with the creative powers of natural selection, since it covers self-organisation in general. But the limitative theorem does entail that natural selection cannot have the powers that are often claimed. In this respect my argument is similar to, for example, Michael Behe’s argument involving the notion of irreducible complexity (e.g. in Darwin’s Black Box).