Natural Selection Can’t Select a Future Function

Author Amy K. Hall Published on 06/05/2015

In this short video from the Discovery Institute, Paul Nelson follows the development of a C. elegans worm from one cell to an adult, showing how “even these little worms, a millimeter long, humble little creatures out there in the compost heap...carry the signal of design unmistakably.”

The successful creation of a live C. elegans requires many intermediate cell divisions, yet the temporary cells created by these intermediate cell divisions play no functional role in the adult worm whatsoever. Instead, they merely serve as stepping stones in a long journey that will eventually reach a functional organism at its conclusion. But natural selection can’t select a future function; it can only select features that are advantageous already.

If something’s going to function in natural selection, it’s got to function now, at this particular moment in time—not five minutes from now, half an hour, a week, a thousand years. So a process that lacks foresight in principle cannot build a[n] unfolding trajectory, an unfolding lineage [of intermediate cells], where you need to know the target. That’s the fundamental difficulty for any undirected process of evolution.

What natural selection and other undirected natural mechanisms cannot achieve, intelligent agents can. Intelligent agents are able to foresee distant functional goals. Intelligent agents can coordinate and choreograph the assembly of many separately necessary parts to achieve a functional end.

When I look at animal development, I see a trajectory. It’s, in a sense, the quintessential end-directed or teleological process in nature. You’re pulling back that bowstring, and you’ve got a target over there fifty yards away, and you want to put that arrow right in the middle of that target. You need to know what you’re aiming at and why, and for that you need a mind.