Home| Letters| Links| RSS| About Us| Contact Us

On the Frontline

What's New

Table of Contents

Index of Authors

Index of Titles

Index of Letters

Mailing List

subscribe to our mailing list:


Critique of Intelligent Design

Evolution vs. Creationism

The Art of ID Stuntmen

Faith vs Reason

Anthropic Principle

Autopsy of the Bible code

Science and Religion

Historical Notes


Serious Notions with a Smile


Letter Serial Correlation

Mark Perakh's Web Site

Arguments out of context

Analyzing a Rhetorical Trick Used by William Dembski

By Jason Rosenhouse

Posted April 22, 2004

The creationist propensity for removing quotations from their proper context has been amply documented elsewhere. More recently, however, they have taken to removing whole arguments from their proper context.

Some of the claims made by ID proponents are purely scientific. They will say, for example, that irreducibly complex machines, by which they mean systems composed of several well-matched, indispensable parts, can not be formed gradually via natural selection. In reply evolutionists point out that there certainly is no theoretical reason why such machines can not be formed gradually, since there are a wide variety of indirect routes through which evolution could have acted. They also point out that for many such systems (the Krebs cycle, circadian clock genes, the blood clotting cascade, and the immune system, among many others) we have a great deal of evidence suggesting how they did evolve. Further, on a small scale we can watch the process happening, as described by Kenneth Miller in his book Finding Darwin's God. There are also the artificial life experiments in which irreducibly complex structures are routinely seen to evolve by way of random variation and selection. Alternatively, an ID proponent will argue that the "No Free Lunch" theorems of optimization theory challenge the validity of evolution. In reply, evolutionists show that the NFL theorems are irrelevant to assessing the validity of evolution. They go on to demonstrate that the hypotheses of the NFL theorem are not satisfied by the goings-on of biological evolution. These are purely scientific criticisms made in reply to purely scientific claims.

But ID proponents also make philosophical claims about how science is blinded by a naturalistic bias. To this we reply in kind that science is about solving problems and that naturalistic theories are the ones that help us do that.  If they could offer some reason for thinking that science would progress by considering supernatural explanations then we would consider them.

Then there is the cultural aspect of ID.  Proponents of ID wish to persuade us that their motives are entirely scientific. This question arises in discussions of the constitutionality of teaching ID in science classrooms. In this context it is perfectly reasonable to point out that they clearly have religious motivations and that in making their case they seem to prefer using the mass media to the normal channels of scientific discourse.

And then they go on to argue that ID is a fully scientific theory that helps us make sense of the natural world. To this we reply that if they are serious then we should take the next step.  Having identified ID in the world, we should ask what could be inferred about the designer from the nature of the design. Doing so, however, forces us to confront the problem of poor design, which casts doubt on either the omnipotence or omnibenevolence of the designer.

And so it goes. But nobody argues that the religious motivations of ID proponents by themselves cast doubt on their scientific claims. No one says that instances of poor design mean the NFL theorems are irrelevant to evolution. No one says that since ID is asking us to consider supernatural explanations we should simply dismiss out of hand their arguments against natural selection.

One of ID's leading proponents is William Dembski.  In much of his writing he has presented arguments made against one aspect of ID as if they were intended to rebut something different. For example, in a recent essay for World magazine, Dembski wrote:

To see how this happened, recall how exchanges between Darwinists and the early design theorists used to go. The design theorists would go to great lengths to analyze a given biological structure, show why it constituted an obstacle to Darwinian and other materialistic forms of evolution, and lay out how the structure in question exhibited clear marks of intelligence. To such carefully drawn lines of scientific argument and evidence, the Darwinist invariably offered stock responses, such as, "There you go with your religion again" "You're just substituting supernatural causes for natural causes" "You just haven't figured out how evolution did it" "You're arguing from ignorance" "You're lazy; get back in the lab and figure out how evolution did it."

