Evolution Icon Evolution
Intelligent Design Icon Intelligent Design

More on Winston Ewert’s “Dependency Graph of Life” — An Important New Paper

Granville Sewell

Winston Ewert

Several articles at Evolution News have explained the significance of Winston Ewert’s important new BIO-Complexity article. I would like to contribute one more. This one is very short and simple but attempts to give a broader overview of the paper’s significance.

Since 1985 I have been arguing that the evolution of life looks much like the evolution of software and other human technology. My main argument, for example in a 2000 Mathematical Intelligencer article, pointed to the fact that according to the fossil record, major new features (new orders, classes, and phyla) appear abruptly, just as major new features in the development of, for example, my PDE solver, appear abruptly, and for the same reason. You only have to think about what gradual development of new organs, or new systems of organs, would look like, or what the gradual development of major new software features would look like, to understand why they must appear abruptly: intermediate stages would usually have to involve incipient new, but not yet useful, features.

Evolution of Life, and of Software

More recently I have called attention to another way in which the evolution of life mimics the evolution of software or other human technology. (For example, in the last segment of this video, or in “I Believe in the Evolution of Life and the Evolution of Automobiles.”) In both cases, similar new features frequently arise independently in distant branches of the “evolutionary tree.” This is called “convergence.” For example, when Ford automobiles and Boeing jets evolve similar new GPS systems, or when bats and whales develop sonar echolocation independently, they are converging and becoming more alike, at least in one attribute, rather than diverging and becoming more distinct. My video notes that the basic types of carnivorous plant traps each evolved multiple times independently.

This phenomenon of convergence is so ubiquitous that it has become a major problem for evolutionists. As Cornelius Hunter points out in “The Real Problem with Convergence,” the problem is not only that it is hard to believe that very different species would develop similar new features independently: anyone who is able to believe that eyes developed though random processes once will find a way to believe they developed multiple times by chance, as Hunter says. The problem is it destroys the tree of life! Contrary to what the textbooks tell us, the similarities among species do not really point to a strict tree structure of common descent. They look more like the way a designer creates new software or technology products: a designer is free to reuse software modules or pieces of engineering technology from multiple previous products, not just from direct “ancestors” of the new product. New species and new products do, nevertheless, often inherit much of their “technology” from one ancestor.

Not the Best Explanation?

Winston Ewert has now provided evidence in a quantitative and objective way, by examining several publicly available protein data sets, that a tree of life may not be the best explanation for the similarities between species. Instead, a dependency graph, like that used to document the dependencies of new software products on modules in earlier products, thus explaining the similarities between these products, appears to be a much better model for the similarities between species.  Further work is needed to confirm his results.

The bottom line is: the pattern of similarities between species looks more like that arising when intelligent humans develop software or other technologies, and less like that expected to arise from Darwin’s tree of descent with modification. And we have not even mentioned the strongest and most obvious connection between the evolution of computer programs and the evolution of the DNA code in living things. That is the absurdity of attributing the information contained in either to anything other than an intelligent programmer!

Editor’s note: For more on Winston Ewert’s paper, see:

Photo: California pitcher plant (Darlingtonia californica), a carnivorous plant, by Noah Elhardt, via Wikimedia Commons.