Reading some of the work of prominent Intelligent Design theorists, it struck me that their coyness about what ID involves is unwarranted. ID theorists repeatedly invoke “intelligence” with knowing smiles about how intelligence could have solved all sorts of design and physical problems, offering this as a simple, easy alternative to figuring out some complex historical sequence that natural selection might have played out.
They further argue that they have no obligation to be more specific about what this intelligence is (what its motives are, what sort of being with what sort of capabilities it has, and so forth). Most critics of intelligent design have from here theistic solution. But in thinking about the problem, I realized that there is another reason to be coy about invoking “intelligence” as a solution to the origin of complex functionality.
It is because you can’t invoke a word and not have it mean anything. Either by “intelligence” ID theorists are invoking some known concept of what it means to be and operate as an intelligent being in a design setting, or they are simply using “intelligence” as semantic sheild for total ignorance of what sort of thing/process they are talking about.
Allow me to supply them with their missing model: a very general sketch of how an idealized intelligence would approach the task of designing some creature or new function for an existing creature. All intelligences we are familiar with, it seems to me, even in our imaginations, would operate along the following general steps.
First of all, they need to collect information from a given environment or physical system: the one into which they want to place their creation or even their modification of an existing creature. They need as much information as possible, and it needs to be accurate: there may be many different interlocking chains of potential cause and effect to take into account.
Next, an intelligence needs to model. Here is where intelligence shows its power: instead of actually trying out each and every possibility in the real world, it figures it out in the abstract, simplifying key elements of a design and thinking about how they might work in practice, measured against the information it collected. This is how virtually any form of intelligent thought that we can concieve of operates: without it, it would be hard to know what was meant by “thinking” or “thoughtful design” at all.
However, as any actual intelligent designer knows, this modeling process is almost never sufficient on the first run through. The abstract can never capture the subtlties of the real, and very minor causal elements or imperfections that might have escaped notice can blow up into serious and unexpected flaws that have to be worked back into and corrected for in the revised model.
Because the intelligent designer must now somehow transfer the lessons concieved back into some real physical structure, hoping that the modeling process has been accurate, that the original information gathered captured enough of the situation to avoid unexpected flaws, and so on. Almost always, it isn’t, requiring many different prototypes and revisions. Finally, the desired functionality can be hit upon, and intelligent design has done its work.
Contrast this “intelligent” method to natural selection, and you’ll notice that it doesn’t measure up very well.
As process, natural selection does away with almost all of the troublesome steps here, most importantly those having to do with information. You might remember that intelligent design theorists make a lot of hay about information: they are deeply incredulous about how it comes to be naturally, how it manages to be preserved, how it becomes modified in ways that increase rather than decrease functionality, and so on. So you’d expect that any ID alternative would offer some sort of ingenious information creation, transfer, and management features over a dumb, accumulative process like natural selection.
Instead, quite the opposite is true. In natural selection, there isn’t anywhere near the sort of convoluted back and forth shuffling of information. The creation of information is a remarkably straightforward event.
Continually accumulating trait variety in a given gene pool provides ambiguity. Natural selection then operates simply by more definitively resolving portions of this ambiguity, non-randomly favoring those traits which happen to increase reproductive fitness within a given environment. In doing so, information about the environment is transferred to gene pool, leaving the individual genomes of the next generation on average better suited to survive and reproduce in the local environment.
That’s it. And yet the process is so powerful that scientists can actually look at a given genome and read out information about the primitive environments the genome’s ancestors had passed through (for instance, that the recent ancestors of mammals were primarily adapted to a nocturnal lifestyle).
In a sense, all that natural selection shares with intelligence is the trial and error aspect, except that this step doesn’t involve the immense information challenges, and it doesn’t take place by juggling both the abstract AND real world applications. Everything happens directly in the real world, interfacing directly with all the subtle elements of the real world rather than having to worry about translation problems.
In short, the information challenges that face evolution via natural selection look, on any fair comparison, to be far less demanding and absurd than those that an even idealized intelligent process faces. If there is any sort of “second law” of information, then the operation of an intelligence violates it many many more times over than natural selection, and in far grosser and sloppier ways.
Could this be why ID theorists refuse to probe very much deeper into the potential operation of their explain-all?
A final addendum: ID theorists might object that an intelligence might not operate by the idealized means I describe. My response is that, well, I’ve at least drawn up a description of how an intelligence works that at least has a legitimate and meaningful claim to the word. If they aren’t going to supply some alternative, then they might as well be declaring that it was done by magic, and they have no right to appeal to “intelligence” as their explanation.
I’d appreciate comments, even from other critics of intelligent design, as to how to make this argument stronger or clearer.