This essay was written as an experiment in "future history." The editors of World asked Dembski, and several other contributors, to hypothesize that by the year 2025 ID will have replaced evolution. The task of the writers was to explain how that happened.

Leaving aside a consideration of Dembski's tone, notice how he presents arguments made in reply to cultural and philosophical points as if they were meant to address scientific claims. Let us consider each of the "stock" responses Dembski mentions.

There you go with your religion again. It is true that evolutionists argue that ID owes far more to religion than it does to science. Consequently, it should not be taught in science classes. This claim is not made in reply to the scientific arguments of ID proponents.

You're just substituting supernatural causes for natural causes. It is indisputable that ID proponents do, in fact, substitute supernatural causes for natural causes. This is relevant because supernatural theories can not be tested in the lab or the field.  Given this, it is difficult to see how they can plausibly be considered scientific. Again, this has nothing to do with the scientific claims of ID.

You're arguing from ignorance. This argument relates to the philosophical apparatus proposed by Dembski for detecting design in biological organisms. His method is purely eliminative. Applying it would require God-like knowledge of all possible naturalistic explanations for a given phenomenon.  Consequently, in its most charitable formulation, Dembski is simply arguing that since no one is currently able to provide a naturalistic explanation for a phenomenon we must chalk it up to supernatural design. This has never been a good argument in the past. Furthermore, the problem confronting evolutionists is the multitude of possible explanations for given structures, not the absence of any possible explanation.

You just haven't figured out how evolution did it. Dembski is so fond of this point, he repeated it in his litany. The point here is the same as in the previous paragraph. In principle, there are many possible naturalistic explanations for biological complexity. Chalking everything up to design is neither warranted nor helpful.

Of course, ID proponents do not go to great lengths to analyze biological structures. Dembski himself has applied his design-detection apparatus to precisely one biological structure: the bacterial flagellum. In doing so he described a probability calculation so transparently nonsensical that evolutionists had no need to resort to rhetorical tricks in replying (you will find this calculation in his book No Free Lunch). Scientists simply pointed out that Dembski's calculation was based on numerous false assumptions, and was consequently worthless. The only other instance of ID proponents considering actual biological systems occurs in the writings of Michael Behe. His analysis consists, in its entirety, of the undisputed assertion that many biological systems require all of their parts to function properly. This observation is hardly an example of "carefully drawn lines of scientific argument and evidence". Dembski has exaggerated the scientific content of ID writing and has misrepresented the arguments of ID proponents.

A second example comes from Dembski's recent book The Design Revolution. Consider this statement:

The word "intelligent" has two meanings. It can simply refer to the activity of an intelligent agent, even one that acts stupidly. On the other hand, it can mean that an intelligent agent acted with skill and mastery. Failure to draw this distinction results in confusion about intelligent design. This was brought home to me in a radio interview. Skeptic Michael Shermer and paleontologist Donald Prothero were interviewing me on National Public Radio. As the discussion unfolded, I was surprised to find that how they used the phrase "intelligent design" differed significantly from how the intelligent design community uses it.

Shermer and Prothero understood the word intelligent in "intelligent design" in the sense of clever or masterful design. They therefore presumed that intelligent design must entail optimal design. The intelligent design community, on the other hand, understands the intelligent in "intelligent design" simply as referring to intelligent agency (irrespective of skill or mastery) and thus separates intelligent design from optimality of design. (P. 57).

Of course, no one thinks that ID entails optimal design. Anyone who has ever looked at an SUV understands that something can be designed by an intelligent agent yet not be optimal. It's not that complicated a point.

Shermer and Prothero were simply taking Dembski seriously when he claims to want to do actual science from an ID perspective. Surely once you have determined that an object was designed by an intelligent agent, it is reasonable to ask what you can infer about the designer from the nature of the design. As soon as we do this we are confronted with all the famous examples of poor design in nature. They point strongly to the sort of stupid designer Dembski admits as a possibility.

