Stephen Meyer’s definition of intelligent design is unfair, according to Barry Arrington

A while back, Barry Arrington challenged critics of intelligent design to define intelligent design, claiming that

I have never seen a fair summary of ID theory come from one of our opponents.

Several ID critics(including me) weighed in with our definitions, but Arrington called all of them “superficial and contemptuous” (my answer was apparently so superficial and contemptuous that it got me banned from commenting at Uncommon Descent). I pointed out at the time that some of these answers were virtually identical to the definitions given by prominent ID proponents.

Stephen Meyer, author of Darwin’s Doubt, founding member of the Discovery Institute, and occasional contributor to Evolution News and Views, has cleared things up for us. Here’s his definition of intelligent design (around 1:58 in this recording):

The theory of intelligent design is the idea that there are certain features of life and the universe that are best explained by a purposive intelligence, rather than an undirected material process such as, in the realm of biology, natural selection acting on random mutations.

Stephen C. Meyer

Dr. Stephen C. Meyer. Discovery Institute press photo from

For purposes of comparison, here are some of the “superficial and contemptuous” definitions provided by intelligent design critics in response to Arrington’s challenge:

The theory of intelligent design holds that certain features of the universe and of living things are best explained by an intelligent cause, not an undirected process such as natural selection. — DaveS

Intelligent design advocates argue that some features of the natural world are best explained by the action of some intelligence, rather than natural and/or undirected forces. In biology this goal is usually pursued by demonstrating that biological systems (including particular proteins, other gene products and interactions among these molecules) could not have have been generated by the biological processes known to generate and filter genetic diversity. — wd400

Intelligent design is the proposition that some features of the natural world, especially of living things, are best explained by the influence of an intelligent agent (in some accounts, the agent in question is supernatural). — me. This is part of the answer that got me banned from Uncommon Descent.

Let’s break that down a bit, shall we? Here’s a piece-by-piece comparison between Dr. Meyer’s definition and those quoted above (in parentheses):

The theory of intelligent design is the idea that (holds that/advocates argue that/is the proposition that)

there are certain features of life and the universe (certain features of the universe and of living things/some features of the natural world/some features of the natural world, especially of living things)

that are best explained by a purposive intelligence (best explained by an intelligent cause/best explained by the action of some intelligence/best explained by the influence of an intelligent agent)

rather than an undirected material process such as, in the realm of biology, natural selection acting on random mutations (not an undirected process such as natural selection/rather than natural and/or undirected forces/my answer didn’t include this).

These are all the same definition. The wording varies a bit, but they are all fairly described as paraphrases of one another.

What does it say about your theory when an honest description, substantially identical to that given by one of its founders, sounds to you like a “superficial and contemptuous” straw man?


  1. Bruce says

    I think the word supernatural threw them. It sounds to some like a story that might involve ghosts, rather than a story that might involve a Holy Ghost, which would obviously be completely different. Suggesting otherwise shows our “contempt”.
    Thanks for speaking up for reason.

    • Matthew Herron says

      Only some of the “superficial and contemptuous” answers said anything about the supernatural, though. Mine did, but it specified that this was “in some accounts,” which is undeniably true.

    • Matthew Herron says

      He doesn’t actually give one himself (at least not in that post or its comments), but he does approve one rather long one by Eric Anderson:

      The argument for design in any particular instance is a combination of at least two points; really two points, with sub-parts.
      1. The positive argument.
      Certain designed things exhibit particular characteristics that are generally recognizable as indicia of design. Many of these characteristics show up in abundance in some biological systems. Such systems would therefore, on their face, appear to be designed, as even the most vociferous naturalistic proponents, such as Dawkins, acknowledge.* Additionally, some intelligent beings are known, on the evidence, to have the ability to produce such features.
      Furthermore, as a sub-argument/observation, we observe that in every instance in which we know the historical provenance of such features, the source inevitably turns out to be an intelligent cause. Therefore, the most reasonable inference is that those features were likely designed.*
      2. The negative argument.
      Purely natural causes have never been shown to produce those kinds of features. Therefore, there is no reason, on the observational evidence, to believe that they can.
      Furthermore, as a sub-argument/observation, there are excellent practical and theoretical reasons to conclude that purely natural causes are not, in principle, capable of producing those kinds of features. Therefore, the most reasonable inference, is that they did not.

      • rjdownard says

        As if being more verbose made the vacuity of the underling concept any less glaring, especially regarding the glaring failure of ID to proactively pioneer any useful knowledge of anything. All we get is a pile of parsing of the data field, which is methodologically what creationists have been doing all along (and just being more broad-ranging in what science disciplines they strive to mangle).

  2. Owlmirror says

    Eric Anderson’s use of the word “design”, without qualification as to intelligence, reminds me that I’ve been thinking of emphasizing that we can reject the implication that design can only ever occur by way of intelligence; that natural processes can and do design without intelligence being involved, and that evolution is an example of such a natural process.

    Hm. Perhaps more rigorously, design can result from iterated, cyclical processes operating over time.

    Daniel Dennett does something similar when he talks about bottom-up design (by nature) vs. top-down design (by humans). I may have read Dennett on the topic, but I didn’t remember it when I first started thinking about it. Also, Dennett seems to have been talking specifically about design by evolution (which sort of takes it for granted), whereas I was thinking more broadly about non-biological processes (so that biology can be understood as a subset of such designing processes).

    Examples chosen to avoid biology:

    Frost on a windowpane is designed by iterated condensation and crystallization of water vapor in freezing temperatures.

    Rivers are designed by the repeated processes of snowpack formation during winter in the highlands the flow down from, spring and early summer meltwater flowing down, the sediment load brought down by the meltwater, and the interaction of those waters with the local geography, as well as tributaries flowing in, and occasional additional inputs from flood events.

    Galaxies are designed by the constant gravitational interactions of the various masses composing the galaxy.

    Probably more can go here.

    • Matthew Herron says

      I think they would argue that while the frost, etc. are possibly complex, they don’t possess ‘specified’ complexity (or complex specified information).

  3. Owlmirror says

    It’s been a while, but I seem to recall that there was a problem on the ID side of defining “specified complexity”/”complex specified information” clearly and unambiguously. The specifications remained nonspecific.

    • Matthew Herron says

      It consistently means whatever allows William Dembski to infer design. On the one hand, he says

      A Shakespearean sonnet is both complex and specified.

      On the other hand, “METHINKS IT IS LIKE A WEASEL” is not, at least not when it’s produced by an evolutionary algorithm.

Leave a Reply