Intelligent-design theorists have long argued that finding complex and specified information (CSI) is the best way to reliably detect intelligent design in nature. Complexity measures the unlikelihood of an event. Specification claims that the event matches some independent pattern. Recently a correspondent wrote to ask whether we can scientifically conclude that a complex event is in fact specified.

I responded that objections to the “specification” part of CSI have been raised for many years, but that I think William Dembski has answered them forcefully and convincingly in The Design Inference and elsewhere.

If you shoot an arrow at a target, it will hit a certain point. Your hitting that precise point is in itself an unlikely event. Is that enough to infer design? It depends. If you draw the bull’s eye around the target after you hit the event, then probably not. But Dembski points out that if you drew the target before shooting the arrow, and then hit the target, then there’s a specification worthy of a design inference. So not all specifications will do.

For example, in cosmology the required specification is an objectively understandable configuration of the physical laws and constants of the universe. Not just any improbable configuration will do. You need one that allows life to exist. The vast vast majority of configurations don’t yield any or all of the following: matter, heavy elements, molecules, galaxies, stars, solar systems, habitable planets, or even elements like carbon. So it’s not hard to understand the specification required for cosmic design: you need a configuration that produces a life-friendly universe. Thus, the laws of the universe exhibit high CSI.

In biology, specification is also easy to understand. The relevant specification in biology is functionality. Folks on both sides of the evolution debate marvel at how biological features are tightly specified to match what is required for functionality. This is not controversial. The fact that it takes agents with relevant background knowledge to discern the specification required for functionality is not a problem in this or most other relevant situations. Here’s a key passage from Dembski:

The irreducibly complex systems [Michael] Behe considers require numerous components specifically adapted to each other and each necessary for function. Such systems are not only highly improbable, but also specified in their function. Biological specification always denotes function. An organism is a functional system comprising many functional subsystems. The functionality of organisms can be cashed out in a number of ways. Behe cashes it out in terms of the minimal function of biochemical systems. Arno Wouters cashes it out globally in terms of the viability of whole organisms. Even the staunch Darwinist Richard Dawkins will admit that life is specified functionally, cashing out functionality in terms of the reproduction of genes. Thus, in The Blind Watchmaker, Dawkins writes, “Complicated things have some quality, specified in advance, that is highly unlikely to have been acquired by random chance alone. In the case of living things, the quality that is specified in advance is … the ability to propagate genes in reproduction.”

(William Dembski, No Free Lunch, p. 151)

Also consider this:

Does the bacterial flagellum exhibit specified complexity? Is such a claim assertible? Certainly the bacterial flagellum is specified. One way to see this is to note that humans developed bidirectional motor-driven propellers well before they figured out that the flagellum was such a machine. This is not to say that for the biological function of a system to constitute a specification, humans must have independently invented a system that performs the same function. Nevertheless, independent invention makes all the more clear that the system satisfies independent functional requirements and is therefore specified. At any rate, no biologist I know questions whether the functional systems that arise in biology are specified.

(William Dembski, The Design Revolution, pp. 110-111)

So specification is related to biological functionality, allowing an organism to survive and reproduce.

I get a sense that some ID critics assume that if something can’t be expressed in mathematical terms, or isn’t otherwise quantitative, then it isn’t science. Clearly the requirement does not always apply — and I don’t think that specification is at all a problem for CSI.

Specification might be qualitative, but being qualitative isn’t necessarily the same thing as being wholly subjective. Specification is both qualitative and objective.

For anyone who doubts that qualitative things can be objective, try this experiment. Go home tonight and tell your wife (or husband), “Because our relationship isn’t a quantifiable mathematically definable entity, it’s entirely subjective. We can’t say it exists in any objective sense.” If you spend the night on the couch, I’m right.

Or perhaps you’d be better advised not to try the experiment after all. Instead, consult this reference list of sources where Dembski has dealt with the issue:

1. The Design Inference, pp. 56-57
2. Intelligent Design: The Bridge Between Science and Theology, p. 149
3. No Free Lunch, pp. 148-149, 250-252
4. The Design Revolution, pp. 100-105, 110-111