The trouble is that ID proponents think they know who the designer is, and stupidity is not one of His attributes. Rather, they believe in an omnipotent, omnibenevolent designer. You would not infer such a designer from nature alone, unless you had a prior religious commitment to such a view.

Not only must ID proponents explain why God (let's drop the subterfuge, OK?) would design so poorly, they must also explain why so many instances of poor design seem to be just the sort of thing that evolution by natural selection produces as a matter of course. The numerous extinctions recorded in the fossil record look like simple waste as the products of ID, but are the expected outcome of natural selection acting over long periods of time. Our weak back muscles are easy to understand if they evolved among creatures that walked on their knuckles and swung through trees, but hard to understand as products of ID.

Curiously, old school creationists have an answer to precisely this objection. They say that the poor design we see does not reflect God's intentions, but rather represents the influence of sin in the world.

Dembski hints at this solution himself:

If we think of evolution as progressive in the sense that the capacities of organisms get honed and false starts get weeded out by natural selection over time, then it seems implausible that a wise and benevolent designer might want to guide such a process. But if we think of evolution as regressive, as reflecting a distorted moral structure that takes human rebellion against the designer as a starting point, then it's possible a flawless designer might use a very imperfect evolutionary process as a means of bringing a prodigal universe back to its senses. But this is an idea to be explored in another book. (P. 62).

Leaving aside the fact that this makes no sense (since human rebellion happened after humans appeared, while most of evolution happened before humans appeared, what could it possibly mean for evolution to reflect a distorted moral structure?) it is a blatantly religious argument and not one upon which ID proponents usually rely.

The only alternative is to argue that the instances of poor design that we see in nature actually reflect good design, or at least optimal design under the circumstances. That we perceive them as poor design reflects some deficiency in our understanding. Thus they will offer this or that hypothesis about why backward eye-wiring is a wonderful thing or why the panda's thumb, in fact, could not really be improved. Dembski avails himself of this option as well:

Design is a matter of tradeoffs. There's no question that we would like to add or improve existing designs by conferring additional functionalities. It would be nice to have all the functionality of the human eye without a blind spot. It would be nice to have all the functionality of the respiratory and food-intake system as well as a reduced incidence of choking. It would be nice to have all the functionality of our backs and a decreased incidence of back pain. ...But when the suboptimality objection is raised, one invariably finds only additional functionalities mentioned but no details how they might be implemented.  And with design, the devil is in the details.  (P. 60).

But this simply will not do. Not as long as we are considering an omnipotent, omnibenevolent designer. And for an obvious reason, too. There are no constraints on omnipotence. Asking God to make our back muscles a little stronger is not exactly like asking him to build a rock so heavy even he can't lift it. What engineering constraints exist are there only because God made them that way. It is a simple fact that animals endure an enormous amount of pain and suffering every day simply because of their poor design. And if God sometimes intervenes to give animals a useful structure (a flagellum, say, or a blood clotting cascade) then there is no obvious reason why He can't intervene to prevent a poor structure from emerging. I suspect it is these sorts of considerations that lead many people to theistic evolution.

Thus, disteleology is a problem for ID's because of what it seems to imply about the designer. If ID's were serious about turning ID into a science they would investigate this question in a serious way. Instead they make irrelevant arguments about suboptimal design not precluding ID.

The scientific claims of ID proponents are false, their philosophical arguments are nonsense, and their cultural arguments are potentially dangerous. Desperate to divert attention from this obvious fact, they blow as much smoke as possible to distract people from the emptiness of their claims. One tactic they use in this regard is to misrepresent the substance of particular anti-ID arguments. Careful reading is required to prevent them from getting away with this.


Dembski, William A. "The New Age of Information, World, April 3, 2004

Dembski, William A, The Design Revolution, InterVarsity Press, 2004.

Dembski, William A, No Free Lunch, Rowman and Littlefield, 2001.