Evolution Icon Evolution
Intelligent Design Icon Intelligent Design

What Does “Life’s Conservation Law” Actually Say?

Winston Ewert

Life's Conservation Law.jpg

In a series of posts, I am examining criticisms from Joe Felsenstein (University of Washington geneticist) and Tom English (Oklahoma City computer scientist) in response to two arguments for intelligent design: specified complexity and conservation of information. This is the second post in the series. Find the first here.

In an earlier article at Evolution News (“These Critics of Intelligent Design Agree with Us More Than They Seem to Realize“), I emphasized that while the theorems of conservation of information show that active information must derive from a non-mechanistic source outside of the universe, this does not necessarily imply that the source was intelligent or even teleological. In response, Tom English accuses me, in his post “The Law of Conservation of Information is defunct,” of attempting to surreptitiously switch claims made in the past about conservation of information for new ones. Felsenstein, in “Why ID Advocates Downplay Our Disagreement with Them,” avoids making this accusation. However, despite my explicit statement that “It is not our intent to argue that all active information derives from an intelligent source,” he states that I’m still in fact arguing that all active information derives from an intelligent source. What, again, is the truth?

Active information is merely bias towards a particular outcome. If a coin turns up heads more often that it turns up tails, that is active information. Demonstrating that active information can exist without teleology or intelligence is straightforward. Consider a universe that begins with a random distribution of matter and operates with a single physical law, gravity. Eventually, all of the matter in this universe will be clumped together roughly in the center. With respect to the target of such a clump, that is an immense amount of active information bounded only by the size of the universe. But such a universe hardly seems teleological. There can be extreme biases towards one outcome or another, without any sort of intelligence deliberately creating that bias.

If my co-authors or I claimed that active information necessarily derives only from an intelligent source, that would be indefensible. However, contrary to what Felsenstein and English say, we did not claim that. Of course, we did use the theorems to argue for intelligent design. We are, after all, proponents of intelligent design. The idea of active information was first developed in William Dembski and Robert Marks’s paper “Life’s Conservation Law: Why Darwinian Evolution Cannot Create Biological Information.” Taken out of context, some statements in that paper could be understood as arguing for the simplistic idea that all active information has to derive from a teleological source. However, if we look at the actual argument there, we find it is more sophisticated than that.

Dembski and Marks wrote:

Where, then, does the LCI [Law of Conservation of Information] Regress end? In fact, it may not end, implying that the information that enables an alternative search to succeed in locating T was present always. Alternatively, it may end because an external information source added the information needed to locate T. One option suggests front-loading of information, the other direct input. Both options evoke intelligent design.

In this passage, they argue for intelligent design on the basis of the conservation of information theorems. However, they do not say that the only possible account of active information is intelligence. Rather, they use words like “suggests” and “evokes.” If Dembski and Marks thought that they had a mathematical proof that only intelligence could be the source, they would have said so. They say only that what they have is suggestive of that explanation.

This is reminiscent of arguments for theism based on Big Bang cosmology. The Big Bang theory indicates a start to the universe, in contrast to the once widely accepted Steady State theory that postulated an eternal universe. A universe with a beginning is suggestive of a creation event as theistic religions teach. This has led to some objections to the Big Bang. An eternal universe would have been a more natural fit for a materialistic worldview, but since a start to the universe at best merely suggests a theistic understanding, the materialist can readily accept the Big Bang while rejecting its theistic interpretation.

Similarly, the conservation of information, since it shows that all information is derived from outside of the universe, is suggestive of intelligent design. Certainly, if the math had turned out that the universe could have started with a small amount of active information which grew slowly over time, akin to the version of Darwinism championed by Richard Dawkins, this would have been a much more natural fit for a naturalistic view. However, since an intelligent source of information is at best only suggested by the theorems, it is completely reasonable for a materialist to accept the source of information as external to the universe, but not intelligent.

Later in “Life’s Conservation Law,” Dembski and Marks wrote:

When we run search algorithms in evolutionary computing, we find that these searches are inherently teleological (Meester readily concedes this point). So, we may ask, do such mathematical models adequately represent biological evolution? In these models, careful tailoring of fitness functions that assist in locating targets is always present and clearly teleological. If these models adequately represent biological evolution, then this teleological feature of fitness ought to be preserved in nature, implying that Darwinian evolution is itself teleological.

Here they argue that Darwinian evolution is teleological. But they do not say that is a necessary consequence of the conservation of information theorems. Instead, they argue for teleology in biology on the basis of teleology in models of evolution. As Dembski and Marks say in the next paragraph, “No nonteleological mathematical models of Darwinian evolution are known.” Thus either Darwinian evolution is teleological or we have no models of Darwinian evolution, making it non-scientific.

But the central claim of “Life’s Conservation Law” is in the subtitle: “Why Darwinian Evolution Cannot Create Biological Information.” Unlike the arguments regarding intelligence or teleology, this result does derive from the conservation of information theorems. Dembski and Marks show that random queries and Darwinian evolution succeed with the same probability, unless the fitness function provides information to Darwinian evolution. This means that Darwinian evolution doesn’t have an informational contribution beyond that of random search, but rather is exploiting information provided by the fitness function. If evolution is going to work, it has to have an information rich fitness function.

The argument, then, isn’t a simplistic claim that all active information has to derive from an intelligent source. Rather, it brings together several points to make a case for intelligent design. Darwinian evolution is not a sufficient explanation of how complex life arises, but it instead depends on an external information source, the fitness landscape, in order to work. All information must eventually derive from a source external to the universe, which is suggestive, but not definitive proof, of intelligent design. Finally, the Darwinian models that have been developed are teleological, incorporating an intelligent agent’s understanding of the problem to be solved. In seeking to understand the universe, then, these arguments together make a case for intelligent design.

Next: “The GUC Bug.”

Image: � BillionPhotos.com / Dollar Photo Club.

Winston Ewert

Fellow, Senior Research Scientist, Software Engineer
Winston Ewert is a software engineer and intelligent design researcher. He received his Bachelor of Science Degree in Computer Science from Trinity Western University, a Master’s Degree from Baylor University in Computer Science, and a PhD from Baylor University in Electrical and Computer Engineering. His specializes in computer simulations of evolution, specified complexity, information theory, and the common design of genomes. He is a Senior Research Scientist at Biologic Institute, a Senior Researcher at the Evolutionary Informatics Lab, and a Fellow of the Bradley Center.